On the Provision of Incentives to Save by jolinmilioncherie


									       On the Provision of Incentives to Save

                                    Philip R. Lane*

    Trinity College Dublin and Centre for Economic Policy Research

                                           January 2001

  Philip R. Lane is Associate Professor of Economics and Fellow at Trinity College Dublin and a Research
Affiliate of the Centre for Economic Policy Research (UK). He holds a PhD in economics from Harvard
University and was previously Assistant Professor of Economics and International Affairs at Columbia
University. His research is in the field of international macroeconomics and he has published widely in top-
ranked journals. He is on the editorial boards of the Journal of International Economics, International Tax
and Public Finance and Economics and Politics. He has acted as a consultant to the International Monetary
Fund and the European Commission and is a research affiliate for the London-based Centre for Economic
Policy Research. He is the lead coordinator of an EU Research Training Network "Analysis of International
Capital Markets: Understanding Europe's Role in the World Economy".The views expressed in this paper
are personal. I was invited to prepare this paper by the Irish Association of Investment Managers (IAIM).
Executive Summary

This paper argues the importance of government intervention in encouraging, through the
taxation system, increased levels of medium-term household savings and a diversification
of household portfolios to include higher-return equity and bond instruments. In addition
to the important goal of saving for retirement, financial prudence requires that households
accumulate sufficient medium-term assets to guard against negative events (e.g. job loss
or family illness) and to achieve important life goals (e.g. education of children or a
deposit for a home). Recent economic theory emphasises that a host of factors can induce
households to under-save and that policy intervention is required to ensure an adequate
level of saving.

Moreover, in pursuing medium-term financial goals, it is generally recommended that
savings be invested in a mixed portfolio that includes higher-return equity and bond
instruments. Such financial diversification can also offer protection against the risk of a
downturn in the domestic economy that would hit labour incomes and property values.
Accordingly, government policy should not discriminate against savings plans that allow
for investment in equities, bonds and other financial instruments by favouring interest-
bearing deposit accounts over alternative investment strategies. Indeed, it is interesting to
note that the UK government goes further by actually offering positive incentives for
investment in equity portfolios relative to cash accounts.

In addition to these efficiency arguments, horizontal equity suggests that “save as you
earn” schemes should be more widely available. At present, such schemes are confined
only to those private-sector firms that also offer a share options scheme. Workers in the
public sector and in other private-sector firms are currently excluded, which is an
unnecessary restriction.

Although the primary goal of savings policy is to assist households in prudent financial
planning, a secondary benefit is that increasing the savings rate and/or improving the
composition of household portfolios by encouraging diversification can help to stabilise
the economy. A pro-savings policy can moderate the tendency for over-consumption
during boom times and protect household incomes during economic downturns, reducing
fluctuations in the level of economic activity and in property prices.

1. Introduction

The goal of this paper is to critically analyse the role of government policy in household
savings formation.

In view of the current policy debate, we focus on medium-term savings. Long-term
saving for retirement is clearly a vital element of overall saving. However, prudent
financial planning requires saving for the medium-term as well as for the long-term.
Moreover, the policy framework is more advanced with respect to the promotion of
pensions than for medium-term savings and, for this reason, the pension issue is not
directly addressed in this paper.

Previewing the main conclusions of the analysis, there are efficiency and equity reasons
for government policy to promote medium-term savings. In addition, economic theory
and the empirical evidence strongly suggests that the policy environment should not
discriminate against medium-term savings instruments that seek to obtain higher returns
by investing in equity markets. Finally, if policy is to stimulate extra savings, it is
extremely important that saving plans be designed that are easily accessible to a wide
range of households.

The structure of the rest of the paper is as follows. In section 2, we review recent savings
trends in Ireland in order to put the current policy debate into context. Section 3 outlines
the rationale for active government policy to promote savings. The appropriate design of
medium-term saving plans is analysed in section 4. Section 5 reviews the international
experience of tax-favoured savings programmes. Finally, some conclusions are offered in
section 6.

2. Recent Savings Trends in Ireland

Figure 1 shows the household savings ratio during the 1990s. Apart from the spike during
the 1992-1993 currency crisis, the savings ratio was relatively stable in the 8-9 percent
range. Since 1998 the savings ratio has sharply declined, from 9.6 percent in 1998 to 6.4
percent in 2000.

Figure 1: The Household Savings Ratio.







      91   92   93     94     95    96     97    98     99     00


Note: Ratio of personal consumption to personal disposable income. Data Source: ESRI Quarterly
Economic Commentary, various issues.

Several factors lie behind the recent decline in the household saving rate. Most obviously,
the strong economic growth performance has improved consumer confidence and
reduced the risk of unemployment. In addition, the sharp improvement in the public
finances has also reduced fears of a future increase in the level of taxation such that

improved expectations about the burden of taxation has also allowed a fall in the savings

However, a third factor behind the decline in the savings rate has been the fall in the real
(i.e. inflation-adjusted) interest rate earned on deposit accounts. The arrival of EMU has
led to a single European nominal interest rate. With Irish inflation in excess of the euro-
zone average, the real interest rate earned by Irish households has actually been negative
in the recent period. Figure 2 shows the inflation-adjusted interbank interest rate since
1991: the return on retail deposits is even lower.

Figure 2: The Real Interest Rate






         91   92   93    94     95     96     97     98     99     00


Note: Interbank rate minus CPI inflation. Data source: Central Bank of Ireland and International Monetary

A decline in real interest rates represents an important saving disincentive. However, the
negative impact can be partly offset by switching from deposit accounts into potentially
higher-yielding investment instruments such as equities or by seeking protection from

  The “Ricardian equivalence” hypothesis postulates that the household savings rate will be a positive
function of the level of the public debt, as households recognize that high public debt implies increases in

inflation by purchasing real assets such as property. Since there are substantial concerns
about potential over-dependence on investment in the property market, it seems desirable
to facilitate access to financial instruments (i.e. equities, bonds and related securities) as a
means for households to achieve satisfactory rates of return on their savings and as a
means of diversification against domestic economic risk. We return to this point in later

Finally, a third indicator of potential over-consumption and/or over-investment in the
property sector is the strong growth in personal credit in recent years. Figure 3 graphs the
inflation-adjusted growth in personal lending over 1993-2000. In real terms, personal
credit grew by more than 250 percent during this period. Although average household
debt may not yet be out of line relative to European norms, it seems likely that some
groups (the under-35 age cohort in particular) may be financially vulnerable in the event
of an economic downturn.

Figure 3. Bank Lending to Personal Sector

                           Personal Credit, 1993-2000
   Personal Credit



                      1993 1994 1995 1996 1997 1998 1999 2000

future taxation. Conversely, a fall in the public debt reduces the need to save.

Note: Author’s calculations based on bank lending and inflation data in Central Bank of Ireland Winter
Bulletin 2000. Figure plots real (inflation-adjusted) growth in bank lending to personal sector, with 1993
base normalised to 100.

3. Why Encourage Saving?

In this section, we investigate the question of whether there is a valid role for government
policy in promoting household saving. In a world of zero tax distortions and perfect
markets, rational forward-looking households would find the right mix between current
consumption and saving for the future. Moreover, they would be able to costlessly design
and manage the correct investment portfolio to achieve their desired savings targets.

Of course, the real world is far from this idealised picture. First, the tax system is heavily
biased against medium-term saving. In general, saving is subject to “double taxation”:
savings are taken from post-tax income and the income and capital gains generated by
accumulating assets are subject to further taxation.

Moreover, the tax system discriminates between different forms of saving and so affects
the composition of household portfolios. For instance, housing is favoured in several
ways: the non-taxation of capital gains on a primary residence; the non-taxation of the
imputed rental income on owner-occupied properties; and the continuing existence of
mortgage interest relief. This special tax treatment is in part responsible for the
predominant role played by property in average household net worth.

An over-reliance on property investment is extremely risky, in that ownership in the
domestic property market does not provide much diversification against the labour
income risks faced by households. For instance, in the event of a downturn of the Irish
economy, labour incomes would suffer, due to slower wage growth and a higher risk of
unemployment, and property owners would experience a further adverse wealth shock by
a fall in the value of their homes. Moreover, accumulating a medium-term buffer stock of

financial assets can offer protection to home owners by providing additional resources
that may be called upon in the event of mortgage repayment difficulties.

Within the class of financial assets, Budget 2001 proposes special tax treatment for
deposit accounts. For instance, it is proposed that the first £500 of dividend (interest)
income earned by credit union members (on funds committed for a minimum period of
five years) be exempt from taxation.2

In summary, the current system of taxation depresses the overall level of saving by
reducing the post-tax rate of return. It also heavily distorts the composition of saving, in
favour of housing and, to some extent, interest-bearing accounts and against investments
in market-based assets such as equities or bonds.

Second, the implementation of optimal portfolios is costly. In addition to the physical
transaction costs (e.g. broking fees for direct investment or management fees to financial
intermediaries), there is an informational cost in acquiring the “comfort level” required to
participate in equity-type investments. Accordingly, an improvement in financial literacy
and a reduction in the barriers to holding equities are important elements in widening
access to balanced savings portfolios.

Third, many households find it difficult to save. Recent developments in economic theory
and in psychology emphasise various factors that can lead to an over-consumption bias.
First, individuals tend to over-emphasise the near future over the medium- and long-term
in their decision-making processes (in technical terms, they apply hyperbolic
discounting).3 Moreover, even when it is recognised that it is desirable to save for longer
horizons, inertia and a lack of self-control prevent many households from acting in a far-
sighted manner. To overcome this tendency towards myopic behaviour, the availability of
commitment devices is important. Commitment devices include “save as you earn”
schemes that automatically withdraw regular savings payments from salaries and similar

 There is a £1000 exemption on capital gains from direct share ownership but such an exemption does not
apply to the returns on investment in diversified asset funds offered by financial intermediaries.

direct-debit schemes. We return to the appropriate design of medium-term savings
schemes in the next section.

During a period of economic boom, group factors can contribute a further bias towards
over-consumption. “Keeping up with Joneses” psychology induces extra pressure to
achieve high consumption during a boom period, as the average households strives to
emulate the consumption patterns of those gaining most from the boom.4

The primary reason for the government to be concerned about under-saving and
imbalance in the composition of individual portfolios is that households lack adequate
financial provision to achieve desired goals (a deposit for a house, replacement income to
finance a return to education or time at home when children are young) and withstand
“rainy day” events such as job loss or illness in the family.5 Bernheim (1995) finds that
the average American household is far below the recommended level of financial
provision. Beverly and Sheridan (1999) emphasise in particular the potentially positive
effects of asset accumulation in low-income households: holding some financial assets
provides some personal security over-and-above the support offered by state welfare
schemes. Banks and Tanner (1999) and Banks and Smith (2000) provide evidence for the
UK. They point out that the median level of wealth held in financial assets in the UK is
only STG£750, suggesting a severe lack of short-term and medium-term financial

However, under-saving also has macroeconomic consequences by increasing the
amplitude of business cycles: booms are intensified but at the price of deeper recessions.
By the same token, under-saving contributes to instability in the domestic inflation rate
relative to the average European inflation rate, with relatively higher inflation during
boom periods and lower than average inflation during downturns (even possibly

    See Thaler (1994) and Laibson (1996).
    See Deaton (1992) and Ljunqvist and Uhlig (2000).

There are also macroeconomic effects from imperfectly-balanced portfolios: over-
reliance on property amplifies fluctuations in house prices; a bias in favour of deposit
accounts over non-bank investments may contribute to greater fluctuations in credit
conditions. The roles of excessive house price and credit booms in creating the conditions
for a prolonged economic slowdown are by now well established.6

A further argument in favour of promoting equity ownership is that the efficiency and
flexibility of the economy is improved if as many households as possible own a stake in
profit-linked assets. At the firm level, this has a direct application in terms of employee
share ownership plans. At the more general level, it should be recognised that this
objective is best served by the promotion of well-diversified equity portfolios. A
diversified portfolio is more desirable than the holding of single shares, which are
excessively risky in terms of optimal household financial planning.

4. The Design of Medium-Term Saving Plans

It is well understood that savers with a medium-term horizon can achieve higher returns
by moving beyond a bank deposit account and investing in a portfolio of equities and
bonds. Many studies have shown that such portfolios consistently achieve higher returns
than cash over the medium- and long-term (Siegel 1998).7 Even small savers can achieve
a diversified portfolio via the collective investment schemes that are offered by financial
intermediaries, since such products reduce transaction costs (both financial and
informational costs) for the individual investor.

Conceptually, a single equity investment is a claim on the profits generated by a firm. By
holding a diversified portfolio across European markets, for instance, the investor has a
claim on aggregate profits in the European economy as a whole. With economic growth,
profits can be expected to increase over time offering a positive average return to

    See also Carroll (1997).
    See Bernanke, Gertler and Gilchrist (1998).

investors. As such, the stock market should not be described as a lottery or zero-sum
game: the natural state of affairs is for earnings and dividends to grow in line with the
economy. In the short term, however, share prices can fluctuate around the "fundamental"
value implied by profit forecasts. For this reason, the role of equity investments is best
suited to medium- and long-term horizons.

The positive contribution of a diversified portfolio that includes equity and bond assets in
achieving higher returns on medium-term savings makes it important that government
policy does not discriminate against such savings products.

It is well understood that reducing the tax on savings returns has both substitution and
income effects. The substitution effect calls forth an increase in savings as the higher
post-tax return induces households to defer consumption. The income effect works in the
opposite direction, in that the higher post-tax return enables a saver to achieve a desired
asset target more quickly. In a wide class of economic models, the substitution effect
outweighs the income effect, with some net improvement in the savings rate.

Moreover, a well-designed savings incentive programme can also raise the overall
savings rate through other channels. By providing extra incentives and encouraging the
provision of low-effort savings plans, inertia and other barriers that lead to under-saving
can be overcome. For instance, it is apparent that the convenience and automatic nature
of “save as you earn” schemes make it easier to save. A workplace-based scheme can
further promote savings by creating positive spillovers among co-workers, through
workplace-based financial education, conversations between employees and other “peer
group” effects (Bernheim 1999). By assisting individuals to save on a regular basis, such
innovations can improve household financial planning.8

  Equities may be understood to also include corporate bonds and related securities that offer higher returns
in exchange for accepting higher short-term volatility.
  Regular savings may be out of weekly or monthly pay packets but also encompasses other regular income
“events” such as annual or semi-annual bonus payments.

Thaler (1994) points out other advantages of special tax-exempt investment accounts.
From a behaviourial perspective, the special status of the investment account may make it
more likely that households do not touch these assets for a longer period. In this way,
outflows from the savings pool may be reduced in addition to any increase in new
inflows. Second, the strong pull of habit formation in savings behaviour means that
households will continue to make payments into the account, once the one-time effort in
setting up a plan has been incurred.

With respect to such schemes, there is an important issue of horizontal equity. At present,
“save as you earn” schemes in Ireland are only available to workers in firms that offer a
share options scheme (workers are encouraged to save in order to purchase the shares
offered at a discounted price by their employer). This excludes public sector workers and
those in firms that are too small or otherwise prevented from offering a share options
scheme. Since the “save as you earn” component is logically distinct from the share
options scheme, it seems an unnecessary restriction (and horizontally unfair) not to offer
similar encouragement for “save as you earn” schemes in other workplaces (in both
public and private sectors). Of course, this is not to discourage the introduction of share
options schemes where feasible and desirable but rather to improve access to “save as
you earn” schemes more widely across the economy.

5. International Lessons

The US and UK experiences is instructive in understanding the responsiveness of
household savings to the introduction of tax-favoured savings products.

The UK currently offers tax exemptions to the returns offered on Individual Savings
Accounts (ISAs) that replaced the older generation of TESSA (deposit accounts) and PEP
(equity plans) products in 1999. ISAs can take the form of cash accounts, equity products
or life insurance products or a combination across these categories. Payments into an ISA

can be made at any time (e.g. a regular monthly payment or, say, the payment of bonus
income on an annual basis) and a wide range of intermediaries offers ISA products.

A maximum of STG£7000 can be invested in the first year and STG£5000 per annum in
subsequent years. However, only STG£3000 in the first year and STG£1000 in
subsequent years can be invested in a cash account and the corresponding restrictions for
life insurance ISAs are STG£1000 and STG£1000 respectively. In contrast, the full
amount can be invested in equity-based ISAs. From this structure, there is a lot of
flexibility and choice offered in the range of qualified investments but the policy offers a
definite positive bias in favour of equity-type investments.

Although it is too early to examine the success of the ISA programme, the earlier TESSA
and PEP programmes were very popular. For example, the number of PEP plans grew
from about 200,000 in 1987/88 to 1.8 million in 1995/96 and the number of TESSAs
grew from 2 billion in 1991 to 4.5 billion in 1995. The average amount held in a TESSA
account had grown to STG£6,180 in 1996. However, the split between new saving and
the switching of assets from taxable funds into tax-exempt accounts is unknown, which is
an endemic empirical problem in assessing the overall impact of any tax-favoured
savings programme.

In designing the ISA programme, much emphasis has been placed on ensuring that
qualified investment products are made attractive to the average household by offering
easy access at low cost and fair terms that do not cause unpleasant surprises.9 Such
concerns emphasise that the success of a savings incentive scheme depends on making
saving as easy as possible for households and by ensuring a high level of transparency in
the marketing of competing investment products.

Although the ISA programme has many attractive features, it is worth pointing out two
potential limitations. First, it is up to the household to actively sign up for such a scheme,
so that the process involves greater effort than workplace-based schemes. Second, the tax

    See HM Treasury (1998).

advantage is that returns are earned tax-free: payments into ISAs are still made out of
post-tax income. For many households, a stronger incentive would be that the payments
into ISAs be made out of pre-tax income.

The US has concentrated on offering tax incentives for retirement savings but these still
provide lessons for the design of medium-term programmes. The two main initiatives
have been Individual Retirement Accounts (IRAs) and 401(k) schemes. Of particular
interest are the 401(k) schemes since there are based in the workplace and contributions
are deducted from pre-tax pay. The “automatic” nature of the savings scheme makes it
attractive for individuals that find it difficult to save out of discretionary income. Finally,
401(k) schemes allow employers to match employee contributions, providing an extra
mechanism by which firms can compensate employees.

Although there is much controversy about the overall impact of IRAs and 401(k)
schemes on net savings, it is agreed that the effects are not confined to just the impact of
the tax breaks on the net rate of return.10 The implicit endorsement implied by
government incentives may increase the awareness of the need for saving. This
awareness is amplified by the financial education and information provided by the
intermediaries competing for the management of these funds. As pointed out in section 4,
the net impact of a scheme on savings is not confined to stimulating new inflows into
savings plans but also by deterring outflows from the savings pool: this is especially the
case for schemes that impose a minimum time limit to take advantage of the tax

6. Conclusions

In this final section, we draw some policy implications from the foregoing analysis. The
primary rationale for policy intervention in promoting medium-term savings is to
encourage adequate levels of personal financial provision at the household level. Such
medium-term savings provide a buffer stock of assets that better enable households to

withstand adverse events and achieve important personal goals. A secondary reason for
intervention is that a higher level of savings during an economic boom can help promote
macroeconomic stability by reducing the risk of overheating and leaving the economy
less exposed in the event of a downturn. Financial prudence requires that households
accumulate sufficient medium-term financial assets, in addition to the long-term goal of
ensuring adequate provision for retirement.

Government policy can encourage greater household saving and promote better portfolio
composition both through tax policy and by widening the availability of schemes, such as
the “save as you earn” programme, that make it easier for households to save on a regular
basis. In particular, reducing the barriers to investment in equity/bond funds can improve
the balance of household portfolios, improving expected rates of return and potentially
achieving greater levels of diversification against key risks to household prosperity, most
notably the risk of a downturn in the domestic economy that would hit labour incomes
and property values. At the very least, tax policy should not discriminate against
equity/bond investments. Indeed, the ISA programme in the UK goes further and actually
promotes equity-type plans as the best mechanism for achieving medium-term savings

Behavioural arguments suggest that the positive impact on overall savings is stronger
from a tax credit on inflows into saving than from a tax exemption on outflows out of

Finally, there is a strong horizontal equity argument to widen the availability of “save as
you earn” schemes: there is no strong reason why these should be confined only to those
private-sector firms that also offer a share options scheme. Workers in the public sector
and in other private-sector firms are currently excluded from “save as you earn”
arrangements, which is an unnecessary restriction.

     See Bernheim (1999) and Poterba (2000).


Banks, James and Sarah Tanner (1999), “Household Saving in the UK,” mimeo, Institute
for Fiscal Studies.

Banks, James and Sarah Smith (2000), “UK Household Portfolios,” Institute for Fiscal
Studies Working Paper 00/14.

Bernanke, Ben, Mark Gertler and Simon Gilchrist (1998), “The Financial Accelerator in
a Quantitative Business Cycle Framework,” NBER Working Paper No. 6455.

Bernheim, B. Douglas (1995), “Do Households Appreciate Their Financial
Vulnerabilities? An Analysis of Actions, Perceptions, and Public Policy,” Tax Policy and
Economic Growth, Washington, D.C.: American Council for Capital Formation, 1-30.

Bernheim, B. Douglas (1999), “Taxation and Saving,” NBER Working Paper No.7061.

Beverly, Sondra G. and Michael Sherraden (1999), “Institutional Determinants of Saving:
Implications for Low-Income Households and Public Policy,” Journal of Socio-
Economics, 28, 457-473.

Carroll, Christopher (1997), “Buffer-Stock Saving and the Life-Cycle/Permanent Income
Hypothesis,” Quarterly Journal of Economics, 112(1), 1-55.

Deaton, Angus (1992), Understanding Consumption, New York: Oxford University

HM Treasury (1998), Making Saving Easy: A Summary of the Proposals.

Institute for Fiscal Studies (2000), Tax Incentives for Saving, Briefing Note.

Laibson, David I. (1996), “Hyperbolic Discount Functions, Undersaving, and Savings
Policy,” National Bureau of Economic Research, Working Paper No. 5635.

Llungqvist, Lars and Harald Uhlig (2000), “Tax Policy and Aggregate Demand
Management Under Catching Up with the Joneses,” American Economic Review, 90(3),

O’Toole, Francis and Pamela Warrington (1998), “Taxation and Savings in Ireland,”
Trinity Economic Paper Series Technical Paper No. 98/5.

Poterba, James (2000), “Taxation and Portfolio Structure: Issues and Implications,”
mimeo, MIT.

Siegel, Jeremy (1998), Stocks for the Long Run, McGraw-Hill.

Thaler, Richard H. (1990), “Saving, Fungibility and Mental Accounts,” Journal of
Economic Perspectives, 4(1), 193-205.

Thaler, Richard H. (1994), “Psychology and Savings Policies,” American Economic
Review, 84(2), 186-192.

Thaler, Richard H. and H. M. Shefrin (1994), “An Economic Theory of Self-Control,”
Journal of Political Economy, 89(2), 392-401.

Thom, Rodney (1998), “The Taxation of Savings,” Foundation for Fiscal Studies
Research Report No.2.


To top