This article is based on the paper Global Market Inefficiencies by Söhnke Bartram and Mark Grinblatt, recently named the winner of the 2017 AAM-CAMRI-CFA Institute Prize in Asset Management, jointly bestowed to an exceptional piece of research by Asia Asset Management, NUS Business School’s Centre for Asset Management Research and Investments (CAMRI), and CFA Institute. The full paper can be accessed through https://www.arx.cfa/post/Global-Market-Inefficiencies-4217.html.
In Global Market Inefficiencies, co-authors Mr. Bartram and Mr. Grinblatt offer evidence that investment strategies, derived from the simplest manipulation of past accounting data, can earn alpha worldwide. Outsized alphas, the authors assert, occur in regions like emerging markets and in the Asia-Pacific region, including that region’s developed markets (e.g. Japan). Areas with the largest alphas tend to have stock markets containing what the authors describe as “greater frictions”, specifically, trading costs, and also prohibitive rules i.e. short sale restrictions that deter sophisticated arbitrageurs. When allowed to fully compete in a country, these traders – who can capture the most fleeting of opportunities and persistently drive stock prices to fair values – leave few scraps for less sophisticated active investors.
Arbitrageurs seem less inclined to attempt their strategies in certain markets where there are forces pushing against them.
What’s the investment issue?
The paper is one of a handful that seeks to address why some markets are less efficient than others. Although conventional wisdom, this premise is not universally accepted.
Some academic research papers support the concept that developed markets are no more efficient than emerging markets. Authors Mr. Bartram and Mr. Grinblatt dispute this. Others have argued that investors rely too heavily on Fama’s Efficient Market Hypothesis (EMH) – that is, that stocks always trade at their fair value – rendering any effort to profit from mispriced stocks an exercise in futility.
On the one hand, the authors ask, if no one can ever profit from active management, then what magical force exists to drive prices to fair value? They argue that a passive investor will buy the index fund, whatever the prices of its component parts. Active managers, who have performed relatively poorly of late, may well have another cycle to show off their talents – and so will want to take note of where and when mispricings and market inefficiencies cross paths.
How do the authors tackle this issue?
Mr. Bartram, of the University of Warwick, and Mr. Grinblatt, of UCLA, began their work by going through a database of company financials that went back more than two decades to capture all the information about the companies that would have been known at the time.
They constructed a robust set of synthetic portfolios – nearly 26,000 stocks from three-dozen countries – to theoretically trade on mispriced companies, with a few unique layers of variables to provide reality checks. Trading signals, suggesting a clearly identifiable deviation of a stock’s price relative to its estimated fair value, were built using international point-in-time accounting data covering 21 company-specific metrics. Trading activity was replicated with transaction cost data from Elkins McSherry; the gold standard for tracking such expenses transaction cost data included explicit commissions and fees as well as harder-to-quantify market impact costs. These costs were converted to alpha reductions using portfolio-turnover approximations, that is, two-way turnover.
What are the findings?
The results of running the mispricing replicating portfolios (and incorporating simulated buy/sell executions) were definitive: Emerging and certain developed Asian markets were shown to be relatively less efficient in countries with quantifiable market frictions – particularly trading costs – that deter arbitrageurs. The paper states: “If profits to trading strategies based on mispricing estimates are a measure of market inefficiency, then profits should vary across countries as a function of transaction costs, short sales restrictions, and other country characteristics that might influence limits to arbitrage, thereby impeding the process that makes a county’s stock prices reflect fair value”.
What are the implications for investors and investment professionals?
In regions where markets are most efficient, Mr. Bartram and Mr. Grinblatt caution that investors need to be aware of the costs of active management, noting that it is unlikely the fees associated with active management will outweigh its value. A caveat here, though, is that the data studied were annual accounting data. Conceivably, some improvement in the alpha may be generated by the strategy, even in the most efficient global markets, when using quarterly data.
And there’s a catch as well: the least efficient markets, where the alpha opportunities may be the largest, can easily be eroded by those frictions – in other words, super-sophisticated investors are steering clear for good reason.
*Rich Blake is a veteran financial journalist and CFA Institute contributor
