Manager-analyst conversations in earnings conference calls


Prior research finds that intraday stock prices move considerably during the discussion period of earnings conference calls. In this study, we explore what features of the manager-analyst dialogue during the discussion drive these price movements. We textually analyze the tone of managers and analysts and find that intraday prices react significantly to analyst tone, but not to management tone, for the full duration of the discussion. This effect strengthens when analyst tone is relatively negative. We then present intraday visual evidence that analysts are more neutral than managers over the call and that the tones of both parties drift downward as the call progresses. Overall, our findings illustrate how manager-analyst information exchanges evolve on earnings calls and indicate that analysts are the participants on earnings calls whose comments move stock prices during the discussion.

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

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7


  1. 1.

    See Appendix A for an illustration and Section 2 for more details.

  2. 2.

    Section 3 assesses in more detail the differential intraday returns effect for positive and negative analyst tone. Our motivation for this test is twofold. First, Davis et al. (2015) argue in their opening section that call participants face increased reputational costs for using negative language in the public earnings call. Second, Huang et al. (2014a) find that investors react more strongly to negative analyst reports than they react to positive reports.

  3. 3.

    Many institutions are clients of analysts who might act on their call remarks (e.g., Irvine et al. (2006)).

  4. 4.

    We discuss these additional results further in Section 3.4.

  5. 5.

    For example, the stock price could rise before the discussion in the presentation section (Matsumoto et al. 2011), which in turn makes analysts more positive on the call. This effect would show up in a two-day return window association test with analyst tone but not in the discussion-period return window. Or analyst tone could have no discussion-period reaction but could be correlated with some future analyst activity, such as a recommendation issued soon after the call. The market, unaware of this correlation, could respond only to the recommendation. This behavior would again show up in the two-day returns test but not in the intraday returns test.

  6. 6.

    In 2008 I/B/E/S stopped providing a matching table for analysts’ names, banks, and I/B/E/S identifiers in the earnings forecast file. We therefore limit our individual analyst earnings forecast tests to the years 2002-2007. Also, because our matching process relies on analyst and broker names from the conference call transcripts, and because these names do not always perfectly align with the I/B/E/S matching database (due to abbreviations, misspellings, etc.), we cannot match every analyst to I/B/E/S.

  7. 7.


  8. 8.

    Larcker and Zakolyukina (2012) also focus on CEOs and CFOs. Including COOs and vice presidents (as well as including all company representatives on the call, regardless of their title) does not affect the mean or standard deviation of the management tone metric, nor does it affect our intraday returns results.

  9. 9.

    One potential concern with our tone measures is that they depend on the number of positive and negative words, not on the total number of words spoken by the analyst. Careful use of firm- and analyst-fixed effects partially control for such heterogeneity; however, there could still be within-firm variation in total words spoken on the call over time. Therefore, in untabulated tests, we construct alternative measures of tone that are identical to Eq. 1, except use as the denominator total words spoken by managers or analysts (depending on the measure). The intraday return results with this measure are similar in terms of sign, magnitude, and p-values. We also control for call length throughout our analysis.

  10. 10.

    Mayew (2008, p. 632) argues that, although management would like to favor pliant analysts on the call, capital market pressures can force management to give airtime to good independent analysts. In addition, Groysberg et al. (2011) document that analysts have strong incentives to be competent and accurate. These considerations further justify our decision to aggregate all analysts on the call.

  11. 11.

    As noted earlier, this sample only includes calls through 2007 for earnings forecasts. I/B/E/S stopped providing the analyst name to I/B/E/S ID matching file for earnings forecasts after this time.

  12. 12.

    The focus on earnings as the metric of financial information is standard in the empirical disclosure literature for two reasons: first, more disaggregated financial items (such as provisions) may vary in importance across firms; earnings by contrast represent a common and important aggregate performance measure for all firms. Second, there exist well-studied measures of analysts’ expected earnings, thus allowing one to compute earnings news.

  13. 13.

    For example, Altinkilic and Hansen (2009) argue that momentum effects could arise from pre-event-day news releases.

  14. 14.

    This data set is the focus of several studies, including Irvine et al. (2006), who describe it in detail.

  15. 15.

    Since we cannot directly uncover the covariance structure of the error terms, we have ensured that our results obtain with alternative clustering methods, including clustering by firm and two-way clustering by firm and year (Gow et al. 2010).

  16. 16.

    While a larger firm is likely to have more remarks, we assume that we can compare individual remarks across firms.

  17. 17.

    We find similar tone patterns across market-value deciles and the ten most populous two-digit SICs in our sample. Note that the very first remark by management is often introductory and thus more neutral than subsequent remarks. Also, the first analyst typically congratulates management on the quarter and is thus more positive than future analyst remarks.

  18. 18.

    By contrast, Jung et al. (2017, Table 6) find significant differences in the dialogue of sell- and buy-side analysts in the discussion period. For example, relative to sell-side analysts, buy-side analysts receive shorter replies from CEOs by a magnitude of about 10 words, on average. Also, sell-side analysts, who are the focus of this study, outnumber buy-side analysts by about a factor of 20 in the discussion period.

  19. 19.

    As described in Section 2, our future analyst revision measures are conservative because a significant portion of analysts make revisions on the conference call date and we measure future revisions from the end of the call date forward. This means that our measurement strategy will bias against finding significance.

  20. 20.

    To allow us to better interpret the economic significance of our results, we use level changes in EPS forecasts; however, our results are similar in sign and statistical significance when we scale EPS by stock price at the quarter’s fiscal end date. Also note that firm-fixed effects control for across-firm variation in EPS.

  21. 21.

    Another feature of our recommendation analysis is that we use OLS, which assumes that the difference between 1 and 0 (upgrade versus no change) is of the same economic magnitude as the difference between 0 and − 1 (no change versus downgrade). While this appears to be a reasonable assumption, we nonetheless replicate the recommendation analysis with ordered and multinomial logit regressions and find significance for the analyst tone measure. However, econometric and computational limitations prevent us from including analyst-firm fixed effects in these logit regressions (Greene 2004). We therefore do not tabulate these results. We also replicate the analysis and find similar results in sign and statistical significance for the full five-point recommendation scale in I/B/E/S, which ranges from sell to strong buy.

  22. 22.

    A potential concern is that returns measured during the Q&A may capture some residual trading from the presentation. Our analysis of the 30-minute period after the call helps to ensure that our findings are not driven by such residual information. In addition, we include management presentation tone as a regressor in all of our returns tests, which will eliminate any effect due to (potentially latent) trades based on the presentation tone.

  23. 23.

    An alternative explanation for the substantial returns results for analyst recommendations is offered by Loh and Stulz (2010), who find that these recommendations overlap with concurrent information events.

  24. 24.

    We recognize that investors may be unable to perfectly predict analysts’ future earnings forecasts, price targets, and recommendations. We therefore use the actual realization of these outputs as the best available proxies.

  25. 25.

    Our findings are similar when using an indicator for recommendation upgrades (as opposed to using the full five-point scale for recommendations).

  26. 26.

    There are significant differences in the nature of our information event and that studied by Tetlock (2007); we cannot execute his tests literally. For example, our interval duration, which we motivate using the argument in Tetlock et al. (2008, p. 1452), is different than that of Tetlock (2007). Furthermore, we use three future intervals, while Tetlock (2007) uses five. However, in untabulated tests, we find that the much longer window of daily returns is also positively correlated with analyst tone, providing further evidence of no reversals.

  27. 27.

    We acknowledge that some intraday time-stamps in I/B/E/S may be inaccurate. Bradley et al. (2014) find that recommendation time-stamps cannot be corrected without manually checking each analyst report, and therefore we do not attempt to adjust the times. However, they find that in their limited sample, time-stamps are delayed on average by 2.4 hours. We thus use three-hour windows when eliminating calls with concurrent analyst revisions.

  28. 28.

    Two realities are possible for analyses three and four: it could be that there is in general more information released on the calls, which might lead to stronger tone effects, or that the additional information substitutes for information on the call, which might lead to weaker tone effects.

  29. 29.

    One exception is our sample split on median management Q&A financial words: the analyst tone effect is marginally larger in the above-the-median sample (coefficient of 0.119 versus 0.086; 10% level).


  1. Altinkilic, O., & Hansen, R. (2009). On the information role of stock recommendation revisions. Journal of Accounting and Economics, 48(1), 17–36.

    Article  Google Scholar 

  2. Altinkilic, O., Hansen, R., Liyu, Y. (2016). Can analysts pick stocks for the long-run? Journal of Financial Economics, 119(2), 371–398.

    Article  Google Scholar 

  3. Bradley, D., Clarke, J., Lee, S., Ornthanalai, C. (2014). Are analysts’ recommendations informative? intraday evidence on the impact of time stamp delays. The Journal of Finance, 69(2), 645–673.

    Article  Google Scholar 

  4. Brav, A., & Lehavy, R. (2003). An Empirical Analysis of Analysts’ Target Prices: Short-term Informativeness and Long-term Dynamics. The Journal of Finance, 58(5), 1933–1968.

    Article  Google Scholar 

  5. Brochet, F., Kolev, K., Lerman, A. (2018). Information Transfer and Conference Calls. Review of Accounting Studies. forthcoming.

  6. Brochet, F., Loumioti, M., Serafeim, G. (2015). Speaking of the short-term: disclosure horizon and managerial myopia. Review of Accounting Studies, 20(3), 1122–1163.

    Article  Google Scholar 

  7. Brochet, F., Naranjo, P., Yu, G. (2016). The capital market consequences of language barriers in the conference calls of Non-U.S. Firms. The Accounting Review, 91(4), 1023–1049.

    Article  Google Scholar 

  8. Brockman, P., Li, X., Price, S.M. (2015). Differences in conference call tones: Managers versus analysts. Financial Analysts Journal, 71(4), 24–42.

    Article  Google Scholar 

  9. Brown, L., Call, A., Clement, M., Sharp, N. (2015). Inside the “black box” of sell-side financial analysts. Journal of Accounting Research, 53(1), 1–47.

    Article  Google Scholar 

  10. Bushee, B., Jung, M., Miller, G. (2011). Conference presentations and the disclosure milieu. Journal of Accounting Research, 49(5), 1163–1192.

    Article  Google Scholar 

  11. Cheng, Q., Du, F., Wang, X., Wang, Y. (2016). Seeing is believing: analysts’ corporate site visits. Review of Accounting Studies, 21(4), 1245–1286.

    Article  Google Scholar 

  12. Davis, A.K., Ge, W., Matsumoto, D., Zhang, J.L. (2015). The effect of manager-specific optimism on the tone of earnings conference calls. Review of Accounting Studies, 20(2), 639–673.

    Article  Google Scholar 

  13. Frankel, R., Johnson, M., Skinner, D. (1999). An empirical examination of conference calls as a voluntary disclosure medium. Journal of Accounting Research, 37(1), 133–150.

    Article  Google Scholar 

  14. Frankel, R., Mayew, W.J., Sun, Y. (2010). Do pennies matter? Investor relations consequences of small negative earnings surprises. Review of Accounting Studies, 15(1), 220–242.

    Article  Google Scholar 

  15. Ge, W., Matsumoto, D., Zhang, J.L. (2011). Do CFOs have style? an empirical investigation of the effect of individual CFOs on accounting practices. Contemporary Accounting Research, 28(4), 1141–1179.

    Article  Google Scholar 

  16. Gintschel, A., & Markov, S. (2004). The effectiveness of Regulation FD. Journal of Accounting and Economics, 37(3), 293–314.

    Article  Google Scholar 

  17. Gow, I.D., Ormazabal, G., Taylor, D.J. (2010). Correcting for Cross-Sectional and Time-Series dependence in accounting research. The Accounting Review, 85(2), 483–512.

    Article  Google Scholar 

  18. Greene, W. (2004). The behaviour of the maximum likelihood estimator of limited dependent variable models in the presence of fixed effects. Econometrics Journal, 7(1), 98–119.

    Article  Google Scholar 

  19. Griffin, J., Harris, J., Topaloglu, S. (2003). The dynamics of institutional and individual trading. The Journal of Finance, 58(6), 2285–2320.

    Article  Google Scholar 

  20. Groysberg, B., Healy, P., Maber, D. (2011). What drives sell-side analyst compensation at high-status investment banks? Journal of Accounting Research, 49(4), 969–1000.

    Article  Google Scholar 

  21. Huang, A., Zang, A., Zheng, R. (2014a). Evidence on the information content of text in analyst reports. The Accounting Review, 89(6), 2151–2180.

    Article  Google Scholar 

  22. Huang, X., Teoh, S., Zhang, Y. (2014b). Tone management. The Accounting Review, 89(3), 1083–1113.

    Article  Google Scholar 

  23. Irvine, P., Lipson, M., Puckett, A. (2006). Tipping. Review of Financial Studies, 20(3), 741–768.

    Article  Google Scholar 

  24. Jung, M.J., Wong, M.H.F., Zhang, X.F. (2017). Buy-side analysts and earnings conference calls. Journal of Accounting Research. forthcoming.

  25. Kasznik, R., & Mcnichols, M. (2002). Does meeting expectations matter? evidence from analyst forecast revisions and share prices. Journal of Accounting Research, 40(3), 727–759.

    Article  Google Scholar 

  26. Kirk, M., & Markov, S. (2016). Come on over: Analyst/investor days as a disclosure medium. The Accounting Review, 91(6), 1725–1750.

    Article  Google Scholar 

  27. Kothari, S., Shu, S., Wysocki, P. (2009). Do managers withhold bad news? Journal of Accounting Research, 47, 241–276.

    Article  Google Scholar 

  28. Larcker, D., & Zakolyukina, A. (2012). Detecting deceptive discussions in conference calls. Journal of Accounting Research, 50(2), 495–540.

    Article  Google Scholar 

  29. Li, E., Ramesh, K., Shen, M., Wu, J. (2015). Do analyst stock recommendations piggyback on recent corporate news? an analysis of Regular-Hour and After-Hours revisions. Journal of Accounting Research, 53(4), 821–861.

    Article  Google Scholar 

  30. Li, F. (2010). The information content of forward-looking statements in corporate filings-A naïve Bayesian machine learning approach. Journal of Accounting Research, 48(5), 1049–1102.

    Article  Google Scholar 

  31. Li, K.H. (2015). What is the value of sell-side analysts? evidence from coverage initiations and terminations. Journal of Accounting and Economics, 60(2–3), 141–160.

    Article  Google Scholar 

  32. Loh, R., & Stulz, R. (2010). When are analyst recommendation changes influential?. Review of Financial Studies, 24(2), 593–627.

    Article  Google Scholar 

  33. Loughran, T., & Mcdonald, B. (2011). When Is a Liability Not a Liability? Textual Analysis, Dictionaries, and 10-Ks. The Journal of Finance, 66(1), 35–65.

    Article  Google Scholar 

  34. Lui, D., Markov, S., Tamayo, A. (2012). Equity analysts and the market’s assessment of risk. Journal of Accounting Research, 50(5), 1287–1317.

    Article  Google Scholar 

  35. Lys, T., & Sohn, S. (1990). The association between revisions of financial analysts’ earnings forecasts and security-price changes. Journal of Accounting and Economics, 13, 341–363.

    Article  Google Scholar 

  36. Madhavan, A. (2000). Market microstructure: a survey. Journal of Financial Markets, 3(3), 205–258.

    Article  Google Scholar 

  37. Madhavan, A. (2015). Exchange-Traded Funds: an overview of institutions, trading, and impacts. Annual Review of Financial Economics, 6, 311–341.

    Article  Google Scholar 

  38. Matsumoto, D., Pronk, M., Roelofsen, E. (2011). What makes conference calls useful? the information content of managers’ presentations and analysts’ discussion sessions. The Accounting Review, 86(4), 1383–1414.

    Article  Google Scholar 

  39. Mayew, W. (2008). Evidence of management discrimination among analysts during earnings conference calls. Journal of Accounting Research, 46(3), 627–659.

    Article  Google Scholar 

  40. Mayew, W., Sharp, N., Venkatachalam, M. (2012). Using earnings conference calls to identify analysts with superior private information. Review of Accounting Studies, 18(2), 386–413.

    Article  Google Scholar 

  41. Minton, B., & Schrand, C. (2016). Institutional investments in pure play stocks and implications for hedging decisions. Journal of Corporate Finance, 37, 132–151.

    Article  Google Scholar 

  42. Orie, B., Byard, D., Kile, C., Riedl, E. (2002). High-Technology Intangibles and analysts’ forecasts. Journal of Accounting Research, 40(2), 289–312.

    Article  Google Scholar 

  43. Soltes, E. (2014). Private interaction between firm management and Sell-Side analysts. Journal of Accounting Research, 52(1), 245–272.

    Article  Google Scholar 

  44. Tetlock, P. (2007). Giving content to investor sentiment: The role of media in the stock market. The Journal of Finance, 62(3), 1139–1168.

    Article  Google Scholar 

  45. Tetlock, P., Saar-Tsechansky, M., Macskassy, S. (2008). More than words: Quantifying language to measure firms’ fundamentals. The Journal of Finance, 63(3), 1437–1467.

    Article  Google Scholar 

Download references


We appreciate the helpful comments and suggestions from two anonymous reviewers, Patricia Dechow (the editor), Andrei Shleifer, and seminar participants at Arizona State University, Carnegie Mellon University, HKUST, London Business School, and the University of Michigan. We also thank Justin Lahart for profiling this study in The Wall Street Journal.

Author information



Corresponding author

Correspondence to Jordan Schoenfeld.



Appendix A: Sample of Remarks from Earnings Conference Calls from 2002 to 2013

Analyst comment Call Positive words Negative words
“... the inside the regional mall stores were the ones that were problematic. Can you break out the average weekly sales or confirm that and sort of give us a sense of that differential?” California Pizza Kitchen, Oct. 23, 2003 0 1
“Al, you’ve done a great job of positioning the company to the future in terms of external changes ... Take us through your growth parameters, growth focus; what do you think you can do?” H.B. Fuller Company, Sept. 24, 2003 1 0
Management Comment    
“Looking forward, we see pullet placements, which is usually a good indicator, being at a small percentage increase, basically keeping up, it would seems to me, like with population growth. So I don’t see any major supply increases going forward. Demand continues to be good.” Gold Kist, Inc., Feb. 9, 2005 2 0
“Thirdly, and this is probably the most honest thing I can give everybody who is on this telephone call, we’ve dropped the ball a little bit this last period. With the end of the overall refi period, our varying programs we usually emphasize in terms of kind of a guarding against prepayments seemed to disappear a little.” BankUnited, FSB, Jan. 19, 2005 0 2
  1. The bolded words represent words from our positive and negative tone dictionaries.

Appendix B: Variable Definitions

Variable* Definition Data source
Tone and Returns Measures
Analyst Q&A Toneit (Analyst positive wordsit - analyst negative wordsit) / (analyst positive wordsit + analyst negative wordsit) Thomson Reuters Call Transcript
Management Presentation Toneit (CEO & CFO positive wordsit - CEO &CFO negative wordsit) / (CEO & CFO positive wordsit + CEO & CFO negative wordsit) Thomson Reuters Call Transcript
Management Q&A Toneit (CEO & CFO positive wordsit - CEO & CFO negative wordsit) / (CEO & CFO positive wordsit + CEO & CFO negative wordsit) Thomson Reuters Call Transcript
TAQ Abnormal Returnsit Holding period return from the start to the end of the time interval being measured, net of the value weighted market return over the same time Trade and Quote (TAQ) Database
Individual Analyst Outputs for Analysts Matched from Conference Call Transcript
Individual Δ EPS Forecastita (Analyst EPS forecast for quarter t + 1 measured at end of day + 20 - analyst EPS forecast for quarter t + 1 measured at end of day 0), 0 if no change I/B/E/S
Individual Δ Price Targetita (Stock price target at end of day + 20 - stock price target at end of day 0) / stock price target at end of day 0, 0 if no change I/B/E/S
Individual Δ Recommendationita Indicator variable that equals 1 if analyst upgrades stock, -1 if analyst downgrades stock, and 0 for no recommendation change from end of day 0 to end of day + 20 (we do not distinguish between different types of upgrades and downgrades) I/B/E/S
Time-Varying Firm Variables
Earnings Surpriseit (Actual EPSit - analyst consensus mean forecast EPSit) / stock price at fiscal quarter end dateit I/B/E/S
Large Positive Surpriseit Indicator that equals 1 if earnings surprise is in top 20% of sample earnings surprisesit I/B/E/S
Large Negative Surpriseit Indicator that equals 1 if earnings surprise is in bottom 20% of sample earnings surprisesit I/B/E/S
Meet/Beat Analyst Forecastit Indicator that equals 1 if actual EPS equals or exceeds analyst consensus mean forecast EPSit I/B/E/S
Sizeit Log of total assetsit Compustat
Market to Bookit Market valueit / book value of assetsit Compustat, CRSP
ROAit Income before extraordinary itemsit / total assetsit Compustat
Log of Analyst Followingit Log of outstanding analyst EPS forecasts at conference call dateit I/B/E/S
S.D. of Analyst EPS Forecastsit Standard deviation of analyst EPS forecasts scaled by stock price on conference call dateit I/B/E/S
Institutional Ownershipit Percentage of common stock held by institutional 13F filers at fiscal quarter end dateit Thomson Reuters 13F Holdings Database
Cumulative Abnormal Returns (CAR)it Firm returns from CRSP net of the value weighted market return CRSP
Abnormal Institutional Tradingit Abnormal daily institutional trading imbalance (net of control period; see Section 2 for precise equations) Ancerno
Conference Call Attribute Variables
Log of Total Wordsit Log of total words spoken by CEO & CFO and analystsit Thomson Reuters Call Transcript
Log of Pres. Wordsit Log of total words spoken by CEO & CFO during presentation portion of callit Thomson Reuters Call Transcript
Log of Mgmt. Q&A Wordsit Log of total words spoken by CEO & CFO during Q&A portion of callit Thomson Reuters Call Transcript
Log of Analyst Q&A Wordsit Log of total words spoken by analysts during Q&A portion of callit Thomson Reuters Call Transcript
Log of Mgmt. Pres. Fin. Wordsit Log of total financially oriented words (Matsumoto et al., 2011) spoken by CEO & CFO during presentation portion of callit Thomson Reuters Call Transcript
Log of Mgmt. Q&A Fin. Wordsit Log of total financially oriented words (Matsumoto et al., 2011) spoken by CEO & CFO during Q&A portion of callit Thomson Reuters Call Transcript
Log of Analyst Q&A Fin. Wordsit Log of total financially oriented words (Matsumoto et al., 2011) spoken by analysts during Q&A portion of callit Thomson Reuters Call Transcript
Log of Mgmt. Pres. FLSit Log of total forward-looking words (Matsumoto et al., 2011) spoken by CEO & CFO during presentation portion of callit Thomson Reuters Call Transcript
Log of Mgmt. Q&A FLSit Log of total forward-looking words (Matsumoto et al., 2011) spoken by CEO & CFO during Q&A portion of callit Thomson Reuters Call Transcript
Log of Analyst Q&A FLSit Log of total forward-looking words (Matsumoto et al., 2011) spoken by analysts during Q&A portion of callit Thomson Reuters Call Transcript
Morning Callit Indicator variable that equals 1 if call starts before noon EST, 0 otherwise Thomson Reuters Call Transcript
  1. *Balance sheet and income statement data are for the fiscal quarter that precedes the call date. Index it represents firm i’s conference call for year-quarter t. Index a represents the individual analyst. Day 0 is the conference call day.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Chen, J.V., Nagar, V. & Schoenfeld, J. Manager-analyst conversations in earnings conference calls. Rev Account Stud 23, 1315–1354 (2018).

Download citation


  • Conference calls
  • Corporate disclosure
  • Financial analysts

JEL Classification

  • G14
  • G20
  • D83