The Business Times, Wednesday 30 May, 2007
The benefits of diversification
Global diversification among asset classes makes the most efficient
portfolios by delivering the highest returns for any given level of
volatility
, says JOSEPH CHONG
MUCH has been written about the benefits of
portfolio diversification. Unfortunately, most of the preaching is
centred on alpha, beta, sigma and correlation coefficients, which is
Greek to most of us. Nonetheless, diversification does bring
benefits and this piece is a brief report card of some of the
interesting insights from our work in diversifying portfolios
globally on behalf of our clients.
One of the recurring client questions is: 'I believe in the concept
but how do I diversify?' The research in finance theory demonstrates
time and time again that global diversification among asset classes
delivers the most efficient portfolios - delivering the highest
returns for any given level of volatility.
Indeed,
this is how our portfolios are allocated globally across equities,
fixed income, commodities and cash.
However, we believe one
important feature that portfolios should have is diversifying by
'perspective'. Table 1 illustrates this concept.
The greatest threat to any investment portfolio is an extreme
event from the blind side, such as the Asian financial crisis.
Multiple perspective global diversification is a 'thatched-weave'
approach which compels a congruency of views and, hence, reduces the
size of the investment blind spot. For example, the current upward
move in global equity markets seems healthier as a sector
perspective check shows that it is broad-based with most sectors
participating; instead of being led primarily by the energy sector
as in 2004-5.
A valuation check shows that the major global
sectors are trading in the 10 to 20 times price-earnings ratio band,
ie, there is no major 'avoid' among the sectors.
This is unlike 1999 when the global information technology (IT) sector
was trading at 150 times earnings versus a market average of 25 times.
Essentially, ensuring congruency of views is a built-in defensiveness,
providing a portfolio with greater survivability; and eventually the
capacity to exploit any severe crisis while grinding out annual gains
consistently. This approach is never expected to give the high-octane
performance of 'hot' single-country markets but achieves regularity
year in, year out.
So what are the results of this approach over the past few years?
Chart 1 shows the performance (in S$) of the model moderate risk
portfolio against the benchmark (55 per cent global equities, 42 per
cent fixed income and 3 per cent cash) before advisory fees. It is
expressed before fees as fees are scaled down as the size of the
portfolio climbs. The annualised rate of return of this model
portfolio from Sept 6, 2002 to Feb 2, 2007 has been 9.8 per cent.
Chart 2 shows the performance of the moderately aggressive versus
moderate risk portfolio. The annualised rate of return of the moderate
risk portfolio from Sept 6, 2002 to Feb 2, 2007 has been 11.3 per
cent. The volatility of the higher risk portfolio has been, as
expected, higher but the swings have been only one-third of the
additional 1.5 per cent earned annually.
That is,
contrary to expectations, a somewhat higher risk portfolio appears to
have been more efficient in the past five years.
A preliminary analysis shows that the culprit may be fixed income -
this asset class has been more volatile relative to returns in the
past five years.
This may have
implications for asset allocation policy-making going forward and
could impact the life of the bull market in equities more than
currently expected. This is over and above other factors like the
on-going massive share buybacks and private equity deals in the
developed markets which are reducing the number of outstanding shares
significantly.
Chart 3 shows the annualised portfolio returns at any point in time as
measured from Sept 6, 2002. This was done for both the moderate and
moderately aggressive portfolios.
What is interesting is the convergence of annualised returns over time
for both portfolio risks - about 9.5 per cent for the moderate
portfolio and 11 per cent for the higher risk one after about two
years. The swings in annualised returns initially (less than one year)
is a mathematical phenomenon - a portfolio that is up 2.5 per cent in
two months would have an annualised return of 16 per cent and vice
versa.
This notwithstanding, we believe that this initial volatility can be
managed. Without going into numerical details, our own experience
shows that by a combination of avoiding overbought conditions (as
dictated by technical analysis) and staggering deployment of fresh
cash, these initial swings can be mitigated.
The writer is CEO of financial adviser New
Independent. He welcomes feedback at : josephchong@ni.com.sg
This article is for information only. Readers should seek independent
advice before making any investment decisions
