Skip to main content
SpringerLink
Log in

Analysts’ annual earnings forecasts and changes to the I/B/E/S database

  • Published:
Review of Accounting Studies Aims and scope Submit manuscript

Abstract

I/B/E/S is a common source of analyst earnings forecast data, and the reliability of these data is important for practice and academic research. Examining a common sample period, we compare annual earnings forecasts across two versions of the I/B/E/S detail file, one made available in 2009 and the other made available in 2015. We find substantial differences in the contents of these two versions of the detail file as well as significant differences in the attributes of the earnings forecasts available in each version. Specifically, the earnings forecasts in the more recent version are more accurate and less biased, and they identify substantially different firms as meeting or just beating analysts’ expectations than those in the older version. To highlight the potential impact of these differences, we show that the economic magnitude of the effects of analyst experience and brokerage size on earnings forecast accuracy change by over 30% when we use the more recent version. Additional analyses suggest that the differences across versions of the detail file are ongoing. In contrast, we find that different versions of the summary file exhibit only minor differences over time. We also find significant differences in the properties of consensus earnings forecasts calculated from the individual earnings forecasts available in the detail file and consensus earnings estimates from the summary file. Finally, we provide guidance to researchers using I/B/E/S for analyst earnings forecast data.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1

Similar content being viewed by others

Notes

  1. Thomson Reuters completed the sale of its majority stake in its Financial & Risk unit, the unit responsible for I/B/E/S, to Blackstone Group LP on October 1, 2018. Thomson Reuters retains a 45% stake in the company, which has been renamed Refinitiv. Refinitiv will be responsible for the maintenance of I/B/E/S (Scuffham 2018).

  2. https://www.refinitiv.com/en/financial-data/company-data/institutional-brokers-estimate-system-ibes

  3. The detail file includes a “currency” variable that represents the currency of the individual earnings forecast. This variable is missing for all 17,227 observations classified as “value changes.” As a result, we cannot determine whether the value changes in our analyses are due to a currency adjustment or some other explanation.

  4. Only 12 of the 905 brokerages contributing to OLD (6.74% of the forecasts in OLD but not in NEW) do not contribute earnings forecasts to NEW. Further, only eight of the 901 brokerages contributing to NEW (56.22% of the forecasts in NEW but not in OLD) did not contribute earnings forecasts to OLD. Other differences between versions of the detail file may be due to changes brokerages impose on individual earnings forecasts, rather than all forecasts contributed by the brokerage. As such, brokerage-initiated changes to the detail file may represent more than 6.74 (56.22) percent of all deletions (additions) to the file over time.

  5. In 2009, the detail file included all individual forecasts (e.g., EPS, CPS, SAL, etc.) for both U.S. and non-U.S. firms.

  6. This step ensures that we do not identify instances where I/B/E/S cleans up the file by removing duplicate and unnecessary observations. There are very few duplicate observations in either file (less than 0.30% of observations).

  7. We impose similar restrictions on the I/B/E/S summary file, which we elaborate on in Section 6.

  8. Although I/B/E/S publishes “release notes” that provide information about the brokerages and firms that are removed from or added to its database, this information is of limited use to researchers. Specifically, I/B/E/S does not identify the new brokerages added to the database, preventing a researcher from deleting these and creating an “as-was” version of the file. Further, while I/B/E/S identifies the brokerages that are deleted from the database, the historical forecasts that are deleted are not made available to researchers seeking to replicate studies. Therefore the “release notes” do not allow researchers to reconcile one version of I/B/E/S to another and do not provide the transparency to allow researchers to observe the frequency and implications of these changes to the database.

  9. To identify value changes that are due to stock splits or stock dividends, we examine whether the forecast is unchanged in both versions of the unadjusted detail file. If the forecast changes across the two versions of the adjusted detail file but is unchanged across the two versions of the unadjusted detail file, we assume the reason for the change in the adjusted detail file is due to a stock split or stock dividend. Our approach may understate the extent of value changes in the adjusted detail file as there may be a value change in the adjusted file for which there is not a corresponding change in the unadjusted file (e.g., due to an error correction that only affects the adjusted file).

  10. We follow Ljungqvist et al. (2009) in summing the deletions, additions, and value changes to arrive at this figure.

  11. We replicate Ljungqvist et al. (2009) and compare the 2008 and 2015 versions of the stock recommendation file, matching observations across the two versions on ticker, brokerage, and recommendation date. Only 0.25% of these recommendations are classified as deletions, 4.38% are classified as additions, and 0.07% are classified as value changes (i.e., “alterations,” per Ljungqvist et al. 2009). Further, only 0.09% of the observations that we classify as additions in our tests share a brokerage and ticker with the additions we identify in the recommendation file, while none of the observations that we classify as deletions (value changes) share a brokerage and ticker with any of the deletions (value changes) from the recommendation file. These results suggest our findings are not driven by any associated problems in the stock recommendation file and that the differences we find are associated with a unique set of analysts and firms.

  12. We re-estimate these analyses at the firm-year-analyst-forecast date level rather than the firm-year-brokerage-analyst-forecast date level (i.e., relaxing the requirement that the observation have the same broker code in both versions of the detail file). We continue to find a large number of deletions and additions (11.47 and 5.79% of OLD, respectively), suggesting that any re-coding of broker codes due to mergers is not driving our results.

  13. As we do not have a copy of the OLD unadjusted detail file for non-earnings forecasts, we cannot estimate the frequency of value changes for these forecasts.

  14. We observe significant differences in the accuracy and bias of quarterly earnings, annual cash flow, and annual sales forecasts across versions of the I/B/E/S file. For example, the median absolute forecast error for quarterly earnings and annual sales (annual cash flow) forecasts is significantly lower (higher) in NEW than in OLD, and the median bias for quarterly earnings, annual cash flow, and annual sales forecasts is significantly lower in NEW than in OLD.

  15. The sample we use in these analyses (1,895,191 observations) is smaller than the number of forecasts available from OLD and NEW combined (2,085,306 = 1,720,161 + 229,776 + 118,142 + 17,227 or summing the unchanged forecasts between the two versions of the file, deletions, additions, and value changes). This difference is due to the requirement that the forecast have an actual value available for the firm-year in both the current year (to determine LOSS_DUM, EPS, and EPS_GROWTH) and the prior year (to determine EPS_GROWTH).

  16. Results are inferentially identical when we estimate Equation (1) using a linear probability model.

  17. As there are more earnings forecasts in OLD relative to NEW, we examine whether this difference is due to delisted firms. In untabulated tests, we augment Equation (1) with an indicator variable to identify firms included in OLD that delist before NEW was published. The coefficient on this variable is significantly negative, suggesting that observations related to firms that delist are less likely to be subject to changes to the detail file.

  18. Only 6.83, 5.35, and 4.49% of the deletions, additions, and value changes, respectively, represent earnings forecasts for which the actual value is missing or the estimate date is on or after the earnings announcement.

  19. The value changes in OLD and NEW are not identical due to data requirements used to calculate forecast accuracy and bias. To calculate these properties, (a) the earnings announcement date must be available, (b) the date of the forecast must precede the earnings announcement, and (c) the actual EPS must be nonmissing and nonzero.

  20. In untabulated tests, we find that the impact of I/B/E/S changes on earnings forecast accuracy and bias is consistent across the years in the common sample period.

  21. Differences in forecast error and bias across the files are not driven by outliers. Our results are qualitatively unchanged when we winsorize or truncate our data at the first and 99th percentiles for analyst forecast error and bias.

  22. We also exclude firm-year observations impacted by stock splits or stock dividends as such adjustments could cause differences in meet or beat classifications across OLD and NEW (Payne and Thomas 2003).

  23. Our findings are inferentially the same if we use the median (rather than mean) consensus earnings forecast.

  24. For purposes of Equation (2), we standardize FIRMEXP to have a mean of 0 and a standard deviation of 1.

  25. For purposes of Equation (3), we standardize BROKER_SIZE to have a mean of 0 and a standard deviation of 1.

  26. https://wrds-www.wharton.upenn.edu/documents/1030/Product_Change_Notification-IBES_Detail_History-_PreApproval_Contributor.._.pdf?_ga=2.263469196.1099027739.1565208724-659584067.1532130746

  27. Our results are similar when we evaluate forecasts between 1993 and 2011 (1.39 and 12.06%, respectively).

  28. The annual percentages for the detail file in Figure 1 do not reflect the sum of the changes to OLD and NEW for each year in Table 2. In Table 2, value changes are considered to be changes to both OLD and NEW (to provide a sense for how much of the detail file changes from one year to the next), whereas for the purpose of Figure 1, value changes are included only once each year.

  29. We note a statistically significant difference in the median absolute forecast error when considering all earnings forecasts in OLD and NEW. However, this difference (0.001) is not economically significant, as it represents less than 1 % of the median absolute forecast error.

  30. We also impose the requirement that the absolute value of both earnings surprise estimates (consensus forecast minus I/B/E/S actual) not exceed 10% of the firm’s stock price.

  31. When we reestimate Equation (4) using maximum likelihood estimation and employ the Bayesian information criterion to compare the two earnings surprise metrics (Chiang et al. 2019), we continue to find that the consensus estimate provided in the summary file is superior. Specifically, in the four comparisons reported in Table 9, the Bayesian information criterion difference is 107, 154, 106, and 106, where a difference greater than 10 is generally considered strong evidence in favor of one model’s superiority over the other.

  32. Brown (1991) finds that the mean of all individual earnings forecasts is less accurate than a mean based on a subset of more recent earnings forecasts, providing guidance to researchers relying on the detail file about which forecasts to include when calculating the consensus earnings forecast. Payne and Thomas (2003) document issues that arise when using I/B/E/S adjusted earnings forecast data and encourage researchers to request the original (pre-split) data from I/B/E/S and create their own split-adjusted data that are not subject to rounding concerns.

  33. These journals are Contemporary Accounting Research, Journal of Accounting and Economics, Journal of Accounting Research, Review of Accounting Studies, and The Accounting Review.

  34. Some papers state which file they use. Many papers do not, but we infer which file they use based on the language in the paper. For example, Cadman et al. (2014, p. 63) indicate that they measure the earnings surprise as the difference between the actual EPS value and “the average of the most recent individual analyst earnings per share forecasts,” which suggests they use the detail file to construct their consensus earnings forecast. On the other hand, Rajgopal and Venkatachalam (2011, p. 14) measure earnings forecast errors as the difference between the actual and “the most recent median consensus earnings forecast immediately prior to the earnings announcement date,” which we infer as evidence that they use the summary file. Many studies do not provide enough information to reliably infer which file is used.

  35. We find similar results when we compare the summary consensus estimate (average analyst following is 9.01) to consensus earnings forecasts calculated from the detail file over 30 days (2.50), 60 days (3.99), and 120 days (8.21).

  36. Specifically, for a common sample period (January 1993 through November 2007) for a version of the surprise file downloaded in 2012 and a separate version downloaded in 2018, we document deletions, additions, and value changes of only 0.53, 1.30, and 0.12%, respectively, as a percentage of the observations in the 2012 version.

References

  • Akbas, F., Markov, S., Subasi, M., & Weisbrod, E. (2018). Determinants and consequences of information processing delay: evidence from the Thomson Reuters institutional brokers’ estimate system. Journal of Financial Economics, 127(2), 366–388.

    Google Scholar 

  • Ayers, B. C., Call, A. C., & Schwab, C. M. (2018). Do analysts’ cash flow forecasts encourage managers to improve the firm’s cash flows? Evidence from tax planning. Contemporary Accounting Research, 35(2), 767–793.

    Google Scholar 

  • Bartov, E., Givoly, D., & Hayn, C. (2002). The rewards to meeting or beating earnings expectations. Journal of Accounting and Economics, 33(2), 173–204.

    Google Scholar 

  • Beaver, W. H., Cornell, B., Landsman, W. R., & Stubben, S. R. (2008). The impact of analysts’ forecast errors and forecast revisions on stock prices. Journal of Business Finance & Accounting, 35(5–6), 709–740.

    Google Scholar 

  • Bhojraj, S., Hribar, P., Picconi, M., & McInnis, J. (2009). Making sense of cents: An examination of firms that marginally miss or beat analyst forecasts. The Journal of Finance, 64(5), 2361–2388.

    Google Scholar 

  • Billings, M. B., & Cedergren, M. C. (2015). Strategic silence, insider selling, and litigation risk. Journal of Accounting and Economics, 59(2–3), 119–142.

    Google Scholar 

  • Bochkay, K., Markov, S., Subasi, M., & Weisbrod, E. (2018). Street earning activation delay. Working paper, University of Miami, University of Texas at Dallas, and University of Maryland. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3219449

  • Bradshaw, M. T. (2011). Analysts’ forecasts: What do we know after decades of work? Working paper, Boston College. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1880339

  • Bradshaw, M. T., Drake, M. S., Myers, J. N., & Myers, L. A. (2012). A re-examination of analysts’ superiority over time-series forecasts of annual earnings. Review of Accounting Studies, 17(4), 944–968.

    Google Scholar 

  • Brown, L. D. (1991). Forecast selection when all forecasts are not equally recent. International Journal of Forecasting, 7(3), 349–356.

    Google Scholar 

  • Brown, L. D., & Rozeff, M. S. (1978). The superiority of analyst forecasts as measures of expectations: Evidence from earnings. The Journal of Finance, 33(1), 1–16.

    Google Scholar 

  • Brown, L. D., Hagerman, R. L., Griffin, P. A., & Zmijewski, M. E. (1987a). Security analyst superiority relative to univariate time-series models in forecasting quarterly earnings. Journal of Accounting and Economics, 9(1), 61–87.

    Google Scholar 

  • Brown, L. D., Richardson, G. D., & Schwager, S. J. (1987b). An information interpretation of financial analyst superiority in forecasting earnings. Journal of Accounting Research, 25(1), 49–67.

    Google Scholar 

  • Brown, N. C., Christensen, T. E., Elliott, W. B., & Mergenthaler, R. D. (2012). Investor sentiment and pro forma earnings disclosures. Journal of Accounting Research, 50(1), 1–40.

    Google Scholar 

  • Brown, L. D., Call, A. C., Clement, M. B., & Sharp, N. Y. (2016). The activities of buy-side analysts and the determinants of their stock recommendations. Journal of Accounting and Economics, 62(1), 139–156.

    Google Scholar 

  • Cadman, B., Carrizosa, R., & Faurel, L. (2014). Economic determinants and information environment effects of earnouts: New insights from SFAS 141(R). Journal of Accounting Research, 52(1), 37–74.

    Google Scholar 

  • Call, A. C., Chen, S., & Tong, Y. H. (2009). Are analysts’ earnings forecasts more accurate when accompanied by cash flow forecasts? Review of Accounting Studies, 14(2–3), 358–391.

    Google Scholar 

  • Call, A. C., Chen, S., & Tong, Y. H. (2013). Are analysts’ cash flow forecasts naïve extensions of their own earnings forecasts? Contemporary Accounting Research, 30(2), 438–465.

    Google Scholar 

  • Chiang, C. H., Dai, W., Fan, J., Hong, H., & Tu, J. (2019). Robust measures of earnings surprises. The Journal of Finance, 74(2), 943–983.

    Google Scholar 

  • Clement, M. B. (1999). Analyst forecast accuracy: Do ability, resources, and portfolio complexity matter? Journal of Accounting and Economics, 27(3), 285–303.

    Google Scholar 

  • Clement, M. B., Hales, J., & Xue, Y. (2011). Understanding analysts’ use of stock returns and other analysts’ revisions when forecasting errors. Journal of Accounting and Economics, 51(3), 279–299.

    Google Scholar 

  • Cornell, B., & Landsman, W. R. (1989). Security price response to quarterly earnings announcements and analysts’ forecast revisions. The Accounting Review, 64(4), 680–692.

    Google Scholar 

  • Degeorge, F., Patel, J., & Zeckhauser, R. (1999). Earnings management to exceed thresholds. The Journal of Business, 72(1), 1–33.

    Google Scholar 

  • Dhaliwal, D., Judd, J. S., Serfling, M., & Shaikh, S. (2016). Customer concentration risk and the cost of equity capital. Journal of Accounting and Economics, 61(1), 23–48.

    Google Scholar 

  • Dugar, A., & Nathan, S. (1995). The effect of investment banking relationships on financial analysts’ earnings forecasts and investment recommendations. Contemporary Accounting Research, 12(1), 131–160.

    Google Scholar 

  • Ertimur, Y., Mayew, W. J., & Stubben, S. R. (2011). Analyst reputation and the issuance of disaggregated earnings forecasts to I/B/E/S. Review of Accounting Studies, 16(1), 29–58.

    Google Scholar 

  • Ertimur, Y., Sletten, E., & Sunder, J. (2014). Large shareholders and disclosure strategies: evidence from IPO lockup expirations. Journal of Accounting and Economics, 58(1), 79–95.

    Google Scholar 

  • Francis, J., & Philbrick, D. (1993). Analysts’ decisions as products of a multi-task environment. Journal of Accounting Research, 31(2), 216–230.

    Google Scholar 

  • Freeman, R. N., & Tse, S. Y. (1992). A nonlinear model of security price responses to unexpected earnings. Journal of Accounting Research, 30(2), 185–209.

    Google Scholar 

  • Fried, D., & Givoly, D. (1982). Financial analysts’ forecasts of earnings: a better surrogate for market expectations. Journal of Accounting and Economics, 4(2), 85–107.

    Google Scholar 

  • Givoly, D., Hayn, C., & Lehavy, R. (2009). The quality of analysts’ cash flow forecasts. The Accounting Review, 84(6), 1877–1911.

    Google Scholar 

  • Jame, R., Johnston, R., Markov, S., & Wolfe, M. C. (2016). The value of crowdsourced earnings forecasts. Journal of Accounting Research, 54(4), 1077–1110.

    Google Scholar 

  • Kaplan, Z., Martin, X., & Xie, Y. (2018). Truncating optimism. Working paper, Washington University in St. Louis. Available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3213833

  • Keung, E. C. (2010). Do supplementary sales forecasts increase the credibility of financial analysts’ earnings forecasts? The Accounting Review, 85(6), 2047–2074.

    Google Scholar 

  • Kirk, M. P., Reppenhagen, D. A., & Tucker, J. W. (2014). Meeting individual analyst expectations. The Accounting Review, 89(6), 2203–2231.

    Google Scholar 

  • Koh, K., Matsumoto, D. A., & Rajgopal, S. (2008). Meeting or beating analyst expectations in the post-scandals world: changes in stock market rewards and managerial actions. Contemporary Accounting Research, 25(4), 1067–1098.

    Google Scholar 

  • Landsman, W. R., Maydew, E. L., & Thornock, J. R. (2012). The information content of annual earnings announcements and mandatory adoption of IFRS. Journal of Accounting and Economics, 53(1–2), 34–54.

    Google Scholar 

  • Lim, T. (2001). Rationality and analysts’ forecast bias. The Journal of Finance, 56(1), 369–385.

    Google Scholar 

  • Lin, H., & McNichols, M. F. (1998). Underwriting relationships, analysts’ earnings forecasts and investment recommendations. Journal of Accounting and Economics, 25(1), 101–127.

    Google Scholar 

  • Ljungqvist, A., Malloy, C., & Marston, F. (2009). Rewriting history. The Journal of Finance, 64(4), 1935–1960.

    Google Scholar 

  • McInnis, J., & Collins, D. W. (2011). The effect of cash flow forecasts on accrual quality and benchmark beating. Journal of Accounting and Economics, 51(3), 219–239.

    Google Scholar 

  • Mikhail, M. B., Walther, B. R., & Willis, R. H. (1997). Do security analysts improve their performance with experience? Journal of Accounting Research, 35(supplement), 131–157.

    Google Scholar 

  • Mohanram, P. S. (2014). Analysts’ cash flow forecasts and the decline of the accruals anomaly. Contemporary Accounting Research, 31(4), 1143–1170.

    Google Scholar 

  • O’Brien, P. C. (1988). Analysts’ forecasts as earnings expectations. Journal of Accounting and Economics, 10(1), 53–83.

    Google Scholar 

  • Payne, J. L., & Thomas, W. B. (2003). The implications of using stock-split adjusted I/B/E/S data in empirical research. The Accounting Review, 78(4), 1049–1067.

    Google Scholar 

  • Rajgopal, S., & Venkatachalam, M. (2011). Financial reporting quality and idiosyncratic return volatility. Journal of Accounting and Economics, 51(1–2), 1–20.

    Google Scholar 

  • Ramnath, S., Rock, S., & Shane, P. (2005). Value line and I/B/E/S earnings forecasts. International Journal of Forecasting, 21(1), 185–198.

    Google Scholar 

  • Ramnath, S., Rock, S., & Shane, P. (2008). The financial analyst forecasting literature: a taxonomy with suggestions for further research. International Journal of Forecasting, 24(1), 34–75.

    Google Scholar 

  • Reichelt, K. J., & Wang, D. (2010). National and office-specific measures of auditor industry expertise and effects on audit quality. Journal of Accounting Research, 48(3), 647–686.

    Google Scholar 

  • Scuffham, M. (2018). Thomson Reuters closes deal with Blackstone. Reuters, October 1, 2018, Retrieved from https://www.reuters.com/article/us-thomsonreuters-m-a-blackstone/thomson-reuters-closes-deal-with-blackstone-idUSKCN1MB3PY

  • Vuong, Q. H. (1989). Likelihood ratio tests for model selection and non-nested hypotheses. Econometrica, 57(2), 307–333.

    Google Scholar 

  • Wu, J. S., & Zang, A. Y. (2009). What determine financial analysts’ career outcomes during mergers? Journal of Accounting and Economics, 47(1–2), 59–86.

    Google Scholar 

  • Yang, H. I. (2012). Capital market consequences of managers’ voluntary disclosure styles. Journal of Accounting and Economics, 53(1–2), 167–184.

    Google Scholar 

  • Yu, F. F. (2008). Analyst coverage and earnings management. Journal of Financial Economics, 88(2), 245–271.

    Google Scholar 

Download references

Acknowledgements

We thank Larry Brown, Michael Clement, John Donovan, Artur Hugon, Phil Lamoreaux, Stephannie Larocque, and Nate Sharp for their comments and suggestions, as well as David Gelinas, Robert Marcucci, and Aaron Mo of Thomson Reuters (now Refinitiv) for valuable discussions. Jessica Watkins acknowledges the generous support of the Deloitte Foundation.

Author information

Authors and Affiliations

  1. Arizona State University, BA 267A, Tempe, AZ, 85287-3606, USA

    Andrew C. Call

  2. The University of Arizona, Tucson, AZ, 85721, USA

    Max Hewitt

  3. University of Notre Dame, Notre Dame, IN, 46556, USA

    Jessica Watkins

  4. Emory University, Atlanta, GA, 30322, USA

    Teri Lombardi Yohn

Authors
  1. Andrew C. Call
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Max Hewitt
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Jessica Watkins
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Teri Lombardi Yohn
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Andrew C. Call.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Table 10 Illustration of deletions, additions, and value changes to the detail file
Full size table

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Call, A.C., Hewitt, M., Watkins, J. et al. Analysts’ annual earnings forecasts and changes to the I/B/E/S database. Rev Account Stud 26, 1–36 (2021). https://doi.org/10.1007/s11142-020-09560-x

Download citation

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11142-020-09560-x

Keywords

JEL classification

Search

Navigation