Impact of regulatory environment on accruals manipulation of bankrupt firms

ABSTRACT

The aims of this study are twofold: first, to determine whether bankrupt firms manipulate signed accruals upwards compared with active firms with similar performance, and second, whether the regulatory environment influences pre-bankruptcy manipulation. Using the performance-matched discretionary accrual model and 500 firm-year observations, this study finds that bankrupt firms manipulate signed accruals upwards. Further, it examines subsamples of positive and negative accruals manipulation in the lightly regulated Alternative Investment Market (AIM) and the heavily regulated Main Market. Analysis of the first subsample reveals that bankrupt firms delisted from AIM manipulate accruals upwards more than bankrupt firms delisted from the Main Market. These results suggest that managers in bankrupt AIM firms have greater incentive and ability to manipulate accruals upwards compared with managers in Main Market firms. Finally, analysis of the negative accruals manipulation subsample shows that AIM bankrupt firms manipulate accruals downwards more than Main Market bankrupt firms.

KEYWORDS:

• Accruals manipulation
• bankruptcy
• Main Market
• Alternative Investment Market (AIM)

JEL CODES:

• M41
• M48

Disclosure Statement

No potential conflict of interest was reported by the authors.

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1. García Lara et al. (2009) use Charitou et al.’s (2004) failure prediction model to divide UK bankrupt firms into two subsamples according to their probability of failure.

2. The period considered in the present study is 1997–2009, whereas that in García Lara et al.’s (2009) study is 1995–2004.

3. As a result of the legal bankruptcy processes, a time lag exists between the last audited financial statements and when firms are declared bankrupt. For example, in this study sample, the last financial statement year was 2009 for a failed firm that was officially declared bankrupt and delisted from the market in 2011.

4. Voluntary arrangement and dissolution are other failure types identified by the Insolvency Act 1986. However, they are much less frequent compared with administration, receivership and liquidation orders (Department for Business Innovation & Skills, 2011).

5. We used the LexisNexis electronic resource to distinguish between failed firms that were listed on the Main Market and on AIM.

6. Following the earnings management literature, we excluded any control group with fewer than six observations in each two-digit SIC industry code/year group (e.g. Campa & Camacho-Miñano, 2014; García Lara et al., 2009).

7. We follow a model similar to that of García Lara et al. (2009) in classifying bankrupt firms into firms with low and high probabilities of failure to compare the present results with that of the only study thus far that has examined discretionary accruals in pre-bankrupt firms and their probability of failure.

8. García Lara et al.’s (2009) study sample consists of 2,801 continuing firms and 268 bankrupt firms from 1995 to 2004. However, they did not distinguish between bankrupt firms that were listed in the Main Market from those that were listed in AIM. In addition, their matching criteria differs from those of this study, as mentioned earlier.

9. Model (1) is the first step in the performance-matched model of Kothari et al. (2005) that we applied to all UK firms. Although this step is similar to the modified June model, the purpose of this first step is to calculate estimated coefficients, which are used to estimate nondiscretionary accruals for both failed and healthy firms.

10. To consider extreme outliers, all the continuous variables are winsorised at the top 3% and bottom 97%. Consistent with Leone et al. (2012), the dependent and independent variable are winsorised, because winsorising only the dependent variables can bias the coefficients.

11. A growing body of research shows that the use of auditor size (Big 4) may reveal audit quality (see L. E. DeAngelo, 1981; Dye, 1993; Hoitash et al., 2007).

12. For example, if the last available financial report before the bankruptcy was for 2005, year −1 is 2005, year −2 is 2004 and year −3 is 2003.

13. $VIF=\frac{1}{1-R_i^2}$ is the formula for computing VIFs.

14. The expected signs of the control variables are inferred from the literature as shown in the discussion on variables.