Xerox: Book-In-Time

SNA VIII Solo, 15 - 16 September 2005

1

THE EFFECT OF TRANSITORY EARNINGS

ON THE USE OF E/P RATIOS IN CORPORATE VALUATION:

EMPIRICAL EVIDENCE FROM INDONESIA

ERNI EKAWATI

Fakultas Ekonomi Universitas Kristen Duta Wacana

ABSTRACT

The purposes of this study are, firstly, to confirm the findings of the previous

study, whether earnings contain transitory components. Secondly, to investigate how E/P

ratios are affected when firms experience transitory earning changes. Thirdly, to examine

whether the differences in E/P ratios across firms due to differences in the magnitude of

transitory earnings will quickly disappear in subsequent years.

Using financial data of companies listed in Jakarta Stock Exchange (JSE) from

the periods of 1993 to 2003, the study finds that earning changes exhibit a transitory

component. Under the condition of transitory earnings, firm's E/P ratio is positively

affected by the changes in earnings. Industry-adjusted earning changes variable is shown

to have a high explanatory power in predicting firm's E/P ratio. As the variation of firm's

E/P ratios are mainly explained by the transitory component of earnings, time series

pattern of E/P ratios reflects transitory deviations due to the transitory component of

earnings changes.

A. INTRODUCTION

Many practitioners subscribe to the view that accounting earnings and earning-toprice

(E/P) ratios are among the most important figures in making investment decisions.

E/P has also drawn much attention from academicians, and has been interpreted in

various ways. Some popular interpretations of E/P ratios have been an earning

capitalization rate (Graham et al., 1962), an indicator of mis-priced stocks (Basu, 1977

and Jaffee et al., 1989), an indicator of transitory earnings (Beaver and Morse, 1978), and

a risk measure (Ball, 1978).

Recent academic evidence documents that changes in earnings and profitability

are to some extent predictable (Lev, 1969; Freeman et al., 1982; Collins and Kothari,

1989; Easton and Zmijewski, 1989; Ou and Penman, 1989; Elgers and Lo, 1994; Basu,

1997; and Fama and French, 2000;). Similar evidence are also found in Indonesia (Assih,

1999; Werdiningsih, 2001; and Febriyanti, 2004). Fama and French (2000) document

that earnings tend to be mean reverting. Large increases in earnings are followed by

subsequent decreases, while large decreases are followed by increases. The changes in

earnings tend to reverse from one year to the next, and large changes of either sign

reverse faster than small changes. They also find that negative changes in earnings

reverse faster than positive changes. Since extreme changes in earnings seem to reverse

faster, it is believed that these changes are due to the transitory component of earnings.

Unlike permanent earning components, transitory components such as

extraordinary gains and losses, changes in earnings due to changes in accounting

standards, and similar items that are not expected to continue in the future, are not

expected to persist for long periods of time. Thus, earnings that contain a large transitory

component lose relevance for making investment decisions (Stunda and Typpo, 2004).

Stock price represented in the denominator of the E/P ratios can deviate from its

value at a given point in time, but it will eventually correct itself to a value that is implied

in some fundamentals such as earnings. As a consequence, time series properties of

earnings have a potentially important implication for corporate valuation using E/P ratio

approach. Should E/P ratios be affected by the transitory component of earnings, the

investment analysts have to adjust the E/P ratios to account for unusual changes in

earnings. Consider a case that a firm has experienced an extreme decline in earnings due

to its poor performance, intuitively, this condition will lead to a higher E/P ratio (lower

P/E ratio). However, if the recent decline in earnings has a transitory component the

SNA VIII Solo, 15 - 16 September 2005

2

correct adjustment may be to decrease the E/P ratio. Furthermore, as the transitory

component of earnings may cause the E/P ratio of such firms temporarily deviate from its

value, the differences of E/P ratios across firms due to differences of the magnitude of

transitory components will quickly disappear in subsequent years.

The purpose of this study is, firstly, to confirm the findings of the previous study,

using data from companies listed in Jakarta Stock Exchange (JSE), whether earnings

contain transitory components or are mean reverting. Secondly, to investigate how E/P

ratios are affected when firms experience transitory