Should the number of bets affect confidence level in a manager’s value-added?

By Dan Hallett, CFA, CFP on July 29th, 2011

In this morning’s Globe & Mail, Shirley Won has an article on portfolio manager Noah Blackstein.  While yours truly is quoted in this piece, I mention it here because the article touches on a topic that I’ve been mulling for some time.

Historical value-added

In the article I commended Blackstein for his excess returns (above his benchmarks) over his tenure, which includes two bear markets.  I also made note of Blackstein’s high turnover.  But I’d like to add some context to both items and propose that there may be a link.

Blackstein began managing money for Dynamic funds in 1998.  (Side-note:  I met him more than a year earlier when he was an analyst at BPI Funds.  But now I’m aging myself.)  He began managing Dynamic Power American Growth in 1998.  Three years later he became lead manager of Dynamic Power Global Growth Class.

In both cases he has posted returns just north of 6% annually since each fund’s respective inception date.  Since Blackstein’s 1998 start in Power American Growth, U.S. stocks (in Canadian dollars) have been flat.  Global stocks, also in Canadian dollars, have also been about flat (-0.2% per year) since the fund’s 2001 birth.  So he has generated excess returns of about 6% annually on his Dynamic funds.

The magnitude of his outperformance is more impressive when you consider that his funds’ returns (as with all mutual funds) are reported net of each fund’s management expense ratio (MER) and brokerage costs incurred on stock trades.  The funds’ MERs, which range from 2.35% to 2.45%, were higher in previous years.

And given Blackstein’s heavy-trading style – turnover has averaged north of 300% annually in recent years – the funds have racked up a significant amount of trading costs (not included in the MER).  Blackstein has posted excess returns exceeding 600 basis points annually net of fees, expenses and trading costs totalling more than 3% annually.  (And his turnover is lower than it used to be.  I first began studying his American Growth fund in 2001 and I was struck by its 2,100% turnover rate for calendar 2000.  That’s no typo.)

The number of bets and confidence levels

Given that our firm advises wealthy families, our clients tend to have most of their wealth outside of tax-deferred and tax-free accounts.  Accordingly, we tend to favour money managers that don’t trade a lot to minimize taxes and fees.  (We also prefer managers with a longer-term view.)  But for the purposes of this discussion I propose that in the case of managers like Blackstein, the combination of high turnover and portfolio concentration may give a higher level of confidence in past excess returns.

Let me use the opposite extreme to illustrate.  Suppose a manager bought a bunch of stocks a dozen years ago and didn’t trade at all over the subsequent 12 years.  Assuming this low-turnover manager generated excess returns of 6% annually over the subsequent 12 years, one might be inclined to attribute success to luck.  All this hypothetical manager needed was to make good picks at the outset with no decisions thereafter.  I note that this is a big assumption since this manager may have been very active ‘behind-the-scenes’ and simply chose not to trade – which is sometimes harder.  But bear with me for a moment.

In Blackstein’s case, he has generated a similar amount of excess returns but he’s done so having made many more decisions or ‘bets’ as implied by his more than 300% annual turnover.  Extending this argument, Blackstein has always held a very concentrated portfolio of 20-25 stocks.  That level of concentration is another form of ‘bet’.  Portfolio managers can play it safe by holding a long list of stocks so that they don’t stand far apart from the index.  In Blackstein’s case, he not only is very concentrated but by trading a lot he’s giving himself many opportunities to make bad, money-losing decisions.  Overall, his trading appears to have enhanced returns.

This doesn’t mean that heavy-traders tend to generate more value added.  But where a money manager has generated returns in excess of his benchmark, it may be reasonable to assign a greater degree of confidence in managers that have achieved success while making a larger number of bets.  I’m not yet convinced of this but it strikes me as a reasonable thought to put forward.  That said, I’d love to hear others’ thoughts on this so don’t be shy to write a comment below.