Event Studies Over Short Time Windows

is hard to control for, and the literature has struggled to introduce convincing instruments.

Sixth, studies usedsettings in which decision-makers are affected dif-ferently by the IFRS mandate. In this case, the analysis is performed at the decision-maker level. For example, Hortonet al. (2013) distinguish between (1) analysts covering only firms that used a single local GAAP before IFRS adoption, though some of these firms use IFRS and some use local GAAP afterward (for them, comparabilitydecreased); (2) ana-lysts covering only firms that used a single local GAAP before adoption, and all used IFRS afterward (for them, comparability wasunchanged);

and (3) analysts covering firms that used different local GAAPs be-fore adoption, but all used IFRS afterward (for them, comparability increased). Thus, IFRS treatment leads to different consequences de-pending on coverage before the IFRS mandate, and identification rests on the extent to which coverage allocation was exogenous. Comparable identification strategies have been applied to loan syndicates (Brown, 2016) and institutional investors (Covrig et al., 2007; DeFondet al., 2011).

We next discuss challenges of studies that test market reactions to (1) regulatory announcements increasing or decreasing the likelihood of IFRS adoption by a jurisdiction, (2) firms’ disclosure of reconciliation adjustments to IFRS, and (3) firm-specific information events.

3.3.1.1 Market Reactions to Key Regulatory Events

New regulation typically evolves in a legislative process that has multiple steps. Event studies accumulate abnormal returns over a series of events that increase or decrease the likelihood of occurrence of the regulatory change.²⁸

Armstrong et al.(2010) assess investors’ perception of net benefits (or costs) of mandating IFRS by investigating the stock price reaction to 16 selected regulatory events between 2002 and 2005 that supposedly increased or decreased the likelihood of the European Union’s adoption of IFRS. While the study’s results are often interpreted as documenting a positive market assessment, the choice of the market-adjustment for abnormal returns is crucial for this interpretation. The average raw market responses are negative, and the majority of events (11 out of 16) yield reactions in a direction opposite to the prediction (Armstrong et al., 2010). Several events were also overlapping with other legislative announcements (De George et al., 2016). The selection of events is subjective, as all analyzed events occurredafterthe European Parliament had passed the regulation in March 2002 that required all listed firms in the European Union to report under IFRS from 2005 onwards (the Parliament’s resolution usually marks the end to the regulatory process in the EU). Somewhat surprisingly, the study does not include any events prior to this decision.²⁹ By considering only events after the general IFRS adoption decision was already made, their

28One limitation is that governments often pass multiple regulations at the same time. It can then be difficult to isolate the effects of one regulatory change. Researchers can still test the validity of expectations about cross-sectional differences in firms’

exposure to the regulatory change if these firm-characteristics are uncorrelated with the effects of other regulations passed at the time (Leuz and Wysocki,2016).

29Plausible events are the announcements by EUROFIN or the European Com-mission earlier in the European Union’s regulatory process that both provided strong support for the ultimate decision; see Armstronget al.(2010), and footnotes 8 and 9.

events relate for the most part to the IFRSendorsement process, and specifically to the discussions of one controversial standard, IAS 39. One could argue that the study therefore primarily captures how the market perceived deviations from IFRS standards as issued by the IASB or, even more narrowly, deviations from IAS 39.

Joos and Leung (2013) apply a very similar idea to the U.S. setting and study the SEC’s deliberations between 2007 and 2009 on whether the United States should adopt IFRS for domestic issuers. Due to the setting in which the adoption process faded out in 2012, the 15 policy announcements capture rather early stages in the SEC’s rule-making process (see Beckeret al.,2020b). The authors find, on average, positive market reactions of +0.86% to events that increased the likelihood of IFRS adoption (marginally statistically significant). However, 13 out of 15 events coincide with the stock market crash of 2008.

Other studies that assess market perceptions based on event returns use events that relate to pronouncements of specific reporting standards.

For example, Onali and Ginesti (2014) study stock price reactions to the IASB’s deliberations on the replacement of IAS 39 by IFRS 9. Their study illustrates the difficulties when applying the event study design to individual IFRS pronouncements. The “events” represent rather technical discussions in the due process of the IASB and a jurisdictions’

endorsement bodies (such as European Financial Reporting Advisory Group (EFRAG) in case of IFRS 9). These events are hard to select, lack visibility, and are often difficult to clearly interpret.³⁰ The reported cumulative abnormal returns of around +10% to the revision of an

30While the authors initially identified 20 events, they ultimately considered only 11, because some were not newsworthy enough to be covered by the business press, while for others it was impossible to infer whether they increased or decreased the likelihood of the standard’s issuance. Furthermore, technical discussions during the development of a new standard affected not only the likelihood of the standard’s issuance but also its content (in case of IFRS 9, there were material changes during its drafting).

individual standard seem economically unreasonably high (compared to the modest reactions to the entire set of IFRS).³¹

All these studies capture investors’ ex-ante perceptions and expecta-tionsof the regulatory change, rather thanrealizationsof its effects after implementation. Expectations are likely to differ substantially from re-alized effects, given that many IFRS adoption deficiencies only became visible over time (e.g., in form of IFRS implementation differences at the firm or jurisdictional level; see Sections4 and 5).

3.3.1.2 Market Reactions to Reconciliation Adjustments

In some settings, regulators mandated the separate disclosure of rec-onciliations from local GAAP to IFRS (or from IFRS to U.S. GAAP), such that researchers can assess, by the market’s reaction to the release, whether these reconciliations contain incremental information.³²

In the United Kingdom, when transitioning from UK GAAP to IFRS, a subset of firms published an IFRS reconciliation separately from other disclosures. Horton and Serafeim (2010) study the market reaction to these announcements, conditional on earnings differences between IFRS and the UK GAAP results that were published earlier. While earnings were, on average, higher under IFRS (thus positive news), only lower IFRS earnings triggered significantly negative market reactions. The authors conclude that IFRS numbers are value relevant for investors.

Christensen et al. (2009) re-examine reactions to negative earnings reconciliations and show that these reactions are more pronounced for firms that face a greater likelihood of covenant violations and higher costs from these violations, due to rolling GAAP covenants (which

31The authors do not report stock price reactions to the individual events, but only aggregated over all events. They do not cover the entire deliberative process.

The latest event used in the study covers the IASB’s issuance of IFRS 9, although the EU’s endorsement process was still ongoing. In a sense, their selection of events stops where Armstronget al.(2010) start.

32A market reaction that is consistent with the direction of the reconciled difference in earnings is often interpreted as evidence for the higher quality of an accounting regime. However, such an interpretation is premature, as reconciliations to some other GAAP are releasedafter the results under the original GAAP had been disclosed.

Thus, independent of the quality of the accounting system, reconciliations may also act as an additional signal of the precision of the initially disclosed earnings.

economically reflects a wealth transfer from shareholders to creditors).

These findings illustrate that interpretations of IFRS market reactions require careful examination. In the UK case, transition (or contracting) costs played a decisive role beyond the mere changes to reporting quality.³³

In the United States, until 2007, reconciliations from IFRS to U.S.

GAAP were a required part of the annual Form 20-F (Harris and Muller, 1999).³⁴ A number of studies examine reactions to the earnings recon-ciliations from IFRS to U.S. GAAP by foreign issuers. Chen and Sami (2008, 2013) examine trading volume responses of American depositary receipts (ADRs) and of shares traded on a firm’s home stock exchange.

They find that the magnitude of reconciliations is associated with ab-normal trading volume during the days surrounding the Form 20-F filing date.³⁵ Kimet al.(2012) and Chen and Khurana (2015) study the stock market reactions to the SEC’s decision in 2007 to eliminate 20-F reconciliations for IFRS issuers. Chen and Khurana (2015) examine reactions to four key events that led to the SEC decision and document an overall positive stock market-reaction that is positively (negatively) associated with proxies for savings of dual reporting costs (the mag-nitude of IFRS reconciliation adjustments). Kim et al.(2012) further refute claims that eliminating 20-F reconciliations harmed investors’

information environment by providing evidence that a large variety of market-based measures were unaffected after reconciliations were lifted.

Overall, the literature concludes that, within the very specific setting, reconciliations from IFRS to U.S. GAAP no longer included material

33This setting entails the limitation that firms self-select to report their IFRS transition documents separately. It’s also possible the market can predict firms’ IFRS-UK adjustments, because similar reconciliations may have already been disclosed by other firms. Thus, the true benchmark for the market reaction to the IFRS reconciliation release would be the predicted IFRS earnings, which are unobservable, rather than the local UK GAAP earnings that the studies use as a proxy.

34The disclosure of first-time reconciliations from IFRS to U.S. GAAP plays a key role in this setting since the market should be able to at least partially anticipate the content of firms’ reconciliations in the future due to the repetitiveness of some reconciliation items.

35However, the form 20-F contains much more information than reconciliations.

Thus, the trading response may be driven by other factors that are correlated with the magnitude of reconciliations.

information after the IFRS had developed into a comprehensive set of reporting standards.

3.3.1.3 Market Reactions to Firms’ Information Events

A last set of event studies examines changes in stock price reactions to firms’ information events, bridging event study and difference-in-differences designs. Information events can be an individual firm’s announcement to adopt IFRS, its own earnings announcement, infor-mation spillovers from competing firms’ earnings announcements, or the disclosure of private information.

A first type of study examines the stock price response to firms’ an-nouncement to voluntarily adopt IFRS before a mandate and persistently finds significant positive abnormal returns. For example, Karamanou and Nishiotis (2009) find an average reaction of +0.73% to 59 firm announcements before 2002 (the EU announcement of the mandate) and attribute this reaction to these firm’s signaling their “high-value”

type and “bonding” to increased transparency (to the extent that vol-untary IFRS adoption is costly and represents a commitment; Leuz and Verrecchia,2000).

A second type of study (for early evidence, see Auer,1996) examines changes in the information content of earnings announcements captured by abnormal return volatility and trading volume. The use of second moments builds on the argument that return volatility and trading volume reflectidiosyncraticinterpretations of the announcement, rather than anaverage change in investors’ beliefs that price changes in first moments would reflect. The greater the information content of an announcement, the more likely investors will interpret the content differently, and the more frequently they will trade. If IFRS improve the information content of firms’ disclosures, return volatility and trading volume around earnings announcements should increase. Using a global sample of firms, Landsman et al. (2012) show a positive association between IFRS adoption and these two measures, suggesting that IFRS deliver more information.

Yet there are limitations to be considered. First, during the sample period of most studies, there had been exceptionally high noise in IBES

announcement dates of international firms (DeFond et al.,2007), and an increase in the precision of announcement dates in the database over time could equally drive these results.³⁶ Second, the underlying theory is ambiguous. Given that IFRS should provide a level playing field on global markets, one could also argue that IFRS should reduce, instead of increase, differences in investors’ assessments.³⁷ Third, a number of innovations in the communication of earnings news overlapped with IFRS adoption, which makes the attribution of the reaction difficult.

For example, around adoption, reporting lag decreased (Landsman et al.,2012), the electronic availability of announcement dates improved (DeFond et al., 2007), earnings guidance was initiated (Li and Yang, 2016), and more detailed financial statements started to be included in earnings announcements (Kimet al.,2019).

A third type of study tests whether IFRS adoption results in in-creased cross-border information transfers and spillovers (Wang,2014;

Yip and Young, 2012). For example, Wang (2014) studies how earnings announcements of global industry leaders (the three largest firms in an industry) translate into price reactions of all non-announcing firms in the same industry domiciled in a country different from that of the an-nouncing firm. The study finds that non-anan-nouncing foreign firms react more strongly to the earnings announcements of a global leader when both firms report under IFRS and transnational information transfers are stronger when firm-pairs are located in countries which economies are more tightly integrated.

A forth type of study looks into changes of market reactions to other information events. For example, in the case of insider trading, Brochetet al. (2013) predict that the adoption of IFRS should result in lower market reactions, because IFRS reduce private information. In fact, they find that mandatory adoption significantly reduces abnormal returns associated with insider purchases after adoption, supporting

36This argument would hold in a difference-in-differences design in case these IBES announcement dates were already precise in the pre-period for the U.S. firms in the benchmark group, which is not unlikely.

37De Georgeet al.(2016) argue that without knowing the identity of the traders, it is difficult to attribute the observed changes to IFRS.

the view that IFRS decreases information asymmetry between insiders and outsiders.

3.3.2 Difference-in-Differences Analysis Over Longer