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.     

Team earnings forecasting

While brokerage houses use both teams of sell-side analysts and individual analysts to conduct earnings research, there is no empirical research examining whether teams and individuals differ with regard to their forecasting performance or purpose. We first examine the most-often researched dimension of forecasting performance, earnings forecast accuracy, and show that teams are less accurate than individual analysts in general and their own individual team members in particular. We conjecture that teams focus their efforts on an alternative dimension of forecasting performance, timeliness, and show that team forecasts are timelier than those of individual analysts in general and their own individual team members in particular. Consistent with the notion that teams trade-off forecast accuracy for timeliness to comply with a market research demand, we show that team forecast revisions are associated with larger market responses than those of individuals. Finally, we illuminate the nature of team assignments by documenting that the firms that teams follow are in greater financial distress and larger in size.     

Additional Evidence on Analysts’ Decision to Issue Disaggregated Earnings Forecasts: Strategic Biasing

This study examines why analysts issue disaggregated earnings forecasts to I/B/E/S. Some recent studies suggest that analysts with superior forecasting ability issue disaggregated earnings forecasts to build reputation in the marketplace and stop forecast disaggregation once their reputation has been established. Based on an analysis of I/B/E/S forecast data for U.S. firms from 1998 to 2008, we find that, in a given year, about 20%–34% of analysts disaggregate for some, but not for all the firms that they follow. This evidence of selective disaggregation by analysts suggests that reputation building alone does not fully explain the decision to disaggregate forecasts. We hypothesize that the decision to disaggregate earnings forecasts is at the firm‐level as well and is systematically related to the analysts’ bias in the issued forecasts. Our findings are that (a) analysts’ overall optimistic bias and forecast errors decrease monotonically with the level of forecast disaggregation, and (b) analysts that selectively disaggregate their forecasts for some firms or who do not persistently disaggregate a given firm's forecasts exhibit more positive bias and larger forecast errors. Our findings are consistent with the notion that the analysts who issue biased forecasts, for example, to curry favour with the management, are less likely to provide disaggregated information as part of the forecast.     

What do professional forecasters' stock market expectations tell us about herding,information extraction and beauty contests?

We study how professional forecasters form equity market expectations based on a new micro-level dataset which includes rich cross-sectional information about individual characteristics. We focus on testing whether agents rely on the beliefs of others, i.e., consensus expectations, when forming their own forecast. We find strong evidence that the average of all forecasters' beliefs influences an individual's own forecast. This effect is stronger for young and less experienced forecasters as well as forecasters whose pay depends more on performance relative to a benchmark. Further tests indicate that neither information extraction to incorporate dispersed private information, nor herding for reputational reasons can fully explain these results, leaving Keynes' beauty contest argument as a potential candidate for explaining forecaster behavior.     

Understanding analysts' use of stock returns and other analysts' revisions when forecasting earnings

We investigate analysts' use of stock returns and other analysts' forecast revisions in revising their own forecasts after an earnings announcement. We find that analysts respond more strongly to these signals when the signals are more informative about future earnings changes. Although analysts underreact to these signals on average, we find that analysts who are most sensitive to signal informativeness achieve superior forecast accuracy relative to their peers and have a greater influence on the market. The results suggest that the ability to extract information from the actions of others serves as one source of analyst expertise.     

Initial public offering prospectus forecast errors and mandatory explanations: evidence from New Zealand

This paper examines management forecast errors in initial public offering (IPO) prospectuses of New Zealand firms and subsequent management explanations for earnings forecast errors in recent years. New Zealand has several unique features and recent changes that are worthy of research, including the requirement for management to make forecasts; a requirement to explain differences between forecast and actual; and a recent change that provides the voluntary opportunity to obtain a negative assurance opinion on forecasts. Using hand-collected IPO data between 1998 and 2014, we find that firms that include a negative assurance opinion on the prospective financial information in the prospectus have more accurate earnings and cash flow forecasts. We investigate the subsequent explanations for forecast errors, which are mandated by a financial reporting standard. We show that management tends to attribute negative and large forecast errors to external causes rather than their own actions. Our findings have implications for regulators and investors in New Zealand and other settings.     

CEO happiness and forecasting

We utilize the IBM Watson Tone Analyzer to measure chief executive officers' (CEOs') levels of joy (happiness) in year-end conference calls, and empirically test how CEOs' happiness affects the properties of their own and analysts' forecasts. We find that joyful CEOs are more likely to issue forecasts, less likely to miss their forecast targets, and exhibit lower optimistic bias in their forecasts. When joyful CEOs issue earnings forecasts, analysts revise their forecasts upwards and produce forecasts that are less dispersed and more accurate. Our results demonstrate that inherent CEO happiness significantly impacts the forecast properties of both managers and analysts, thus supporting upper echelons theory.     

Identifying unsound firms in Italy: An attempt to use trend variables

The aims of the present study are (i) to develop a predictive model, using discriminant analysis to forecast the solvency of Italian firms and (ii) to compare the ability of this model with that of a second one in which static ratios have been replaced by values that take account of their own trends. The results obtained by considering the trends of ratios over three years are encouraging and indicate that the second model not only performed better than the first, but also better than any other model using the balance sheet data of Italian firms.     

Stock market openness and analyst forecast bias

Biases in analysts’ forecasts can be reduced not only through regulation but also through market mechanisms. In 2014, China launched the Shanghai-Hong Kong Connect program, which opened part of its domestic equity market to foreign investors. The implementation of this program provides a quasi-natural experimental setting to explore whether stock market openness plays a governance role in brokerage firms and minimizes their affiliated analysts' forecast biases. We find that the participation of foreign institutional investors mitigates the forecast biases of affiliated analysts. We also show that these analysts exert more significant effort by conducting more site visits. Our findings suggest that market liberalization can help improving the quality of analysts’ forecasts.     

Heterogeneous Beliefs, Speculation, and the Equity Premium

Agents with heterogeneous beliefs about fundamental growth do not share risks perfectly but instead speculate with each other on the relative accuracy of their models' predictions. They face the risk that market prices move more in line with the trading models of competing agents than with their own. Less risk‐averse agents speculate more aggressively and demand higher risk premiums. My calibrated model generates countercyclical consumption volatility, earnings forecast dispersion, and cross‐sectional consumption dispersion. With a risk aversion coefficient less than one, agents' speculation causes half the observed equity premium and lowers the riskless rate by about 1%.     

On the Determinants of Optimism in Financial Analyst Earnings Forecasts: The Effect of the Market's Ability to Adjust for the Bias

This paper examines whether the documented bias in analyst earnings forecasts is intentional by examining whether it is related to the market's ability to adjust for this bias. For intentional bias to exist it is not enough for analysts to face incentives but rather, analysts should also be willing to respond to these incentives. As the market's ability to adjust for the bias increases, its market effects decrease while analyst reputation costs increase reducing analyst willingness to bias their forecasts. The paper utilizes a firm‐specific design that allows for both the bias component of the forecast error and the market's ability to adjust for the bias to be computed at the firm level. Results suggest that even though forecast error is positive in the latter part of the period under review reflecting overall analyst pessimism, the bias embedded in the forecasts is optimistic throughout the period. More importantly, I find that analyst forecast bias is decreasing in the market's ability to adjust for it. This result provides further evidence that analysts knowingly bias their forecasts and provides support for the existence of reporting bias, in particular. Thus, the evidence provides justification for recent regulatory efforts to increase the objectivity of analyst research reports.     

Rational Expectations and Security Analysts' Earnings Forecasts

This paper investigates the rationality of security analysts' forecasts. The forecasts of analysts participating in Lynch, Jones, and Ryan's Institutional Brokers Estimate System (I/BE/S) data base are evaluated relative to past values of their own forecast errors, past values of forecasted earnings per share, and quarterly percentage changes in publicly available macroeconomic and financial time series. The publicly available series include the consumer price index, unemployment rate, oil prices, stock prices, gross national product, and corporate profits. The authors conduct a generalized orthogonality test and include only information available to analysts at the time the forecasts are made. The empirical results reject analyst forecast rationality, but not without exception.     

The Effects of Disclosure and Analyst Regulations on the Relevance of Analyst Characteristics for Explaining Analyst Forecast Accuracy

We posit and find an effect of disclosure and analyst reporting regulations implemented from 2000 through 2003 (including Regulation Fair Disclosure, the Sarbanes‐Oxley Act and the Global Settlement Act) on the importance of analyst and forecast characteristics for analyst forecast accuracy. Following the enactment of these regulations, more experienced analysts and All‐Star analysts do not maintain their superior forecast accuracy, and analysts employed by large brokerage houses perform worse than other analysts. In addition, we find a decrease in the importance of analyst effort, the number of industries and firms followed, days elapsed since the last forecast, and forecast horizon. While the importance of bold upward forecast revisions does not change, bold downward revisions lose their relevance for forecast accuracy after 2003. Finally, we find an increase in the importance of prior forecast accuracy. We find that the importance of these characteristics varies with the precision of publicly available information. Specifically, the decrease in the importance of most analyst and forecast characteristics and the increase in the importance of prior forecast accuracy are greater when the precision of publicly available information is low. Overall, our results suggest that the positive effects of experience, effort, brokerage house size and All‐Star status on forecast accuracy in the pre‐regulation period were because of the information advantages that these analysts enjoyed (rather than their ability to generate private information). In contrast, our results suggest that prior forecast accuracy is related to analysts’ ability to generate private information.     

Evidence That Management Earnings Forecasts Do Not Fully Incorporate Information in Prior Forecast Errors

Abstract:  This paper investigates whether managers fully incorporate the implications of their prior earnings forecast errors into their future earnings forecasts and, if not, whether this behavior is related to the post-earnings announcement drift. I find a positive association in consecutive management forecast errors, suggesting that managers underestimate the future implications of past earnings information when forecasting earnings. I also find that managers underestimate the information in their prior forecast errors to a greater extent when they make earnings forecasts with a longer horizon. Finally, I find that, similar to managers, the market also underreacts to earnings information in management forecast errors, which leads to predictable stock returns following earnings announcements.     

High-Technology Intangibles and Analysts' Forecasts

This study examines the association between firms' intangible assets and properties of the information contained in analysts' earnings forecasts. We hypothesize that analysts will supplement firms' financial information by placing greater relative emphasis on their own private (or idiosyncratic) information when deriving their earnings forecasts for firms with significant intangible assets. Our evidence is consistent with this hypothesis. We find that the consensus in analysts' forecasts, measured as the correlation in analysts' forecast errors, is negatively associated with a firm's level of intangible assets. This result is robust to controlling for analyst uncertainty about a firm's future earnings, which we also find to be higher for firms with high levels of internally generated (and expensed) intangibles. Given that analyst uncertainty increases and analyst consensus decreases with the level of a firm's intangible assets, we also expect and find that the degree to which the mean forecast aggregates private information and is more accurate than an individual analyst's forecast increases with a firm's intangible assets. Finally, additional analysis reveals that lower levels of analyst consensus are associated with high-technology manufacturing companies, and that this association is explained by the relatively high R&D expenditures made by these firms. Overall, our results are consistent with financial analysts augmenting the financial reporting systems of firms with higher levels of intangible assets (in terms of contributing to more accurate earnings expectations), particularly R&D-driven high-tech manufacturers.     

14-Week quarters

Many firms define their fiscal quarters as 13-week periods so that each fiscal year contains 52 weeks, which leaves out one or two day(s) a year. To compensate, one extra week is added every fifth or sixth year and, consequently, one quarter therein comprises 14 weeks. We find evidence of predictable forecast errors and stock returns in 14-week quarters, suggesting that analysts and investors do not, on average, adjust their expectations for the extra week. The ease with which 14-week quarters can be predicted, and expectations adjusted, suggests a surprising lack of effort on the part of analysts and investors.     

Optimistic bias in management forecasts by Japanese firms to avoid forecasting losses

Listed firms in Japan are effectively compelled to report management forecasts of sales, ordinary income, and net income along with actual earnings and sales each year. Prior studies report that Japanese managers tend to announce optimistic forecasts of earnings. We show that a large part (61.6%) of the overall optimistic bias in management earnings forecasts in Japan can be explained by loss forecast avoiding behavior of a small fraction (5.25%) of firms. Such behavior is caused in part by the view of the main bank and power group that the management forecast of earnings is the manager's earnings target. Our findings suggest that the Japanese stock market recognizes such loss forecast avoidance and accordingly discounts new information in management forecasts.     

Does foreign exchange risk matter to equity research analysts when forecasting stock prices? Evidence from U.S. firms

We study the relationship between the foreign exchange risk and analyst target price forecast error using U.S. firm-level data for the sample period between 1999 and 2014. We find that the target price forecast error is higher when foreign exchange risk is higher. The relationship is stronger for smaller firms and less pronounced among financial firms. Collectively, the findings suggest that analysts make fewer errors when forecasting for firms that are more capable of managing foreign exchange risks. These findings imply that either analyst does not make a significant effort in incorporating foreign exchange risk into their forecast models, or they do not have the skills for that task.     

Assessing exchange rate dynamics of East Africa: fragmented or integrated?

This article investigates the dynamics of the currency markets of the East African Community, using forecast error variance decompositions from vector autoregressions. It shows that the exchange rates of Kenya, Tanzania and Uganda have been mainly driven by shocks to their own economies, while those of Burundi and Rwanda have been increasingly dictated by spillovers from the dollar and euro since the global financial crisis. Interactions within the region are limited, although there is some sign of elevation. This makes a clear contrast with European currency markets prior to the euro, where spillovers from the German mark dominated the markets.     

How do financial analysts interpret industrial firms' corporate refocusing announcements?

This study investigates how analysts perceive the effect of corporate refocusing announcements on UK industrial firms' future earnings by examining current-year and one-year-ahead earnings forecast revisions, current-year target price revisions and earnings forecast errors in the five years surrounding a refocusing announcement year. The results reveal that analysts adjust their earnings forecasts downward in a refocusing announcement year and the following two years, predicting that operating performance in the post-refocusing period is likely to decline relative to their former earnings forecasts. Secondly, there is no evidence that analysts issue biased earnings forecasts after refocusing announcements or that their forecasts appear less accurate. Thirdly, they adjust their earnings forecasts downward in a refocusing announcement year with downward market movement. However, they do not similarly adjust their earnings forecast upward with upward market movement. The magnitude of downward adjustments exceeds that of upward adjustments. They also adjust current-year target price forecasts downward with downward market movement in the year prior to a refocusing announcement.