This paper investigates the effect of venture capitalist (VC) quality on earnings management in firms conducting initial public offerings of their equity stock, focusing on manipulation of both accruals and real activities. I develop a measure of VC quality based on a principal components factor analysis using data that are obtainable for virtually all VC firms. This metric is highly correlated with VC funds’ financial returns, and with the likelihood of successful exits through initial public offerings or trade sales. After going public, companies backed by higher quality VCs have lower abnormal accruals, lower earnings management through real activities manipulation, and a lower likelihood of financial restatement. Companies backed by top-quartile VCs do not appear to engage in real activities manipulation as a substitute for accruals manipulation. As for companies backed by lower-tier VCs, their earnings management behaviors are indistinguishable from those of non-VC-backed companies. The results continue to hold when controlling for endogeneity. Overall, the results suggest that higher quality VCs are better able to constrain opportunistic financial reporting by their portfolio companies going public.

The returns that institutional investors realize from private equity investments differ dramatically across institutions. Using detailed and hitherto unexplored records of fund investors and performance, we document large heterogeneity in the performance of different classes of limited partners. In particular, endowments’ annual returns are nearly 21% greater than average. Funds selected by banks lag sharply. Analysis of reinvestment decisions suggests that endowments (and to a lesser extent, public pension funds) are better than other investors at predicting whether a follow-on fund will have high returns. We find that the results are not primarily due to endowments’ greater access to established funds, since they also hold for young or under-subscribed funds. Our results suggest that limited partners vary in their level of sophistication and also their objectives.

This paper examines the impact of stock market misvaluation on corporate bankruptcy prediction. Prior research has documented that market-based variables enhance the accuracy of hazard models for bankruptcy prediction. However, when market prices deviate from fundamentals, the predictive accuracy of models using these market-based variables may be affected. We document three main sets of findings: (1) the difference in predictive accuracy between properly-valued and misvalued firms is greater for market-based hazard models than for accounting-based hazard models; (2) the forecasting accuracy of market-based hazard models applied to misvalued firms improves significantly if forecasting variable coefficients are allowed to vary in the presence of misvaluation; (3) the difference in predictive performance of accounting-based and market-based hazard models is greater for misvalued than for properly-valued firm-quarters. In additional tests, we find that the predictive accuracy of structural bankruptcy risk models is not affected by misvaluation to the same extent as reduced form hazard models. Our results suggest that users should modify reduced form hazard models used for bankruptcy prediction to account for misvaluation, and confirm the impact of misvaluation on the quality of market signals.

We investigate the effect of shareholder activism on debtholders. We examine a sample of bank loans taken out by firms targeted by hedge fund shareholder activists between 1995 and 2008. We find that, after the arrival of hedge fund activists, banks charge higher spreads when the activist’s objective is to force the target into being acquired or to increase payouts to shareholders, and lower spreads when the activist’s objective is to block a takeover of the target or to oust an entrenched CEO. We also find that the proportion of general covenants to total covenants increases when the target is likely to be sold after entry of the activist shareholder. Overall, the results suggest that activist hedge fund intervention affects the risk faced by lending banks. However, unlike prior studies, we show that the effects vary with the nature of actions undertaken by the shareholder activist.

We examine the relationship between shareholder activism and managerial quality at firms targeted by activists. We decompose variations in different measures of firm performance into firm and manager components and use principal component analysis of the manager fixed effects to construct a measure of managerial ability. We find that this measure of CEO quality is a significant predictor of shareholder activism. In addition, conditional on a firm being targeted, we also find that the likelihood of the activist’s demand being control-related is higher for firms with lower quality CEOs. Finally, hedge fund activists appear no different than other activists in their sensitivity to CEO quality when making control-related demands, but hedge funds are more sensitive to their percentage shareholding in the target firm when making control-related demands.