IMC Official Training Manual updates: Version 7 to 7.1
The IMC Official Training Manual authors have created the following updates to
Version 7 of the Manual. These are to reflect industry developments since Version 7 was printed and in the
light of recent market change. This does not constitute a substantial ‘syllabus review’. However, candidates
should familiarise themselves with these updates, which could affect a small number of questions in IMC
examinations from 1 December 2009.
5a4 Add as additional bullet point after Financial stability of the provider
Stability, independence and standing of trustees, fund custodians and
The collapse of the brokerage firm Bernard L Madoff Investment Securities (BMIS) in December 2008
provides an important illustration of the requirement to undertake due diligence in the fund selection
process. Madoff effectively operated a giant Ponzi scheme whereby existing investors were paid from new
money coming in. Investors had to invest in an approved feeder fund which in turn opened up a brokerage
account with BMIS. The feeder fund had to delegate to BMIS the full trading authority of their portfolios.
A typical hedge fund uses a network of service providers which can include an investment manager to
manage the assets, a broker to execute trades, a fund administrator to calculate the NAV and
custodian(s)/prime broker(s) to custody the assets. These service providers should normally be independent
of each other. With BMIS all the functions were performed internally with no independent oversight. This
should have provided warning signals to investors as it allows for performance manipulation and
misappropriation of assets.
The auditors of BMIS were a small accountancy firm that was virtually unknown within the investment
management industry. The choice of such a small auditor with little track record for such a large fund should
have been questioned by investors. Additional investigations would have shown up that the accounting firm
had declared in writing in its annual returns to the American Institute of Certified Public Accountants that it
had not conducted any audit since 1993.
6a4 Add to the end of the section
(f) The implications of assuming returns are normally distributed
Conventional asset allocation theory makes a range of assumptions about the normality of asset returns, the
most problematic of which are that returns are independent from period to period and normally distributed.
To implement an asset allocation framework based on normal asset return distributions, investors need only
to make two assumptions for each asset class (mean and standard deviation) and one assumption for each
The normal distribution has several attractive properties: it is easy to use and produces tractable results in
many analytical exercises as it is completely characterized by its first two moments (mean and standard
deviation), thus establishing the link with the mean-variance optimization theory.
The normal distribution arises as the limiting distribution of a whole class of statistical testing and estimation
procedures, and therefore plays a central role in empirical modeling exercises. One of the main
characteristics of the normal distribution is that its tails decay exponentially toward zero; thus extreme
realizations are very unlikely. However, this seems to contradict empirical findings on asset returns, which
evidence that returns' distribution generally exhibits leptokurtic behaviour, i.e. has fatter tails than the normal
distribution. This means that extreme returns of either sign (but particularly of a negative sign) occur far
more often in practice than predicted by the normal model. In other words, it is commonly believed that
optimal asset allocations under the assumption of normally distributed returns have a higher Value at Risk
(VaR) than the model suggests if returns in reality follow a leptokurtic distribution.
OTM update - October 2009
Another problem of non-normality is correlation breakdown. It has been found in recent times that in many
cases correlations under extreme conditions are quite different to those under normal conditions. Relying on
linear correlation matrices tends to overestimate the benefits of portfolio diversification during periods of high
6b4 Delete section (iv) and replace with:
Performance attribution - return due to choice of benchmark
Relative performance measures such as the information ratio are used to evaluate performance relative to a
benchmark. Performance attribution aims to identify the causes of deviations between the return of the
benchmark and the actual portfolio. Here we discuss performance attribution methods that analyze both the
portfolio composition and returns in identifying causes of deviations from the benchmark. Most methods
used in practice are derived from the framework used by Brinson and Fachler. The Brinson and Fachler
framework is based on a topdown investment management process.
The framework starts from a general investment plan that describes planned portfolio weights for asset
classes and assigns benchmarks to asset classes. The framework provides room for decisions on tactical
issues regarding deviation from the asset allocation. This decision is called the tactical asset allocation
decision. Portfolio managers within asset classes also have different levels of discretion to deviate from their
benchmarks. Actual deviations from the benchmark with an asset class are called stock selection decisions.
The analysis is based on four different portfolios:
I. The benchmark portfolio;
II. The stock-selected portfolio;
III. The timing portfolio;
IV. The actual portfolio.
Portfolio I is the overall benchmark portfolio, which is derived directly from the general investment plan.
Portfolio I defines the desired asset allocation also known as the strategic asset allocation and the
benchmarks for individual asset classes.
Portfolios II and III are the outcomes of a ‘what if’ analysis that aim to measure the impact of decisions in
isolation from the other decisions. The return on portfolio I is the result of investing the portfolio exactly
according to the strategic asset allocation and the benchmarks for the asset classes. The return of this
portfolio is defined as:
R(I) = ∑wiP RiP
where wi is the strategic weight of asset class i and Ri is the return of the benchmark for asset class i.
Portfolio II measures the return of implementing the stock selection decision, whilst ignoring the tactical
asset allocation decision. Let the actual return of an investor in asset class i be Ria, then the return on
portfolio II is:
R(II) = ∑ wi Ri
Portfolio III measures the return of implementing the tactical asset allocation decision without the stock
selection decision. The asset class weights for this portfolio are equal to the actual weights wia, and the
securities within an asset class are exactly equal to the benchmark for the asset class. Thereturn on this
R(III) = wia RiP
Portfolio IV is the actual portfolio subject of the analysis. The return on the actual portfolio is:
R(IV) = wi Ri
OTM update - October 2009
The difference between the return of portfolio I and portfolio III is the so-called ‘timing effect’, and represents
the additional return due to the tactical asset allocation. The return difference between portfolio I and
portfolio II is the so-called ‘selection effect’ and shows the additional return due to stock selection. Since the
timing effect and selection effect are calculated independently, we also need to capture the joint impact of
both decisions, which is called the ‘interaction effect’. This effect is calculated as the
sum of the return of portfolio I and IV minus the sum of the return of portfolio II and III.
Since the interaction effect cannot be attributed to a single person or department, some practitioners
allocate this effect in equal parts to both the timing and selection effect. The effects and their calculations
are summarized below:
Components of the attribution model
Suppose for example that an investor utilizes a strategic asset allocation of 50% bonds, 20% domestic
stocks and 30% foreign stocks. Furthermore, assume that the actual allocation, the actual returns, and the
benchmark returns are presented in table 6.3 below
Table 6.3: Actual portfolio weights, benchmark returns and benchmark portfolio weights
Table 6.4: Outcomes of the attribution model
This analysis shows that the investor in this example outperformed the overall benchmark by 2.5%, which is
mainly due to considerable timing skills and not to selection skills.
8b2 Add to the end of this section
The monetarist school of thought began in the late 1940s with Milton Friedman. Instead of rejecting macro-
measurements and macro-models of the economy, the monetarist school embraced the techniques of
treating the entire economy as having a supply and demand equilibrium. However, because of Irving
Fisher's equation of exchange, they regarded inflation as solely being due to the variations in the money
supply, rather than as being a consequence of aggregate demand. They argued that the "crowding out"
effects discussed above would hobble or deprive fiscal policy of its positive effect. Instead, the focus should
be on monetary policy, which was largely ignored by early Keynesians. Monetary policy is discussed further
in the next section.
OTM update - October 2009
8b4 Add to end of section
Other tools used by central banks to manage the economy and in particular
As noted above, central banks use interest rates to target inflation. As interest rates approach zero this
method of implementing monetary policy is constrained by the inability of interest rates to go below zero.
The central bank is unable to provide a stimulus to the economy by reducing interest rates. This is obviously
a problem when inflation is below target or worse inflation is negative. An alternative approach to providing a
stimulus is to target the quantity of reserves rather than its price (interest rate). This approach is sometimes
referred to as quantitative easing.
The two main channels of monetary transmission open to a Central Bank when changes in short term
interest rates are not effective is:
(1) Expand bank reserves – banks should use these reserves to fund credit creation when they can only
earn a zero interest rate on holding reserves. The expansion of credit will boost economic activity
(2) Purchase long term securities to drive down long term rates of interest
The first channel requires banks to believe that the increase in reserves is long lasting (so they don’t simply
hold on to them) and that they have sufficient capital to support an increase in credit extension. The second
channel of directly targeting long term rates requires the Central Bank to make large purchases of long term
securities to drive up their price and hence push down yields.
OTM update - October 2009