آیا بازار سهام با رفع نگرانی ها واکنش نشان می دهد؟ مدارک و شواهد از ایالات متحده آمریکا و استرالیا

We examine 12-month returns following disclosure of first-time going concern (GC) opinions in the U.S. and Australia. We find no evidence of significant negative abnormal returns associated with GC opinions in Australia. In the U.S., negative abnormal returns subsequent to GC opinions are sensitive to choice of expected returns—notably, there are no significant negative abnormal returns when using factor models or after controlling for momentum. Overall, contrary to Taffler, Lu, Kausar's [2004. In denial? Stock market underreaction to going-concern audit report disclosures. Journal of Accounting and Economics 38, 263–285.] U.K. results, we are unable to document a market anomaly in the U.S. or Australia associated with GC opinions.

Taffler et al. (2004) document that U.K. firms receiving first-time going concern audit opinions (henceforth GC opinions) on average underperform various return benchmarks by 28–35% during the 12 months following the opinion announcement. The Taffler, Lu and Kausar (henceforth TLK) study is a part of a burgeoning literature that documents anomalous market behavior in general and underreaction to bad news events in particular.1 Much of this literature concludes that the documented returns’ patterns are unlikely explained by risk or other rational considerations but represent anomalous market behavior arising from human information processing biases (Kahneman et al., 1982). In a similar vein, TLK conclude that their evidence suggests anomalous market underreaction to the GC opinions that is inconsistent with conventional notions of market efficiency. Specifically, TLK observe that they “cannot reject an irrational investor explanation for [their] results with investors apparently underreacting to or, in effect, denying the bad news conveyed by a going concern audit opinion”.2 Ceteris paribus, if the principal source of the GC opinion anomaly is human information processing bias, one would expect to observe similar inefficiency in other markets (countries). On the other hand, if the TLK's evidence is limited to the U.K. stock market, then country-specific institutional features or other market peculiarities are a more likely explanation for their findings. Accordingly, in this paper we investigate whether the GC opinion anomaly identified by TLK in the U.K. is present in other markets such as those of the U.S. and Australia. Our sample comprises 1159 GC opinions in the U.S. over the 1993–2004 period and 91 GC opinions in Australia over the 2000–2004 period. As in TLK, we examine returns’ patterns over the 12 months succeeding the GC opinion. To verify the robustness of our results, we examine several alternative measures of risk-adjusted returns, including those using benchmarks (reference portfolios and control firms) matched on risk characteristics, as well as abnormal returns using conventional factor models. Overall, we are unable to find evidence that stock markets in the U.S. and Australia underreact to GC opinions. For the Australian sample, we are unable to find significant negative returns under any of our returns’ definitions.3 We do find that our U.S. sample firms underperform benchmarks based on certain risk characteristics (i.e., book-to-market and size). However, we find no evidence of negative abnormal returns using well-accepted factor models. The inability of the GC opinion strategy to generate significant negative abnormal returns when using factor models casts serious doubts on the underreaction hypothesis. Additionally, there are no significant negative returns in the U.S. following GC opinions after controlling for momentum using either the control firm or the factor model approaches. TLK note that high transaction costs likely preclude arbitragers from exploiting market mispricing associated with the GC opinion in the U.K. Further analysis reveals that our sample firms are qualitatively similar to TLK's U.K. firms in terms of size and liquidity, suggesting that similarly high transaction costs are present in our U.S. or Australian samples. Therefore, our differential results are unlikely attributable to differences in arbitrage effectiveness, but rather suggest that there is unlikely investor irrationality associated with the pricing of the GC opinion in the U.S. or Australian markets. Our study has significant implications for the GC opinion anomaly. TLK find significant market underreaction associated with the GC opinion in the U.K. and conclude that their evidence is consistent with irrational investor behavior. In contrast, we are unable to document any systematic mispricing associated with the GC opinion in the U.S. or Australia.4 Overall, the inability to find irrational investor behavior associated with the GC opinion outside the U.K. suggests that the TLK anomaly unlikely represents some universal phenomenon related to human information processing bias.

We examine 12-month returns following disclosure of GC opinions for a sample of U.S. and Australian firms. We find little evidence suggesting that the stock market underreacts to the GC opinion in either the U.S. or Australia. In Australia, abnormal returns following GC opinions are insignificant under a variety of expected return definitions. In the U.S., we do find that GC opinion firms underperform portfolios or control firms matched on book-to-market and size. However, there is no evidence of negative abnormal returns when we use factor models. Also, there are no significant abnormal returns when we control for momentum, either through factor models or through matched control samples. Overall, contrary to Taffler et al. (2004) who report significant underreaction to the going concern reports in the U.K., we do not find a pricing anomaly associated with the going concern opinion either in the U.S. or in Australia. Our results have the following implications. While acknowledging that limited arbitrage could explain their results, TLK conclude that they cannot rule out irrational investor behavior as the likely explanation for their findings. Our inability to document a similar anomaly in markets outside the U.K. suggests that the TLK's results are unlikely attributable to systematic errors in human information processing. Rather, their results likely represent country specific phenomenon that may be caused by either specific institutional features or other peculiarities of the U.K. market.