Jun 23rd 2015, 15:38 by Buttonwood
WHEN it comes to investing, those who buy tracker funds are sometimes regarded
with a patronising air; of course, those are fine for the mass market, but
sophisticated people pick the outperforming funds. And for those, you have to
pay more.
But how does one know which fund will outperform? Any assessment of the past
record carries the implication that performance is persistent. But that
assumption is highly doubtful. The last study I quoted came from Vanguard,
which manages index trackers; some doubted the results on those grounds. This
time, the survey comes from S&P Dow Jones Indices, so may be criticised for the
same reason. The trouble is that active managers are unlikely to produce
research in this area as it is not in their interest. And it is worth noting
that Morningstar found similar results.
So let us take the 682 domestic US equity funds that were in the top quartile
as of March 2013. How many were still in the top quartile a year later? If
performance was random, one would expect a quarter to do so; the actual number
was 21.3%. By the time one reached March 2015, randomness would suggest 6.25%
of funds would remain in top quartile (a quarter of a quarter); the outcome was
5.28%. One might as well toss a coin or spin a bottle.
Aha, you might say, those numbers are skewed by the large cap funds. Stocks in
the S&P 500 are widely followed and the index is difficult to beat. The real
skill is in the smaller stocks; there a manager can add value. Alas, those
numbers look random too; 22.7% smallcap managers remained in the top quartile
after two years and 4.67% over three. Take it to five years and no smallcap
manager maintained top quartile performance.
A single year is just too short to measure a manager's skill, you say? OK, let
us go back over five years. What proportion of the top quartile over the five
years ending March 2010, were also top performers in the five years ending
March 2015? Again, a random result would be 25%; the actual result is 24.8%.
Nearly 22% of such funds ended up in the bottom quartile.
Similar results were found in the fixed income category, where a passive
approach might seem less sensible (a passive bond allocation implies the
biggest weights in the most-indebted countries). Nevertheless, in many fixed
income categories, no manager achieved top quartile performance in five
consecutive years. Only in short-dated investment grade bonds was there any
sign of persistence.
If the numbers do not convince you, then go back to logic. If it were easy to
find fund managers who persistently outperform, why would anyone give money to
all the other managers? And yet there is no shortage of fund management
companies or mutual funds.