The logic of positive accounting research

Accounting Organizations and Society, Vol. 14, No. 5/6, pp. 455--468. 1989. Printed in Great Britain

0361-3682/89 $3.00+.00 Pergamon Press pie

THE LOGIC OF POSITIVE A C C O U N T I N G RESEARCH*

PAUL F. WILLIAMS

North Carolina State University

Abstract Positive accounting research consists of tests of positive accounting theory; it is a substantial body of important accounting literature. The discussion in this paper is aimed at demonstrating that a problem in experimental procedure exists with such research that is created when positive accounting researchers translate the theoretical language of positive accounting theory into an empirical language that permits experimentation. Experimentally, positive accounting theory is transformed into a tautolo~'. Experimenters define a causal variable in terms of the phenomenon they wish to explain with it. This transforms accounting practice into its own explanation, which is not quite the improvement in our understanding envisioned by positive accounting theory, in addition, the paper includes the demonstration that any plausible argument available to defend positive accounting research against the accusation that it creates a tautology requires that damage be done to the theory that informs such research.

Positive a c c o u n t i n g research (PAR) has prod u c e d a sizeable b o d y of literature ( s e e Table 1 for e x a m p l e s ) c o n s i s t i n g of r e p o r t s o n experim e n t s investigating the a c c o u n t i n g p r o c e d u r e s choice problem. The researchers performing these e x p e r i m e n t s have b e e n i n f o r m e d by what is called positive a c c o u n t i n g t h e o r y (PAT); specifically, the a p p l i c a t i o n o f J e n s e n & Meckling's ( 1 9 7 6 ) a g e n c y t h e o r y to e x p l a i n m a n a g e r s ' ( a n d others) choices of accounting procedures. T h e s e c h o i c e s m a y b e e i t h e r actual c h o i c e s of p r o c e d u r e o r p r e f e r e n c e s for p r o c e d u r e s revealed t h r o u g h l o b b y i n g of s t a n d a r d setting bodies. The p u r p o s e of any a c c o u n t i n g theory, acc o r d i n g to PAT's m o r e p r o m i n e n t p r o p o n e n t s (Watts & Z i m m e r m a n , 1986, p. 2), "... is to explain a n d p r e d i c t [emphasis in original] a c c o u n t ing practice." Positive a c c o u n t i n g t h e o r y is a n att e m p t to p r o v i d e s o m e e x p l a n a t i o n o f a c c o u n t ing p r a c t i c e r o o t e d in the p u r p o s e s o f managers. Simply stated, these p u r p o s e s of managers, as initially d e s c r i b e d b y Watts & Z i m m e r m a n

( 1 9 7 8 ) , are r e d u c i b l e to e c o n o m i c self-interest. That is, w h e n c h o o s i n g a c c o u n t i n g p r o c e d u r e s m a n a g e r s c o n s i d e r o n l y the effect of the proced u r e s o n their wealth; c h o i c e s are wealthm a x i m i z i n g given the c o n s t r a i n t s i m p o s e d by o t h e r w e a l t h - m a x i m i s i n g agents (e.g., shareholders, b o n d h o l d e r s ) . Further, it is p r e s u m e d that m a n a g e r s m a x i m i z e their wealth if they c h o o s e those a c c o u n t i n g p r o c e d u r e s that m a x i m i z e the value o f the firm a n d / o r m a x i m i z e m a n a g e r comp e n s a t i o n via c o m p e n s a t i o n a g r e e m e n t s tied to a c c o u n t i n g n u m b e r s . T h e list of wealth-affecting costs is familiar, e.g., political costs, b o o k k e e p i n g costs, taxes, c o n t r a c t i n g costs. T h e type o f study w i t h w h i c h this p a p e r is conc e r n e d is identified as positive a c c o u n t i n g research, s i n c e all studies of this type involve testing p o s i t i v e a c c o u n t i n g theory. T h e p u r p o s e of this p a p e r is to d e m o n s t r a t e that these PAR s t u d i e s all suffer from a logical flaw in their design w h i c h makes ascribing u n a m b i g u o u s meaning to the m e a s u r e s used to test the theory very p r o b l e m a t i c . T h e logical p r o b l e m is c r e a t e d

*Expressions of gratitude are due to Ed Arrington,Jon Bartley,Jere Francis, Katherine Frazier,participants at the accounting workshops at UNC-CH and Florida State University and two anonymous reviewers. The author also wishes to thank the Department of Economics and Businessof North CarolinaState Universityfor providing the financialsupport for this project. 455

456

PAULF. WILLIAMS

when the theoretical language of positive accounting theory is translated into the empirical language of PAIL The result of the translation process is to define self-interest in terms of accounting practice, the p h e n o m e n o n self-interest is to explain. Experimentally PAT is made tautologous. Further, it will be demonstrated that avoiding that tautology requires that one of the two fundamental propositions of PAT be either false or undecidable.

POSITIVE ACCOUNTING THEORY The demonstration that PAR has a problem of incoherency must begin with the development of the logic of PAT in an acceptable form. If the logical form of PAT is to be acceptable, it should be constructed from the blueprint explicitly given by the individuals w h o are currently constructing and testing it; this is the procedure that will be followed. Watts & Zimmerman, as previously noted (1986, p.2), claim that positive accounting theory is to provide a scientific explanation of accounting practice. They contend that their view of theory is the view of theory in science; to quote them ( 1986, p.2), "The preceding view of theory, explicitly or implicitly, underlies most empirical studies in economics. It is also the view of theory in science (e.g. Poincar~, 1905; Popper, 1959; Hempel, 1965)." Whether this view is appropriate to accounting or the social

sciences in general is questionable; it is perhaps presumptuous to impose a set of generally accepted scientific standards on accounting. The discussion in this paper is not to be taken as an endorsement of this view of theory; it is c o n c e r n e d only with demonstrating that a logical incoherence exists within those experiments designed as tests of some PAT propositions. The status of PAT qua theory is not at issue. The theory itself has been criticized, e.g., Tinker etal. (1982), Lowe et aL (1983), Schreuder ( 1985); the theorists have been criticized, e.g., Christenson (1983); and the epistemology forming the basis of the theory has been criticized, e.g., by Arrington ( 1 9 8 6 ) in accounting, McCloskey ( 1 9 8 5 ) in economics, and Rorty ( 1 9 7 9 ) in philosophy. It is apparent that in spite of such extensive criticisms, PAT still has the status of a serious theory for a number of accounting researchers. But if that status is granted, the evidence thus far offered by positive researchers is flawed evidence. The focus of this paper is on explaining why such experimental evidence is of doubtful value, thus the view of theory held by positivist accounting researchers will be maintained throughout. PAT is an explanation of accounting practice. According to Watts & Zimmerman ( 1986, p. 2), "(O)ur definition of accounting practice is broad." How broad is not explicitly stated but most certainly broad enough to include the accounting procedures that may be observed to be in effect at some particular time and place.

TABLEI. Examplesof PARstudies Authors

Dependent variables

Watts&Zimmerman(1978)

1. Supportor not support GPLA

Hagerman&Zmijewski(1979)

I. lnventor,/method 2. Depreciationmethod 3. Pension m e t h o d

4. ITCmethod

Independent variables I. 2. 3. 4. 5. 6. 1. 2.

Depreciationexpense + mkt. value" Net monetarya.ssetposition + mkt.value* Sales× effectof GPLAon income* Sales+ total sales in SICgroup x GPLAeffect* Compensationplan:yes or no Regulated:yes or no Totalassets° Sales* 3. Beta 4. Fixedassets + sales* 5. C o n c e n t r a t i o n ratio °

6. Compensationplan:yes or no

THE LOGIC OF POSITIVE ACCOUNTING RESEARCH Dhaliwal ( 1980 )

1. Oil and gas development cost method

1. Sales ( control variable )" 2. Debt/equity"

Salamon & Dhaliwal ( 1 9 8 0 )

1. Disclosure of segmental data

I. Total assets* 2. New capital issues: yes or no

Bowenetal. ( 1981 )

1. Capitalization of interest

1. 2. 3. 4. 5. 6.

Holthausen ( 198 i )

I. Abnormal return, proxy for depreciation change

1. 2. 3. 4.

Leftwich et aL ( 1981 )

1.

Zmijewski & Hagerman ( 1981 )

1. Income strategy

1. 2. 3. 4. 5. 6.

Chow (1982)

1. Hire external auditing

1. Mkt. value of owners' equity + book value of debt" 2. Debt + (measuredefined in I above)" 3. Number of accounting measures used in debt convenants 4. % management ownership 5. Stock registration: NYSEor OTC

Dhaliwawl et aZ (1982)

1. Depreciation method

1. O w n e r controlled: yes or no 2. Total assets* 3. Debt + equity*

Kelly(1982)

I. Reaction to SFAS # 8

1. 2. 3. 4. 5. 6.

Frequency ofinterim reporting

457

Compensation plan: yes or no Dividends + unrestricted R/E* Current interest + interest expense * Net tangible assets + long term debt" Sales* Unrestricted R/E: yes or no*

Forecast error EPS - stock price" Compensation plan: yes or no Impact ofdep, change on EPS + stock price* Bookvalueofpublicdebt + (stock price x shares)* 5. Book value ofprivate debt + (stock price × shares)* 6. Inventor)'ofpayable funds .--" (stock price x shares)" 7. Bookvalueofpubliedebt + book value of private debt + (stock price x shares)" 2. Net property + firm value* 2. Mkt. value ofstock + book value ofcurrent liabilities, long term debt and preferred stock* 3. Book value of bank loans, public and private debt + firm value" 4. Book value ofpreferred equity + firm value" 5. Outside director 6. Frequency of reporting in 1937 7. Stock exchange listing Compensation plan: yes or no Concentration ratio* Beta Log of net sales" Gross ftxed assets + sales* Total debt + total assets*

Compensation plan: yes or no Debt + equity" Total consolidated assets* % management ownership Foreign sales + total consolidated sales" Remuneration percentage*

458

PAUL F. WILLIAMS

KeLly(1985)

1. LohbyonSFAS #8

Wong(1988)

1. Disclosure ofcurrent cost

1. 2. 3. 4. 1.

financial statements 2. 3. 4. 5. 6.

Foreign sales + total consolidated sales* Total debt + total assets* % management ownership Total assets* Income tax expense (net ofdeferred tax) + net income before taxes" Long-term liabilities .-"( total assets current liabilities)* Market concentration ratios measured using net income* Net income before interest and taxes + total assets* Gross f'Lxedassets + totalassets" Net income after taxes before extraordinary items*

"Denotes on accounting measure.

M o s t o f t h e s t u d i e s l i s t e d in T a b l e 1 a r e c o n cerned with accounting procedures; Watts & Z i m m e r m a n ( 1 9 7 8 , p. 1 1 2 ) s t a t e that, "Ultim a t e l y , w e s e e k to d e v e l o p a p o s i t i v e t h e o r y o f the determination of accounting standards." Of course, accounting standards are accounting p r o c e d u r e s , i.e., t h e r u l e s t h a t a r e a p p l i e d for integrating transaction events into the financial s t a t e m e n t s . It s e e m s t o b e t h e c a s e w i t h p o s i t i v e t h e o r i s t s that t h e a c c o u n t i n g p r a c t i c e t o b e e x p l a i n e d b y a p o s i t i v e t h e o r y m u s t c e r t a i n l y include the procedures accountants employ to c a l c u l a t e s u c h t h i n g s as t o t a l assets, t o t a l liabilities, o w n e r s ' e q u i t y , n e t i n c o m e , etc. If accounting procedures are not part of the phenomenal domain of accounting practice then most o f w h a t has b e e n w r i t t e n u n d e r t h e r u b r i c s o f positive theory and positive research would. m a k e little sense. As C h r i s t e n s o n ( 1 9 8 3 , p . 1 7 ) n o t e s , t h e e m p h a s i s in PAT is o n e x p l a n a t i o n . T o e x p l a i n an e v e n t , in t h e v i e w o f s c i e n c e e n d o r s e d b y posit i v e a c c o u n t a n t s , is t o c o m e t o a c c e p t t h e e v e n t ' s c a u s e s . T h e i m p o r t a n c e o f c a u s a l i t y in an e x p l a n a t i o n is n o t e d b y W a t t s & Z i m m e r m a n ( 1 9 8 6 , p.4):

The public accountant or corporate manager may observe an association between variables such as changes in procedures and changes in stock prices, but cannot tell whether the association is causal... To make the causality interpretation, the practitioner requires a theory that explains the relation between the variables. The theory enables the practitioner to attach causality [emphasis added} to a particular variable, such as a procedure change. C l e a r l y , in W a t t s & Z i m m e r m a n ' s v i e w it is t h e r o l e o f t h e o r y to p e r m i t t h e a t t r i b u t i o n o f causality. t If PAT is to b e e x p l a n a t o r y in t h e s e n s e env i s i o n e d b y p o s i t i v e t h e o r i s t s , it m u s t c o n t a i n at least o n e p r e m i s e o r p r o p o s i t i o n that p e r m i t s causal attribution. What the theoretical proposit i o n s a r e for p o s i t i v e t h e o r y c a n b e i n f e r r e d f r o m t h e p o s i t i v e l i t e r a t u r e ; t h e r e a r e t w o that a r e apparent. One proposition pertains to the motives of m a n a g e r s ; it is t h e s e l f - i n t e r e s t p r o p o s i t i o n . In t h e i r s e m i n a l p a p e r , W a t t s & Z i m m e r m a n ( 1978, p. 1 1 3 ) s t a t e "...we a s s u m e t h a t i n d i v i d u a l s a c t to m a x i m i z e t h e i r o w n utility." Utility p r o v i d e s m a n a g e m e n t w i t h t h e m o t i v e for p r e f e r e n c e o n a c c o u n t i n g p r o c e d u r e s , i.e., " ( T ) h e o b v i o u s imp l i c a t i o n o f this a s s u m p t i o n is t h a t m a n a g e m e n t

~lt should be emphasized here that positive theorists have been perhaps rather facile in their treatment of causality. One of the reviewers of this paper noted that the inadequacy of PAT's notion of causation is still one more difficulty not overcome. Though this issue may be somewhat germane to the argument in this paper, particularly with respect to the problem of the "similarity of events", incorporating it would mean venturing too far afield. For a brief elucidation (with bibliography) of the philosophical problems of causation see Taylor (1972).

THE LOGIC OF POSITIVEACCOUNTING RESEARCH l o b b i e s o n a c c o u n t i n g s t a n d a r d s b a s e d o n its o w n self-interest" (p. 113). Watts & Z i m m e r m a n e q u a t e serf-interest o r utility w i t h e c o n o m i c interests: " W e a s s u m e that m a n a g e m e n t ' s utility is a positive function of the expected compensation in f u t u r e p e r i o d s ( o r w e a l t h ) and a n e g a t i v e function of the dispersion of future compensation ( o r w e a l t h ) . T h e q u e s t i o n is h o w d o accounting standards affect management's wealth?" (p. 114). W e c a n thus state p o s i t i v e acc o u n t i n g t h e o r y ' s first t h e o r e t i c a l p r o p o s i t i o n as: M a n a g e r s ' e c o n o m i c self-interests d e t e r m i n e t h e i r p r e f e r e n c e s for a c c o u n t i n g p r o c e d u r e s . " T h e m a n a g e r i a l self-interest o r m o t i v e p r o p o s ition is b y itself insufficient to p r o v i d e t h e causal e x p l a n a t i o n o f a c c o u n t i n g p r a c t i c e that p o s i t i v e t h e o r i s t s seek, s i n c e m a n a g e m e n t p r e f e r e n c e is not accounting practice. What positive accounting t h e o r i s t s have d o n e to p r o v i d e for an explanation o f a c c o u n t i n g p r a c t i c e is to a t t r i b u t e to m a n a g e m e n t a causal role. 3 W a t t s & Z i m m e r m a n ( 1 9 7 8 ) state q u i t e e x p l i c i t l y a b e l i e f in t h e causal efficacy o f m a n a g e m e n t : " M a n a g e m e n t , w e believe, plays a c e n t r a l r o l e in t h e determination [ e m p h a s i s a d d e d ] o f s t a n d a r d s " ( p . l 1 3 ) . This a t t r i b u t i o n o f causality has b e e n r e i t e r a t e d by subsequent researchers putatively testing the theory. F o r e x a m p l e , H o l t h a u s e n ( 1 9 8 1 , p. 7 3 ) d e s c r i b e d a m o t i v a t i o n for his s t u d y as follows: "Recently, h o w e v e r , r e s e a r c h e r s have b e g u n to e x a m i n e m a n a g e m e n t ' s i n c e n t i v e s to influence the m e n u [ e m p h a s i s a d d e d ] o f a c c e p t e d acc o u n t i n g t e c h n i q u e s a n d to c h o o s e a m o n g availa b l e alternatives." In a similar vein, Z m i j e w s l d & H a g e r m a n ( 1981, p. 1 2 9 ) i m p l y a causal r o l e for m a n a g e m e n t , i.e.: "These a c t i o n s ( m a n a g m e n t ' s l o b b y i n g a c t i v i t y ) a r e d e s i g n e d to influence the

459

set [ e m p h a s i s a d d e d ] o f g e n e r a l l y a c c e p t e d acc o u n t i n g p r i n c i p l e s ( G A A P ) from w h i c h a firm m a y c h o o s e . " Later in the s a m e a r t i c l e t h e s e t w o a u t h o r s a l l u d e to b o t h p r o p o s i t i o n s ( p p . 1 2 9 130): The question is, assuming economic rationality, what are the benefits justifying these costs (management's lobbying and changing accounting principles)? In response to this question a positive theory of the determination [emphasis added} and choice of accounting principles is being developed. O b v i o u s l y , if m a n a g e r s are rational, t h e y w o u l d not i n c u r costs to influence a d o p t i o n o f a c c o u n t ing p r o c e d u r e s if t h e r e w e r e n o p r o s p e c t t h e y c o u l d s u c c e e d . If m a n a g e r s a r e n o t in s o m e , e v e n limited, s e n s e causal o f a c c o u n t i n g p r a c t i c e , att e m p t i n g to u n d e r s t a n d t h e i r b e h a v i o r m a k e s little sense for an e x p l a n a t o r y t h e o r y o f a c c o u n t i n g p r a c t i c e . W e can state s u c c i n c t l y PAT's s e c o n d t h e o r e t i c a l p r o p o s i t i o n as: Managers' prefere n c e s for a c c o u n t i n g p r o c e d u r e s affect t h e obs e r v a b l e set o f e x i s t i n g a c c o u n t i n g p r o c e d u r e s . 4 Having i d e n t i f i e d t h e t w o c e n t r a l p r o p o s i t i o n s o f PAT, it is n o w p o s s i b l e to c o n s t r u c t a c r i t i q u e of the experiments of those researchers who p u r p o r t to b e t e s t i n g t h e theory. T h e b e h a v i o r o f s u c h p o s i t i v e r e s e a r c h e r s can b e s h o w n to i m p l y that t h e t w o f u n d a m e n t a l p r o p o s i t i o n s o f PAT c a n n o t b e b e l i e v e d b y t h e m to b e t r u e simultaneously. For e c o n o m y o f e x p o s i t i o n , t h e dem o n s t r a t i o n o f t h e logical i n c o h e r e n c e o f PAR in t h e n e x t s e c t i o n will b e d e v e l o p e d b y focusing o n o n l y o n e r e p r e s e n t a t i v e PAR s t u d y listed in Table 1; t h e i m p l i c a t i o n s e x t e n d to all s t u d i e s listed there.

2Alltests of positive accounting theory have focused exclusively on wealth variables. If other ~utility'-yieldingvariables are deterministic of managers' choices of procedures they have yet to be incorporated into positive theory. -~Foran explanatory theory of accounting practice to have almost singularly focused on the choices of management must imply that managers' choices of accounting procedure must be viewed as important determinants of the set of accounting procedures we are able to observe. 4It is admittedly the case that many "groups," e.g., regulators and CPAs, participate in the process of creating accounting procedures. The theoretical proposition is not meant to imply that management is strictly causal, for such is probably not factual. It does imply that knowledge of management preferences provides knowledge about what permissible accounting procedures one is likely to observe, i.e., management is a causal agent, even if only one among many.

460

PAULF. WILLIAMS THE LOGICAL PROBLEM W I T H PAR

T h e logical p r o b l e m w i t h PAR will b e e x p l a i n e d t h r o u g h t h e use o f a syllogism c r e a t e d f r o m t h e t h e o r e t i c a l p r o p o s i t i o n s o f PAT d e r i v e d in t h e p r e v i o u s s e c t i o n . T h e s e t w o p r o p o s i t i o n s are t h e self-interest p r o p o s i t i o n ( h e n c e f o r t h l a b e l e d TP1 ): Managers' economic self-interests determine their preferences for accounting procedures; and t h e causal a g e n t p r o p o s i t i o n ( h e n c e f o r t h l a b e l e d T P 2 ): Managers' preferences for accounting procedures affect the set of existing accounting procedures. T h e s e t w o p r o p o s i t i o n s are c o n j u n c t s and c a n b e c o m b i n e d to f o r m t h e c o m p o u n d s t a t e m e n t o f PAT ( h e n c e f o r t h l a b e l e d TP): Managers' economic self-interests determine their preferences for accounting procedures and managers' pt'eferences for accounting procedures affect the set of existing accounting procedures. This s t a t e m e n t is a t r u t h - f u n c t i o n a l c o m p o u n d s t a t e m e n t (s ee, e.g., Copi, 1979). If b o t h c o m p o n e n t s (TP1 and T P 2 ) are "true", t h e n p o s i t i v e t h e o r i s t s m a y h a v e t h e e x p l a n a t o r y t h e o r y o f acc o u n t i n g p r a c t i c e ( d e f i n e d as a c c o u n t i n g p r o c e d u r e s ) , that t h e y seek. But it is also t h e case that

if e i t h e r TP1 o r TP2 is n o t "true", t h e n TP is also n o t "true". s T h e n e x t step in c o n s t r u c t i n g t h e syllogism is to c o m b i n e t h e t h e o r e t i c a l p r o p o s i t i o n TP w i t h a factual s t a t e m e n t ( h e n c e f o r t h l ab el ed FS): T h e set o f e x i s t i n g a c c o u n t i n g p r o c e d u r e s d e t e r m i n e s e x i s t i n g a c c o u n t i n g measures. 6 This statem e n t is c a l l e d factual s i n c e if it is n o t c o n s i d e r e d factual t h e s p e c i f i c a t i o n o f t h e p h e n o m e n o n PAT is c o n s t r u c t e d to e x p l a i n is substantially altered; t h e "facts" PAT deals w i t h disappear. If m a n a g e r s ar e rational and if t h e s e l e c t i o n o f a c c o u n t i n g p r o c e d u r e s d o n o t d e t e r m i n e t h e r esu l t i n g accounting measures of managements' performance, t h e r e w o u l d b e n o a c c o u n t i n g p r o c e d u r e s c h o i c e p r o b l e m to explain. It is b e c a u s e differ e n t a c c o u n t i n g p r o c e d u r e s p r o d u c e different a c c o u n t i n g m e a s u r e s that m a n a g e r s e x p e n d e n e r g y trying to c r e a t e and c h o o s e a m o n g procedures. C o m b i n i n g TP w i t h FS p e r m i t s d e d u c i n g a t h e o r e t i c a l c o n c l u s i o n ( l a b e l e d TC), w h i c h c o m p l e t e s t h e syllogism. T h e c o m p l e t e set o f logically r e l a t e d st at em en t s, w i t h c o n c l u s i o n , is as follows: TP: Managers' economic self-interests determine their preferences for accounting procedures and managers' preferences for accounting procedures affect the set of existing accounting procedures. FS:The set of existing accounting procedures detertrdnes existing accounting measures.~ TC: Managers" economic serf-interests affect existing accounting measures.

~Since the term "true" is used in reference to propositions that are purportedly testable by experimental means, for these propositions to be "true" to positive researchers the ideas expressed by them must correspond to the facts. The activity of positive researchers is directed toward establishing whether such correspondence exists i.e., are the implications for belief of the PAT propositions publically acceptable. The discussion in this paper is concerned not with truth conditions but with whether the "facts" being offered as evidence by positive researchers can be reasonably accepted as such. °The term "accounting measure ~ is understood in the sense in which accountants use it. The number appearing after "Accounts Receivable" on a balance sheet is a "measure ~ of the asset, accounts receivable. The ~bottom line" of an income statement is an accounting ~measure". Financial ratios are accounting ~measures". VSome may take issue with the use of the verb "determines" in the factual statement. Do not other ~things" also "determine" accounting measures? The answer to that may be yes or no. Given a set of accounting procedures, other things determine accounting measures, but given a distinct setof other "things", accounting procedures determine accounting measures. The only thing required for the argument being developed in this paper is to grant what positive researchers must grant if they are to continue to have a problem to research: accounting procedures determine accounting measures to an extent greater than ~trivially', i.e., the effect is one that cannot be ignored.

THE LOGIC OF POSITIVE ACCOUNTING RESEARCH

The theoretical conclusion is a logically t r u e statement if both TP and FS are true. s Equipped with the set of logically related PAT propositions, it is n o w possible to demonstrate that positive accounting researchers have introduced an incoherency to their experiments by making it logically impossible for the components of the theoretical proposition ( T P I and TP2) to both be true simultaneously while preserving unambiguous meanings for the measurements made to test PAT. To give meaningful interpretation to the results of PAR experiments requires denying one of the c o m p o n e n t propositions. The logical p r o b l e m arises w h e n positive researchers translate the language of the theory into a language that permits empirical testing. H o w this translation process occurs and the logical result it produces will be illustrated using one paper representative of those informed by PAT, that by Hagerman & Zmijewski ( 1 9 7 9 ) (henceforth HZ).9 In the introduction to their p a p e r (p. 142), HZ indicate the paper's purpose: "(T)his paper is designed to determine if economic motives are determinants in the choice of alternative accounting principles." In effect, the e x p e r i m e n t reported in the paper was a test of the truth value of TPI, as are all of the other studies listed in Table 1. Because their study was informed by PAT, management's choices of accounting procedures w e r e the dependent variables. For HZ or any other positive researchers to test w h e t h e r "... economic motives are determinants in the choice of alternative accounting principles," they must alter the language of TP so that it has empirical meaning. Since managers' economic self-interests are not observable, HZ

461

define certain economic variables they believe can be defined operationally and substituted for managers' economic self-interests. These economic variables are linked to managers' e c o n o m i c self-interests by the logical arguments provided by PAT. In their study, HZ identify five such e c o n o m i c variables: size, risk, capital intensity, competition, and incentive plans. They link these variables to managerial self-interest via the standard arguments. For example, size is associated with the political costs firms may incur w h e n subjected to political attack from groups lobbying to transfer wealth from firms, and thus from managers, to themselves. HZ argue (pp. 142-143): Firms have a variety of tactics they can employ to reduce these costs.., the management may reduce reported net income in order to avoid drawing the lobbyists' attention to themselves. The reasonableness of this tactic is clear when one remembers h o w many n e w (sic) broadcasters have reported spectacular earnings increases of large firms. Thus large firms will have an incentive to choose accounting standards ~'hich reduce net income in order to avoid publicity and opprobrium.

Similar arguments are made to link the other variables to economic interests. They are: 1. Risk: "... riskier firms will appear to make excessive profits and thus be subject to negative wealth transfers... so we expect higher risk firms to choose income deflating alternatives" (p. 142); 2. Capital intensity: ~Thus we hypothesize that firms that are relatively capital intensive and subject to political costs will have an incentive to reduce reported in. come b v selecting the appropriate accounting principles" (p. 143); 3. Competition: "Therefore, we assume that the more concentrated the industry is, the greater is the likelihood

a'I'he derivation may easily be demonstrated if the propositions are put in functional form. Functionally, TP 1 is: Management preferences = f ( Economic self-interests ), and TP2 is: Existing accounting procedures --- g (Management preferences), and FS is: Existing accounting measures = h (Existing accounting procedures). By substitution: Existing accounting measures = h (g ( f ( . ) ) ) . 'q'his paper was selected for two reasons. The first is that it provides an example of the logical problem with PAR with which other PAR studies can be equated. The second is that it is an important paper. According to Brown & Gardner's ( 1 9 8 5 ) citation analysis, it is the fifth most influential paper in accounting published since 1979.

462

PAULF. WILLIAMS

of either anti-trust or entry, hence the greater the incentive to choose accounting methods that reduce reported Prolits~ CP. 145); and 4. Incentive plans: "If management incentive schemes are related to accounting earnings we expect that management has an incentive to use accounting principles that increase accounting earnings if part of their income is derived from incentive plans" (p. 145).

O n c e positive researchers like HZ have identiffed e c o n o m i c variables w i t h e c o n o m i c serf-interest, they s u b s t i t u t e these variables into the t h e o r e t i c a l p r o p o s i t i o n of PAT. T h e p r o p o s i t i o n ( l a b e l e d TP-I) n o w reads:

TP-I: Economic variables determine managers' preferences for accounting procedures and managers' preferences for accounting procedures affect the set of existing accountingprocedures. " E c o n o m i c variables" stand-in for the u n o b s e r v able e c o n o m i c self-interests of managers. After e c o n o m i c variables have b e e n e n u m e r ated, it is n e c e s s a r y to identify p r o x i e s for t h e m to define t h e m operationally, t° W i t h o u t operational d e f i n i t i o n s for e c o n o m i c variables, empirical testing of PAT is n o t possible. Positive researchers have t e n d e d to rely v e r y heavily o n existing a c c o u n t i n g m e a s u r e s c r e a t e d from the financial s t a t e m e n t s of firms as t h e i r p r o x i e s for e c o n o m i c variables. For the studies listed in Table 1, o v e r 70% of all p r o x i e s used w e r e c r e a t e d from r e p o r t e d a c c o u n t i n g n u m b e r s . C o n t i n u i n g w i t h the HZ e x a m p l e , these researc h e r s followed the p r a c t i c e o f u s i n g existing acc o u n t i n g m e a s u r e s for their proxies, i.e.: Size = Total assets o r Sales ( a c c o u n t i n g m e a s u r e s ) ; Risk = Beta; Capital I n t e n s i t y = Fixed assets + Sales ( a c c o u n t i n g m e a s u r e s ) ; C o m p e t i t i o n = Conc e n t r a t i o n ratio ( a c c o u n t i n g m e a s u r e s ) ; a n d Inc e n t i v e Plans = C o m p e n s a t i o n p l a n exists. W h e n these existing a c c o u n t i n g m e a s u r e proxies are s u b s t i t u t e d for e c o n o m i c variables i n t o TP-I, the original t h e o r e t i c a l p r o p o s i t i o n of PAT is further altered and b e c o m e s ( l a b e l e d TP-II):

TP-II: Existing accounting measures determine managers' preferences for accounting procedures and managets' preferences for accounting procedures affect the set of existing accounting procedures. "Existing a c c o u n t i n g measures" n o w stand-in for e c o n o m i c serf-interests and serve as m e a s u r e s of preference. T h e theoretical p r o p o s i t i o n , thus altered, has b e e n g i v e n empirical m e a n i n g , i.e., managers' serf-interests are defined in t e r m s of o b s e r v a b l e s that are quantified. T h e altered p r o p o s i t i o n is ind e e d the o n e tested by positive researchers. Every s t u d y listed in Table 1 was a test of TP1 a n d e v e r y o n e used existing a c c o u n t i n g measures as the vast majorit3, of i n d e p e n d e n t variables to p r e d i c t managerial p r e f e r e n c e s for acc o u n t i n g practices, e.g., p r o p o s e d standards, acc o u n t i n g p r o c e d u r e s , e x t e r n a l auditing. T h e p r o b l e m that positive a c c o u n t i n g researc h e r s create for the i n t e r p r e t a t i o n of their exp e r i m e n t a l results is revealed w h e n TP.II is subs t i t u t e d for the original theoretical proposition. TP, to yield an empirical ( o b s e r v a b l e ) c o n c l u sion: TP-II: Existing accounting measures determine managers' preferences for accounting procedures and mana. gets' preferences for accounting procedures affect the set of existing accounting procedures. FS:The set of existing accounting procedures determines existing accounting measurs. Empirical Conclusion (EC): Existing accounting measures affect existing accounting measures. This empirical c o n c l u s i o n appears to b e a t a u t o l o g o u s statement. That a c c o u n t i n g measures affect o t h e r a c c o u n t i n g m e a s u r e s is a reco g n i z e d fact. Do positive researchers still possess a t h e o r y w h o s e p r o p o s i t i o n s e n a b l e the ded u c t i o n of a fact a b o u t a c c o u n t i n g practice or have t h e y m e r e l y m a d e the theory, empirically, t a u t o l o g o u s by d e f i n i n g m a n a g e r s ' e c o n o m i c serf-interests in t e r m s of a c c o u n t i n g practice? If "existing a c c o u n t i n g measures" as the s u b j e c t of EC is reliably i n t e r p r e t a b l e as "measures of econo m i c serf-interest" t h e n the f o r m e r may likely b e

t~l'he issue of selecting better or worse proxies is a problem in its own right but it not germane to the issue at hand. This paper is concerned with the conditions under which a possible proxy can even be considered a reasonable candidate.

THE LOGICOF POSITIVEACCOUNTINGRESEARCH the case. But if the ability to sustain confidence in that interpretation of "existing accounting measures" is compromised then the latter may be the more likely case. In the following sections an attempt is made to demonstrate from the behavior of positive researchers that the latter is indeed the case.

SUSTAINING THE MEANING OF "EXISTING ACCOUNTING MEASURES" Positive researchers may be excused from having pursued the expedient path of malting PAT true, and thus guaranteeing some significant correlations, by defining managerial serf-interest in terms of accounting practice if their research can be interpreted as valid attempts to establish the truth value of TP1 and TP2. Because, as previously noted, for PAT to be an explanatory theory of accounting practice, its two fundamentai propositions must be believable. They must be "true." All of the studies listed in Table 1 are tests of TP1. The studies are reports of experiments each designed to test the hypothesis of whether the economic self-interests of managers determine their preferences for accounting practices. The results of these studies have yielded rather mixed results but supportive enough of TP1 that it is still maintained as a hypothesis. PAT has not yet been altered to suggest some other explanation for managers' preferences for accounting practices. The question is whether these attempts to test TP1 provide valid results within the context of PAT. As noted in the previous section, the validity of these results depends upon whether existing accounting measures are reliable indicators of managements' economic selfinterests; i.e., do they connote preferences? The reliability of these measures can be interpreted within two contexts: TP2 is "true" or it is "false".

463

I f TP2 is true For the purpose of the argument to follow, TP2 is true if knowledge of the preferences of managers for accounting procedures alters the probability one would assign to observing a particular procedure being among the set allowed in practice. If TP2 is true, two interpretations of PAR results are possible depending on interpretations of managers' behavior in pursuit of their economic serf-interests. One interpretation of serf-seeking behavior is that it implies management is manipulative. That means that managers' preference for accounting procedures are for those that allow them to tell stories economically beneficial to them. IfTP2 is true with such manipulative managements, then h o w do positive researchers attribute unambiguous economic meaning to existing accounting measures? There is a reflexivity; managers are manipulating those variables to represent the "economic" facts in a manner beneficial to themselves, tl How can positive researchers, with the reliability required of a "scientific" measurement, interpret those variables as containing information of economic substance? Accounting measures in such circumstances can convey little more than information about managers' past preferences for accounting procedures. This point can be illustrated with a simple example from positive theory. Sales and Total Assets are accounting numbers used to measure political costs. Positive theory predicts that, ceterispartbus, firms that are "larger", measured by Sales or Total Assets, will have higher potential political costs and their managements will, therefore, attempt to create and choose incomereducing accounting prQcedures. This strategy, is deemed rational because of the assumed importance of profits (namely "excess profits") as an event prompting the imposition of increased political costs. But h o w can managers affect income-reducing procedures without affecting either Sales or Total Assets? By the arithmetic of

HOne might argue there are no economic "facts"conveyed by accounting numbers. The response to this as if there is no extant idea of communicatingreliable economic data then manipulatingaccountingmeasures makesno sense. For example, the notion of incomesmoothingis irrational unless one believes somethingof economicsignificanceis beingcommunicated other than that management is acting to smooth income.

464

PAULF. WILLIAMS

t h e i n c o m e e q u a t i o n , m a n a g e r s c a n r e d u c e inc o m e b y d e c e l e r a t i n g t h e r e c o g n i t i o n o f Sales rev e n u e . Or, t h e y c a n a c c e l e r a t e t h e r e c o g n i t i o n o f e x p e n s e s , w h i c h in t u r n d e c e l e r a t e s t h e g r o w t h o f T o t a l Assets. Since t h e p r o c e s s o f m a n a g e r i a l creation and selection of accounting procedures is ongoing, if m a n a g e r s a r e i n d e e d effective mani p u l a t o r s o f a c c o u n t i n g m e a s u r e s t h e r e is n o relia b l e w a y to k n o w w h a t s u c h p u b l i c a c c o u n t i n g m e a s u r e s mean. D o t h e y say s o m e t h i n g a b o u t t h e r e l a t i v e p o l i t i c a l c o s t s o f firms o r s i m p l y t h e r e l a t i v e effectiveness o f m a n a g e r s as mani p u l a t o r s o f a c c o u n t i n g n u m b e r s ? T h e r e is n o justifiable c r i t e r i o n o f f e r e d for d e c i d i n g . T h e original e x p l i c a t o r s o f PAT, W a t t s & Zimm e r m a n , s u b s c r i b e t o t h e m a n i p u l a t i v e charact e r i z a t i o n o f m a n a g e m e n t b e h a v i o r . In d e s c r i b ing t h e value o f t h e i r t h e o r y ( W a t t s & Z i m m e r man, 1986, p. 356), t h e y m a k e t h e f o l l o w i n g claim: The theory provides investors and financial analysts with a useful predictive model of the accounting procedures underlying the financial statements. Using the theory, investors or analysts do not [emphasis addedl interpret balance sheet and earnings numbers as unbiased estimates of firm value and changes in firm value. Instead, they recognize the effect of the contracting and political processes on the calculation of earnings and balance sheet numbers. For example, the manager's incentives to choose earnings increasing/decreasing accounting methods depend on the existing compensation and debt contracts With knowledge of these contracts, the analysts can aftjust the reported numbers [emphasis added], in particular, if Healey's (1985) evidence regarding the effect of compensation plans on accounting accruals (seef:hapter 11 ) is confirmed, an investor or analyst could adjust the earnings number for expected management manipula. tions [emphasis added] in deriving cash flow estimates. This would help the investor or analyst better predict the market value of nontraded stocks or bonds. W h a t W a t t s & Z i m m e r m a n a r e asserting is that for t h e p u r p o s e s o f financial analysis, a c c o u n t i n g m e a s u r e s a r e b i a s e d b y t h e m a n i p u l a t i v e beh a v i o r o f m a n a g e m e n t a n d t h a t this b e h a v i o r is p r e d i c t e d b y t h e i r t h e o r y ( m o r e specifically b y tests o f TP1). Yet this t h e o r y has b e e n t e s t e d using t h e s e same, unadjusted a c c o u n t i n g measures. W h a t o n e is r e q u i r e d to b e l i e v e is that, t h o u g h m a n a g e m e n t affects a c c o u n t i n g m e a -

s u r e s that a r e b i a s e d and, t h e r e f o r e , u n r e l i a b l e for p u r p o s e s o f assessing t h e value o f t h e firm, t h e s e s a m e a c c o u n t i n g m e a s u r e s are u n b i a s e d a n d r e l i a b l e i n d i c a t o r s o f t h e e c o n o m i c self-int e r e s t s o f managers. T o e n t e r t a i n t h e i d e a that o n e is c o n d u c t i n g scientific e x p e r i m e n t s req u i r e s that t h e s c i e n t i s t p r o v i d e s o m e assura n c e s that t h e measurementglafi- is u s i n g a r e reliable indicators of the magnitude of treatment factors. His m e a s u r e m e n t s o f i n d e p e n d e n t varia b l e s m u s t b e i n d e p e n d e n t o f his t h e o r y , i.e., h o w h e i n t e r p r e t s his o b s e r v a t i o n s o f i n d e p e n d e n t variables s h o u l d n o t d e p e n d u p o n t h e status o f his t h e o r y . In W a t t s & Z i m m e r m a n ' s case, a n o t h e r t h e o r y is n e e d e d that p r e d i c t s that no matter how much management manipulates a c c o u n t i n g m e a s u r e s to its e c o n o m i c advantage, t h e a c c o u n t i n g m e a s u r e s so p r o d u c e d a r e alw a y s u n b i a s e d m e a s u r e s o f t h e e c o n o m i c self-int e r e s t s m o t i v a t i n g m a n a g e r s to manipulate. If TP2 is true, w i t h m a n i p u l a t i v e m a n a g e m e n t , s k e p t i c i s m a b o u t t h e " e v i d e n c e " p o s i t i v e researc h e r s have p r o d u c e d s e e m s to b e in o r d e r until s u c h a t h e o r y is p r o d u c e d . It w o u l d s e e m for n o w that if c o n v i n c i n g tests o f TP1 are to b e m a d e u n d e r t h e a s s u m p t i o n o f successfully manipulative managers, t h e m e a s u r e s o f the i n d e p e n d e n t variables m u s t e x c l u d e m e a s u r e s p r o d u c e d b y t h o s e managers. T h e o t h e r i n t e r p r e t a t i o n o f m a n a g e r i a l selfs e e k i n g b e h a v i o r u n d e r t h e c o n d i t i o n that TP2 is t r u e is that it is n o t m a n i p u l a t i v e . That is equival e n t to g r a n t i n g to t h e p o s i t i v e r e s e a r c h e r his ass e r t i o n that a c c o u n t i n g m e a s u r e s u s e d as indep e n d e n t variables d o p r o v i d e r e l i a b l e m e a s u r e s o f m a n a g e r s ' e c o n o m i c self-interests. Thus, managers' p r e f e r e n c e s for a c c o u n t i n g p r o c e d u r e s p r o d u c e a c c o u n t i n g m e a s u r e s f_hat h a v e an e c o n o m i c i n t e r p r e t a t i o n ; t h e y tell t h e researcher something about what managers' econo m i c self-interests are. T h e r e f o r e r e g a r d l e s s o f t h e i r motives, m a n a g e r s a r e s o m e h o w g u i d e d to c r e a t e a n d p r e f e r a c c o u n t i n g p r o c e d u r e s that y i e l d e c o n o m i c i n t e r p r e t a t i o n s that n o t o n l y tell a b o u t t h e i r o w n b e h a v i o r b u t also p u b l i c i z e t h e i r e c o n o m i c interests. T h e i m p l i c a t i o n o f g r a n t i n g this status to a c c o u n t i n g m e a s u r e s is t h e a p p a r e n t loss o f self-interest as an explana-

THE LOGIC OF POSITIVE ACCOUNTING RESEARCH

tion. If the accounting procedures preferred by managers p r o d u c e accounting measures believed by positive researchers to be reliable indicators o f e c o n o m i c constructs, then it can be said that in a very real sense the choices of managers are not manipulative; they are veracious. Since managers' past choices are used as measures of independent variables and those measures are assumed to be unbiased measures, then serf-interest and veracity b e c o m e indistinguishable. It is e x t r e m e l y difficult to discern w h e t h e r managers' accounting preferences are motivated by e c o n o m i c serf-interest or by a desire to tell the truth, at least to positive researchers. Perhaps another way of stating the above conclusion will add some clarity. Experimentally positive theory states that managers' preferences for accounting procedures are functionally determined by accounting measures of managers' e c o n o m i c serf-interests, i.e.: Managers' current preferences = f (accounting measures).

But the accounting measures are believed to measure in an unbiased way the e c o n o m i c variables that tell the researcher about managers' economic serf-interests, so: Managers'current preferences = f (accounting measures that are unbiased), where unbiased means that the researcher believes them to convey knowledge about economic p h e n o m e n a motivating managers. The assumption that TP2 is true permits the following statement: Managers' current preferences = f (managers' past preferences which produced unbiased measures). A n experiment confirming this relationship that unbiased accounting measures predict managers' current preferences, i.e., coefficients of the variables are statistically significant, would imply that m a n a g e r s ' behaviors are consistent with their economic self-interest, but it also implies that their current preferences will produce unbiased measures in the future. This last impli-

465

cation appears to be a necessary assumption for the following reason. Since scientists are concerned with answering questions of the general form, What should one expect (Toulmin, 1986)?, the experimenter as a scientist asserts that his result permits him, when he performs a similar experiment for the next accounting preference, to expect to obtain a similar result. His confidence that his next experiment will produce the e x p e c t e d result is justiffed only if he believes that the preference he has just investigated results in unbiased measures. He has already implicitly assumed that all preferences that p r e c e d e d his recently completed experiment did so. If managers' current preferences do not produce unbiased measures in the future, measurement error of the independent variables in that future experiment has been induced. This would seriously compromise the confidence of any e x p e r i m e n t e r in his ability to replicate his result in the future. With biased measures, future experiments may not result in economic variables predicting managers' preferences. Only if managers' choices lead to unbiased measures is the experimenter's confidence in his result as a scientific result justified. But if managers' choices always lead to unbiased measures, then it b e c o m e s m o o t whether economic serf-interest or a desire to disclose truthfully what their serf-interests are motivates managers in their accounting preferences. H o w does one tell the difference empirically? The positive researcher seems to sacrifice his ability to establish empirically the truth of TP1. If managers' preferences .always produce unbiased measures of their e c o n o m i c self.interests I then those measures cannot distinguish what the motives for those preferences are. Some accounting measures that would be different between serf-interest and veracity as motives are not available. In conclusion, if TP2 is true, then the PAR results produced thus far are not meaningfully interpretable. If managers manipulate, then accounting measures cannot just be assumed to be unbiased measures of managers' economic selfinterests. If the positive researcher wishes to continue to assert that he is entitled to regardac-

466

PAULF. WILLIAMS

counting measures as unbiased measures of managers' economic serf-interests then whether or not TP1 is true does not matter. Accounting measures produced by managers' preferences are the same regardless of motives; the truth of TP1 cannot be determined using accounting measures. I f TP2 is n o t true The second context within which to consider the positive researcher's belief in the reliability of accounting measures is that of TP2 being untrue, i.e., managers' preferences for accounting procedures do not affect the existing set of accounting procedures permitted in practice. This argument is a more plausible one than that under the condition that TP2 is true. The assumption that management does not affect accounting practice implies that some regular (or irregular) process or mechanism, potentially involving a multitude of participants (living and dead), is being assumed to affect accounting practice instead of management. Various analogies of this process are available, e.g., a cartographic analogy (Solomons, 1978) or accounting practice as analogous to a commodity sold in a competitive market (agency theory is one such characterization). Whatever the analogy, always tacitly evoked, the researcher relies upon it to provide the assurance needed to ascribe economic meaning to his independent variables. This dependence on some assumed process which always produces reliable accounting measures has equally troublesome implications for the positive researcher. Reliance on "something" other than management implies that the process of creating accounting practice does not require management participation. Accounting measures would be the same whether management prefers one accounting procedure or another, t2 This must be asserted, for if the measures were not the same we would be back to TP2 being true and the arguments given in the previous section would hold. Results of the experiments listed in Table 1

might then be legitimately claimed to be tests of TP 1; managers' economic serf-interests do determine their preferences for accounting practices. But that assurance comes only by destroying PAT as an explanatory theory of accounting practice. If positive researchers want to argue that their measures of managers' economic selfinterests are reliable because TP2 is not really true, then the theory that informs them no longer has an explanatory proposition. Managers are no longer causal. PAR is n o w the same kind of research accountants have been doing for many years, i.e., Lens-model research. Accounting variables, independently produced, are used to predict choice; accounting practice is being used to explain managers' choices which is the opposite o f what PAT purports to do. Positive researchers deprive PAT of its serf-proclaimed status as an explanatory theory of accounting practice.

CONCLUSION The conclusion of this paper is that an explanatory theory of accounting practice cannot be convincingly tested by resorting to accounting practice as its own explanation. It was demonstrated that the practice of using existing accounting measures as proxies for managements' economic serf-interests makes PAT tautologous at the experimental level. This practice makes it logically impossible for the two fundamental theoretical propositions of PAT to be true simultaneously. If the results of PAR (tests of the serf-interest hypothesis) are to be accepted with any degree of confidence, then managers cannot be assumed to be causal. If managers are assumed to be causal, then either the results of PAR are uninterpretable or the self-interest proposition becomes undecidable. It is not logically possible, using accounting measures, to establish simultaneously the truth value of the two propositions. What would seem to be minimally necessary

*ZHere"same"does not necessarilymean identical. "Same"implies immaterial difference,where immaterialityis decided by the positive researcher.

THE LOGICOF POSITIVEACCOUNTINGRESEARCH for positive researchers to actually test the propositions of PAT would be the design of experiments in which the truth of one proposition is not crucial to the test of the other and "facts" do not have to be assumed. For example, the way "political costs" have been tested thus far is to include an accounting measure of size as an argument in a linear decision model. What "facts" are implicitly assumed? One is that a linear decision model is appropriate; two is that larger firms actually bear higher political costs (an assertion, not an empirically tested proposition); and three is that accounting measures are unbiased estimates of "size". If results are not significant, is it because managers are not self-interested, or that linear models do not capture very well the decision process, or that size is a poor indicator of exposure to political costs, or that accounting measures are poor measures of "size"? There is no way to tell. Independent tests of TP1 and TP2 are possible, hut may very well require accounting researchers to largely abandon the equating of "scientific" behavior with sophisticated financial statement analysis that is the characteristic of so much current empirical work in accounting. PAT's substantive contribution has been its emphasis on accountability, not predictive usefulness, as the organizing principle of accounting. It, perhaps properly, made financial statements the dependent variables, but has confused things empirically by resorting to the comfortable financial analysis simile and making the same statements its independent variables. Given the current institutional make-up of the process by which accounting procedures are created, the PAT perspective would seem to lead some researchers to consideration of political p o w e r and the question of managements' abilities to affect the outcomes of this process. This requires experimental designs necessitating that data be gathered from sources other than those shared with financial analysts. For example, one experiment to test TP2 could take advantage of the d o c u m e n t e d history of the 97 FASB standards actually adopted. Some of these standards underwent major changes from initial exposure to final form. In addition to

467

adopted standards, some proposed standards were "killed". With a sample of issues that large, it should be possible to assess the relative likelihoods of positions advocated by management (weighted, perhaps) being adopted versus not being adopted. Such a study might provide some evidence about the extent of management's p o w e r in creating the story told about it. Another conclusion of this paper is the question it raises of how PAR could be conducted for nearly a decade without anyone noting that defining managerial self-interests in terms of the p h e n o m e n o n such self-interest is to explain does serious damage to the believability of any results? Notable about the studies listed in Table 1, which span a decade, are that they employ the same "method", and their results, when considered i n toto, are inconclusive. No consistent results emerge. No variable persists in significance across studies and each situation investigated produces significant variables (if any) that are unique. Results thus far indicate a great amount of situational specificity to PAT with, as yet, no real evidence to know whether it is poor theory or poor tests of the theory. This decade-long use of the same method, which produces no consistent results, to investigate a theory that in spite of those results has successfully resisted any emendment suggests two things. The first is that relying so heavily on COMPUSTAT tapes may not be the best way to subject PAT to rigorous testing. Perhaps case studies of procedure choice or lobbying behavior are in order to begin building a meaningful data base with which to test such theories of accounting practice. The second thing that PAR's decade-long constancy suggests is that a potential problem may exist with quality control. Whitley ( 1984, p.66) notes the tendency in academic disciplines for c o m p e t e n c e to b e c o m e defined in terms of people's abilities to employ various techniques. Zeff( 1983, p. 134) alludes to this same occurrence in accounting: "When modeling problems, researchers seem to be more affected by technical developments in the literature than by their potential to explain phenomena". After ten years there is some reason to believe that the tech-

468

PAUL F. WILLL~MS

n i q u e s t h u s l~ar u s e d t o t e s t P A T a r e i n a d e q u a t e

t o t h e task.

BIBLIOGRAPHY Arrington, C, The Rhetoric of Inquiry and Accounting Research, Unpublished Working Paper (The University of Iowa, 1986). Bowen, R., Lacey, J. & Noreen, E., Determinants of the Corporate Decision to Capitalize Interest,Journal of Accounting and Economics ( August 1981 ) pp. 151 - 179. Brown, L D. & Gardner, J. C., Applying Citation Analysis to Evaluate the Research Contributions of Accounting Faculty and Doctoral Programs, The Accounting Review (April 1985) pp. 262-277. Chow, C., The Demand for External Auditing: Size, Debt and Ownership Influences, TheAccounting Review (April 1982) pp. 272-291. Christenson, C., The Methodology of Positive Accounting, The Accounting Review 0anuary 1983) pp. 122. Copi, 1. M., Symbolic Logic (New York: Macmillan Publishing Company, 1979). Dhaliwal, D., The Effects of the Firm's Capital Structure on the Choice of Accounting Methods, The Accounting Review (January 1980) pp. 78--84. Dhaliwal, D., Salamon, G. & Smith, E,, The Effect of Owner Versus Management Control on the Choice of Accounting Methods,Journal of Accounting and Economics (July 1982) pp. 41-53. Hagerman, IL & Zmijewski, M., Some Economic Determinants of Accounting Policy Choice, Journal of Accounting anaEconomics (August 1979) pp. 141-161. Holthausen, R., Evidence on the Effect of Bond Covenants and Management Compensation Contracts on the Choice of Accounting Techniques: The Case of the Depreciation SwitchbackJournal of Accounting and Economics (March 1981 ) pp. 73-109. Jensen, M. & Meckling, W., Theory of the Firm: Management Behavior, Agency Costs and Ownership Structure,Jourttal of Financial Economics (1976) pp. 305-360. Kelly, L, Corporate Lobbying and Changes in Financing or Operating Activities in Reaction to FAS No. 8, Journal of Accounting and Public Policy (Winter 1982) pp. 153---173. Kelly, L, Corporate Management Lobbying on FAS No. 8: Some Further Evidence,Journal of Accounting Research (Autumn 1985) pp. 619-632. Leftwich, R., Watts, R. & Zimmerman, J., Voluntary Corporate Disclosure: The Case of Interim Reporting, Journal of Accounting Research (Supplement 1981) pp. 50-77. Lowe, E. A., Puxty, A. G. & Laughlin, R. C., Simple Theories for Complex Processes: Accounting Policy and the Market for Myopia,Journal of Accounting and Public Policy (Spring 1983) pp. 19-42. McCloskey, D. N., The Rhetoric of Economics (Madison: The University of Wisconsin Press, 1985). Rorty, R., Philosophy and TbeMirror of Nature (Princeton: Princeton University Press, 1979). Salamon, G. & Dhaliwal, D., Company Size and Financial Disclosure Requirements with Evidence from the Segmental Reporting IssueJournal of Busines~ Finance and Accounting (Winter 1980 ) pp. 55 5-568. Schreuder, H., Suitable Research: On the Development of a Positive Theory of the Business Suit, Accounting Organizations and Society (1985) pp. 105-108. Solomons, D., The Politicization of Accounting,Journal of Accountancy (November 1978) pp. 65-72. Taylor, R., Causation, in The Encyclopedia of Philosophy, Reprint ed. (New York: Macmillan, 1972 ). Tinker, A. M., Merino, B. & Neimark, M. D., The Normative Origins of Positive Theories: Ideology and Accounting Thought, Accounting Organizations and Society (1982) pp. 167-200. Toulmin, S., The Place of Reason in Ethics (Cambridge: Cambridge University Press, 1950). (Reprinted Chicago: University of Chicago Press, 1986). Watts, R. & Zimmerman, J., Towards a Positive Theory of the Determinants of Accounting Standards, The Accounting Review (January 1978 ) pp. 112-134. Watts, R. & Zimmerman, J., Positive Accounting Theory (Englewood Cliffs, NJ: Prentice-Hall, 1986). Whitley, IL, The Intellectual and Social Organization of the Sciences (New York: Oxford University Press, 1984). Wong, J., Economic Incentives for the Voluntary Disclosure of Current Cost Financial Statements,Journal of Accounting and Economics (1988) pp. 151-167. Zeff, S. A., Editorial: A Retrospective, The Accounting Review (january 1983) pp. 129-134. Zmijewski, M. & Hagerman, IL, An Income Strategy Approach to the Positive Theory of Accounting Standard Setting~Choice,Journal ofAccounttng andEconomtcs (August 1981 ) pp. 129-149.