# Introduction

he value relevance of earnings is referred to the association between earnings and security market values. This association was explored as early as in 1968 by Ball and Brown (1968), who examine the usefulness of accounting income numbers to investors. In the past two decades, a large number of empirical articles examine this association in different contexts. For example, Alford et al. (1993) find empirical evidence that this association between earnings and stock returns is stronger in countries where capital is traditionally raised in capital markets and there are weaker links between financial and tax reporting. Chan and Seow (1996) examine this association under foreign GAAP versus U.S. GAAP. They find a stronger association using foreign GAAP earnings than using earnings adjusted to U.S. GAAP. Ali and Hwang (1999) examine the relationship between value relevance and country specific factors related to financial reporting. They find that the value relevance of financial reports is lower for countries where the financial systems are bank-oriented rather than market-oriented.

Recently, in the literature, a group of articles focuses on examining the value relevance of accounting information under mark to market accounting. Barlev and Haddad (2003) advocate that historical cost accounting hides the real financial position and the fair value accounting is more value relevant than historical cost accounting. Ahmed and Takeda (1995) supports fair value accounting in their examination of the relation between unrealized or realized gain/losses and bank returns in normal period. Beatty et al. (1996) also find empirical evidence supporting fair value accounting in analyzing the bank share prices around the adoption of SFAS 115. Petroni and Wahlen (1995) argue that property-liability share prices can be explained by fair value of equity investments and U.S. Treasury investments, even after controlling for historical costs. Venkatachalam (1996) shows that the fair value estimates for derivative help explain cross-sectional variation in bank share prices and that the fair values have incremental explanatory power over and above notional amounts of derivatives. Gebhardt et al. (2004) find strong evidence in favor of fair value accounting within the German banking sector, although there are additional problems to overcome when extending fair value measurements to more classes of financial assets and liabilities.

Another group of articles in the literature examines the value relevance of accounting information under conservative accounting. For example, Kousenidis, Ladas and Negakis (2010) examine how conservatism affects value relevance of earnings. They use the data in Greece from the period 1989 to 2003. Their empirical evidence suggests that conservatism helps earnings to exhibit sufficient information content for security prices through alleviating measurement errors in earnings. Brown et al. (2006) find evidence in their international study among 20 countries that conservatism is associated with a higher level of value relevance in countries with high accrual intensity. However, the empirical results of Balachandran and Mohanram (2005) indicate that accounting conservatism is not related with the value relevance of accounting information. In their study on whether conservatism has caused the value relevance of accounting information to decline, the data over a twenty-five year period from 1978-2002 is used. They find that the value relevance of accounting information has declined only for firms with the least conservative accounting and changed insignificantly for firms with the most conservative accounting. Hellman (2008) in his study of how the conservatism principle is applied in Europe under IFRS, reports that the mixing of consistent and temporary conservatism practice can reduce the value relevance of earnings information. standard provide more value relevant accounting information. This paper takes an analytical approach to formally investigate whether earnings under conservative accounting is more value relevant than the ones under mark to market accounting.

Specifically, this paper sets up a one-period model of a competitive security market with one risk-free asset and one risky asset. The payoff of the risky asset is normally distributed. Rational traders have correct prior knowledge about the normal distribution of the risky asset's payoff. Earnings fixated traders; on the other hand, underestimate the mean and variance of the asset's payoff under historical cost accounting. The reason for this is as follows. Due to the nature of U.S. accounting standards being conservative, accounting earnings numbers do not incorporate the expected future profits from the positive net present value projects until they are realized while the expected future losses from the negative net present value projects are incorporated in the current accounting earnings numbers. The failure of the historical cost recognition rules to recognize the expected future profits of positive net present value projects in current accounting earnings implies that accounting earnings numbers underestimate the mean and variance of the asset's payoff (economic earnings). This means that earnings fixated traders who are fixated on accounting earnings would underestimate the mean and variance of the asset's payoff.

Under an alternative accounting of mark to market, earnings fixated traders are indistinguishable from rational traders. Also, with mark to market, the payoff of the assets is the same as the accounting earnings number generated under mark to market accounting.

Before any trade takes place, an informational signal about the risky asset's payoff is released to the market. Based on the informational signal received, rational and earnings fixated traders both rationally update their estimates of the mean and variance of the asset's payoff given their prior knowledge. The utilities of both rational and earnings fixated traders are exponential function of their wealth. In this competitive market, both types of traders behave as price takers. They trade against each other or against noise traders whose demand is completely random. The demand for risky asset of both rational and earnings fixated traders is generated from the maximization of their utilities.

Note that the notion of earnings fixated traders is originated from the Functional Fixation Hypothesis in the financial markets. This hypothesis claims that investors who are unfamiliar with different accounting methods of generating accounting output rely on reported accounting numbers for their investment strategies without paying attention to the procedures used in producing these numbers. There are some empirical evidence showing the existence of such fixated traders. For example, Bradshaw et al. (2001), Khurana et al. (2003) and Hermann et al. (2007) show that analysts do not pay enough attention to the differential persistence of earnings components and they tend to focus on total earnings. Brown et al. (2000) present evidence that neither analysts nor investors make proper use of note disclosure in financial reports.

This paper proves that the reported earnings are value relevant under both conservative accounting and mark to market accounting. Furthermore, if the proportion of earnings fixated traders lies in the upper range of the interval [0, 1], the reported earnings under conservative accounting are more value relevant than the ones under mark to market. Otherwise, the reported earnings under mark to market are more value relevant than the ones under conservative accounting.

The remainder of the paper consists of three sections. The next section presents the model. The results are discussed in Section three. Section four concludes the paper.


# II.


# The Model

This section constructs a one-period model of a competitive asset market. There is one risk-free asset and one risky asset. The payoff for the risk-free asset is one and the payoff for the risky asset (denoted as ) is normally distributed with the mean of and variance of There are three types of traders: earnings fixated traders, rational traders and noise traders.

Earnings fixated traders are functionally fixated on accounting earnings and they view the accounting earnings numbers as the asset's payoff (economic earnings). Since conservatism in accounting requires more verifiability for the recognition of gains than for the recognition of losses, the accounting earnings numbers do not incorporate the expected future profits from the positive net present value projects until they are realized while the expected future losses from the negative net present value projects are incorporated in the current accounting earnings numbers. The failure of the historical cost recognition rules to recognize the expected future profits of positive net present value projects in current accounting earnings implies that accounting earnings numbers underestimate the mean and variance of the asset's payoff (economic earnings). Denote earnings fixated traders' prior belief about the mean and variance of the risky asset's payoff as and respectively, where and Rational traders have correctly estimated the mean and variance of the asset's payoff.

An informational signal about the risky asset's payoff is released to the market before any trade takes modeled according to where is normally
? ? ? 2 ? ? ? ? ? 2 ? ? ? ? ? ? ? 2 ? ? ? 2 ? ?
place. The informational signal (denoted as ) is Based on the informational signal received, rational traders update their prior beliefs about the asset's payoff according to and respectively, where the subscript indicates rational traders and
�" �" = ? + ? ? Global Journal of
The derivations of equation ( 1) and ( 2) are presented in the appendix. Similarly, given their prior beliefs about the mean and variance of the asset's payoff, earnings fixated traders rationally update their conditional mean and variance about the asset's payoff according to and respectively, where the subscript indicates earnings fixated traders and Note that presented in the appendix. The utility functions of both rational and earnings fixated traders are exponential functions of their wealth. Their initial wealth for both rational and earnings fixated traders are assumed to be Since the asset market is assumed to be perfectly competitive, rational and earnings fixated traders both behave as price takers. In other words, they take the risky asset price as given. Denote trader demand for the risky asset as where Trader wealth at the end of the period is the summation of their initial wealth and profits from trading the risky asset in the period. That is, where is the asset price. Trader demand for the risky asset is generated from the maximization of where and is the coefficient of absolute risk aversion. With normality assumption, the demand for the risky asset for trader that maximizes his or her exponential utility function is obtained from solving the following optimization problem: Substituting equations ( 1) through (4) into the optimization problem (5), the demand for the risky asset for trader (where ) is solved as, Noise traders' demand for the risky asset is modeled to be normally distributed with the mean of zero and variance of

The supply of the risky asset is assumed to be zero. The risky asset price in this competitive market is determined by the market clearing condition below:
? 2 ? ? ? ? ? ? (? |ï?" ) = ? + ? ? ¡ ï?" ? ? ¢ ?(1)? ?? ? (? |ï?" ) = ? 2 ? ? ? 4 ? ? 2 ? + ? 2 ? = ? ? ? 2 ? ? (2) ? ? = ? 2 ? ? 2 ? +? 2 ? ? ? ? (? |ï?" ) = ? ? + ? ? ¡ ï?" ? ? ? ¢ ?(3)? ?? ? (? |ï?" ) = ? 2 ? ? ? 4 ? ? 2 ? + ? 2 ? = ? ? ? 2 ? ?(4)? ? ? = ? 2 ? ? 2 ? +? 2 ? . ? ? ? ? ? 2 ? ? ? 2 ? ?? ? 0 ? ? ? ? ? = ?? ? . ? 0 ? ? ? = ? + ? ? (???)? ? ? 0 ? ? (? ? ) = ?? ??? ? ? ? ? 0 ? ? (? = ?? ?) max ? ? ? ? (? ? |ï?" ) ? ? 2 ? ?? ? (? ? |ï?" )? ???? ? ? = ? + ? ? (? ? ?)?(5)? ? = ?? ? ? ? = ? ? + ? ? ¡ ï?" ? ? ? ¢ ? ? ?? ? ? 2 ? ? (6) ? ? = ?? Where ? 2 ? ? ?? ? + (1 ? ?)? ? + ? = 0? ?(7)
III.


# The Result

This section compares the value relevance of the reported earnings under conservative accounting and market to market accounting. It proves that the reported earnings are value relevant under both types of accounting methods. In addition, it also proves that the reported earnings under one of the accounting methods can be more value relevant than the ones under the other accounting method provided that the right model parameter value restrictions are imposed. Substituting equation ( 6) into (7), the asset price is solved as the following:

Taking a derivative of equation ( 8) with respect to yields the following:

Under an alternative accounting of mark to market, earnings fixated traders become indistinguis-
? = ?(1 ? ? ? ) ? ?? ? (? ? ? ? + ? ? ? ? ? ?? ? ) ? ? ? ?? ? + ?? ? + ? ? ? ? (�" + ??? 2 ? ) ? ? ? ?? ? + ?? ? = ?? ? (? ? (? ? 1) ? ? + 1) + ?? ? ? ? (1 ? ? ? ) + ? ? ? ? (�" + ??? 2 ? ) ? ? ? ?? ? + ?? ? ? (8)(9)? ? ?? ?? ? = ?? ? (1 ? ? ? ) ? ? ? ?? ? + ?? ? ? ? ? ? 1 . ? ? = ?, ? ? = ? ?
(due The derivations of equations ( 3) and (4) to

).

able to rational traders. In other words, Where denotes the proportion of earnings fixated traders in the population of rational and earnings fixated traders. (10) Note from equation (10) a positive association between earnings under conservative accounting and the asset price due to D and Hence, using equation ( 8), the asset price (denoted as ) is solved as Taking a derivative of equation with respect to earnings under mark to market accounting results in the following: which is positive due to 1 This means that an increase in earnings under markto market leads to an increase in the asset price.

Note from equations ( 10) and ( 12), 
? = 0? ? 0 ? 0 = ?(1 ? ? ? ) + ? ? (�" + ??? 2 ? )? (11) ?? 0 ?? = 1 ? ? ? ? ? ? ? ? ?? ?? ? ? ?? 0 ?? if 1 ? ? ? ? ? ?? ? ? ? ? ? +? 2 ? ?2? ? ? ? ; and(12)?? ?? ? ? ?? ?? if ? ? ? ? ?? ? ? ? ? ? +? 2 ? ?2? ? ? ? ?Theorem? * ? * �"= ?+?? ? ? ? ? ? ? ? ; ? ? �" ?
This means that if the proportion of earnings fixated traders lies in the upper range of the interval [0, 1], the reported earnings under conservative accounting are more value relevant than the ones under mark to market. Otherwise, the reported earnings under mark to market are more value relevant than the ones under conservative accounting.

IV.


# Concluding Remarks

This paper compares the value relevance of earnings under two different accounting methods (namely, conservative accounting and mark to market accounting) in a competitive security market. The earnings information under conservative accounting is transmitted into the market through the trading of earnings fixated traders. The results of the model suggests that under certain model parameter restrictions, the asset price incorporates more information from the earnings number generated from conservative accounting; on the other hand, under other model parameter restric-tions, the asset price incorporate more information from earnings generated from mark to market accounting.

The future research agenda in this direction is to investigate the impact of accounting standards on the value relevance of earnings under other market settings. One example of such market setting is the one that involves traders acting strategically instead of taking price as given.

V.  


# Global Journal of Management and Business Research
![mean of zero and variance of The random variables and are independent.](image-2.png "")


			© 2013 Global Journals Inc. (US)
		
		
			

With the results (a), (b) and (c), equations ( 1) and ( 2) follows from Theorem 1.

Derivation of equation ( 3) and ( 4): Notice that where is estimated by conservatism traders as normally distributed with mean

deviation of and is also normally distributed with mean zero and standard deviation of furthermore, and standard

and are independent. Hence, the following are true, (a) and are jointly normal distributed;
			
			

				


* 
	
		Stock market valuation of gains and losses on commercial banks' investment securities: an empirical analysis
		
			ASAhmed
		
		
			CTakeda
		
	
	
		J. Accoun. Econ
		
			20
			
			1995
		
	




* 
	
		The Relative Informativeness of Accounting Disclosures in Different Countries
		
			AlfordA
		
		
			JJones
		
		
			RLeftwich
		
		
			MZmijevski
		
	
	
		Journal of Accounting Research
		
			31
			
			1993
		
	




* 
	
		Country-specific factors related to financial reporting and the valuerelevance of accounting data
		
			AliA
		
		
			LHwang
		
		
			1999
		
		
			University of Arizona
		
	
	working paper




* 
	
		Conservatism and the value relevance of accounting information
		
			SVBalachandran
		
		
			PSMohanram
		
		
			2006
		
		
			Columbia University
		
	
	working paper




* 
	
		empirical Evaluation of Accounting Income Numbers
		
			RBall
		
		
			PBrown
		
	
	
		Journal of Accounting Research
		
			
			1968
		
	




* 
	
		Fair value accounting and the management of the Firm
		
			BBarlev
		
		
			JRHaddad
		
	
	
		Critical Perspectives Accoun
		
			14
			
			2003
		
	




* 
	
		An empirical analysis of the economic implications of fair value accounting for investment securities
		
			ABeatty
		
		
			SChamberlain
		
		
			JMagliolo
		
	
	
		J. Accoun. Econ
		
			22
			
			1996
		
	




* 
	
		Do Analysts and Auditors Use Information in Accruals?
		
			MTBradshaw
		
		
			SARichardson
		
		
			RGSloan
		
	
	
		Journal of Accounting Research
		
			39
			1
			
			2001
		
	




* 
	
		Acquisition-Related Provision-Taking and Post-Acquisition Performance in the UK Prior to FRS 7
		
			SBrown
		
		
			MFinn
		
		
			O-KHope
		
	
	
		Journal of Business Finance & Accounting
		
			27
			9
			
			2000
		
	




* 
	
		Conditional conservatism and the value relevance of accounting earnings: An international study
		
			WJBrown
		
		
			HHe
		
		
			KTeitel
		
	
	
		Eur. Accoun. Rev
		
			15
			
			2006
		
	




* 
	
		The Association between Stock Returns and Foreign GAAP Earnings versus Earnings Adjusted to US GAAP
		
			KChan
		
		
			GSeow
		
	
	
		Journal of Accounting and Economics
		
			21
			
			1996
		
	




* 
	
		Accounting for financial instruments in the banking industry: Conclusions from a simulation model
		
			GGebhardt
		
		
			RReichardt
		
		
			CWittenbrink
		
	
	
		European Accounting Review
		
			13
			
			2004
		
	




* 
	
		Accounting conservatism under IFRS
		
			NHellman
		
	
	
		Accoun. Europe
		
			5
			
			2008
		
	




* 
	
		The Effect of Changes in Japanese Consolidation Policy on Analyst Forecast Error
		
			DHermann
		
		
			TInoue
		
		
			WBThomas
		
	
	
		Journal of Accounting and Public Policy
		
			26
			1
			
			2007
		
	




* 
	
		Does Analyst Behavior Explain Market Mispricing of Foreign Earnings for U.S. Multinational Firms?
		
			IKKhurana
		
		
			RPereira
		
		
			KKRaman
		
	
	
		Journal of Accounting, Auditing and Finance
		
			18
			4
			
			2003
		
	




* 
	
		Value relevance of conservative and nonconservative accounting information
		
			DVKousenidis
		
		
			ACLadas
		
		
			CINegakis
		
	
	
		International Journal of accounting
		
			44
			
			2009
		
	




* 
	
		Fair values of equity and debt securities and share prices of property-liability insurers
		
			KRPetroni
		
		
			JMWahlen
		
	
	
		J. Risk Insurance
		
			62
			
			1995
		
	




* 
	
		Value-relevance of banks' derivatives disclosures
		
			MVenkatachalam
		
	
	
		Journal of Accounting and Economics
		
			22
			
			1996