The Roles of Data Providers and Analysts in the Production, Dissemination, and Pricing of Street Earnings

• Published date: 01 December 2022
• Author: KHRYSTYNA BOCHKAY,STAN MARKOV,MUSA SUBASI,ERIC WEISBROD
• Date: 01 December 2022
• DOI: http://doi.org/10.1111/1475-679X.12457

DOI: 10.1111/1475-679X.12457

Journal of Accounting Research

Vol. 60 No. 5 December 2022

Printed in U.S.A.

The Roles of Data Providers and

Analysts in the Production,

Dissemination, and Pricing of Street

Earnings

KHRYSTYNA BOCHKAY,∗STAN MARKOV,†MUSA SUBASI ,‡

AND ERIC WEISBROD §

Received 18 January 2021; accepted 29 May 2022

ABSTRACT

In September 2009, Thomson Reuters (TR) discontinued its practice of

relying on analysts to determine the treatment of unexpected charges and

gains in favor of their immediate exclusion from GAAP earnings. Adopting

a difference-in-differences approach, we show that this plausibly exogenous

∗University of Miami; †University of Texasat Dallas; ‡University of Mar yland; §University of

Kansas

Accepted by Philip Berger.We would like to thank two anonymous reviewers for their help-

ful comments and suggestions. We also thank Sam Bonsall, Indraneel Chakraborty, Roman

Chychyla, Ed deHaan, Atif Ellahie, Fabrizio Ferri, Kurt Gee, Eric Holzman, Peter Joos, Zach

Kaplan, Steve Karolyi, Kalin Kolev, Stephannie Larocque, Edward Li, DJ Nanda, Jed Neilson,

Joseph Pacelli (discussant), Sundaresh Ramnath, Thomas Ruchti, Harm Schutt, Erin Towry,

Brady Twedt, Ben Whipple, and seminar participants at the University of Miami, Ohio State

University, Erasmus University Rotterdam, Northwestern University, Penn State University,

Southern Methodist University,University of Rochester, Temple University,University of Geor-

gia, 2019 Hawaii Accounting Research Conference, 2019 FARS Midyear Meeting, University of

California Riverside, Washington University, Dartmouth College, University of Houston, Uni-

versity of Kansas, CUNY Baruch, University of Utah, and American University for their help-

ful comments and suggestions. Previous versions of this paper were circulated with the titles

“Street Earnings Activation Delay” and “The Dissemination and Pricing of Street Earnings.”

An online appendix to this paper can be downloaded at https://www.chicagobooth.edu/jar-

online-supplements.

1695

© 2022 The Authors. Journal of Accounting Research published by Wiley Periodicals LLC on behalf of The

Chookaszian Accounting Research Center at the University of Chicago Booth School of Business.

This is an open access article under the terms of the Creative Commons

Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium,

provided the original work is properly cited, the use is non-commercial and no modifications or

adaptations are made.

1696 k. bochkay, s. markov, m. subasi, and e. weisbrod

change in TR’s methodology resulted in street earnings that are more predic-

tive of future performance; and timelier, more accurate, and less dispersed

analyst forecasts of future earnings, consistent with TR enhancing the prop-

erties of street earnings and analyst forecasts. Finally, using path analysis we

show that a significant portion of TR’s effect on price discovery is through its

effect on analysts; and that the change in TR’s treatment of unexpected items

increased (decreased) the relative influence of TR (analysts) on the pricing

of street earnings. We conclude that forecast data providers like TR are more

than a conduit of information from analysts to investors.

JEL codes: D62, D83, D84, G14, G24, M40, M41

Keywords: analyst information production; forecast data providers;

I/B/E/S; price discovery; street earnings processing

1. Introduction

A nascent but growing literature examines the role of forecast data

providers (FDPs) in capital markets and concludes that they add value

by aggregating and disseminating information produced by analysts

such as earnings estimates, stock recommendations, and street earnings

(Akbas et al. [2018], Kaplan, Martin, and Xie [2021], Schaub [2018]). In

this study, we exploit a plausibly exogenous change in how a prominent

FDP, Thomson Reuters (TR), produces and distributes street earnings to

examine whether TR also influences analyst information production activ-

ities and, more broadly, to deepen our understanding of how FDPs add

value in capital markets.

Street earnings are GAAP earnings adjusted for non-cash or transitory

items and distributed by FDPs (Black et al. [2018]).1Street earnings are

prevalent, more predictive of future performance and more impactful than

GAAP earnings (e.g, Black et al. [2018], Bradshaw et al. [2018], Doyle,

Lundholm, and Soliman [2006]). With the properties of street earnings

dependent on which items are excluded from GAAP earnings, there is a

substantial interest in understanding forces that shape analyst exclusions.

Prior literature has thoroughly documented the role of managers in in-

fluencing analyst exclusions and shaping the properties of street earnings

(Black et al. [2019], Bentley et al. [2018], Christensen et al. [2011], Doyle,

Jennings, and Soliman [2013]), but largely overlooked the role of FDPs

for two main reasons. First, FDPs claim to report street earnings as deter-

mined by analysts. For example, TR states that its “goal is to present actuals

on an operating basis, whereby a corporation’s reported earnings are ad-

justed to reflect the basis that the majority of contributors use to value the

stock” (Thomson Reuters [2009]). Second, neither FDPs’ interactions with

analysts nor FDPs’ own activities are directly observable. Stymied by lack of

1More specifically, street earnings in our study is the Earnings per Share (EPS) actual from

the I/B/E/S database. Street earnings often differ from GAAP earnings, but not in all cases

(Bentley et al. [2018]).

the roles of data providers and analysts 1697

observational data, researchers have accepted FDPs’ self-declared objective

to report street earnings as determined by equity analysts as the represen-

tation of FDPs’ actual role.

Drawing on institutional evidence, we suggest that reporting street earn-

ings as determined by analysts may sometimes impose a cost on FDPs that

they may not want to bear. Specifically, GAAP earnings often include un-

expected gains and charges whose treatment by analysts is not known un-

til after earnings are released. Delaying dissemination until all, or most,

analysts process the earnings release to assess the nature of each of these

line items and to determine street earnings increases the likelihood that

investors obtain earnings information from alternative sources (analysts

or media), reducing FDPs’ usefulness to investors as a source of street

earnings.

A 2009 change in how TR produces and distributes street earnings exem-

plifies TR’s unwillingness to bear the cost of delayed street earnings report-

ing. According to TR’s own policy documents, some earnings releases in-

clude unexpected charges and gains whose treatment by analysts—exclude

from GAAP earnings or not—is known only after earnings are announced

and processed by analysts (unexpected line items, henceforth). With the

express goal of improving the timeliness of its reporting of street earn-

ings, TR discontinued its policy of reporting street earnings after ascer-

taining analysts’ treatment of unexpected line items in favor of reporting

street earnings immediately and exclusive of unexpected line items. This

plausibly exogenous change in TR’s activities presents a unique opportu-

nity to identify TR’s influence on street earnings and analyst information

production.2

We begin by investigating the effect of TR’s methodology change on

two key properties of TR-reported street earnings: timeliness and ability

to predict future performance. Adopting a difference-in-differences (DiD)

approach, we assign earnings announcements containing unexpected line

items to the treatment group and all other announcements to the con-

trol group; and we define the pre- and post-periods as 2005–September

2009 and October 2009–2019. We find that the timeliness of street earn-

ings dissemination in the treatment sample increases by approximately 6.1

hours relative to the control sample following the methodology change. We

also find that the ability of street earnings to predict future cash flows, as

captured by the slope coefficient in a typical cash flow prediction model

(e.g., Doyle, Lundholm, and Soliman [2003]), is 35% lower in the treat-

ment group prior to the methodology change, but increases approximately

21% relative to the control group as a result of the methodology change.

This finding suggests that analysts did not exclude all unexpected line items

2Our conversations with TR staff reveal that the change in methodology occurred as part of

TR’s integration of its First Call and I/B/E/S databases, an event whose timing is likely exoge-

nous to how analysts treat unexpected line items. See subsection 2.2 for institutional details.