A recent article by finance professor George Athanassakos in The Globe and Mail alerted individual investors to the fact that financial markets are laden with conflicts of interest. Among the culprits listed were institutional investors who indulge in “window dressing” of portfolio holdings at the end of the quarter, rating agencies that are compensated by the very companies they are reviewing and the media, which may be sensitive to advertising and sponsorship revenues.
For many investors, however, the most obvious conflict of interest and the hardest to avoid is that of the sell-side brokerage analyst. As Dr. Athanassakos points out, they favour buy over sell recommendations by a factor of seven to one. They are understandably motivated to write positive stories about companies because they need to maintain access to management and hope to be involved in corporate finance deals.
There isn’t much we can do about the behaviour of portfolio managers or rating agencies, but is there a simple fix to reduce, if not eliminate the obvious conflict by sell-side analysts? The European Union financial regulators certainly thought so. In January, 2018, they introduced the Markets in Financial Instruments Directive II (MiFID II). The regulations covered a wide range of subjects affecting the investment process, but the section of interest here is the one that required asset managers and broker dealers to unbundle the cost of equity research from that of trade execution. The intent was to force equity analysts to justify their value-added and for the buyers of investment research to put a separate price on the value of the product delivered.
This sounds fine on paper, but did it achieve its objective or was it just another regulatory change that created paperwork but missed its mark? The latest issue of the Financial Analysts Journal (Fourth Quarter 2022) contains an article that tackles this very question, so we have some answers.
The article, Analyst Incentives and Stock Return Synchronicity: Evidence from MiFID II was written by two finance professors and a PhD candidate – Yihan Li, Xin Liu and Vesa Pursiainen, so the mathematics are heavy duty, but the conclusions seem straightforward. They start out with two broad assumptions about the consequences of MiFID II: that the number of analysts covering European firms will decrease and that the quality of the remaining research will improve. The first assumption is quickly confirmed: There were about 3,000 analysts covering European stocks in 2015. Thirteen per cent left the industry in 2017 and another 9 per cent in 2018, presumably in response to a more rigorous evaluation of their research output. For a control group of U.S. firms, the annual attrition rate of analysts over the same time frame was in a range between 3 per cent to 5 per cent, so it seems reasonable to attribute the shrinkage to MiFID II.
Assessing the improvement in quality of the remaining research output is trickier. The authors’ solution is to calculate the correlation of daily stock returns with the daily returns from the market index. They believe that a high degree of correlation between the two reflects less firm-specific information being incorporated into the stock price. In other words, a stock that closely tracks the index is responding mainly to market moves and not to company-specific events. If analyst research coverage improves in quality, then stock prices will more likely respond to company-specific news events uncovered by the analysts rather than the overall market. So, the correlation will diminish.
With this framework in place, the article shows that relative to the U.S. control group, correlation with the market decreased by more than six percentage points for European firms following MiFID II. This doesn’t sound like much, but the result is “statistically significant and economically large” according to the authors. Equally important, there had been virtually no difference in the market correlation between European and matched U.S. firms prior to MiFID II, which implies that it was the new regulation that had a positive impact on European research quality.
The research study is detailed, but obviously has some limitations, which the authors concede. The period covered is the two years leading up to MiFID II and the two years after so as to avoid factors related to COVID-19, but the time frame is certainly short. The study also excludes the bottom 10 per cent of stocks measured by market capitalization, so the conclusions may not hold for the small cap universe.
The authors believe their results demonstrate that the regulatory changes introduced by MiFID II have created an improved information environment in the European stock market. Fewer analysts are providing better quality information. This sounds like a classic definition of productivity improvement – a better quality product produced with fewer inputs. It is difficult to argue with this outcome, though you may feel that only a finance professor would need to do a correlation study to discover that buyers are more discriminating when they have to pay for a product that was formerly “free.”
MiFID II did not have a direct application in Canada, although its introduction did lead to a flurry of comments from domestic law firms and securities commissions at that time. Their conclusion was that the MiFID II unbundling requirement did not create an immediate need to amend the current regulatory regime in Canada, which did permit bundled services after appropriate disclosure.
This was a perfectly reasonable response at the time when no data were available to assess the cost/benefit trade-off of the proposed change in payment for investment research. Now that we have credible research from the Financial Analysts Journal, it may be time to revisit the topic.
Robert Tattersall, CFA, is co-founder of the Saxon family of mutual funds and the retired chief investment officer of Mackenzie Investments.