How Is This Research Useful to Practitioners?
The author contends that because of a lack of liquidity within certain asset classes and higher perceived returns from those asset classes, investment professionals may be inclined to recommend a greater portfolio allocation to these private assets than may be justified. He finds that once the true amount of systemic and liquidity risk is accounted for, the resulting alpha provided by asset managers declines significantly. Using a lagged beta approach results in a better estimate of systemic risk, liquidity risk, and the associated alpha.
Investment professionals should be fully aware of the key drivers of return for each type of investment. Having this detailed level of understanding will help them better detect both risks and opportunities in the valuation process. Moreover, they will be in a better position to access and evaluate the historical and prospective returns of such illiquid investments. Furthermore, when evaluating the performance of investment professionals, whether internal or external, the evaluator should be aware of the findings within this study, particularly as they relate to overstated alpha. A true assessment of alpha generation will enable the evaluator to assess performance within each asset class as well as compared with that of other asset classes.
How Did the Author Conduct This Research?
Relying on prior research and considering the methods used within prior liquidity studies, the author focuses on such private asset classes as private equity, venture capital, and real estate. The historical data used for this research originate from Cambridge Associates, which collects data related to private equity and venture capital returns from foundations, pension funds, and endowments. To determine historical liquidity premiums, business development company (BDC) data are used—specifically, mezzanine debt data because duration risk and credit risk can be quantified and removed, thereby isolating the liquidity premium. Thirteen such BDCs are identified to allow for this liquidity premium analysis.
The author also tests whether private asset values lagged behind public markets. He uses various statistical techniques, including regression analysis and t-statistics, to test and support the findings; he also tests for serial correlation within the liquidity premium data. Once the full amount of market risk is identified and quantified, the author uses the traditional capital asset pricing model to isolate performance. To capture the lagged beta phenomenon, he uses a lagged beta four-factor model—that is, market beta, small-cap factor, value factor, and momentum.