國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
10
improved comparability associated with the adoption of IFRS are two mechanisms behind the reduced cost of equity.
3. Hypotheses development
Given the evidences regarding the association between the improved information environment and the reduced mispricing of accruals, I predict that the mandatory IFRS adoption reduces the mispricing of foreign firms cross-listed in the U.S. for three reasons. First, IFRS is recognized as a high quality set of accounting standards, which can improve the information environment and provide relevant information for investors so that they can better distinguish the difference between the earnings components. Second, IFRS requires more extent to which financial and nonfinancial information are disclosed and prior researches have suggested that high-quality disclosure can reduce the mispricing of securities. As a result, with more required disclosure in IFRS, the mispricing could also decrease due to the mandatory IFRS adoption. Finally, since prior literature suggests that the IFRS-based and the US GAAP-based amounts are comparable, implying that IFRS can fit well in the U.S.
economic system, the mandatory IFRS adoption in foreign firms’ home countries could potentially affect the accruals anomaly of those firms cross-listed in the U.S.
Thus, the hypothesis is:
H1: Foreign firms cross-listed in the U.S market exhibit lower mispricing of discretionary accruals following the mandatory IFRS adoption.
4. Research design
To test the prediction that the mandatory IFRS adoption would affect the mispricing of discretionary accruals, I perform hedge portfolio tests and
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
11
return-discretional accruals regressions and compare the results of firms cross-listed in the U.S. from countries that mandate IFRS with their comparable U.S.-domiciled firms and with cross-listers from non-IFRS adoption countries.
4.1. Hedge portfolio tests
The hedge portfolio test mimics trading strategies based on accruals that are implemented by investors. I develop portfolios by independently sorting stocks into terciles based on levels of discretionary accruals, where discretionary accruals are estimated from the modified Jones (1991) model following Kothari et al. (2005). The hedge portfolio returns are the returns spread between the lowest and highest discretionary accruals terciles. Based on the modified Fama-French (1996) four-factor model (FFM), the portfolio abnormal returns are equivalent to the intercept of the following time-series regression:
𝑅𝑝,𝑡 − 𝑅𝑓𝑡= 𝛼𝑝+ 𝛽𝑝(𝑅𝑚𝑡− 𝑅𝑓𝑡) + 𝑠𝑝𝑆𝑀𝐵𝑡+ ℎ𝑝𝐻𝑀𝐿𝑡+ 𝑢𝑝𝑈𝑀𝐷𝑡+ 𝜀𝑝,𝑡 (1) where 𝑅𝑝,𝑡 is the return of the test tercile portfolio p in month t, 𝑅𝑓𝑡 is the risk-free return in month t proxied by the U.S. 30-days treasury bill yield, 𝑅𝑚𝑡 is the return of market portfolio in month t proxied by NYSE value-weighted yield, 𝑆𝑀𝐵𝑡 is the month t value-weighted return measured by the return of (S)mall (M)inus (B)ig companies, 𝐻𝑀𝐿𝑡 is the month t value-weighted return constructed by the return of (H)igh (M)inus (L)ow book-to-market value companies, 𝑈𝑀𝐷𝑡 is the momentum factor, factor coefficients 𝛽𝑝, 𝑠𝑝, ℎ𝑝, and 𝑢𝑝 capture portfolio risk exposures, and 𝛼𝑝 is interpreted as the portfolio abnormal return after controlling four risk-factors. I annualize the alpha by multiplying 12 and expect that there is a decline in hedge portfolio abnormal returns from the pre- to the post-IFRS period.
To mitigate the possibility that the evidence is driven by some confounding factors, I also report the results of two sets of control samples. The first set of control
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
12
firms is cross-listed companies from non-IFRS adoption countries. The second set of control groups is matched U.S.-domiciled firms. I identify a set of comparable U.S.
firms for each cross-listers from IFRS countries by matching with year, industry, and sales growth. I do not expect to find any abnormal return differences from the pre- to the post-IFRS period among these two control groups.
4.2. Regression-based tests
This regression provides evidence for time-series unobservable impact and controls for additional variables argued in the existing literature to be associated with the formation of accruals anomaly. I perform the following regression:
𝑆𝐴𝑅𝐸𝑇𝑖,𝑡+1 = 𝛾0+ 𝛾1𝑅_𝐷𝐴𝐶𝐶𝑖,𝑡+ 𝛾2𝑅_𝑂𝐶𝐹𝑖,𝑡 + 𝛾3𝑃𝑅𝐸𝑖,𝑡+ 𝛾4𝑅_𝐷𝐴𝐶𝐶𝑖,𝑡× 𝑃𝑅𝐸𝑖,𝑡 + 𝛾5𝑅_𝑂𝐶𝐹𝑖,𝑡 × 𝑃𝑅𝐸𝑖,𝑡 + 𝛾6𝑀𝑉𝑖,𝑡+ 𝛾7𝐵𝑀𝑖,𝑡+ 𝛾8𝐶𝑃𝑖,𝑡+ 𝛾9𝑅𝑉𝐴𝑅𝑖,𝑡 + 𝜀𝑖,𝑡 (2) where 𝑆𝐴𝑅𝐸𝑇𝑖,𝑡+1 is the annual size-adjusted return calculated as the difference between the raw buy-and-hold return of each firm and the return of matched portfolio to which each firm is assigned, 𝐷𝐴𝐶𝐶𝑖,𝑡 is discretionary accruals estimated from the modified Jones (1991) model following Kothari et al. (2005), and 𝑂𝐶𝐹𝑖,𝑡 is operating cash flow divided by total assets, R_DACC and R_OCF are the annual decile rank of discretionary accruals and operating cash flow, scaled to range from 0 to 1, 𝑃𝑅𝐸𝑖,𝑡 is a dummy variable assigned to 1 for pre-IFRS period, , 𝑀𝑉𝑖,𝑡 is the log of market value, 𝐵𝑀𝑖,𝑡 is the book-to-market ratio, 𝐶𝑃𝑖,𝑡 is the cash flow-to-price ratio, and 𝑅𝑉𝐴𝑅𝑖,𝑡 is the residual variance estimated from the Fama and French (1996) three-factor model. Coefficient 𝛾1 captures the return predictability of discretionary accruals in the post-IFRS period and 𝛾4 indicates the incremental effect in the pre-IFRS period.
Following Lang et al. (2006), I identify matched U.S.-domiciled firms for each
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
13
firm-year observation of cross-listers based on matching year, industry, and sales growth. Because the U.S. matched firms are not experiencing the mandatory IFRS adoption but will be affected by any other factors in the U.S. capital market, they serve as a direct control group for those identified foreign firms. Moreover, to control for other factors that U.S.-domiciled firms may not able to capture but could still influence cross-listers, I also perform regression-based tests separately on a sample consisting of foreign firms cross-listed in the U.S. from countries that have not mandated IFRS.