I vehemently disagree with David Stevenson on factor investing

view original post

Citywire recently published an article by David Stevenson headlined Why factor investing has failed.

The article has triggered a response from Research Affiliates director of research Vitali Kalesnik, who said he ‘vehemently’ disagreed with Stevenson’s conclusion. 

Kalesnik took issue with Stevenson’s view that factor investing was a fad and that investors are best advised to play to strengths of index funds for ‘mega-cap equities and individual segments of the bond markets’, while picking active managers in a few areas where they can outperform. 

In this article for Citywire, Kalesnik dissects Stevenson’s argument.

Where we agree

I do agree with several points David Stevenson made. The observation that three to five years ago, factor investing was overhyped and investors were overpromised is excellent. But many of us in the industry did realise this and rang the alarm bell. In 2018, my colleagues at Research Affiliates, Rob Arnott, Cam Harvey, Juhani Linnainmaa, and I published ‘Alice’s adventures in factorland: Three blunders that plague factor investing’, in which we warned of the likely consequences from the overselling of factor investing.

Specifically, we identified three major blunders likely to hurt investors if they are not prepared:

  • exaggerated expectations about factor performance as a result of data mining, crowding, and unrealistic trading cost expectations, among other issues;
  • unrealistic risk expectations and underappreciation that factor investing is prone to sharp drawdowns and prolonged periods of underperformance;
  • and unrealistic expectations about the perceived diversification benefits of buying several factors, when in the real world, factors often crash in a correlated manner.

Largely due to these blunders, many investors have suffered through a period of factor performance far worse than the promises made by marketers of factor investing.

We did not expect that soon after we published the article, the global economy would enter the worst recession since World War II, caused by the pandemic-induced lockdown.

That precise sequence of events was impossible to predict. But what was possible to predict was that some shocks would occur and that investors were largely unprepared to handle them.

Don’t throw the baby out with the bathwater

What was the genesis of factor investing and why did it become so popular?

The most common and well-known factors were discovered between the 1970s and 1990s. The popularity behind factor investing, however, was not based on that financial research. Instead, the main driver behind factor investing’s growth in popularity has been that the strongest sources of return, which can differentiate a good fund from a bad fund, were threefold: one, fees charged by the fund; two, transaction costs incurred by the fund; and three, factor exposures of the fund.1

Active fund managers’ claims of unique skill imply that investors should sell after a fund underperforms (because the skill wasn’t delivered and likely won’t be in future) and buy after a fund outperforms (because the skill was delivered and likely will continue to be in future).

This logic is flawed because many of the performance drivers are beyond a manager’s control. The rotation from a recent loser to a recent winner means investors sell recently cheap assets and buy recently expensive ones. As the asset’s value reverts to the mean, research shows that investors suffer an average performance gap of 2% a year versus a capitalisation-weighted benchmark.2

The fundamental reason behind factor investing’s more widespread adoption over the last few years has been low fees and transparent exposures to the key ex-ante identifiable drivers of return (ie, systematic factors).

Patient investors who stick to the robust factors — that is, those supported by solid economic research, such as value or low beta — are likely to outperform a cap-weighted benchmark, which tends to overweight overpriced stocks and underweight underpriced stocks.

The low-cost aspect of factor investing means that investors and not managers can benefit from these investment strategies. The transparency associated with factor investing renders the need to chase alpha, and thus lose 2% a year from poor entry/exit timing of fund assets, obsolete.

Can investing in mega-caps and active growth go wrong?

Stevenson concluded by recommending a return to active investing in concert with a passive investing strategy tilted heavily towards mega-caps and specific segments of the bond market. The irony is that today this particular recommendation is prone to be one of the worst possible.

In 2016, my colleagues at Research Affiliates Rob Arnott, Noah Beck, John West and I published ‘How can smart beta go horribly wrong?’. We observed that smart beta (ie, factor investing) had the potential to deliver very disappointing results.

Some very popular strategies, such as low volatility, quality, and momentum, were trading at expensive valuations relative to their respective histories. By 2021, many investors had come to view factor investing as going wrong for them because they had invested in expensive strategies and underappreciated the risks of their underperformance.

To follow Stevenson’s recommendation means investing in a market-cap index dominated by very few and very expensive (based on their valuations) Fang+Tesla companies. The cyclically adjusted price-to-earnings ratio (aka, the Shiller PE or Cape) shows that the MSCI All Country World index is more expensive today (99.7%) than at any other time in its history since 1995.

An expensive starting valuation means a low return going forward. Most fixed-income securities available to investors today are yielding less than inflation. Active management? If investors were to pick an active manager showing recent outperformance, likely they would be allocating to the same very expensive Fang+Tesla companies as well as paying high management fees.

Sticking with factor investing

By June 2021, in the developed markets, the value factor was trading at valuation discounts cheaper only than in the summer of 2020, when the global economy was in total lockdown. Momentum, illiquidity, and some definitions of quality (for example, conservative investments) were trading at valuations cheaper than 90% of their respective histories.

Factor investing can also offer low management fees and transparent exposures to assets currently trading at record-low valuation multiples. That said, investors should not expect miracles from these strategies as, like any investment strategy, they can and will underperform at times. This risk is exactly the reason they will tend to reward patient investors with a premium over the long term.

Vitali Kalesnik is the director of research at Research Affiliates

1 The key conclusions from Mark Carhart’s 1996 article ‘On persistence in mutual fund performance’. Beyond the this three drivers, Carhart’s research found very little persistence in active fund managers’ skill. Later research showed that it is possible to identify skilled managers with measures such as active share, but these three drivers still remain the strongest drivers of fund-performance persistence.
2 In 2016, Hsu, Myers, and Whitby showed in their article ‘Timing poorly: A guide to generating poor returns while investing in successful strategies’ that on average a dollar-weighted allocation lags a buy-and-hold return by nearly 2% a year.

Any opinions expressed by Citywire or its staff do not constitute a personal recommendation to you to buy, sell, underwrite or subscribe for any particular investment and should not be relied upon when making (or refraining from making) any investment decisions. In particular, the information and opinions provided by Citywire do not take into account your personal circumstances, objectives and attitude towards risk.