The authors examine whether investment banks add or subtract value compared with nonbank financial conglomerates for three types of investment portfolios: mutual funds, hedge funds, and institutional funds. They find that investment bank ownership reduces alpha by 46 bps a year, regardless of the fund type.

Conflicts of interest can cause harm if not controlled by various mechanisms. The authors address the question of how effectively conflicts of interest in asset management are controlled for in investment banks versus nonbank financial services groups. They look at alphas for three types of delegated portfolios—mutual funds, hedge funds, and institutional funds—for the two types of ownership and conclude that ownership does matter.

In the sample they study, investment bank ownership reduces alphas by 46 bps per year, with the loss being derived largely from the dispersion of fees among portfolios. But participation in lead loans on the part of the investment bank increases alphas. The form of the contract—mutual fund, hedge fund, or institutional fund—does not seem to matter once control variables are taken into consideration; competition appears to even out returns for the three contract forms over time.

The reduced alphas related to investment bank ownership add up to $93 billion for the 19-year period (1990–2008) studied by the authors. For 14 of the 19 years in the sample, the costs of investment bank ownership were higher than the benefits.

The authors compare investment performance of three types of delegated portfolios: mutual funds, hedge funds, and institutional funds. Their sample is composed of all financial groups—investment banks and nonbank financial services conglomerates—that managed all three types of portfolios at the same time for at least one year during the 1990–2008 sample period and reported performance data to readily available databases. The sample includes 23 investment banks and 48 nonbank financial services conglomerates. To control for the effect of omitted variables, they limit the comparison to financial groups.

The authors estimate alphas on unsmoothed returns using the moving average process, accounting for differences in portfolio exposure to various risk factors by using a seven-factor model with time-varying alphas. They examine the hypothesis that investment banks produce different alphas compared with nonbank conglomerates by looking at the cross-sectional regression of fund alphas on control variables and type of organization.

Competition appears to equalize the impact of the three contract forms across time: The form of the contract offered to investors is not found to be a statistically significant control variable. But ownership does seem to matter in the authors’ analysis. They find that, on average, investors receive an alpha that is lower by 46 bps a year when a fund is offered by an investment bank rather than by a nonbank financial conglomerate. The effect amounts to a $4.9 billion loss per year over the 19-year period, varying by time: In five of the years, investment banks add value to their funds, whereas in 14 years, they subtract value. Mutual fund portfolios appear to suffer a large part of the loss.

Given the size of the economic loss estimated by the authors, their research seems worthy of attention. In addition to the evidence that ownership does indeed affect alphas, the finding that the type of contract offered to investors does not make a difference in performance is noteworthy. The sample size is, of necessity, small, and the existence of several years during which investment banks actually add rather than subtract value raises intriguing questions and opportunities for further research.