Past Success…is in the past!

I am piggybacking off a Dan Solin article here which picked up on a Morningstar article about the inclusion of actively managed funds in portfolios. 

And from the Morningstar article the following quote is relevant:

Investors can easily have the best of both worlds—topnotch actively managed funds and low-cost index trackers–in their portfolios.”

The Morningstar article suggested that low fees and strong 10 year track records are two key decision factors for fund selection, and Dan pointed out (perhaps miffed by the “can easily have” belief of the article) that 10 year track records are just that. 

Dan and his MPT brethren believe that it is impossible for most actively managed funds to consistently add value, to beat the market.   I would agree, but for different reasons:

A fund that has outperformed may have done so for reasons specific to the fund’s style (market cap, growth/value, sector focus etc), and even significant but short term performance boosts, that are unlikely to ever be meaningfully repeated, can warp a 10 year track record.   Very strong performance should raise more questions than answers!

The manager may have also been unwittingly in the right area at the right time.   Some will have performed by design, others by accident, while factors that had driven styles and sectors and hence boosted historical performance (tech, large cap, small cap) may for one reason or another be set for a decline. 

You have to know your valuation parameters, you have to know what drives and what has driven valuations in the market place.   You also need to understand the importance of significance – small valuation differentials are often not significant enough to justify an active position, more often than not because small valuation differentials have an insufficient margin of error built into them.   Under uncertainty not all relative price movements can be deemed incorrect!  

One of the key truisms of MPT, that is relevant for all asset management, is that money flows/demand drives correlations and that part of management is getting the right allocation to these relative flows – an MPT approach would state that all assets are correctly valued (those with in and those with out flows), while a value biased investor might believe that the two are contrary movements of relative value.  Since not all relative price movements are movements out of or into value, the universe of real active, potentially rewarding, investment opportunities is fairly limited, most of the time.  

Fund selection based on past performance can be a graveyard for the naive and ignorant and an opportunity for those sufficiently conflicted (the sellers of funds) as to ignore the realities of such decisions. 

In reality you should not be looking for performance, but value, which means future relative out performance, which is more than likely (if you are like me a contrarian, value biased investor) to be found in areas which have under performed, and to be able to accept the risks of such allocation, those areas which have under performed significantly. 

Really, you should be eschewing the top performing funds of the last 10 years.   The better option, whether you are an MPT adherent, or a value biased investor who believes opportunities do exist, is to buy the index rather than the top performing funds of the last 10 years

In fact, because humans/markets are not uniformly rationale/efficient,and do not provide effective intrinsic valuation guidance, at key points, that past performance is even less of a predictor of future performance than it would be in an efficient market scenario: sectors can become wildly over and under valued, markets euphoric and depressed from one moment to the other, while economies and financial systems can be pushed way outside their equilibrium points.  

Performance is an awkward dimension: anyone who was active during the late 1990s and who was also aware of the mind blowing extreme valuations, and the eye popping valuation differentials between “old” and “new” sectors will know that you can out perform while engaging in reckless behaviour and under perform while making sensible valuation decisions.   In a way, it has been the imperfection of markets and human decision making that lays the path for the existence of far too many actively managed funds – most funds are graveyards, but there is always new blood chasing the past.

It should be clear that there are far far too many actively managed funds, and those managed funds that do exist should really be focussed on the margins of the portfolio.   If there is a rationale for active, it should be marginal and it should be driven by out of equilibrium positions in markets.   There may also at times be little rationale whatsoever for any active allocation. 

Unfortunately most humans lack the discipline and the expertise to make such effective asset allocation decisions, and most investors and most advisors would indeed be better off with the indexed allocation.   I must admit to having some concerns over the use of the word “investors” in the Morningstar article, because most individual investors should not be going anywhere near the “candy store”/“sweet shop”.   This is not good advice! 

Unfortunately, MPT has not really helped in the past performance stakes: in an efficient market paradigm, past performance profiles, are, on average, meant to be predictive and have been used, rightly or wrongly, for such purposes: note the endemic use of mean variance optimisers with irresponsible long term return assumptions for equity market returns.   MVOs allocate more to higher expected returns for a given level of risk.

Leave a Reply