“High conviction”. The very term conjures up images of Martha Stewart, Bernie Ebbers or some hedge fund blow up. But the term is also being used to describe an emerging alpha-centric approach to investing. In this case, however, “high conviction” refers to the manager’s belief in their stock picks, not to the likelihood of them getting nailed for insider trading.
While the idea isn’t rocket science, it captures the essence of alpha-centric investing – shunning the benchmark and going out on a limb instead. Successful or not, it’s why you should be paying your manager. As Alexander Ineichen points out, there’s no skill required to manage a benchmark.
We found this white paper by Fortis, a European banking and insurance firm, makes the case for conviction. Fortis launched its high conviction offering in response to what it says was a “disappointment with perceived ‘benchmark hugging’ by supposedly actively managed funds.” To which we say “Amen”.
Fortis’ philosophy has three parts:
- Markets are inefficient
- It’s easier to exploit inefficiencies at a security-level than at a sector-level.
- Sector performance is becoming less disparate – reducing the opportunity to simple pick winning sectors (see chart below)
After dancing around the efficient market hypothesis for a while, the Fortis paper lands its first blow:
“…despite well-publicised data that most equity managers do not beat their benchmarks, implying that stock markets are in fact efficient and that it is hard for managers to generate alpha. For example, Kung and Poulman found that over the ten years to 2003, the median US large cap managers generated only 0.58%.
“However other studies support the view that certain managers can outperform (or underperform) on a persistent basis – there are superior fund managers out there. Kosowski et al find that ‘superior funds that beat their benchmarks by an economically and statistically significant amount do exist. Notably, our tests also show strong evidence of inferior funds.’ They conclude that this ‘strongly supports that many of these fund managers have genuine skills in picking stocks. Thus, our study supports the value of active mutual fund management, although it also highlights the drawbacks of funds actively managed by those who cannot pick stocks’.”
(For those who are in the mood to read an extensive white paper, you can check out Kosowski’s tome here.)
The paper follows with a flurry to the solar plexus of efficient markets. Inefficiencies, it says, take time to exploit. As a result, mutual fund managers – who are compensated for short-term performance – never get an opportunity to catch the big one.
“Most managers are judged on their short term, relative performance: they face short term pressures to outperform the benchmark on a quarterly basis. Inevitably, this encourages ‘benchmark hugging’ – managers are unwilling to take large-scale bets on ideas which might take more than a year to come good, or to take a contrarian stand and find themselves alone and on the wrong side of the market for a couple of quarters. This short term focus means that managers may well make small, short term upside from small, short term bets, but it also means that there are significant mispricing opportunities which are left unexploited because they take longer to play themselves out.”
Fortis takes it on the chin, however, when it recalls a study proving no more than 20 stocks are needed to diversify a portfolio. This, after taking great pains to acknowledge that high conviction portfolios contain truckloads of idiosyncratic (manager-) risk.
“Statman calculated the optimal number of stocks for a fully diversified portfolio as 120. He also concluded that the lion’s share of the benefits of diversification are achieved with a portfolio of just 20 stocks – the marginal benefit of adding to the portfolio declines rapidly thereafter.”
Meir Statman’s 2002 paper, How Much Diversification is Enough?, actually lamented the lack of diversification in most brokerage portfolios. But it did make passing reference to a “rule of thumb” that 20 stocks ought to do it.
Much of the academic rationale for a high conviction portfolio is reminiscent of the empty statements contained in many asset managers’ marketing brochures: “invest for the long-run”, “ignore short-term volatility”, and “don’t compare us to a benchmark”. At the end of the day, the decision to invest with “high conviction” is a subjective and behavioral one.
By coincidence, Meir Statman authored another paper on the cognitive biases that can turn overconfidence into a dangerous lack of diversification. That paper explores the real-life case of one individual’s personal portfolio. Showing great conviction, that individual allowed a huge unrealized gain to build up in one particular stock. Then, when security-specific problems hit that stock, the portfolio was threatened. Clearly panicking, the individual hit the sell button a little prematurely, it now turns out. That stock was ImClone. And that investor was Martha Stewart.
“High conviction” indeed…