As an alternative way to shed light on the debate over whether accruals quality is a priced risk factor, we examine the effect of seasonality on the pricing of the modified Dechow and Dichev (2002) accruals quality measure (AQ). We find that (1) high AQ stocks outperform low AQ stocks only in January; (2) during the rest of the calendar year, high AQ firms underperform low AQ firms such that there is no AQ premium on an annual basis; (3) about half of the January AQ premium occurs during the first five trading days of January; (4) a January AQ premium is observed in almost every year of our sample period; and (5) the January AQ premium reflects, at least partly, the stock price effects of tax loss selling around the turn of the year. Taken together, these findings are difficult to reconcile with a risk interpretation of accruals quality.A fundamental issue in this debate is whether there is a robust, positive relation between AQ and future returns. 1 If high AQ firms do not outperform low AQ firms on average, then AQ cannot represent risk and additional asset-pricing tests, as in CGV, are moot. However, prior evidence on the relation between AQ and future returns is inconclusive, differing across studies, sample periods, and methodologies. For example, CGV conclude that there is no robust positive relation between AQ and future returns after controlling for market microstructure biases and/or transaction costs. On the other hand, Ogneva (2008) and Kim and Qi (2010) argue that high AQ firms do outperform low AQ firms after controlling for cash flow shocks or removing low-priced stocks, respectively.Since traditional asset-pricing tests (as in the above studies) have yielded mixed results on AQ's pricing, and thereby on the risk debate, we take a different approach and examine how seasonality affects AQ's pricing. In particular, we investigate the role of January versus the rest of the calendar year. As we show, viewing the pricing of AQ through a different lens-that of seasonality-is a powerful way of shedding light on whether AQ reflects risk. The seasonality approach also helps to explain why prior evidence on AQ's pricing (based on tests using all calendar-year months) has so far been ambiguous.Our exploration of January's role in AQ's pricing is motivated by two considerations. First, the nature of the AQ measure strongly suggests that it could proxy for tax loss selling around the turn of the year. If so, then January is likely to play an important role in AQ's pricing because tax loss selling causes (at least partly) the well-known January effect in stock prices (Roll 1983b;Reinganum 1983;Sias and Starks 1997;Poterba and Weisbenner 2001). 2 Prior research argues that firms with high operating or return volatility-such as small firms-are more likely candidates for tax loss selling because they are statistically more likely to experience operating losses, and therefore negative stock returns, in any given year (e.g., Roll 1983b; Constantinides 1984). This reasoning suggests that AQ could also proxy for tax los...

No abstract

No abstract

We investigate whether multiemployer defined-benefit pension plan (MEPP) underfunding is priced by shareholders and creditors. Prior to the FASB's new MEPP standard (effective December 2011), when the disclosures on such plans were sparse, we find evidence (some evidence) that our estimate of a firm's share of MEPP underfunding is credit (value) relevant. We also find some evidence that a proxy for the funded status of a firm's MEPPs is incrementally value relevant over and above the firm's cash contributions, but no evidence that it is credit relevant. Furthermore, an estimate of MEPP underfunding that incorporates the additional disclosures required under the new MEPP standard is value and credit relevant, both individually and incrementally, over and above our old estimate. Overall, our findings suggest that shareholders and creditors view MEPP underfunding as a debt-like obligation and that the additional MEPP disclosures under the new standard are useful to market participants.

We predict and find that short-selling constraints combined with investor disagreement cause prices to respond more strongly to bad earnings news than to good earnings news, an asymmetry characterized by Skinner and Sloan as the ''torpedo effect.'' However, in the absence of short-sales constraints, the price reaction to good and bad news is entirely symmetric, regardless of the level of investor disagreement. Our findings contribute to the ongoing debate about the existence and causes of the torpedo effect. In particular, we extend Skinner and Sloan's explanation of this effect by showing that short-selling constraints are essential for such an effect to occur; in their absence, there is no torpedo effect, even if investors are overoptimistic. Moreover, this asymmetric effect is not intrinsic to growth stocks; even value stocks are torpedoed in the presence of short-selling constraints and disagreement.