Table A1

Summary of the key variables (with value measured in RMB for all variables)

VariableDefinitionData source
BHAR [t1, t2]Market-adjusted BHAR over the event window [t1, t2] is calculated by first compounding the raw returns of firm i and its benchmark (the corresponding value-weighted market index returns) over the event window [t1, t2] and then calculating the BHAR as the difference between the compounded returns of firm i and its benchmarkCSMAR
MBEA dummy variable equal to 1 if a firm's reported EPS equals or exceeds the latest analysts' forecasted EPS (i.e. the earnings surprise is 0 or positive), and 0 otherwise. For all variables that involve analyst forecasts (e.g. MBE, SMBEAT and BIGEAT), the latest analyst EPS forecast should be made at most 150 days before the firm's annual earnings announcement date and at least 1 day before the beginning of the event window of the corresponding BHAR in the regression modelCSMAR
MBE_EMA dummy variable equal to 1 if an MBE firm is identified as having engaged in income-increasing earnings management to achieve MBE, and 0 otherwise. We classify a firm as engaging in income-increasing earnings management through discretionary accruals, production, related-party sales, related-party purchases, expenditures or CFO if it is in the highest quintile of AAcruals, AProduction, ARPS or ARPP, or it is in the lowest quintile of AExpenditures and ACFO during the yearCSMAR, Wind
SMBEATA dummy variable equal to 1 if a firm's actual EPS equals or exceeds the latest analyst forecast by 1 cent per share or less, and 0 otherwiseCSMAR
BIGBEATA dummy variable equal to 1 if a firm's actual EPS equals or exceeds the latest analyst forecast by more than 1 cent per share, and 0 otherwiseCSMAR
AAcrualsAbnormal accruals are estimated using the modified Jones model (equation (A1) in  Appendix 2)CSMAR
AProductionAbnormal production cost is estimated using equation (A2) in  Appendix 2CSMAR
AExpendituresAbnormal expenditures are estimated using equation (A3) in  Appendix 2CSMAR
ACFOAbnormal cash flows from operations are estimated using equation (A4) in  Appendix 2CSMAR
ARPSAbnormal related-party sales are estimated using equation (A5) in  Appendix 2Wind
ARPPAbnormal related-party purchases are estimated using equation (A6) in  Appendix 2Wind
SurpriseEarnings surprise scaled by the stock price per share at the beginning of the year. Earnings surprise is measured as the actual EPS released on the annual earnings announcement date minus the latest analyst EPS forecast made at most 150 days before the firm's annual earnings announcement date and at least 1 day before the beginning of the event window of the corresponding BHAR in the regression modelCSMAR
AnalystAnalyst coverage of a firm, which is measured as the total number of analysts following the firm during the yearCSMAR
GVTA dummy variable equal to 1 if a firm is ultimately controlled by either the central government or a local government, and 0 otherwiseAudited annual reports
ROANet profit scaled by year-end total assetsCSMAR
ETPThe ratio of earnings to price, which is calculated as actual EPS scaled by the closing share price 3 days before the annual earnings announcement dateCSMAR
SizeThe natural logarithm of total assets at year endCSMAR
MTBThe ratio of market value of equity to book value of equity at year endCSMAR
LeverageThe ratio of total liabilities to total assets at year endCSMAR
CFOCash flows from operating activities divided by year-end total assetsCSMAR
Prior stock returnMarket adjusted BHAR over the window [−210, −11] before the annual earnings announcement dateCSMAR
Posi∆ProfitA dummy variable equal to 1 if a firm has a positive increase in earnings during the year, and 0 otherwiseCSMAR
ProfitableA dummy variable equal to 1 if a firm reports a positive net profit in its annual earnings announcement, and 0 otherwiseCSMAR
Sales growthChanges in net sales revenue from year t−1 to year t divided by net sales revenue in year t−1CSMAR
SEOA dummy variable equal to 1 if a firm makes a share issue application between year t + 1 and year t + 3, and 0 otherwiseCSMAR
Delisting riskA dummy variable equal to 1 if a firm is issued a delisting risk warning, and 0 otherwiseCSMAR
Analyst dummyA dummy variable equal to 1 if a firm is covered by at least one analyst in a given year, and 0 otherwiseCSMAR
IndAnaCovThe proportion of firms with at least one analyst in the same industry in a given yearCSMAR
Inverse MillsInverse Mills ratio, calculated based on the Heckman (1979) two-stage modelAuthors' construction
Surprise_rankAn ordinal variable obtained by ranking Surprise into deciles, subtracting 1 and then dividing by 9CSMAR

or Create an Account

Close subscription notice
Close access options