Abstract
This study investigates financial analysts’ revenue forecasts and identifies determinants of the forecasts’ accuracy. We find that revenue forecast accuracy is determined by forecast and analyst characteristics similar to those of earnings forecast accuracy—namely, forecast horizon, days elapsed since the last forecast, analysts’ forecasting experience, forecast frequency, forecast portfolio, reputation, earnings forecast issuance, forecast boldness, and analysts’ prior performance in forecasting revenues and earnings. We develop a model that predicts the usefulness of revenue forecasts. Thereby, our study helps to ex ante identify more accurate revenue forecasts. Furthermore, we find that analysts concern themselves with their revenue forecasting performance. Analysts with poor revenue forecasting performance are more likely to stop forecasting revenues than analysts with better performance. Their decision is reasonable because revenue forecast accuracy affects analysts’ career prospects in terms of being promoted or terminated. Our study helps investors and academic researchers to understand determinants of revenue forecasts. This understanding is also beneficial for evaluating earnings forecasts because revenue forecasts reveal whether changes in earnings forecasts are due to anticipated changes in revenues or expenses.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
We measure experience and portfolio complexity as revenue specific; that is, we employ analysts’ firm-specific and general experience in forecasting revenues and the complexity of analysts’ revenue forecast portfolio (measured from the numbers of covered firms and industries).
In 1998, 13.92% (572/4110) of the analysts in I/B/E/S provide at least one revenue forecast. In 1997, it is only 1.57% (53/3371) of the analysts.
If too few analysts cover a specific firm-year, then the range-adjusted variables might yield weak differentiation criteria. For example, if a firm is covered by only two analysts, they are assigned the values 0 and 1.
We can directly compare the coefficient estimates because all variables are adjusted and range between 0 and 1.
In contrast, we find the expected effect for the other portfolio complexity measure \( Industries \) (coefficient estimate − 0.0090, t value − 3.66); that is, revenue forecasts are less accurate if the analyst covers a larger number of different industries. This finding is reasonable because the practices of revenue recognition might differ between industries.
Keung (2010) conducts a related analysis and finds that 79.20% of the forecasts are consistent. However, Keung (2010) aims to compare earnings forecasts that are not accompanied by revenue forecasts versus consistent pairs versus inconsistent pairs. He finds that earnings forecasts are most accurate when accompanied by consistent revenue forecasts and least accurate when not accompanied by revenue forecasts at all.
We excluded 51,295 observations for the following reasons: the analyst’s prior forecast equals the current one; the consensus forecast equals the current one; or the consensus immediately before the forecast could not be determined.
We naturally employ all revenue forecast data available in I/B/E/S to determine whether the analyst keeps issuing revenue forecasts, and we do not restrict this determination to our employed sample.
Similar to Clement and Tse (2005, 325), we make this inference because \( Accuracy_{ijt - 1} \) is scaled to range between 0 (for the analyst with the worst revenue forecasting performance for firm \( j \) in fiscal year \( t - 1 \)) to 1 (for the analyst with the best performance for firm \( j \) in \( t - 1 \)). The odds ratio (0.292) is the difference in odds that a one-unit difference in \( Accuracy_{ijt - 1} \) has on the probability to observe \( Firm\_Stop_{ijt} = 1 \). By construction, a one-unit difference in \( Accuracy_{ijt - 1} \) corresponds to the difference between the worst and the best revenue forecaster for firm \( j \) in fiscal year \( t - 1 \). The probability that the best forecaster for firm \( j \) in \( t - 1 \) stops forecasting in year \( t \) is only 0.292 times the probability that the worst forecaster for firm \( j \) in \( t - 1 \) stops. This corresponds to 70.8% smaller odds.
We employ \( \overline{Accuracy}_{it - 1} \), instead of \( \frac{1}{J}\mathop \sum \limits_{j = 1}^{J} Lagged\_Accuracy_{ijt} \), as we attempt to measure the average revenue forecast accuracy of all firms analyst \( i \) covers in year \( t - 1 \). \( \frac{1}{J}\mathop \sum \limits_{j = 1}^{J} Lagged\_Accuracy_{ijt} \) can only be calculated for firms that the analyst covers in the two succeeding years.
This inference is conducted from the odds ratio (0.065), which means that an analyst with \( \overline{Accuracy}_{it - 1} = 1 \) has only 0.065 times the odds to stop forecasting revenues compared to an analyst with \( \overline{Accuracy}_{it - 1} = 0 \). This corresponds to 93.5% smaller odds.
Carter and Manaster (1990) base their ranking on the hierarchy of an underwriter’s position in the stock offering “tombstone” announcements. Each underwriter is assigned a rank between 0 (low prestige) and 9 (high prestige). Loughran and Ritter (2004) update the rankings in regular intervals and assign values between 1.1 and 9.1 to the brokerage houses (the attached 0.1 helps to differentiate their ranking from Carter and Manaster’s (1990)). We use the data supplied on Jay Ritter’s homepage. For discussions on the ranking process see Carter and Manaster (1990) or Carter et al. (1998).
Ertimur et al. (2011) conduct a related analysis. However, they employ a multivariate logit model, instead of two different logit models and measure the impact of revenue forecasts on promotion and demotion simultaneously. Furthermore, Ertimur et al. (2011) use a notable smaller sample (8954 vs. 25,668 observations) and a shorter time span (1999–2006 vs. 1999–2014).
References
Baginski S, Hassell J, Kimbrough M (2004) Why do managers explain their earnings forecasts? J Acc Res 42(1):1–29. doi:10.1111/j.1475-679x.2004.00127.x
Bilinski P (2014) Do analysts disclose cash flow forecasts with earnings estimates when earnings quality is low? J Bus Financ Acc 41(3–4):401–434. doi:10.1111/jbfa.12056
Bradshaw MT (2011) Analysts’ forecasts: what do we know after decades of work? Working paper. Boston College. doi:10.2139/ssrn.1880339
Brown LD (2001) How important is past analyst forecast accuracy? Financ Anal J 57(6):44–49. doi:10.2469/faj.v57.n6.2492
Brown LD, Mohammad E (2010) Is analyst earnings forecast ability only firm specific? Contemp Acc Res 27(3):727–750. doi:10.1111/j.1911-3846.2010.01025.x
Brown LD, Mohd E (2003) The predictive value of analyst characteristics. J Acc Audit Financ 18(4):625–647. doi:10.1177/0148558X0301800412
Call AC, Chen S, Tong YH (2009) Are analysts’ earnings forecasts more accurate when accompanied by cash flow forecasts? Rev Acc Stud 14(2):358–391. doi:10.1007/s11142-009-9086-7
Call AC, Chen S, Tong YH (2013) Are analysts’ cash flow forecasts naïve extensions of their own earnings forecasts? Contemp Acc Res 30(2):438–465. doi:10.1111/j.1911-3846.2012.01184.x
Carter R, Manaster S (1990) Initial public offerings and underwriter reputation. J Financ 45(4):1045–1067. doi:10.1111/j.1540-6261.1990.tb02426.x
Carter R, Dark FH, Singh AK (1998) Underwriter reputation, initial returns, and the long-run performance of IPO stocks. J Financ 53(1):285–311. doi:10.1111/0022-1082.104624
Clement MB (1999) Analyst forecast accuracy: do ability, resources, and portfolio complexity matter? J Acc Econ 27(3):285–303. doi:10.1016/S0165-4101(99)00013-0
Clement MB, Tse SY (2003) Do investors respond to analysts’ forecast revisions as if forecast accuracy is all that matters? Acc Rev 78(1):227–249. doi:10.2308/accr.2003.78.1.227
Clement MB, Tse SY (2005) Financial analyst characteristics and herding behavior in forecasting. J Financ 60(1):307–341. doi:10.1111/j.1540-6261.2005.00731.x
Clement MB, Koonce L, Lopez TJ (2007) The roles of task-specific forecasting experience and innate ability in understanding analyst forecasting performance. J Acc Econ 44(3):378–398. doi:10.1016/j.jacceco.2007.04.002
Curtis AB, Lundholm RJ, McVay SE (2014) Forecasting sales: a model and some evidence from the retail industry. Contemp Acc Res 31(2):581–608. doi:10.1111/1911-3846.12040
Edmonds CT, Leece RD, Maher RD (2013) CEO bonuses compensation: the effects of missing analysts’ revenue forecasts. Rev Quant Financ Acc 41:149–170. doi:10.1007/s11156-012-0305-0
Elgers PT, Lo MH, Xie W, Xu LE (2016) A contextual evaluation of composite forecasts of annual earnings. Rev Pac Basin Financ Mark Polic 19(3):1–40. doi:10.1142/S0219091516500144
Ertimur Y, Livnat J, Martikainen M (2003) Differential market reactions to revenue and expense surprises. Rev Acc Stud 8(2–3):185–211
Ertimur Y, Mayew WJ, Stubben SR (2011) Analyst reputation and the issuance of disaggregated earnings forecasts to I/B/E/S. Rev Acc Stud 16(1):29–58. doi:10.1007/s11142-009-9116-5
Givoly D, Hayn C, Lehavy R (2009) The quality of analysts’ cash flow forecasts. Acc Rev 84(6):1877–1911. doi:10.2308/accr.2009.84.6.1877
Gleason CA, Lee CMC (2003) Analyst forecast revisions and market price discovery. Acc Rev 78(1):193–225. doi:10.2308/accr.2003.78.1.193
Hirst DE, Koonce L, Venkataraman S (2007) How disaggregation enhances the credibility of management earnings forecasts. J Acc Res 45(4):811–837. doi:10.1111/j.1475-679x.2007.00252.x
Hong H, Kubik JD (2003) Analyzing the analysts: career concerns and biased earnings forecasts. J Financ 58(1):313–351. doi:10.1111/1540-6261.00526
Hong H, Kubik JD, Solomon A (2000) Security analysts’ career concerns and herding of earnings forecasts. RAND J Econ 31(1):121–144. doi:10.2307/2601032
Hsu D, Chiao CH (2011) Relative accuracy of analysts’ earnings forecasts over time: a Markov chain analysis. Rev Quant Financ Acc 37:477–507. doi:10.1007/s11156-010-0214-z
Hutton A, Miller G, Skinner D (2003) The role of supplementary statements with management earnings forecasts. J Acc Res 41(5):867–890. doi:10.1046/j.1475-679x.2003.00126.x
Jackson AR (2005) Trade generation, reputation and sell-side analysts. J Financ 60(2):673–717. doi:10.1111/j.1540-6261.2005.00743.x
Jacob J, Lys TZ, Neale MA (1999) Expertise in forecasting performance of security analysts. J Acc Econ 28(1):51–82. doi:10.1016/S0165-4101(99)00016-6
Ke B, Yu Y (2006) The effect of issuing biased earnings forecasts on analysts’ access to management and survival. J Acc Res 44(5):965–999. doi:10.1111/j.1475-679X.2006.00221.x
Keung EC (2010) Do supplementary sales forecasts increase the credibility of financial analysts’ earnings forecasts? Acc Rev 85(6):2047–2074. doi:10.2308/accr.2010.85.6.2047
Lehavy R (2009) Discussion of “Are earnings forecasts more accurate when accompanied by cash flow forecasts?”. Rev Acc Stud 14(2–3):392–400. doi:10.1007/s11142-009-9090-y
Loughran T, Ritter J (2004) Why has IPO underpricing changed over time? Financ Manag 33(3):5–37
Marks J (2008) Revenue forecasts as a mechanism for analyst reputation development. Working paper. Seton Hall University. doi: 10.2139/ssrn.1095248
Mest DP, Plummer E (2003) Analysts’ rationality and forecast bias: evidence from sales forecasts. Rev Quant Financ Acc 21:103–122. doi:10.1023/A:1024841531461
Mikhail MB, Walther BR, Willis RH (1997) Do security analysts improve their performance with experience? J Acc Res 35:131–157. doi:10.2307/2491458
O’Brien PC (1988) Analysts’ forecasts as earnings expectations. J Acc Econ 10(1):53–83. doi:10.1016/0165-4101(88)90023-7
Pae J, Yoon SS (2012) Determinants of analysts’ cash flow forecast accuracy. J Acc Audit Financ 27(1):123–144. doi:10.1177/0148558x11409148
Penman SH (2013) Financial statement analysis and security valuation, 5th edn. McGraw-Hill, New York
Rees L, Sivaramakrishnan K (2007) The effect of meeting or beating revenue forecasts on the association between quarterly returns and earnings forecast errors. Contemp Acc Res 24(1):259–290. doi:10.1506/g767-1674-5686-5375
Stickel SE (1992) Reputation and performance among security analysts. J Financ 47(5):1811–1836. doi:10.1111/j.1540-6261.1992.tb04684.x
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1
See Table 10.
Appendix 2
See Table 11.
Appendix 3
See Table 12.
Rights and permissions
About this article
Cite this article
Lorenz, T., Homburg, C. Determinants of analysts’ revenue forecast accuracy. Rev Quant Finan Acc 51, 389–431 (2018). https://doi.org/10.1007/s11156-017-0675-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11156-017-0675-4