Quarterly Reporting and Market Liquidity

Trading in U.S. equity markets is fast and cheap.  While proponents of ending quarterly reporting point to the dangers of short-termism, less frequent disclosure is also likely to lead to a decline in liquidity and to greater trading costs.  The SEC should carefully consider the effect of increasing asymmetry on the flow of investment capital in secondary markets.

For decades, economists have pointed out that asymmetric information inhibits trade, beginning with George Akerlof’s famous observation that nobody will buy used cars if there is a high enough chance that some are lemons.[1]  In modern financial markets, the risk of transacting against an informed trader is reflected in the “bid-ask spread,” the difference between the price at a which a market maker is willing to buy and sell stock, respectively.  By definition, market makers lose to informed traders: If a counterparty is buying a stock because it is undervalued, a market maker trades the wrong way by selling that stock to the buyer, and vice-versa if the market maker buys an overvalued stock from an informed seller.[2]  The bid-ask spread thus compensates market makers for providing liquidity in the face of losses from transacting against informed traders.

Because bid-ask spreads are paid by all market participants, they function as a kind of transaction tax.  And as many of us learned in introductory microeconomics, transaction taxes can impose “deadweight losses,” making everyone worse off for no apparent reason.  In a recent article, Larry Glosten and Tālis Putniņš show formally how increasing bid-ask spreads leads to social welfare losses.[3]  The key idea in their paper is that market participants will not trade when they believe the value of the stock lies within the bid-ask spread—and this inhibits gains from trade.

To illustrate this point, it is helpful to consider a hypothetical stock with a bid price of $9.50 and an ask price of $10.50; that is, a bid-ask spread of $1.  The midpoint of the spread is $10, which reflects the market maker’s best guess as to the value of the firm.  Suppose a bullish buyer believes the firm is actually worth $10.25 per share.  It is clear that he or she will not buy the stock, because it would cost $10.50 to purchase a share valued at $10.25, leading to a loss of $0.25 per share.  Similarly, suppose a bearish seller believes the firm is worth $9.75 per share.  The seller will not trade, because he or she would only receive $9.50 for a share valued at $9.25, leading to a loss of $0.25 per share.  Thus, even though both the bullish buyer and bearish seller would be better off by trading, the bid-ask spread prevents these transactions from taking place.

Now consider what happens if bid price is $9.90 and the ask price is $10.10, i.e., the bid-ask spread is $0.20.  The midpoint is the same ($10), i.e., market makers do not value the firm any differently than before.  But now the bullish buyer will trade, because he or she can pay $10.10 to acquire a stock that he or she values at $10.25.  This transaction adds $0.15 per share of value for the buyer.  Similarly, the bearish seller will sell, because he or she receives $9.90 for a stock that he or she thinks is worth $9.75.  From the seller’s standpoint, that transaction adds $0.15 per share of value.  In other words, there are gains from trade: A smaller bid-ask spread makes both the bullish buyer and bearish seller better off.

The accounting literature has identified an empirical link between quarterly reporting and lower bid-ask spreads.  In a 2012 article published in the Journal of Accounting and Economics, Renhui Fu, Arthur Kraft, and Huai Zhang examine disclosure behavior prior to the SEC mandating quarterly reporting in 1970.[4]  During the years 1959 to 1973, as many as 70 percent of publicly traded companies voluntarily chose to report at a quarterly interval.  Fu, Kraft, and Zhang begin by showing a simple negative correlation between reporting frequency and measures of information asymmetry, including the bid-ask spread, i.e., firms that reported at a quarterly interval had a lower spread.[5]

Of course, there can be other explanations for this simple correlation other than information asymmetry.  For example, firms with low trading volume or high levels of risk/uncertainty are likely to experience higher spreads, simply because market makers are less sure about how much the firm’s stock is worth.  Fu, Kraft, and Zhang address these alternative explanations three ways. First, they estimate a firm fixed effect model, which considers disclosure behavior and bid-ask spreads within-firms.  This design controls for differences between firms that are unchanging over time.  Second, they employ an instrumental variables (IV) design, using the reporting year as an instrument.  Finally, they estimate a matched control sample design, comparing firms within the same year and industry that are closest in size.

To be sure, none of these provides conclusive proof that information asymmetry is responsible for the increase in bid-ask spreads arising from quarterly reporting.  The firm fixed effect model is subject to the critique that trading volume or risk/uncertainty vary over time, which is not controlled for in that design.  The IV design seems particularly weak, as it is contaminated by time trends in macroeconomic reporting frequency.  The matched control sample is more persuasive.  Firms of a similar size, in the same year-industry pair, are more likely to have similar trading volume and risk/uncertainty.  All in all, the persistence of this link between greater disclosure frequency and lower bid-ask spreads across all of these designs strongly suggests that quarterly reporting reduces information asymmetry, as measured by bid-ask spreads.  The authors also present additional measures of information asymmetry and cost of capital, and similarly find that quarterly reporting reduces these as well.[6]

What to make of this evidence?  Many have argued that semi-annual reporting will help managers focus on long-run shareholder value—though as my colleague Professor Coffee points out, even that conclusion is questionable.[7]  And while much of the policy conversation around securities disclosure focuses on IPOs, secondary markets are no less important.  The presence of a robust, liquid market for a firm’s shares reassures employees that equity-linked compensation can be readily converted to cash in the event of an unexpected financial shock.  Greater trading strengthens the incentive to ferret out information about a firm and profit by arbitraging away temporary mispricing—thereby making prices more accurate.  In short, quarterly reporting makes markets more efficient.  The benefits of quarterly reporting for market liquidity should be weighed against the possible costs highlighted by proponents of less frequent disclosure.

ENDNOTES

[1] George A. Akerlof, The Market for “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q. J. Econ. 488 (1970).

[2] Merritt B. Fox, Lawrence R. Glosten & Gabriel V. Rauterberg. The New Stock Market: Sense and Nonsense, 65 Duke L.J. 191 (2015).

[3] Tālis J. Putniņš & Lawrence R. Glosten, Welfare Costs of Informed Trade (2016), available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2827158.

[4] Renhui Fu, Arthur Kraft & Huai Zhang, Financial Reporting Frequency, Information Asymmetry, and the Cost of Equity, 54 J. Acct’g & Econ. 132 (2012).

[5] Id. at 144.

[6] Id. at 145.

[7] John C. Coffee, Jr., What Really Drives “Short-Termism”?, Colum. L. Sch. Blue Sky Blog, Aug. 27, 2018.

This post comes to us from Joshua R. Mitts, an associate professor at Columbia Law School.