비용의 가속 또는 지연인식, 이익변동성 및 이익지속성 = Accelerated or delayed recognition of expenses, earnings volatility and persistence

다국어 초록 (Multilingual Abstract)

This study examines the effect of accelerated or delayed recognition of expenses on earnings volatility and earnings persistence. It is documented that the matching between revenues and expenses deteriorates over time and that neutral matching, measured as the contemporaneous association between revenues and expenses, induces earnings to have higher quality than non-neutral matching does. The non-neutral matching could be classified as either accelerated or delayed matching between revenues and expenses. Accelerated recognition of expenses means that future-period expenses are recognized in the current period so that it is of conservative nature. On the other hand, delayed recognition of expenses indicates that current- period expenses are recognized in the future period such that it is aggressive in nature. However, the different effect of accelerated or delayed matching on earnings quality has not been explored yet. Using Dichev and Tang`s (2008) model where current-period revenues are regressed on past-period, current-period, and future-period expenses, I measure three types of recognition of expenses: (i) the neutral recognition of expenses is proxied by the standardized regression coefficient of current-period revenues on current-period expenses, (ii) the accelerated recognition of expenses is proxied by the standardized regression coefficient of current-period revenues on past-period expenses, and (iii) the delayed recognition of expenses is proxied by the standardized regression coefficient of current-period revenues on future-period expenses. Among three, the latter two types of recognition of expenses are of interest. Further, earnings volatility is defined as firm-specific variances of earnings and earnings persistence is defined as the firm-specific slope coefficients from the regression of future-period earnings on current-period earnings. The recognition of expenses, earnings volatility and earnings persistence are estimated using the rolling ten-year-period time-series data for each firm as consistent with Francis et al. (2004). In the accelerated recognition, part of current-period expenses is recognized in the past period. On the other hand, in the delayed recognition, part of current-period expenses is recognized in the future period. Both types of recognition introduce an intentional or unintentional interruption in the time-series of earnings by switching part of current-period expenses into past or future periods. Assuming that the switched portion of current-period expenses is not systematically different between the accelerated and delayed recognition, I do not expect that earnings volatility differ between two types of recognition. To the extent that earnings persistence is measured as the association between current-period and future-period earnings, earnings persistence in the delayed recognition is reduced by the negative correlation between current-period earnings and future-period earnings. This negative correlation is induced by the fact that delayed recognized current-period expenses are negatively correlated with current- period earnings. Accordingly, I expect that earnings are less persistent for the delayed recognition than for the accelerated recognition. The sample consists of 6,350 non-financial firm-years with December fiscal year that are traded over Korea Exchange for 1990-2008. Empirical findings generally support my expectation. Specifically, earnings volatility is not significantly different between the accelerated and delayed recognition but earnings persistence in the delayed recognition is significantly lower than that in the accelerated recognition. The result in this study provides a more complete understanding of the matching principle by documenting the effect of different types of expense recognition on earnings attributes such as earnings volatility and earnings persistence. Note that higher quality of earnings enables firms to raise external funds at lower costs and at the same time it enhances financial reporting transparency. In this vein, managers who are interested in earnings quality to raise cheaper external funds or standard- setters who are interested in earnings quality to enhance transparency of financial reporting will benefit from evidence in this study regarding potential consequence of such different types of expense recognition.

참고문헌 (Reference)

1 백원선, "회계환경의 변화와 수익비용대응원칙" 한국경영학회 40 (40): 29-49, 2011

2 백원선, "수익비용대응원칙과 이익의 질" 한국회계학회 36 (36): 101-127, 2011

3 Cryer, J, "Time Series Analysis" Duxbury Press 1986

4 Gaio, C, "The relative importance of firm and country characteristics for earnings quality around the world" 19 : 693-738, 2010

5 Dechow, P., "The quality of accruals and earnings: The role of accrual estimation errors" 77 : 35-59, 2002

6 Dichev, I, "On the balance sheet-based model of financial reporting" 22 : 453-470, 2008

7 Dichev, I, "Matching and the changing properties of accounting earnings over the last 40 years" 83 : 1425-1460, 2008

8 Pindyck, R, "Econometric Models and Economic Forecasts. 3rd ed" McGraw-Hill 1991

9 Dichev, I, "Earnings volatility and earnings persistence" 47 : 160-181, 2009

10 Sloan, R, "Do stock prices fully reflect information in accruals and cash flows about future earnings?" 71 : 289-315, 1996

11 Francis, J., "Costs of equity and earnings attributes" 79 : 967-1010, 2004

12 Paton, W, "An Introduction to Corporate Accounting Standards" American Accounting Association 1940