IFRS and accounting quality: The impact of enforcement

Abstract

The aim of this study is to investigate the impact of enforcement (greater monitoring of auditors and more regulatory scrutiny of financial reporting) on accounting quality under IFRS using measures of earnings smoothing, managing towards earnings targets, timely loss recognition and value relevance. Our sample consists of 495 UK listed firms in 2000-2009 under UK GAAP and IFRS (2,356 and 1,823 firm-years respectively), including 246 cross-listed firms (predominantly listed in Germany and the US). We find value relevance improves under IFRS for all firms and that firms are less likely to manage towards earnings targets. However measures based on earnings smoothing and timely loss recognition improve only for cross listed firms, pointing to a favourable impact of changes in the regulatory scrutiny of cross listed firms in the IFRS period.

JEL Codes: M40, M41 Key words: IFRS, accounting quality, public enforcement, auditor oversight, security market regulators

IFRS and accounting quality: The impact of enforcement

1. Introduction

The aim of this study is to investigate the impact of enforcement on accounting quality under International Financial Reporting Standards (IFRS). IFRS have been adopted in many countries with the aim of improving financial reporting and consequently increasing market efficiency and promoting cross border investment (IFRS, 2012). However, it is widely recognised that the quality of financial reporting is affected by many factors including firm incentives and the institutional setting in which financial reporting takes place (Ball, 2006; Brown, 2011; Brggemann, Hitz and Sellhorn, 2012).

Studies show national legal systems are linked to capital market development (La Porta, Lopez-de-Silanes, Shleifer and Vishny, 1998) and the quality of financial reporting (Leuz, Nanda and Wysocki, 2003). Investigations of adoption of IFRS commonly find country-level institutional setting is associated with the extent to which benefits from IFRS are realised (Daske, Hail, Leuz and Verdi, 2008; Beuselink, Joos, Khurana and Van der Meulen, 2009; Flourou and Pope, 2011; Landsman, Maydew and Thornock, 2012). The institutional setting has many elements including legal

requirements (eg company law, accounting standards), listing rules of stock exchanges, corporate governance statutory and best practice requirements, the ability of parties to protect their rights through legal action, and the strength of enforcement of the law, via public and private mechanisms.

Aspects of the institutional setting of particular relevance for the quality of financial reporting under IFRS relate to the enforcement of compliance with IFRS via the activities of auditors and public or private sector regulatory bodies with the responsibility for promoting auditor independence and compliance with accounting standards. Practitioners and regulators claim that enforcement of accounting standards is an integral part of achieving accounting quality under IFRS (Schipper, 2005; Whittington, 2005; Leuz, 2010; SEC, 2002; CESR, 2003). 1 In addition, in many countries, auditors have traditionally played a key role in promoting accounting quality
1

We define enforcement as the monitoring and sanctioning activities of public and private entities to promote quality of audit and compliance with accounting standards.

(FEE, 2000; FEE, 2001; Baker, Mikol and Quick, 2001; Fearnley and Hines, 2003). The World Bank and the IFRS Foundation have pointed to a sound financial reporting infrastructure as essential for achieving the hoped-for benefit of IFRS (Hegarty, Gielen and Barros, 2004; IFRS, 2012).

In many jurisdictions, the oversight and monitoring of auditors has increased (Baker et al., 2001; Fearnley and Beattie, 2004). Following the events relating to Enron and Arthur Andersen in the US, countries have introduced initiatives to improve auditor independence and the quality of auditors work including the creation of auditor oversight boards (IFAC, 2011; Brown, Preiato and Tarca, 2012). In addition, it is argued that high quality financial reporting cannot be achieved without supervisory activity by independent enforcement bodies (SEC, 2000; CESR, 2003). Consequently, the adoption of IFRS has been accompanied by activities to promote compliance with IFRS by regulatory bodies in the countries where IFRS are adopted. It is expected that the activities of auditors and enforcement bodies have an important role in promoting accounting quality. However, there is lack of evidence about the contribution of these entities in realising benefits of IFRS.

To test the research proposition that more auditor oversight and monitoring of financial reporting by independent bodies is associated with higher quality reporting under IFRS we focus on cross listed firms from the United Kingdom (UK). The UK listed firms adopted IFRS from 1 January 2005, however there were also changes in the monitoring of compliance with accounting standards at the same time in the UK (FRRP 2004; FRRP 2005). Because we cannot distinguish between the effects of IFRS and the effects of enforcement for UK firms in general, we focus on UK cross listed firms that experienced heightened enforcement in the IFRS period because of increases in oversight of auditors and monitoring of financial reporting in the countries where they are cross listed.

We study UK firms listed on the London Stock Exchange because the UK is an economically important capital market with a strong legal system and high levels of enforcement activity in the period prior to IFRS adoption (Kaufmann, Kraay and Mastruzzi, 2010; Jackson and Roe, 2009). In addition, UK firms did not voluntarily adopt IFRS prior to 2005 (thus we avoid issues related to self-selection bias in relation to IFRS) and many UK firms have foreign stock exchange listings. Our sample

consists of 495 firms for five years under UK GAAP (2000/01-20004/05) and four years under IFRS (2005/6-2008/9) giving 2,356 firm-year observations under UK GAAP and 1,823 under IFRS. The sample includes 246 cross-listed firms, of which all except one are listed in Germany. Among the 246 firms, 131 (53%) are additionally cross listed in the US including 24 that are listed on the NYSE. All cross listed firms were cross listed for the whole period and adopted IFRS from 1 January 2005.

Lang, Raedy and Yetman (2003) report that foreign firms listed in the US display less earnings management, a tendency for more timely recognition of bad news and stronger associations between accounting data and share price than firms that are not cross-listed in the US. Supporting Coffees (2002) bonding hypothesis, they link higher accounting quality to the effect of firms placing themselves under the jurisdiction of the US market regulator, the Securities and Exchange Commission (SEC), and within the US legal system.

We investigate whether a similar cross listing effect is observed for UK firms with foreign listings, but we focus on the effect of increases in scrutiny for cross listed firms that occurred around the time of adoption of IFRS. We compare accounting quality for cross listed firms before and after adoption of IFRS, using each firm as its own control. We also compare the cross listed firms to a control sample of non cross listed firms that adopted IFRS in the same period. Sun, Cahan and Emanuel (2011) report improvements in accounting quality (less managing to earnings targets) for UK and Australian firms cross listed in the US but they do not explicitly consider the changes in enforcement that occurred in the time period of their study. We explore whether changes in accounting quality are linked to heightened enforcement. If the greater enforcement activity in the IFRS period is effective (for example in Germany and the US) we should observe higher accounting quality for cross listed firms under IFRS compared to their accounting quality under UK GAAP, and compared to non cross listed firms.

Consistent with prior studies, we use accounting quality metrics based on earnings smoothing, managing towards earnings targets, timely loss recognition and value relevance (Lang et al., 2003; Van Tendeloo and Vanstralen, 2005; Lang, Raedy and Wilson, 2006; Barth, Landsman and Lang, 2008; Paananen and Lin, 2009; Sun, Cahan and Emmanuel, 2011). First, we compare these measures for UK firms before and

after adoption of IFRS. Next, we compare the measures for cross listed and non cross listed firms. We find value relevance increases for all UK sample firms under IFRS. Despite high quality UK GAAP and a strong institutional setting throughout the study period, we show value relevance (measured by models testing the association of prices, returns and accounting numbers) is greater under IFRS.

We also examine factors that may have contributed to improved value relevance. On the one hand, more enforcement activity in the UK may have affected accounting quality. However we are unable to isolate the effect of changes in enforcement in the UK from the effects of the new standards themselves. Considering the standards, we investigate whether greater use of fair value measurement (for financial and intangible assets) and more disclosure requirements under IFRS (compared to UK GAAP) contributed to the increase in value relevance. Generally, we were unable to link these specific changes to value relevance although we do observe less managing towards earnings targets by firms with larger amounts of financial assets, financial liabilities and intangible assets.

In relation to the other measures of accounting quality, only small loss avoidance indicates an improvement in accounting quality for UK firms under IFRS. However, we find strong evidence that the accounting quality of cross listed firms is higher in the IFRS period. For the 246 cross listed firms that used UK GAAP then IFRS (2,099 firmyears), we observe improvements in accounting quality metrics based on earnings smoothing, managing towards earnings targets and timely loss recognition under IFRS. The same pattern is not observed for the 249 firms that are not cross listed.

The results for cross listed firms could be due to change in enforcement, changes in standards, or differences in firm characteristics between cross listed firms and others. Because we do not observe an increase in accounting quality for the control sample of non cross listed firms that also adopt IFRS at the same time (and we include controls for differences between firms in size, leverage, turnover, growth, capital raising activity and audit firm in all models used to measure accounting quality), we conclude the change in accounting quality is associated with the additional regulatory scrutiny experienced by the cross listed firms.

Our study makes several contributions to the literature about the impact of IFRS adoption, an event with far reaching effects for capital markets and financial reporting. First, we consider the issue of the quality of reporting under IFRS. Regulators, standard setters and practitioners want to know about the extent to which the costs of IFRS have generated benefits. Our evidence suggests the value relevance of financial reporting for UK firms has increased. The result is particularly interesting as it suggests IFRS benefits are possible even for firms from a developed market with strong enforcement. Prior evidence about the effect of IFRS on accounting quality in a number of countries, for example Chen, Tang, Jiang and Lin (2010) and Ahmed, Neel and Wang (2012), is mixed, possibly because some studies consider only a short time period or a small number of firms. Using a relatively large sample over a nine year period for UK firms, we demonstrate that some benefits have been realised, but they are more likely to be observed in relation to accounting quality measures using market measures and accounting numbers rather than accounting numbers alone.

We also show a beneficial effect for accounting quality under IFRS that we argue is related to greater regulatory scrutiny. There have been extensive activities to improve external audit and monitoring of financial statements by independent enforcement bodies in the IFRS period (CESR, 2009; Brown et al., 2012). Indicating the importance of enforcement activity, Christensen, Hail and Leuz (2012) report IFRS benefits (proxied by greater liquidity) occur in countries that introduced proactive review of financial statements. Sun et al. (2011) comment that some evidence of an increase in accounting quality under IFRS for UK and Australian firms cross listed in the US is somewhat surprising given the firms come from jurisdictions with strong legal enforcement. We add to these studies by considering changes in enforcement, specifically related to monitoring of auditors and increased activity by independent enforcement bodies.

Ernstberger, Stich and Volger (2012) suggest changes in enforcement in Germany in the mandatory IFRS adoption period have been beneficial. They find less earnings management, greater liquidity and some evidence of an increase in market valuation for German firms. Hitz, Ernstberger and Stich (2012) also report an impact of the German enforcement bodies because firms named for non-compliance with standards experience a negative effect on share price when this information is released. We add to these studies, based on evidence relating to UK firms in listed in Germany. Our

results suggest the increase in accounting enforcement activity is associated with higher quality reporting. Consistent with the view of Jackson and Roe (2009) that public enforcement is a key mechanism for promoting protection of investors and growth of securities markets, our evidence shows benefits from activity by public bodies to promote the quality of audit and compliance with accounting standards.

2. Literature and hypotheses

The adoption of IFRS aims to improve financial reporting by increasing its transparency and comparability. The statement announcing adoption in the European Union (EU) member states proposed that adoption of IFRS would help eliminate barriers to cross-border trading on securities markets by ensuring consolidated accounts are more reliable and transparent and can be more easily compared, with benefits for market efficiency, cost of capital, competiveness and growth (EC1606/2002). Reliability (i.e. representational faithfulness), transparency and comparability are elements of the qualitative characteristics of accounting information (IASB, 2010). The underlying proposition is that when the qualitative characteristics are present, accounting information is of higher quality because it is more useful for decision making.

Researchers have developed several approaches to measuring the concept of accounting quality, based on the proposition that accounting information has higher quality information when it is more useful for decision making (IASB, 2010). Accounting quality metrics based on earnings smoothing, managing towards earnings targets, timely loss recognition and value relevance have been developed and applied in many prior studies (Lang et al., 2003; Van Tendeloo and Vanstralen, 2005; Lang et al., 2006; Barth et al., 2008; Paananen and Lin, 2009; Sun et al., 2011). Studies have used the value relevance approach to compare financial reporting under US GAAP, IFRS and various national GAAP, notably in Germany in the period prior to mandatory adoption (Harris and Muller, 1999; Bartov, Goldberg and Kim, 2005; Hung and Subramanyan, 2007).

Agency theory (Jensen and Meckling, 1976) suggests insiders incentives and behaviours may not be aligned with the interests of parties external to the firm, in particular capital providers. Managers have incentives to hide poor performance to

protect their reputation and compensation (Moses, 1987). The process of adjusting reported earnings to be greater or lower than expected is referred to earnings smoothing (Beidleman, 1973). Earnings smoothing produces information that is biased, that is, it does not faithfully represent underlying economic performance. The usefulness of the information to firm outsiders is reduced, thus it is argued that earnings smoothing reduces accounting quality (Lang et al., 2003; Barth et al., 2008). Chen et al. (2010) study listed EU firms in 15 countries prior and subsequent to adoption of IFRS (2000-2007). They report higher levels of income smoothing and less timely recognition of losses, although the authors argue that others measures, based on discretionary accruals, suggest accounting quality has improved.

Another aspect of hiding poor performance relates to avoiding small losses. Although managers have incentives to avoid losses of any magnitude, their ability to cover up large losses is lower than for small losses. Thus there is a tendency for insiders to manage earnings to a small positive net income (Leuz et al., 2003). This results in biased information and thus lower accounting quality. Jeanjean and Stolowy (2008) report that earnings management (measured as the ratio of small reported profits to small reported losses) did not decrease in a sample of firms from France, the UK and Australia following mandatory adoption of IFRS.

Recognition of losses is considered to be timely if they are included in the financial statements as they occur instead of being spread over future periods. Ball and Shivakumar (2005) explain that timely recognition of losses increases the usefulness of financial information by influencing the behaviour of managers and capital providers. Timely loss recognition holds managers accountable for their investment decisions and promotes action to stem the losses relating to poor investment decisions. It also provides timely information to lenders, who can take action when debt covenants are breached. Thus, timely recognition of losses benefits both investors and creditors by improving the usefulness of accounting information for decision making. Considering the last year of UK GAAP and the first year of IFRS (2004-2005), Iatridis (2010) studied 241 UK listed firms (excluding financial institutions) and finds less scope for earnings manipulation via earnings smoothing, timely loss recognition and lower frequency of small profits which should result in more value relevant information for decision making. The author studied the potential rather than the actual effect as he used the 2004 figures reinstated in IFRS.

Financial information that better captures the underlying business economics of an entity is more useful. A closer association between accounting numbers and market measures such as share prices and returns is argued to mean accounting is more relevant and therefore more useful for investor decision making (Amir, Harris and Venuti, 1983; Francis and Schipper, 1999; Ali and Hwang, 2000; Hung, 2000). More value relevance is assumed to represent higher accounting quality. Devalle, Onali and Magarini (2010) report an increase in value relevance of earnings (but not book value of equity) in five European countries following adoption of IFRS.2 They find increased value relevance for earnings and equity for UK firms. In contrast, Aubert and Grudnitski (2011) find neither increase in value relevance nor timeliness for firms from nine European countries including the UK.

Ahmed et al. (2012) investigate firms from 20 countries where there was mandatory IFRS adoption. They report increases in earnings smoothing, aggressive accruals recognition and a reduction in the timeliness of loss recognition, compared to a benchmark sample of countries that did not adopt IFRS. The authors attributed the general decrease in accounting quality under IFRS to the inability of enforcement mechanisms to combat the relative increase in flexibility in IFRS compared to domestic GAAP.

Within any individual country, accounting quality could be enhanced by changes to accounting standards or to the way they are used. IFRS adopted in 2005 (the stable platform) introduced many changes to financial reporting requirements in various national GAAPs (Nobes, 2001; PwC, 2005). For example, the recognition and measurement of financial and intangible assets changed. More financial assets were recognised on the balance sheet and the relevance of the items increased as more fair value measurement was used under IAS 39 Financial Instruments: Recognition and Measurement. Similarly, more intangible assets were separately recognised and measured at fair value on initial recognition under IFRS 3 Business Combinations. IFRS were also accompanied by more extensive disclosure requirements, for example in relation to financial instruments under IAS 32 Financial Instruments: Presentation and IAS 39. Another area of change was segment reporting, where the basis of

Germany, Spain, France, Italy and the UK.

reporting was changed under IAS 14 Segment Reporting to reflect the way the business was managed (Nobes, 2001; Cairns, Massoudi, Taplin and Tarca, 2011).3

However, a key aspect of the impact of IFRS is the extent to which the new standards represented a change from existing GAAP. In the UK, accounting standards changed in the ways outlined in the previous paragraph. However, it is possible for some firms that changes were few, as many UK GAAP standards had been harmonised with IFRS prior to its adoption in 2005. Bae, Tan and Welker (2008) compare national GAAP and International Accounting Standards (IAS) in 2002 between countries, constructing measures based on the absence of IAS recognition, measurement and disclosure rules in national GAAP and inconsistencies between GAAP and the IAS at that time (based on Nobes, 2001). Considering accounting practices under 21 standards, they report only one difference between UKGAAP and IAS, the lowest result among 49 countries.

Based on the above discussion, we propose that accounting quality improves for UK firms following adoption of IFRS. We recognise that this may not occur, due to the similarities of UK GAAP and IFRS and the strong enforcement in place in the UK in the study period. Nevertheless, IFRS introduced some important changes thus we expect them to increase accounting quality.

In addition, the adoption of IFRS was accompanied by considerable attention by practitioners, the media and the business community. The UK enforcement body responsible for promoting compliance with financial reporting requirements, the Financial Reporting Review Panel (FRRP) introduced proactive enforcement (systematic review of financial statements) from 2005 (FRRP, 2004; FRRP, 2005). Thus earnings management could have been seen as more potentially more costly in the IFRS period, leading to fewer instances of earnings management practices. For the UK sample as a whole, we are unable to distinguish the effects of changes in the standards and changes in enforcement because they occurred concurrently (however, this question is addressed in our second hypothesis described below). The first hypothesis can be formally stated as:

There have been subsequent changes to recognition and disclosure requirements for financial instruments (IFRS 7 Financial Instruments: Disclosures and IFRS 9 Financial Instruments) and segment reporting (IFRS 8 Operating Segments) however these changes are applicable outside the time period of our study.

10

H1 Accounting quality is higher under IFRS compared to UK GAAP.

Ball, Kothari and Robin (2000) and Ball, Robin and Wu (2003) argue that accounting quality can also be affected by the way standards are used, that is, the institutional setting in which financial reporting occurs. FEE (2001) identifies several elements in the institutional setting for financial reporting, including regulations (eg company law, accounting standards, listing rules of stock exchanges) and enforcement, which encompasses the activities of several parties involved in financial reporting. The enforcement activities include: self-enforcement by preparers (reflecting corporate governance statutory and best practice requirements); external audit; the institutional oversight system (eg activities of securities market regulators); legal action in courts; and public exposure and sanction via public media.

Studies have demonstrated the importance of these various elements of the institutional setting for accounting quality. Beekes and Brown (2006) show that better governed firms release more informative disclosures in a timely manner. Verriest, Gaeremynck and Thornton (2012) report that firms with stronger governance mechanisms provide more transparent IFRS restatements, provide better disclosure quality and comply with IFRS more rigorously than other firms. The importance of effective independent auditors is confirmed by the regulatory activity post-2003 to improve the quality of audit through additional monitoring and oversight, carried out through such bodies as the Public Companies Auditing Oversight Board (PCAOB) in the US and the Professional Oversight Board (POB), the Audit Inspection Unit (AIU) and the Auditing Practices Board (APB) in the UK (FRC, 2010; AIU, 2005). Suggesting that quality of audit is important, Zghal, Chtourou and Sellami (2011) report lower earnings management (based on discretionary accruals models) for French firms following IFRS adoption when the firms have Big 4 auditors.

Regulators have identified a key role for independent enforcement bodies in promoting the quality of financial reporting (SEC, 2002; CESR, 2003). The EU called for the establishment of independent enforcement bodies in EU member states to promote the quality of reporting under IFRS as part of the IFRS adoption announcement (EC1606/2002). The Committee of European Securities Regulators (CESR) and subsequently the European Securities Market Authorities (ESMA) have been active in promoting compliance with IFRS (CESR, 2009). Comparability of enforcement has

11

been promoted through the European Enforcers Sessions (EES), where national security market regulators can discuss and compare enforcement decisions (Berger, 2010). Another aspect of enforcement is the ability to seek redress through the legal system. FEE (2001) reports that courts can have an important role in protecting investors. Jackson and Roe (2009) document the resources consumed and outputs achieved by the Financial Services Authority (FSA), the UK market regulator, arguing the FSA is an active regulator. However, Armour (2008) concludes that other mechanisms (such as the activity of the FRRP and the media) are more important in the UK setting than formal enforcement through court actions. Consistent with this view, evidence suggests that use of the media (a name and shame approach) has been an effective way to promote compliance with accounting standards in the UK (Fearnley, Hines, McBride and Brandt, 2000).

Germany and the US are the most popular cross listing locations (Banalieva and Robertson, 2010). There have been increases in regulatory scrutiny that have coincided with the adoption of IFRS in both Germany and the US. In Germany, there were changes to the activities of public regulatory and the extent of scrutiny of financial reporting. In December 2004 the financial statement monitoring act

(Bilanzkontrollgesetz, BilKoG) was passed establishing the legal basis for creating a two tier external financial reporting enforcement structure. The first tier private sector institution Financial Reporting Enforcement Panel (Deutsche Prfstelle fr Rechnungslegung DPR e.V. or DPR) monitors compliance with financial reporting requirements, focusing on the consistent and faithful application of IFRS. The second tier, the Federal Financial Supervisory Authority (Bundesanstalt fr

Finanzdienstleistungsaufsicht or BaFin), can intervene as required,. for example, where firms refuse to cooperate with the DPR or there are serious concerns about the accuracy of an assessment made by DPR. BaFin is a government agency and has the executive power to direct firms to disclose the DPR findings (Hitz, Ernstberger and Stich, 2012). DPR began its operations in July 2005. During the 2005 to 2009 period, the DPR initiated and completed 507 examinations of which 116 cases were found to be erroneous (DPR, 2010). Once the DPR has established that there has been an error, firms are required to disclose the DPRs findings in a press release. Hitz et al. (2012) reveal there is significant negative investor reaction around the release of the DRPs findings, indicating the name and shame strategy is effective.

12

In addition, auditor oversight reforms were carried out in Germany that established the Auditor Oversight Commission (Abschlussprferaufsichtskommission or AOC). The AOC began operations in January 2005 with the intent of enhancing the oversight of auditors and ensuring quality assurance of auditors (AOC, 2010; Ernstberger, Stich and Volger, 2012). The AOC proactively carries out inspections and makes announcements on the areas it will focus on in the forthcoming year. Thus in Germany, following the adoption of IFRS, there have been significant changes in the reporting environment that could improve accounting quality through more regulatory scrutiny of financial reporting and greater oversight of the work of external auditors, promoting higher audit quality.

During the period of this study, there have also been improvements in the regulatory and enforcement regime in the US. In reaction to a number of high profile corporate collapses including Enron and WorldCom, the Sarbanes Oxley Act (SOX) was introduced in the US (Zingales, 2009). SOX also applies to foreign firms under the jurisdiction of the SEC in the US (i.e. foreign firms listed on the NYSE, AMEX and NASDAQ exchanges). Although it was signed into law on 30 July 2002, foreign firms only have to meet all the requirements of SOX by fiscal years ending on or after 15 July 2006 (Piotroski and Srinivasan, 2007). Thus for the 24 SEC registrants (NYSE, AMEX or NASDAQ listed firms) in this study, the impact of SOX changes takes effect in the post 2005 IFRS period.

SOX requires more stringent governance practices and greater disclosure. For example, it requires (i) the CEO and the Chief Financial Officer (CFO) certification of financial statements (which makes them liable for civil and criminal liabilities), (ii) management and the external auditor to report on the adequacy of the firm's internal control over financial reporting, (iii) the prohibition of loans to officers and directors, and (iv) the inspection of foreign auditors by the Public Company Accounting Oversight Board (PCAOB) (Zhang, 2007; Hart, 2009). A major development under SOX is that the CEO and the CFO can be subject to criminal proceedings if they knowingly sign off on misleading financial statements (Hart, 2009). In addition, SOX has expanded oversight of the work of auditors, increasing their motivation to carry out their role effectively. A possible outcome of these changes is to reduce managers propensity to engage in earnings management, because the costs of doing so have increased.

13

The quality of financial reporting under IFRS is of great importance to the SEC as it considers whether to adopt IFRS for US domestic companies. SEC staff provided a review of IFRS reporting by foreign listed companies (SEC, 2011), indicating the SEC has been closely monitoring IFRS reporting by cross listed firms. The effect of this additional regulatory scrutiny may be to improve accounting quality under IFRS, as cross listed firms act to protect their reputations.4

Lang et al. (2003) report that foreign firms (from 21 countries) cross listed in the US display less earnings management, a tendency for more timely recognition of bad news and stronger associations between accounting data and share price than firms that are not cross-listed in the US. They proposed that better accounting quality reflects the US legal environment and the monitoring and sanctioning activities of the SEC. Sun et al. (2011) also investigate a number of measures of earnings management for firms cross listed in the US. They find a lower frequency of small positive earnings post-IFRS for foreign listed firms from the UK and Australia. The authors note the finding is surprising, since these firms come from countries with strong legal enforcement. However, they find no impact of adoption of IFRS on timely loss recognition for foreign firms listed in the US.

Daske et al. (2008) provide early evidence of benefits of IFRS in a number of countries, notably in those with a stronger legal environment. Christensen et al. (2012) extend Daske et al. (2008) to show the improvements in liquidity associated with IFRS are limited to the five EU countries that introduced proactive enforcement around the time of adoption of IFRS (the UK, Germany, Finland, Netherlands and Norway). Consistent with this view, Ernstberger et al. (2012) conclude the additional enforcement activity in Germany has had some benefits, in terms of a decrease in earnings management, an increase in stock liquidity and to some extent an increase in market valuation for German firms after adoption of IFRS.

Attracting the attention of the SEC in relation to financial reporting quality is not seen as a positive event for firms. Firms required by the SEC to restate their financial statements experience negative capital market consequences (Dechow, Sloan and Sweeney, 1996). Most (105) of the UK firms listed in the US are OTC firms and as such do not have to meet SEC accounting requirements applicable to firms listed on the regulated exchanges (NYSE, AMEX, NASDAQ). All firms listed in the US are subject to other aspects of the US legal setting such as relatively greater litigation risk. We argue these elements are part of regulatory scrutiny and can act towards promoting accounting quality.

14

However, the incentives of cross listed firms to improve accounting quality by the introduction of stricter regulations can be mitigated by the fact that the regulators such as the SEC and BaFin may be less likely to prosecute foreign firms than domestic firms. In addition, shareholders may face obstacles enforcing US or German court rulings against UK firms. The increased enforcement risk associated with a cross listing may be exaggerated (Siegel, 2005). Nevertheless, we argue that UK firms with a foreign listing have an incentive to increase their accounting quality as they come under the scrutiny of not only the institutional oversight system in the UK (including the Financial Services Authority (FSA) and the FRRP) but also the enforcement mechanisms of foreign countries, specifically Germany and the US. Therefore we propose that accounting quality improves for firms that receive greater regulatory scrutiny. The hypothesis can be formally stated as:

H2 Improvements in accounting quality under IFRS are greater for firms subject to more regulatory scrutiny.

3. Method

3.1 Sample Selection

We study firms listed on the LSE for several reasons. First, the UK market is the largest capital market where IFRS is mandatory for domestic companies. Second, UK listed firms adopted IFRS for the preparation of consolidated accounts from 1 January 2005 and early voluntary adoption was not permitted. Therefore we include a large number of firms in our study, including economically and internationally important firms. By excluding early adopters, our tests are not affected by the early voluntary use of IFRS. Another important element to our selection is that these firms previously used UK GAAP, which were considered high quality standards with much in common with the principle-based standards of IFRS. In addition, the institutional setting for financial reporting (as discussed in the previous section) was arguably one of the strongest in the world both before and after adoption of IFRS. Because the UK setting was already strong, we can be more confident the any observed increases in accounting quality are due to the additional effects of enforcement associated with cross listing.

15

From the 1,610 listed firms (as at 31 October 2006) we include 495 (31%) (Table 1). We include only MAIN board firms and exclude foreign firms, firms with preference shares only, firms not reporting under UK GAAP and those not providing consolidated accounts. Another 508 firms are excluded because of missing data (163 were not listed for the full period and 345 have missing accounting data in Datastream. Firms were drawn from a range of industry sectors based on GICS: 138 financial firms (including six banks and other entities), 104 industrial, 93 consumer discretionary, 53 information technology, 37 materials, 26 consumer staples, 19 health care, 13 energy, seven utilities and five telecommunication services.

Data is collected from Datastream to prepare the accounting quality measures discussed below. We require at least three years of data under UK GAAP as well as an additional three years under IFRS. Observations for UK GAAP (n = 2,356) were pooled from five years (adoption years -4 to 0, which generally corresponds to fiscal years 2000 to 2004) and observations under IFRS (n = 1,823) were pooled for four years (adoption years 1 to 4, which generally corresponds to fiscal years 2005-2009).5 The number of cross listed firms (CROSS) is 246 (50%) and non-cross listed NONCROSS) is 249. For CROSS (NONCROSS), there are 1,178 (1,178) UK GAAP observations and 921 (902) IFRS observations.67

Sample firms are more likely to be listed in Germany and the US. Out of the 246 CROSS firms, 245 (99.59%) are listed in Germany. This may be due to geographic proximity, low listing costs and the Frankfurt Stock Exchange (Deutsche Brse) being the second largest capital market in Europe (after the LSE). Of the 246, 131 (53%) firms are also listed in the US: 105 are listed on the Over the Counter (OTC) market and 24 firms are listed on the NYSE, NASDAQ or AMEX markets. There are 10 (4%), nine (4%) and eight (3%) firms listed in Ireland, France and South Africa respectively. Four are listed in the Netherlands and Switzerland and three are listed in Australia.
Sample firms are most likely to have a 31 December financial year-end (44%) meaning transition to IFRS occurred on 31 December 2004. Next, 101 firms (20%) had March year ends, 7% have June and 7% have September. The year ends of other firms are spread throughout the years.
6 5

NONCROSS firms have fewer observation in the IFRS period due to missing data.

Ideally, we would match the CROSS and NONCROSS firms on size and industry to control of differences between firms. However matching is not possible due to our design choice to include all CROSS firms and the lack of NONCROSS firms with similar characteristics with which each firm could be matched. Nevertheless we control for firm attributes in the earnings management models, as discussed in the next section.

16

One firm is listed in each of Canada, Hong Kong, Japan, Luxembourg, Malaysia, Namibia and Sweden. Cross listing data is sourced from Datastream and checked against stock exchange websites and data from Reuters and the Bank of New York.8 A total of 105 (54%) firms are listed in one non-UK country, while 116 (40%) are listed in two countries. There are 21 (5%) firms listed in three countries and three (1%) firms listed in four countries. Rio Tinto Plc is the one firm that is listed in five countries.

3.2 Measuring accounting quality Prior studies use a number of measures to capture accounting quality based on earnings smoothing, managing towards earnings targets, timeliness of loss recognition and value relevance. Table 2 provides a list of the measures and empirical models used. The four sets of accounting quality measures are use to compare accounting quality under UKGAAP and under IFRS (H1) and to compare accounting quality of subsamples of CROSS and NONCROSS firms (H2).

Earnings smoothing The first set of measures relates to earnings smoothing. A higher variance in the residuals in Model 1.1 and 1.2 and a lower negative correlation of the residuals from the cash flows and the accruals model (Model 1.3) are taken as indicative of higher accounting quality (Lang et al., 2005; Barth et al., 2006; 2008, Paananen and Lin, 2009; Sun et al., 2011).9 The models used as described below.

In Model 1.1 (Table 2) we first determine the relationship of change in net income (NI) and a number of explanatory factors. NI is the change in net income available to ordinary shareholders at fiscal year end scaled by end of year total assets. The other variables are: SIZE is the natural logarithm of market value of equity in billions of pounds at the end of the financial year; GROWTH is annual percentage change in sales; EISSUE is annual percentage change in book value of equity at the end of the financial year; LEV is end of year total liabilities scaled by end of year book value of

According to Banalieva and Robertson (2010) the reasons for Germany being such a popular destination for cross listing include the German equity market being a major force within the EU, Germany offering the benefit of several regional stock exchanges such as the Frankfurt and Berlin Stock Exchanges, less stringent listing requirements compared to the US, and offering a useful platform for entering other European markets such as XETRA. Models 1.1-3.1 are calculated with industry fixed effects, to control for impact on the accounting quality measures of differences between industries.
9

17

equity; DISSUE is annual percentage change in total liabilities; TURN is annual sales scaled by end of year total assets; OCF is annual net cash flow from operating activities scaled by end of year total assets; AUD is an indicator variable set to one if the firms auditor is PricewaterhouseCoopers, KPMG, Ernst & Young or Deloitte Touche Tohmatsu and zero otherwise; NUMCROSS is the number of countries in which a firms stock is listed.

Next we test the cross sectional variation in the residuals from equation 1.1 (NI*) for each firm over the UK GAAP and IFRS periods. We test whether the difference in medians for variance of NI* is significantly different under IFRS and UK GAAP using Wilcoxon rank tests.10 A higher variability of NI* is indicative of less earnings smoothing and therefore better accounting quality. We repeat this procedure for the second earnings smoothing model (Model 1.2) which uses change in operating cash flow as the dependent variable (OCF) and other variables as listed above.

The third measure of earnings smoothing is the Spearman correlation between accruals (ACC) and cash flows (OCF) (Model 1.3). Accruals and cashflows generally have a negative correlation but a larger negative correlation is assumed to capture earnings smoothing as managers react to poor cash flows by increasing accruals or concealing better than expected performance by decreasing accruals (Land and Lang, 2002; Drake et al., 2009). First, the residuals from the equations in Model 1.3 (OCF* and ACC*) are calculated. (ACC is annual net income available to ordinary shareholders at financial year end less annual cash flow from operating activities, scaled by end of year total assets. OCF is defined above). Second, we compute a Spearman rank correlation test of the correlation of ACC* and OCF*. A weaker negative correlation is indicative of less earnings smoothing and thus better accounting quality.

Managing towards earnings targets The second measure of accounting quality is based on the extent to which managers use accounting discretion to avoid showing small losses (Burgstahler and Dichev, 1997; Degeorge, Patel and Zeckhauser, 1999). This measure examines a firms tendency to manage earnings towards targets, namely towards small positive net incomes. Following Burgstahler and Dichev (1997) small positive net income is when

10

Kolmogorov-Smirnov tests of the variance of NI* reject the null hypotheses of a normal distribution.

18

income scaled by total assets is between 0 and 0.01. First, Model 2.1 (Table 4) is fitted, where FIRM (0, 1) is an indicator variable given a value of one for observations under IFRS and zero for observations under UK GAAP. SPOS is indicator variable set to one for observations when annual net income scaled by total assets is between 0 and 0.01 and zero for other observations (Lang et al., 2003). A negative coefficient for SPOS indicates that there is a lower frequency of small positive net incomes under IFRS compared to UK GAAP, therefore better accounting quality.

Timely loss recognition The third accounting quality measure relates to whether firms recognise large losses as they occur, thereby providing investors with more value relevant information and incentivising managers to stem the losses faster. Following Lang et al. (2003), Lang et al. (2006) and Barth et al. (2008), large losses (LNEG) are observations where annual net income scaled by total assets is less than -0.2 and zero otherwise. A higher frequency of LNEG is indicative of, ceteris paribus, better accounting quality as it shows that managers are recognising large losses as they occur. A firms tendency to report large negative losses could be affected by a variety of factors unrelated to earnings management. Rather than comparing the frequency of LNEG directly between firms, Model 3.1 (Table 2) is used to calculate the coefficient of LNEG. A positive coefficient for LNEG indicates firms recognise large losses more frequently under IFRS compared to UK GAAP, and therefore have better accounting quality.

Value relevance The final set of accounting quality measures is based on value relevance models. They differ from the above models in that they consider market measures (prices and returns) as well as accounting numbers. Model 4.1 measures accounting quality by the association (adjusted R2) of stock prices and earnings and equity book value using a model derived from Ohlson (1995). A stronger association between stock prices and earnings suggest earnings better reflect a firms underlying economics and are therefore of higher quality (Ali and Hwang, 2000).

In Model 4.1, stock price (P) is regressed on industry fixed effects in order to control for mean differences in stock prices across industries. Second, the residuals from this regression (P*) are regressed on equity book value per share (BVEPS) and net income per share (NIPS). P is calculated three months after year-end to ensure that all relevant

19

accounting information is in the public domain. Differences in adjusted R2 are determined using Cramers test (1987) (Lang et al., 2003). A higher adjusted R2 indicates a closer association between earnings and stock prices, therefore greater usefulness of financial information to users. Thus, a higher adjusted R2 is indicative of better accounting quality.

The second value relevance model, Model 4.2, is based on the assumption that firms with higher accounting quality will show a higher association between earnings (net income per share scaled by price) and returns (Barth et al., 2008). In our study, Model 4.2 follows the Khan and Watts (2009) approach, which augments the Basu (1997) model to control for firm characteristics (size, leverage and market to book ratio). The Khan and Watts (2009) approach is to model NIPS/P (net income per share available to ordinary shareholders at fiscal year end scaled by share price at the beginning of the fiscal year) as the dependent variable and SIZE, LEV (defined above) and MB (end of year market value of equity scaled by end of year book value of equity) as the independent variables. The residuals from this equation (NIPS/P*) are used in Model 4.2 (Table 4), where DT is an indicator variable set to one if returns are negative and zero if returns are positive. The DT variable is included in the regression because differences in accounting quality are more prominent in bad news firms than good news firms because the latter have a lower incentive to engage in earnings management (Ball et al., 2000). To ensure that investor responses to previous years earnings are excluded, we calculate the twelve month stock return (RET) starting from nine months before fiscal year end and ending three months after fiscal year end (Easton and Harris, 1991). Model 4.3 (Table 4) investigates value relevance based on the adjusted R2 from a regression of returns on earnings. If investors find information provided in the annual reports useful, there should be a greater association between returns and earnings. First, returns (RET) are regressed on industry fixed effects in order to control for mean differences in returns across industries. Next, the residuals from this regression (RET*) are regressed against level of earnings (NIPS) and changes in earnings (NIPS) (Easton and Harris, 1991). NIPSit/Pit-1 is the change in net income per share available to common shareholders at fiscal year end scaled by beginning of year share price. Cramers test (1987) is used to compare adjusted R2 for UK firms under UK GAAP and IFRS. A higher adjusted R2 is indicative of better accounting quality.

20

4. Results 4.1 Descriptive statistics

Table 3 reports the mean and median values for variables used in the accounting quality models (Table 2). The observations under UK GAAP have been pooled over five years (adoption years -4 to 0) and the observations under IFRS have been pooled over four years (adoption years 1 to 4). Table 3 includes t-statistics for differences in means and z-scores for the Wilcoxon rank sum test for differences in medians. Focusing on median values, we observe significant changes in eight of the ten variables included in the models. We observe several variables were significantly lower in the IFRS period (NI, OCF and RET), reflecting the impact of the financial crises at the end of the IFRS period. (Untabulated data show marked decreases in years 3 and/or 4 of the IFRS period, i.e. 2008 and 2009). These variables show that change in net income and cash flows were more likely to be negative in the IFRS period and they were reflected in lower market returns. However, other variables increased in the IFRS period (ACC, PRICE, BVEPS and NIPS) although all except BVEPS were lower in the final year of the IFRS period (untabulated). OCF and NIPS/P were not significantly different between the periods. Considering the control variables, six were significantly larger (LEV, GROWTH, EISSUE, DISSUE, SIZE and MB) and TURN was significantly smaller in the IFRS period. These firm attributes suggest that firms were growing and making more debt and equity issues in the IFRS period. Values for EISSUE, DISSUE and MB were lower in the final IFRS year reflecting the effects of the financial crisis at the end of the IFRS period (untabulated).

Overall, the descriptive statistics in Table 3 reveal that significant changes have occurred in most of the test and control variables. While earnings measures such as NIPS show an increase in the post IFRS adoption period, cash flows (OCF) have not increased, indicating an increase in accruals (ACC). In addition, the volatility of earnings (NI) has increased but so has the volatility of cash flows (OCF). Market based measures such as share price (P), market value of equity (SIZE) and market to book ratio (MB) have increased. However, cumulative stock returns (RET) have declined significantly in the IFRS period compared to UK GAAP. The increases in the median values of the equity issue (EISSUE), debt issue (DISSUE) and growth (GROWTH) variables indicate that sample firms are expanding at a higher rate under

21

IFRS compared to under UK GAAP. Pearsons Chi-square tests reveal the proportions of the dichotomous variables (AUD, SPOS and LNEG) were not significantly different between the two time periods (untabulated).

Table 4 provides mean and medians for test and control variables and results of tests of differences. Focusing on medians, cross listed firms are larger (SIZE), with higher growth (GROWTH), higher market to book ratio (MB) and higher leverage (LEV). They are more likely to have issued debt (DISSUE) or equity (EISSUE) in the study period. Cross listed firms have higher share prices (PRICE) and book value of equity (BVEPS) but returns (RET) are not significantly higher. Accruals (ACC) are lower and operating cash flows (OCF) are higher than non-cross listed firms. Net income per share (NIPS/P) and change in net income (NIPS/S) are higher for cross listed firms but only NIPS/S is significantly greater.

4.2 Regression results

Table 5 presents the results for H1, which proposed that UK firms will have higher accounting quality under IFRS compared to UK GAAP. H1 is supported if the UK firms show (1) less earnings smoothing, (2) less managing towards earnings targets, (3) more timely loss recognition and (4) higher value relevance under IFRS compared to UK GAAP. We discuss the results for each set of accounting quality measures for the full sample of firms and for the subsamples of CROSS and NONCROSS firms. Earnings smoothing If accounting quality is higher under IFRS, the median variance of NI* should be higher under IFRS than UK GAAP. Table 5 (Model 1.1) shows median variance of NI* is higher under IFRS compared to UK GAAP (0.0013 and 0.0010). However, the difference is not significant. Similarly, if accounting quality is higher under IFRS, the median variance for NI*/OCF* should be higher under IFRS than UK GAAP. Table 5 (Model 1.2) shows the median variance of NI*/OCF* is higher under UK GAAP at 1.1565 compared to 1.0235 under IFRS (p < 0.10). This is not consistent with the prediction.

If accounting quality is higher under IFRS, there should be a smaller negative correlation between ACC* and OCF*. Table 5 (Model 1.3) shows the correlation

22

coefficient is -0.4910 under UK GAAP and -0.5410 under IFRS, which is not consistent with the prediction. Therefore, all three measures of earnings smoothing do not support H1. They do not provide evidence that, overall, UK firms reduced their level of earnings smoothing in the IFRS period.

Further insights can be obtained from the subsample of CROSS and NONCROSS firms. Table 6 (Model 1.1) shows CROSS firms experience an increase in the variance of NI*, from 0.0012 under UK GAAP to 0.0014 under IFRS (not significant). NONCROSS firms do not show an increase in variance of NI*. They show a significant decrease in their variance from 0.0011 under UK GAAP to 0.0010 under IFRS (p < 0.10). In addition, Table 7 (Model 1.1) reveals while there is no significant difference between CROSS and NONCROSS firms under UK GAAP, the CROSS firms display a significantly higher variance in NI* of 0.0014 compared to 0.0010 for the NONCROSS firms (p < 0.05).

The second earnings smoothing measure considers the variance of NI*/OCF*. The results in Table 6 (Model 1.2) show CROSS firms have a significantly higher variance under IFRS at 1.2300 compared to 1.2069 under UK GAAP (p < 0.10). However, the NONCROSS firms do not show an increase in their variance of NI*/OCF* under IFRS. Similar to the first measure, Table 7 (Model 1.2) shows there is no significant difference between CROSS and NONCROSS firms under UK GAAP while the CROSS firms display a significantly higher variance for NI*/OCF* of 1.2300 compared to 0.6993 for the NONCROSS firms (p < 0.01) under IFRS.

For the third measure of earnings smoothing (the correlation between accruals ACC* and operating cash flows OCF*), a larger negative correlation is indicative of earning smoothing and lower accounting quality. Table 6 (Model 1.3) shows the correlation coefficients for both the subgroups of CROSS and NONCROSS firms are not significantly different between the two time periods. However, the CROSS firms had a significantly smaller negative correlation compared to the NONCROSS firms under UK GAAP (-0.4510 and -0.5430, p < 0.01) and continued to do so under IFRS (-0.4900 and -0.5790, p < 0.01) (Table 7, Model 1.3).

Managing towards earnings targets

23

If firms are less likely to manage towards earnings targets under IFRS, the coefficient for SPOS should be negative. Results are as predicted. Table 5 (Model 2.1) shows the coefficient for SPOS is negative and significant (-0.2270 p < 0.01). The result is consistent with Chen et al. (2010) for UK firms. In the subsample analysis, we see that this result is attributable to the CROSS firms. The coefficient for SPOS for the CROSS firms is negative and significant (-0.5360, p < 0.01). This reflects a lower frequency of SPOS under IFRS compared to UK GAAP for CROSS firms (Table 6, Model 2.1). Table 6 (Model 2.1) shows no significant changes in the frequency of SPOS under UK GAAP and IFRS for NONCROSS firms. Table 7 (Model 2.1) shows that while there was no significant difference in the frequency of SPOS under UK GAAP between the two categories of firms, CROSS firms have a significantly lower frequency of SPOS under IFRS (-0.7200, p < 0.01).

Timeliness of loss recognition A positive coefficient for LNEG indicates that firms recognise large losses more frequently under IFRS. Results in Table 5 (Model 3.1) do not support the prediction. The coefficient for LNEG is negative (-0.0080) and not significant. However, Table 6 (Model 3.1) shows the coefficient for LNEG for CROSS firms is positive and significant (0.3750, p < 0.10), indicating more timely recognition of losses by the CROSS firms under IFRS compared to UK GAAP. The coefficient for LNEG for NONCROSS firms is not significant. They do not show a significant difference in the frequency of LNEG between the two time periods. Table 7 (Model 3.1) shows CROSS firms have a significantly higher frequency of LNEG under IFRS compared to NONCROSS firms (1.364, p < 0.01) but there was no significant difference in the frequency of LNEG between the two groups under UK GAAP.

Value relevance The final set of accounting quality measures investigates value relevance under IFRS and UK GAAP. A higher adjusted R2 suggests a closer relationship between accounting amounts and market measures and therefore better accounting quality. Table 5 shows the adjusted R2 for all models are greater under IFRS (based on Zscores for Cramers test). For Model 4.1 the adjusted R2 is 0.5349 under IFRS and 0.3576 under UK GAAP while in Model 4.2 the adjusted R2 under IFRS is 0.1177 and 0.0597 under UK GAAP. With Model 4.3, the adjusted R2 is 0.1673 under IFRS and 0.0654 under UK GAAP.

24

Similar results are observed for the CROSS and NONCROSS firms (Table 6). The adjusted R2 for Models 4.1, 4.2 and 4.3 for both the CROSS and NONCROSS groups are higher under IFRS compared to UKGAAP. In addition, Table 7 shows NONCROSS firms had higher adjusted R2 under UK GAAP (p < 0.01, for all models). For Model 4.1 the adjusted R2 is 0.3046 for CROSS and 0.4219 for NONCROSS; in Model 4.2 the corresponding adjusted R2 is 0.0484 and 0.0613; and for Model 4.3 adjusted R2 for CROSS is 0.0582 and 0.0741 for NONCROSS. A similar pattern is observed under IFRS. Table 7 shows NONCROSS firms had higher adjusted R2 under IFRS (p < 0.01, for all models). For Model 4.1 the adjusted R2 is 0.4710 and 0.5361 for CROSS and NONCROSS respectively; in Model 4.2 the adjusted R2 are 0.0996 and 0.1400 and for Model 4.3 the adjusted R2 is 0.1534 for CROSS and 0.1786 for NONCROSS.

In summary we find support for H2, based on accounting quality measures relating to earnings smoothing, managing towards earnings targets, and timeliness of loss recognition. Our results are consistent with Sun et al. (2011) who report less managing to earnings targets by UK firms cross-listed in the US under IFRS. We find a consistent pattern of an improvement in accounting quality for the subsample of CROSS firms but not for NONCROSS firms. We show the two groups of firms were not significantly different on the accounting quality measures under UK GAAP but under IFRS the CROSS firms have better accounting quality based on several measures. This supports the view that stronger (more effective) enforcement of requirements of accounting standards by the local regulators and auditors is related to improvements in accounting quality.

Among the accounting quality measures, the first three sets of measures are based on accounting information and firm attributes. The fourth set of measures, the value relevance tests, use the relationship of accounting numbers and market measures (prices and returns). The value relevance tests provide strong support for benefits from IFRS for UK firms including both the CROSS and NONCROSS firms. The result of greater value relevance for UK firms is consistent with Devalle et al.s (2010) and Aubert and Grudnitskis (2010) findings for UK firms.

4.3 Further investigation

25

The increase in value relevance in the IFRS period, for both CROSS and NONCROSS firms is perhaps surprising given that it was not obvious that financial reporting relevance would increase as a result of the change from UK GAAP to IFRS. We investigated possible explanations for the increase in value relevance, relating to recognition and measurement of financial assets, financial liabilities, intangible assets and disclosure via segment reporting. We created two sub-samples of companies based on having greater than or less than the median value for (a) financial assets, (b) financial liabilities11 (c) intangible assets and (d) number of segments. Our proposition was that firms with larger amounts amount of financial assets, financial liabilities and intangible assets were making greater use of fair value measurement and providing more disclosure. For segment reporting we expected a larger number of segments represented more disclosure. We tested whether the accounting quality measures were greater for firms in the above-the-median group, indicating the usefulness of IFRS measurement and disclosure in these specific areas.

For financial assets two accounting quality measures were better for firms with more financial assets (Model 2.1 managing towards earnings targets; Model 4.3 value relevance). For financial liabilities, only one measure was better for firms with more financial liabilities (Model 2.1 managing towards earnings targets). Similarly, for intangible assets, only one measure was better for firms with more intangible assets (Model 2.1 managing towards earnings targets). For segment reporting, value relevance was higher (Models 4.1, 4.2 and 4.3) for firms with fewer segments under IFRS. In summary, these tests did not provide evidence useful in identifying the areas of IFRS that are associated with improvements in value relevance under IFRS.

5. Conclusion

The adoption of IFRS aims to improve the quality of financial reporting by improving recognition, measurement and disclosure practices of companies throughout the world. IFRS has the potential to provide more transparent and comparable information,

Datastream did not provide an amount for financial assets or liabilities. We used the amount of derivative assets and liabilities at FYE 2007 to proxy for the magnitude of a firms financial assets and liabilities.

11

26

leading to more efficient capital markets and more cross border investment and economic growth (IFRS, 2012).

At the time of adoption of IFRS, many jurisdictions introduced new enforcement mechanisms (Christensen et al., 2012; Brown et al., 2012). Although research shows institutional setting is important for realising benefits of IFRS, the impact of changes in oversight of auditors and enforcement of financial reporting requirements by independent enforcement bodies is not often specifically addressed. We identify changes in the oversight of auditors that may strengthen the motivation and ability of auditors to ensure compliance with IFRS. We also describe the independent review of IFRS financial statements and associated sanctioning for non-compliance that may motivate firms to comply with IFRS and to avoid earnings management practices.

We predict that after adoption of IFRS cross listed firms will be under greater scrutiny because they fall under regulation in more than one jurisdiction where enforcement activities increased in the post 2005 mandatory IFRS period. Consistent with our prediction, we find accounting quality measures improve for cross listed UK firms after they adopt IFRS. A corresponding improvement is not observed for non cross listed firms, after controlling for firm differences in the calculation of accounting quality measures.

Our study extends the literature by focusing on enforcement of financial reporting requirements, considered an important aspect of the application of IFRS (IASB, 2012; SEC, 2011). We add to studies of accounting quality of cross listed firms (Lang et al., 2003; Sun et al., 2011) because we investigate the impact of changes in enforcement that occurred in the IFRS adoption period. The findings add to studies of the effects of heightened enforcement for German firms (Hitz et al., 2012; Ernstberger et al., 2012). We also add support to Jackson and Roes (2009) claim that public enforcement activity has important outcomes for capital markets. Our findings provide evidence to support the call for policy initiatives to promote a global infrastructure to support adoption of IFRS (Hegarty et al., 2004; IFRS, 2012) that includes effective auditing and greater financial reporting scrutiny via public enforcement bodies.

27

References ABSCHLUSSPRFERAUFSICHTSKOMMISSION (AOC) 2010, Auditor Oversight Commission issues comment letter to the EU-Green Paper on Audit Policy and publishes its work plan for 2011. Available from: <http://www.apakaoc.de/english/pdf/APAK-AOC-Press_Release_EU_Green_Letter_Work_Plan_2011.pdf>. [20 September 2012]. Ahmed, A., Neel, M. and Wang. E. 2012, Does Mandatory Adoption of IFRS Improve Accounting Quality? Preliminary Evidence. Contemporary Accounting Research, Forthcoming. Ali, A. and Hwang, L. 2000, Country-Specific Factors Related to Financial Reporting and the Value Relevance of Accounting Data. Journal of Accounting Research, vol. 38, no. 1, pp. 1-21. Amir, E., Harris, T. and Venuti, E. 1993, A Comparison of the Value-Relevance of U.S. Versus Non-U.S. GAAP Accounting Measures Using Form 20-F Reconciliations. Journal of Accounting Research, vol. 31, no. Supplement, pp. 230-264. Armour, J. 2008, Enforcement Strategies in UK Corporate Governance: A Roadmap and Empirical Assessment. European Corporate Governance Institute (ECGI) Working Paper Series in Law, no. 106/2008. Aubert, F. and Grudnitski, G. 2011, The Impact and Importance of Mandatory Adoption of International Financial Reporting Standards in Europe. Journal of International Financial Management and Accounting, vol. 22, no. 1, pp. 1-25. Audit Inspection Unit (AIU) 2005, 2004/5 Audit Quality Inspections - Public Report, June 2005. Aussenegg, W., Inwinkl, P. and Schneider, G. 2008, Earnings Management and Local vs International Accounting Standards of European Public Firms, Available from: SSRN. [21 September 2012]. Bae, K., Tan. H. and Welker. M. 2008, International GAAP Differences: The Impact on Foreign Analysts. The Accounting Review, vol. 83, no. 3, pp. 593-628. Baker, R. C., Mikol, A., & Quick, R. (2001). Regulation of the statutory auditor in the European Union: a comparative survey of the United Kingdom, France and Germany. European Accounting Review, 10(4), 763-786. doi: 10.1080/09638180120069124

28

Ball, R. 2006, International Financial Reporting Standards (IFRS): pros and cons for investors. Accounting and Business Research, vol. 36, no. Supplement 1, pp. 527. Ball, R. and Shivakumar, L. 2005, Earnings quality in UK private firms: comparative loss recognition timeliness. Journal of Accounting and Economics, vol. 39, no. 1, pp. 83128. Ball, R., Kothari, S. and Robin, A. 2000, The effect of international institutional factors on properties of accounting earnings. Journal of Accounting and Economics, vol. 29, no. 1, pp. 1-5. Ball, R., Robin, A. and Wu, J. 2003, Incentives versus standards: properties of accounting income in four East Asian countries. Journal of Accounting and Economics, vol. 36, no. 1-3, pp. 235-270. Banalieva, E. and Robertson, C. 2010, Performance, diversity, and multiplicity of foreign cross-listing portfolios. International Business Review, vol. 19, no. 6, pp. 531-547. Barth, M., Landsman, W. and Lang, M. 2008, International Accounting Standards and Accounting Quality. Journal of Accounting Research, vol. 46, no. 3, pp. 467498. Bartov, E., Goldberg, S. and Kim, M. 2005, Comparative Value Relevance Among German, U.S. and International Accounting Standards: A German Stock Market Perspective. Journal of Accounting, Auditing and Finance, vol. 20, no. 2, pp. 95-119. Basu, S. 1997, The conservatism principle and the asymmetric timeliness of earnings. Journal of Accounting and Economics, vol. 24, no. 1, pp. 3-37. Beekes, W. and Brown, P. 2006, Do Better-Governed Australian Firms Make More Informative Disclosures? Journal of Business Finance and Accounting, vol. 33, no. 3/4, pp. 422-450. Beidleman, C. 1973, Income Smoothing: The Role of Management. The Accounting Review, vol. 48, no. 4, pp. 653-667. Berger, A. (2010). The development and status of enforcement in the European Union. Accounting in Europe, 7(1), 15-35. Beuselinck, C., Joos, P., Khurana, I., & Van der Meulen, S. (2009). Mandatory IFRS Reporting and Stock Price Informativeness. SSRN eLibrary.

29

Brown, P., Preiato, J., & Tarca, A. (2012). Measuring country differences in enforcement of accounting standards: An audit and enforcement proxy. Working paper, University of Western Australia. Brggemann, U., Hitz, J.-M., & Sellhorn, T. (2012). Intended and Unintended Consequences of Mandatory IFRS Adoption: Review of Extant Evidence and Suggestions for Future Research. European Accounting Review, forthcoming. Brown, P. 2011, International Financial Reporting Standards: what are the benefits? Accounting and Business Research, vol. 41, no. 3, pp. 269-285. Burgstahler, D. and Dichev, I. 1997, Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics, vol. 24, no. 1, pp. 99-126. Bushman, R. and Piotroski, J. 2006, Financial Reporting Incentives for Conservative Accounting: The Influence of Legal and Political Institutions. Journal of Accounting and Economics, vol. 42, no. 1-2, pp. 107-148. Cairns, D., Massoudi, D., Taplin, R. and Tarca, A. 2011, IFRS fair value measurement and accounting policy choice in the United Kingdom and Australia. The British Accounting Review, vol. 43, no. 1, pp. 1-21. Cearns, K. and Ferran, E. 2008, Non-Enforcement Led Public Oversight of Financial and Corporate Governance Disclosures and of Auditors. Available from: SSRN. [21 September 2012]. Chen, H., Tang, Q., Jiang, Y. and Lin, Z. 2010, The Role of International Financial Reporting Standards in Accounting Quality: Evidence from the European Union. Journal of International Financial Management and Accounting, vol. 21, no. 3, pp. 220-278. Christensen, H. B., Hail, L., & Leuz, C. (2012). Mandatory IFRS Reporting and Changes in Enforcement. SSRN eLibrary. doi: 10.2139/ssrn.2017160 Coffee, J. 2002, Racing towards the Top?: The Impact of Cross-Listings and Stock Market Competition on International Corporate Governance. Columbia Law Review, vol. 102, no. 7, pp. 1757-1831. Committee of European Securities Regulators (CESR) 2003, Standard No. 1 on financial information: Enforcement of standards on financial information in Europe, CESR/03-073. Committee, European, Securities, Regulators, T. C. o. E. S., & CESR. (2009). CESR re-assesses application of its Standard No. 1 on Financial Information in Europe. Paris, France.

30

Cramer, J. 1987, Mean and variance of R2 in small and moderate samples. Journal of Econometrics, vol. 35, no. 2-3, pp. 253-266. Daske, H., Hail, L., Leuz, C. and Verdi, R. 2008, Mandatory IFRS Reporting around the World: Early Evidence on the Economic Consequences. Journal of Accounting Research, vol. 46, no. 5, pp. 1085-1142. Dechow, P. M., Sloan, R. G., & Sweeney, A. P. (1996). Causes and Consequences of Earnings Manipulation: An Analysis of Firms Subject to Enforcement Actions by the SEC. Contemporary Accounting Research, 13(1), 1-36. Degeorge, F., Patel, J., & Zeckhauser, R. (1999). Earnings management to exceed thresholds. Journal of Business, 72(1), 1-33. DEUTSCHE PRFSTELLE FR RECHNUNGSLEGUNG (DPR) 2010, Annual Activity Report 2009, Herbert Meyer (ed.), Berlin. Devalle, A., Onali, E. and Magarini, R. 2010, Assessing the Value Relevance of Accounting Data After the Introduction of IFRS in Europe. Journal of International Financial Management and Accounting, vol. 21, no. 2, pp. 85-119. Easton, P. and Harris, T. 1991, Earnings as an explanatory variable for returns. Journal of Accounting Research, vol. 29, no. 4, pp. 19-36. Ernstberger, J., Stich, M. and Vogler, O. 2012, Economic Consequences of Accounting Enforcement Reforms: The Case of Germany. European Accounting Review, vol. 21, no. 2, pp. 217-251. Fearnley, S. and Beattie, V. 2004, The Reform of the UKs Auditor Independence Framework after the Enron Collapse: An Example of Evidence-based Policy Making. International Journal of Auditing, vol. 8, no. 2, pp. 117-138. Fearnley, S. and Hines, T. 2003, The regulatory framework for financial reporting and auditing in the United Kingdom: the present position and impending changes. The International Journal of Accounting, vol. 38, no.2, pp. 215-233. Fearnley, S., Hines, T., Mcbride, K. and Brandt, R. 2000, A peculiarly British Institution: An analysis of the contribution made by the Financial Reporting Review Panel to accounting compliance in the UK, Center for Business Performance of the Institute of Chartered Accountants in England and Wales, London. Fdration des Experts Comptables Europens (FEE) 2000, The auditors report in Europe. Available from: <http://www.fee.be/fileupload/upload/Auditor%20s%20Report%20in%20Europ e%2000061632005101241.pdf>. [22 September 2012].

31

Fdration des Experts Comptables Europens (FEE) 2001, Enforcement mechanisms in Europe : a preliminary investigation of oversight systems. Available from: <http://www.fee.be/fileupload/upload/Enforcement%20Mechanisms%20in%20E urope153200520952.pdf>. [22 September 2012]. Fdration des Experts Comptables Europens (FEE) 2002, Discussion paper on enforcement of IFRS within Europe. Available from: <http://www.fee.be/fileupload/upload/DP%20on%20Enforcement%20of%20IFR S%20within%20Europe%20020416320058118.pdf>. [22 September 2012]. Fdration des Experts Comptables Europens (FEE) 2003, Discussion paper: European enforcement coordination. Available from: <http://www.fee.be/fileupload/upload/DP%20European%20Enforcement%20Co ordination%200311153200553111.pdf>. [22 September 2012]. Financial Reporting Council (FRC) 2010, Key Facts and Trends in the Accountancy Profession. Available from: <http://www.frc.org.uk/getattachment/cb6ae423f1ed-4e0b-8cee-c661d54bf7b5/Key-Facts-and-Trends-in-the-AccountancyProfession-28-June-2010.aspx>. [23 September 2012]. Financial Reporting Review Panel (FRRP) 2004, Financial Reporting Review Panel Announces 2005 Risk-based Proactive Programme, 21 December, FRRP Press Notice: 81. Financial Reporting Review Panel (FRRP) 2005, Financial Reporting Review Panel Report on Pro-Activity: 1 January 2004 to 31 March 2005. Financial Reporting Council, London. Florou, A., & Pope, P. F. (2009). Mandatory IFRS Adoption and Investor Asset Allocation Decisions. SSRN eLibrary. Francis, J. and Schipper, K. 1999, Have Financial Statements Lost Their Relevance. Journal of Accounting Research, vol. 37, no. 2, pp. 319-352. Hail, L., & Leuz, C. (2007). Capital Market Effects of Mandatory IFRS Reporting in the EU: Empirical Evidence. SSRN eLibrary. Harris, M. and Muller, K. 1999, The market valuation of IAS versus US-GAAP accounting measures using Form 20-F reconciliations. Journal of Accounting and Economics, vol. 26, no. 1-3, pp. 285-312. Hart, O. 2009, Regulation and Sarbanes-Oxley. Journal of Accounting Research, vol. 47, no. 2, pp. 437-445.

32

Hegarty, J., Gielen, F., & Farros, A. (2004). The implementation of international accounting and auditing standards: Lessons learned from the World Bank's Accounting and Auditing ROSC Program: The World Bank. Hines, T., McBride, K. and Page, M. 1999, An investigation into the stock market reaction to press notices issued by the Financial Reporting Review Panel. Journal of Applied Accounting Research, vol. 5, no. 2, pp. 54-82. Hines, T., McBride, K., Fearnley, S. and Brandt, R. 2001, Were off to see the wizard: An evaluation of directors and auditors experiences with the Financial Reporting Review Panel. Accounting, Auditing and Accountability Journal, vol. 14, no. 1, pp. 53-84. Hitz, J.-M., Ernstberger, J. and Stich, M. 2012, Enforcement of Accounting Standards in Europe: Capital-Market-Based Evidence for the Two-Tier Mechanism in Germany. European Accounting Review, vol. 21, no. 2, pp. 253-281. Hung, M and Subramanyam, K. 2007, Financial Statement Effects of Adopting International Accounting Standards: The Case of Germany. Review of Accounting Studies, vol. 12, no. 4, pp. 623-657. Hung, M. 2000. Accounting standards and value relevance of financial statements: An international analysis. Journal of Accounting and Economics, vol. 30, no. 3, pp. 401-420. Iatridis, G. 2010, International Financial Reporting Standards and the quality of financial statement information. International Review of Financial Analysis, vol. 19, no. 3, pp. 193-204. Iatridis, G. 2012, Hedging and earnings management in the light of IFRS implementation: Evidence from the UK stock market. The British Accounting Review, Forthcoming. International Accounting Standards Board (IASB) 2010, The Conceptual Framework for Financial Reporting. IFRS Foundation, London. International Federation of Accountants (IFAC) 2011, Member Body Compliance Program. www.ifac.org. Jackson, H. and Roe, M. 2009, Public and private enforcement of securities laws: Resource-based evidence. Journal of Financial Economics, vol. 93, no.2, pp. 207-238. Jeanjean, T and Stolowy, H. 2008, Do accounting standards matter? An exploratory analysis of earnings management before and after IFRS adoption. Journal of Accounting and Public Policy, vol. 27, no. 6, pp.480-494.

33

Jensen, M. and Meckling, W. 1976, Theory of the firm: Managerial behavior, agency costs and ownership structure. Journal of Financial Economics, vol. 3,no. 4, pp. 305-360. Kaufmann, D., Kraay, A., & Mastruzzi, M. (2010). The Worldwide Governance Indicators Project: The World Bank. www.govindicators.org. Khan, M. and Watts, R. 2009, Estimation and empirical properties of a firm-year measure of accounting conservatism. Journal of Accounting and Economics, vol. 48, no. 2-3, pp. 132-150. KPMG 2003, Implementing IAS Extract from: IAS compared with US GAAP and UK GAAP. KPMG International, Amstelveen. La Porta, R., Lopez-De-Silanes, F., Shleifer, A. and Vishny, R. 1998, Law and Finance. Journal of Political Economy, vol. 106, no. 6, pp. 1113-1155. Landsman, W, Maydew, E. and Thornock, J. 2012, The information content of annual earnings announcements and mandatory adoption of IFRS. Journal of Accounting and Economics, vol. 53, no. 1-2, pp. 34-54. Lang, M., Raedy, J. and Wilson, W. 2006, Earnings management and cross listing: Are reconciled earnings comparable to US earnings? Journal of Accounting and Economics, vol. 42, no. 1-2, pp. 255-283. Lang, M., Raedy, J. and Yetman, M. 2003, How Representative Are Firms That Are Cross-Listed in the United States? An Analysis of Accounting Quality. Journal of Accounting Research, vol. 41, no. 2, pp. 363-386. Leuz, C., Nanda, D. and Wysocki, P. 2003, Earnings Management and Investor Protection: an International Comparison. Journal of Financial Economics, vol. 69, no.3, pp. 505-527. Mcconnell, D. and Banks, G. 2003, How Sarbanes-Oxley Will Change the Audit Process. Journal of Accountancy, vol. 196, no. 3, pp. 49-55. Nobes, C. (ed.) 2001, GAAP 2001: A survey of national accounting rules benchmarked against International Accounting Standards, Anderson, BDO, Deloitte Touche Tohmatsu, Ernst and Young, Grant Thorton, KPMG, PricewaterhouseCoopers. Ohlson, J. 1995, Earnings, Book Values and Dividends in Equity Valuation. Contemporary Accounting Research, vol. 11, no. 2, pp. 661687. Paananen, M. and Lin, H. 2009, The Development of Accounting Quality of IAS and IFRS over Time: The Case of Germany. Journal of International Accounting Research, vol. 8, no. 1, pp.31-55.

34

Piotroski, J. and Srinivasan, S. 2007, The Sarbanes-Oxley Act and the Flow of International Listings. Available from: SSRN. [21 September 2012]. PriceWaterhouseCoopers (PWC) 2005, PwC inform IFRS/UK main differences indicator, McCann, H. (ed.), PricewaterhouseCoopers LLP. Regulation (EC) No 1606/2002 of the European Parliament and of the Council of 19 July 2002 on the application of International Accounting Standards. Schipper, K. (2005). The introduction of international accounting standards in Europe: Implications for international convergence. European Accounting Review, 14(1), 101-126. Securities and Exchange Commission (SEC) 2002, SEC Concept Release on International Accounting Standards. Available at http://www.sec.gov.rules/concept.shtml. Securities and Exchange Commission (SEC) 2011, Work plan for the consideration of incorporating International Financial Reporting Standards into the Financial Reporting System for U.S. Issuers: An analysis of IFRS in Practice. November. www.sec.gov. (November 111611). Seetharaman, A., Gul, F. and Lynn, S. 2002, Litigation risk and audit fees: evidence from UK firms cross-listed on US markets. Journal of Accounting and Economics, vol. 33, no. 1, pp. 91-115 Siegel, J. 2005, Can foreign firms bond themselves effectively by renting U.S. securities laws? Journal of Financial Economics, vol. 75, no. 2, pp. 319359. Sun, J., Cahan, S. and Emanuel, D. 2011, How would the mandatory adoption of IFRS affect the earnings quality of U.S. firms? Evidence from cross-listed firms in the U.S. Accounting Horizons, vol. 25, no. 4, pp. 837-860. Van Tendeloo, B. and Vanstraelen, A. 2005, Earnings management under German GAAP versus IFRS. European Accounting Review, vol. 14, no. 1, pp. 155-180. Verriest, A., Gaeremynck, A. and Thornton, D. 2012, The Impact of Corporate Governance on IFRS Adoption Choices. European Accounting Review, Forthcoming. Watts, R. and Zimmerman, J. 1986, Positive accounting theory, Prentice-Hall, Englewood Cliffs. Whittington, G. 2005, The Adoption of International Accounting Standards in the European Union. European Accounting Review, vol. 14, no. 1, pp. 127-153.

35

Zghal, D., Chtourou, S. and Sellami, Y. 2011, An analysis of the effect of mandatory adoption of IAS/IFRS on earnings management. Journal of International Accounting, Auditing and Taxation, vol. 20, no. 2, pp. 61-72. Zhang, I. 2007, Economic consequences of the Sarbanes-Oxley Act of 2002. Journal of Accounting and Economics, vol. 44, no. 1-2, pp. 74-115.

36

Table 1 Sample of UK firms listed on the London Stock Exchange

Number of firms Total number of firms listed as at the 31 October 2006 Exclusions: Foreign firms Firms which only listed preference shares Late listings Non-consolidated accounts Firms which are not covered by Datastream or insufficient data Firms which reported under US GAAP Total firms available for analysis 316 134 163 154 345 3 495 20% 8% 10% 10% 21% 0% 31% 1,610 100%

This table shows 1,610 firms were listed on the Main Board of the LSE at 31 October 2006. Firms were removed from the sample if they were of foreign origin, had only preference shares, did not provide consolidated accounts or did not have sufficient data for all the tests to be carried out. Firms listed on the LSE on or after 31 December 2003 did not have sufficient data and were deleted from the sample as late listings.

37

Table 2 Measures of accounting quality

Accounting quality measure 1. Earning Smoothing 1.1. Variance of residuals from the NI model NIit = 0 + 1SIZEit + 2GROWTHit + 3EISSUEit + 4LEVit + 5DISSUEit + 6TURNit + 7OCFit + 8AUDit + 9NUMCROSSit + it 1.2. Variance of residuals from the NI model scaled by the variance of residuals from the OCF model 1.3. Correlation between the residuals of the OCF and ACCRUALS models OCFit = 0 + 1SIZEit + 2GROWTHit + 3EISSUEit +4LEVit + 5DISSUEit+ 6TURNit+ 7 OCFit + 8AUDit + 9NUMCROSSit + it OCFit = 0 + 1SIZEit + 2GROWTHit + 3EISSUEit + 4LEVit + 5DISSUEit + 6TURNit + 7AUDit + 8NUMCROSSit + it ACCit= 0 + 1SIZEit + 2GROWTHit + 3EISSUEit + 4LEVit + 5DISSUEit + 6TURNit + 7AUDit + 8NUMCROSSit + it 2. Managing towards earnings targets 2.1. Frequency of small positive net income (SPOS) FIRM (0,1)it = 0 + 1SPOSit + 2SIZEit + 3GROWTHit + 4EISSUEit + 5LEVit + 6DISSUEit + 7TURNit + 8OCFit + 9AUDit + 10NUMCROSSit + it 3. Timeliness of loss recognition 3.1. Frequency of large negative net income (LNEG) FIRM (0,1)it = 0 + 1LNEGit + 2SIZEit + 3GROWTHit + 4EISSUEit + 5LEVit + 6DISSUEit + 7TURNit + 8OCFit + 9AUDit + 10NUMCROSSit + it 4. Value relevance Adjusted R2 from a regression of : 4.1. Stock prices on equity book value and earnings 4.2. Earnings on stock returns 4.3. Stock returns on earnings P*it = 0+ 1BVEPSit + 2NIPSit + it [NIPSit / Pit-1] * = 0+ 1RETit + 2DTit + 3RETit* DTit + it RET*it = 0+ 1NIPSit/Pit-1 + 2NIPSit/Pit-1 + it Higher adjusted R2 Higher adjusted R2 Higher adjusted R2 Higher frequency of LNEG Lower frequency of SPOS Lower negative correlation Higher variance Higher variance Model Better Accounting quality

NI is the change in net income available to common shareholders at fiscal year end scaled by end of fiscal year total assets. SIZE is the market value of equity in billions of pounds as of the end of the fiscal year. GROWTH is annual percentage change in sales. EISSUE is annual percentage change in book value of equity. LEV is total liabilities divided by book value of equity at fiscal year end. DISSUE is annual percentage change in total liabilities. TURN is annual sales divided by end of year total assets. OCF is annual net cash

38

flow from operating activities scaled by end of year total assets. AUD is an indicator variable is set to one if the firms auditor is PricewaterhouseCoopers, KPMG, Ernst & Young or Deloitte Touche Tohmatsu and zero otherwise. NUMCROSS is the number of countries in which a firms stock is listed. OCF is change in annual net cash flow from operating activities scaled by end of fiscal year total assets. ACCRUALS is annual net income available to common shareholders at fiscal year end less annual cash flow from operating activities, scaled by end of year total assets. FIRM (0,1) is an indicator variable which is set to one for observations under IFRS and zero under UK GAAP. SPOS is indicator set to one for observations for which annual net income scaled by total assets is between 0 and 0.01 and set to zero otherwise. LNEG is an indicator variable set to one for observations for which annual net income scaled by total assets is less than -0.2 and zero otherwise. P* is the residuals from a regression where stock price three months after fiscal year end (PRICE) is regressed on industry fixed effects. BVEPS is book value of equity in pounds per ordinary share at fiscal year end. NIPS is net income available to common shareholders at fiscal year end per ordinary share. NIPS/P* is the residuals from the following regression: NIPSit / Pit-1 = 0+ 1SIZE it + 2LEV it + 3MB it + it, where NIPS/P is net income available to common shareholders per ordinary share at fiscal year end, scaled by beginning of year share price; MB is end of year market value of equity scaled by end of year book value of equity. RET is cumulative stock return computed over 12 months, starting from nine months before fiscal year end and ending three months after fiscal year end. DT is an indicator set to one for observations for which RET are negative and zero otherwise. RET* is the residuals from a regression where RET is regressed on industry fixed effects. NIPS/P is change in net income per share at fiscal year end scaled by beginning of year share price.

39

Table 3a

Descriptive statistics for test and control variables

Mean Median Min Max Std Dev

Dependent Variables NI OCF ACC OCF RET NIPS/P PRICE BVEPS NIPS NIPS/P Control Variables LEV GROWTH EISSUE DISSUE TURN SIZE MB 2.176 0.106 0.111 0.115 0.898 1.688 2.371 1.118 0.074 0.074 0.058 0.791 0.204 1.657 0.024 -0.416 -0.492 -0.622 0.017 0.006 0.470 19.409 0.978 1.251 1.468 2.957 23.026 10.643 3.604 0.253 0.313 0.375 0.771 4.409 2.141 -0.005 0.000 -0.037 0.069 -0.039 0.059 3.441 2.375 0.219 0.003 -0.001 -0.001 -0.027 0.064 0.043 0.056 2.210 1.350 0.113 0.006 -0.225 -0.167 -0.270 -0.129 -1.281 -0.279 0.160 0.083 -0.404 -0.423 0.181 0.174 0.127 0.261 0.764 0.487 16.540 13.984 1.557 0.387 0.067 0.064 0.077 0.080 0.445 0.126 3.626 2.938 0.368 0.129

This table presents descriptive statistics for the continuous variables for the 495 firms in the pooled sample. The total number of firm year observations is 4,179. All variables (defined in Table 2) have been winsorised at the 2.5 and 97.5 percentiles.

40

Table 3b

Univariate comparison of continuous test and control variables under UK GAAP and IFRS

Means
UKGAAP IFRS

Medians t- statistic
UKGAAP IFRS

Wilcoxon zstatistic 3.25 *** 2.00 ** 4.67 *** 0.50 11.04 *** 0.20 6.67 *** 4.50 *** 5.87 *** 4.26 ***

n = 2,356 Dependent Variables NI OCF ACC OCF RET NIPS/P PRICE BVEPS NIPS NIPS/P Control Variables LEV GROWTH EISSUE DISSUE TURN SIZE MB 2.066 0.096 0.093 0.084 0.917 14.662 2.328 -0.001 0.001 -0.042 0.069 0.029 0.064 2.955 2.221 0.184 0.001

n = 1,823 -0.010 -0.002 -0.032 0.068 -0.125 0.053 4.069 2.574 0.263 0.006 4.08 *** 1.56 -4.23 *** 0.33 11.14 *** 2.85 *** -9.57 *** -3.81 *** -6.71 *** -1.14

n = 2,356 0.000 0.000 -0.032 0.065 0.091 0.053 2.000 1.264 0.095 0.004

n = 1,823 -0.002 -0.002 -0.022 0.063 -0.035 0.058 2.620 1.485 0.137 0.009

2.319 0.119 0.134 0.156 0.873 19.747 2.426

-2.26 ** -2.93 *** -4.23 *** -6.03 *** 1.83 * -3.62 *** -1.46

1.084 0.058 0.053 0.044 0.828 1.769 1.551

1.178 0.091 0.108 0.084 0.755 2.622 1.819

2.82 *** 5.58 *** 7.32 *** 5.70 *** 2.01 ** 6.71 *** 4.93 ***

***significant at the p < 0.01 level (two-tailed). **significant at the p < 0.05 level (two-tailed). *significant at the p < 0.10 level (two-tailed). This table presents the means and medians of the continuous variables for the 495 firms in the final sample under UK GAAP and IFRS. Observations for UK GAAP have been pooled over five years (adoption years -4 to 0). Observations for IFRS have been pooled over four years (adoption years 1 to 4). T-statistics for the differences in means, z-scores for the Wilcoxon rank sum tests and p values (two-tailed) are also presented. All variables (defined in Table 2) have been winsorised at the 2.5 and 97.5 percentiles.

41

Table 4: Descriptive statistics for firms cross listed (CROSS) and not cross listed (NONCROSS) firms
Means
NONCROSS CROSS

n = 2,080 n = 2,099

Differences in means tp value statistic

Differences in medians NONCROSS CROSS Wilcoxon p value n = 2,080 n = 2,099 Z statistic

Medians

Test Variables NI -0.006 -0.004 -0.763 0.445 -0.001 -0.001 -0.701 0.483 OCF 0.000 -0.001 0.279 0.781 -0.001 -0.001 -0.347 0.729 ACC -0.035 -0.040 2.047 0.041 -0.020 -0.033 4.620 <0.001 OCF 0.064 0.074 -4.314 <0.000 0.053 0.072 -5.840 <0.001 01 RET -0.044 -0.033 -0.857 0.392 0.037 0.046 -1.131 0.258 NIPS/P 0.060 0.059 0.275 0.783 0.054 0.057 -0.744 0.457 PRICE 2.654 4.221 -14.331 <0.001 1.735 2.880 -14.330 <0.001 BVEPS 2.358 2.393 -0.386 0.699 1.268 1.438 -2.474 0.013 NIPS 0.163 0.273 -9.772 <0.001 0.082 0.157 -9.198 <0.001 NIPS/P 0.000 0.006 -1.427 0.154 0.004 0.007 -1.935 0.053 Control Variables LEV 1.669 2.679 -9.170 <0.001 0.894 1.369 -13.977 <0.001 GROWTH 0.098 0.115 -2.192 0.028 0.063 0.083 -3.212 0.001 EISSUE 0.096 0.126 -3.037 0.002 0.066 0.086 -3.455 0.001 DISSUE 0.104 0.126 -1.838 0.066 0.040 0.074 -4.018 <0.001 TURN 0.916 0.881 1.464 0.143 0.804 0.774 -1.213 0.225 SIZE 0.281 3.082 -42.708 <0.001 0.081 0.649 -36.081 <0.001 MB 1.970 2.769 -12.290 <0.001 1.267 2.078 -16.814 <0.001 This table presents the means and medians of the continuous variables for the 249 cross listed and 246 non-cross listed firms. Observations have been pooled over the sample period (adoption years -4 to 4). T-statistics for the differences in means, z scores for the Wilcoxon rank sum test and p values (two tailed) are also presented in this table. All variables have been winsorised at the 2.5 and 97.5 percentiles. rcentiles.

42

Table 5 Comparison of accounting quality for UK firms under UK GAAP and IFRS
Panel A:Earnings smoothing Prediction UKGAAP (n=495) 1.1 Variance of NI* 1.2 Variance of NI*/ OCF* UKGAAP < IFRS UKGAAP < IFRS 0.0010 1.1565 IFRS (n=495) 0.0013 1.0235 0.701 1.302* z- score

(n =2,356) (n=1,823) 1.3 Correlation OCF* and ACC* Panel B:Managing towards earnings targets UKGAAP < IFRS Prediction -0.4910 -0.5410 2.184**

Coefficient Wald- stat (n=4,179)

2.1 Frequency of SPOS Panel C:Timeliness of loss recognition

Negative Prediction

-0.2270

3.345**

Coefficient Wald- stat (n=4,179)

3.1 Frequency of LNEG Panel D:Value relevance R for association between: 4.1 P* and accounting numbers 4.2 NIPS/P* and RET 4.3 RET* and earnings
2

Positive Prediction UKGAAP

-0.0080 IFRS

0.002 z- score

(n =2,356) (n=1,823) UKGAAP < IFRS UKGAAP < IFRS UKGAAP < IFRS 0.3576 0.0597 0.0654 0.5349 92.156*** 0.1177 32.978*** 0.1673 68.455***

***significant at the p < 0.01 level (one-tailed). **significant at the p < 0.05 level (one-tailed). *significant at the p < 0.10 level (one-tailed).This table compares accounting quality measures under UK GAAP and IFRS for 495 firms. Sample firms reported under UK GAAP for five years (adoption years -4 to 0) and reported under IFRS for four years (adoption years 1 to 4). Models are presented and variables defined in Table 2.

43

Table 6

Comparison of accounting quality for under UK GAAP and IFRS for cross listed (CROSS) and non-cross listed firms (NONCROSS)

CROSS (N = 246)
Panel A: Earnings smoothing Prediction UK GAAP (n=246) 1.1 Variance of NI* 1.2 Variance of NI*/ OCF* UK GAAP<IFRS UK GAAP<IFRS 0.0012 1.2069 (n=1,178) 1.3 Correlation between OCF* and ACC* Panel B: Managing towards earnings targets UK GAAP<IFRS Prediction -0.4510 IFRS (n=246) 0.0014 1.2300 (n=921) -0.4900 Coefficient (n=2,099) 2.1 Frequency of SPOS Panel C: Timeliness of loss recognition Negative Prediction -0.5360 Coefficient (n=2,099) 3.1 Frequency of LNEG Panel D: Value relevance R for association between: 4.1 P* and accounting numbers 4.2 NIPS/P* and RET 4.3 RET* and earnings UK GAAP<IFRS UK GAAP<IFRS UK GAAP<IFRS
2

NONCROSS (N = 249)
Z- score Prediction UK GAAP (n=249) 0.770 1.329* no difference no difference 0.0011 1.0378 (n=1,178) 1.137 Wald- stat no difference Prediction -0.5430 IFRS (n=249) 0.0010 0.6993 (n=902) -0.5790 Coefficient (n=2,080) 7.418*** Wald- stat no difference Prediction -0.0140 Coefficient (n=2,080) 1.995* Z- score no difference Prediction UK GAAP (n=1,178) 44.861*** 12.575*** 31.520*** no difference no difference no difference 0.4193 0.0619 0.0676 -0.3870 IFRS (n=902) 0.5361 0.1400 0.1786 29.706### 21.942### 36.846### 1.894 Z- score 0.007 Wald- stat 1.186 Wald- stat 1.938# 1.712# Z- score

Positive Prediction UK GAAP (n=1,178) 0.3091 0.0559 0.0609

0.3750 IFRS (n=921) 0.4710 0.0996 0.1534

This table compares the accounting quality measures under UK GAAP and IFRS for cross listed (CROSS) and non-cross listed firms (NONCROSS). Firms reported under UK GAAP for five years (adoption years -4 to 0) and under IFRS for four years (adoption years 1 to 4). Models are presented and variables defined in Table 2. Significant differences in adjusted R2 for the value relevance measures in Panel D are tested using Cramers test (1987). ***significant at the p < 0.01 level (one-tailed). **significant at the p < 0.05 level (one-tailed). *significant at the p < 0.10 level (one-tailed). ###significant at the p < 0.01 level (two-tailed).

44

Table 7

Comparison of accounting quality for cross listed (CROSS) and non-cross listed firms (NONCROSS) under UK GAAP and IFRS
Prediction

IFRS UK GAAP CROSS NONCROSS z- score Prediction CROSS NONCROSS z- score (n=246) (n=249) (n=246) (n=249) CROSS>NONCROSS 0.0014 0.0010 1.566** no difference 0.0012 0.0011 0.200 1.1 Variance of NI* CROSS>NONCROSS 1.2300 0.6993 3.288*** no difference 1.2069 1.0378 0.871 1.2 Variance of NI*/ OCF* CROSS>NONCROSS (n=921) (n=902) (n=1,178) (n=1,178) 1.3 Correlation between OCF* and ACC* CROSS>NONCROSS -0.4900 -0.5790 -2.662*** no difference -0.4510 -0.5430 2.968### Panel B: Managing towards earnings targets Prediction Coefficient Wald- stat Prediction Coefficient Wald- stat (n=1,823) (n=2356) 2.1 Frequency of SPOS Negative -0.7200 7.357*** no difference -0.1530 0.601 Panel C: Timeliness of loss recognition Prediction Coefficient Wald- stat Prediction Coefficient Wald- stat (n=1,823) (n=2,356) 3.1 Frequency of LNEG Positive 1.3640 13.418*** no difference 0.0710 0.066 Panel D: Value relevance Prediction CROSS NONCROSS z- score Prediction CROSS NONCROSS z- score R2 for association between: (n=921) (n=902) (n=1,178) (n=1,178) 4.3 P* and accounting numbers CROSS>NONCROSS 0.4710 0.5361 14.795*** no difference 0.3046 0.4219 38.875### 4.1 NIPS/P* and RET CROSS>NONCROSS 0.0996 0.1400 10.194*** no difference 0.0484 0.0613 4.283### 4.2 RET* and earnings CROSS>NONCROSS 0.1534 0.1786 7.400*** no difference 0.0582 0.0741 6.372### This table compares the accounting quality measures under UK GAAP and IFRS for cross listed (CROSS) and non-cross listed firms (NONCROSS). Firms reported under UK GAAP for five years (adoption years -4 to 0) and under IFRS for four years (adoption years 1 to 4). Models are presented and variables defined in Table 2. Significant differences in adjusted R2 for the value relevance measures in Panel D are tested using Cramers test (1987). ***significant at the p < 0.01 level (one-tailed). **significant at the p < 0.05 level (one-tailed). *significant at the p < 0.10 level (one-tailed). ###significant at the p < 0.01 level (two-tailed). Panel A: Earnings smoothing

45

46