with Andrew Metrick, Yale School of Management

DOWNLOAD COMPLETE ARTICLE (PDF)

ABSTRACT This paper analyzes the economics of the private equity industry using a novel model and dataset.  We obtain data from a large investor in private equity funds, with detailed records on 238 funds raised between 1993 and 2006. We build a model to estimate the expected revenue to managers as a function of their investor contracts, and we test how this estimated revenue varies across the characteristics of our sample funds. Among our sample funds, about two-thirds of expected revenue comes from fixed-revenue components that are not sensitive to performance. We find sharp differences between venture capital (VC) and buyout (BO) funds.  BO managers build on their prior experience by increasing the size of their funds faster than VC managers do.  This leads to significantly higher revenue per partner and per professional in later BO funds. The results suggest that the BO business is more scalable than the VC business, and that past success has a differential impact on the terms of their future funds.

With Massimo Massa and Lei Zhang, INSEAD

DOWNLOAD ARTICLE (PDF)

ABSTRACT We examine the effects of the capital supply uncertainty (CSU) of institutional bond investors (e.g., mutual funds and insurance companies) on the leverage of the firm using a novel dataset. Our main finding is that the withdrawal risk (or stability) of the firm's bond investor base (as measured by (i) the average portfolio turnover of investors holding the firm's bonds, or (ii) the prevalence of mutual funds among the firm's bondholders as opposed to insurance companies) has a negative and significant effect on the leverage of the firm. The CSU also has a negative and significant effect on the firm's probability of issuing bonds, and a positive and significant effect on the firm's probability of issuing equity and borrowing from banks. These results are robust to controlling for the potential endogeneity of CSU. Taken together, the findings suggest that the capital supply uncertainty arising from institutional investors' withdrawal risk significantly affects the firm's capital structure.

with Lily H. Fang, INSEAD

DOWNLOAD COMPLETE ARTICLE (PDF)

ABSTRACT We examine the relation between analyst reputation and their recommendation values. We measure personal reputation using the Institutional Investor All-American (AA) awards. Daily-rebalanced portfolios of AA analysts' recommendations yield significantly better risk-adjusted performance than those of non-AA analysts.  AAs' alphas are concentrated in the top-ranked (1st- and 2nd-place) AAs; the lower-ranked (3rd-place and runner-up) AAs do not perform better than non-AAs.  Alphas for all analyst groups fall as investors¹ access to recommendation information is delayed.  But the assessed performance differential between the top-ranked AAs and other analysts is only marginally larger for investors who have private, pre-release access to recommendations than for public investors. The top-ranked AAs are also more experienced and have longer tenure as AAs than the lower-ranked AAs.  Our results hold for both tech and non-tech sectors, are moderately strengthened by trading-cost adjustments, and are robust to alternative portfolio construction methods.   We conclude that the top-ranked AAs are more skilled and produce more valuable information than other analysts.  While the institution-granted AA status is a noisy indicator of analyst skill overall, the top two ranks of this award are useful signals from which public investors can benefit.

with Lily H. Fang, INSEAD

DOWNLOAD COMPLETE ARTICLE (PDF)

ABSTRACT Using 1983-2002 U.S. data, we examine whether the quality differentials in earnings forecasts between reputable and non-reputable analysts vary as the severity of conflicts of interest varies. We measure personal reputation using the Institutional Investor All-American (AA) awards, and bank reputation using Carter-Manaster ranks. While both personal reputation and bank reputation are associated with higher-quality forecasts overall, their effectiveness against conflicts of interest differs. The severity of conflicts (proxied by the aggregate volume of new equity issues) has a negative and significant effect on the performance of non-AAs at top-tier banks relative to both AAs at top-tier banks and non-AAs at lower-tier banks. In contrast, the severity of conflicts has a positive and significant effect on the performance of AAs at top-tier banks relative to both non-AAs at top-tier banks and AAs at lower-tier banks. These findings suggest that personal reputation is an effective disciplinary device against conflicts of interest, while bank reputation alone is not.

DOWNLOAD COMPLETE ARTICLE (PDF)

ABSTRACT This paper examines the effects of bank relationships on underwriter choice in the Japanese corporate-bond market following the 1993 deregulation. Bank relationships have significant positive effects on a firm's underwriter choice. Relationship firms receive a small but significant fee discount and, consistent with mitigating effect of bank competition on holdup cost, multiple-relationship firms receive a significantly deeper discount than solo-relationship firms. Bank shareholding alone negatively affects underwriter choice, whereas shareholding together with loans have significantly more positive effects than loans alone. Finally, existing relationships reduce a Japanese firm's switching probability by 32%, in contrast to only 6% for U.S. firms.

DOWNLOAD COMPLETE ARTICLE (PDF)

ABSTRACT This paper studies the effect of bank relationships on underwriter choice in the U.S. corporate-bond underwriting market following the 1989 commercial-bank entry. I find that bank relationships have positive and significant effects on a firm’s underwriter choice, over and above their effects on fees. This result is sharply stronger for junk- bond issuers and first-time issuers. I also find that there is a significant fee discount when there are relationships between firms and commercial banks. Finally, I find that serving as arranger of past loan transactions has the strongest effect on underwriter choice, whereas serving merely as participant has no effect.