The relationship between returns and unexpected earnings: A global analysis by accounting regimes

Numerous studies have documented a long-term association between earnings and returns. Surprisingly, few attempts have been made to internationally examine market reactions to earnings releases over return windows less than 12 months. This paper globally explores the market reaction to unexpected earnings defined by both the change in earnings per share (EPS) and analyst forecast errors (AFE) using a 1-month return window. First, the existence of the earnings-returns relationship is examined using a sample of firms from 32 countries grouped into accounting regimes. Accounting regimes represent groups of countries that exhibit similarities in accounting standards, stock market characteristics, corporate governance mechanisms, and economic conditions. Thus, similar reactions to earnings are expected within regimes. Next, the incremental information content of analyst forecasts, a proxy for investors’ earnings expectations, is examined. Finally, changes in the structure of the earnings-returns relationship over time are investigated. Results support the existence of a relationship between earnings and returns in all accounting regimes. In addition, analyst forecast errors appear to be incorporated into earnings expectations in most developed countries. Finally, evidence suggests that the significance and explanatory power in the earnings-returns relationship has increased in recent years.     

Aggregate earnings and stock market returns: The good,the bad,and the state-dependent

Prior research documents a negative aggregate earnings-returns relation. In contrast, we posit that the sign of the relation varies, depending upon the macroeconomic and financial market conditions that exist in the earnings announcement quarter. We argue that the existing macroeconomic and financial market conditions influence market participants’ frame of reference, which in turn affects whether they interpret aggregate earnings surprises to be informative about the expected inflation component of the discount rate, the market risk premium component of the discount rate, or aggregate future cash flows. Consistent with this, we find that the sign of the aggregate earnings-returns relation changes numerous times across our sample period. We also find that market participants interpret aggregate earnings to be informative about changes in expected inflation (market risk premium) when the sign of the aggregate earnings-returns relation is negative (positive). Finally, we identify macroeconomic and financial market conditions under which the aggregate earnings-returns relation is more (less) likely to be negative (positive).     

THE USE OF MULTIPLE INSTRUMENTS FOR MEASUREMENT OF EARNINGS FORECAST ERRORS, FIRM SIZE EFFECT AND THE QUALITY OF ANALYSTS'FORECAST ERRORS

This paper develops an instrumental variables framework to form a better proxy for earnings forecast errors. The key aspect of the approach is to extract information from alternative proxies for the same underlying variable, namely a portion of realized earnings signals unexpected by the market. We use signs of various proxies for earnings forecast errors obtained from different time-series forecasting models as multiple instruments. The results show that the instrumental variables approach is effective for reducing measurement errors inherent in various proxies for earnings forecast errors. It produces not only a smaller magnitude but also a narrower dispersion of earnings forecast errors. The paper provides evidence that the instrumental variables approach performs better for small-firm samples than for large-firm samples. Finally, we observe that analysts' forecast errors seasoned with the signs of various time-series forecast errors (as well as the signs of their own forecast errors) outperform those without seasoning. This indicates that analysts' forecast errors can still be improved by employing the instrumental variables technique.     

Is there Information in an Earnings Announcement Delay?

Using a sample of announcements drawn from the 1980s and early 1990s, we reassess the relation between earnings news and earnings announcement timing. Using analyst forecast errors to proxy for news, we find that early announcements are associated with good news relative to late announcements. The relation between news and timing, however, does not appear to be strictly monotonic. Furthermore, we find that unexpected earnings explain 4% or less of the variation in timing. Finally, we assess whether abnormal returns behave in a manner that is consistent with a good news early, bad news late relation.     

Reverse stock splits and earnings performance

This paper presents evidence that reverse stock splits are preceded by significantly poorer earnings performance for splitting firms compared to a sample of matched control firms. Interestingly, the overall earnings-returns relationship becomes significantly stronger following the reverse stock split. I interpret this as evidence that reverse splits communicate to market participants that sub-par earnings performance before the split is not transitory and that it is expected to persist in the future. Together, the evidence in this paper provides an explanation as to why reverse splits, which are employed for reasons that are seemingly beneficial to shareholders, are assessed negatively, on balance, by market participants.     

Effect of foreign GAAP earnings and Form 20-F reconciliations on revisions of analysts' forecasts

A controversial area of U.S. securities regulations involves the Securities and Exchange Commission's (SEC) financial reporting requirements for foreign firms, specifically, the necessity of providing a quantitative reconciliation to U.S. GAAP (Form 20-F). The results of earnings-returns research to date indicate that the release of foreign GAAP earnings provides important information. However, the results of earnings-returns studies using reconciled information are mixed. Instead of using an earnings-returns methodology adopted in prior research, this study utilizes analysts' revisions as a market indicator of the effect of information released in foreign GAAP earnings and the reconciled information in Form 20-F. Additionally, the study investigates the influence of four firm-specific variables in the firm's information environment—similarities of accounting systems, analyst following, difference between reconciled earnings and foreign GAAP earnings, and dispersion of analysts' expectations—on positive abnormal revision activities of financial analysts at the time of filing Form 20-F.The results indicate that the release of foreign GAAP earnings (at earnings announcement dates) and reconciled information (at the time of filing Form 20-F) contains relevant information as measured by analysts' revisions. Further, variables representing analyst following, change in reconciled earnings, and dispersion of analysts' expectations are significant in explaining the variation observed in positive abnormal revisions.     

Earnings Persistence, Losses and the Estimation of Earnings Response Coefficients

One of the major themes of capital markets accounting research concerns mapping the relation between accounting earnings and security returns. There is still not agreement on the functional form of this relation. The models analysed here are those where: the level of earnings alone, the change in earnings alone, or both, scaled by price, are used as explanatory variables for returns. This article demonstrates that if earnings are either completely permanent or entirely transitory, the earnings response coefficients (ERCs) estimated by levels and changes models should coincide. However, if earnings comprise a mixed process of permanent and transitory components, the ERC estimated by the levels will differ from that estimated by the changes model. Using losses to identify transitory components in earnings, empirical evidence consistent with these predictions is provided.
A combined model using both the level of, and change in, earnings is justified as a weighted average of an earnings and a book value valuation model (e.g., Ohlson, 1989). An alternative rationalization concerns the mitigation of an errors-in-variables problem associated with the estimation of unexpected earnings (Ali and Zarowin, 1992). The results for the combined model are more consistent with the latter. In this context, some previous empirical studies perceive the levels variable as a useful addition to the changes variable when there are transitory components in earnings. However, the evidence reported here suggests that the level of earnings, scaled by price, appears to be the fundamental earnings explanatory variable for returns (Ohlson, 1991, p. 1). The changes variable can, when the errors-in-variables problem is not mitigated by other methods, be a useful addition to the levels variable.     

Value relevance of value-at-risk disclosure

The SEC issued FRR No. 48 in 1997 to enhance public disclosure of firms’ exposures to market risk. We examine whether the quantitative value-at-risk (VAR) estimates disclosed by 81 non-financial firms during the period 1997–2002 are value-relevant using the earnings-returns relation. The empirical results indicate that high VAR is associated with weaker earnings-returns relation. Further analysis shows that VAR is positively and significantly associated with future stock return volatility. Our evidence suggests that investors perceive the earnings of firms with substantial market risk exposure to be less persistent, and adjust the future abnormal earnings for the higher risk exposure. Thus, this results in a lower expected rate of return.

Earnings, cashflows and returns: Functional relations and the impact of firm size

The relation between stock returns, earnings and cashflows is of importance because it directly addresses the issue of whether accounting data provide value relevant information. The empirical evidence to date, however, has documented low explanatory power for earnings and inconclusive incremental information content for cashflows. This research re-evaluates the incremental information content debate using Australian data. Our research is motivated by: recent innovations in research design, including the specification of nonlinear functional relations between accounting variables and prices, and the fact that differences in firm size characteristics may influence the relative information content of the accounting variables. We observe that: (i) a nonlinear functional relation provides greater explanatory power for both earnings and cashflows;(ii) the results are consistent with more transitory earnings components for smaller firms; and (iii) contrary to received theory, cashflows add greater incremental explanatory power for large firms.     

The Relation between Stock Prices and Accounting Information

The functional relation between expected stock prices and accounting information is analyzed through the theory of inverse probability. The approach models the mean of the posterior distribution for price, given the information that the accounting process provides. The implications of alternative assumptions about accounting measurement error and the unconditional price distribution are discussed. Our most refined model is consistent with recent empirical evidence showing convexity in the relationship between price and accounting information. Empirical tests, while exploratory, provide further evidence of a nonlinear relation between stock price and accounting measures of earnings and book value.     

Corporate Financial Policy,Information, and Market Expectations: An Empirical Investigation of Dividends

This paper documents a relationship between announcements of unexpected changes in financial policy and unexpected changes in performance of the firm. Using a new methodology that combines analysis of stock price movements and earnings forecast data, the authors provide evidence that analysts revise their earnings forecasts following the announcement of an unexpected dividend change by an amount positively related to the size of the unexpected dividend change. They also provide evidence that these revisions are positively related to the change in equity value surrounding the announcement. Further, they find that these revisions are consistent with rationality. Their results therefore provide direct evidence consistent with the hypothesis that unexpected dividend changes signal information about firm performance to market participants.     

The effect of changes in Japanese consolidation policy on analyst forecast error

Our broad research objective is to investigate whether convergence towards international standards improves the decision usefulness of information. Recent changes in Japanese consolidated reporting practices to better align with international standards provide an excellent setting to investigate this research objective. Specifically, we examine the effect of changes in Japanese consolidation policy on financial analysts’ perceptions of the persistence of subsidiary earnings. Previous research provides evidence that, prior to the change in consolidation policy, consolidated financial information was not used efficiently in the Japanese capital market. Prior research finds a positive relation between subsidiary earnings and future stock returns in Japan, indicating that investors underestimate the persistence of subsidiary earnings. Consistent with prior research using stock returns, we find that financial analysts also underestimate the persistence of subsidiary earnings in Japan. We document a significant positive relation between subsidiary earnings and future forecast errors of consolidated earnings. However, following the changes in consolidation policy in Japan, we find that financial analysts no longer underestimate the persistence of subsidiary earnings. Changes in Japanese consolidation policy in conformance with international standards increase decision usefulness by improving the ability of financial analysts to predict overall firm performance. One limitation of our research design relates to the adoption of mandated accounting policy changes by all sample firms in the same calendar time. This makes it difficult to control for the impact of correlated omitted variables. While this concern can never be completely eliminated, we provide additional tests that examine sample partitions by firm size and industry. These additional tests support the primary findings that Japan’s efforts to converge consolidation rules with international standards have improved analysts’ consolidated earnings forecasts.     

Information Content of Interim Earnings Components—Evidence from Finland

This study examines the information content of the interim earnings of listed Finnish firms. The degree of association between returns and interim earnings is studied. The assumption is that, compared to transitory changes, permanent changes in earnings have a greater association with returns. Unexpected returns are regressed over unexpected permanent earnings and unexpected transitory earnings. Three return measurement periods are used to examine any potential asynchrony between prices and earnings. To reduce the errors-in-variables problem expected in single-security-level studies, observations are grouped into portfolios. When the data are divided into portfolios, the results give evidence of the association as hypothesized.     

Is ‘benchmark beating’ by Australian firms evidence of earnings management?

We investigate the extent to which Australian firms that report small profits and/or small increases in earnings (i.e. benchmark beaters) have done so by the upward manipulation of these earnings. Although evidence of an unusually large number of firms managing to just beat such earnings benchmarks has been interpreted as evidence of earnings management, this approach fails to identify those firms that are the manipulators from those where unbiased earnings fall naturally into the benchmark beating group. Our results suggest that caution is required in interpreting benchmark beating as an indicator of the extent of earnings management. Using several methods for estimating the unexpected accrual component of earnings, we show that although benchmark beaters have larger positive unexpected accruals than other firms, a similar result holds when firms with small losses or earnings declines (i.e. ‘just miss’ firms) are compared with other firms. Moreover, there is no statistically significant difference between unexpected accruals for the benchmark beating and just miss groups. At a minimum, we reject the joint hypothesis that unexpected accruals capture earnings management and that an unusual kink around zero in the distribution of earnings levels or earnings changes is caused by earnings management.     

The pricing of conservative accounting and the measurement of conservatism at the firm-year level

This paper analyzes the relation between equity prices and conditional conservatism and introduces a new measure of conservatism at the firm-year level. We show that the asymmetric properties of conservative accounting, the existence of non-accounting sources of information, and the properties of GAAP related to special items combine to generate a nonlinear relation between unexpected equity returns and earnings news (the shock to expected current and future earnings). Based on this model, we construct a conservatism ratio (CR) defined as the ratio of the current earnings shock to earnings news. CR measures the proportion of the total shock to expected current and future earnings recognized in current year earnings. Ranking firms according to CR, we show empirically that higher CR firms have more leverage, increased volatility of returns, more incidence of losses, more negative accruals, and increased volatility of earnings and accruals, consistent with the literature on conservative accounting.     

Cost behavior and executive bonus compensation

Prior literature provides compelling evidence of an asymmetric relation between executive bonus compensation and earnings performance. In particular, this literature reports that compensation committees assign greater weight to good (positive) earnings performance than poor (negative) earnings performance. Taken together, the prior literature provides strong support for critics who claim that compensation committees blindly protect executives from earnings underperformance. We further examine this issue by investigating whether a firm's cost behavior (i.e., the relation between expenses and sales) provides an explanation for the apparent inefficiency in executive compensation contracts. Our evidence suggests that executives are rewarded more for increases in ROA that arise from normal cost behavior than other increases in ROA consistent with these increases being perceived as more persistent. In contrast, we do not find such a relationship for decreases in ROA which suggests that executives are largely shielded from decreases in ROA that follow normal cost behavior. We examine two factors suggested by the prior literature, expected future sales and the extent of capacity utilization, which may provide an explanation for why executives are shielded from normal cost behavior decreases in ROA. When these additional factors are included in our empirical models, our evidence suggests that the asymmetric relation between changes in CEO bonus compensation and increases and decreases in earnings performance documented in prior literature goes away. That is, our results suggest that compensation committees do not blindly protect executives for earnings underperformance. On the contrary, our evidence suggests that these committees take into account other non-earnings information when deciding how much weight to give to a decrease in earnings and that executive compensation may not be as inefficient as suggested by prior research.     

Voluntary Causal Disclosures: Tendencies and Capital Market Reaction

In this study, we provide empirical evidence on whether voluntarily disclosed causal attributions made in management earnings forecasts are credible by investigating the conditions under which such attributions are made and the extent to which security price responses are associated with attribution existence. We find that causal attributions are more likely to be made when forecast news is bad (relative to good), and that the type of attribution made is more likely to be external (internal) for bad (good) forecast news. Incorporating the existence and type of attribution into models that explain announcement period three-day cumulative abnormal returns yields significant effects for attribution incidence and type after controlling for unexpected earnings and forecast type (e.g., point, range, etc.). Consistent with the idea that attributions enhance the credibility or precision of management forecasts, attribution disclosure enhances price reactions per dollar of unexpected earnings conveyed in a management forecast.     

Analysts' forecasts as earnings expectations

I examine three composite analyst forecast of earnings per share as proxies for expected earnings. The most current forecast weakly dominates the mean and median forecasts in accuracy. This is evidence that forecast dates are more relevant for determining accuracy than individual error. Consistent with previous research, I find analysts more accurate than time-series models. However prior knowledge of forecast errors from a quarterly autoregressive model predicts excess stock returns better than prior knowledge of analysts' errors. This is inconsistent with previous research, and is anomalous given analysts' greater accuracy.     

THE ASSOCIATION BETWEEN ANALYSTS’ FORECAST REVISIONS AND EARNINGS COMPONENTS: THE EVIDENCE OF FRS 3

This study examines whether the earnings components as required by FRS 3 help UK analysts to predict firms’ earnings changes by investigating the statistical association between analysts’ forecast revisions and firms’ unexpected earnings components. I find that analysts’ forecast revisions made in different time horizons are consistently associated with unexpected earnings components as required by FRS 3. UK analysts are able to incorporate current-year unexpected earnings components into their current and future earnings forecasts even before firms officially release this information. However, empirical results also show that current-year unexpected earnings components are not fully incorporated into analysts’ forecasts of future earnings. Analysts appear to wait for more information releases regarding firms’ future earnings and delay their revisions of future earnings forecasts. This is consistent with the evidence that the cumulative revisions of current earnings forecasts are generally associated with prior-year unexpected earnings components, and the association appears to be stronger as time progresses. Overall, this study provides evidence suggesting that the earnings components as required by FRS 3 help UK analysts to identify firms’ permanent and transitory earnings changes over different forecast horizons. This study also provides strong evidence supporting the informativeness of earnings components for analysts’ forecasts and the information set perspective of FRS 3 that highlights the importance of earnings components in predicting a reporting entity’s future performance.     

Business Cycles and the Relation between Security Returns and Earnings

This paper examines business cycle variation in the earnings-returns relation. Earnings are more persistent when growth rates are high (i.e., in an expansion) than when growth rates are low (i.e., in a recession). Earnings are more persistent when production is high (i.e., in a credit crunch period) than when production is low (i.e., in a reliquification period). Relatedly, earnings response coefficients are larger in expansions (credit crunch periods) than in recessions (reliquification periods). Thus, earnings persistence and earnings response coefficients are positively associated with the rate of growth in economic activity and the level of economic activity.