For eligible charities
Investment Perspectives 2011
Section 1: 2011 Outlook
1.1 2011: A year in the Global Economy Keith Wade Chief Economist and Azad Zangana European Economist 4
1.2 What got you here, won’t get you there Alan Brown Group Chief Investment Officer 6
Section 2: Investment themes
2.1 Ageing gracefully: What the West can learn from Japan Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate 10
2.2 China’s five-year plan: The start of a new economic era? Laura Luo Fund Manager, Asian ex Japan Equities 14
2.3 The American way: The impact of ageing on the US economy Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate 15
2.4 “Nothing is permanent but change” The Economic Impacts of demographics Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate 21
2.5 The impact of climate change and resource shortages Sir David King Senior Scientific Advisor to UBS, Director of the Smith School of Enterprise 26
2.6 Biofuels Karen Shaw Climate Change Specialist 29
2.7 Currency Unions: Breaking up is not so hard Alan Brown Group Chief Investment Officer 39
2.8 Currency wars: The new weapon of choice Keith Wade Chief Economist 41
2.9 The engines of global growth beyond 2010 Virginie Maisonneuve Head of Global and International Equities 42
Section 3: Modern investment approaches
3.1 A strategic approach to investing: An alternative to passive John Marsland Fund Manager, QEP Global Equities 48
3.2 Can investors rely on equities for long term growth? Mark Humphreys Strategic Solutions and Stephen Bowles Head of Defined Contribution (DC) 53
3.3 Diversification in times of crisis Neil Walton Head of Global Strategic Solutions and Jonathan Smith Strategic Solutions Analyst 55
3.4 The power of technical analysis Jamie Fairest Fixed Income Fund Manager 60
3.5 The role of equity investing in a modern portfolio Neil Walton Head of Global Strategic Solutions 64
Section 4: Risk management
4.1 Risk Management is more than just tracking error QEP Team 72
4.2 The Cyclical Framework: Cycling Around Asset Allocation Harish Vekaria Quantitative Analyst, Multi-Asset 77
4.3 Lies, damn lies and (risk) statistics for pension funds Anthony Earnshaw LDI (Liability Driven Investment) Fund Manager 85
The views and opinions contained within these articles are of the person(s) quoted, and may not necessarily represent views expressed or reflected in other Schroders communications, strategies or funds.
“Eligible charities” for these purposes are existing Schroders Charities clients or charitable organisations who are registered with the Charities Commission, or exempt charities, or excepted charities, or other persons eligible to participate in collective
investment schemes constituted under the Church Funds Investment Measure 1958 or section 24 of the Charities Act 1993, or equivalent organisations in Scotland or Northern Ireland. Certain investments or services to which this material relates may
only be available to such persons. This material must not be relied upon or acted upon by any other persons in the United Kingdom.
There was a definite sense of déjà vu with 2010 ending in a similar way to how it began, with a rally in Our third section is focused on modern investment approaches. Here we learn about a strategic approach
markets and a recovery in the world economy. 2010 has provided a real roller coaster ride. There was a to investing in equities from John Marsland. Neil Walton and Jonathan Smith talk about the art of
hopeful start in the first few months of the year with it looking like we were in for a fair recovery and with diversification and how it was affected during the financial crisis and Jamie Fairest discusses the
anticipation that interest rates would rise and the central banks would begin to initiate exit strategies from power of technical analysis.
loose monetary policies.
Our fourth and final section is on risk management, where the QEP Global Equities team discuss the
However, by spring, the euro crisis had erupted with the world economy showing signs of slowing down and process of risk management and what factors go into it. Harish Vekaria tells us about the cyclical framework
any optimism was soon replaced with pessimism. The forecast changed to the possibility of a double dip and provides an update on the development and drivers of the recovery phase. Finally to finish the journey,
and authorities were planning what they could do to provide stimulation. The approach to the end of the Anthony Earnshaw discusses how a Pension Scheme’s funding level will change over time and how this can
year has seen a second bout of quantitative easing, in an attempt to aid the recovery and revive markets. be measured to test the future financial health of a scheme.
With this background, our economists and Chief Investment Officer look to the year ahead and give their We hope that these articles will provide a comprehensive and informative snapshot of what we can expect
outlook for 2011 and beyond. in 2011 and beyond.
Following these we have a set of articles on current investment themes. They include, Virginie Maisonneuve and Charities Team
Katherine Davidson sharing with us what the West can learn from Japan and James Gautrey educates us on
Organic Light Emitting Diodes (OLEDs).
1.1 2011: A year in the Global Economy 4
1.2 What got you here, won’t get you there 6
1.1 2011: A year in the Global Economy
Keith Wade Chief Economist and Azad Zangana European Economist
Austerity bites: more to come and what does it mean?
– Austerity will be a key theme for 2011, with the UK and peripheral Europe Austerity is going to be the big theme of 2011, easily taking pole position as the main barrier to global
being the main focus growth. The focus will continue to be on the peripheral parts of Europe – Greece, Ireland, Spain and
Portugal. In particular, the impact of needing to unwind some of the huge borrowing they’ve amassed over
– Any benefit from QE2 to the US is largely through the fall in the dollar the last decade. This tightening of fiscal policy is going to be a headwind to global growth, but particularly
in the UK and eurozone where we expect to see a slowdown in economic activity. It’s a key question as to
– China has also benefitted from this, but will struggle to prevent liquidity from
whether they can withstand tightening of fiscal policy.
spilling over into asset bubbles and inflation
The exception to that austerity is in the US, where we’re looking at an extension of tax cuts. There’s very little
– The recovery, albeit a weak one, will continue, but low interest rates will force debate there at the moment about fiscal tightening.
investors to continue seeking yield.
Quantitative easing: does it still have a place?
It’s very much a case of déjà vu with 2010 ending pretty much the way it started, with a rally in markets As the markets expected, quantitative easing restarted in the US but a surprise this time around was the
and a recovery in the world economy. That’s not to say it’s been an easy or enjoyable ride. 2010 has been impact it had on the rest of the world. Currency appreciation and an ever increasing risk of growth stalling
a real roller coaster. At the beginning of the year, it looked like we were in for a decent recovery and people elsewhere led to the start of the ‘currency wars’.
began to anticipate interest rate rises and the central banks initiating exit strategies from their loose
“Austerity is going to be the big theme of 2011, easily taking pole position as the main barrier
to global growth.”
But by the springtime, the euro crisis had erupted; the world economy showed signs of slowing down and
Beyond the fall in bond yields and the dollar, we don’t see there being much further impact from QE2,
any optimism soon faded. Talk began to move towards a double dip and what the authorities could do in
certainly not while banks refuse to lend and more importantly while households and corporates continue to
terms of more stimulation. As we approach the end of the year, with more quantitative easing coming
have little appetite for borrowing. Emerging markets have fast become the destination for investors forced to
through, we’re beginning to see the recovery pick up again and markets have revived.
search for yield. However, the creation of huge capital flows is causing liquidity problems and inflation
is beginning to pick up too.
1 2011 OUTLOOK
Currency wars: who will win the war?
The focus of the currency war has really been on the fall in the dollar, but against that we’ve seen a rise in This continued recovery in growth is important for supporting real assets, but it’s not a strong recovery so
the euro, the yen and the Australian dollar. That increase in currencies is going to squeeze growth in the interest rates are going to stay very low as we go through 2011. That means investors are still going to be
eurozone, Japan and Australia, so there’s a real battle going on where everybody is trying to devalue their searching for yield. Following the rally in bonds we expect investors to rotate out of credit and move into
currency in order to get a bigger share of a shrinking pie. So far it seems the US is winning, while Europe, equities and commodities. Emerging markets are likely to continue to attract funds given their growth
where there is a desperate need for growth is fast becoming the biggest casualty. It is only recently with the prospects. These will be our main asset allocation plays as we go into 2011.
crisis in Ireland that the euro has begun to fall again providing some relief to the embattled region.
The views and opinions contained herein are those of Keith Wade and Azad Zangana and may
China is benefiting from the fall in the dollar with a currency that is becoming cheaper against the euro and
not necessarily represent views expressed or reflected in other Schroders communications,
the yen. But the benefits are not without problems. Big capital inflows are creating liquidity problems and
strategies or funds.
asset bubbles, not to mention inflation – and that’s something the authorities have to deal with in 2011.
The recovery: where will the search for return be focused?
Going into 2011, we expect to see the recovery continue globally, but especially in the US, which has been
boosted by strong profits and healthy balance sheets in the corporate sector. This should see increased
capital expenditure and employment, raising growth in the world economy as confidence returns to US
consumers. In emerging markets, an enviable fiscal position capable of increasing expenditure in
infrastructure will continue to boost growth. Nevertheless, the Fed will be on hold through 2011.
“Going into 2011, we expect to see the recovery continue globally, but especially in the US,
which has been boosted by strong profits and healthy balance sheets in the corporate sector.”
We are expecting a slowdown in Europe and in the UK, mainly because of the spending cuts and tax
increases, which will begin to bite into 2011. Given the fragility of the periphery, we do not expect any
increases in euro policy rates until 2012.
1.2 What got you here, won’t get you there
Alan Brown Group Chief Investment Officer
The 10 years that followed saw the gradual unwinding of that bubble. If you look at the equity risk premium
today (i.e. the earnings yield minus the 10-year government bond yield), in most places, equity risk premia
When asked in 1971 to discuss the consequences of the French Revolution, look fairly attractive by historic standards. This is partly why we don’t believe investors should give up on
then Chinese Premier Zhou Enlai famously replied that it was too early to say. equities just yet. It is worth bearing in mind, also, that the reason why equities look attractive is not entirely
to do with equities themselves. It is, in part, because bond yields are so abnormally low (see Graph 1).
Fortunately he wasn’t a fund manager! With the worst of the financial crisis only In fact, today’s equity risk premium is merely showing up an asset price bubble that is developing
recently behind us, the investment community cannot afford to ignore the lessons elsewhere – in government bonds.
of the last two and half years. In this article, Alan Brown, Group Chief Investment
Officer, outlines how and why investors must change the way they think. Graph 1: Are equities cheap?
UK Equity yield and Government Bond yield meet again
Latest reading: FTSE All share dividend yield = 3.0%
Faith in diversification 18
UK 10-year Govt. Bond = 3.5%
‘In the fourth quarter of 2008, there was no place to hide’
During the worst of the downturn most asset classes were moving in the same direction – down – and this 14
caused many to question the value of diversification. It is true that diversification certainly doesn’t help in an 12
environment when virtually every asset class is moving south. In the fourth quarter of 2008, there was no
place to hide in terms of spreading the risk to your assets. Correlations went as close to one as we’ve ever
seen and this did make people question the value of diversification – albeit for a short period. As soon as we 8
got past that tail event and the market moved back towards slightly more normal conditions, diversification 6
made a comeback. In fact, we would argue that given the uncertainty out there at the moment, people
should make diversification a priority. 4
The end of equities? 0
19 29 39 49 59 69 79 89 99 09
‘Today’s equity risk premium is merely showing up an asset class bubble developing elsewhere
– in government bonds’ UK 10-year Govt. Bond Yield UK Equity Market Dividend Yield
Generally, investors appear to have become weary of the prospects of being rewarded for investing in Annual Data: UK Equity Market Dividend Yield; FTSE All shares DY (2010 – 1965), UK Equities DY from global Fin Data
equities. We think this nervousness is somewhat justified given the fact that the last ten years has seen (1964 – 1919) UK 10-yr Govt. bond yield; DataStream yield (2010 – 1980), UK 10-yr from Global Fin Data (1979 – 1919).
stocks go absolutely nowhere, other than up and down in an extraordinarily painful fashion. However, the Source: Global Fin Data, Datastream, Schroders, 29 November 2010.
true story is a bit more complicated: the last decade began under the shadow of a significant equity bubble.
1 2011 OUTLOOK
Pension fund pain
‘Some funds are beginning to think about implementing a glide-path strategy to gradually If you look at Sainsbury’s, Marks & Spencer or GlaxoSmithKline – all of them yield significantly more than
de-risk the portfolio’ 10-year gilts. It’s very rare to have equity indices yielding the same or more than government bonds – and you
As defined-benefit schemes close and, with time horizon’s shortening, pension fund investors will be forgiven have to go back to the early seventies to see this on any significant scale. Investors might reasonably ask
for fearing that they won’t have enough time to get rewarded for investing in equities. The truth is that most themselves: who has got the more predictable revenues and expenses, Sainsbury’s or the UK government?
schemes – even if they are closed to new members and accruals – probably do have a life of 40 years left in
What I’m getting at is that notions of what is safe probably need to change. It may well be that a portfolio of
front of them. They are, however, becoming increasingly mature, so you would expect them to begin the
quality, decent-yielding, fairly-priced companies may be a safer bet than government bonds.
process of de-risking along the way. In recognition of this reality, some funds are beginning to think about
implementing a kind of glide-path strategy that gradually de-risks the portfolio whenever the opportunity
presents itself. At Schroders, we are certainly talking to clients about this as we aim to support them in Developed versus developing nations
constructing portfolios that are more dynamic and able to respond quickly to what the market is doing. ‘Do you want to be in the fast part of the river or do you want to silt up on the bank?’
The crisis clearly triggered a shift in the relationship between developed and emerging nations. In our view,
Efficient market hypothesis the emerging market story has not yet been built into asset allocation structures to the extent that it should.
We’re not suggesting that emerging and developed markets dance to different drum beats, neither are we
‘We’d like to put benchmarks back into the box from whence they came’
dismissing the fact that emerging markets will continue to be more volatile than developed markets.
The events of the last two and a half years have challenged the dominance of many major stock market
However, on every kind of metric that you consider, they look better off than the developed world: whether
theories. The efficient market hypothesis (EMH), for example, has been shown up as little more than
it’s debt to GDP (see Graph 2), whether it’s external or internal accounts, whether it’s leverage, whether it’s
what it is – an abstraction of reality. In saying that, the theory has served us quite well in the sense that
their business and banking systems, whether it’s demographics. There are hardly any measures you can
it has never been easy to outperform capitalisation-weighted indices; a theory which the EMH supports.
find where they don’t look better than us. Yes, the ride may be more rocky, but do you want to be in the fast
On the other hand, when you look at the volatility of equity markets compared to the volatility of
part of the river or do you want to silt up on the bank?
underlying economies – which are driving earnings – markets are much more turbulent.
If markets were to soar because they correctly anticipated a surge in earnings – this would be good. Graph 2: Are we going bust?
If they went down because they correctly anticipated a collapse in earnings – again, this would be good.
G-20 countries: general government debt ratios
Unfortunately, there is little evidence to suggest that markets are doing either of these things. This has
profound implications because the moment you start questioning the EMH, you begin to also question % of GDP
the importance our industry has placed on capitalisation-weighted benchmarks. As we already know,
in following a capitalisation-weighted approach, you are guaranteed to overweight the expensive and
underweight the cheap. Although we have no failsafe means of identifying which securities are cheap 100
and which are otherwise, we are increasingly finding more reliable ways of spotting the difference.
When Dow Jones first created indices, it was intended merely to offer a broad reference point to help
investors identify how well their investments were performing in relation to other people’s. It wasn’t 60
supposed to be the start of portfolio construction or – in the case of a passive fund – the end of portfolio
construction. In short, we’d really like to put indices back into the box from whence they came!
The attitude to risk 20
‘The concept of what is safe probably needs to change’
On the topic of whether we can still call anything a risk-free asset anymore, it might be more appropriate 0
at the moment to talk about return-free risk! More seriously, when you look at government bonds, they are 90 92 94 96 98 00 02 04 06 08 10 12 14
phenomenally expensive as a result of all the unusual policy actions being taken (from zero interest rates, Advanced G-20 economies Emerging G-20 economies
to quantitative easing) which is dragging down the yield curve. Source: World Economic Outlook (WEO), IMF, October 2010.
Low interest rates, high inflation
‘Investors don’t necessarily fear inflation today, but they do fear inflation tomorrow’ When you think about how asset prices have rolled around in the last decade, and how institutions only
The coming years are likely to be characterised by the clash of low interest rates and high inflation, and there really change their asset allocations at the margin, you have to wonder whether – as an industry – we were
is evidence that investors are beginning to position their portfolios accordingly. An increasing number of our asleep at the wheel. I believe that the industry must sit down and relearn the basics: think about the
clients are now interested in real asset funds because – although they do not fear inflation today – they do plausible scenarios surrounding what asset prices might do. Then, as your conviction increases around a
fear inflation tomorrow. There used to be a widespread view that this would be the decade in which we particular scenario, have the courage to change your asset allocation. The truth is, this is what investors
would flirt with deflation at the beginning of the period and – with the risk of the Federal Reserve being too used to do. If you go back to the end of the 80s, investors were using every device under the sun to reduce
slow to take stimulus off the table – we would end up with inflation at the end of the decade. It is clear that the weight of Japan in their international portfolios. They weren’t afraid to move away from capitalisation
this analysis is too simplistic. At Schroders, we now think the global economy will actually flirt with inflation benchmarks in their bid to pare back their exposure to Japan. However, by the end of the 90s this spirit had
and deflation simultaneously, in different parts of the world. In the developed world you already see all but disappeared – the industry was almost afraid to do the same thing when we were faced with a tech
countries like Ireland, Greece and Japan stuck in deflation. What complicates matters globally is that a lot bubble; the industry doesn’t appear to be doing it now as the shadow of a government bonds bubble
of countries in Asia link their exchange rates and monetary policy – either formally or informally – to the US. gathers over us.
Therefore, the ‘easy-money’ policy, which might be appropriate to the US, is certainly not appropriate to
We must learn to use the tools available to us more effectively. One can do some reasonable analysis on
these developing economies, and there is a danger that this will fuel an asset price bubble in Asia.
what an asset may or may not deliver. Academics rarely comment on whether the market is cheap or
Obviously, the first country that people worry about is China.
expensive, but two American academics – Eugene Fama and Ken French – in 2001, published a paper that
Earlier this year, we saw money supply growth in China shoot up to nearly 40% pa, and the Chinese called it right. They said that equity markets were completely overblown. They arrived at this conclusion by
authorities seemed to be doing a good job of reigning in the inflation tiger. Most recent data indicates that simply getting into a time machine, going back to 1950, and asking what an intelligent, rational academic
they can’t relax their guard, however, and inflationary pressures are rising once again. Overall, we think the would have said about equity market returns at the time. They then compared that prediction to the
world is entering a very complex economic era. outcome 50 years later, and they found that the returns you would’ve earned were much higher than you
would have expected. They then asked: so what did we get wrong in 1950? Was it that economies or
What’s the alternative? earnings or dividends grew much faster than expected? No. The truth was, in the last five years of the
century, there was a massive expansion in price-earnings multiples. Therefore, based on that
‘Investors have to be sure that the benefits don’t just go to the fund manager’
straightforward observation, they concluded that if you believe in the same kinds of returns going forward,
In the post-crisis world, many investors are considering alternatives as a means of diversifying their
you have to assume yet more price-earnings expansion. How likely is that? Not very! And if price-earnings
portfolios. For the small- to medium-sized fund the only realistic way to access these instruments is to
multiples revert back to their historic norms – as they recently have done – then, according to Fama and
go through diversified growth funds. However, in this respect, investors do have to be cost sensitive.
French, you will be in for a decade of miserable returns. If only we’d listened to them!
It is important to remember that we are in a world of relatively low nominal rates of returns and some
alternative products can be pretty expensive. Investors, therefore, have to be sure that a superior gross The views and opinions contained herein are those of Alan Brown and may not necessarily
return translates to a superior net return – i.e. that the benefits don’t just go to the fund manager. represent views expressed or reflected in other Schroders communications, strategies or funds.
As much as we do like fund managers!
The crisis: how was it for you?
‘You have to wonder whether – as an industry – we were asleep at the wheel’
So what are the final lessons that we must take from this crisis, in order to move on? Clearly there are a lot
of lessons that we need to learn out of this. I actually had a Pauline conversion on the road to Damascus
over the course of this downturn. Our industry has been founded on doing strategic asset allocation studies
for the last 30 years – and maybe we need to stop. The reason for saying this is: if you look at the last four
decades, the numbers that you’d have plugged into your strategic asset allocation study at the start of each
of those decades would have born no relation to the return you’d have earned at the end of those decades.
It seems the pseudo science has provided the industry with a dangerous comfort blanket that masks the
fact that if you put garbage in, you get garbage out.
2.1 Ageing gracefully: What the West can learn from Japan 10
2.2 China’s five-year plan: The start of a new economic era? 14
2.3 The American way: The impact of ageing on the US economy 15
2.4 “Nothing is permanent but change”
The Economic Impacts of demographics 21
2.5 The impact of climate change and resource shortages 26
2.6 Biofuels 29
2.7 Currency Unions: Breaking up is not so hard 39
2.8 Currency wars: The new weapon of choice 41
2.9 The engines of global growth beyond 2010 42
2.1 Ageing gracefully:
What the West can learn from Japan
Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate
As the world’s ‘oldest’ country, Japan provides a unique example of how countries and companies can respond stands at just under 200% of GDP. But because of a feature of the Japanese government’s accounting
to the challenges of ageing. As we know, it is difficult to reverse or meaningfully alter unfavourable demographic system, this figure significantly overstates actual liabilities. In commonly used ‘unified’ systems, the revenues
trends, and pre-emptive reforms are essential for successful adaptation. This has, to a certain extent, occurred and expenditures of all departments are pooled and government bonds issued to cover the shortfall. In the
in Japan, meaning the country is now in relatively good shape to bear the costs of ageing in comparison to other Japanese system, however, departments keep their accounts separate, and those in surplus formally lend
developed markets economies. As much of the world begins to face the challenges of demographic change, money to those in deficit via bond issuances or the creation of government deposit accounts. This means an
we explore the lessons that governments, companies and investors can learn from Japan. intra-governmental loan stays on the balance sheet as a separate asset and liability, while a unified system
would just show the net figure. The ‘gross debt’ figure includes this liability, while the ‘net debt’ figure (assets
Japan is the ‘oldest’ country in the world, with a median age of 44.7 years: about 5 years higher than the
minus liabilities) strips it out.2
developed country average, and more than 15 years higher than the global average.1 By 2050, the elderly
dependency ratio will be 75%, meaning there will be three retirees for every four people of working age. Figure 2 shows the OECD’s forecasts for gross and net debt in 2010. On the net measure, Japan’s debt is
expected to be 105% of GDP – still very high, but less shocking than the 200% figure.
Figure 1: Age structure of the Japanese population
Figure 2: Government debt, 2010
80% % of GDP
20 10% -50
0 0% -100
1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Aged 0-14 Aged 15-64 Aged 65 or Over Old-age Dependency (RHS)
Gross Debt Net Debt
Source: UN World Population Prospects, 2008 Revision.
Source: OECD (2009).
It also has the most government debt, but neither as much, nor by as large a margin, as is often thought.
The number most commonly used when discussing Japanese debt is the ‘gross debt’ figure, corresponding
1 UN World Population Prospects, 2008 revision. In case you were wondering, the youngest country is Niger, with a
to the total liabilities of central and local governments, including the social security system. Gross debt
median age of just under 15 years!
2 As explained in Broda and Weinstein (2004), Happy News from the Dismal Science: Reassessing Japanese Fiscal Policy
10 and Sustainability.
2 INVESTMENT THEMES
On the challenges generated by demographic trends, the key point is that Japan recognised its ageing The Japanese system, in contrast, was carefully redesigned to avoid these pitfalls. Prices are stringently
problem early, and took steps to reform both the health and pension systems to contain age-related controlled and set biannually, with high-tech care priced below cost and basic care above cost to realign
spending. It also helps that both employees and companies look favourably on working beyond retirement. incentives. Though this may be counterintuitive, it has been shown to result in more appropriate incentives:
As a result, the prospects for Japanese solvency are better than the demographics imply and not as bad as quality basic care (cheaper for the government) is provided in most cases, and high-tech care only ‘when
many presume, even compared to other developed countries. appropriate’.6 Bundled pricing (flat reimbursement for a condition/illness rather than per treatment) gives
providers an incentive to switch to cheaper alternatives, and post-utilization reviews prevent improper use
Lessons for governments: health and pensions and fraud. Co-payments account for 20-30% of each claim, meaning users have some ‘skin in the game’.
Wages for doctors are lower than in the United States, partly because malpractice suits are less common.7
Japan undertook large-scale reforms of its public healthcare system as far back as the 1980s, after just 20
years of universal coverage. As a result, Japan spends around 8.1% of GDP on healthcare, below the OECD All of these measures help keep healthcare relatively ‘cheap’ for the government, reducing the burden of
average of 8.9% ($2581 versus $2964 on a per capita basis). In Germany, which is the next oldest country ageing. Crucially, lower spending is not reflected in the quality of healthcare: Japan has the longest life
by median age, spending is 10.4% of GDP. Spending in Japan has also grown at just 2.5% p.a., while in the expectancy and the lowest infant mortality rates in the OECD. It has the lowest incidence of heart disease
US the rate is consistently GDP growth plus 2.5%.3 and obesity in the world and cancer survival rates among the highest, despite low screening rates.8
We now turn our attention to the pension system. To begin again with OECD indicators: Japan’s public
Figure 3: Health expenditure
pension spending currently stands at 8.7% of GDP, higher than the average (7.2%) but lower than several
% of GDP, 2007 ‘younger’ countries in Western Europe.9 On a per capita basis, this equates to an outlay of $2637 in Japan
18.0 compared to an OECD average of $2150 but, interestingly, the spending per retiree (i.e. over 60) in Japan is
16.0 16.0 almost 15% lower than the average figure.
Reform efforts in Japan began in the 1980s, with the most recent round occurring in 2004. The explicit
OECD Average 8.9% 9.9 10.1 10.1 10.2 10.4
10.8 11.0 target was to “establish a sustainable and secure pension system that can support the ageing population
10.0 9.3 9.6 9.8 9.8
8.7 8.9 8.9 9.0 9.1 in the future”.10 The key cost-containment features of the system are: relatively low replacement rates
8.0 8.1 8.2 8.4 8.5
7.3 7.4 7.6 7.7
6.3 6.4 6.8 (generosity), gradual raising of the pensionable age from 60 to 65 (and possibly further), and price
6.0 5.7 5.9
– rather than earnings – indexation. The latter is particularly important given Japan’s persistent deflationary
environment. Benefits are automatically adjusted (downwards) to reflect increases in life expectancy,
though a minimum payment is guaranteed.11
Source: OECD Health at a Glance 2009.
Japan and the United States both have ‘fee-for-service’ systems, where the government or private insurer
reimburses the provider for care received. Unless carefully regulated, these systems can provide perverse
5 Carey et al (2009), Healthcare Reform in the United States; Anderson et al (2003), It’s the Prices Stupid: Why the United
incentives, resulting in general over-use of healthcare and possibly a disproportionate use of high-cost
States is Different from Other Countries.
services.4 This is one of the key reasons why US healthcare expenditure is almost twice the OECD average,
6 Commentators suggests this is because high-tech care builds reputation and is more professionally rewarding for
while healthcare outcomes (life expectancy, infant mortality) are no better.5 doctors, so doctors continue to offer it despite making a small loss on each treatment. Wagstaff, (2005), Health Systems
in East Asia: What can Developing Countries Learn from Japan and the Asian Tigers?
7 Fukuwa & Izumida (2004), Japanese Healthcare Expenditures in a Comparative Context.
3 OECD Health at a Glance 2009; Reinhardt (2009), Economic Trends in US healthcare. 8 OECD Health at a Glance 2009.
4 As long as the payment to the provider exceeds their cost, their incentive is to supply, and encourage patients to use, 9 OECD Pensions at a Glance 2009.
additional services. Reimbursements are generally based on ‘cost-plus’, so there is no incentive to provide cheaper 10 Japanese Ministry of Health, Labour and Welfare (2009), Pension Overview.
services. Also, because insured patients bear very little of the cost of healthcare, they are inclined to over consume. 11 Ibid. 11
Figure 4: Public pension spending Figure 5: NPV of fiscal impact of the financial crisis and population ageing
% of GDP, 2005 % of GDP Cost of crisis/cost of ageing
12.6 12.4 700
12 11.5 11.4 11.4
10 9.0 8.7 8.5 8.4
8.1 7.8 7.7 OECD Average 7.2%
8 7.3 7.2 500 15%
6.2 6.0 5.7
6 5.4 5.0 400
4 3.5 3.4 10%
Canada Korea Spain US Australia UK Germany France Mexico Turkey Italy Japan
Crisis Cost of crisis/cost of ageing Ageing
Source: OECD Pensions at a Glance 2009. Source: IMF (2009).
Like much of the developed world, Japan has actually enjoyed favourable demographic trends for much of Lessons for companies: workers and customers
the last 50 years, with a large working age population and a small number of dependents (as lower fertility
resulted in fewer children). This means that the Japanese pyramid during the period has been the ‘optimal’ One of the other factors contributing to Japan’s robust old age is the willingness – and ability – of older people
shape for growth and low expenditures. To take advantage of this ‘demographic dividend’, since the last to continue working. Despite the fact that most companies have a mandatory retirement age of 60, the majority
world war Japanese pension contribution rates have been set higher than dictated by contemporary needs, of workers remain active until at least 65, even once they are entitled to a full state pension.15 Though Japan’s
and the surplus fed into a reserve fund. This now stands at $1.2 trillion, second only to the United States. life expectancy is the highest in the world, time spent in retirement is below the OECD average.16
According to the latest available data, this equates to a per capita value of approximately $9100 for the
Japanese fund, while the US stands at $7,938.12 Official estimates are for this reserve fund to last until 2100,
though Japanese economists are now saying that the financial crisis (and embezzlement equivalent to about
1% of fund assets, discovered in 2007) have brought this forward to 2050.13 This means that the buffer
provided by the reserve fund compares favourably with many European retirement funds and the US Social
Security Trust Fund, which the latest CBO projections anticipate will be exhausted by 2043.14
Although Japan’s demographic trends are clearly not propitious for economic growth, these health and
pension reforms mean that the country is in some respects better prepared for the consequences of a
rapidly ageing population than much of the Western world. A recent IMF report on the financial crisis uses
estimates from multinational projection exercises to compare the cost of ageing with that of the current
financial crisis. The chart below shows that, while the financial impact of the crisis, as a percentage of GDP,
is broadly similar across the developed world, the lower net present value of the cost of ageing implies that 12 OECD (Oct 2009), Pension Markets in Focus.
Japan is expected to manage its future costs better than most. 13 2100 figure from MHLM (as above), revision from personal correspondence with local sources.
14 European Commission (2006), The Impact of Ageing on Public Expenditure; (2009), The 2009 Ageing Report;
Congressional Budget Office (2009), Updated Long-Term Projections for Social Security.
15 International Labour Organisation. Economic activity rate is the labour force (employed or unemployed and looking for
work) divided by the population.
12 16 The Economist (27 June 2009), Work Till You Drop.
2 INVESTMENT THEMES
Figure 6: Economic Activity Rate, 2009
80 Japan (along with Singapore) is exceptional in this regard. A 2007 survey found that, of 28,000 employers in
70 25 countries, only 14% had a strategy for employing older workers.20 There has been very little written on
how to manage older workers, and an ageing workforce can lead to spiralling wage bills if salaries are driven
60 by seniority. As legal and effective retirement age increases, Western companies will need to become more
50 cognizant of the costs and benefits of older workers.
40 The Japanese experience demonstrates that most companies will need to make ‘subtle and difficult
adjustments’ as buyers age.21 Niche businesses have sprung up catering to the ‘grey market’: cosmetics,
30 robot pets, even indoor vegetable patches and colourful incontinence pads! 22
Even global leviathans are adapting their products and services: McDonald’s branches in Tokyo have
10 sections with seating designed for older diners, and a number of consumer goods companies have begun
to sell smaller packs for smaller retired households. More accustomed to targeting the young, some large
companies are now employing managers to focus exclusively on the silver dollar.23 Financial firms in Japan
Canada France Germany Greece Italy Japan United United
are offering additional services and perks to attract the business of asset-rich older customers.24 The
beneficiaries are not always in the obvious places: for example, the heavily regulated healthcare market
Men – 60-64 Years Men – 65 Plus
means that branded pharmaceuticals are struggling.25
Source: ILO (2010).
Lessons for investors
This is partly because the public pension is ungenerous, but also because attitudes and legislation are more
conducive. Older Japanese employees see their work as a source of pride, and around 60% say they would Painful reforms in the 1980s and ‘90s mean that Japan is now in comparatively good shape to bear the
like to stay at the same job when they turn 60.17 Fortunately for them, a 2004 law requires companies to cost of ageing. Yet, despite its best efforts, demographic change will still present serious challenges for
either raise their mandatory retirement age (in line with the rising pensionable age) or re-hire workers who the country: economically, politically and socially. Most significantly, its dramatic ageing profile will seriously
want to stay on. There are also financial incentives for hiring older workers. The majority of firms, including hamper economic growth. In this respect, the impact of reform is limited and Japanese policymakers are
well-known names such as Toyota, Aeon and Mitsubishi, have embraced re-hiring programmes, and now now taking a more direct approach, attempting to raise the birth rate by offering a ¥26,000 monthly childcare
retain 50-70% of their employees over 60.18 Employees are often hired part-time and usually on lower pay, allowance and free schools. But even if this is successful, it will be several generations before it improves
allowing for greater flexibility and keeping wages in line with productivity.19 At a more macro level, higher dependency and economic outcomes.
activity should reduce the impact on economic growth of a falling working age population.
The same central lesson holds for investors. To profit from changing demographics, investors need to
recognise – well in advance – those markets, industries and companies that are best-placed to meet the
challenge. Because demographic trends are regarded as very long-term, most investors neglect to include
them in their analysis of companies’ operating environment and earnings sustainability. Yet demographic
change will alter the competitive landscape for each and every company worldwide. Investors must be able
to identify the ones that will adapt, survive, and thrive.
17 Japanese Institute of Policy and Training.
The views and opinions contained herein are those of Virginie Maisonneuve and Katherine Davidson
18 Gross and Minot (2008), Effects of Japan’s Ageing Population on HR Management.
and may not necessarily represent views expressed or reflected in other Schroders communications,
19 This is a matter for debate, but the weight of evidence suggests productivity declines in old age e.g. Skirrbeck (2003);
Kotlikoff and Wise (1989); Hansen (1993); Meghir & Whitehouse (1996).
strategies or funds.
20 Manpower (2007), cited in Nomura (2008), The Business of Ageing.
21 FT (11 August 2009), Japanese shoppers never retire from aspiration.
22 Economist (June 2009), The Silver Dollar.
24 FT (10 August 2009), Fireflies and seminars: how Japan’s older investors are being courted.
25 FT (06 August 2009), Japan’s drugmakers find rich pickings elusive. 13
2.2 China’s five-year plan:
The start of a new economic era?
Laura Luo Fund Manager, Asian ex Japan Equities
In November China released the details of its 12th five-year plan (2011 – 2015), outlining the government’s These intentions mark a significant change for China, but they also intensify some of the concerns that
strategy for creating a more balanced economy and sounding the starting gun for a next phase of the already linger over the economy. Any income increase, for example, would lift labour costs and subsequently
country’s growth. Here, we address the major themes to emerge from the plan and what the new economic stoke the flames of inflation. The government therefore faces a significant challenge of managing inflation
era will mean for investors. while working to close the income gap between urban and rural inhabitants.
In a document dominated by the topic of balance, the key themes of the five-year plan were as follows: If China can make some progress towards achieving this rebalance, it is the broad consumption sectors that
are likely to benefit most. These include the consumer staples and discretionary sectors, insurance/wealth
– Rebalancing economic growth, in particular, stimulating consumption (relative to exports and investment)
management services, healthcare, tourism and education, as well as the subsidised housing sector. On the
– Adjusting the supply structure by upgrading traditional industries, developing new strategic industries, other hand, the effects are likely to be less positive for the region’s low-end exporters.
and promoting the service industry, clean energy and environmental protection
With technology upgrades and the promotion of the service industry, clean energy and environmental
– Balancing regional growth by accelerating the development of inland areas, promoting regional
protection also mentioned in the plan, we believe interesting investment opportunities are likely to emerge
integration, and fostering urbanisation in small- and medium-sized cities.
from the following areas of the market:
“If China can make some progress in achieving its rebalancing objective, the impact will be – alternative-energy plays (such as natural gas, wind-power and solar)
positive for the broad consumption sectors.”
– high-end capital goods manufacturers
The emphasis on rebalancing economic growth – or the structural adjustment of the economic growth – IT services
model (from investment and export-led growth to a consumption-led model) – are likely to have a profound
– and logistics companies.
impact on China’s economy and equity markets going forward. Although this is not something new, the
issue has now moved to the very core of the government’s focus. As the government continues its urbanisation efforts – promoting faster development in inland areas – it will
boost regional producers of building materials (in particular cement), retail/property focusing on the second
While the 11th five-year plan also called for greater emphasis on consumption, we saw little progress on the
and fourth tier cities, as will as some infrastructure companies that will also benefit from the continued
rebalancing of the economic growth model over the past five years. This is partially due to the massive fiscal
investment spending by governments.
spending package and, therefore, the surge in fixed-asset investment during the financial crisis. However,
given the sluggish global economic outlook for the next few years, along with current levels of over-
The views and opinions contained herein are those of Laura Luo and may not necessarily
investment/over-capacity in China, significant and renewed effort must be made to rebalance the economy
represent views expressed or reflected in other Schroders communications, strategies or funds.
if China is to sustain its future growth. The good news is that the government appears ready to address this.
In an effort to achieve the rebalancing, the government has identified two key policy measures:
– raising household disposable income and reducing income disparity;
– building a better social security net.
2 INVESTMENT THEMES
2.3 The American way:
The impact of ageing on the US economy
Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate
The second reason is that the US accepts far more immigrants. And this, in fact, explains a large part of its higher
fertility rate as migrant populations tend to have more children than natives. The net migration rate in the US
In summary: has fallen in the last decade but remains almost double the rate of western Europe and six times that of Japan,
where immigration policies are notoriously stringent. The number of (net) immigrants into the US is expected to
– The United States has the most favourable demographic profile in the
be just over 1 million a year according to the UN, or as many as 1.5m according to the US Census Bureau.3
developed world and, with it, offers some of the best prospects for investors
Figure 1: US Population Pyramid, today (left) and 2050 (right)
– As baby boomers retire, the changes in the US labour force and the cost of
ageing will weigh heavily on both government and corporate balance sheets
– In this article, we will argue that America has a ‘country competitive advantage’, but 80-84 80-84
this will not necessarily translate into success at the industry or company level 70-74 Baby boom: 70-74
– Identifying the best opportunities in this diverse and changing economy 60-64 60-64
requires close analysis and strong stock-picking skills. 50-54 50-54
30-34 born 1965-75 30-34
What makes the US different? 20-24
Echo boom: 20-24
born 1976-2001 10-14
Like the rest of the developed world, America is ageing. By 2050, the median US citizen will be 5 years older
than today, and the proportion of the population over 65 will rise from 13% to 21.6%. However, by western 0- 4 0-4
-4% -3% -2% -1% -0% 1% 2% 3% 4% 5% 6% -4% -3% -2% -1% -0% 1% 2% 3% 4%
standards, the US looks positively youthful. The proportion of retirees in Japan, and the median age in
western Europe, are already higher than the 2050 estimates for the US. Old-age dependency in the US will Male Female Male Female
be just 35% in 2050, compared to 51% in western Europe and a whopping 74% in Japan. The US even The age structure in the US will not change dramatically over the next 40 years, but we can see from the
compares favourably with China, where the proportion of the population of working age will converge with population pyramids above that there will be a ‘smoothing’ of the age distribution as the abnormally large ‘baby
that of the US over the next 40 years, and dependency will be 38% by 2050.1 boom’ and ‘echo boom’ cohorts age and dissipate (i.e. die). As this transition takes place, there will be significant
In terms of overall population growth, the US is also exceptional in that it is growing throughout the forecast implications for the labour force, economic growth, and government spending on health and pensions.
period at 0.77% per year. In Japan and some western European countries, the population has already begun
to shrink, and several others (including China) are expected to follow suit in coming decades.
So why is the US different from a demographic standpoint? The first reason is that the fertility rate has been,
and is expected to remain, higher than in most developed countries. After dropping sharply from its
baby-boom levels, the average number of children per woman has remained fairly constant over the last 20
years, just below the replacement rate of 2.1. The average fertility rate for western Europe is around 1.6 and
1 UN World Population Prospects, 2008 Revision
some countries, such as Spain, Italy and Japan, have reached ‘lowest-low’ fertility levels of less than 1.3.2
3 Bureau of Labour Statistics (Nov 2009), Labour Force Projections to 2018. 15
Greying of the labour force
Our first observation is that labour force growth in the US is slowing, from 1.1% p.a. in the early 2000s to This may be all the more true in the wake of the financial crisis, which has forced some baby boomers to delay
0.8% this decade, and even slower thereafter.4 This is primarily due to lower population growth and the retirement out of economic necessity. The annual Retirement Confidence Survey finds that Americans’ confidence
gradual retirement of 78 million baby boomers: more than the entire current population of the UK. This in their ability to afford a comfortable, or even financially secure, retirement fell to record lows in 2009 and remained
brings us to our second observation, regarding labour force ageing. Today, the Boomers range from 46-64 there in 2010.5 The proportion of baby boomers that intends ‘to work until they die’ rose from 15% before the crisis
years of age, but next year the first 3.4 million will become eligible for retirement, and by 2020 they will all be to 25% in mid-2009, though it is likely that this figure will fall as sentiment about the economy improves.6
over 55. This transition, combined with an increasing participation rate (discussed below), will cause the
Another issue that has arisen during the crisis is that, as baby boomers delay retirement, they are seen to be
ranks of older workers to swell while younger cohorts are stagnant or shrinking. Consequently, over 55s will
‘hogging’ jobs and driving up youth unemployment. The unemployment rate for 16-19 year olds currently
account for 24% of the workforce by 2018, almost double the proportion in the 1990s. Labour force ageing
stands at 26.1%, nearly three times the headline rate of 9.5%, though there is little evidence to suggest this
is illustrated in the charts below by the increasing ‘top-heaviness’ of the orange segment.
is due to ‘crowding out’ by older workers.7
Figure 2: Age structure of the population of working age. L-R: 1998, 2008, 2018
Ageing and productivity
75 and older 75 and older 75 and older
The economic impact of ageing is an issue that is extensively discussed, with the key conclusion that an
65 to 74 65 to 74 65 to 74 older labour force represents a headwind to productivity and economic growth. Indeed, BLS forecasts are
for 1.8% annual productivity growth to 2018, below the 2.6% average over the last decade. However, this
55 to 64 55 to 64 55 to 64
should not detract from the fact that the US is one of the most productive economies in the world, and faces
45 to 54 45 to 54 45 to 54 fewer headwinds than the majority of developed countries.8 Furthermore, BLS forecasts do not take account
of technology or innovation, and have become notorious for upward revisions.
35 to 44 35 to 44 35 to 44
A new and unusual angle on the productivity debate in the US concerns the impact of the population
25 to 34 25 to 34 25 to 34
age structure on management quality. Recent research suggests that this explains a significant portion
16 to 24 16 to 24 16 to 24 (about one-fifth) of the slowdown in US productivity in the 1970s and recovery in the 1990s.9 The boomer
generation is much larger than their elders, so their entry into the labour force prompted the recruitment
Millions Millions of new managers (borne out by the data). These marginal managers were necessarily of lower quality,
-20 -10 0 10 20 -20 -10 0 10 20 -20 -10 0 10 20 otherwise they would have already been promoted. Furthermore, the mean and median age of management
Female Male Female Male Female Male fell in the 1980s, implying that boomers were promoted to management earlier than other cohorts. All in all,
this resulted in lower management effectiveness until the boomers reached the ‘normal’ management age
Blue: total population. Yellow: labour force (includes employed and unemployed, excludes economically inactive).
in the 1990s. We could see a repeat of this phenomenon in the 2020s after the retirement of the baby
boomers, with the smaller ‘baby bust’ cohort occupying the management role for the larger ‘echo boomers’.
Since the 1990s, there has been a secular trend towards increasing participation among older cohorts,
implying an increase in the average retirement age. There are many logical reasons for workers to retire later:
longer life expectancy, improving health, and a reduction in the number of workers that are engaged in 4 Ibid; BLS (March 2010), Long-term Labour Force Projections to 2050.
physically demanding work such as farming, mining and manufacturing. Pension reform in 2000 eliminated 5 EBRI (March 2010), The 2010 Retirement Confidence Survey. Lowest level since records began in 1993.
the mandatory retirement age, introduced (dis)incentives for (early) delayed retirement, and initiated a 6 FT (08 May 2009), Desperate ‘baby boomers’ return to work. Note, however, that the aggregate retirement age has not
gradual increase in the statutory retirement age. Some commentators have suggested that baby boomers’ actually changed during the recession, as some workers have been forced into early retirement by redundancy. Gustman et al
entrepreneurial spirit and higher education levels than their predecessors are also relevant. Delayed (2009), What the Stock Market Decline Means for the Financial Security and Retirement Choices of the Near-Retirement
retirement, however, is not always a matter of choice: spiralling healthcare and insurance costs, the shift Population. Coile et al (2009), The Market Crash and Mass Layoffs: How the Current Economic Crisis may Afffect Retirement.
from defined-benefit (DB) to defined-contribution (DC) pension plans, and inadequate private savings mean 7 US Bureau of Labour Statistics, 06 August 2010; Kalwij et al (2009), Early Retirement and the Employment of the Young;
Banks et al (2008), Releasing jobs for the young? Early retirement and youth unemployment in the United Kingdom;
that many workers can simply no longer afford to retire when they would like.
Jousten et al (2008), The Effects of Early Retirement on Youth Unemployment: The Case of Belgium; Brugiavini et al
(2008), Youth Unemployment and Retirement of the Elderly: The Case of Italy.
8 In 2009, only Ireland, Luxembourg and Norway were more productive in terms of GDP per labour hour, though other
countries had faster growth (OECD).
16 9 Feyrer (2009), The US Productivity Slowdown, The Baby Boom, and Management Quality.
2 INVESTMENT THEMES
Number of jobs in selected occupations in health and education, 2008-18 (projected)
Millions This observation has led to fears of an impending shortage of as many as 35 million workers.11 In our view,
the chances of this materialising are remote, because a tight domestic labour market would induce greater
immigration, or outsourcing, offshoring and FDI (Foreign Direct Investment) by American firms. Today, it is
possible to object that foreign workers lack the skills to make them directly substitutable for US workers,
0.8 but the skills gap is closing fast. The US now accounts for less than one-sixth of global university enrolments
0.6 and doctorates, down from one-third and one-half respectively in 1970.12 During the 1990s, US firms hired
0.4 increasing numbers of scientists and engineers without an increase in either wages or the number of US
0.2 graduates, demonstrating the ability of global sourcing to introduce slack into the US labour market.
0 There are, nevertheless, certain occupations where talk of a shortage is credible. These are mainly in service
Home health aides
Child care workers
Personal and home
industries where work cannot be outsourced or offshored, and where it is hard to improve productivity using
technology. Public sector jobs will face the greatest near-term pressure, because the majority of workers are
allowed to retire at 55 and are covered by DB pension plans, which incentivise early retirement.13
The one example that appears time and time again in the literature is nursing, because demand is growing
as the population ages but stretched healthcare budgets prevent wages from rising.14 Teaching and public
Employment growth Replacement needs administration face similar problems as they are also publicly funded, plus both employ a large proportion
of baby boomers.15 Furthermore, both health and education are female-dominated, and women tend to
leave the labour force more rapidly after the age of 55.
An impending labour shortage? Baby boomer retirements over the period 2008-18 are expected to result in replacement demand for over
However, ageing is only half the story. As noted above, lower population growth and Boomer retirements two million workers in health and related services such as home help and medical aides, and roughly the
will also drive a gradual slowing of labour force growth. Despite higher participation within older cohorts, same number in education.16 This is before even considering employment growth, which is expected to be
the overall participation rate in the United States will fall as the population ages, completely wiping out the rapid for both industries.
contribution of higher female participation over the last 50 years. The combination of lower participation
Another particularly vulnerable area is manufacturing. New employment in secondary industry is expected
and Boomer retirements means that, in just a few years, there will be more non-workers than workers in
to decline over the next ten years, with the only job openings resulting from baby boomer retirements. The
relative unattractiveness of these careers to younger workers means that what vacancies there are may not be
filled.17 Despite current slack in the labour market, one-third of US manufacturing firms say they are suffering
from a shortage of qualified workers, while 38% expect shortages in the future.18 At some large-cap names, as
much as 80% of the workforce will be eligible for retirement in the next five years.19 The National Association of
Manufacturers has previously warned of a skills gap of 21 million workers by 2020.20
Industry participants expect a ‘war for talent’ as the economy recovers, pushing up labour costs and forcing
10 BLS (2009), as above. some companies to ‘make do’ with less qualified workers.21 The sector has been criticised for under-investing in
11 Freeman (2006), Is a Great Labour Shortage Coming? human capital: the US produces fewer manufacturing graduates every year, and has long since been overtaken
12 Ibid. by India and China. As the baby boomers retire, the US is at even greater risk of losing its competitive advantage
13 BLS (2000), as above.
14 Ibid; Freeman, as above; Cappelli (2003), Will There Really be a Labour Shortage?
in manufacturing. Its share of global manufacturing has already fallen from 26%, to under 18%, in the last
15 BLS (2000), as above. decade, with China gaining proportionally. There is no intrinsic economic reason why this should be a problem,
16 BLS (2009), Employment Projections for 2008-18. but the political stakes are high: the perception is that manufacturing and innovation are “what made our country
17 Ibid. great”.22 US carmakers were bailed out during the crisis, but government support is hardly a sustainable
18 FT (01 March 2010), Concern as Baby Boomers prepare for retirement. response in the face of structural decline.
20 Freeman, as above.
21 FT, as above.
22 Quotation from Jim Whaley, president of the Siemens Foundation. FT, as above. 17
The fiscal cost of ageing The real national debt
The US healthcare system is already the most expensive in the world, consuming a massive 16% of GDP, A report commissioned – but never published – by the US Treasury, details the present value of all the US’s
compared to the OECD average of 8.9%. This is partly due to the higher prevalence of costly chronic conditions unfunded future obligations, including health, pensions and debt service costs. The final figure is $65.9
in the US compared to Europe, but mostly down to higher costs. The US has higher salaries for medical trillion, almost five times the size of GDP (in constant prices).31 This is consistent with IMF work on the
professionals, more administrative staff, higher utilization of new and expensive technologies, higher financial crisis that estimates the present value of the cost of ageing at 495% of GDP.32 To plug this hole
pharmaceutical prices, and extremely costly malpractice insurance.23 Most of these are the result of perverse would require an immediate and permanent doubling of personal and corporate income taxes, or a
incentives in its fee-for-service system, which encourages over-consumption of healthcare without producing two-thirds cut in social security and Medicare benefits.33 Every year of inaction by the US government adds
superior outcomes: US life expectancy is lower than would be expected for its GDP level, as is infant mortality.24 an estimates $1.5 trillion to the projected shortfall. Note that these estimates precede the crisis: the fiscal
gap is now thought to have reached an astonishing $202 trillion.34
Figure 3: Health expenditure as a share of GDP, 2007
One issue that often goes unnoticed is that the Social Security Trust fund is invested exclusively in
Health expenditure as a share of GDP, 2007 specially-issued US government bonds. These are redeemable at par at any time, so there is no liquidity
18.0 risk, but issues may arise when the fund – the single biggest buyer of US debt – becomes a seller.
16.0 Trust fund holdings currently constitute nearly 20% of total US government debt, which will have to be
14.0 absorbed on top of new issuances.35
12.0 10.8 11.0
9.9 10.110.1 10.2 10.4
9.3 9.6 9.8 9.8
OECD Average 8.9%
7.8 7.4 7.6 7.7
8.1 8.2 8.4 8.5
8.7 8.9 8.9 9.0 9.1
Occupational and private pensions
6.3 6.4 6.8
6.0 The outlook for Social Security is particularly bleak, but other sources of retirement income do not look
4.0 much better. Recent research finds that DB pension plans covering public workers are underfunded by
2.0 more than $1 trillion, or 30%.36 Even those funds that do appear solvent are often basing their calculations
0.0 on outdated mortality tables, high discount rates, and unrealistic rates of return.
Source: OECD Health at a Glance 2009.
Over the past 30 years, despite its favourable demographic profile, public healthcare spending in the US 23 Anderson et al (2003), It’s the Prices, Stupid: Why the United States is so Different to Other Countries; OECD Health
Data 2009; Carey et al (2009), Healthcare Reform in the United States.
has significantly outpaced GDP, growing at over 5% per year.25 As baby boomers become eligible for
24 Carey et al, as above; OECD.
state-funded Medicare, this is likely to accelerate. The latest official projections anticipate an increase in total 25 Hagist and Kotlikoff (2005), Who’s Going Broke: Comparing Healthcare Costs in 10 OECD Countries; Reinhardt (2009),
healthcare spending from the current 16% to 25% of GDP by 2025, and 37% by 2050. This is based on the Economic Trends in US Healthcare: Implications for Investors.
assumption that cost inflation and consumption growth will slow: extrapolating historic rates would imply 26 CBO (2007), The Long-Term Outlook for Health Care Spending.
healthcare expenditure reaching 100% of GDP by 2080! 26 Projections of public spending in isolation 27 CBO (2010), The Long-Term Outlook for Mandatory Spending on Health Care.
estimate that Medicare and Medicaid expenditures will rise from 5.5% of GDP today to 10% in 2035 28 OECD (July 2010), Pension Markets in Focus.
in the baseline scenario; higher if costs grow faster than GDP.27
30 CBO (2010), The Long-Term Outlook for Social Security
The other big age-related expenditure is pensions. We will focus here on the public pension system (social security), 31 Gokhale and Smetters (2005), Measuring Social Security’s Financial Problems.
32 IMF (2009), Fiscal Implications of the Global Economic and Financial Crisis.
addressing occupation and private schemes below. At first glance, the US pension system seems in relatively good
33 Gokhale and Smetters (2005), as above; Gokhale (2009), Measuring the Unfunded Obligations of European Countries;
shape, given that it holds the biggest public pension reserve fund in the world, at $2.5 trillion and 18% of GDP.28 Kotlikoff (2007), Is the US Bankrupt?
However, despite not making losses during the recession, plummeting tax revenue did mean that assets had to 34 Kotlikoff (Bloomberg, August 2010), The US is Bankrupt and we don’t even know it.
be drawn down in 2009, the first time since the fund was founded.29 Even if the economy recovers, social security 35 OECD (2010), as above. OECD government debt statistics 2009.
will face a permanent shortfall as early as 2016, and the trust fund will be completely exhausted by 2039.30 36 Bloomberg (10tAugust 2010), Hidden Pension Fiasco May Foment Another $1 Trillion bailout. Note that the $1 trillion
figure is not comparable with the $65.9 trillion figure in the previous section, because the former is the current shortfall
while the latter is the present value of all future shortfalls: measured on the same basis, the gap in state and local
18 pensions would look much bigger.
2 INVESTMENT THEMES
In the private sector, DB plans still contain around 45% of pension assets, and many listed companies have In light of this, we expect to see an increase in old-age poverty and inequality. Much is made of income
large unfunded pension liabilities.37 At the end of 2009, the average pension deficit of US-listed companies inequality in the US – with the top 1% of earners raking in over 23% of income – but financial wealth
was 20%. This was already among the worst in the world, but preliminary estimates are that funding could (including pension assets) is even more skewed. Before the crisis, the top percentile held 43% of financial
deteriorate further, to a deficit of 32% ($651 billion) by the end of 2010.38 Industrials and airlines have the wealth and the top quintile 93%, and this appears to have worsened as a result of the crisis.43 The US pension
largest unfunded liabilities, and the single largest underfunded plan is, unsurprisingly, General Motors: the system is thoroughly progressive: 40% of private sector workers have no access to employee-sponsored
hole in its finances is estimated at $44 billion, more than ten times its market cap at the depths of the crisis. plans, with the burden falling disproportionately on lower-income groups, who are more likely to be
The funding situation appears even more grave if we look at the assumptions underlying the current pension part-time or casual workers.44 Furthermore, high earners are the biggest beneficiaries of tax incentives
liabilities: a number of industrial and pharmaceutical names are using expected return assumptions of close for private saving: by definition, people paying no income tax do not benefit from a tax credit. In 2005,
to 9% and discount rates well in excess of bond yields.39 more than 73 million workers had income low enough to be eligible for Saver’s Credit, but two-thirds
failed to qualify because they had no federal income tax liability.45
Our key conclusion is that workers will have to take greater individual responsibility for their retirement
income, leading to a further extension of working lives. Company plans are continuing to shift from DB This implies that inequality will be aggravated as an increasing proportion of the population moves into
to DC (including 401k plans), and social security benefits are likely to become less generous. This is retirement. Around one-third of current retirees are completely reliant on social security, and almost the
recognised by workers, a minority of whom are very (7%) or even somewhat (23%) confident that social same share again rely on it for the majority of their income.46 This has implications for consumption.
security will continue to provide benefits at current levels.40 All workers, and particularly those without Retirees have lower per capita expenditure, so increasing dependency is a headwind for the consumer
access to an employer-sponsored scheme, are being encouraged to save money in an Individual sector, particularly discretionary. This will be even more relevant if a large number of those retirees are
Retirement Account (IRA), and incentivised by tax relief on investment returns and payouts. reliant on dwindling social security benefits, making consumption extremely vulnerable to benefit cuts.
In the past, lower-income households boosted their purchasing power by borrowing, but this was clearly
A comfortable retirement? not a sustainable solution and will be even less viable for retirees.
However, there is a worrying amount of evidence to suggest that US workers are not willing or able to take
on this responsibility. According to the EBRI Retirement Confidence Survey, only 60% are currently putting
anything aside for retirement, and almost 30% of respondents had virtually no savings.41 Before the crisis,
it was estimated that 70% of baby boomers had not accumulated enough assets to maintain their
pre-retirement lifestyles; since then, households with a head aged 45-64 have seen their net worth
fall by more than 45%.42
37 Towers Watson 2010 Global Pension Asset Study.
38 ISI (31 August 2010), Accounting and Tax Report. Estimates are based on YTD asset returns (more or less flat for a
standard pension fund asset allocation) and a 100bps fall in the Aa corporate bond yield, which determines the
discount rate for pension plan liabilities. A lower discount rate implies a higher liability, so funding deteriorates despite
the asset side remaining stable.
40 EBRI (2010), as above.
41 EBRI (2010), as above.
42 Allianz (2008), as above; Employee Benefit Research Institute (Feb 2009), The Impact of the Recent Financial Crisis on
401k Account Balances.
43 Domhoff (2010), Wealth, Income and Power.
46 Allianz (2009), as above. 19
The issues discussed not withstanding, the US does have the most favourable demographic profile of any
country in the developed world. Higher fertility and a (relatively) relaxed attitude to immigration imply that the
prospects for sustained economic growth and consumption are better than in Europe, Japan and, in the
long-run, even China.
However, the ageing population poses significant problems for some companies and industries.
Occupations with the greatest replacement needs as the boomers retire include airline pilots, engineers,
construction workers and mechanics. As discussed, the US manufacturing sector is likely to face skills
shortages and rising labour costs, eroding competitiveness versus low-cost Asian economies (although
clearly relative currency values will also play a key role over the next decade). Unsurprisingly, given their
relatively old workforce, these are also the industries with the largest unfunded pension liabilities. GM, Ford,
Boeing, GE and Delta Airlines all appear in the top ten most underfunded plans. Given declining bond yields
and volatile stock market returns, current accounting assumptions appear untenable, and adjusting these will
result in higher pension costs that could jeopardise earnings by as much as 10% for some large-cap names.
In our opinion, the US is very much a stock-picker’s market. Some companies are well-positioned to benefit
from demographic trends both domestically and internationally, and have demonstrated the ability to
manage their labour costs and pension liabilities. Others are simply value traps. Its demographic profile
bestows on America a ‘country competitive advantage’, but only some companies will be able to translate
this into shareholder returns.
The views and opinions contained herein are those of Virginie Maisonneuve and Katherine
Davidson and may not necessarily represent views expressed or reflected in other Schroders
communications, strategies or funds.
2 INVESTMENT THEMES
2.4 “Nothing is permanent but change”1
The Economic Impacts of demographics
Virginie Maisonneuve Head of Global and International Equities and Katherine Davidson Research Associate
The world is undergoing a dramatic demographic transition. While the global population is set to swell to They travel less on public transport and buy far fewer cars, but are the main consumers of package holidays and
over 9 billion over the next forty years, falling fertility and continued improvements in life expectancy will also cruises. Academic work shows that an investment strategy based on these simple observations can be highly
result in rapid population ageing. profitable, as investors ignore long-term variables and persistently under-react to slow-moving trends.2
Such seismic shifts in the size and shape of the world’s population will affect consumption patterns, Our proprietary work suggests that global healthcare spending will grow twice as fast as most other areas
economic growth, and international relations. However, while the potential macro-economic consequences from 2010, concentrated in areas such as dialysis, orthopaedics, and treatments for age-related diseases
of demographic change are often discussed, the medium to long-term impacts on the global competitive (heart disease, stroke, dementia etc). Another key beneficiary of ageing is the financial sector, as public
landscape and individual companies are rarely considered. We believe that changes in global wealth pension reforms will necessitate, and incentivise, private saving for retirement. Life insurers, asset managers
dynamics will present new and exciting investment opportunities, so it is essential for investors to have and financial advisers are all well placed, especially those targeting older customers by offering annuities
a thorough understanding of the ways in which demographic themes are evolving. and equity release products.
In emerging markets, economic development will also shape consumption patterns. The general ‘rule’ is that,
Demographics – the danger of over-simplification as income increases, consumption shifts away from necessities, especially food, towards discretionary items
Over the coming decades, most developed countries will see their workforce shrink, and many their whole such as education, travel and recreation.3 Consumers in Asia and Africa spend 50-60% of their income on
population. In some southern European states the fertility rate is so low that, if sustained, the population will food, compared to the developed world average of 11.5%.4 At the other end of the scale, the proportion of
halve by 2050. Developing countries will, on the other hand, account for 98% of global population growth spending on recreation and culture in developed countries is twice that in China and five times that in India.
over the next forty years, with many receiving an economic boost as their young populations mature. As income rises, companies providing discretionary and aspirational products and services will benefit.
A simplistic interpretation of these facts could suggest that investors should favour emerging over developed As well as influencing domestic consumption, demographic trends will alter the composition of consumption
countries. The reality, however, is far less clear cut: the outlook for individual countries depends on complex globally. By 2050, 85% of the world’s population will be in countries that are currently considered ‘less
interactions between demographics and other factors, notably government policy. In this article, we explore economically developed’, meaning that, for every Western consumer, there will be six in Asia, Africa or Latin
some of the likely impacts of demographic change and explain how we, as investors, are seeking to benefit America.5 Some global brands will benefit as globalisation has broadened recognition, but we should still
from these trends. expect new products to be demanded and the global brand hierarchy to change over time. While wealthy
Chinese consumers currently favour well-recognised foreign brands (BMW, Coke etc.), this may well change
in coming decades with the accession of Chinese brands to the global market. It is therefore important for
Consumption and demographics investors to be conscious of how well brands can adapt to the changing global marketplace and recognise
One of the simplest, but most important, ways demographic change matters for companies and investors that local brands, though often underestimated, can constitute a competitive threat.
is via its impact on consumption. Consumption patterns vary by the age of the consumer, so as the age
But, of course, demographic change will not only affect consumption, but all areas of the economy via its
structure of a population changes, the types of goods and services that are demanded will also change.
impact on economic growth. The mechanisms are fairly complicated but we now explore some key points.
For example, older people spend less on education, clothing and entertainment than younger people, but
more on healthcare, books and utilities.
1 Heraclitus (535-475BC).
2 Dellavigna and Pollet (2005), Attention, Demographics and the Stock Market.
3 The relationship between food and income is known as ‘Engel’s Law’, and was first observed in 1856.
4 World Bank International Comparison Programme (2005), Schroders.
5 UN World Population Prospects. 21
What are the implications for growth?
Academics have found that neither population growth nor size has a significant impact on GDP, but a As the labour supply shrinks, the second factor- labour productivity- will become the most important driver
stronger relationship exists for population age structure. Per capita GDP growth is closely correlated with of economic growth.7 The impact of demographics on productivity is hard to detangle, and depends
the proportion of the population that are of working age, so countries with a young population reap a crucially on government policy. Consequently, it will be crucial for investors to understand the prospects for
‘demographic dividend’ as it matures.6 The necessary corollary is that rising dependency will tend to different countries to implement long-term strategies for productivity and flexibility, despite the inherent
dampen economic growth, which bodes ill for rapidly ageing countries such as Japan. short-termism of political incentives.
There are three main channels through which demographics can influence economic growth: labour supply, The third factor- capital intensity- could either be positive or negative for growth. On the one hand, a
labour productivity, and capital intensity (the quantity of physical capital relative to labour). The most shrinking labour force will cause firms to use more capital instead; capital deepening raises labour
straightforward of these is labour supply, which generally moves in line with the working age population. productivity and, consequently, GDP per capita. On the other, ageing can affect capital intensity by
Forecasts for the working age population in eight major global economies are shown in the chart below. changing savings behaviour and the supply of funds for investments. The seminal ‘life cycle model’ predicts
The outlook is negative for the developed world, joined by Russia and China in the near future, while Brazil that individuals will save during working life and run down savings in retirement.8 The aggregate savings rate
will enjoy a demographically-induced boost until 2025 and India until beyond 2040. will then rise with the proportion of the population in working age, and population ageing actually limit capital
deepening by reducing funds available for investment. There are reasons to be sceptical of this conclusion,
Figure 1: Proportion of the population aged 15-64, % as we shall see below.
Working age population, % of total The most important take home is that the large and unambiguously negative impact of labour supply in
75 ageing societies implies lower economic growth. A recent European Commission report estimates that
potential GDP growth in Western Europe will average 1.6% p.a. to 2050, which will be driven entirely by
70 anticipated productivity improvements. In Eastern Europe the change will be more dramatic, with growth
falling from over 4% to just 0.9%, and the outlook is similarly bleak for Japan.9 More favourable
demographics (higher birth rate and immigration) will sustain growth in the United States: GDP per capita
will be 30% higher than in Europe and Japan by 2050.10 But it’s not all bad news: slower global population
growth could improve our chances of tackling global warming and, at the risk of sounding Malthusian, ease
60 the pressure on natural resources and food supplies.
2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Brazil China India Japan
Russian Federation United Kingdom United States of America Western Europe
Source: UN World Populations Prospects, 2008 Revision.
6 Bloom, Canning and Sevilla (2003), The Demographic Dividend.
7 Batini et al (2006), The Global Impact of Demographic Change; European Commission (2009), The 2009 Ageing Report.
8 Modigliani and Brumberg (1954), Utility Analysis and the Consumption Function: An Interpretation of Cross-Section Data.
9 European Commission (2009), as above; Canone et al, as above; Oliveira Martins et al (2005), The Impact of Ageing on
Demand, Factor Markets and Growth.
22 10 Oliveira Martins et al, as above.
2 INVESTMENT THEMES
Who’s going broke?
As well as reducing economic growth and government revenues, an ageing population is more expensive to In the absence of major fiscal reform, ageing will put substantial pressure on government budgets. It is
support. The combined cost of public pensions, health and long-term care in Europe is expected to reach extremely hard to make reliable country-level projections because of changes in government policy, but
an average of 28% of GDP by 2050.11 In the US, while the demographics are more favourable, rapid benefit recent work by the OECD indicates that pension spending alone will cause the government deficit to worsen
growth could drive public sector spending on healthcare alone to 40% of GDP by 2060.12 Japan, conversely, by an average of 3.3% of GDP by 2050, and as much as 8% in Spain.15
reformed its healthcare system in the 1980s and 90s, and spending is only expected to reach 10% of GDP
Factoring the costs of the current crisis and demographic decline, the IMF expects average government debt
despite rapid ageing.13
in the advanced G20 countries to exceed 300% of GDP by 2050, which could spark fears of government
A recent IMF report on the financial crisis has used these estimates to compare the cost of ageing with that insolvency.16 While the burden of ageing falls disproportionately on the West, emerging markets face the
of the current financial crisis. Figure 2 shows total current and future spending on each in net present value additional challenge of sustaining growth and balancing society’s demands without sacrificing fiscal restraint.
terms, meaning that figures in the future have been discounted by a theoretical interest rate to find the Needless to say, investors will not tolerate the same levels of debt in developing as developed economies.
equivalent sums in today’s money. They find that the fiscal burden of the crisis (including fiscal stimulus
packages, financial sector support and automatic stabilisers) amounts to less than 11% of anticipated Figure 3: Government debt, advanced G20 countries, % of GDP
agerelated spending.14 350
Figure 2: Net present value of fiscal impact of the financial crisis and population ageing
% of GDP Cost of crisis/cost of ageing
800 25% 250
500 15% 150
300 10% 100
2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Source: IMF (2009).
Canada Korea Spain US Australia UK Germany France Mexico Turkey Italy Japan
Crisis Ageing In this environment, governments must manage the delicate balance between fiscal prudence and
Source: IMF (2009). maintaining the correct incentives. Higher taxes and social security contributions would reduce incentives
to work, precisely when the need for higher productivity and participation is greatest. Furthermore, higher
taxes could limit saving and government borrowing could crowd out private investment.
The message for investors: always keep one eye on government policy when assessing the macroeconomic
investment environment. The observations in this sector also support our earlier recommendations on
11 European Commission (2009), as above.
healthcare and financials companies: the former because government spending will rise as well as personal
12 Hagist and Kotlikoff (2005), Who’s Going Broke: Comparing Healthcare Costs in Ten OECD Countries; Congressional
Budget Office (2007), The Long-Term Outlook for Health Care Spending. consumption, and the latter because the burden of saving for retirement will be shifted onto individuals as
13 OECD (2006), Projecting OECD Health and Long-Term Care Expenditures: What are the Main Drivers? government budgets come under increasing pressure.
14 IMF (2009), Fiscal Implications of the Global Economic and Financial Crisis.
15 OECD (2001), Fiscal Implications of Ageing: Projections of Age-Related Spending.
16 IMF (2009), as above. 23
Ageing and asset prices: Don’t panic! Ageing and asset allocation
Enough of the nuanced arguments, we hear you say, what of the relationship between demographics and So we have dispensed with the meltdown hypothesis: demographics should have no impact on asset prices
asset prices? If there is a direct link, it operates through the ‘life cycle hypothesis’. As described above, as a whole. But what about relative returns on different asset classes? It is a common viewpoint that
economic theory tells us that, in retirement, households should be running down their savings and wealth individuals should allocate an increasing proportion of their portfolios to bonds as they age, as they become
to fund consumption. The difficulty is that most wealth is held in the form of securities or property, which more reliant on investment income and therefore more risk averse.23 However, there is also a counter
cannot simply be spent but must be sold to provide disposable income. This observation has spawned the argument that risk aversion actually falls with age, because young people are less able to bear short-term
so-called ‘market meltdown hypothesis’. investment losses because of their limited financial wealth. Either of these hypotheses would, if correct,
have implications for multi-asset portfolios and pension funds.
“The words “Sell? Sell to whom?” might haunt the baby boomers in the next century. Who are the buyers
of the trillions of dollars of boomer assets? [They] threaten to drown in financial assets…”17 Empirical studies show that today’s older generation are indeed wary of equities, but this is generally
thought to be due to their experience of the Great Depression rather than their age per se. There is no
The argument is that the higher supply and lower demand for financial assets will drive down asset prices
reason to think that Baby Boomers, who have seen equities generate positive returns over most of their
in the developed world as boomers retire and pension funds become net sellers. Some studies suggest
lives, will have the same asset allocation. There is no evidence that portfolio shares change with age in the
that baby boomers accounted for as much as 30% of the rise in the US stock market in the 1980s-90s
United States, and most individuals don’t actually change their asset allocation at all over a ten-year period.24
and predict massive falls in shares over the next 20 years, potentially to P/E ratios as low as five.18
It therefore seems unlikely that ageing will cause a structural shift out of equities. Conversely, we would
While the market meltdown hypothesis makes good headlines (and some intuitive sense) there are several
expect both individual and institutional investors to shift their portfolios towards equities to meet the higher
reasons why we would not expect it to hold.19 Most fundamental is the lack of evidence for the core
return objectives necessitated by longer life expectancy and mounting pension liabilities.
assumption of life-cycle savings behaviour, especially in relation to the elderly. A number of studies find that
retirees continue to save, presumably due to bequest motives, uncertainty about their life expectancy, and
the possibility of high healthcare costs.20
The East will hold the purse strings
A further reason to be suspicious of the ‘meltdown’ scenario is that the underlying models almost always
assume a closed economy, with no trade or capital flows. In reality, capital moves reasonably freely within
the OECD, and indeed globally, breaking the identity between domestic savings and investment.21
Modelling exercises have found that an increase in the proportion of population in working age induces
net capital outflows (investing abroad), because saving is higher than is necessary to finance domestic
investment. As the population ages, savings decline and interest rates rise, attracting capital inflows (foreign
companies investing here) and supporting returns to capital. Assuming free capital flows within the OECD
alone changes the prediction from ‘meltdown’ to a modest decline in the rate of return to capital: less than 17 Siegel (2007), Impact of an Ageing Population on the Global Economy.
1% by 2070.22 18 Bergantino (1997), Life Cycle Investment Behaviour, Demographics and Asset Prices; Geanakoplos et al (2002),
Demography and the Long-Run Predictability of the Stock Market.
According to these models, the Western world will see past capital outflows revert to net inflows when 19 Poterba (2004), The Impact of Population Ageing on Financial Markets, finds weak evidence for a relationship between
population ageing becomes more pronounced. Developing countries, such as India, Indonesia and Turkey, T-bills, bonds and the age-structure in the United States over the last 70 years, and no statistically significant
will become net capital exporters. Since the capital account and current account (trade in goods and relationship for stocks.
services) are opposite sides of the same coin, this means the developed world will have to run trade 20 In the UK, saving even appears to rise over the age of 70. This could be because ill-health and infirmity make the act of
consumption more difficult for the very old, though part of the observed results can probably be attributed to a sample
surpluses and the developing world deficits: a reversal of the current situation. This could have dramatic selection problem: life expectancy and wealth are positively correlated, so at higher ages the sample is increasingly
implications for global power balances. It is anyone’s guess how international relations will play out when drawn from wealthy persons who have a higher propensity to save throughout the life cycle. [Demery and Duck (2001),
the developing world holds the purse strings. Saving Age Profiles in the UK.]
21 Borsch-Supan (2005), Demographic Change, Saving and Asset Prices: Theory and Evidence
23 Malkiel (1996) and Hanna and Chen (1997), cited in Wang and Hanna (1997), Does Risk Tolerance Decrease with Age?
24 24 Ameriks and Zeldes (2004), How Do Household Portfolio Shares Vary with Age?
2 INVESTMENT THEMES
Demographic change has serious implications for the global economy and any investment strategy,
yet remains poorly understood. Too often regarded as a distant issue, many investors neglect to
consider demographics in their analysis of a company’s operating environment or, indeed, the
sustainability of its earnings.
We believe that a good grasp of the topic improves the ability of investors to capitalise on changing wealth
dynamics and new business opportunities. A simplistic understanding could lead investors to aggressively
expand their emerging markets exposure, but our research suggests that dichotomy of developed and
developing countries does not map neatly onto either demographic trends or growth prospects. Some
emerging markets will age rapidly; others will struggle to manage their younger populations. Meanwhile,
there will still be profitable pockets of opportunity in those developed countries with a robust policy
framework – regardless of their demographic profile.
We are looking for companies with the ability to adapt to and cater for changing patterns of demand in their
markets. The challenge is to identify successful companies early, anticipating how shifts in consumption will
affect the relative power of global and local; established and nascent brands.
Demographic change is a process of adaptation and evolution – a process that we believe necessitates
an innovative and pragmatic investment strategy. As Heraclitus observed in 500BC, “nothing is permanent
The views and opinions contained herein are those of Virginie Maisonneuve and may not
necessarily represent views expressed or reflected in other Schroders communications,
strategies or funds.
2.5 The impact of climate change and resource shortages
Sir David King Senior Scientific Advisor to UBS, Director of the Smith School of Enterprise
While human beings have been present for only a fraction of the earth’s several billion years of evolution, Unique challenges
we have nevertheless coevolved over a long period with the geology of the planet. As a result, we have an
ecosystem with the right balance for human habitation and yet, an amazing period of economic and This level and density of the global population presents major challenges that are unique to the 21st Century:
industrial development has resulted in the profligacy and mismanagement of our natural resources.
– Food Production
For some commentators, including Sir David King, Senior Scientific Advisor to UBS and previously the – Conflict and Terrorism
UK Government’s Chief Scientific Adviser and the Head of the Government Office of Science, this unique
– Water Resources
ecosystem means there is only one chance to get things right.
– Energy Security and Supply
We invited Sir David to Schroders’ Secular Market Forum, a medium for debating the issues that will impact
– Health and Development
markets through the cycle. The following provides a summary of his thoughts on the environmental
challenges we face in the 21st Century and how effective management of the world’s resources means it is – Ecosystems
still possible to successfully tackle the climate change phenomenon. – Climate Change
Dealing with the legacy of huge population growth These are all interconnected.
The human race has a track record of mismanaging the earth’s natural resources, as illustrated by the
destruction of China’s Loess Valley ecosystem during the 1400s. Overfarming and a lack of land Water
management resulted in extensive top soil erosion and sand clouds that still impact Beijing, Korea and
Japan today. At that time, the solution was to relocate to a more fertile region. Clearly, with a global – Without managing our supplies more effectively, the world is set to face a global freshwater shortage
population of 6.8 billion people, this option is not open to us today. by mid century. Shortages are already appearing on a regional level, with the ‘green belt’ of South
Australia a prime example. Traditionally a significant area for crop production, the state of Victoria has
The population problem results from the huge economic, scientific and industrial developments of the 18th, faced eight successive years of drought, such that one third of domestic water is sourced through a
19th and 20th centuries, which led to significant wealth creation and a doubling of life expectancy. process of desalination.
Population growth tends to slow in the third and fourth generations (indicating a plateau by around 2060),
– Desalination is not a simple solution given the highly energy intensive production process, which is often
but this still means we should be planning for a total global population of approximately 9 billion by the mid
fuelled by coal. This not only creates energy supply and security concerns, but the burning of coal also
contributes to climate change and to desertification.
“Some eminent cosmologists think that the future for humanity is to get into a spaceship and – We haven’t yet mastered solar-driven desalination, but this innovation challenge would be a big win
find another planet, but that planet will not have the right levels of ecosystem services that this going forward.
planet provides. We only have one chance.”
2 INVESTMENT THEMES
– Climate change poses a challenge for health in the sense that diseases are moving to new locations as – Wealth creation in the developing economies is increasing meat consumption, which necessitates further
temperatures rise. Furthermore, an increasingly globalised economy means that air travel can facilitate crop production and results in greater water consumption.
the spread of a new infectious disease to the rest of the world within three months. – Fish consumption per head has increased and massive overfishing means that there could be no more
large fish in the oceans by the mid 21st Century. Recent experiments to establish fishing protection zones
Minerals (such as in the Cape Cod region) have been proven to work and should be replicated on a larger scale
– Dwindling resources and the uneven distribution of minerals around the world create economic elsewhere in the world.
challenges, but also the potential for conflicts of interest. – We also need to be sensible about the management of disease. Incidences of TB in UK cattle, for
– Estimated copper reserves, for example, would be exhausted in 50 years if current consumption trends example, have increased significantly since 1986, threatening the UK’s dairy industry. Badgers have
are extrapolated. Demand continues to grow, however, notably from China. Its resource needs cannot played an important role in spreading the disease and culling their population may have contained its
be met internally and Africa is proving an important investment destination. This has parallels with the spread. Yet badgers remain a protected species, despite being far from an endangered species.
European colonisation of Africa during the Industrial Revolution.
Ecosystems – The temperature of the earth went through a maximum 50 million years ago, at around 12 degrees higher
– Given the role the forests play in controlling carbon dioxide and in providing the biodiversity necessary than pre-industrial temperatures. Greenhouse gas levels were significant at that time.
for human survival, there is a conflict of interest with food production. – The development of civilisation in the past 12,000 years, with the introduction of farming processes and
– Food production needs to increase by 50% within 15 years to satisfy current growing demand. The challenge deforestation, has been accompanied by a significant increase in greenhouse gas emissions.
is whether this demand can be met without removing the ecosystems on which we rely (which is what is Greenhouse gases have risen from the 275 parts per million (ppm) typically seen during a ‘warm period’
happening at the moment). (as opposed to an ice age) to 388ppm today, and are still rising at 2ppm per year. However, we have not
yet seen a corresponding increase in temperature levels, with the global average only about 0.8 oC higher.
Food supply (While greenhouse gases are by no means the only factor behind temperature rises, it seems evident that
they are a significant contributor.) According to data from the Hadley Centre, if we reach 450ppm, there
– Given the need to increase food production, we need to be far smarter about capacity per hectare than is a 20% chance that the global temperature will rise by over 3.5oC – significantly beyond the global
we have been. This means farming efficiently, soil management and the implementation of crop agreement to prevent a temperature rise greater than 2oC. If we follow a business-as-usual path and
development techniques. levels exceed 650ppm, a rise of over 3.5oC becomes the most likely outcome.
– After 15 years of research and sophisticated selected breeding, we have developed the means to grow – The long-term challenge posed by this is clear – the melting of the Greenland ice sheet alone would
flood resistant rice crops. The same result could have been achieved more rapidly with the use of genetic result in a six metre rise in sea levels around the world.
modification techniques. However, the European aversion to GM crops is holding back the commercial
application of these important new methods. To meet the rising demand for food these techniques will be
required to generate sufficient yields. The technology is available to provide flood resistance, drought
resistance and saline resistance crops, but the social acceptability of genetic modification is not.
Energy security and supply Where now?
– With transportation systems heavily dependent on oil, the production capacity associated with The financial crisis has provided somewhat of a breather for the environment by reducing the volume of
conventional oil is not sufficient to meet rising demand. Furthermore, only around 15% of the remaining emissions produced globally. However, business as usual will put us back on a dangerous track.
conventional oil reserves are in the hands of international oil companies – 85% resides in the hands of Encourageingly, the Copenhagen Climate Change conference and the surrounding publicity meant that 76
national companies. The concentration of oil and mineral resource in certain nations and the consequent countries arrived with their proposals, and highlighted their commitment to tackling climate change. These
potential for conflict is very concerning. 76 nations currently account for 85% of greenhouse gas emissions.
– The dependency on oil and the risk of conflict means we need to look at alternatives within
unconventional oils (such as tar sand oil), of which there is a vast supply, but significant difficulties Country by country, there seems to be a real understanding of the likely impacts of climate change, and it is
associated with their use. Indeed, demand trends are already pushing beyond the boundaries China that appears to be doing the most. The country is already the biggest producer of nuclear energy and
appropriate for the technology that currently exists. the biggest producer of photovoltaics, playing a key part in driving prices down. Whilst much more needs to
be done in terms of tackling climate change and effectively managing our resources, it would seem that this
– The key message is that we need to remove our oil dependency as quickly as possible and, in general,
widespread international recognition sets us off to a good start.
we need to de-fossilise our economy by the middle of the century. If we don’t meet the oil challenge,
we look set to face an oil price crisis with the potential to trigger another financial shock.
The views and opinions contained herein are those of Sir David King, and may not necessarily
– Our oil production processes to date have been simple and wasteful given the perception of oil as a
represent views expressed or reflected in other Schroders communications, strategies or funds.
cheap commodity. Data from Cambridge University highlights massive inefficiencies in the current
conversion rate from primary energy to total useful energy (of the total primary energy burnt globally
per annum, only 12% is produced as ‘useful energy’).
– In this sense, we are facing one of the greatest innovation and wealth creation challenges since the
beginning of the industrial revolution. There are massive opportunities for energy efficiency gains and
for finding alternative means of energy production, but these also present huge technical and economic
challenges. If we are going to tackle the oil challenge effectively and optimise GDP growth, we need
to focus on large scale infrastructure like grids and energy production processes.
Loess – an excellent example of environmental management
To come full circle, it is worth returning to China’s Loess plateau. 12 years ago, the Chinese government
began replanting the area on an enormous scale. To date, an area the size of Belgium has been regreened,
with the full plateau (approximately the size of France) set to be regreened by 2020. The terracing of the area
and the creation of arable land has been a source of wealth creation for the local farmers, and other
biodiverse systems are returning. This is an astonishing example of effective land management.
2 INVESTMENT THEMES
Karen Shaw Climate Change Specialist
Biofuel is an alternative transportation fuel to fossil fuels, derived from renewable resources and currently In Brazil, the biofuel market is competitive, primarily based on sugar cane, and anhydrous bioethanol which
almost entirely derived from food crops or their residues. Theoretically, biofuel can be produced from is mandated at a blending rate of 20-25%. Biodiesel blending was launched in 2004 and raised to a
cellulosic biomass, municipal solid waste, forestry waste, energy crops and novel crops such as algae. In mandatory 5% in 2010. Advanced biofuel production is not commercialised yet either. The US Environmental
practice, these technologies have not been scaled up commercially and many are still under development. Protection Agency (EPA) has designated that Brazilian sugar cane meets a 61% reduction on GHG and it’s
likely the EU will also. With increasing investments from the US and EU in the Brazilian biofuel industry it
Despite this, the global biofuel market, dominated by Brazil until 2000, has grown rapidly and is now focused
appears that Brazilian bioethanol exports could increase from the circa 5 billion litres of export in 2009, of
around three markets: the US, Brazil and the European Union (EU). These three markets in 2009 produced
which approximately 50% is estimated to have reached the US and another large percentage to the EU.
circa 90 billion litres of biofuel globally; the US just less than 44 billion litres (96% bioethanol, 4% biodiesel);
Brazil 31 billion litres (96% bioethanol, 4% biodiesel) and the EU 15 billion litres (27% bioethanol, 73%
biodiesel),(Cheuvreux 2010). The US and EU markets have emerged largely since 2000, driven by Biofuel
government regulation, industry targets and financial support. However, new stringent sustainability criteria
What is it?
threaten the future growth potential of the existing EU and US biofuel industries, discussed below.
Currently, the transport sector is largely dependent on petrol and diesel derived from crude oil as a fuel
The US is the largest producer of bioethanol, which is almost entirely made from corn starch from the mid source; in 2009 it was 98% dependent, contributing about 25% of global energy related carbon emissions
west of the country. The government has set a target for corn starch based biofuel production to be no and 50% of global oil consumption (Cheuvreux 2010).
more than 56 billion litres. This target could be reached as early as the end of 2010 (United States
Biofuel is an alternative transportation fuel to petrol and diesel fuel. It is made from renewable resources (plant
Department of Agriculture 2010). Under the Environmental and Social Impact Assessment (ESIA) 2007
biomass) and theoretically can be made from food crops, (sugar cane, corn, wheat etc) cellulosic biomass, (food
regulation, another 60 billion litres has to be produced from cellulosic biofuels, (meeting a minimum 60%
crop residues, forestry residues etc.) municipal solid waste or more dedicated energy crops, such as switch
Green House Gasses (GHG) life cycle reduction requirement). As this technology is still under review, many
grass, Jatropha, fast growing trees and novel crops such as algae. Today, it is mainly made from food crops, as
expect federal targets to be revised downwards or the shortfall to be made up with Brazilian imports (if the
biofuels from these other feed stocks are generally not produced yet in commercially viable quantities or prices.
import tax is lowered). The US biodiesel industry is awaiting the retroactive extension of federal support
which expired at the end of December 2009; the industry will disappear without this support. Although biofuel has been used in Brazil since the 70s, there has been a more recent shift towards biofuel globally,
an attempt to move the transport industry away from fossil fuel dependency, thereby reducing exposure to high oil
In the EU, 73% of all biofuel is biodiesel and made from rapeseed oil. The rest is bioethanol, produced from
prices (improving national energy security) and reducing carbon emissions and combating climate change.
wheat and corn. Currently production is also based on food crops which meet the minimum 35% GHG life
cycle reduction requirement. However, this rises to a 50% reduction requirement in 2017, including gases
Climate change – mitigation
such as methane and di-nitrous oxide which may only be met by more advanced cellulosic biofuel
Research has found that when the GHG life cycle of biofuels is compared to the GHG life cycle of crude oil
processes or potentially bioethanol imports from Brazil. At the moment, only Germany, France and Austria
derived from petrol and diesel, the biofuel life cycle can (though not always) demonstrate significant net GHG
have met, or are in line to meet, the EU target for 5.75% biofuel usage of all used traffic fuel by 2011.
savings. One component of this research is that plants are considered energy sources that do not contribute
European oil and gas companies who are struggling to meet the EU fuel quality directive blending
to the accumulation of GHG in the atmosphere. The GHG generated when they are used are reabsorbed
requirements are finding it cheaper to pay the penalties than find biofuel. Shell and BP have set up joint
during the growing of the next harvest, there is a balance between the emission and absorption of pollutants
ventures with Brazilian bioethanol companies to manage this going forwards.
and close to net zero emissions. Fossil fuels, on the other hand, release circa 80% of carbon dioxide in the
GHG life cycle upon combustion.
The specific GHG savings obtained by biofuel instead of fossil fuels are under current international debate. In the future, the use of prime agricultural land for biofuel will continue to be contentious because of political
Results depend on many variables implicit to the choice of feedstock, choice of inputs (such as fertiliser concern that scarce prime agricultural land is being used to grow biofuels not food, causing energy versus
use), conversion technology and methodological assumptions. As a result no consensus has been food security concerns.
established internationally. Whilst the EU and US governments have adopted GHG life cycle methodologies
these are acknowledged to be imperfect and still under development. Figure 1, is taken from a United Climate change – adaptation
Nations Environmental Programme (UNEP) 2009 study and shows the wide range of net potential GHG From an adaptation perspective, if previously unused ‘brown’ land or non ‘green’ land was planted with
savings that can be achieved. The highest are clearly for sugarcane, and potentially also for biogas from biofuel crop (e.g. forest), this might provide an additional carbon sink or offsetting option. However, a
manure or biodiesel from waste oil and biofuel from agricultural and forestry residues. scientific GHG life cycle analysis would be required to assess the GHG life cycle of the land prior to use and
to compare this to proposed use. Algae grown on a huge commercial scale on specialist ‘farms’ might also
Figure 1: GHG savings of biofuels compared to fossil fuels (UNEP 2009) be net positive in GHG emissions but further scientific research is required here.
% GHG emission saving compared to fossil fuels
110% 110% 174% Biofuel definitions
Biodiesel from In the last ten years biotechnology has developed greatly as an industry and there are now varying types of
palm oil FT diesel biofuel investments on offer in the market. Definitions of biofuel types are not consistent globally and vary
from wood from being crudely simple to very complex. The EU Commission broadly distinguishes biofuels between
60 Bioethanol from Bioethanol from crops that ‘can be eaten’ as 1G, and crops that ‘cannot be eaten’ as 2G. A more analytical version is used
sugar cane agriculture or
forestry residues in the US which defines biofuels from corn starch as 1G; biofuels from other parts of the corn plant as
40 ‘advanced biofuels’ (2G) and biofuels from non corn plants as ‘cellulosic biofuel’ (2G).
Bioethanol from sunflower Industry prefers to use a more analytical version still which suggests four generational types of biofuel or
Bioethanol Biodiesel from Biomethane 1G-4G. The 3rd Generation (3G) and 4th Generation (4G) technologies are largely in research and
0 from wheat rapeseed from manure development still (not even 2G has been fully commercialised), which is why regulators have not introduced
more advanced terms to date. However, the definitions are interesting as they include some sustainability
criteria and a reference to the type of land use. In the future, it is likely that once GHG life cycle emission
soya beans methodologies take into account land use change and supply chains become more traceable and certified
that more specific definitions of biofuel types will emerge in the form seen below.
-868% to -2070%
– 1G – biofuels from the conversion of crops on prime crop land into biodiesel or bioethanol
Sources: Own compilation based on data from Menichetti/Otto 2008 for bioethanol and biodiesel, IFEU (2007) for sugarcane – 2G – biofuels from marginal crop lands unsuitable for food production or using non food crops and
ethanol, and Liska et al. (2009) for corn ethanol; RFA 2008 for biomethane, bioethanol from residues and FT diesel. residues (including cellulosic ethanol technology and energy crops such as Jatropha, also Fischer
Tropsch and biomass gasification)
The extremely wide range of GHG emissions savings or losses for biodiesel from soya or palm oil reflects – 3G – biofuels made using non arable land, based on integrated technologies that produce feedstock and
concern about the use of land prior to it being converted to biofuel production. The EU GHG life cycle fuel and require the destruction of biomass – e.g. algae
methodology has created controversy as it does not fully consider land use change (though this is likely to
– 4G – biofuels which are made using non arable land and do not require the destruction of biomass.
be rectified in 2011). So there has been concern that land used to grow biofuel could be sourced from areas
of previously high biodiversity value (a carbon sink), the loss of which would partially or fully negate the
potential savings made from using biofuel rather than fossil fuels. In Figure 1, biodiesel produced from soya Clearly, the more evolved the biofuel technology from 1-4G the smaller the environmental footprint and the less
beans and palm oil can produce a net loss of GHG emissions savings (i.e. more emissions than fossil fuels), impact on food security and pressure on habitat loss, as arable land is not required for 3G and 4G technology.
depending on prior land use (e.g. if rainforest was taken down in Indonesia or Brazil). In this report, where we are not referring to regulatory terms we have used the following definitions:
Our research indicates that current biofuel production is predominantly grown on agricultural land in the US 1G – biofuels made from corn starch (in the USA) or from edible food crops in the EU and rest of the world.
and EU or on unused land or pasture land in central Brazil. Nonetheless, verification procedures for the 2G – biofuels made from anything else (including biomass, energy crops and algae).
industry are still under development. The GHG life cycle methodologies that are apparently conducted by
companies may not be accurate and are not audited at present by any government regulatory body.
2 INVESTMENT THEMES
Market overview Table 1: World biofuel production/capacity (2009)
Million litres EU US Brazil
Current world fuel ethanol production
Figure 2 below, shows historical world biofuel production and places regional production in a global context. Biodiesel Bioethanol Biodiesel Bioethanol Biodiesel Bioethanol
Clearly Brazil was the major world biofuel producer (with sugar cane based bioethanol) until a few years ago. Production 10,872 3,935 1,682 42,013 1,386 29,816
However, the US has ramped up production in corn based bioethanol and has become the global leader in
Capacity 23,546 6,131 10,200 49,730 N/A 34,000
biofuel production. The EU contribution to global production is still relatively small and the rest of the world
also makes a minor contribution. 473 billion litres 85 billion litres
Petroleum consumption by transport 726 billion litres
(IEA 2006) (transport fuel consumption)
Figure 2: Global biofuel market (1975-2007) IEA estimates for 2020 27,000 113,000
Mio Litres Source: Cheuvreux (2010).
Figure 3, below, shows the historical and current breakdown in bioethanol and biodiesel production
60,000 between the regions of the US, Brazil and the EU. Clearly, the EU is a relatively significant biodiesel producer
50,000 and insignificant bioethanol producer. In contrast, both the US and Brazil are significant biofuel producers,
from corn and sugar cane respectively.
Figure 3: Bioethanol and biodiesel production (2009)
Million litres Million litres
Brazil USA EU Rest of the world
Source: eBIO & F.O. Licht.
Today production has risen still further (see Table 5) with EU biofuel production in total reaching just under
15,000 million litres (15 billion) in 2009, Brazil with production of 31,000 million tonnes and US biofuel 10,000
production just under 44,000 million litres (44 billion). Table 1, below, also illustrates the Cheuvreux view that 0
there is much greater capacity in the industry to produce more in certain regions, particularly in the EU and 07 08 09 07 08 09 07 08 09
in US Biodiesel, but this is dependent on financial support continuing in these regions. EU US Brazil USA Brazil EU
Source: Cheuvreux (2010), OECD, FRA. 2007 2008 2009
In summary, Table 2, below, shows the percentage split between bioethanol and biodiesel production for the Only a small target has been set for biomass based biodiesel (largely produced from soya beans). Most targets
three major producing regions. It is worth highlighting that for the US and Europe, this growth has been are for bioethanol, which is currently mainly produced from Midwest corn starch (1G), but some bioethanol is
dependent on financial incentives. A quick look at Table 3, for example, highlights the average feedstock costs produced from sugarcane (some imported from Brazil) and some from US sorghum. Table 4, highlights the
versus average prices in Q1 of 2010. Though these are volatile, biodiesel margins are generally tighter than year on year total renewable fuel requirement targets to 2022. The cellulosic biofuel requirements are a sub
bioethanol margins in both the US and EU. The high feedstock prices in Brazil reflect the volatile high price for division of the advanced biofuel requirement. So the total renewable fuel requirement, minus the advanced
sugar cane and sugar in Q1 of this year. The following pages review the regional markets in more detail. biofuel and biomass based biodiesel requirement gives the permitted allowance for 1G corn starch based
biofuel. Any 1G or corn starch based biofuel produced exceeding this amount does not serve the regulatory
Table 2: World biofuel production 2009 (Cheuvreux 2010) mandate and is not eligible for financial support.
Country Total production % Bioethanol % Biodiesel The RFS and total renewable fuel requirement cover all transportation fuel, including gasoline and diesel fuel
intended for use in highway vehicles and engines, and nonroad, locomotive and marine engines.
Brazil 31 billion litres 96% 4%
US 44 billion litres 96% 4% Table 4: ESIA fuel volume requirements (billion gallons)
Europe 15 billion litres 27% 73%
Year Cellulosic biofuel Biomass-based diesel Advanced biofuel Total renewable fuel
requirement requirement requirement requirement
Table 3: Biofuel and fossil fuel price trends 2008 n/a n/a n/a 9.0
Per litre UE (EUR/litre) US (USD/litre) Brazil 2009 n/a 0.5 0.6 11.1
Bioethanol 2010 0.1 0.65 0.95 12.95
Biodiesel Bioethanol Biodiesel Bioethanol
(anhydrous ethanol) 2011 0.25 0.80 1.35 13.95
Current price (August 2010) 68.00 0.54 0.97 0.74 0.57 2012 0.5 1.0 2.0 15.2
H1-10 prices (min-max-avg) 0.59-0.71-0.64 0.43-0.57-0.49 0.86-0.98-0.94 0.66-0.77-0.70 0.45-0.78-0.58 2013 1.0 a 2.75 16.55
H1-10 feedstock prices (min-max-avg) 0.49-0.73-0.61 0.17-0.28-0.20 0.73-0.97-0.84 0.13-0.18-0.15 0.68-1.3-0.99 2014 1.75 a 3.75 18.15
Source: Cheuvreux (2010). 2015 3.0 a 5.5 20.5
2016 4.25 a 7.25 22.25
US market overview 2017 5.5 a 9.0 24.0
2018 7.0 a 11.0 26.0
US regulatory policy
2019 8.5 a 13.0 28.0
The Energy Independence and Security Act (EISA, 2007), set up the national Renewable Fuel Standard
(RFS 2) programme and renewable fuel consumption targets were ambitiously set at 36 billion gallons by 2020 10.5 a 15.0 30.0
2022. The act broke this target down and set specific annual volume requirements for: 2021 13.5 a 18.0 33.0
– total renewable fuel requirements (which is all renewable fuel) 2022 16.0 a 21.0 36.0
– biomass based biodiesel 2023+ b b b b
– advanced biofuel (defined as non corn starch or kernel – i.e. the part of the plant which is edible) and, a To be determined by EPA through a future rulemaking, but no less than 1.0 billion gallons.
b To be determined by EPA through a future rulemaking.
– cellulosic biofuel (a sub division of advanced biofuel which is not corn based and is essentially just biomass).
2 INVESTMENT THEMES
The specific volume requirements for biofuels are ambitious and have not been met in 2010. The current Of the fuel pathways already modelled those which pass the required minimum GHG standards include:
cellulosic and biomass production levels do not meet the requirements in the above EISA Act. Table 5 shows
– corn based ethanol plants using new efficient technologies
the revised 2010 standards which reflect a more current and up to date market review. An estimated 98% of
all biofuel in the US today is made from corn starch (Gerson Lehrman Group 2010). – soy based biodiesel
– biodiesel made from waste grease, oils and fats
Table 5: Standards for 2010
– sugarcane based ethanol.
Fuel Category Percentage of Fuel Volume of Renewable Fuel The RFS also increases the total amount of biofuels to be blended with US gasoline. The US Environmental
required to be Renewable (in billion gallons)
Protection Agency (EPA), under the Energy Policy Act of 2005, has responsibility for ensuring that a
Cellulosic biofuel 0.004% 0.0065 minimum volume of renewable fuel is used for motor fuel in the US. The blending rate is fixed by the EPA
Biomass-based diesel *1.10% *1.15 each year to ensure that ethanol consumption attains the volumes required under the legislation. Previous
Total Advanced biofuel 0.61% 0.95 legislation set a 10% limit on ethanol blending with fuel. However, in 2010, the percentage of fuel required to
be renewable is 12.95 billion gallons which according to the EPA is equivalent to a blending rate of 8.25% of
Renewable fuel 8.25% 12.95
all fuel. In 2011 this may exceed the 10% blending rate. It is expected that the EPA will push for an increase
*Combined 2009/2010 biomass-based diesel volumes applied in 2010 in the current 10% legal limit on ethanol in motor fuel to 15% in order to ensure the RFS can be met going
forwards to 2022 or the government will need to introduce more Flex Fuel Vehicles (FFV), which currently
The RFS programme also sets specific life cycle GHG targets for each fuel type. As of December 2007, account for circa 3.5% of all vehicles on US roads. However, the EPA has to take into consideration the
renewable fuel must achieve a 20% reduction in carbon emissions when compared to a 2005 baseline life impact of ethanol, which is highly corrosive on engines, to ensure that automakers will continue to provide
cycle GHG emissions benchmark using traditional fuel. Advanced, cellulosic and biodiesel biofuel must a warranty on engines which are essentially FFV. This is currently being debated.
achieve higher reductions when compared to this baseline (see Table 6).
Table 6: Lifecycle GHG thresholds (EISA 2010) (percent reduction from 2005 baseline) The government has set a target for the production of corn starch bioethanol at 15 billion gallons (or 56 billion
litres). Over this level, corn starch bioethanol will neither count towards the renewable mandate (RFS 2) nor be
Renewable fuel** 20% eligible for government support. Currently, production of corn starch bioethanol is circa 12 billion gallons
Advanced biofuel 50% (45 billion litres) per annum. There are 201 plants in the US already with a capacity to produce 13.5 billion
gallons and additional facilities are under construction that will add another 1.2 billion gallons of capacity
Biomass-based diesel 50%
(USDA 2010). So the 15 billion gallons (or 56 billion litres) of corn starch target for 2022 is effectively
Cellulosic biofuel (ethanol and diesel) 60% already met.
**The 20% criterion generally applies to renewable fuel from new facilities that commenced construction Currently the government supports bioethanol production through the Volumetric Ethanol Excise Tax Credit
after 19 December 2007.
(VEETC) at $0.45 per gallon of ethanol blended with gasoline, though this is due to expire at the end of
December 2010. There is also a small producers’ credit at $0.10 per gallon on the first 15 million gallons
Compliance with each threshold requires a comprehensive evaluation of renewables and the baseline produced a year.
gasoline or diesel. The assessment includes aggregate GHG emissions, direct and indirect, including land
use changes, all stages of fuel and feedstock production, distribution and use by the ultimate consumer. Advanced bioethanol
“The term ‘lifecycle greenhouse gas emissions’ means the aggregate quantity of greenhouse gas emissions To meet the RFS, the US government has requested that 16 billion gallons (60 billion litres) are produced
(including direct emissions and significant indirect emissions such as significant emissions from land use from cellulosic biofuels (general biomass) and another 4 billion gallons (15 billion litres) from advanced
changes), as determined by the Administrator, related to the full fuel lifecycle, including all stages of fuel and biofuels (from other parts of the corn starch plant). Out of the 4 billion gallons required from corn residues,
feedstock production and distribution, from feedstock generation or extraction through the distribution and 0.95 were produced in 2010. The feedstocks identified as appropriate to meet advanced biofuel targets
delivery and use of the finished fuel to the ultimate consumer, where the mass values for all greenhouse include switch grass, crop residues, wood biomass and more. Table 7 shows the USDA estimates of how
gases are adjusted to account for their relative global warming potential.” ESIA 2007, EPA website 2010. these obligations might be met.
Table 7: USDA feedstock estimates to RFS US biofuel requirements
EPA expects the following feedstocks and the associated number of gallons by 2022: This leaves some room for the US RFS targets to be met with imported bioethanol, particularly from Brazil,
but the import tariff at $0.54c a gallon is currently keeping most Brazilian bioethanol out. Nonetheless,
Switchgrass (perennial grass): 7.9 bg
Brazilian ethanol is eligible under the US legislation to meet ‘advanced biofuel’ status because the
Soy biodiesel and corn oil: 1.34 bg conversion rate of sugar to ethanol is highly efficient.
Crop residues (corn stover, includes bagasse): 5.5 bg
It is estimated that to meet the RFS requirements the US government will need to use agricultural cropland
Woody biomass (forestry residue – data does not include short-term woody crops): 0.1 bg and around 6.5% of the total 406.4 million acres of cropland reported in the 2007 census of agriculture.
Corn ethanol: 15.0 bg The ESIA states that feedstocks must come from agricultural land and planted trees/tree residue from land
Other (municipal solid waste (MSW)): 2.6 bg
cleared and cultivated prior to 19 December 2007.
Animal fats and yellow grease: 0.38 bg Biodiesel
Algae: 0.1 bg The RFS also has a requirement to produce no less than 1 billion gallons (3.8 billion litres) of biodiesel per
Imports: 2.2 bg
annum. The US biodiesel market is mainly based on soybean biodiesel but also corn oil. Historically, US
farmers were producing surpluses of soybean, which were estimated to be capable of producing circa 700
USDA estimates the following feedstocks and the associated gallons by 2022: (this count does not include tallow, MSW, algae)
million gallons of biodiesel per annum. Some of this biodiesel was even exported to Europe, but this has
Dedicated energy crops: perennial grasses, energy cane, biomass sorghum: 13.4 bg now been stopped by EU ‘anti dumping’ rules. The USDA states the government produced 650 million
Oilseeds (soy, canola): 0.5 bg gallons of biodiesel from 173 plants in 2010. A further 29 companies have reported that plants are under
Crop residues (corn stover, straw): 4.3 bg
construction. This combined capacity, if it continues, will provide another 427.8 million gallons per year of
biodiesel production, thus fulfilling the 1 billion gallons target in the RFS 2.
Woody biomass (logging residues only): 2.8 bg
Com starch ethanol: 15.0 bg
However, the US biodiesel industry is currently in financial limbo. The US $1/gallon biodiesel tax credit and
mixture tax credit and $0.5/gallon alternative mixture tax credit expired at the end of December 2009,
leaving the industry without government support (Michigan State University (MSU) 2010). Due to conflicting
However, the USDA (2010) estimates that to meet cellulosic and advanced biofuel requirements will mean
industry views and budgetary concerns, the intention to extend the credit has not manifested. With the
building 527 bio-refineries at a cost of $168 billion. The assumptions here include $8 per gallon to build a
elections in November, a number of experts in the industry believe that the US biodiesel industry is dead
cellulosic plant, a capacity of 40 million gallons a year for each plant built and certain energy yields per acre.
already, that the credits will not be extended this year and that by the time they might be it will be too late.
The capital expenditure costs are high, the operating costs are likely to be uneconomic, and the time frame
However, Washington analysis remains more optimistic that a credit will be extended and will be passed
to build these plants likely to be 3-4 years. Novozymes is targeting the US markets with its corn stover
retroactively at the end of this year. In the meantime, most biodiesel plants, circa 160 out of 180, are
enzyme technology but under the RFS more biofuel is to come from other cellulosic crops than from corn
apparently on hold waiting for this retrospective $1/gallon tax credit. Some believe it will not happen and that
stover. We believe cellulosic biofuel is even less commercially ready to meet RFS targets, so we view these
the US will import Brazilian, Argentinean, and Asian biodiesel which is at a much lower cost and there is no
with some caution. The government is supporting the production of cellulosic biofuel with the credit for
tariff so biodiesel can be cheaply imported.
production of cellulosic biofuel at a maximum of $1.01 per gallon produced (this is reduced by the amount of
VEETC to avoid duplication when applicable) also due to expire end of December 2012. In addition, in the
first year of operation, cellulosic plants may take a 50 percent additional depreciation deduction before
01 January 2013.
2 INVESTMENT THEMES
European market overview
EU regulatory policy The EC produced a press release in June 2010 on a system for certifying biofuels (Europa 2010). The
There are three directives which are relevant to the production of biofuels within the EU: system introduces a Sustainable Biofuel Certificate, which will guarantee that all biofuels sold under the
label are sustainable and produced under the criteria set in the renewable energy directive. Natural forests,
– 2003 biofuels directive, which set an indicative target of 5.75% biofuel usage of all used traffic fuel by each protected areas, wetlands, peatlands, and palm oil from deforestation are explicitly ruled out. In addition,
member state by 18 May 2010, but the date was changed to 2011 under the 2009 renewable energy directive only biofuels with 35% GHG savings compared to petrol and diesel will be permitted. This will rise to 50%
– 2003 energy taxation directive, which allows member states to reduce or exempt excise duties on biofuels in 2017 and include methane and di-nitrous oxide.
– 2009 renewable energy directive, repeals 2003 biofuels directive, (with the exception of the target of The majority of biofuels are produced in the EU, but in 2007 we know that 26% of biodiesel was imported
5.75% by 2011 for each member state. and 31% of bioethanol was imported from Brazil and the US (Europa 2010). It’s also noted in Europa, that
the 10% target for the share of energy from renewable sources to be in all forms of transport will require
It sets a mandatory target for 20% renewables in the EU’s gross final energy consumption by 2020. So 20%
some 2-5 million hectares of land which the EC believes is available on land previously used for arable
is to be the average over all Member States.
production but not currently (possibly implying set aside land). Currently, if the oil and gas companies
It sets national targets which vary from 10% in Malta to 49% in Sweden. cannot meet blending requirements it is cheaper for them to pay the penalties than find biofuels and
meet blending requirements.
In addition, the share of energy from renewable sources in all forms of transport has to be at least 10% of
the final consumption of energy in transport by 2020. These targets could partially be met by electrical forms EU market overview
of transport, however. Today, it is unlikely that the 2003 biofuels directive target will be met, expectations are for around 4.2%
2009 fuel quality directive sets a mandatory 6% decarbonised target for suppliers of transport fuels per unit (up on the 3.4% on average in 2008), see Table 8. So by 2010, Germany, France and probably Austria will
of energy sold compared to a baseline in 2010. This is likely to be set as 86g CO2/MJ. It also increases the have met the targets but the majority of EU states will have fallen far short of it (International Institute of
maximum share of ethanol in blended petrol fuels to 10%. Sustainable Development (IISD) 2010). The EU can start infringement procedures if it wishes for EU member
states that do not comply with the renewable energy directive but it is not anticipated that it will do so. In the
The renewable energy directive also states sustainability criteria that biofuels need to comply with or they short term, it is expected that some production increase will occur due to investments which have already
may not be taken into account when the European Commission (EC) calculates national targets, EU targets commenced and will come on line, however, the EC is not committed to the European target, more
or be eligible for financial support. The minimum GHG life cycle reduction for biofuels is 35% as of 2009, embarrassed by it, and financial incentivisation is falling.
50% by 2017 and 60% for new installations from 2017 onwards. For plants in operation in January 2008, the
GHG requirement will take effect in April 2013. There is also a bonus of 29g CO2/MJ for biofuels from
degraded/contaminated land. Qualifying biofuels are not allowed to be produced from land of high
biodiversity value, including wetlands, forested areas, peatland, and some reporting is required to show that
biofuel production is done in a responsible way. Currently, the incorporation of the ‘indirect land use impact’
on carbon emissions (from deforestation, or a shift of land from food to fuel production) is still under debate.
It does not figure yet in the formula to calculate GHG life cycle emissions. However, the GHG methodology is
proposed for review by the end of 2010 to take into account land use change and it is expected that this
new methodology will need to be adopted by member states by 2011.
Table 8: Member state biofuel production 2003 and 2008
Member State Biofuels share in 2003 Biofuels share in 2008 producing up to circa 10 million tonnes of biodiesel annually. These plants are mainly located in Germany,
EU25 EU27 Italy, Austria, France and Sweden. However, compared to bioethanol, import protection tariffs on biodiesel
Germany 1.2 6.0 are generally quite low, whilst the tariffs on bioethanol are high. The EU maintains a tariff of around 45% ad
France 0.7 5.7 valorem on ethanol whilst tariffs on biodiesel are around 0-5% (IISD 2010).
Austria 0.1 5.5
Sweden 1.3 5.0 According to a Citigroup analyst, global production of palm oil is circa 45 million tonnes per annum but only 2.5 million
Lithuania 0.0 4.3 tonnes (6.25%) of global production is certified palm oil, produced every year. Only a few companies, now members
Poland 0.5 3.7 of the Sustainable RoundTable for Palm Oil (SRPO), are certified as sustainable producers of palm oil in Asia.
Hungary 0.0 3.5 Table 9 below, shows which companies these are, the number of mills and palm oil produced in million tonnes.
The Netherlands 0.0 3.3 In addition, about 4-5% of biofuel in the EU is produced from palm oil. This is about 1% of palm oil
Slovakia 0.1 2.7 production. More than 95% of global palm oil is used by the food industry (Europa 2010). Malaysia and
Romania – 2.35 Indonesia produce 90% of palm oil globally, with a little coming from Papua New Guinea. So at this stage,
Finland 0.1 2.1 the end market for palm oil, for biofuel, is not considered significant.
Luxembourg 0.0 2.0
Table 9: Certified palm oil production, under (SRPO)
Portugal 0.0 2.0
Spain 0.4 2.0 Member Country Mills Palm oil Palm kernels
(million tonnes) (million tonnes)
UK 0.0 2.0
United Plantations Bhd Malaysia 6 200,456 53,608
Ireland 0.0 1.6
New Britain Palm Oil Ltd PNG 5 277,524 60,503
Cyprus 0.0 1.3
Sime Darby Malaysia 13 435,312 107,654
Czech Republic 1.1 1.3
Kulim (Malaysia) Berhad Malaysia 3 88,914 24,943
Slovenia 0.0 1.2
Wilmar International – PPB Oil Palms Berhad Malaysia 5 227,450 49,892
Belgium 0.0 1.1
Wilmar International – PT Mustika Sembuluh Indonesia 1 81,350 16,640
Greece 0.0 1.0
PT Musim Mas Indonesia 2 143,459 34,047
Estonia 0.0 0.6
PT PP London Sumatra Indonesia Tbk Indonesia 4 169,480 30,017
Italy 0.5 0.6
IOI Group Malaysia 3 155,447 36,234
Malta 0.0 0.44
SIPEF: HOPL PNG, Indonesia 4 173,168 20,447
Denmark 0.0 0.23
Cargill: PT Hindoli ,PT Hindoli SS Indonesia 4 186,892 42,097
Bulgaria – 0.22
Kuala Lumpur Kepong Berhad: KDC Malaysia 2 92,000 22,000
Latvia 0.2 0.2
EU 0.5% 3.4% JC Chang Group – Carotino Sdn Bhd Malaysia 1 30,300 7,700
PT Bakrie Sumatera Tbk Indonesia 1 36,438 7,436
According to the European Bioethanol Association (EBA), 80% of all biofuel in the EU is biodiesel. This is PT Agrowiratama Indonesia 1 46,635 11,635
primarily made from rapeseed oil within the EU (the EBA will not release current feedstock details). Biodiesel PT Berkat Sawit Sejati Indonesia 1 54,166 12,584
international trade has been inhibited as a result of trade and sustainability issues in palm oil, policies such Felda Malaysia 2 102,884 24,391
as B-99 anti-dumping duties (EU commission number 193/2009 imposed a provisional anti-dumping duty
TOTAL 2,398,991 561,828
on imports of biodiesel originating in the US) stopped the rise in biodiesel imports in the US. Instead, there
has been increased national production of biodiesel. Today, there are approximately 120 plants in the EU Source: http://www.rspo.org/?q=node/520
2 INVESTMENT THEMES
The EU production capacity is steadily increasing. In 2008, the installed capacity amounted to 6.1 billion do Sul in the southwest, and São Paulo in the southeast, where cane fields are taking over pastureland.
litres, while production capacity under construction was 2.4 billion litres. The top 4 EU producers of ethanol Sugarcane is not grown in areas where the rainforest remains. There is no need as there are extremely large
are France, Germany, Spain and Poland, followed by Sweden and the UK. Total imports of bioethanol (fuel areas of unused land elsewhere. In addition, the climate, seasonal changes and composition of the soil is
and non-fuel) are estimated to have reached 1.9 billion litres in 2008, increasing by 400 million compared to also not optimal for sugarcane growth in rainforested areas (IPS News 2010). To expand sugarcane, the
2007. Thereof, between 1.4 and 1.5 billion litres came from Brazil only. Approximately 50% of total imports Brazilian government has conducted a study of areas that are suited for production and has included credit
have been used for the fuel sector (circa 1 billion litres). This equals 39% of total EU production. In 2008 the lines to favour underused areas or degraded pastureland (OECD/IEA 2010).
preferred raw material was grain and, in particular, wheat and maize. Only 2% of total EU cereal production
In terms of biodiesel, the government launched the National Program of Biodiesel Production and Use in
was destined for bioethanol production in 2008. (EBA 2010).
2004, to blend 2% biodiesel by volume into diesel fuel on a mandatory basis in 2008. This blending mandate
Whilst the agricultural commission wants to keep bioethanol out of the EU, Shell, BP and Petrobas have has now increased to 5% in 2010 and will require about 2.3 billion litres of production per annum. Large
recently set up joint ventures with Brazilian/European operators and now have access to both EU lobbying scale soya bean production fuels the majority of biodiesel production and is being grown in the central west
abilities and the World Trade Organisation (WTO) to potentially support importation. It is expected that parts of Brazil. Petrobras aims to produce 1 billion litres of biodiesel B2 per year by 2012, rising to 2.4 billion
excise duties on importation of bioethanol will have to be lowered to avoid conflict with WTO rules. by 2013. Around 92% of the plants authorised for commercialisation have the ‘Social Fuel’ seal, which is a
Therefore, in the medium term, there could be an influx of Brazilian bioethanol into the EU market. certificate given to the companies that purchase a minimum amount of feedstock from family farmers and
that have social and environmental concerns. There are reductions in excise duty along the biodiesel chain
Brazilian market overview according to the type of feedstock used, according to the geographic region of production and according to
the supplier (whether family farm or agribusiness).
2G biofuels are currently not competitive so the government has not introduced policies to support this
About 45% of Brazil’s total primary energy supply is derived from renewable sources, mainly biomass (31%)
as yet. Despite there being no commercialisation of this industry at present, there is an expectation that
and hydropower (13.8%). In Brazil, bioethanol has been used for decades in fuel consumption. Brazil
cellulosic ethanol will be in production by 2020.
launched the ‘National Proalcool’ programme in the 1970s, which amongst other things promoted the use of
hydrous (E100) fuel and vehicles that could use this. In the 1990s, Law 8.723/93 was introduced and made
mandatory an anhydrous bioethanol blend that could range from 20-25%. As a result, most Brazilian Chinese market overview
bioethanol is used internally to supply the renewable transport market - nearly eight out of every ten cars China is the world’s third largest bioethanol producer after the US and Brazil (European Biofuels Technology
sold in Brazil are flex fuel or bioethanol. To date, Brazil has replaced circa 25% of its gasoline fuel with sugar Platforms 2010). The country began producing bioethanol in 2000 using surplus grain. According to the
cane based fuel grown on <5% of its land area. IEA/OECD (2010) China produced 1.51 billion litres of bioethanol in 2008 and 362.8 million litres of biodiesel.
Brazil is the second largest producer of bioethanol in the world, after the US. Currently there are 423 According to UBS (2010), the Chinese government has a development plan for renewable energy which
bioethanol plants, and it is estimated that annual production will reach around 43.4 billion litres in 2013 nominates biomass as a priority sector and sets targets of 2 million tonnes by 2010 and 10 million tonnes by
(International Energy Agency (IEA), 2009b (Organisation for Economic Cooperation and Development 2020 for bioethanol (making circa E10 (a mixture of 10% ethanol and 90% gasoline) usage prevalent) and
(OECD)/IEA 2010). It is also the world’s largest exporter of fuel-grade bioethanol, exporting in 2008 circa 0.2 million tonnes for 2010 and 2 million tonnes by 2020 for biodiesel. However, these figures have
5% of total production at 27 billion litres). Today, the primary destinations for ethanol exports are the US apparently been changed many times and are currently under reconsideration, yet again (GLG).
(including exports under the Caribbean Basin Initiative and Central American Free Trade Agreement) and China originally supported the set up of five state-certified fuel ethanol production plants which run today.
Europe, (OECD/IEA 2010). However, China does not view biofuels as having large scale commercial application in the near future and
Brazilian bioethanol does meet US advanced bioethanol sustainability criteria. In 2010, the US EPA federal support is wavering; nuclear, hydro, wind and solar are perceived as far more attractive. This is
designated Brazilian sugarcane ethanol as an advanced biofuel due to its 61% reduction of total life cycle because there are some issues in the industry which present real challenges, not least that fact that China
GHG emissions, including direct indirect land use change emissions. In the EU, it is likely that Brazilian has such a large population to feed and the use of corn starch to make biofuel was stopped in 2006 in light
bioethanol will also be accepted as compliant with 2G bioethanol sustainability criteria, when the directive of food security concerns. This means that currently, bioethanol must be made from cellulosic waste, such
comes into force in December 2010. Oil companies, such as Shell and BP have few concerns in this respect as corn waste and through 2G processes. Currently, gasification is preferred locally to produce electricity.
and have made large investments in Brazilian bioethanol companies with a view to importing more Brazilian The government sees issues with the large scale collection of corn or biomass remains from crop harvest
bioethanol into the EU. According to Brazil’s Ministry of Agriculture, sugarcane plantations are permitted only (i.e. feedstock stock cost), transport logistics/costs and also production costs. This means that the
on eight percent of Brazilian territory. Most sugar cane is grown in the Goias in central Brazil, Mato Grosso economics are not viewed as attractive.
There is some interest in non food crops from the private sector, and China National Cereals, Oils and – For those companies which are above the market capitalisation, these are normally 1G biofuel
Foodstuffs Corp has invested 50 million Yuan (USD $6.5 million) to build a cellulosic ethanol pilot plant technology companies and consideration needs to be given to sustainability criteria. For example, in the
of a capacity of 5,000 tonnes. A further plant with a production capacity of 10,000 tonnes is planned. EU, the processes for the certification of the sustainability of biofuels are still emerging and becoming
http://www.biofuels.apec.org/ In addition, Novozymes, Sinopec and COFCO are working together to set up more robust, so companies need to be considered on a case by case basis.
a pilot plant with a capacity of 500 tonnes per year. The aim is to develop a commercially viable 2G ethanol – Large capitalisation companies invested in various stages of the biofuel value chain generally have a very
plant by 2010. small exposure to this industry, because it is such a niche market (e.g. oil and gas companies and specialist
chemical or engineering companies). So investment on a biofuel rationale alone is difficult to justify.
Indian market overview Therefore, the only companies that we believe are potentially investible are those that exceed the market
In February 2009, India and the US exchanged a memorandum of cooperation on biofuels development, capitalisation threshold (i.e. are sufficiently large capitalisation) and which have material exposure to the biofuel
covering the production of utilisation, distribution and marketing of biofuels in India. Research and industry. To date, this only appears to be a few Brazilian sugar cane companies and pure play biofuel companies.
development on cellulosic ethanol is increasing with PRAJ Industries developing technology for cellulosic
ethanol. Reliance Life Sciences is also active in developing biodiesel (from Jatropha and other non-food References
oil seed crops), ethanol (from cellulosic biomass) and biobutanol. In February 2009, India and the US
exchanged a memorandum for cooperation on biofuels development, covering the production, Asia Pacific Economic Cooperation (2010) Website on Asian use of biofuels. http://www.biofuels.apec.org/
utilisation, distribution and marketing of biofuels in India. Petrobras (2010) Brazil’s Biofuels. 50 questions and answers. http://www2.petrobras.com.br/petrobras/ingles/pdf/
The future of biofuels Biofuels Barometer (2009) http://www.eurobserv-er.org/pdf/baro198.pdf
Cheuvreux (2010) Biofuel Sector – Green Tech Research, March 2010.
Projections for biofuel production to 2020 are mainly an extension of today’s production patterns. The IEA CICERO (2008) Centre for International Climate and Environmental Research. Biofuels Market Development: International
estimates that European biofuel consumption in 2020 will be 27 billion litres and US consumption 113 billion Trend and China’s Challenges. http://www.cicero.uio.no/fulltext/index_e.aspx?id=6868
litres by the same year, including a 40 billion litre production of 2G biofuels. In Brazil the ethanol production Environmental Protection Agency (EPA 2010) EPA Finalises Regulations for the National Renewable Fuel Standard Program
foreseen in 2020 is expected to be 32.40 billion litres for the domestic Brazilian market and 18.6 billion litres for 2010 and Beyond. http://www.epa.gov/otaq/renewablefuels/420f10007.htm
for export (OECD/ IEA 2010). Similarly, the OECD and United Nations Food and Agriculture Organisation European Biofuels Technology Platform (2010) http://www.biofuelstp.eu/legislation.html
have projected that global ethanol production will double between 2007 and 2017 reaching 125 billion litres. Europa (2010) European Commission sets up a system for certifying Biofuels. http://europa.eu/rapid/pressReleasesAction.
Biodiesel is expected to increase from 11 billion to 24 billion litres over the same period (assuming no or do?reference=MEMO/10/247&format=HTML&aged=0&language=en&guiLanguage=en
little change in government support through tax incentives and blending targets). On a sales basis, Pike European Bioethanol Fuel Association (2010) http://www.ebio.org/EUmarket.php
Research estimated that biofuel markets (combined) could reach $247 billion in sales by 2020 up from
European Biodiesel Board (2010) http://www.ebb-eu.org/
just $76 billion in 2010. This report is optimistic though, anticipating the commercialisation of waste oils,
Mississippi State University, Sustainable Energy Research Centre (2010) Summary of Federal Biofuel Incentives.
Jatropha based fuels and algae-based biodiesel by this time.
IISD (2010) Biofuels but at what cost? Government support for bioethanol and biodisesel in the European Union. International
However, the World Energy Outlook 2009 and the Blue Map Scenario of Energy Technology Perspectives Institute for Sustainable Development. http://www.globalsubsidies.org/files/assets/bf_eunion_2010update.pdf
2008, produced by the IEA, show a radically different approach by 2050 with biofuels providing 26% of total IPS News (2010) Tierramerica Brazilian Biofuels run into EU Obstacles. August 2010. http://www.tierramerica.info/nota.
transportation fuel in 2050 and second generation biofuels accounting for 90% of all biofuel. Furthermore, php?lang=eng&idnews=3471
more than half of the 2G biofuel production in the blue map scenario is projected to occur in non-OECD Roundable Table for Sustainable Palm Oil (2010) http://www.rspo.org/
countries, with China and India accounting for 19% of the total production. Currently the US is the largest USDA Biofuels Strategic Production Report 23 June 2010, A USDA Regional Roadmap to Meeting the Biofuels Goals of
global producer of 1G biofuel. the Renewable Fuels Standard by 2022
UNEP (2009) Towards Sustainable Production and use of resource: Biofuels. http://www.unep.fr/scp/rpanel/pdf/
Conclusions and companies assessing_biofuels_full_report.pdf
Most companies in the biofuel industry at present are not investible. This is due to a number of specific reasons.
– Pure play biofuel companies investing in 1G and 2G are typically private or under the USD 200million The views and opinions contained herein are those of Karen Shaw and may not necessarily
market capitalisation and below the investment threshold of our strategy. represent views expressed or reflected in other Schroders communications, strategies or funds.
2 INVESTMENT THEMES
2.7 Currency Unions: Breaking up is not so hard
Alan Brown Group Chief Investment Officer
There are entrenched views on both sides as to whether the euro can survive in its current form or not. On the The Irish Pound
continent, many regard speculation on the break up of the euro as Anglo Saxon shenanigans. And in the other
camp, blind faith in the euro’s stability is seen as a denial of economic realities. I want to step aside from that The Irish pound in history was originally linked to France (997) and then to sterling (in the 1180s at an
argument for a moment and instead want to consider what history can tell us about the dissolution of currency exchange rate of 13 Irish pounds = 12 pounds sterling) before it was replaced by British currency in 1826.
unions. First, there have been a surprising number of currency union dissolutions. Since the Second World War a A new Irish pound was created in 1928 but still linked to sterling at an exchange rate of 1:1.
total of 69 unions have come to an end while only 61 entities have remained in unions throughout the same period.1
It was the introduction of the European Monetary System that brought an end to the link with sterling as the
Admittedly, many of these were quite ‘odd’ currency unions that many of us probably were never aware of, Irish decided to join in while the British stayed out. Thereafter there were substantial exchange rate shifts
Iraq and the UK (1967) for example, but others were more relevant such as Ireland and the UK (1979) or the with the Irish pound ranging between 74 and 110 pence between its introduction in 1979 and its entry into
Czech and Slovak Republics (1993 and not included in the list of 69). I shall come back to the Irish and the euro in 1999. Yet, because the Irish already had their own notes and coins, breaking the link with sterling
Czechoslovakian examples later. presented relatively few difficulties. One of the biggest changes was that until 1986 Irish coins were the
same size and shape as British coins. After 1986, their shape and size changed, a necessary move given
Bordo and Jonung in their National Bureau of Economic Research paper2 note that “The causes of the the wide usage of vending machines at the time!
break-up of national monetary unions are foremost found in political developments. Political unity is the glue
that holds a monetary union together. Once it dissolves, it is most likely that the monetary union will
dissolve.” The euro has often been described as a political project. We would agree and, as long as the The Czech and Slovak Republics
political will exists, so will the euro. However, it is economic developments above all else that have the Perhaps the most relevant recent example of the dissolution of a currency union is the break up of
potential to undermine that political will. We should recall the Clinton/Bush presidential campaign of 1992, Czechoslovakia into the Czech and Slovak Republics in January 1993. The decision to form separate
“It’s the economy, stupid”! Republics was essentially a political one. The two governments agreed to retain the Czechoslovak koruna
as their common currency but that agreement lasted only 38 days (See Jiri Pehe’s article for a contemporary
Definition account of the split ).3 On 08 February the two governments decided to establish separate currencies citing
divergent economic developments as the reason.
First though we need a definition as there are all sorts of currency unions, national, multi-national or currency
board to name but three. National unions typically share the same physical currency (as was the case with With just a little bit of imagination you can cast the Czech Republic in the role of the EU and the Slovak
Czechoslovakia). Following Rose, for multi-national unions, I am only considering currency unions where the Republic as Greece. If you look at the economic starting conditions in the two Republics, it is not difficult to
exchange rate was fixed at 1:1 so that both currencies can be accepted for payment. Fixed exchange rates see where the strains were coming from. Productivity in Slovakia was 8% lower than in the Czech Republic
such as Hong Kong or currency boards do not qualify. Rose also ignores political dissolutions such as and GDP per capita was 24% lower, but wages only a marginal 2% lower.4
Czechoslovakia. We shall consider this later. Rose, in his Checking Out: Exits from Currency Unions paper,
finds that “countries leaving currency unions tend to be larger, richer and more democratic; they also tend to
experience somewhat higher inflation. Most strikingly, there is remarkably little macroeconomic volatility
around the time of the currency union dissolutions…”. As a generality that may be true, but the case of the
Czech and Slovak Republics paints a rather different picture.
Note that the euro is an interesting case as it is a multi-national currency union and yet participating 1 Andrew K. Rose, Checking Out: Exits from Currency Unions, MAS, April 2007
countries have the same central bank and share the same notes and coins as if in a national union. 2 Michael D. Bordo, Lars Jonung, The Future of EMU: What does the history of monetary unions tell us, NBER, working
paper 7365, September 1999
3 Jiri Pehe, The Czech-Slovak Currency Split, RFE/RL Research Report, Vol 2, no 10, 5 March 1993
4 Josef Tosovsky, Governor of the Czech National Bank, Managing Financial Turbulence: Czech Experience, 1998 39
There was an annual budget transfer from the Czech Republic of CSK25billion ($900 million). Unemployment was In the first decade of the euro’s existence domestic demand grew unchecked in Greece by about 37%
2.6% in the Czech Republic, 10.6% in Slovakia. 93% of all Foreign Direct Investment went to the Czech Republic. versus just 3% in Germany.5 In 2007 and 2008 Greece was running a current account deficit of nearly 15%
The Czech Republic had a sharply rising trade surplus with Slovakia. It doubled in one year to CSK16 billion. of GDP, albeit more recently the deficit has narrowed as a result of the severity of the economic downturn.
Then in January 1993, in just one month, Slovakia lost 60% of its $1billion foreign currency reserves making Relative to Germany, since joining the euro, Greece has lost 20% of its competitiveness (Ireland, Spain,
action almost inevitable. Italy and Portugal have all also lost competitiveness by between 18 and 33% 6 ).
While separate currencies were always likely, there was no plan to create them at the start of January. How While Greece attempts to go through its austerity programme, Germany, rather unhelpfully, is also intent on
did the two countries introduce two separate currencies so quickly? On 02 February the two governments balancing its budget over the next few years. In an integrated economy with policy coordination, Germany
voted to create two separate currencies on 08 February, just six days later! It may seem extraordinary but would be stimulating domestic demand to offset the contraction in Greece. Without pretty rapid economic
the actual currency split went through without any major hitches. 13,000 employees in the two countries growth in the eurozone, Greece’s adjustment programme looks positively Herculean. To stabilise its
stuck stickers on the large denomination notes. Individuals were encouraged to hold the minimum amount debt/GDP at 150% or so will likely require Greece to run a primary surplus of 8% of GDP.
of cash to facilitate the process.
Markets, which had blithely ignored these diverging trends for years, woke up with a vengeance in the
A common accounting unit was introduced, the koruna and the exchange rate between the two currencies second quarter. Official response to the burgeoning crisis was flat footed and ultimately required a €750
and the koruna was set at 1:1. All financial obligations prior to 08 February were fixed at the level of the billion bail out package and emergency loans from the ECB to put a lid on the crisis. But unlike BP’s well in
koruna/ECU exchange rate on 05 February. Obligations after 08 February were not at a fixed rate of the the Gulf, no one can be sure that the ‘lid’ will hold. Greece is making quite good progress to date, but the
koruna to the ECU and the exchange rates of the Czech and Slovak currencies to the koruna were fixed daily Euro crisis has already rolled on to engulf Ireland. The immediate major funding needs of Spain, Portugal,
by their central banks. Ireland and Greece look quite manageable, at least in the very near term, but the longer-term picture
remains extremely unclear. If contagion continues to spread, and particularly if it were to reach Spain,
Very quickly pressure grew for the Slovak government to devalue. While this was still under discussion and
well then all bets would be off...
the official rate was held at 1:1, de facto commercial banks devalued the Slovak koruna by nearly 20%
quoting the Czech koruna on the 17 February at 80.15 to the Slovak unit.
In an extraordinarily short period of time, economic realities and market pressures had taken control of the
political agenda. All the underlying structural weaknesses in the eurozone persist. A lack of political and economic integration
means that the region will remain vulnerable to external shocks. As a political project, the euro will survive in
its current form as long as the political will continues to be there. But no one can be sure that that ‘will’ will
The Euro and Germany and Greece hold. Will the Greeks be prepared to go through a decade of austerity as they rebalance their economy? If
It is not difficult to see parallels between the experience with the Czech and Slovak Republics and tensions Greece’s double-dip extends to Spain and Portugal, will their citizens accept their lot as the Irish appear to
within the euro area. Until recently, I had been arguing the possibility that we could flirt with deflation at the be doing? We live in democratic countries (thankfully). Is it possible that a party could come to power on the
beginning of this decade and inflation at the end of the decade. I now realise that we can flirt with inflation back of a restructuring mandate? Or alternatively, will Germany finally run out of patience and the will to bank
and deflation simultaneously. Double-dip is a reality in Greece with two negative quarters of growth roll the europroject?
year-to-date, and the economy shrinking by 2.8% in the first half in contrast to Germany growing by 2.7%.
The message of the 70 or so currency union exits of the last 65 years is that unbreakable marriages can
At the same time, in other parts of the world, notably China, inflation pressures are very real and have been
indeed get dissolved. Dire economic and political consequences need not follow, and indeed, to the extent
the recent driver behind policy moves.
that divorce recognises economic realities, adopting separate currencies can come as a relief to both parties.
The euro, while a multi-national currency union, has important features of a national currency union, notably
As I have said before, this tragedy (or pantomime) has many more acts to come. Stay alert.
all countries share the same central bank and the same currency. Yet the euro region lacks the high degree
of political and economic integration so fundamental to the stability of national currency unions. Instead of
The views and opinions contained herein are those of Alan Brown and may not necessarily
economic convergence, the euro area has experienced significant divergence.
represent views expressed or reflected in other Schroders communications, strategies or funds.
5 Economist Intelligence Unit
40 6 European Central Bank
2 INVESTMENT THEMES
2.8 Currency wars: The new weapon of choice
Keith Wade Chief Economist
This is not new: growth in Asia is based on an export-led model which requires a competitive currency.
Foreign exchange reserves have been rising for years. When the US and Europe were booming on the
– Global co-operation weakens as policy makers’ options to stimulate their flagging back of the credit boom this was not a problem as emerging countries fed the appetite of the western
economies dwindle. The exchange rate is becoming the ‘weapon’ of choice consumer. Today though it is a different story as consumers focus on repairing their balance sheets
rather than shopping.
– The need for export-led economies in Asia to maintain a competitive currency
In our view we can expect a gradual appreciation of the emerging currencies including the yuan.
holds little scope for any significant appreciation in the short-tem This will help the process of rebalancing the world economy, but is no panacea.
– With no resolution in sight we expect currency wars to continue and First, the benefit to the west will be limited in the short run as higher import prices will simply push up
protectionist pressure to build. the costs of electronics and clothing. Consumers will be poorer and inflation higher. Second, the long run
effects may be marginal as the competitiveness gap between China and economies like the UK is vast.
Talk of currency wars has moved to the top of the global agenda following comments from Dominique Remember that even after their recent strikes when Honda workers in China achieved a 25% pay rise, they
Strauss-Kahn. The head of the International Monetary Fund said that countries were no longer co-operating still earned less than $3,500 a year. Massive currency moves would be needed to close the cost gap even
as well as they had during the financial crisis and that there were signs that countries were trying to use their adjusting for productivity differences.
currencies ‘as a weapon’.1 If the UK and US are not going to start exporting cheap manufactures to Asia then the key to resolving the
The debate reflects concerns about growth in the US and Europe where activity remains sluggish and policy imbalances will be for China to open its markets in areas where there are clear needs which can be met by
makers are running out of options to stimulate their economies. With interest rates at rock bottom levels the western companies. For example, as Chinese incomes rise, the demand for services increases providing
exchange rate is seen as one tool that can be used to boost trade performance and take the economy scope for a range of western products. The recent 5-year plan indicated that China would be looking to
forward. In the UK, Bank of England Governor Mervyn King has made no secret of the fact that the pound emphasise consumer spending going forward as a source of growth.
can play an important part in rebalancing the economy. Unfortunately, such developments take time and in the meantime patience is running low: policymakers may
The problem is that the emerging countries, who are seen as being on the other side of the bargain by surprise us but watch for currency wars to continue and protectionist pressure to build.
collectively running a large trade surplus, are unwilling to allow their currencies to rise. Attention has
focussed on China which is accused of running a large trade surplus on the back of an undervalued yuan. The views and opinions contained herein are those of Keith Wade and may not necessarily
However, in recent meetings with the US and Europe, officials have indicated that they will continue to peg represent views expressed or reflected in other Schroders communications, strategies or funds.
the yuan to the US dollar and see little scope for a significant appreciation in their currency.
Consequently, intervention in the foreign exchange markets continues with China now holding foreign
exchange reserves of $2,250 billion, up 15% over the past year. The rise in reserves reflects both the trade
surplus and the investment capital which is currently pouring into the emerging markets. China is not alone;
arrangements to fix or link currencies to the dollar are common across Asia. Foreign exchange reserves
have soared across the region.
1 Source: Strauss Kahn, head of the International Monetary Fund, 05 October 2010. 41
2.9 The engines of global growth beyond 2010
Virginie Maisonneuve Head of Global and International Equities
Please note: This article was written in March 2009.
While equity markets have rallied strongly since the lows in March 2009, and economic conditions have improved The real question for us as investors is – which companies should we select in this environment to build
in a number of ways, things still aren’t quite what they used to be. For several reasons, the environment in which attractive medium to long-term portfolios?
companies and consumers face heading into 2010 is far different to that experienced prior to the crisis. Since
visibility around conditions through 2010 is still limited, investors will be tempted to focus on the short-term which Shifts in global competitiveness
may create some volatility. This volatility could, in turn, produce some excellent buying opportunities for those
investors who are willing to be patient and to extend their time horizon beyond the next six months. An important aspect of the crisis has been its dynamic impact on relative global competitiveness. Not only has
access to loans for most companies changed dramatically during the last 12 months, but the range of cost of
debt has also varied a lot from one company to the next. Sharp moves in currencies during the past year have
Volatility and 2010 also altered the competitive landscape, quite noticeably, in favour of the US and the dollar zone. Currencies
The governments of those countries in the ‘eye of the storm’ (i.e. the UK, USA and Europe) will be extremely wary such as the Australian dollar, the real, the euro and the yen have appreciated during the past 12 months.
of tightening policy too soon – the memory of Japan in the 1990s still looms large. As such, we believe that the
more likely risk is that governments will err on the side of keeping monetary policy too loose for too long. Focus on long-term structural growth trends
Over the coming months, markets are likely to remain focused on the effect of the withdrawal of quantitative In this environment of short-term volatility, investors must expand their time horizon in order to take advantage
easing in those countries most affected by the credit crisis. It may also be difficult for economists to judge of volatility to buy quality companies opportunistically and build strong long-term portfolios. We believe it is
the real shape of final demand until April 2010 or, possibly, later, when the base effect (year-on-year important to anticipate the impact that long-term trends can have on companies’ earnings in the future (and,
comparison) becomes less favourable. importantly, the sustainability of those earnings) in order to identify growth gaps i.e. pockets of growth that will
surprise the markets. This allows patient investors to buy targeted companies at attractive valuations.
The timing of the reversal in the quantitative easing (QE) measures will most likely have an impact on currencies
and we should expect volatility in this area over the course of 2010. Another potential source of volatility is the We believe that key trends such as climate change, demographics and ‘Supercycle’ will have a long lasting
lack of clarity (particularly in Europe) over the exact expectations of bank regulators with regards to capital impact on the environment companies operate in. Failing to comprehend how such trends will impact
rules (and timeframe) required beyond the general principle that ‘more capital is needed’. Although one companies’ ability to sustain earnings growth could be hazardous for stockpickers.
understands the argument for a soft approach given how many banks would still fail to pass requirements
such as Tier 11 ratio above 8%, this lack of clarity will still create volatility in the markets over the next few Climate change
months. Moreover, the various measures such as tax increases – either politically motivated or in an attempt to
Climate change is one of the most important challenges facing the world today and is presenting investors
fund large deficits – will have an impact on countries’ relative competitiveness, as will decisions on tax policies.
with an abundance of opportunities for the near, medium and long term. We see climate change-related
Another source of volatility for companies’ earnings will be linked to capital costs. Those will most likely issues as driving the next industrial revolution and believe they will provide a key source of innovation for the
increase in 2010 but the impact on company earnings will be more than ever linked to balance sheet quality. global economy. Despite the financial and economic crisis, governments around the world are recognising
the challenges of a growing global population with limited resources and the potential impact of sharp
In our view, any volatility in 2010 should be used to buy equities as the prospects over the medium term continue
climate changes on the current economic and social balance.
to be attractive. We also believe that the liquidity that has been injected by governments around the world in an
unprecedented synchronised manner during the past 12 months will continue to provide support to markets.
1 The Tier 1 ratio is one measure of a bank’s financial strength. Generally speaking, the ratio is defined as a bank’s core
42 equity capital to its risk weighted assets (which are composed in part of its loans to its customers).
2 INVESTMENT THEMES
As a result, measures attempting to tackle climate change have been given a significant boost as many of the Adaptation is also important. Even if the world’s governments and companies are successful in managing
economic stimulus plans unveiled by governments have included significant financial backing for green the global transition to a low carbon economy, approximately two degrees of global warming seems
initiatives. In China around 40% of the country’s four trillion yuan budget was earmarked for the development inevitable. This will have a negative impact on agricultural yields, potentially pushing up food prices globally.
of green technologies, such as solar and wind energy, and electric car production. The long term opportunities Droughts and flooding around the world are already reducing the amount of arable land globally. In our view,
for investors are numerous and can be split between companies that help to mitigate the impact of global given this trend, there is also an opportunity for companies aiding agricultural productivity: from those
climate change and those that can help the world to adapt to changes that are, by now, inevitable. supplying irrigation systems to those selling fertiliser and crop protection products.
Mitigation efforts include the use of alternative energy sources that produce lower carbon emissions than the more
traditional energy sources such as coal and oil. These include solar, wind, hydro electric, geothermal, nuclear and
natural gas. As such, we expect to see a long period of spending in the nuclear industry as ageing plants are Global demographics are shifting rapidly. Population growth over the next 40 years will be very rapid as the
replaced. There are also advances being made in carbon capture and sequestration (CSS) – a technology that world reaches nine billion people in 2050 from six billion currently. Powerful shifts in population growth,
effectively removes the carbon dioxide from fossil fuels, such as coal, and stores it safely underground (in an empty ageing and urbanisation provide compelling long-term opportunities for companies in a variety of industries.
gas well, for instance).
Importantly, the source of growth arising from the global consumer will shift dramatically as marginal income
Figure 1: IEA roadmap to a reduction in greenhouse gas emissions dynamics adjust in accordance with wealth and spending patterns at different stages of life.
CO emissions (Gt CO /yr) Overall, although the global population will grow dramatically, the world is getting older at a rate
unprecedented in its history. Europe, for example, is currently the world’s oldest continent with a median age
of almost 40. Overall population decline in Europe will begin in the next 10-15 years, but for the population of
Baseline Emissions 62 Gt
CCS (carbon capture) industry and transformation 9% working age, this could begin as early as 2010. Currently, dependency ratios are relatively similar across the
CCS (carbon capture) power generation 10% region, but will diverge over the next 40 years, ranging from under 38% in the UK to over 62% in Italy (see
50 Nuclear 6% figure 2 below).
40 Figure 2: IEA roadmap to a reduction in greenhouse gas emissions
Power generation efficiency and fuel switching 7%
End use fuel switching 11% Old-age dependency ratio
End use electricity efficiency 12%
End use fuel efficiency 24% 60
10 Blue Map Emissions 14 Gt 55
WE0 2007 450 ppm case ETP 2008 BLUE Map scenario 50
2005 2010 2015 2020 2025 2030 2035 2040 2045 2050 45
Source: International Energy Agency (IEA), 2008. 35
Energy efficiency is another key consideration and can involve anything from the development of building 30
insulation, to fluorescent lighting, to lighter materials for cars and planes. It can also involve companies that 25
provide, for example, video conferencing technology which are, even now, benefiting from firms cutting 20
travel costs and, at the same time, their carbon footprints. 2000 2005 2010 2015 2020 2025 2030 2035 2040 2045 2050
Greener, sustainable transport is another focus with leading carmakers such as Honda and Toyota already ––– France ––– Germany ––– Italy ––– Spain ––– Sweden ––– United Kingdom ––– Western Europe
marketing hybrid and electric cars. While these vehicles are still under development – and we expect many
advances to be made over the coming years – we are currently finding good opportunities with the Source: UN World Population Prospects, 2008 Revision.
manufacturers of the batteries used to power them.
In the absence of major fiscal reform, the impact of ageing on government budgets will be substantial, ‘Supercycle’
especially in terms of the cost of public pensions, health and long-term care. It also provides opportunities
for investors. For instance, as the population ages, the types of goods and services in demand will change: We define the impact of the growth in large emerging market economies on the rest of the world
older people will spend more of their income on food and, as they become more health-conscious, as ‘Supercycle’.
nutritious and local produce will win out over fast food or eating out. Evidence suggests that older people
As these emerging economies grow stronger (representing over 50% of the global economy on a
also tend to spend less on clothing, furniture and entertainment, but more on home help, books and utilities.
purchasing power parity basis) their industrialisation, urbanisation, and thirst for oil and resources are
While this section of the population spends less on public transport, and buys fewer cars, it is the main
increasingly impacting global prices as they become the key originators of marginal demand.
consumer of package holidays and cruises. It perhaps goes without saying that healthcare is the net
beneficiary of an ageing population, while education loses out. As some of these large emerging market economies have reached a ‘sweet spot’ in their development
where they can increasingly rely on domestic demand to support their economic growth and a stronger
Our proprietary research suggests that spending on healthcare will grow twice as fast as most other areas
middle class develops, the long-term demand for resources, energy and key industrial products
from 2010 and will be particularly concentrated in areas such as dialysis, orthopaedics and treatments for
(particularly linked to urbanisation and industrialisation) will continue.
agerelated disease (heart disease, strokes, dementia etc.). As governments attempt to contain healthcare
costs with stricter regulation, demand will shift towards generic drugs. The financial sector could also stand Today China, India and the US together account for 35% of the world’s urban population; by 2050 almost
to benefit from this trend as public pension reforms will necessitate and encourage private saving for 30% will live in India and China alone. The number of people moving to cities in China alone will be more
retirement. Life insurers, asset managers and financial advisers are all well-placed – especially those than the entire population of the US today. By 2025 three of the world’s most populous cities will be in India,
targeting older customers by offering annuities and equity release products. with over 70 million inhabitants between them.
A large portion of the growth in the population worldwide over the next 40 years will be concentrated in Chart 3: Strong demand growth forecast across the energy complex
some emerging markets economies and by 2050, over 85% of the world’s population will live in emerging
markets. China and India are, in particular, set to experience dramatic levels of population growth over the
next 40 years – a quarter of the growth of the entire world. However, how both of these key countries arrive 800
at these levels – and the opportunities for investors – differs significantly. 700
India’s population is forecast to continue to grow over the next 40 years, albeit at a gradually slowing rate. By 600
contrast, it is estimated that China’s population growth will peak in 2030 and gently decline thereafter. We predict
that healthcare will be the fastest growing sector in China as household spending on health by people over the
age of 60 is twice that of consumers in their 20s and 30s. Meanwhile, India will enjoy a demographic boost during 400
this period as a large group enters the most economically active period of their lives. This will have a profound
effect on discretionary spending in India, which is generally unappreciated by today’s investors.
Rapid urbanisation in some of those rapidly growing countries creates clear challenges and opportunities for
governments and investors respectively – although these may differ from country to country. Infrastructure 100
spending will necessarily increase to facilitate the expansion of transport networks and housing. Between 0
now and 2025, China will pave up to five billion square metres of road and add around 28,000km of metro 2004 2010 2015 2020 2025 2030
rail. To provide homes and offices for its new 350 million urban inhabitants, China will build almost 40 billion
square metres of floor space, adding a number of skyscrapers equivalent to ten New York cities.
Clearly, the way to benefit from such demographic trends is to find companies which have identified the Source: Energy Information Administration (EIA) Data, for illustration only. As at 30 September 2009.
source of marginal income growth and can cater to them. Those companies can equally be based in *British Thermal Unit.
emerging or developed markets. As global brands become more well known and more affordable in
emerging markets economies, some global companies are benefitting very strongly.
2 INVESTMENT THEMES
Chart 4: Per capita consumption of energy in developing markets only set to rise
bbl/year per capita
India China Brazil Russia UK Japan USA
Source: Barclays Capital, for illustration only. As at 30 September 2009.
Understanding how companies’ earnings can be affected by the challenges and opportunities presented
by long-term structural trends is essential when analysing their operating environment. Awareness of the
potential long-term implications of the way our world is evolving helps us to distinguish strong leadership
and vision in management teams, and to find those companies that are best positioned to deliver strong
earnings growth in the future. 2010 will most likely be a volatile year as a third phase of the crisis unfolds.
In this environment stock pickers will benefit from identifying businesses with positive earnings growth and
attractive valuations that are leveraging from pockets of opportunities across the globe, in emerging and
The views and opinions contained herein are those of Virginie Maisonneuve, and may not
necessarily represent views expressed or reflected in other Schroders communications,
strategies or funds.
MODERN INVESTMENT APPROACHES
3.1 A strategic approach to investing: An alternative to passive 48
3.2 Can investors rely on equities for long term growth? 53
3.3 Diversification in times of crisis 55
3.4 The power of technical analysis 60
3.5 The role of equity investing in a modern portfolio 64
3.1 A strategic approach to investing:
An alternative to passive
John Marsland Fund Manager, QEP Global Equities
Comprehensive investment solutions
We do not believe that investors must necessarily choose between diversified passive or concentrated There is also little diversification at a regional level. Investment opportunities are limited to those stocks listed
active approaches. Rather, we believe that it is possible to be both ‘diversified’ and ‘active’– maximising the within the 23 developed countries that constitute the index. What is more, almost half of the total regional
potential return opportunity without taking undue stock-specific risk, while also avoiding the pitfalls of exposure lies in US equities, dramatically overstating the importance of the US market compared with the
allocating to stocks simply to satisfy index constraints. rest of the world.
Schroders QEP team offers a suite of complementary strategies that can each hold over 400 stocks at any In addition, stock weightings in market cap-weighted indices are a function of the company’s past success,
one time. This level of stock diversification could lead to the assumption that these are effectively ‘passive’ with index weightings being skewed towards current market leaders.
strategies - in fact, they are anything but. Indeed, the number of stocks in our strategies belies their ‘active’
That is to say, a handful of so called ‘mega-cap’ stocks with larger market capitalisations command a
nature, targeting returns in excess of the index while also limiting potential risk exposure.
greater proportion of the index in percentage terms.
Significant opportunities exist outside of the index MSCI World Exposure1
Indices can restrict your investment universe
For passive investors, market cap-weighted tracker funds can provide a cost-effective investment option
which employs a systematic, repeatable and transparent process and offers a level of diversification.
However, this notwithstanding, we would argue that adhering to index constraints may limit the potential
return opportunity for investors and can lead to the inefficient allocation of funds to satisfy index constraints. United States 49.2%
Which begs the question, ‘is passive investing the only option for investors seeking a tried and tested, Continental Europe 23.9%
systematic approach, or could there be an alternative that offers a similar level of transparency but with Japan 10.3%
higher repeatable returns, while still limiting risk exposure?’ United Kingdom 9.7%
Typical indices, such as the MSCI World Index, predominantly cover large cap stocks from developed Pacific ex Japan 2.0%
markets. By adhering to index constraints, investors forgo significant opportunities to invest in attractive
stocks within the emerging markets, and among small- and mid-cap names. To put this into perspective,
we believe there is a global investible universe of up to 15,000 stocks compared with only 1,646 stocks1 in
MSCI World Index.
48 1 Source: Schroders, Factset. MSCI World Index constituents at 16 April 2010.
3 MODERN INVESTMENT APPROACHES
For market cap-weighted tracker funds, these stocks crowd out other investment opportunities as too Technology/Telecoms bubble of 1999/2000 Japanese stock market bubble of the late 1980s
much of the portfolio gets allocated to them by virtue of their size. If we take the UK market, for example, % weight % weight
the largest stock constitutes around 9% of the index and the top three stocks by market cap amount to 40% 50%
over 20% of the index total, as illustrated below:
36% of MSCI World Index is 44% of MSCI World Index is
% market cap comprised of tech stocks here comprised of Japanese stocks here
94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 87 89 91 93 95 97 99 01 03 05 07
MSCI World MSCI World
Source: Schroders, MSCI.
Largest stock 2nd and 3rd
Combating inefficiency through ‘smarter’ portfolio construction
Source: Schroders, Factset. MSCI All Country World Index constituents as at 24 August 2010.
We have all too recently seen the consequences of this ‘concentration risk’ in practice. Remember the ‘tech Market-cap weighting follows momentum, encouraging investors to buy high and sell low – effectively
bubble’ of 1999/2000, where 36% of the MSCI World Index was comprised of TMT (telecom, media and backing the ‘winners’ of yesterday and the ‘losers’ of tomorrow. Put simply, we believe these index-relative
technology) stocks; or the Japanese stock market bubble of the late 1980s. Japan’s weighting in the MSCI constraints can cost investors returns. Conversely, removing constraints increases the breadth of investment
World Index reached an astounding 44% in 1988. Astonishingly, in hindsight, Japan was at that time only a opportunity and can greatly increase the chance of benefiting from the skill of a manager. By introducing a
15% share of the GDP of the MSCI World constituent2. When this bubble deflated it was at enormous cost rebalancing mechanism and trading against these inefficiency drivers, investors can reduce this ‘mega-cap’
to most global funds, since even those managers who had a negative view on Japan could not avoid a drag. This rebalancing ‘anchor’ can be anything measurable that is not price sensitive, such as equal
substantial weighting. The unwinding of this overvaluation took 15 years and cost managers 2.6% per year3. weighting, but is most likely better if linked to a return driver like Value or Quality.
2 PPP (purchasing power parity) adjusted weight sourced from IMF.
3 The annualised difference between the MSCI World and the MSCI Kokusai from1988 to 2006 was 2.58% p.a. 49
Can active managers add value? Market breadth and active performance Market breadth and median active performance
Annual excess returns
Market breadth helps to determine the opportunity 5% 2%
Global equity markets offer more opportunities for active managers in some years than others. We believe 2
R = 0.24
that the so-called breadth of opportunity, measured simply by the percentage of stocks outperforming the 1.2%
market index, is related to the ability of an active manager to deliver outperformance in any particular period. 3% 1%
Percentage of stocks outperforming the index 0.4%
Higher active opportunity
20% 30% 40% 50% 60% 70% Low Average High All
Breadth (% of companies outperforming MSCI World) Breadth Breadth Breadth Periods
Source: Schroders, Mercer MPA database. All Global managers with equity funds domiciled in Australia. Performance is
measured as the annual excess return relative to MSCI World ex Australia (before fees). Breadth is calculated as the
proportion of global stocks outperforming the World MSCI All Country Index using the QEP Global universe ex Small Caps
Lower active opportunity over twelve month periods through time.
88 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10
Can active managers justify the fees?
Source: Schroders as at September 2010. The proportion of global stocks outperforming the World MSCI All Country
Index using the QEP Mega to Mid universe over six month periods through time. Compounding of returns adds value for the long-term investor
Not only does the risk/return profile for any investment need to be thoughtfully managed, but the burden of
Volatility in markets can be an attractive aspect for active managers, providing a wealth of opportunities to cost must also be carefully considered. Passive investing generally attracts a smaller fee from investors, but
add value and increase returns through different market environments. The lower percentage of stocks maybe a more relevant question should be ‘does the return justify the fee?’
outperforming the wider market in 2008 limited the opportunity set for active managers. However, as the We know that compounding returns adds value. In our view, the most consistent way to accumulate returns
breadth of outperforming stocks widened in 2009, active managers were poised to take advantage of this through time is by investing in a strategy that offers repeatable performance and can work across different
growing pool of strong performers. Analysis suggests that median performance by active global managers market environments.
hit 1.2% versus the negative returns sustained when less than 40% of global stocks were outperforming
(see the following graphs). As an example, let us consider the enhanced-index approach below. The strategy targets modest
outperformance against the MSCI World Index (+1% p.a.). Multiple investment strategies are spread across a
large number of small stock positions to capture broad themes and limit stock-specific risk, and portfolios are
constructed with the aim of outperforming the benchmark across all major market environments e.g. rising
markets, failing markets, markets driven by value stocks, growth stocks, large companies and small companies.
The key strength of this investment approach is the proven ability to produce consistent, repeatable performance
while taking limited index-relative risk4. Generating small, incremental returns which compound significantly
over time has seen the strategy outperform the MSCI World Index by a good margin since inception.
4 The product has typically beaten the index in two out of every three months with a win rate of over 50% in all major
50 market environments.
3 MODERN INVESTMENT APPROACHES
Repeatable performance across Cumulative performance since inception (%) Maximising the breadth of opportunity
different market environments
% % All cap investing
100 40 As we move away from the restrictions of market cap-weighted indices, the breadth of investment
opportunities across the market cap spectrum becomes that much greater. The chart below effectively
illustrates the dominance of the large ‘mega-cap’ names in the MSCI World Index, constituting 65%.
75 30 The allocation to large cap stocks may not be too dissimilar, and obviously liquidity issues must be
25 considered when allocating to small- and micro-cap stocks. However, the mid-cap space offers the greatest
number of opportunities for stocks to deliver strong returns while, at the same time, offering ample liquidity.
The QEP unconstrained approach is to weight stocks based upon their fundamentals and liquidity.
15 We believe this is more balanced and, again, reduces the problem of the ‘mega-cap’ drag. The building
25 10 blocks of our approach are ‘Value’ and ‘Quality’.
Overall Rising Falling Value Growth Large Small MSCI World (Passive Strategy) Enhanced Index Strategy 80%
Source: Schroders. Schroder QEP Global Core composite ■ Net QEP ■ Fees
Source: Schroders. Schroder ReturnGlobal Composite
compared with MSCI World NDR, since inception of the compared with MSCI World NDR in GBP from 31 January 60%
QEP Global Core composite 31 January 2000 to 31 2000 to 31 October 2010. Active management fees 0.35%;
October 2010. Market environment (rising, falling, value, passive management fee assumed 0.15% and other
growth, large and small) defined using MSCI indices. expenses assumed 0.04% (for both strategies). 40%
Past performance is no guarantee of future results.
It is important to remember that no investment is free and we believe that the expense incurred should be 20%
commensurate with the expected return. If an active manager can put clients’ money to good use –
consistently and efficiently outperforming global equity indices while also minimising risk – the resulting 0%
compound returns may very well justify the cost of supporting such an investment. Mega > $20bn Large $5bn – $20bn Mid $1bn – $5bn Small > $250m Micro < $250m
■ Global Value ■ MSCI World
Unconstrained investing Source: Schroder QEP Global Active Value Fund (Representative) at 31 October 2010
Alternatives to passive investing
We would advocate a more ‘unconstrained’ approach to deliver higher returns, while still minimising risk Bottom-up emerging market allocation
exposure. A diversified, unconstrained strategy affords the freedom to invest in the best opportunities – On a regional basis, an unconstrained approach can also take advantage of off-benchmark investments
wherever they are found – and not be forced to invest in a region, sector or stock simply because it forms like emerging markets opportunities, which can offer significant growth potential beyond the scope of
part of an index. Unconstrained investment does not imply a disregard for risk, but instead recognises that developed market investments. Having the ability to dip into emerging markets when they offer value and
constraints are a very costly way of managing risk. We believe risk should be managed more strategically at to avoid them when expensive is preferable to simply including these markets in the benchmark – where
the overall fund level, rather than micro-managed by over-constraining individual components of the portfolio. the investor is effectively forced to own them regardless of price. This ‘self-managing’ process within a
systematic, unconstrained approach is an excellent way to take advantage of a huge breadth of potential
Most people would agree with the benefits of diversification, but few realise that it can also be a way to return opportunities.
generate high returns through unconstrained investing. At any given time, there is a wealth of investment
themes to exploit around the world, but many of these stocks fall outside mainstream indices. Therefore, in
order to capture these, you may need to widen the investment scope, drop index-relative constraints, and
invest in a greater number of stocks than a more conventional portfolio is able to invest in.
% of total fund We think our comprehensive range of complementary investment solutions offer a compelling option for
18% investors looking to generate higher return growth without taking on undue index-relative or stock-specific risk.
Schroders QEP Global Equity Strategies
The below table highlights a selection of QEP Global Equity investment strategies that we believe offer an
attractive alternative to passive investing.
6% QEP Product Range
Global Value Global Value is an index unconstrained, value-based investment strategy designed to deliver higher long
run returns. A focus upon stock weighting, diversification and trading costs means we can capture the long
term premium to value investing efficiently. Targets a gross return of 3-4% per annum above Global indices.
Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Global Quality Global Quality invests in quality companies that offer stable growth, are profitable and are financially strong
05 05 05 06 06 06 06 07 07 07 07 08 08 08 08 09 09 09 09 10 10 10 10 while avoiding the disappointment associated with glamour stocks. Targets a gross return of 2-3% per
■ Emerging Markets Asia ■ Emerging Markets EMEA ■ Emerging Markets Latin America – – – Long Run Average of 10.9% annum above Global indices.
Global Blend A combination of Global Value and Quality to offer the best of both worlds. A 50/50 allocation would target
Source: Schroders. The weightings shown are those from 01 April 2005 through 08 October 2010 for a representative 3% per annum above Global indices in the medium term. The allocation can also be tailored to match
account of the Global Value strategy that is not available to US investors. This illustration should not be viewed as an specific investor time horizons, risk attitudes and return targets. We can also dynamically blend the strategy
investment recommendation. to take advantage of current market conditions.
Global Value Extension A highly flexible approach which can go up to 150% long of Value opportunities and up to 50% short Low
Conclusion: review of an alternative approach versus passive Quality, this strategy has been designed to avoid some of the problems of the rigidly-defined 130/30 funds. In
addition, we look to manage important collateral bets such as currency and beta exposure which are produced
from the unconstrained long portfolio. Targets a gross return of +4% per annum above Global indices.
As an alternative solution to passive investing, an unconstrained, bottom up, systematic and pragmatic
approach can ensure that portfolios are well-diversified and avoid the pitfalls associated with excessive Core Equity Global, US and Japanese enhanced indexation strategies targeting gross outperformance of 1% per annum
for Global & Japan and 0.75% for US that has the benefits of index-based investing in terms of risk and cost
stock or market cap concentration. Schroders QEP investment decisions are based on hard evidence but provides the opportunity to achieve consistent outperformance.
and our process is transparent. By broadening our global opportunity set and deviating from market
cap-weighted approaches, we avoid allocating ‘dead money’ to expensive stocks to satisfy index
For further details of our product range and to access proprietary research, please visit our dedicated
constraints, and are able to really maximise the potential return opportunity.
We believe that investors should consider well-diversified, unconstrained equity strategies as a means
to enhance returns and reduce risk. If efficiently implemented, these should exhibit low concentration, The views and opinions contained herein are those of John Marsland and do not necessarily
while still offering all the benefits of a high-conviction strategy. represent views expressed or reflected in other Schroders communications, strategies or funds.
Passive Index Alternative Solution
Systematic and transparent
Low management fees (If we can justify the fee)
Concentration (too concentrated) (reduced concentration)
Breadth (limited breadth) (maximises opportunity)
Return Drivers (buy high, sell low) (clear investment rationale)
3 MODERN INVESTMENT APPROACHES
3.2 Can investors rely on equities
for long term growth?
Mark Humphreys Strategic Solutions and Stephen Bowles Head of Defined Contribution (DC)
This article considers the role of equity investment. In particular, the investment industry has a common An important way of looking at asset returns is to consider the chance of success of achieving a particular
assumption that equities are ‘safe’ over the long run insofar as they will generate returns well in excess of rate of return. A consideration of the chances of investment return success allows for better planning.
inflation. Many charity trustees choose entirely equity based strategies.
Commonly quoted evidence provided to back-up this assumption includes: What target?
– The average return on UK equities over the last 100 years has been around 5% p.a. above inflation The long-term average return for UK equities is around inflation (RPI) + 5% p.a. This average level does not
vary too much with the time horizon, for example, the average return over all ten year periods is not very
– The UK has never had a 20 year period where the average annual return was less than inflation.
different from the average return over all 30 year periods.
In fact once we look at this assumption more closely the picture becomes less clear. This should have
considerable implications for setting investment strategy. An RPI +5% p.a. target is commonly used as an expected equity return planning investment strategy.
The charts below show the probability of achieving a return greater than an RPI plus target (in this case, 1%,
Equity Performance 2% and so on) over an investment horizon of 20 and 30 years. In the chart below you can see that over the
previous 109 years, with an investment horizon of 20 years, you would have outperformed an RPI plus 5%
We can investigate the performance of equity markets over different time frames because developed equity
p.a. target 57% of the time. If you have an investment horizon of 30 years this improves to 68%.
markets have high quality returns data going back over a century. We are going to look at the long-term
returns from UK equities as we have a very long return history for this region, however the basic ideas are Chart 2: Probability of success versus RPI + margin over 20 and 30 years to end 2008
equally valid for other regions.
Chart 1: Real return compared to inflation plus 5% over one year periods (1900-2008) 100% 100%
100% 94% 85%
80% 80% 86%
-60% + 1% p.a. + 2% p.a. + 3% p.a. + 4% p.a. + 5% p.a. + 6% p.a.
Outperformance over 30 years Outperformance over 20 years
Source: Dimson, Marsh, Staunton: Credit Suisse Global Investment Return Sourcebook.
Returns above inflation +5% Returns below inflation +5% Inflation +5%
Source: Dimson, Marsh, Staunton: Credit Suisse Global Investment Return Sourcebook. UK Equities and RPI.
What is this telling us? Analysis summary
Equities are, on average, a great investment with average returns of around RPI plus 5% p.a. The analysis outlined above has serious implications for trustees when considering a long term investment
strategy. Many would be surprised to know that their chances of achieving an RPI+5% growth target is just
But the shorter the investment time horizon, the more exposed an investor is to prolonged periods of
over two thirds if they invest for 30 years and even lower for shorter periods.
This last point may sound obvious, as we all know that from year to year equity returns can vary a great deal. What practical steps can Trustees take?
What may come as a surprise is that even over the longer term timescales which are relevant to many Reduced spending rates and explicitly extending the investment period will help address the potential
charities, there is still a substantial possibility of failing to meet the target level of return. shortfalls, but there are ways to manage investment risk more efficiently mainly through better designed
default investment strategies.
Where did the idea of ‘safe’ equities come from? Solutions that we have witnessed increased Trustee interest in include:
One of the key reasons for what could be described as an overly rosy view of equities can be identified by
– Diversification for growth – looking for smarter ways to deliver the growth associated with equities but
looking at historic returns.
without the volatility that accompanies this single asset class
The graph below shows the 30 year annualised real return with the real return of less than 5% marked in yellow. – Managing the downside risk associated with equity investment – maintaining equity exposure but rather
than looking towards expensive and permanent guarantees, putting in place quantitative overlays to help
Chart 3: Analysis of 30 year real returns versus inflation to end 2008 switch portfolios into safer assets when they most need it
(rolling returns below inflation plus 5% p.a. in yellow)
– Holding growth assets for longer
– Differentiating between different uses of their investments and applying different levels of investment
risk – essentially the segmentation of the investment assets based on a measure of their risk appetite.
The views and opinions contained in this article are those of Mark Humphreys and Stephen
Bowles and may not necessarily represent views expressed or reflected in other Schroders
communications, strategies or funds.
1929 1935 1941 1947 1953 1959 1965 1971 1977 1983 1989 1995 2001 2007
Below inflation +5% p.a. Above inflation +5% p.a.
Source: Dimson, Marsh, Staunton: Credit Suisse Global Investment Return Sourcebook.
It can be clearly seen that those investors whose 30 year investment period ended between 1977 and 2007
have seen unusually good returns when compared with the longer history for UK equities.
Thus it is likely investors are influenced by the recent past. It may well be that the next 30 years resembles
the previous 30, however there is nothing certain about this.
3 MODERN INVESTMENT APPROACHES
3.3 Diversification in times of crisis
Neil Walton Head of Global Strategic Solutions and Jonathan Smith Strategic Solutions Analyst
Diversification is sometimes referred to as a ‘free lunch’ – a way of reducing risk without sacrificing return. The Figure 1: Getting paid by different risk premia
last 18 months were certainly a stern test of this claim. In this article we look back at how well diversification
strategies fared during the credit crunch and recovery. Diversification may indeed have protected investors
from some of the worst of the volatility seen; however not all diversification strategies would have been
effective as most growth asset class returns suffered to some extent during the credit crunch.
Implementing a successful diversified strategy requires consideration of a wide range of asset exposures
and attention to implementation. This implies that precious governance must be allocated to manage a
diversified strategy. In a world where most charity endowments or trusts are governance constrained,
trustees must weigh the benefits of diversification against other calls on their governance budget. We will
also discuss ways in which investors may be able to do diversification better. We suggest that monitoring
exposure to underlying market and financial factors can offer an alternative perspective to the traditional risk
premia model, and importantly that a dynamic approach to diversification can add significant value.
Passive Private Property Infrastructure High Yield Commodities Hedge Funds Currency
Diversification – the theory
Fundamental to the concept of diversification is that different asset classes provide different sources of Equity Risk Premium Illiquidity Term Credit Insurance/volatility Active Skill
return (we summarise these sources of return, or ‘risk premia’ below, by indicating the driver of the excess
return over a risk free asset). The core of the theory is that diversification reduces risk because these premia Source: Schroders, for illustration purposes only.
react to economic factors in different ways, so for example, when equity markets are falling some non-equity
assets will fall by less or even rise in value.
Therefore diversification should reduce volatility and in times of crisis, a well diversified portfolio is expected
to outperform a pure equity portfolio.
The amount of diversification benefit a particular asset brings can be measured by its ‘correlation’ with other
asset classes. As Figure 3 shows, correlations can change over time and in particular, tend to increase in
Risk Premia times of market turmoil. This is because, despite being driven by a variety of risk premia, all asset classes
Term risk premium: investing in longer-dated instruments than cash ultimately depend on one or both of the following: a growing global economy and functioning financial
Credit risk premium: compensation for the risk associated with lending money markets. In 2008/09 both of these were clearly in short supply. Thus, for many asset classes market
Equity risk premium: compensation for the risks associated with providing capital to back an enterprise sentiment and liquidity problems overwhelmed the fundamentals which limited the benefits of diversification.
Liquidity risk premium: a higher return is expected for locking away an investment
Insurance/volatility premium: compensation for the risk of providing insurance. For example,
in the case of commodities, the ‘insurance’ is against adverse price movements. Sellers of Correlation
commodities usually use forward contracts to lock in a price at some future date. They will often Correlation is a measure of how “alike” two assets are. The higher the correlation, the less
accept what they believe will be a small loss compared to the expected market price, in exchange diversification benefit an asset class brings. If two assets have a correlation of 1 then they
for the benefit of receiving a guaranteed price in the future act in exactly the same way as each other. A negative correlation means the assets act in
Skill premium: a return from the successful application of investment management skill an opposite way to each other. Therefore diversification is increased by holding assets
with low or negative correlations.
As shown in Figure 2 replacing equities with a static diversified portfolio of growth assets did not necessarily
protect funding positions during the credit crunch. For example, a fixed allocation to a portfolio of equities, high
Figure 3: Correlations increase in times of crisis
yield bonds, emerging markets debt, property and hedge funds will have faired much the same as a pure equity
portfolio during the period of market crisis, liquidity squeeze and forced selling from certain groups of investors. Historical correlation ranges (December 1997 – November 2009) compared to correlations over 2008
Figure 2: Diversification will not necessarily have benefited pension schemes during the credit crunch 1.0
UK Equities and UK Equities and UK Equities and UK Equities and
50 UK Direct Property UK Corporate Bonds Hedge Funds Commodities
May 06 May 07 May 08 May 09
Correlations between asset classes over 2008
MSCI AC World Diversified Index*
The marroon bars represent the range between the 25th and 75th percentiles (i.e. correlations were in this range around
*Diversified Index: 50% MSCI AC World TR $,10% HFRI Fund of Funds Composite TR $, 7% JPM EMBI Global Composite half of the time). The black lines represent the range to the highest and lowest correlations recorded over the period.
TR $, 10% ML Global High Yield TR $, 10% FTSE EPRA/NAREIT Developed TR $, 3% LPX50 TR $, 10%DJ UBS Future Source: Schroders, Datastream. For illustration only. Indices used, FTSE All Share, ML Sterling Non-Gilts, Credit Suisse
Commodity Index ER $. Rebalanced on monthly basis. All shown currency hedged to Sterling Tremont Multi-Strategy Index, IPD Property Index. In GBP except Hedge Funds and Commodities in USD. Rolling annual
Source: Thomson Datastream, Schroders. Updated 30 June 2009. correlations based on monthly data.
3 MODERN INVESTMENT APPROACHES
The implication of this is that it becomes much harder to reduce risk via diversification in times of market
turmoil. This is not to say that diversification does not have value in times of crisis, but rather that we need to
work harder to find those diversifying assets that genuinely will reduce risk. This implies a more ‘dynamic’ Dynamic Asset Allocation (DAA)
approach to asset allocation is required. DAA is a medium-term approach to the adjustment of the asset mix, looking out 1 to 3
years or longer. DAA uses broad market signals and the relationships between asset
A dynamic approach to diversification classes and their macroeconomic environment, valuation and sentiment (for timing) to
inform allocation decisions.
Once a scheme has put in place a diversified selection of growth assets it can be tempting to think ‘job
done’ – however this would be a mistake. As the historic analysis below shows, different asset classes have
on average performed well at different stages of the economic cycle. Furthermore, as we saw above, the
Building a dynamic approach into your scheme’s investment strategy will clearly eat into your governance
performance of risk assets tends to converge in times of crisis, and of course the characteristics of any crisis
budget. Note however that, as characterised here, dynamic asset allocation is not the same as tactical asset
are unique, even if there are similarities with past crises and recessions. Therefore, a dynamic approach to
allocation (TAA). TAA normally implies a large number of trades with the aim of identifying areas for short
portfolio management can add significant value and manage risk.
term gain, and is indeed very governance intensive. Dynamic asset allocation should be seen as more of a
Figure 4 gives the average returns for a range of asset classes for each part of the economic cycle. medium term approach, perhaps involving taking major positions that are intended to provide value as the
Interestingly, equities have historically only been the best performing asset of those shown in one phase of economic cycle evolves over months and years.
the cycle – the recovery phase.
Figure 4: Varying your asset allocation through the economic cycle can add significant value
Diversification in a governance constrained world
As the complexity of the investment universe increases trustees are increasingly finding it a challenge to find
Phase/Asset Recovery Expansion Slowdown Recession the governance budget to achieve all of their investment aims. This means that trustees may be forced to
Equity 8.5 8.8 -8.1 10.0 prioritise those strategies that have the potential to offer the biggest gains, delegate more of their activities
or ask their asset managers to take on more responsibility.
Government Bond 1.4 -1.0 -0.2 1.7
High Yield Bond 6.2 1.8 -10.3 13.2 These considerations beg the question – how does diversification rank in terms of its ability to reduce risk
in times of crisis compared to other strategies? The answer, of course, depends on the specifics of your
Investment Grade 3.8 -0.1 -4.2 5.1
circumstances and of the form of the crisis. Arguably a permanent endowment fund with a large allocation
Commodity 1.9 16.2 7.7 1.0 to growth assets will benefit more from diversification than a portfolio of short-term assets with a large
Cash 4.7 4.8 5.9 5.6 allocation to bonds.
Source: Thomson Datastream, Schroders, Global Financial Data.
All return figures are in % and annualised from monthly data. Equity: S&P 500 composite (1950-2008), Bonds: US Govt.
10yr (1950-2008), US Lehman HY Corp. bond (1983-2008), US Lehman Corporate Investment grade (1973-2008), Cash:
US 3M T-Bill (1950-2008), Commodity: S&P GSCI commodity Futures TR Index (1970-2008).
Can we have our cake and eat it? An alternative way of ‘looking through’ a portfolio is by using factor exposure analysis. This is a process
whereby we look at how correlated a portfolio has been historically to a number of different economic
Diversification funds have been around for a few years and offer schemes with limited governance budgets factors, with the intention of identifying any hidden relationships.
a way to access diversification without the need to select and monitor a large number of managers. Many
diversification funds contain an element of dynamic asset allocation, so can offer a low governance way of The charts below show a traditional risk premia analysis and a factor exposure analysis for a portfolio with
ensuring that the portfolio reflects market conditions. a combination of bonds, UK and overseas equities, property and private equities.
The first chart highlights how equity risk premia dominates return, however, as the second chart shows,
Figure 5: Example asset allocation of a diversification fund
it understates the extent to which the equity exposure dominates risk. Figure 6 shows contribution of
some factors (not an exhaustive list) to the risk in the portfolio. For example, according to the analysis
approximately 80% of the contribution to total risk (i.e. volatility or standard deviation) comes from equities.
High Yield Debt 18%
In other words, even though equities account for 60% of the allocation (including private equity) they
Hedge Funds 17%
contribute almost 80% of the risk.
Emerging Markets Debt 13% Figure 6: Breakdown of contribution to expected portfolio return by risk premia
Private Equity 5%
Leveraged Loans 5%
Active Currency 5%
Convertible Bonds 3%
Catastrophe Risk 3%
Source Schroders, for illustration only.
Diversification – an alternative perspective UK Equities Global Equities Private Equity Property
The most common way of assessing how well diversified a portfolio is, is to look at the composition of Equity Risk Premium Illiquidity Credit Active Skill
its risk premia (effectively considering the mix of exposures indicated in Figure 1). In this case we treat
government bonds as if they are a risk-free asset. Whilst this is an intuitive method of presenting the
underlying characteristics of a portfolio, it does have limitations. In particular its relatively simplistic
presentation can mask a number of ‘hidden’ exposures within a portfolio. Furthermore, it gives no
indication of the contribution to overall risk (which is unlikely to be in proportion to the contribution to
return) of each asset exposure and risk premia.
3 MODERN INVESTMENT APPROACHES
Figure 7: Factor exposure analysis gives an alternative perspective
Emerging US Treasuries US Equities US Credit Trade Commodities US Small Cap
Market Premium 5-10Y Premium Weighted Dollar Premium
Factor Contribution to Risk
Source: Schroders, for illustration purposes only. Assets are 30% fixed interest gilts, 25% global equities, 25% UK equities,
10% property and 10% private equity.
For many portfolios diversification will have added significant value during the credit crunch; however the last
18 months have also highlighted some of its limitations. In particular, in times of crisis, “correlations move
towards 1”. This reduces the effectiveness of diversification at reducing risk and highlights the importance
of a more ‘dynamic’ approach to asset allocation.
The use of diversification funds as a ‘one stop shop’ to bring genuine diversification to a portfolio of growth
assets is a useful way to cut through the governance constraints faced by many schemes.
Trustees should also consider if they can ‘do diversification better’, for example by looking at the factor
exposures in their portfolios, not just the risk premia.
The views and opinions contained herein are those of Neil Walton and Jonathan Smith and may
not necessarily represent views expressed or reflected in other Schroders communications,
strategies or funds.
3.4 The power of technical analysis
Jamie Fairest Fixed Income Fund Manager
Technical analysis can prove a useful tool in the search for additional sources of return. The power of Figure 1: Alternative representations of price
technical analysis can be utilised over any time scale and across all asset classes. The benefits of technical
analysis and how profits can be realised from using these techniques are discussed in this article along with
current projections. High
Technical analysis encapsulates the combined psychology of the different market participants that rule the Open
supply and demand dynamic of price. The three key guiding principles of technical analysis are:
1. History repeats itself
2. Prices move in trends
3. Price discounts everything.
The art of technical analysis Low
One of the cornerstones of technical analysis is the belief that history repeats itself. Looking back through
Candlestick Bars Point and Figure
time, it is clear that certain patterns emerge time after time. Based on the balance of probability that a
complete pattern will occur again, these patterns can be used in real time to assist trading. The real ‘art’
comes in the correct recognition and interpretation of these patterns. For example, two analysts can look at
the same chart and come to very different conclusions as to the future direction of price due to the
subjective nature of establishing trends, patterns and key levels. This refutes the often held perception that
technical analysis is a self fulfilling prophecy and the common belief that price will move in the same Common chart patterns
direction as everyone is trading.
A few of the more common patterns shown below (Figures 2, 3 and 4) are easy to spot with the benefit of
hindsight but only a thorough knowledge of technical analysis will give an investor the edge to trade in real
Chart construction time when patterns are still forming. Technical analysis can be used to assist trading in favour of a trend or
The method of how price data is presented in charts varies with dramatically differing results (Figure 1). as a contrarian tool when the trader opposes the market direction.
Candlesticks are arguably the oldest technique borne out of Japan in the mid-1600s using the high, low,
open and close data points to construct a ‘candle’. This method makes it easier to identify movements in
price and identify patterns. The western alternative, bar charts, use the same data points but are
represented in a different manner, simply as a bar. In fact, candlesticks are now more widely used due to the
extra flexibility they provide and quicker analysis that can be achieved. Before computers, point and figure
charting was used due to its simplicity and ease of drawing by hand, with price movements annotated with
an X (up) or O (down) in columns. But importantly this method has no timescale, so only actual changes in
price are recorded.
3 MODERN INVESTMENT APPROACHES
Figure 2: Double top in US 5-year yield Figure 3: Inverted head and shoulders in EUR/USD
31 Dec 2009 5 Apr 2010
2.700 High 2.7531 High 2.7533 2.7530 1.3300
1.3200 27 Jul 2010
2.600 1.3100 High 1.3046
2.500 1.3000 1.3007
21 Jun 2010
1.2600 High 1.2488
2.100 5 Feb 2010 1.2400 Sloping Neckline
2.000 1.2300 1.2248
1.2200 Left Shoulder Right Shoulder
1.900 1.2100 19 May 2010 29 Jun 2010
1.800 1.2000 Low 1.2144 Low 1.2151
1.700 1.1800 7 Jun 2010
21 Jul 2010 Low 1.1876
1.600 Low 1.6205 1.1700
Dec 09 Jan 10 Feb 10 Mar 10 Apr 10 May 10 Jun 10 Jul 10 May 10 Jun 10 Jul 10 Aug 10 Sep 10
Source: Updata/Bloomberg. Source: Updata/Bloomberg.
The US 5-year yield hit a high in December 2009 and again in April 2010 at 2.75% with a middle low of For an inverted head and shoulders the middle low (head) makes a new low, but the subsequent move
2.19% (February 2010). down (right shoulder) fails to reach the same level of conviction.
On the break of this level in May 2010 a measurement of the next move from the break can be applied as a This pattern indicates a reversal may be likely, which occurred when EUR/USD broke the neckline of 1.24 at
target level of 1.63%. This was achieved in July 2010. the beginning of July.
(Target = 2.19-(2.75-2.19) = 1.63) This pattern gives us a measured target being the distance from the head to the neckline projected from the
break. The target of 1.29 was accomplished on 16 July 2010.
(Target = 1.24+(1.24-1.19) = 1.29)
Figure 4: Support becomes resistance in Japan 10-year yields
1.80 Combining the art and science of technical analysis gives investors valuable tools to build scenarios where
profitable trades can be implemented. The added advantage of technical analysis techniques is that definitive
target and stop-loss levels can be subjectively calculated and so an acceptable risk/reward ratio is known
1.60 pre-trade. For example, a risk/reward ratio of three would yield a return of at least £3 with a maximum loss of £1.
1.50 Technical analysis helps with trade timing
1.40 The timing of trades is crucial for investors. A directionally correct trade that is poorly timed can lead to
unnecessary losses. Technical analysis enables investors to capitalise on their trades by giving them the
tools to enter the market at the start of trends and exit when the trend starts to reverse. For example, a
resistance level shows where bearish investors are the dominant force and continue to push prices lower.
When the level is broken these investors will quickly cover their loses by pushing the market higher.
1.10 The euphoria of more bullish investors will continue to push the trend higher until the bearish investors
regroup and try again at a higher level to push price down.
Nov 08 Jan 09 Mar 09 May 09 Jul 09 Sep 09 Nov 09 Jan 10 Mar 10 May 10 Jul 10 Sep 10
Current analysis (at 04 August 2010)
Figure 5: US 10-year yields
A common occurrence is when a previous support level becomes a resistance level. This highlights a 4.000
change in the psychology of bullish and bearish investors. Conversely, this is also true when resistance 4.00283 11/09/09
Here, the 1.16% in the Japanese 10-year yield had been tested but held going back to 2005 (not shown on 3.500 23.6%
the chart). When this finally broke, it was swiftly retested before heading lower again.
The use of indicators 3.000
3.10144 -0.90139 -22.51%
11/09/09 0 3.09477 11/09/09
A further tool that can aid understanding price movements is the use of indicators. There are numerous
indicators available ranging from signals for trend-following trades to identifying over-bought or oversold 61.8%
conditions for contrarian trading. The moving average is the most widely used indicator which in its simplest 2.500
form averages the last ‘x’ number of days. For example, a 200-day moving average is the average of the last 78.6%
200 days’ closing prices. The earliest price is removed each day and the current closing price added.
2.20006 -0.89471 -28.91%
11/09/09 0 DT target = 2.20%
An understanding of each indicator’s characteristics and how it is calculated are essential in using them 2.000
successfully. It is important to remember that indicators are a derivative of price and while they can provide May 09 Jul 09 Sep 09 Nov 09 Jan 10 Mar 10 May 10 Jul 10 Sep 10
good signals, they should remain secondary to the thorough analysis of price charts. Source: Updata/Bloomberg.
The science of technical analysis The trend of lower yields in the US 10-year continues with a clear trend line established. The break of the 3.10%
With the development of systematic investment rules, signals can be triggered by the interaction between level was significant and subsequent pushes higher have failed to overcome this technically important level.
price and a number of different indicators. A simple example would be a buy signal being generated when A continuation of the existing trend is expected with targets of 2.75% and 2.45%. The target measured from
the price crosses the 200-day moving average. the double top at 4% comes in at 2.2% but will hopefully not be achieved.
3 MODERN INVESTMENT APPROACHES
Figure 6: UK 10-year yields
Mar 09 May 09 Jul 09 Sep 09 Nov 09 Jan 10 Mar 10 May 10 Jul 10 Sep 10 Nov 10 Jan 11 Mar 11
A similar picture is presented in the UK with the 3.30% level breaking on 03 August 2010. If this level holds,
after an expected retest from below, this will open ground to a rapid move lower in yields.
Interestingly, the gilt yield has now broken the third upward trend line which is also technically significant and
warns of lower yields.
Below 3.30% a target of 3.10% is given with 2.95% signifying a full retracement of the March 2009 to
February 2010 move.
Technical analysis is a powerful tool that can be incorporated into the investment decision-making process
to complement fundamental analysis. In addition to the identification of new trades, technical analysis can
also be utilised to identify entry and exit levels for existing trades (both profit taking and stop-loss levels).
However, a thorough understanding of the various techniques is essential to maximise the full potential
gained from technical analysis. This suggests that technical analysis can be used successfully to provide
additional sources of return for investors.
The views and opinions contained herein are those of Jamie Fairest and may not necessarily
represent views expressed or reflected in other Schroders communications, strategies or funds.
3.5 The role of equity investing in a modern portfolio
Neil Walton Head of Global Strategic Solutions
The role of equities in a portfolio is to harvest the equity risk premium, the extra return from equities over The insights that have developed around some of the issues with market capitalisation indices allow a
lower risk assets, and for this to play a part in maintaining the real value of the assets and allowing higher systematic approach to be followed that addresses the large size bias and the potential for ‘market cap’
spending levels over time. to overpay for past success. The core of such an approach is:
Success on both these criteria is the true measure of progress. Maintaining the real value of the portfolio, – unconstrained investing – remove the benchmark constraints and invest in the widest possible
whilst spending a defined amount, is thus the real benchmark and so the choice of equity benchmark, universe of companies
although important, is secondary to the overall investment objective of the scheme. – non-market capitalisation weighting – don’t weight stocks based on their market capitalisation
Market benchmarks have an interesting history. Initially they were an indicator of broad market sentiment and – have a clear driver of return – for example a focus on ‘value’ by choosing a portfolio of companies
price movements. The financial theories of the 1950s and 1960s, gave market capitalisation benchmarks that represent tangible financial worth.
more credibility and led to the development of a substantial passive (index-tracking) investment industry.
In the context of meeting an inflation plus target, the role of equity investing remains to capture the equity
Long term evidence is not supportive of the financial theories that support index-tracking, as both cheap risk premium. However this outcome driven view reduces the ‘headline’ nature of the equity benchmark and
stocks (value investing) and less market sensitive stocks have outperformed over time suggesting that there potentially allows additional return to be achieved by removing the damaging constraints associated with
is more to real world investing than the theories indicate. market capitalisation indices.
Behavioural finance draws out some of the human nature of investment markets and through the concepts
of ‘following the herd’, short term decision making and being drawn to interesting stories starts to explain The importance of benchmarks
some of the less rational elements of investment markets, including stock market bubbles. The choice of equity benchmark, or potentially, the choice to down play the benchmark via ‘unconstrained’
Concern over the effectiveness of mainstream market indices has led to a search for improvements such as equity strategies, sends a very strong message to the investment manager. This is a crucial decision which
equal weighting of companies in an index or approaches that focus on business measures such as sales or will impact both risk and returns from the equity portfolio.
dividends rather than company size as determined by the current market price. These approaches address The original benchmark dates back to Charles Henry Dow who published a weighted average of 11 railroad
some of the problems with traditional indices, but raise the question – do they go far enough? stocks back in 1884. This became the Dow Jones index in the USA. A significant step was in 1923 when the
Standard Securities Corporation created an index of 223 securities where the weighting of each company
reflected its size. This index became the S&P 500 in 1957. The significance of this development is in the use
of the size of each company (market capitalisation) to weight each company in the index. Closer to home the
FTSE Actuaries All Share Index was launched in 1962 with an initial value of 100. It’s now around 3,800
(using the total return variant).
3 MODERN INVESTMENT APPROACHES
As these benchmarks were initially developing they were really only used to indicate the broad movement of Value investing, efficient markets and market
the stockmarket. A revolution was then set in motion in the 1950s and 1960s from the work by a series of
forward thinkers, including Harry Markowitz, William Sharpe, Jack Treynor and other, primarily US based, capitalisation benchmarks
academics. Unfortunately as time progressed through the following 40 years the differences between theory and practice
Harry Markowitz is the father of modern portfolio theory. Markowitz believed that investors should be started to emerge and the restrictive assumptions made in these financial theories have been questioned.
concerned about risk as well as return. This allowed him to introduce the concept of efficiency – investors The key assumptions of CAPM include:
should aim to maximise return for a given level of risk.
– no transaction costs (cost have a material impact on real world portfolios)
William Sharpe and Jack Treynor built on this work in the early 1960s when they introduced the Capital
– investors can buy or sell (even take short positions) in any security without moving the price (market
Asset Pricing Model (CAPM). The insight of this work is that an investor will not receive a return for taking
impact from adjusting positions is an issue to be managed by many significant institutional investors)
any risk that can be diversified away, so the expected return of any share reflects only the company’s
non-diversifiable or specific risks. – investors have the same time horizon (which is unlikely); and
– investors can borrow or lend at the risk free rate (the credit crunch has demonstrated that borrowing
Sharpe then went on to show that if a number of key assumptions were made, a portfolio of the whole
is not always straight forward).
market weighted by company size (market capitalisation) is the portfolio that every investor should hold
because it is efficient. As with any theory or model of the world, it is a simplification and should be treated as such.
This work provided the impetus for the development of market capitalisation based indices and the Possibly the biggest problem with the theories is that over long periods it has been observed that cheap
subsequent idea of passive equity investing or index tracking. Over the next 40 years passive investing and stocks deliver better returns than the market. This is value investing and the long observed value effect.
management of equity portfolios relative to market capitalisation benchmarks would take hold across the
globe, becoming one of the dominant approaches in investment management. Cheap stocks have outperformed expensive stocks over time
The benchmark given to an investment manager is now a key part of the way an institutional portfolio is Log Scale
constructed. Market capitalisation indices, such as the FTSE All Share or the S&P 500 are not just used to 100,000 Benjamin Graham was an economist and is
indicate the broad direction of the market; they have become the foundation of actual portfolios. The idea viewed as an initial proponent of value
that the market is difficult to beat and that a low cost passive investment is therefore efficient seems very investing. He set out the difference between
sensible; it has become the standard, but is it truly effective for longer term investors? investment and speculation; ‘An investment
1000 operation is one which, upon thorough
analysis, promises safety of principal and an
adequate return. Operations not meeting
these requirements are speculative.’
Working with David Dodd he set out a
framework for investing that focused on the
intrinsic value of a company and aimed to
0.1 invest in companies where the market price
was below this intrinsic value, bringing a
1926 1936 1946 1956 1966 1976 1986 1996 2006
margin of safety to the investment.
Source: Schroders, Ken French.
Some would argue that the outperformance of value stocks reflects the fact that cheap companies are Technology/Telecoms bubble of 1999/2000 Japanese stock market bubble of the late 80s
cheap for a reason and this extra return reflects additional risk. It has been well documented that over long
% Weight % Weight
periods, both emerging market stocks and smaller companies have delivered better results than the broad
global stockmarket as measured by the mainstream indices. The higher risk argument applies for these
sub-sections of the market and so arguably risk and return are related. But, unfortunately for the efficient 44% of MSCI World is
comprised of Japanese stocks here
market theory a body of evidence is growing that a portfolio of high risk stocks (measured by an increased 40% 40%
36% of MSCI World is
sensitivity to movement in the overall market; ‘beta’) actually underperforms the global index. comprised of tech stocks here
Under the theories the expected return was supposed to be related to this measure of risk (beta) and so this
result, which is robust across many markets and over different time periods, indicates that this extra risk
(from the high beta) has not been rewarded with higher returns.
Similar research into the historic performance of portfolios of ‘cheap’ value stocks shows that these have 10% 10%
added value over the global index, but this has been achieved with less volatility or risk. The theories fail to
explain how value stocks have delivered this extra return across this period with lower volatility, lower beta 0% 0%
and in market downturns with lower falls in share prices than the overall market. 1986 1989 1992 1995 1998 2001 2004 2007 1986 1989 1992 1995 1998 2001 2004 2007
The outperformance of cheap stocks over long periods is one of the most serious attacks on the credibility MSCI World MSCI World
of the efficient market theories (including CAPM). This is crucial because the CAPM is the cornerstone that
supports the use of market capitalisation weighted indices and benchmarks (and of course passive investing). Source: Schroders, MSCI.
The concept of value investing stands in contrast to the idea of efficient markets. Value investing is directly
linked to avoiding overvaluation or stock market bubbles. Weighting companies by their market capitalisation Behavioural finance
actually does the exact opposite. There are many examples: the technology bubble of the late 1990s pushed The observations above have significantly dented the credibility of modern portfolio theory and the CAPM.
investors into the technology and telecommunications sectors and many UK pension funds found that around It is rare to find an investor who believes that markets are totally efficient. A different area of academic work
15% of their UK equity portfolio was invested in one company – Vodafone. Ten years earlier the success of helps to explain the problems between theory and practise for the efficient market theories.
the Japanese economy using Kanban (just in time manufacturing) and Kaizen (continuous improvement)
techniques led to the Japanese stock market becoming an astonishing 44% of the global stock market index Behavioural finance recognises that investment decisions are taken by people and this brings a
(MSCI World), even though Japan only had a 15% share of the GDP of the MSCI World constituent countries1 psychological element to investment markets. Psychologists have put forward the idea that there are two
As this bubble unwound over 15 years the performance drag was a huge 2.6% per annum.2 systems in the brain; an emotional system (Dr McCoy from Star Trek) and a logical system (Mr Spock).3
The McCoy system developed much earlier as it was needed to survive. For example at a 3D movie,
if a car comes flying out of the screen at the audience, they will flinch and dive for cover (McCoy) before
the Spock system catches up and reminds them that they are in the cinema.
For important decisions, we all think that we will follow a Spock approach, but McCoy is likely to be in there
somewhere (try the exercise in the appendix).
There is a wide body of work, but some of the key findings are that we are ‘hard wired’ to follow the herd.
Classic experiments that date back to the 1950s involve a group of people where only one is the subject of
the experiment, the others work for the experimenter. In one example the subject is shown a line on a piece
of paper and then other lines alongside; one is the same length, one clearly too short and one longer.
1 purchasing power parity adjusted weight sourced from the IMF
2 the annualised difference between the MSCI World and the MSCI Kokusai from 1988 to 2006
66 3 Source Behavioural Investing – A practitioner’s guide to applying behavioural finance – James Montier.
3 MODERN INVESTMENT APPROACHES
The experiment is to ask the subject which of the three lines is the same length? If the others in the group In the world of investing these behavioural traits of herding with the group, following interesting stories and
go first and suggest the shorter line, amazingly about two thirds of the time the subject agrees! seeking short term rewards have implications. Could they be linked to stock market bubbles with a great
story accepted by the majority?
Asch Conformity Experiments
Alternatives to current benchmarks
If we put aside the use of market capitalisation as the measure of how large a company should be in the
benchmark we can firstly consider a benchmark of equally weighted companies (which was the approach
adopted by the early index constructors over 100 years ago). We can see if avoiding some of the problems
with market capitalisation makes a difference.
MSCI versus Equally Weighted MSCI
A B C 2.7% p.a.
Source: Asch, S. E. (1951). Effects of group pressure upon the modification and distortion of judgment. In H. Guetzkow
(ed.) Groups, leadership and men. Pittsburgh, PA: Carnegie Press.
In this form of experiment, neuroscientists have actually found that going against the group involves the 88 89 92 94 96 98 00 02 04 06 08 10
same part of the brain as fear, so going against the crowd makes us scared and can also be painful. MSCI World total return Real S&P Index Level (rhs)
Other investigations have highlighted a bias towards short term results. For example within a study4 a Source: Schroders, MSCI. Schroders has estimated Equally Weighted benchmark prior to January 1999.
majority of people preferred to receive $10 today rather than $11 tomorrow, but when framed differently,
preferred $11 in 365 days time rather than $10 in 364 days time. One theory that has gained much empirical support relates to the idea of pricing noise leading to the inefficient
allocation of capital in a market cap index. Basically if a stock is worth $10, but trading at $12 in the market
We are also drawn to interesting stories, the McCoy system reacts to the narrative when making decisions. today, this (positive) pricing error leads to the stock being overweighted in the market cap index. Similarly
A good example comes from the world of medicine.5 undervalued stocks are underweighted. The overall result is a significant drag on investment returns.
The participants in a study were told that a drug, Tamoxol, was 90% effective, but could have material side By avoiding the noise in market cap weightings (which is similar to chasing momentum) investors can add value
effects. One group were then told a negative story about only one individual along the lines of ‘the worm was by avoiding bubbles. This thinking led George Keane (President of the New York Common Retirement Fund) to
not completely destroyed, the disease came back; the patient was blinded and lost the ability to walk’. Another approach Rob Arnott (Chairman of the Californian-based Research Affiliates) in 2002 to explore a more effective
group received a positive story, ‘the worm was destroyed, the doctors were confident that the disease would not index. Mr Arnott took up the challenge; the first thing he and his team of researchers did was to quantify the
re-appear, the patient has fully recovered’. Remember the story only covers one person of the thousands who impact of valuation bubbles in large stocks over time. One point they looked at was the performance of the
have been treated with the drug. There are two important pieces of information (Spock), firstly that the treatment largest companies in the years following their growth into the top ten percent of the market.
is 90% effective and secondly it can have material side effects. However 88% of the group who were told the
positive story said they would take the treatment, but only 39% would from the group with the negative story.
4 Department of Psychology and Center for the Study of Brain, Mind, and Behavior, Princeton University, Princeton, NJ08544,
USA. Department of Economics, Havard University, and National Bureau of Economic Research, Cambridge, MA 02138, USA.
5 A Freumuth and G Ronan, ‘Modelling Patient Decision-Making: The Role of Base Rate and Anecdotal information – Journal of
Clinical Psychology in Medcial Settings (2004). 67
Frequency top 10% of stocks beat the average stock Equally Weighted MSCI versus Cap Weighted MSCI versus Global RAFI
20% 8 2.7% p.a.
1 year 3 year 5 year 10 year 88 90 92 94 96 98 00 02 04 06 08 10
Equally Weighted MSCI MSCI World total return Global RAFI
-3.6% per annum performance on average
Source: Robert D, Arnott, FAJ March/April 2005. Source: Schroders, MSCI. Schroders has estimated Equally Weighted benchmark prior to January 1999 and Global RAFI
(Research Affiliates Fundamental Index) prior to January 2000.
This analysis showed that 71% of the time the largest stocks lag behind the average stock over rolling 10
year periods, indicating an element reflecting ‘yesterday’s winners’ within the largest companies measured Although the fundamental index is determined from a set of transparent rules referencing the ‘main street’
by market capitalisation. variable mentioned above, is it an index or is it an investment approach?
Rob Arnott went on to write an article in the Financial Analysts Journal in 2005 which set out many of the
criticisms of standard benchmarks and proposed an approach called fundamental indexation. This was
Index or strategy
positioned as weighting companies according to ‘Main street’ factors such as sales and dividends The strength of a fundamental weighting approach has been to question the efficiency of market
measured over the medium term, rather than ‘Wall Street’ measures. capitalisation indices during valuation bubbles and offer a solution that reduces this impact while offering
many of the best features of passive approaches. It is of course an ‘index’ approach with the problems of
The idea is that the weighting in a portfolio should reflect real world measures of company size, such as 5
fixed and rules based re-balancing process.
year histories of sales and dividends, cash flow and book value measure. This avoids the measures that rely
on current price and are thus driven by the market’s view of a company’s worth. Does the fundamental approach go far enough? We do not think so. If there are synergies from value and
non-cap weighting and even a simple, rules-based strategy that fixed re-balancing can perform so well
The use of sensible measures to counteract the behavioural biases of the market capitalisation indices is an
against a cap-weighted index then surely a carefully constructed, well-diversified, fully active strategy
improvement over just equally weighting companies in the index.
focused on value as a driver of return could do even better?
In a similar form to the previous charts if we add in value investing and remove the constraints related to the
market cap indices we see very interesting long term results.
3 MODERN INVESTMENT APPROACHES
Adding value over the MSCI World Conclusions
The role of benchmarks in investment management has evolved over the years from being one that indicates
a broad market result to a control mechanism that drives the shape of the equity portfolio. The ascendancy
2000 of capitalisation weighted indices was a direct result of breakthroughs in academic thinking throughout the
second half of the 20th century that lauded the efficiency of such an approach. But whilst the academic
1500 theory supporting them has been heavily questioned in the more recent past, traditional market
capitalisation weighted benchmarks have become the mainstay of modern investment practice for both
active and passive managers.
1000 8% p.a.
More recently, the experience of living through some extreme valuation bubbles – largely caused by herding
behaviour – has exposed many of the pitfalls associated with benchmarking and in particular the
performance drag that results from chasing yesterday’s winners. Behavioural finance is an additional body
of academic work aiming to capture the human side of financial markets and helps us to understand the
0 problems with traditional benchmarking.
88 90 92 94 96 98 00 02 04 06 08 10
Both equally weighted and fundamentally weighted indices share the characteristic of avoiding yesterday’s
MSCI World Value Non-Cap Weighting Unconstrained
winners, but the key benefit of the fundamental idea is that they are very scalable and are far more appropriate
Source: Schroders, Proprietary QEP systems. Estimated data shown is representative of 1987-2009. Past performance is to serve the purpose of a benchmark for active managers. The main criticism levelled against a fundamentally
no guarantee of future results. The value of an investment can go down as well as up and is not guaranteed.
weighted approach is that it is not an index but instead a semi-active value investment strategy.
These numbers are based on a realistic backtest that avoided the bubble in Japan and lagged behind in the We view capitalisation weighted benchmarks as just another constraint that can be loosened in order to
development of the technology before more than making it back in the tech-wreck of the early 2000s. Such generate excess returns as part of a funding level improvement strategy for the pension scheme’s
a strategy is certainly not replicating Fundamental indices, as there is only a 15% overlap between the investments. The core of such an investment approach is for the equity portfolio to be;
stocks in the unconstrained strategy and those in the Fundamental Index. Instead, it amply illustrates how an – unconstrained – remove the benchmark constraints and invest in the widest possible universe
active manager can exploit the same insight about avoiding valuation ‘noise’ as part of a broader investment
process that also loosens all restrictions on the investible universe and weighting scheme. – non-market capitalisation weighting – don’t weight stocks based on their market capitalisation
– have a clear driver of return – for example a focus on ‘value’ by choosing a portfolio of companies that
represent tangible financial worth.
Finally this suggestion of an unconstrained equity approach does not imply a disregard for risk, but
recognises that unnecessary constraints are a very costly way of managing performance. Instead, we
believe that risk should be managed more strategically by considering the portfolio outcome rather than
micro managed by over-constraining individual components of the equity portfolio.
The Shane Frederick developed Cognitive Reflection Task (CRT)
[S Frederick, Cognitive Reflection and Decision Making, Journal of Economic perspectives 19 (2005)]
Three simple questions to determine the influence of the McCoy system:
A bat and a ball cost £1.10 in total. The bat costs £1 more than the ball. How many pence does the ball cost?
If it takes five machines, five minutes to make five widgets, how many minutes would it take 100 machines to make 100 widgets?
In a lake there is a patch of lily pads. Every day, the patch doubles in size. If it takes 48 days for the patch to cover the entire
lake, how many days would it take for the patch to cover half the lake?
To arrive at the right answer candidates need to suppress the first response that springs ‘impulsively’ to the mind (McCoy)
and instead work it out logically (Spock).
1 bat + 1 ball = £1.10
1 bat – 1 ball = £1.00
Add the two equations so that 2 bats = £2.10 so 1 bat is £1.05 and the ball must be 5p.If it takes five machines five minutes to
make 5 widgets, the output is one widget per machine very 5 minutes, so 100 widgets with 100 machines also takes 5 minutes.
Question 3 is most commonly answered incorrectly. If it doubles everyday then the day before the whole lake is covered
half will be covered, so one day less, 47 days. Of the 3,500 people that Professor Frederick tested only 17% got all three
right. Around a third got none right! This illustrates that people are prone to decision making using the McCoy, emotional
system even when the logic is reasonably straightforward.
The views and opinions contained herein are those of Neil Walton and do not necessarily
represent views expressed or reflected in other Schroders communications, strategies or funds.
4.1 Risk Management is more than just tracking error 72
4.2 The Cyclical Framework: Cycling Around Asset Allocation 77
4.3 Lies, damn lies and (risk) statistics for pension funds 85
4.1 Risk Management is more than just tracking error
It has long been recognised that risk management is a critical and sometimes challenging aspect of any We are also particularly concerned that models built over historical periods may not be very representative
investment process. For many investors this means setting some broad constraints on the portfolio such as going ahead. One recent example of how risk models can easily miss a key fundamental relationship is the
maximum sector or country exposures relative to a benchmark as well as specific limits on individual stock recent behaviour of Value and Quality (Profitability, Stability and Financial Strength) compared to the past
holdings. Quantitative investors are in theory much better placed to provide a robust description of their risk twenty years. We have long highlighted the limited commonality between Value and Quality stocks and
management process as the majority tend to use a risk model and an optimisation engine to build their suggested that Quality (not ‘Growth’) offers strategic diversification from Value. This is largely because
portfolios. Both are however, means to control active risk which is better known as tracking error – an cheap companies tend, on average, to be characterised by more variable profitability, lower growth and
estimate of the volatility of relative portfolio returns compared to a benchmark. higher leverage. However, most risk models will have failed to capture this relationship since this is not
suggested by the historic correlation.
Whilst tracking error appears to have become the industry standard for assessing risk we believe that there
is a huge gulf between tracking error management and portfolio construction in the real world. The former The strong returns to Quality and the subsequent sharp fall in correlation since the start of the global
may tick the box but frequently leaves investment opportunities unexploited while the latter requires a more financial crisis is, in fact, more in line with the longer-term historical average (1956-2009) as the chart
pragmatic perspective of the evolving nature of risk. More specifically, we have two primary objections to the below clearly highlights.
over-reliance on tracking error as means of managing risk, one technical and one philosophical:
– Many of the key assumptions underlying the tracking error calculation are fundamentally flawed Quality/Value: Low long run correlation
– A single risk measure is appealing (and very neat) but fails to embrace the multi-faceted nature of risk. 0.5
Sector, Country and stock constraints are an inefficient means of capturing returns and do not
recognise that the primary risks to a portfolio are thematic in nature and vary over the course of the
cycle. We believe that the ‘best’ portfolio managers are often the skillful and imaginative portfolio 0.3
constructors who understand when opportunities are attractive on a risk–adjusted basis.
In this paper we summarise some of the key problems that can result from relying too heavily on tracking
error and also outline our own risk management framework that has been developed over the past decade 0.1
in the process of managing a variety of global equity portfolios.
The problem with tracking error -0.1
The tracking error calculation itself relies upon a number of assumptions that are frequently violated in the -0.2
real world (i.e. returns are normally distributed, constant volatility and independence of returns through time) 1956 – 2009 1956 – 1988 1989 – 2007 2008 – 2009
-especially during periods of market stress which is precisely the time when risk estimates are most valuable.
Some risk model builders have acknowledged the limitations of their models in recent years and made Correlation (North American Universe)
adjustments with varying degrees of success without addressing the problems with the underlying building
Source: Schroders. Correlation estimates based on the North American Compustat Data (1955-2009) based on a
blocks of their risk estimates. This has only highlighted more clearly the many issues surrounding measurement composite measure of Value and Quality.
in practice through the building and estimation of a risk model (see the technical box in the Appendix).
4 RISK MANAGEMENT
Enhancing our approach to risk – more flexibility
To put it simply, a risk model based on the 1990s or early 2000s would have completely missed the strong Over the past decade, we have enhanced our approach to risk management in order to address a range of
diversification opportunity offered by Quality during the global financial crisis. The message here is clear – real world issues that we face when managing global equity portfolios. Crucially, we do not regard any one
managers need to take a more forward-looking view of risk within an intuitive fundamental framework. approach as sufficient but prefer instead to utilise a broad range of insights that together provide a more
all-encompassing picture of the risk in our portfolios.
Indeed, many of the risk factors used in even the most popular commercially available risk models are often
based on historic precedents for stock characteristics. For example, few risk models include a significant We have researched and developed our own global risk model which side-steps many of the issues
role for quality factors before the financial crisis although some introduced terms during 2008 in recognition mentioned earlier. For example, it permits us to stress-test the results by choosing different periods or risk
of its ‘newfound’ relevance to markets. A better approach would have been to incorporate a wider range of regimes, thereby supplying multiple estimates of risk across a range of environments (e.g. up/down markets,
risk factors in the first place. Such factors may have not been particularly significant over the prior decade or large cap dominated periods, Value markets etc). The range of estimates downplays the emphasis on a
so but they would have nevertheless made the model far more robust and effective. single tracking error number that may not be appropriate if market conditions change dramatically (see table
overleaf). As with our approach to researching stock selection, our risk model is not ‘over fitted’ to the past;
A second example would be ‘Momentum’ which often forms a part of many risk models in one way or another.
we also incorporate a wide range of factors that have not necessarily been important in the recent past but
The very different experience of this strategy in the recent past suggests that its volatility and correlation
which are, nevertheless, likely candidates to become key drivers of returns in certain market environments.
characteristics are not as predictable as a simple analysis of the prior decade might suggest (see chart
below). Once again, a longer term perspective would have highlighted several periods since the 1920s when Regime analysis on one of the QEP Core portfolios (including risk decomposition)
momentum strategies suffered significant reversals of a similar order of magnitude as that observed in 2009.
Predicted Risk Selected Dates All periods Up Markets Down Markets Value Growth Small Large
BARRA’s Global Momentum factor
Tracking Error 1.1% 1.2% 1.2% 1.2% 1.3% 1.2% 1.1% 1.1%
Beta 0.98 0.98 0.99 0.99 0.98 0.98 0.98 0.97
Absolute Fund Volatility 14.1% 12.7% 8.1% 13.3% 12.5% 12.9% 11.7% 15.5%
1.8 Benchmark Volatility 14.4% 12.9% 8.1% 13.3% 12.7% 13.2% 12.0% 15.9%
1.6 Stock Specific 48.2% 39.6% 42.1% 43.0% 36.9% 43.4% 46.6% 50.8%
Global Factors 1.1% 0.3% 0.0% 0.8% 0.3% 0.2% 0.2% 3.9%
Country 2.6% 1.7% 3.7% 3.0% 1.6% 3.2% 3.7% 4.2%
1.2 Industry Group 4.1% 2.8% 1.1% 4.9% 3.5% 1.5% 3.9% 4.6%
All Factors 35.9% 32.1% 33.1% 25.1% 30.2% 33.0% 37.0% 29.1%
1.0 – Value Factors 8.3% 8.3% 8.8% 7.5% 8.9% 6.4% 10.1% 3.0%
– Quality Factors 15.0% 12.9% 10.1% 15.4% 13.8% 11.2% 14.4% 13.5%
– Growth Factors 0.5% 0.2% 0.0% 0.3% 0.3% 0.0% 0.4% 0.5%
Dec 96 Dec 97 Dec 98 Dec 99 Dec 00 Dec 01 Dec 02 Dec 03 Dec 04 Dec 05 Dec 06 Dec 07 Dec 08
– Glamour Factors 0.2% 0.1% 0.0% 0.1% 0.1% 0.1% 0.2% 0.1%
Source: Schroders, MSCI Barra (Reference: Consumer Sentiment and the Momentum Factor Nov 2009). – Momentum Factors 1.2% 1.5% 1.1% 0.7% 1.5% 1.6% 0.7% 2.1%
*Momentum Cumulative Factor Returns, Index=1 in 12/96
– Size Factors 10.7% 9.1% 12.9% 1.0% 5.6% 13.6% 11.2% 9.8%
Source: QEP Global Core Fund as at end January 2010.
Thematic Industry Risk Management
Useful as this model is, our primary issue with tracking error is that it is purely concerned with short-term Core Decorative
relative volatility. Our view is that beta is more predictive and has a much broader relevance than its typical “Wider range allowed”
use as a measure of overall market sensitivity. The distinction between volatility and beta is important, Staples Retailing
Media Property & Casualty Insurance
particularly for more benchmark aware approaches, as managing beta involves a consideration of the thematic Real Estate Consumer Durables
influences that can impact upon index relative performance over time and across different market regimes. Health Eg Chemical
Aerospace Trading Co’s Food Products Life & Health Insurance Simple Bank
This encompasses a far broader concept of beta than its more traditional meaning of market sensitivity. Construction Containers Integrated Telcos
Apparel Asset Management
Electrical Equipment Hotels Restaurants & Leisure
Household Products Gas Utilities
The incorporation of a time dimension into risk management can make a significant difference at times when Beverages Road & Rail Retailing
risk factors exhibit inertia such as in narrow and trending markets. For example, being underweight the IT Servces Autos & Components
technology sector in early 1999 may not have seemed particularly risky in a short-term relative volatility sense Electric & Muli-Utilities Wireless Telecos
Integrated Oil & Gas Health Providers
but it would have led to a dramatic impact upon performance over the course of several months as these Construction Material Consumer Finance Paper & Forest Products Comms Equipment
stocks trended higher. The same argument could be applied to materials stocks more recently. By contrast, Software Steel & Aluminium Biotechnology
Complex Bank Refining
virtually all standard risk models assume that risk only plays out over a single period (typically a month, which Tobacco Air Freight & Logistics Fertilizers & Agri Chemicals
Internet & Catalog Retail
is then annualised to a yearly estimate). Over the years we have come to the conclusion that it is inadequate Marine Semiconductors Housebuilders
beta management that tends to destroy active returns over time rather than poor volatility management. Mining
Transport Infrastructure Oil & Gas Exploration & Production
What are the alternatives? – Applying a ‘real world’ perspective Energy Equipment & Services
Gold & Precious Metals “Risk manage”
1. Assumption free approaches. We monitor the live performance of our strategies on a daily basis Satellite Higher “Repeatability” Lower Thematic
across a range of recent market environments with a view to assessing how ‘typical’ recent performance
has been. Unusual behaviour, either at the fund level or within certain areas of the strategy, are quickly Source: Schroders. QEP propriety indicators.
identified and managed if necessary. Unlike tracking error, this approach is not based upon any statistical
assumptions (often called a non-parametric approach) and provides us with a more accurate estimate of 3. Simple risk metrics. We believe in also looking at ‘uncomplicated’ risk metrics such as ‘active share’.
short-term risk than a more traditional risk model. For example, our hit rate model indicated that the Active share is, essentially, a measure of the degree of overlap between a fund and its benchmark.
‘beta’ of Value was rising strongly during 2006 and 2007, far earlier than traditional models A number of 100% would represent no overlap, 0% would be an index fund and a number above
70% would be regarded as a high conviction portfolio with a strong tendency to deviate from the index.
2. Thematic risk management. We are very aware of the need to manage ‘top-down’ thematic risks,
One attractive feature of active share is that all fund managers can be assessed on the same statistic.
particularly those manifest in industry behaviour. We estimate the natural tendency of industries to behave
Moreover, the measure is assumption free and does not rely upon past correlations. Moreover, research
differently from the wider market and actively manage those that have a greater likelihood of deviating over
into the performance of 2,500 US equity retail funds over more than two decades has highlighted a much
a period of time (resources and materials industries are a good example). In our experience, it is the lack of
stronger correlation between active share and outperforming managers than that between tracking error
management of these ‘thematic’ risks that can lead to significant downside performance for managers. For
and subsequent performance (Cremers and Petajisto, 2009).
example, even initially small positions in mining stocks would have had a dramatic impact upon relative
returns over the past few years. In short, industries are an important tool for risk management and portfolio 4. Stock level diversification. We typically hold over 500 stocks in our global equity portfolios.
exposures should vary in proportion to the risk of each industry (see chart below). The benefit of this approach is far more than avoiding the permanent loss of capital that would occur
if an individual company went bankrupt. Within a global universe, breadth of ownership also helps to
maximise our exposure to opportunities and the likelihood of capturing our investment insights is greatly
enhanced when compared to more concentrated approaches which are more likely to suffer from
4 RISK MANAGEMENT
5. Fundamental diversification. For us, diversification is not limited to the stock level: we ensure that our Effective risk management cannot, in our view, be achieved by relying solely upon tracking error. It’s far too much
portfolios are highly diversified by geography, sector and theme. We also believe that the building blocks of an important job to be left to a single model and the industry’s over-reliance on managing tracking error
of our stock selection models: Value and Quality are more ‘fundamental’ in nature in that they offer strong has almost certainly left money on the table. Ironically, these forsaken opportunities have been particularly
strategic diversification over time. In particular, Quality as a strategy tends to perform very well during apparent in those investment processes that are most tightly managed with respect to a benchmark.
adverse market conditions when Value strategies can underperform. An exposure to both Value and
We believe instead in taking a more holistic view of risk, using a variety of tools to monitor and manage risk
Quality within our portfolios plus a wide toolkit coupled with on-going research allows us to have
in its many forms. Truly good portfolio constructers understand the interplay of both return and risk and how
confidence that we have a framework in place which allows us to maximize our chance to perform across
this evolves over time. It is a skill that has been neglected in our opinion.
a broad range of market environments.
6. Actively manage the outputs not the inputs. This requires a more multi-faceted approach than simply
targeting a single output variable such as tracking error which often conceals more than it reveals and
encourages overconfidence. Investors often impose very tight index relative constraints on their portfolios in Assumptions underlying tracking error calculation are violated in the real world.
order to manage tracking error but, in our view, this is a very inefficient way to manage risk and can
endanger returns. We believe a more appropriate response is to work within the investment process and Assumption One: Normally distributed returns
understand the array of opportunities offered at any moment in time on a risk adjusted basis. Implies that negative stock movements are just as likely as positive stock movements and that extreme
market events should rarely occur (e.g. September/October 2008, the crash of 1987 and many others).
Perhaps the best manifestation of the success of this framework is found in the strong track record of our Whilst we ‘should’ observe global market movements of greater than 4% a day, twice every century, in
Global Core portfolios over the past decade, which has performed very consistently across a broad range of practice this has occurred over 100 times since 1900. This non-normal return structure is not new and has
market environments (see chart below). been known since the ground breaking work of Mandelbrot (1963) and Fama (1965) who showed that stock
prices followed a non-normal distribution which exhibited fat-tails and negative skew. Downside risk is
Win rates across various market regimes (QEP Global Core, 2000-2010) therefore understated by tracking error as a normal distribution is not necessarily a good approximation to
100% the occurrence of risk events, particularly over shorter time periods which need to incorporate non-
symmetric measures that are less sensitive to such rigid and potentially unrealistic assumptions.
75% Assumption Two: Independent of returns through time
75% Implies that we should not observe trends in the market other than those created by random events.
66% 67% 64% 67% 65%
However, there is a wealth of evidence supporting the prevalence of momentum in financial markets, e.g. De
Bondt and Thaler (1985, 1987), Jagedeesh and Titman (1993, 2002). The phenomena is frequently manifest
50% in a particular theme such as the recent strong momentum in emerging markets or materials or negative
momentum in financials during 2008. Momentum can lead to annual tracking error estimates being
significantly understated when funds have exposure to these themes. This was particularly apparent in 1999
25% when tracking error estimates understated risk by a factor of three due to the strong momentum effects
evident in technology stocks at the time (Scowcroft and Sefton, 2001).
Assumption Three: Constant volatility through time
0% Implies that volatility or variance is constant through time (although increasingly commercially available risk
Overall Rising Falling Value Growth Large Small models are attempting to correct for this). Again there is a wealth of evidence on the time varying nature of
Source: Schroders. Performance is calculated in accordance with Global Investment Performance Standards. Schroder stock market volatility (Bollerslev, 1987, Bollerslev & Engle, 1993 and reviews by Bollerslev, et. al. 1992, 1994
QEP Global Core Composite relative to MSCI World NDR, inception 31 January 2000 with performance to 31 December and Bollerslev, 2001). The violation of this assumption can cause Risk Models to substantially underestimate
2009. Market environment (Rising, Falling, Value, Growth, Large and Small) defined using MSCI indices.
tracking error if markets head into a higher volatility environment (Scowcroft and Sefton, 2001). The
converse is of course also true. Again at Schroders we believe that we need to model tracking error in
various market environments to deal with this
Technical box: Risk models References
There are two broad types of ‘standard’ risk model, cross sectional and time series based but both require
an estimation of a covariance matrix (volatility and correlation) and an estimate of individual stock specific Ang, A., and J. Chen, 2002, Asymmetric Correlation of Equity Portfolios, Journal of Financial Economics, 443-494.
risks. Typically, the covariance matrix is ‘shrunk’ so that it is more tractable and a subset of style factors, Bollerslev, T., 1987, ‘A Conditional Heteroskedastic Time Series Model for Speculative Prices and Rates of Return, Review
countries and industries are used to proxy risk in the market. Time series risk models do not prescribe what of Economics and Statistics, 69, 542-547.
these factors are in advance but the danger with this approach is that the factors may not be intuitive Bollerslev, T., 2001, ‘Financial econometrics: Past developments and future challenges’ Journal of Econometrics, 41-51.
because there is no economic intuition behind the selection of the factors. Using a risk model therefore Bollerslev, T., and R. Engle, 1993, ‘Common persistence in conditional variances’ Econometrica, 166-187.
leads to several issues when estimating ex-ante tracking error. Bollerslev, T., R. Chou, and K. Kroner, 1992, ‘ARCH modeling in finance.’ Journal of Econometrics, 5-59
Bollerslev, T., R. Engle, and D. Nelson, 1994, ‘ARCH models.’ In: Engle and McFadden (Eds.) Handbook of Econometrics
– As each model uses different methods to estimate the covariance matrix, the same portfolio would
Vol IV. North Holland, 2959-3038.
almost certainly generate different estimates of tracking error – there is no one right answer.
De Bondt, W., and R. Thaler, 1985, ‘Does the stock market overreact?’, Journal of Finance, 793-805
– Risk models that use factor models assume that each factor is independent which implies that
De Bondt, W., and R. Thaler, 1987, ‘Further evidence of investor overreaction and stock market seasonality’, Journal of
correlations fully explain co-movements between factors and that it is unlikely for these correlations to Finance, 557-581
vary over time. However, we regularly see this assumption breaking down in financial markets, Estrada, J., 2008, ‘Black Swans and Market Timing: How not to generate Alpha’, Journal of Investing, Fall. 20-34
particularly instability in correlations (Ang & Chen, 2002, Patten 2004).
Fama, E., 1965, ‘The Behaviour of Stock Market Prices,’ Journal of Business, 34-105.
– Further, ex-ante tracking error assumes that current portfolio weights will not change over the time period Jagedeesh, N., and S. Titman, 1993, ‘Returns to Buying Winners and Selling Losers: Implications for Stock Market
that tracking error is calculated. However, portfolio weights are constantly changing in line with price Efficiency’, Journal of Finance, 65-91.
movements (as well as the benchmark weights) which will lead to ex-ante tracking error being different to Jagedeesh, N., and S. Titman, 2002, ‘Cross-Sectional and Time-Series Determinants of Momentum Returns, Review of
ex-post tracking error as shown by Satchell & Hwang (2001) and Lawton-Browne (2001). Financial Studies, 143-157.
Lawton-Browne, C., 2001, ‘An alternative calculation of tracking error,’ Journal of Asset Management, 223-234.
Mandelbrot, B., 1963, ‘The Variation of Certain Speculative Prices’, Journal of Business, 394-419.
Patten, A., 2004, ‘On the Out-Of-Sample Importance of Skewness and Asymmetric Dependence for Asset Allocation’,
Journal of Financial Econometrics, 130-168.
Satchell, S., and S. Hwang, 2001, ‘Tracking error: Ex ante versus ex post measures,’ Journal of Asset Management,
Scowcroft, A., and J. Sefton, 2001, ‘Do tracking errors reliably estimate portfolio risk?’ Journal of Asset Management, Dec,
Cremers, K and Petajisto, A, 2009, ‘How Active Is Your Fund Manager? A New Measure That Predicts Performance’
The views and opinions contained herein are those of the QEP Team and may not necessarily
represent views expressed or reflected in other Schroders communications, strategies or funds.
4 RISK MANAGEMENT
4.2 The Cyclical Framework: Cycling Around Asset Allocation
Harish Vekaria Quantitative Analyst, Multi-Asset
Introduction The Cyclical Framework… Recovery on track
– In this article, we provide an update on the development and drivers of the recovery phase. Cyclical Framework: An overview
– We also introduce our global earnings and Institute of Supply Management (ISM) cycles, which we have We define the economic cycle in terms of the position of the US economy relative to its long-term trend.
incorporated into the cyclical framework. We discuss the recent developments in these cycles and what This enables us to specify four distinct phases depending on where the economy is in relation to the output
their signals mean from an asset allocation perspective. Our backtests show that when they are gap and whether growth is above or below its trend.
combined appropriately with the economic cycle, they have resulted in better asset allocation decisions.
Each phase is characterised by different behaviour in interest rates, inflation and activity.
Summary Model illustration: Four distinct phases and the output gap
– Our economic cycle moved into the ‘recovery’ phase Schroders Economic Cycle Slowdown Recession Recovery Expansion
in mid-April 2010 and the path of recovery is moving
in the right direction, but we expect this phase to be Output below trend, Output below trend,
longer than normal. Growth decelerating, Growth accelerating,
Inflation falling Inflation falling
– The drivers of the economic cycle are so far intact
with a low inflation environment, loose policy stance
(reinforced by quantitative easing or ‘QE’) and positive Recovery Expansion
earnings growth contributing to an improvement in
Output above trend, Output above trend,
equity market performance. Growth decelerating, Growth accelerating,
– Our newly-introduced global earnings cycle indicates Inflation rising Inflation rising
that we are in ‘recovery’ mode, and this started in
March 2010. Moreover, the pace of recovery in
Recession Slowdown Output
earnings has been strong.
– The ISM cycle, based on the US ISM manufacturing Source: Schroders, for illustration only.
survey, remains above the 50 level. Our backtest
indicates that we should only be cautious on risk
assets when the ISM falls below 50 and is decelerating.
– Meanwhile, there is a strong correlation between the
improvement in M&A (mergers and acquisition) Source: Schroders, for illustration only.
activity and the outperformance of equities over
corporate bonds. This corresponds to the asset
allocation view suggested by our economic cycle.
Where are we in the cycle? The length of the cycle
We transited into the ‘recovery’ phase in mid-April 2010. The last time we were in this phase was between In general, the average length of a completed economic cycle is between 5-6 years. The last completed
October 2003 and April 2006. cycle seems to have been 6.5 years long, resulting from a longer than expected recovery phase, as well as
the deepest recession since 1950.
The trend of the current recovery phase has been gradual, rather than the usual ‘V’-shaped recovery.
This is in line with our baseline view of steady growth. Table 1: Details of full cycle and transition;
Chart 1: The Schroders Economic Cycle Model (The model is calibrated on US monthly data back to 1950)
Transition to Date of transition Length (in years)
Percentage of GDP (%)
Recovery October 2003 2.6
Expansion May 2006 1.4
Slowdown October 2007 0.7
Recession June 2008 1.8
-2 Length of complete cycle 6.5
-4 Start of new cycle
-6 Recovery April 2010
-8 Source: Schroders, Datastream.
80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 What drives the cycle and market performance?
Recovery Recession Slowdown Expansion Estimated US Output Gap Baseline forecast
Behaviour of the economy in different phases of the cycle
Source: Schroders, Datastream. We find that each phase is characterised by different behaviour in economic factors. As a result, investors
focus on different factors at different times, in order to build a picture of potential drivers of markets in each
stage of the cycle.
The length of the ‘recovery’ phase
The approximate average length of the recovery phase, based on data since 1950, is 17 months.
Table 2: Drivers of the cycle
However, a deeper ‘recession’ in general has meant a longer ‘recovery’ phase, such as the 1980s episode.
For this cycle, not only has the recession been deeper, but so far, the recovery has been slower than usual. Slowdown Recession Recovery Expansion
This in turn increases the risk of a double-dip or mid-cycle pause scenario. Output level Above Below Below Above
Growth Trend downwards Trend downwards Trend upwards Trend upwards
Inflation Rising/High Falling Falling/Low Rising
Policy Stance Tight Loosening Loose Tightening
Interest rates Stable Falling Stable Rising
Equity Fundamentals Earnings/Valuations Valuations Earnings Earnings
4 RISK MANAGEMENT
The economic cycle and the interest rate regime
Historically, a sustainable recovery is usually accompanied by a low inflation environment and a
relatively loose/ stable policy stance. As a result, positive growth in earnings drives equity market returns. Where are we?
The re-rating of markets normally occurs in the ‘recession’ phase, which has already taken place. We are currently in the recovery phase where monetary policy is loose and US interest rates are as low as
they can go. As it is not possible to cut rates below 0%, quantitative easing has been employed, which in
What drives equity market returns? reality reinforces a falling interest rate regime.
We gain insights into the drivers of the equity markets by breaking down equity returns into changes in
earnings per share and changes in the ratings of those earnings as measured by the price-earnings ratio Chart 2: Stage of the cycle and US interest rate regime
(PE). Please note that this is labelled ‘valuations’ in the table.
% occurrence of different interest rates ragime
Table 3: Financial markets and the economic cycle 100%
Light colour refers to % of times Dark colour refers to % of
Equity market performance is strong in the recovery phase with a Sharpe ratio of 0.63. 80% when rates remained unchanged times policy rates actually changed
Assets/Phase Statistics Recovery Expansion Slowdown Recession 60%
Equity Excess return 1
8.6 8.4 -8.4 7.5 40%
Volatility 13.7 12.4 17.4 16.2
% PE Monthly Average -0.6 -0.2 -0.4 1.7
% EPS 1.4 1.2 -0.1 -0.9 Raising Falling Rising Falling Rising Falling Rising Falling
Trajectory Trajectory Trajectory Trajectory Trajectory Trajectory Trajectory Trajectory
Success rate2 100% 83% 22% 80%
Recovery Expansion Slowdown Recession
Average length of the phase Months 17 15 9 15
We are here Interest rate and the economic cycle
Note: Rising trajectory occurs when rates are on a course to rise from trough to peak in the interest rates cycle whilst falling
In this phase, growth in earnings is trajectory occurs when rates are on a course to fall from peak to trough. Rising and falling trajectory identified by
a key driver of equity performance comparing current month to previous month of US Federal Reserve rate policy.
Source: Schroders, Datastream. Data (1950-2010).
As mentioned previously, the re-rating of the PE tends to drive market returns in the ‘recession’ phase. In
contrast, growth in earnings drives equity gains in recovery as, on average, markets tend to de-rate slightly.
What to expect?
We found that during the recovery phase, rates were on a falling trajectory (mostly unchanged) 42% of the
time. Meanwhile, rates were on a rising trajectory (mostly rising) over the remaining 58% of the time.
This suggests that we should expect rate rises around the mid recovery path.
1 S&P 500 TR (1950-2010), excess over cash.
2 Success rate measured as number of times equity market outperformed cash at the complete recovery phase, based on
nine cycles since 1950. 79
When to expect rates rises in the recovery phase? Chart 4: Equity market performance by stage and interest rate cycle
Examining previous transitions from recession to the recovery phase, we found that, on average, 20.18
interest rates tend to start rising after the tenth month3 (i.e. February 2010 for the current phase).
Thereafter, we see a gradual increase in interest rates. 12.64
Chart 3: Examining the previous recovery phase and interest rate policy
Fed funds target rate at start of recession*
23% Rates Falling
Dates refer to begining of 8.53
the “Recovery” phase Rates Rising
Recovery Expansion Slowdown Recession
8% 1983 (Stagflation) (Disinflation)
3% 1972 1961
1992 Past episodes of the recovery phase compared
Baseline forecast (End 2011) Looking at previous recovery phases, the US equity market has tended to outperform cash by around 10%
-2% April 2010 Expect 1st rate rise in Q1 2012 by the end of the phase, on average.
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40 42 44 46 48 50 52 54 56 58 60
Subsequent months into the recovery phase Chart 5: Performance in the recovery phase
*Recession and recovery definition based on Schroders economic cycle. Dotted line refers to the ‘Recession’ phase and Performance index of the S&P500 excess of cash
solid line refers to ‘recovery’. Average length
In the recovery phase, equities perform well in both rate falling and rate rising regimes. Cumulative performance of
110 the average recovery phase
Our analysis shows that in an environment of falling rates, markets produce their best returns compared to
the periods of rising rates. However, their historical returns are still strong compared to other environments 105
where rates are on a rising trajectory.
95 Cumulative performance of the current
phase (updated as to 18 November
90 2010) (First signal April 2010)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Subsequent months into the recovery phase
80 3 Average based on seven cycles where rates remained stable as we switched into the recovery phase.
4 RISK MANAGEMENT
The current phase had an extraordinary start in terms of equity market performance. Equities fell 10% in the Chart 7: Earnings growth in the recovery phase is key
first three months at the start of the recovery phase, as they started pricing in a double-dip. This concern
S&P 500 Earning growth MoM (indexed=100)
appears to have eased given the recovery in markets over recent months.
What’s driving the performance? 150 2003 1992 1976
Chart 6: Equity market performance 2010
Cumulative S&P 500 return excess of cash
140 1958 1961
120 1971 1958 1992 1996
110 1976 80
1996 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Subsequent months into the recovery phase
90 Earnings growth from the beginning of the current recovery phase has remained robust, similar to the levels
1961 seen in 2003.
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 Chart 8: P/E multiples in the recovery phase
Subsequent months into the recovery phase S&P 500 Expansion MoM (indexed=100)
In the two of the nine recovery phases, there was a correction in the equity market at the mid recovery 140
stage. However, by the end of the recovery phase, equities had outperformed cash. 130
60 2010 2003
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30
Subsequent months into the recovery phase
Updated: 18 November 2010.
P/E multiples are in de-rating mode, as seen in 2003. Some de-rating in valuation is normal in this phase as
long as there is support from earnings growth.
The earnings & ISM cycle Chart 10: ISM cycle (The model is calibrated on monthly data back to 1970)
ISM Purchasing managers index (smoothed EWMA)
Chart 9: The global earnings cycle
(The model is calibrated using MSCI World EPS on monthly data back to 1970) Our analysis suggests that
an ISM above 50 is less of
% Differential between real EPS and a 20 year trend a concern for risk assets
20 ISM falling below 50 would
40 be a concern for risk assets
as probability of the economic
0 cycle moving into recssion phase
will be high
80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10
-40 Earning recovery
is ‘V’-shaped Recovery Expansion Slowdown Recession
Source: Schroders, Datastream.
80 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10
Recovery Expansion Slowdown Recession We perform a similar exercise using a smoothed version of the ISM Purchasing Managers’ Index that is
widely seen as a good leading indicator of US economic activity. We classify the movement of the ISM into
Source: Schroders, RIMES, MSCI World data.
four categories and examine how different asset classes have performed in each stage since 1970.
We define the earnings cycle with reference to the growth in global real earnings per share (EPS) adjusted Where are we?
for its long-run trend (i.e. rolling twenty-year history). This enables us to specify four distinct states or phases The ISM cycle suggests that we have returned to a state of ‘expansion’ after four months of slowdown in activity.
in relation to the earnings cycle.
Our analysis indicates that when the ISM is in the ‘recovery’ or ‘expansion’ phase the model tends to favour
Where are we? risk assets, in particular equity, high yield debt and commodities. During the ‘slowdown’ phase, the direction
The current signal from the model suggests that we are in the ‘recovery’ phase, which started in March of the performance of these assets is less clear with positive commodity performance and equities
2010. Since then, global corporate earnings have recovered strongly in a ‘V’-shaped fashion, such that they delivering relatively flat returns (see table 2 in the appendix). However, when the ISM falls below 50 and is
are now almost halfway below their 20-year trend. decelerating (i.e. in ‘recession’ mode), risk assets – particularly commodities – underperform significantly
while all assets enter a phase of high volatility.
What does this mean in terms of asset allocation?
Our analysis suggests that the current environment is favourable for equities and high yield bonds (see table This is a short- term model that generally has multiple cycles during one business cycle.
1 in the appendix). Moving to the expansion phase indicates a tilt towards an increase in equity exposure
and commodity assets would also be favoured.
This is a medium-term model and we do not expect the transition between the phases to be frequent.
4 RISK MANAGEMENT
The Cyclical Framework & M&A activity Appendix:
Recent developments in the mergers & acquisition activity The Cyclical Framework and top-down asset allocation
Chart 11 has been included to draw a comparison between M&A activity and the cyclical framework. It Asset allocation and the economic cycle
shows that the correlation between the M&A activity trend and the relative performance of equities versus
corporate bonds tends to be strong. Adjusting portfolio exposure for the recovery phase: We find that asset performance varies in each
stages of the economic cycle and successful active allocation can be achieved by:
At the end of 1990 and 2001 recessions, M&A activity troughed and showed signs of improvement as we
moved into the recovery phase (around 1992 and 2003 respectively). – Identifying the current stage of the cycle
– Increasing (decreasing) exposure to the assets expected to outperform (underperform) in the current stage
This time, the outcome looks similar in that we are in the recovery phase and there are some signs of M&A
activity improving. As we move into the recovery phase, equity markets become the top pick from a broad range of asset
classes. Property also performs well as investors seek real assets to play the recovery. Nominal assets such
Favour equity over corporate bonds… as bonds tend to underperform.
If this is the trough in M&A activity, we are likely to see equities outperform corporate bonds. This in turn
reaffirms the asset allocation view suggested by our economic cycle to favour equities over corporate bonds Slowdown Recession Recovery Expansion
in the recovery phase.
Output below trend, Output below trend,
Chart 11: The rise in M&A activity correlates with outperformance of equity over corporate bonds Growth decelerating, Growth accelerating,
Inflation falling Inflation falling
280 Output above trend, Output above trend,
1.4% Growth decelerating, Growth accelerating,
1.2% Inflation rising Inflation rising
130 0.4% Asset performance ranked by preference and stage of the cycle.
0.2% TIPS Credit bonds Equity Private Equity
1990 Recession 2001 Recession 2008 Recession Commodity Hedge Funds Property Commodities
90 92 94 96 98 00 02 04 06 08 10 Government Equity Hedge Funds Property
Bonds Government Private Equity Hedge Funds
% M&A activity* (RHS) US equity V’s corporate bond relative return (LHS)
Cash Bonds High yield EMD
**M&A activity measured as six-month moving average of the monthly value of deals in US$ as a percentage of market value. Infrastructure Infrastructure Equity
Source: Datastream, Schroders. Updated December 2010.
Other sub-asset classes
Value stocks, Financials
Large Cap, Defensive Sectors Small Cap, Cyclical Sectors Growth stocks, Cyclical Sectors
& Energy Sector
Energy/Precious commodity Energy/Precious commodity Industrial commodity Industrials/Energy commodity
Data: S&P 500, GSCI commodity, CSFB hedge fund, US 10 yr Government bonds, UK IPD property, Barclays (ML) credit
bonds, US DS Sectors, LPX50 Private equity index, MSCI World Infrastructure index. Source: Schroders, Datastream.
Table 4: Asset performance in different phases of the earnings cycle (1970-2009) Table 5: Asset performance in different phases of ISM cycle (1970-2009)
Assets/Phase Statistics Recovery Expansion Slowdown Recession Assets/Phase Statistics Recovery Expansion Slowdown Recession
Equity Excess return 4
8.5 14.0 -15.7 7.2 Equity Excess return 4
17.7 4.3 -0.6 -2.6
Volatility 13.0 11.6 19.2 16.9 Volatility 15.2 14.8 14.3 19.4
% PE Monthly Average -0.7 0.1 0.8 3.6 % PE Monthly Average 3.2 -0.5 -0.7 2.6
% EPS 1.6 1.6 -1.3 -0.7 % EPS -1.2 1.4 1.1 -1.4
Government Bonds Excess return 2.2 0.0 0.9 4.7 Government Bonds Excess return 1.4 -0.1 4.0 1.2
Volatility 8.0 8.0 10.4 8.5 Volatility 8.5 7.5 7.0 11.8
High Yield Excess return 10.0 3.4 -5.8 12.4 High Yield Excess return 13.6 4.8 3.1 0.4
Volatility 6.9 4.5 14.8 9.8 Volatility 8.7 6.2 6.6 17.3
Investment Grade Excess return 4.1 0.3 -1.7 9.4 Investment Grade Excess return 6.5 1.1 2.0 1.4
Volatility 6.4 7.5 8.5 7.2 Volatility 7.9 5.1 6.8 11.7
Commodity Excess return 2.2 10.6 -17.9 9.1 Commodity Excess return 10.9 9.2 9.4 -17.3
Volatility 14.2 18.5 24.3 14.9 Volatility 18.4 14.9 21.1 23.5
Note: We compare performance of the asset classes in each stage based on the Sharpe ratio. Source: Schroders, Datastream.
Data: Equity S&P 500, Government bond US 10 Yr, High Yield = Barclays HY, Investment Grade = Barclays corporate, Data: Equity S&P 500, Government bond US 10 Yr, High Yield = Barclays HY, Investment Grade = Barclays corporate,
Commodity = S&P GSCI Index. Commodity = S&P GSCI Index.
How we define the phases:
ISM below 50 and decelerating: Recession
ISM below 50 and accelerating: Recovery
ISM above 50 and accelerating: Expansion
ISM above 50 and decelerating: Slowdown
The views and opinions contained herein are those of Harish Vekaria and may not necessarily
represent views expressed or reflected in other Schroders communications, strategies or funds.
84 4 Excess of cash return. In US$.
4 RISK MANAGEMENT
4.3 Lies, damn lies and (risk) statistics for pension funds
Anthony Earnshaw LDI (Liability Driven Investment) Fund Manager
Due to the effects of changes in market values on both assets and liabilities, a Scheme’s funding level will “Prediction is very difficult, especially about the future”
change over time. Funding level risk management involves measuring the magnitude of potential changes
in the financial health of a Scheme. These estimates play an important part in Trustees’ decision making Inevitably a key input into any funding risk model is the volatility of the individual components that determine
process, so the quality of these estimates is key. the Scheme’s funding level. The issue therefore is how should we estimate this volatility? Here we take
equities as an example and look at two potential sources of estimates:
Most managers and consultants have a risk model and use the results to inform Trustees’ decision-making
process. However, the numbers coming out of a model are only as good as the information put in! This article – Option-implied volatility
examines the effect of different inputs into a risk model and shows how different assumptions can lead to vastly – Historic volatility.
different estimates. In essence Trustees should be as interested in the inputs as the outputs of the models and
should never forget that “Assumptions are the mother of all…”, I think you know the rest! One common method of sourcing a volatility estimate is to use the implied volatility that can be derived from
an option price. The benefit of this measure is that it shows the implied market forecast of future risk and as
a result is a forward looking measure. The implication is that this is better than an estimate based on historic
“If you risk nothing then you risk everything” pricing. However, as the chart below shows the volatility implied by option prices is actually driven more by
Before rushing headlong into an examination of risk measurement and management, it is worth reminding the amount investors are prepared to pay for the protection an option strategy gives them than by a best
ourselves what risk is. It is a measure of the likelihood of an outcome. This is often quantified in terms of guess of the future risk of the underlying asset.
percentages or a likelihood of an outcome occurring one in so many years. For the purposes of this article
we will use a ‘Value at Risk’ (VaR) measure. This shows the potential size of deterioration in the funding level1 Historic rolling 10 year volatility of S&P 500 (annualised monthly data) and option-implied volatility
that may occur once in every 20 years (i.e. there is a 1-in-20 or 5% chance that the funding level deteriorates 60%
by more than this in any one year). Calculating the effect of different strategies on the funding risk allows
Trustees, the Sponsor and their Advisors to choose an appropriate mixture of growth and LDI assets that 50%
will generate both a rate of excess return over the liabilities and an acceptable level of funding risk.
“Never assume the obvious is true” 30%
To demonstrate the power of assumptions we will examine how a risk estimate is produced.
To estimate a funding level risk estimate three key elements are required:
– an understanding of how the value of the liabilities is estimated 10%
– the assets (and potential assets) held by the Scheme
– a risk model that can estimate the likely performance of the assets relative to their liabilities in the future
There are myriad ways to generate a risk estimate. However the key inputs into any risk model are essentially
Historic Volatility Implied Volatility
the same – volatility and correlation.
Source: Schroders. For illustration purposes only.
1 It does not take into account changes in this ratio due to changes in the Scheme Actuary’s estimates
(e.g. with regard to longevity). 85
The historic relationship between liabilities and growth assets
As the chart shows, 10 year rolling realised volatility has remained very stable over the period. However, Correlation
option based implied volatility can experience short term spikes in times of market distress, perhaps around 50%
30% p.a. It is unlikely that the heightened volatility experienced over the past two years will persist for the
next 10 years – i.e. a similar outcome to the Great Depression. The 1930s saw unemployment of 25% in the
US and the S&P 500 fell over 85% from peak to trough. Whilst deep (and potentially prolonged), the current
recession is unlikely to be this severe.
“The quality of your life is the quality of your relationships”
When armed with some estimates of asset and liability volatility we also need to estimate the nature of the -25%
relationships between each of these components. The relationship between two variables is often quantified
through the use of ‘correlation’. This number shows the direction and strength of the relationship between
two sets of returns. It can vary between -1 and 1. A relatively large absolute number indicates a strong -50%
relationship whilst a number close to zero indicates that the two variables don’t appear to have any
Liabilities versus diversified growth assets
The power of the correlation numbers used in a risk model lies in the effect different correlation numbers will
have on a portfolio of two assets. Intuitively we already know that diversification can reduce the overall risk
Source: Schroders. Please see notes on risk modelling.
of a portfolio. This occurs when two assets have a relationship that is imperfectly correlated so that at times
the performance of one asset ‘offsets’ the performance of the other. The table below illustrates how a
different correlation estimate affects the overall risk of an equally weighted portfolio of two assets. The graph shows that for most of this period the correlation between the liabilities and growth assets has
been close to zero or negative. However, as a result of the extreme (and probably anomalous) market
The effect of correlation on portfolio risk movements during the credit crunch, correlations have become positive. This instability highlights how
important the use of skill, judgement and experience (or perhaps just a healthy dose of common sense!)
Asset 1 Risk Asset 2 Risk Correlation between assets Portfolio Risk should be applied when selecting appropriate inputs to a risk model.
25% 30% -1.0 3% One way to overcome this ‘over-reliance’ on one set of correlation and volatility assumptions is to generate
25% 30% -0.5 14% a range of different volatility and correlation estimates.
25% 30% 0.0 20%
“There is no such thing as failure, there are only results”
25% 30% 0.5 24%
Risk estimates are heavily dependent on the volatility and correlation assumptions used. To illustrate this
25% 30% 1.0 28% point we have taken a ‘typical’ Scheme and generated a set of risk estimates by applying a range of volatility
and growth assumptions. To do this we have used a single risk model and generated:
Source: Schroders. For illustration purposes only.
– A worst case2 estimate (i.e. a set of assumptions that will create a high risk number)
What is immediately apparent is the enormous influence correlation (or diversification) can have on portfolio – A best case estimate (i.e. a set of assumptions that will create a low risk number)
risk. An added complication is that the relationship between different variables is not constant over time.
To illustrate this point the chart below shows the rolling 5 year correlations between a set of liabilities and – Blend (i.e. some – ‘reasonable’ assumptions in between the best and worst cases).
a portfolio of diversified growth assets.
2 The worst case scenario represents the highest risk number given the assumptions we have used. The change in the funding
86 level could be more extreme than the numbers generated. Please see notes on risk modelling at the end of the document.
4 RISK MANAGEMENT
The graph below shows the effect of the different assumptions on a Scheme’s risk estimate at different “It’s more fun to arrive at a conclusion than to justify it”
levels of liability coverage.
Understandably Trustees are concerned with the magnitude of future changes in their Scheme’s funding level.
The effect of LDI coverage on funding level risk
However, how can Trustees quantify how far their funding level could fall? We have shown that risk modelling
Funding Level VaR can help. But we have also highlighted that there are several ‘risks’ to risk modelling:
– We can manufacture just about any desired risk number by adjusting our assumptions
30.0% – Too much emphasis can be placed on a risk number without understanding that it is actually a mid-point
estimate and has some (very) wide bands of error around it
– Focusing on the estimate without also considering whether the inputs into the model are reasonable can
Should trustees lose all hope? Risk modelling can have its enormous benefits. The estimates are a very
15.0% useful guide as long as they are used with the appropriate care. The risks highlighted above can be
mitigated by the following measures:
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% – The generation of a range of likely risk outcomes to accompany a ‘reasonable’ estimate
Liability Coverage – The publication of the volatility and correlation figures that have been used to generate the estimate
Best case/Worst case funnel Blend As well as the method used to generate these estimates. This gives the user the chance to gauge
Source: Schroders. Please see notes on risk modelling. whether the key drivers of the estimate are sensible.
Different risk models will inevitably generate different results and this article has demonstrated the effects
From the previous chart we can see that, irrespective of how much liability coverage the Scheme chooses,
of different assumptions on the estimates of the Scheme’s funding level risk. It has also armed readers
both its starting risk and the benefit of adding liability coverage depends on the on the risk and correlation
with an understanding of the pitfalls awaiting innocent users of “Lies, damned lies and (risk) statistics.”
assumptions applied. The ‘funnel’ between the best case and worse case estimates shows the magnitude
of the impact of the assumptions. Apologies to Disraeli et al.
In the table below we highlight some key differences between the funding level risk estimates and results
given the levels of liability coverage. These risk estimates include the Scheme’s growth assets: Notes on risk modelling
Risk estimates are from our internal modelling based on the data provided and the assumptions stated.
The assumptions and risk estimates of the Scheme
The risk forecasts are the result of statistical modelling, based on a number of assumptions. There is no
assurance or guarantee that the forecast will be achieved and it should not be considered as a prediction
Assumption/Liability risk coverage (%) Best case Blend Worst case
of actual returns that may be realised in the future from the portfolio. Assumptions may change materially
Growth asset volatility 11.5% 13% 16.5% with changes in underlying assumptions that may occur, among other things, as economic and market
Liability volatility 12% 12% 12% conditions change. Risk numbers exclude other non-investment risks (such as longevity).
Growth/liability correlation 35% -10% -35%
Risk when liability coverage is: The views and opinions contained herein are those of Anthony Earnshaw and may not necessarily
0% 19% 24% 31% represent views expressed or reflected in other Schroders communications, strategies or funds.
50% 13% 17% 23%
80% 12% 16% 20%
100% 12% 14% 18%
Source: Schroders. Please see notes on risk modelling.
SCHRODERS CHaRitiES invEStmEnt PERSPECtivES 2011
Important Information: This document is intended to be for information purposes only and it is not intended as promotional material in any respect. The material is not intended as an offer or solicitation for the purchase
or sale of any financial instrument. The material is not intended to provide, and should not be relied on for, accounting, legal or tax advice, or investment recommendations. Information herein is believed to be reliable but
Schroder & Co Limited (Schroders) does not warrant its completeness or accuracy. No responsibility can be accepted for errors of fact or opinion. This does not exclude or restrict any duty or liability that Schroders has to
its customers under the Financial Services and Markets Act 2000 (as amended from time to time) or any other regulatory system. Reliance should not be placed on the views and information in the document when taking
individual investment and/or strategic decisions. The forecasts included in this document should not be relied upon, are not guaranteed and are provided only as at date of issue. Our forecasts are based on our own
assumptions which may change. We accept no responsibility for any errors of fact or opinion and assume no obligation to provide you with any changes to our assumptions or forecasts. Issued by Schroder & Co Limited,
100 Wood Street, London, EC2V 7ER, which is authorised and regulated by the Financial Services Authority. For your security, communications may be taped or monitored. w35636