Corporate Governance

Corporate Fraud and Business Conditions: Evidence from IPOs

By Harvard Law School Program on Corporate Governance

In our paper Corporate Fraud and Business Conditions: Evidence from IPOs, which is forthcoming in the Journal of Finance, we use a sample of firms that went public between 1995 and 2005 to test a set of theories modeling how a firm?s incentive to commit fraud when raising external capital varies with investor beliefs. Instead of a strictly increasing relationship between investor beliefs and fraud propensity as highlighted in Hertzberg (2005), we find evidence more consistent with the predictions of Povel, Singh, and Winton (2007): a firm is more likely to commit fraud when investors are more optimistic about the firm?s industry?s prospects, but in the presence of extreme investor optimism, the probability of fraud becomes lower as the firm is able to obtain funding without misrepresenting information to outside investors.

Further analysis suggests that both investor monitoring and executive pay structure play a role in the relationship between investor beliefs and fraud. Using venture capitalists as specialized investors with lower monitoring costs than other institutional investors, we find evidence supporting the prediction of Povel et al. Fraud is less likely for low investor beliefs but more likely for high investor beliefs for firms backed by venture capitalists than non-VC-backed firms, and for firms backed by venture capitalists of a higher level of industry expertise. Also, investor beliefs about business conditions have a positive impact on short-term compensation, which in turn has a positive impact on a firm?s fraud propensity, consistent with the predictions in Hertzberg. Nevertheless, the level of investor beliefs continues to have an independent, hump-shaped impact on the incidence of fraud even after controlling for executive compensation. This suggests that the mechanisms in both Hertzberg and Povel et al. are relevant for IPO fraud.

We also find that the monitoring incentives of underwriters differ from those of venture capitalists. Lower underwriter monitoring costs reduce fraud for all levels of investor beliefs about business conditions, though more so for low beliefs; thus, underwriters? monitoring choices appear to be more concerned with preventing fraud per se so as to protect their reputations. We interpret this as evidence in support of Sherman (1999).

Our findings suggest that the monitoring mechanism modeled in Povel et al. (2007) help better understand the effect of investor beliefs on firms? fraud propensity, and thus have implications for regulators and auditors concerned with rooting out fraud. As we noted before, corporate fraud is likely to have negative externalities, particularly in the IPO market; widespread fraud can turn investors off from IPOs, depriving young firms of a critical source of funding. Although some have argued that it should be up to investors to prevent fraud, our findings support Povel et al. (2007)?s argument that investors are focused on finding good investments rather than preventing fraud per se. Since fraud seems to peak in relatively good times, and even underwriter expertise is least effective in preventing fraud in such times, this suggests that regulators and auditors should be especially vigilant during booms.

The full paper is available for download here.

Full post as published by The Harvard Law School Corporate Governance Blog on February 18, 2010 (boomark / email).