Financial Markets Evidence File

Nov18: New Financial Market Information: Indicators and Surprises
Why Is This Information Valuable?
Empirical Asset Pricing via Machine Learning” by Gu et al.
Excellent overview of the asset pricing accuracy of different ML techniques. Key insight: the best performing methods to a better job than traditional approaches of capturing non-linear interactions between key variables
Index Proliferation Adds Choice But Fuels Confusion” by Pauline Skypla in the Financial Times
“A recent survey by the Index Industry Association revealed its 14 member companies publish 3.29m indices, of which 3.14m cover stock markets. Only 5.6 per cent of these 3.14m are factor or smart beta indices, which are based on factors other than companies’ market capitalisation. However, that modest percentage still works out at more than 175,000.”

“These measures may not all be investable indices — benchmarking, where investors use an index to assess their own performance, is also a driver of proliferation. Even so, the choice facing investors can be confusing. “The proliferation of indices, and the way providers calculate indices that sound the same differently, makes the job of investors more difficult,” says Deborah Fuhr, managing partner at ETFGI, a London-based consultancy. Problematically, there is no standard set. “Smart beta is a space that isn’t well defined or owned by a couple of index providers,” says Ms Fuhr.”

“A check of five well-known providers in the field (ERI Scientific Beta, Vanguard, State Street Global Advisors, FTSE Russell Global Factor Index Series, MSCI Factor Indexes) shows they all include the criteria of value, momentum and volatility that are among the top half dozen filters commonly applied to smart beta products.”
The above column highlighted the growing number of indexes that underlie so-called “smart beta” products. This brought to mind a number of previous papers on the smart-beta approach, which concluded that many investors were likely to be disappointed. These include Rob Arnott’s “How Can Smart Beta Go Horribly Wrong”, by Rob Arnott, “Quantifying Backtest Overfitting in Alternative Beta Strategies”, by Suhonen et al, and “Smart Beta Herding and Its Economic Risks: Riding the Dragon” by Krkoska and Schenk-Hoppé.
Since the smart beta products first appeared, The Index Investor, we have emphasized that they are active management products (see, The Confusing World of Factor (“Smart Beta”) Models and Indexes, from our August 2003 issue). While it is possible that they will return lower returns with less risk, or higher returns with more risk than a broad market index fund, a belief that they will produce higher returns with lower risk rests on three hiqhly questionable assumptions:

(1) The mispricing of factor risks that smart beta products claim to exploit is a durable phenomenon – e.g., one caused by investors’ systematic cognitive or emotional biases;

(2) There are durable barriers that prevent other investors (including algorithmically driven funds) from arbitraging away the mispricing of one or more factor risks that smart beta products exploit; and

(3) Investors are able to identify in advance smart-beta funds that are based on those factors to which assumptions (1) and (2) apply.
Unfortunately, once it becomes widely recognized, the failure of smart-beta funds to deliver superior returns is likely to further shake investor confidence in active management.
It Was the Worst of Times: Diversification During a Century of Drawdowns” by AQR Capital Management
AQR highlights a critical distinction between diversification and hedging. The former involves investing in assets whose returns have a low correlation to equities. There is no guarantee that the returns on diversifying investments will be positive when those on equities are negative. And as we saw in 2008, correlations across asset classes can substantially increase during periods of extreme uncertainty and system stress.

In contrast, hedges are deliberately designed to increase in value when returns on equities decline –put options being the classic example.

However, for this very reason, hedging investments will tend to be more expensive than diversifying investments.
Challenging the Conventional Wisdom on Active Management: A Review of the Past 20 Years of Academic Literature on Actively Managed Mutual Funds”, by Cremers et al
Cremers presents a good summary of arguments in favor of active management. At Index Investor, we have never denied that over some periods of time many active managers will outperform an appropriate passive benchmark index.

What we have always questioned, however, is (1) their ability to sustain that superior forecasting performance (or luck) over time; (2) their ability to identify and implement profitable investment opportunities as their funds grow in size; and (3) investors’ ability to identify these superior active managers in advance, rather than in hindsight. If you consider this a joint probability and assume that each of these probabilities is slightly better than luck – say, 55% -- then the joint probability – which essentially equals the probability of an active management strategy outperforming a passive strategy over the long term – is only equal to about 17% -- or a one in six chance.
Private equity deals fail to keep up pre-crisis successFinancial Times 17Oct18
This column is a good example of the argument against active management outlined above.

The proportion of winning private equity deals — those that deliver more than three times the original investment — has seen a sharp decline in the years since the financial crisis as buyout groups struggle with record-high valuations and fierce competition, an analysis has shown.”

“On average, 35 per cent of deals produced healthy returns between 2002 and 2005 compared to roughly 20 per cent of winning transactions between 2010 and 2013, an analysis by Cambridge Associates and Bain & Company showed.”
One week after the above story this one appeared: “Private equity set to surpass hedge funds in assetsFinancial Times, 24Oct18
Private equity will overtake hedge funds as the largest alternative asset class within the next five years as investors flock to private rather than public markets in search of returns, according to a new analysis.”

There are at least two possible explanations for this: (a) optimism, overconfidence, and conformity biases on the part of the institutional investors committing more funds to private equity in spite of declining recent returns; or (b) a rational decision to take on more risk in pursuit of higher returns, even though the probability of the latter being realized has significantly declined.

In the latter case, I have in mind the no-win situation faced by public sector pension funds in the United States, most of which are badly underfunded. Their managers must choose between hoping to reduce underfunding by earning high investment returns, or telling public sector employers that they must increase their annual pension fund contributions, which in turn will necessitate either cuts in spending in other areas, and/or an increase in taxes on the public.

Oct18: New Financial Market Information: Indicators and Surprises
Why Is This Information Valuable?
Stories on 10th anniversary of the Lehman Bankruptcy
A recurring theme is how little has changed. While very aggressive monetary, and in some cases fiscal policy, staved off a severe and prolonged downturn, little was done to address the underlying causes of the 2008 crisis, which in some ways have arguably become worse over the past decade (e.g., debt levels, inequality, low productivity, corporate concentration, and declining labor share of national income).
The Next Financial Crisis Won’t Come from a Known Unknown” by Robin Wigglesworth in the Financial Times

No Deal Brexit has Big Implications for Europe’s Derivatives Market” by John Dizard in the Financial Times

How Hedge Funds Keep Markets Trading in a Crunch” by Gillian Tett in the Financial Times

Wigglesworth highlights that in a complex adaptive system like the global financial markets, crises are most likely to emerge from unanticipated combinations of apparently benign factors. He also notes that since “high frequency trading, quants, passive funds, and options now account for about 90 percent of US equity trading volumes”, this structural change in the market is likely to rapidly accelerate, and potentially exponentially increase the damage caused by whatever combination of causes trigger the next global financial crisis.

Dizard suggests one potential cause that is easily overlooked – the post 2008 concentration of derivative trading in a small number of clearinghouses that lack sufficient capital to make good on a rapidly increasing number of failed trades, as might occur if the next crisis produced, as the last one did, an exponential increase in funding/liquidity problems – e.g., for leveraged hedge funds that, as Tett reports, have, since Dodd-Frank imposed higher limits on bank capital, become much more important sources of market liquidity than they were in 2008.
The FTSE All World ex US index results compared to US equity market returns. Through September, the rest of world is down (5.26%), while the US is up 8.96%. But gains in the US are narrowly concentrated: FTSE Health, up 16.1%; Consumer Services, 18.5%, and Technology, 19.1%
Narrow markets imply a high degree of social learning and imitation, which is a hallmark of situations characterized by high uncertainty and elevated potential for sudden and substantial changes and regime shifts.