Determinants of discretionary fair value measurements: the case of Level 3 assets in the banking sector

The objective of our research was to respond to the call of Barth and Taylor ( 2010 ) for more research to examine the role of discretion in fair value estimates. Specifically, we investigate factors that explain banks’ accounting choices to use Level 3 valuation inputs from the fair value measurement hierarchy. Using hand‐collected data from a sample of international banks during 2009–2013, we find that incentives to use discretionary Level 3 valuation inputs, which can provide an opportunity to manage earnings, are associated with both firm‐level and country‐level determinants. Additional tests provide evidence that Level 3 ‘transfer‐in’ behaviour is related to changes in bank characteristics.     

Adjustment of valuation inputs and its effect on value relevance of fair value measurement

The fair value accounting standards; i.e., FAS 157, FAS 157-3 and FAS 157-4, specify the circumstances where firms need to adjust valuation inputs to fair value measurements in response to changes in market conditions. Such an adjustment inherently involves substantial management judgment and is accompanied with transfers of assets and liabilities among the different levels of the fair value hierarchy. We study the effect of adjusting valuation inputs to reflect market variations on value relevance of fair value measurements by comparing the value relevance of fair value assets between the banks that make transfers of assets and the banks that make no transfers. Overall, we find a significant increase in value relevance of fair value measurements for banks that transferred assets into/out of the Level 3 category. Our study examines a challenging situation in the application of fair value standards; i.e., determining fair value when there is a change in market conditions. Fair value measurement under such a situation involves substantial management judgment and potential estimate errors and manipulation. Our findings provide useful information for researchers, regulators and accounting professionals to assess the market’s perception of the reliability of fair value information when management exercises substantial discretion in adjusting valuation inputs under changing market conditions.     

Understanding accounting discretion in China: An analysis of fair value reporting for investment property

This study sheds light on the discretionary accounting practices in China, the largest emerging market in the world. In particular, we focus on whether Chinese firms use discretion in investment property fair values to manage reported performance. We examine whether firms’ ex ante needs for accounting discretion affect their decisions to adopt fair value reporting for investment property and the ex post performance management by these fair value adopters. Our findings show that the voluntary adoption of fair value reporting for investment property in China remains low. However, the fair value option for investment property is significantly more likely to be chosen by firms with greater needs for accounting discretion. Consistent with the conjecture that firms choose the fair value model to manipulate reported earnings, we show that fair value adopters use the unrealized gains and losses associated with investment properties to smooth earnings and that these firms are also more likely to meet or beat earnings benchmarks after adoption. Overall, our findings indicate that the use of fair value reporting for investment property in the emerging Chinese market is driven by managerial opportunism.     

Changes in the measurement of fair value: Implications for accounting earnings

With the FASB's issue of staff position papers in 2009 and the relaxation of how fair value standards are applied, there has been a change in the practice of how fair value is measured. Since the FASB staff position papers in 2009, fair value measurement by financial institutions has increasingly relied on managerial assumptions. This study examines the impact of this change on the quality of earnings. Consistent with attribute substitution theory that emphasises reliability over relevance, we find that an apparent increase in managerial discretion in fair value measurement is associated with a higher probability of earnings management and lower earnings informativeness. The results indicate that allowing more managerial discretion in fair value measurement adversely affected the quality of financial reporting. Our study highlights the issue of reliable measurement in the debate among academics and practitioners of increasing the use of fair value accounting.     

Voluntary fair value disclosures beyond SFAS 157’s three-level estimates

Some firms voluntarily make disclosures about the controls and processes in place to ensure the reliability of fair value estimates. Consistent with these disclosures being driven by investors’ concerns about the reliability of their SFAS 157 estimates, we find that firms with more opaque estimates are more likely to provide such disclosures. We then examine whether these disclosures improve investors’ perception about the reliability of fair value estimates. We find that they are associated with higher market pricing and lower information risk for Level 3 estimates. Further analyses of the disclosures reveal that the following types of information are particularly important to investors: discussion of the external and independent pricing of fair value estimates and their proper classification according to the SFAS 157 hierarchy. Overall, our results suggest that the voluntary reliability disclosures that firms provide beyond SFAS 157’s three-level estimates help reduce investors’ uncertainty toward the more opaque fair value estimates.     

The effect of additional guidance on fair value measurement and disclosure in illiquid or inactive markets

This paper examines fair value accounting – specifically, the application of FASB FSP 157-4 in the US. Data is analyzed from financial firms before and after FSP 157-4 was implemented to examine how this standard changed fair valuations and disclosures. We consider whether managers took advantage of the flexibility in the new standard by classifying their assets at level 3. We find that there is no significant change in the amount of assets that are transferred into level 3 after FSP 157-4 as compared to before. We also find a significant increase in the extent of disclosures as measured by word count. Fair value disclosures increased by an average of 52%. After further partitioning the sample based on size, we find that both main results hold for small and big firms in our additional sample. There is no evidence managers used the flexibility of the new standard to classify more financial assets at level 3; however, managers responded to the new standard with a significantly longer disclosure.     

Do measurement-related fair value disclosures affect information asymmetry?

Using a sample of European real estate firms over the 2007–2010 period, this study provides some evidence that measurement-related fair value disclosures reduce information asymmetry. We find a negative association between the extent of fair value disclosures and the bid-ask spread, but no association with two additional measures of information asymmetry (zero returns and price impact). Contrary to our expectation, we fail to find evidence that firms using model estimates exclusively benefit the most from such additional disclosure. Analysing measurement errors (the absolute difference between the selling price of an asset and its fair value prior to sale), we find that firms that use model estimates exclusively and provide more measurement-related disclosures have lower errors and more accurate fair value estimates. In other words, if our lack of results is due to investors not using this additional disclosure this is to their detriment.     

Fair value accounting: Current practice and perspectives for future research

A fundamental issue debated in the accounting literature centres on the appropriate basis for measuring firms’ assets and liabilities. During the last several decades, scholars have generated a growing body of important insights about the use of the fair value measurement attribute in financial reports around the globe. In this paper, we provide an overview of the institutional background of fair value accounting and the associated accounting standards that prescribe the use of fair value measurements under International Financial Reporting Standards and Generally Accepted Accounting Principles in the US. We discuss and document the extent to which firms across different industries and accounting regimes recognize and disclose in their financial reports assets and liabilities measured at fair value and we reflect on aspects of the fair value accounting literature. In doing this, we identify several areas in which additional research can further our understanding of fair value measurements and disclosures.     

Asset value management in the Portuguese real estate sector

This paper analyzes whether real estate investment fund managers use asset valuation discretion strategically to achieve financial reporting objectives. Portuguese real estate investment funds represent a unique opportunity to investigate executive behavior regarding accounting choice, as fund managers may choose to use historical cost, fair value or a mixed system (historical cost with internal revaluations) to value fund properties. We also investigate the factors that influence this strategic behavior. Empirical results confirm that fund managers manage asset valuations in order to avoid net asset value declines, particularly in a period of financial distress. We also observe that funds with a higher level of past unconditional conservatism are more likely to manage asset values. With respect to corporate governance issues, we conclude that audit quality reduces managerial discretion and that the conflicts that may arise between fund management company shareholders and fund participants due to management fees do not seem to have impact on fund managers’ opportunistic behavior.     

The financial reporting of fair value based on managerial inputs versus market inputs: evidence from mortgage servicing rights

This research examines whether the fair value of mortgage servicing rights (MSRs) based on managerial inputs (Level 3) better reflects the cash flow and risk characteristics of the underlying assets than the fair value of MSRs based on market inputs (Level 2). Using mortgage servicing fees as a proxy for the underlying cash flows, we find that the valuation multiples for MSRs based on Level 3 inputs are more positively associated with the persistence of future servicing fees compared with the fair value of MSRs based on Level 2 inputs. We also document that only the valuation multiples based on Level 3 fair values are negatively associated with proxies for risk factors. Our results suggest that, although unobservable inputs are subject to managerial discretions, managers can generate higher quality fair value estimates than market inputs due to their information advantage, especially when the market for the underlying asset is inactive.     

Information asymmetry of fair value accounting during the financial crisis

We examine whether US banks’ fair value net assets, measured according to the three-level hierarchy introduced in SFAS 157, are associated with information asymmetry during the 2008 financial crisis. Our results show that bid–ask spread, a proxy for information asymmetry, is positively associated with fair value net assets, and the degree of association is contingent upon the three-level hierarchy, with bid–ask spreads being lowest for Level 1 (the most transparent valuation inputs) and highest for Level 3 (the least observable). Also, there is some evidence that SFAS 157 led to a reduction in bid–ask spread, and we find that quarterly changes in Level 1 and Level 2 fair value net assets are significantly associated with changes in bid–ask spread in 2008 when the spread was rapidly rising, but not in 2009 when it was falling. Our findings suggest that the three-level hierarchy under SFAS 157 provides investors with useful information, and fair value is associated with uncertainty, as measured by bid–ask spread, before and during the financial crisis.     

IFRS fair value measurement and accounting policy choice in the United Kingdom and Australia

This study investigates the use of fair value measurement by 228 listed companies in the UK and Australia around the time of adoption of IFRS from 1 January 2005. We test whether within and between country comparability in policy choices (as measured by T indices) has changed in relation to (a) mandatory and (b) optional use of fair value measurement. Mandatory requirements related to financial instruments (IAS 39) and share-based payments (IFRS 2) have increased comparability, with a weaker effect for biological assets (IAS 41). In relation to the optional use of fair value, comparability increased in relation to property (IAS 16) because some companies discontinued fair value measurement. Under IAS 39, the fair value option for other financial assets and other financial liabilities decreased comparability. Options to use fair value in other areas (intangible assets, plant and equipment and investment properties) are not generally taken up, either for on-going measurement or on IFRS adoption (under the ‘deemed cost’ option). The results suggest a conservative approach and/or lack of incentives to use fair value measurement for most companies. Exceptions include some banks and insurance companies (for other financial assets and liabilities) and companies holding investment properties.     

Evidence on the use of unverifiable estimates in required goodwill impairment

SFAS 142 requires managers to estimate the current fair value of goodwill to determine goodwill write-offs. In promulgating the standard, the FASB predicted that managers will, on average, use the fair-value estimates to convey private information on future cash flows. The current fair value of goodwill is unverifiable because it depends in part on management??s future actions (including managers?? conceptualization and implementation of firm strategy). Agency theory predicts managers will, on average, use the unverifiable discretion in SFAS 142 consistent with private incentives. We test these hypotheses in a sample of firms with market indications of goodwill impairment. Our evidence, while consistent with some agency-theory based predictions, does not confirm the private information hypothesis.     

Measurement Model or Asset Type: Evidence from an Evaluation of the Relevance of Financial Assets

This study focuses on the operation of the Level 1, 2, and 3 measurement uncertainty hierarchy embedded in the SFAS 157 accounting for financial assets. Prior studies conclude the SFAS 157 fair value measurement model and prevailing financial market conditions are causal factors for the lower value relevance of the Level 3 financial assets. The contribution of our paper is to provide evidence on an additional, hitherto undocumented source of measurement uncertainty impacting the relevance of SFAS 157 financial assets to investors: the type of asset appearing in Level 3 financial assets as a result of asset securitizations and SFAS 140 securitization accounting. The paper also presents evidence that suggests the SFAS 166 amendments were unable to fully address informational transparency for financial assets arising from securitizations. The key contribution is evidentiary insights suggesting the prescribed measurement model has a relatively lower impact on measurement uncertainty and relevance of financial assets compared to the effects of the asset type.     

Value relevance and timeliness of transitional goodwill-impairment losses: Evidence from Canada

Focusing on transitional goodwill-impairment losses (losses) recorded by Canadian firms following the adoption of revised standards on purchased goodwill, we investigate the value relevance and timeliness of mandatory changes in accounting principles accounted for using the retroactive method. We find a negative relationship between reported losses and share price. Such a finding is consistent with investors perceiving losses as being sufficiently reliable measurements of a reduction in the value of goodwill to incorporate them in their valuation assessments. We find also that investors put a higher valuation weight on losses reported by firms that are expected to record a loss. In addition, we show that investors perceive that there are reduced opportunities for managerial discretion when there is a more effective audit committee. Finally, our results show that returns lead losses, i.e., that losses represent a catch-up adjustment to reflect the cumulative effect of using the impairment approach for the first time. Overall, our evidence supports U.S. standard setters' decision, through SFAS 154, to favour enhanced comparability and consistency over the potential costs of frequent restatements. Our results also show that fair-value measurements can be relevant even when the financial statement elements of interest are inherently bound to measurement error and subject to significant managerial discretion. They support the notion that reliability is about faithful representation, not precision.     

How have US banks adopted the Financial Accounting Standards Board's Level 3 fair value disclosure rules?

We examine how banks have complied with the Financial Accounting Standards Board's disclosure rules on Level 3 recurring fair value measurements. We document widespread noncompliance with the basic disclosure requirements. We also find that the noncompliant banks are smaller in size and are associated with lower audit quality, lower institutional ownership and less effective internal controls. Our results should be of use to regulators, auditors and audit committees in the United States, Australia and other countries for assessing the likelihood of noncompliance with fair value disclosure rules and improving the quality of fair value disclosures provided to investors.     

Value Relevance of Fair Values—Empirical Evidence of the Impact of Market Volatility

This study examines empirically the effects of market volatility on the value relevance of fair values. Using the modified Ohlson model ( 1995 ) and a sample of U.S. financial companies for the period of 2008 to 2013, this study shows that fair values are priced at a significant discount when market volatility is high. Song ( 2013 ) shows analytically that the effectiveness of fair value accounting is negatively affected by market volatility. Findings of the current study suggest that investors understand the effects of market volatility on fair values and price them accordingly. The study extends the research on the determinants of the usefulness of fair values by looking beyond factors associated with the reliability of estimated fair values (Level 2 and Level 3 fair values). This study has practical implications: current accounting standards for fair value measurement acknowledge the limitations of the market as a source of fair values by offering a three‐level fair value hierarchy with provisions for fair values to deviate from market prices. Findings of this study shed light on a previously little studied factor, that is, market volatility, on the usefulness of fair values.     

The association between firm characteristics and CFO's opinions on the fair value option for non-financial assets

SFAS 157 provides a common definition for fair value while SFAS 159 expands the applicability of the fair value option. This paper analyzes the responses of 209 CFOs of U.S. firms to a survey asking whether they would choose the fair value option for non-financial assets (FVONFA) and investigates the determinants of CFOs' responses to the option. One of our results suggests that CFOs in the U.S. are resistant to the FVONFA, consistent with prior studies based on firms in Europe and Australia. Our results also suggest that firm size, leverage, the amount of non-financial assets, and expertise in fair value measurements all positively affect the CFOs' responses to the FVONFA.     

The Effect of Fair Value Adjustments on Dividend Policy Under Mandatory International Financial Reporting Standards Adoption: Australian Evidence

Using a sample of firms from the financial sector of the Australian Securities Exchange, we examine the effect of the fair value adjustments of financial instruments on firms’ dividend distributions in the context of mandatory International Financial Reporting Standards (IFRS) adoption. We find a positive relationship between the fair value adjustments of financial instruments and firms’ dividend payouts, suggesting that the frequent use of fair value adjustments of financial instruments by financial firms following mandatory IFRS adoption has the potential to increase the proportion of transitory earnings in reported earnings and cause changes in dividend policies. Our results add to the ongoing debate on the unintended economic consequences of fair value accounting (FVA) and provide empirical support for regulators’ concerns that unrealized FVA gains from asset revaluation during booms may encourage the distribution of those unrealized gains.     

Fair Value Measurements of Control Premiums

As the overview of the current state of research within this paper shows, the debate around fair value measurements is far from over. This paper analyzes fair value measurement requirements in a controversial scenario, namely when a control premium exists. The analyses of the paper show that, while measurement rules around control premiums could have a material impact on fair value measurements and the financial statements as a whole, significant fair value measurement issues remain unresolved. The conclusion is that fair value measurements should include or exclude control premiums consistently. It is argued that including control premiums for all fair value measurements is the most faithful representation of the underlying economic phenomenon. This paper contributes to the fair value measurement debate by comparing the merits of alternative fair value measurements for control premiums and highlights an area where researchers, investors, and other users should exercise caution when evaluating financial statements.