How do analysts forecast earnings

In summary, we find that the CM represents market expectations most consistently among all tested models. The ICC is a popular proxy for expected returns see e.

Better earnings estimates should improve the correlation between the ICC and subsequent realized returns leading to more useful ICC estimates. First, we compute the ICC on an aggregate level and evaluate its ability to predict realized returns over time. Then, we analyze the cross-sectional correlation between ICC and ex-post forward returns.

There is evidence that the ICC at an aggregate level is a good proxy for time-varying expected returns e. In this section, we test whether the slopes from a regression of realized market returns on the ICC, computed using different methods to forecast earnings, are greater than zero. Panel A of Table 6 presents the results. For the one-year-forward return predictive regressions, we document that the ICC estimated with earnings from the CM offers the largest number of significant regression slopes.

In the previous section, we compared the predictive power of the ICC over time. Now, we analyze whether the ICC has a positive correlation with the cross-section of stock returns.

The results are reported in Panel B of Table 6. However, this finding might be driven by small and micro-cap stocks as the FM regressions weight the observations equally Novy-Marx An additional shortcoming of FM regressions is that they are sensitive to outliers.

To address these potential issues, we analyze the performance of value-weighted portfolios sorted by their ICC. Table 7 presents annual excess returns in excess of the risk-free return. The stocks are sorted into quintiles and deciles based on their respective ICC at the end of June each year from to In addition, we include a Composite ICC, which is the average of the above-mentioned approaches.

To compute excess returns, we use the one-month Treasury bill rate. This may be due to a different return frequency used to compute t-statistics, different sample periods, and different stock universes. In summary, the ICC estimated with the CM reports a stronger correlation with returns compared to the other models. The results hold for both dimensions, over-time and cross-sectionally. We evaluate whether a set of firm characteristics that have been used to explain the cross-sectional variation of expected returns proxied by average realized returns also have the same relation when the ICC as a proxy for expected returns is used.

The independent variables are firm characteristics available prior to the end of June of year t. We estimate the ICC Footnote 30 based on different proxies of earnings forecasts at the end of June of each year.

We use the following firm characteristics. Size is the natural logarithm of market equity at the end of June in year t. Gross profitability is the ratio of gross profits to total assets e. Leverage is book value of debt divided by book equity. In Table 8 , we provide the average of the FM regression coefficients estimated yearly for the period from June to June and the respective t-statistics with Newey-West adjustment.

These results are similar to Hou et al. All proxies of expected returns have positive coefficients for idiosyncratic volatility. However, the coefficients are statistically significant only for the ICC with earnings forecasts derived from the CM t-statistic of 2. The results for asset growth are interesting since we are able to confirm the negative cross-sectional relation of asset growth and returns, also shown in Aharoni et al.

The size effect is stronger when we use the ICC as a proxy for expected returns than when realized returns are used. The ICCs based on any of the tested earnings forecasts methods show significant coefficients at the 0. When we analyze the relation of size and forward realized returns, the coefficient is not statistically significant.

Concerning the value effect, the coefficients of ln beme are positive and statistically significant for all proxies of expected returns, but the t-statistics are higher when the ICC is used as a proxy for expected returns than when the ex-post realized returns are used. This is not surprising as the ICC is a more sophisticated value measure and is therefore highly correlated with the value factor e.

According to Novy-Marx , gross profitability has a positive and significant relation to returns. In our study, we confirm these results when using realized returns as the dependent variable as the corresponding coefficient is 5. However, when we analyze the ICCs with earnings forecasts from the HVZ model and the RI model, the results show a negative and significant relation, with a t-statistic of 6.

In this section, we address two potential points of criticism regarding our CM. To test if these improvements deliver similar results than our CM, we implement the procedure from Mohanram and Gode henceforth MG model. Then we compute forecast bias, accuracy and ERC based on street earnings.

Mohanram and Gode propose a model to remove the predictable errors, which can be used to estimate ICC. The authors show that the ICC estimated with the MG model has a stronger cross-sectional correlation with returns. We closely follow Mohanram and Gode Footnote 34 when implementing their model and subsequently compare the results to our analyzed models.

Regarding bias and accuracy, we do not find any statistically significant one-year or two-year ahead mean bias, but we find a positive and statistically significant median bias for two-year-ahead earnings forecasts 0.

These results are in line with the findings from Hou et al. Moreover, we use the long-term growth LTG forecasts to estimate the three-year ahead earnings forecasts and set the short-term growth STG forecast equal to LTG in the estimation of the OJ and MPEG methods when the two-year ahead forecast is below the one-year-ahead one. Table 10 displays the results from cross-sectional tests of realized returns and ICC estimates based on the MG model.

From the five ICC approaches tested, two report a statistically significant relation with the cross-section of returns. Only one of the long-short strategies based on the ICC from the MG model is statistically different from zero, while three strategies based on the CM display positive and significant returns. To sum up, although the tested adjustments improve forecast performance compared to their unadjusted implementations, we confirm that the CM still scores better in terms of accuracy, bias, ERC, and when looking at the correlation between ICC and realized returns.

In this study, we develop a new method to forecast corporate earnings. First, we include gross profits, as Novy-Marx finds a strong association with earnings. Second, we follow Ashton and Wang , who show that stock price changes drive earnings, by including recent stock market performance. This also mitigates the fact that analysts, on average, need longer to incorporate new information into their earnings forecasts than it takes the stock market to incorporate new information into the share price Guay et al.

We compare our new approach, the CM, to several methods from the literature, namely raw analyst forecasts, the model by Hou et al. We find that our CM has the lowest bias and highest accuracy among all the tested models.

Regarding market expectations, we show that the CM also performs better than the other benchmark models. Furthermore, we compute the ICC based on the different earnings forecast models and find that the CM leads to ICC estimates that have the strongest association with subsequent realized stock returns.

We confirm that this improves their forecast performance, but they still trail behind our CM. This new method makes a strong case for combining two different approaches to forecast earnings: human forecasts made by financial analysts and cross-sectional forecasts based purely on financial data. These two approaches have distinct advantages and disadvantages. On the other hand, cross-sectional forecasts are unbiased, but not as accurate.

Combining them into one model mitigates both disadvantages while conserving the advantages. Our findings are relevant for practitioners working with earnings forecasts, as well as academics employing earnings forecasts as inputs in valuation models, such as the ICC. We recommend the use of our CM to improve the accuracy and unbiasedness of earnings forecasts, which benefits methods that build on these forecasts and applications thereof.

Another shortcoming of time-series models is that they suffer from survivorship bias as only firms with a long history of earnings can be included in the model.

Guay et al. They suggest to use short-term stock returns to mitigate the effect of sluggish analysts forecasts. For instance, Bradshaw et al. Furthermore, Novy-Marx shows that gross profits explain many earnings-related asset pricing anomalies, such as return on assets, earnings-to-price, asset turnover, gross margins, and standardized unexpected earnings.

Claus and Thomas , p. Easton and Monahan analyze the cross-sectional correlation between returns and different ICC approaches and find that none of the ICC estimates has a positive association with returns. The authors conclude that the ICC estimates are unreliable for the entire cross-section of firms. We include the RW based on evidence that at a one-year horizon, the RW model performs as well as more sophisticated estimation methods Gerakos and Gramacy According to Hou et al.

We include the Earnings Persistence and Residual Income Models because Li and Mohanram find that these models outperform the HVZ model in terms of forecast bias, accuracy, earnings response coefficient, and correlation of ICCs with future earnings and risk factors.

We use dollar level data for our estimations and not share level data because our subsequent applications ICC computation rely on the clean surplus relation. The assumption of clean surplus accounting is more likely to be violated at the share level than at the dollar level.

In unreported results, we perform the regressions of the mechanical models with street earnings instead of GAAP earnings and, in most cases, it leads to lower accuracy, weaker earnings response coefficients, and higher bias. We use the change in market value instead of stock returns because we aim to estimate earnings at the dollar level.

By multiplying lagged market value with the stock return we ensure that this measure is not affected by any net stock issues. Although the variable is significant in the in-sample regressions, the inferences are unchanged if the variable is removed.

In unreported tests we confirm that the main results still hold when performing the regression at the per share level. Although the CT and GLS approaches are both based on a residual income valuation model, the methods have an important difference. While the CT model is designed to compute the market-level cost of capital, the GLS model is better suited to compute the firm-level cost of capital. We estimate one, two, and three-year-ahead forecast bias for the periods —, —, and , respectively.

According to Clement et al. We estimate one-, two-, and three-year-ahead forecast accuracy for the periods —, —, and —, respectively. Brown , p. Following Li et al. We use a one-sided test to analyze the correlation between ICC and returns over-time as well as cross-sectionally.

We do not include the RW model because this method does not allow for earnings growth and is, therefore, not suitable for estimating the ICC. For the sake of brevity, following Hou et al. We focus our analysis on the sign and the significance of the coefficients since there is no evidence that a larger or smaller coefficient would lead to a better model in such a regression.

As in the CM, we benchmark these forecasts against street earnings, which is the earnings standard adopted by the analysts. In unreported results, we perform the regressions with GAAP earnings as realized earnings, and it leads to lower accuracy, weaker earnings response coefficients, and similar bias. Although there are other models that remove predictable forecast errors to estimate ICC, such as Larocque , we follow the methodology from Mohanram and Gode because the authors show a significant increase in correlation between ICC and realized returns when using their method.

The MG model is based on a two-stage regression. In the first stage, the authors regress one-year SURP1 and two-year ahead SURP2 earnings surprise on lagged underreaction and overreaction variables based on Hughes et al. In the second stage, the coefficients of the first-step regression are multiplied by the current underreaction and overreaction variables. Following Mohanram and Gode , we truncate the variables yearly at the 0.

The model is estimated in June of each year and one-year two-year ahead forecasts are estimated as the sum of one-year two-year ahead analysts forecasts and the estimated SURP1 SURP2. The residual income model is also commonly referred to as the Edwards—Bell—Ohlson valuation model. Total accruals are based on Richardson et al. Abarbanell J, Lehavy R Biased forecasts or biased earnings? J Account Econ 36 1—3 — Google Scholar.

J Account Econ 14 2 — J Financ Econ 2 — J Finance 61 1 — Ashton D, Wang P Terminal valuations, growth rates and the implied cost of capital. Rev Account Stud 18 1 — Ball R, Brown P An empirical evaluation of accounting income numbers.

J Account Res 6 2 — Management Science, Forthcoming. Rev Quant Finance Account 52 3 — J Finance 72 2 — Rev Account Stud 17 4 — J Account Res 40 1 — Brown L, Richardson G, Schwager S An information interpretation of financial analyst superiority in forecasting earnings. J Account Econ 25 1 — J Account Res 53 1 :1— Int J Forecast 9 3 — Carhart MM On persistence in mutual fund performance.

J Finance 52 1 — Create a personalised ads profile. Select personalised ads. Apply market research to generate audience insights.

Measure content performance. Develop and improve products. List of Partners vendors. Your Money. Personal Finance. Your Practice. Popular Courses. Fundamental Analysis Tools for Fundamental Analysis. Table of Contents Expand. Forecasting Revenue. Forecasting Growth.

Growth Rate Calculation. Impact of Forecasts on Valuation. The Bottom Line. Compare Accounts. The offers that appear in this table are from partnerships from which Investopedia receives compensation.

This compensation may impact how and where listings appear. Investopedia does not include all offers available in the marketplace. Related Articles. Fundamental Analysis Analyzing Retail Stocks.

Entrepreneurs Valuing Startup Ventures. Partner Links. What Is Fundamental Analysis? Many investors rely on earnings performance to make their investment decisions.

Stocks are assessed according to their ability to increase earnings as well as to meet or beat analysts' consensus estimates. This influences a company's implicit value i. The basic measurement of earnings is earnings per share.

This metric is calculated as the company's net earnings—or net income found on its income statement—minus dividends on preferred stock, divided by the number of outstanding shares.

So, why does the investment community focus on earnings, rather than other metrics such as sales or cash flow? Any finance professor will tell you the only proper way to value a stock is to predict the long-term free cash flows of a company, discount those free cash flows to the present day and divide by the number of shares.

But this is much easier said than done, so investors often shortcut the process by using accounting earnings as a "good enough" substitute for free cash flow. Accounting earnings certainly are a much better proxy for free cash flow than sales. Besides, accounting earnings are fairly well defined and public companies' earnings statements must go through rigorous accounting audits before they are released. As a result, the investment community views earnings as a fairly reliable, not to mention convenient, measure.

To read more, see: Getting The Real Earnings. Earnings forecasts are based on analysts' expectations of company growth and profitability. To predict earnings, most analysts build financial models that estimate prospective revenues and costs. Many analysts will incorporate top-down factors such as economic growth rates , currencies and other macroeconomic factors that influence corporate growth. They use market research reports to get a sense of underlying growth trends.

To understand the dynamics of the individual companies they cover, really good analysts will speak to customers, suppliers and competitors. The companies themselves offer earnings guidance that analysts build into the models.

To predict revenues, analysts estimate sales volume growth and estimate the prices companies can charge for the products. On the cost side, analysts look at expected changes in the costs of running the business.

Costs include wages, materials used in production, marketing and sales costs, interest on loans, etc. Analysts' forecasts are critical because they contribute to investors' valuation models. Institutional investors, who can move markets due to the volume of assets they manage, follow analysts at big brokerage houses to varying degrees.

Consensus estimates are so consistently tracked by so many stock market players that when a company misses forecasts, it can send a stock tumbling; similarly, a stock that merely meets forecasts might get sent lower, as investors have already priced in the in-line earnings. Consensus estimates are so powerful that even small deviations can send a stock higher or lower. If a company exceeds its consensus estimates, it is usually rewarded with an increase in stock price.

If a company falls short of consensus numbers—or sometimes if it only meets expectations—its share price can take a hit. With so many investors watching consensus numbers, the difference between actual and consensus earnings is perhaps the single most important factor driving share-price performance over the short term.

This should come as little surprise to anyone who has owned a stock that "missed the consensus" by a few pennies per share and, as a result, tumbled in value. For better or for worse, the investment community relies on earnings as its key metric. Stocks are judged not only by their ability to increase earnings quarter over quarter but also by whether they are able to meet or beat a consensus earnings estimate. Like it or not, investors need to keep an eye on consensus numbers to keep tabs on how a stock is likely to perform.

Zacks Investment Research. Business Essentials. Financial Ratios. Tools for Fundamental Analysis.