How, if at all, do the recommendations of modern portfolio theory vary as a function of an investor's time horizon? Can "patient" investors such as endowments, universities, sovereign wealth funds, and public and private pension plans capture extra returns by investing in relatively illiquid asset classes? Does the measurement and definition of risk differ for long- and short-horizon investors? These are among the core questions that are explored by the NBER's Project on Long Term Asset Management. With the generous support of the Norwegian Finance Initiative, the NBER has convened a series of three research meetings on this topic, bringing together researchers who are exploring the theory and practice of long-term investing.
The most recent meeting, held in New York City on May 3-4, 2018, brought together nearly 50 researchers and a small group of investment practitioners to explore some of the issues that distinguish long-term from short-horizon investing. The researchers who gathered at this meeting examined a wide range of issues, including the optimal risk allocation for endowment investors, the source and persistence of asset pricing anomalies that seem to have offered opportunities for above-market risk-adjusted returns over some historical time periods, and the role of cross-border investors in affecting market efficiency. The meeting also included a keynote address from former U.S. Secretary of the Treasury and Harvard University Professor Lawrence Summers, who addressed the implications of chronic weakness in aggregate demand for long-term asset returns.
Kelly, Pruitt, and Su propose a new modeling approach for the cross section of returns. Their method, Instrumented Principal Components Analysis (IPCA), allows for latent factors and time-varying loadings by introducing observable characteristics that instrument for the unobservable dynamic loadings. If the characteristics/expected return relationship is driven by compensation for exposure to latent risk factors, IPCA will identify the corresponding latent factors. If no such factors exist, IPCA infers that the characteristic effect is compensation without risk and allocates it to an "anomaly" intercept. Studying returns and characteristics at the stock-level, the researchers find that four IPCA factors explain the cross section of average returns significantly more accurately than existing factor models and produce characteristic-associated anomaly intercepts that are small and statistically insignificant. Furthermore, among a large collection of characteristics explored in the literature, only eight are statistically significant in the IPCA specification and are responsible for nearly 100% of the model's accuracy.
Baruch and Zhang develop a conditional capital asset pricing model (CAPM) that maintains the rationale for index investment: In equilibrium, it is optimal for nonindex investors to index. The model demonstrates that index investment is not benign. As more nonindexers become indexers, the proportion of an asset's idiosyncratic risk to total risk increases, correlation in asset prices increases, correlation in returns decreases, and for any portfolio other than the market portfolio, the Sharpe ratio decreases and the conditional variance of payoff increases. As a limiting case of the model, the researchers present an equilibrium in which all investors are indexers.
Gupta and Sachdeva document the role that inside investment plays in managerial compensation and fund performance. Merging against a comprehensive and survivor bias-free dataset of U.S. hedge funds, they find that funds with greater investment by insiders outperform funds with less "skin in the game" on a factor-adjusted basis. The researchers emphasize the role of capacity constraints in explaining this result: insider funds are smaller, are less likely to accept inflows in response to positive returns, and are more likely to be closed to outside investors. These results suggest that managers earn outsize rents by operating trading strategies further from their capacity constraints when managing their own money. Gupta and Sachdeva findings have implications for optimal portfolio allocations of institutional investors and models of delegated asset management.
This paper was distributed as Working Paper 26113, where an updated version may be available.
Poor intermediary health coincides with low asset prices and high risk premia, but it is unclear how much fluctuations in intermediaries' health matter for aggregate asset prices rather than simply being correlated with aggregate risk aversion. Haddad and Muir argue that relative predictability of more vs less intermediated asset classes by intermediary health allows to quantify how much variation in risk premia they can ascribe to intermediaries. Intermediary health should matter relatively more for assets that households are less willing to hold directly, whereas frictionless aggregate risk aversion should, if anything, exhibit the opposite pattern. The researchers provide direct empirical evidence that this is the case and hence argue that intermediaries matter for a number of key asset classes including CDS, commodities, sovereign bonds, and FX. Haddad and Muir's findings suggest that a large fraction of variation in risk premia in these asset classes is related to intermediary risk appetite.
This paper was distributed as Working Paper 28692, where an updated version may be available.
Kacperczyk, Sundaresan, and Wang study the importance of foreign institutional investors in global capital allocation. They find that stocks that are held by a higher fraction of foreign institutional investors have more informative prices both unconditionally, and conditional on the same level of domestic institutional ownership. Foreign institutions contribute to price informativeness about as much as domestic ones, especially for stocks in developed countries. The effect is robust, and is stronger for stocks held more by active investors, by foreign investors from countries with advanced financial systems, and by foreign investors from countries with looser capital controls. Finally, the researchers show that the documented effects are most likely not due to increased firm information disclosure, but rather to real efficiency gains.
This paper was distributed as Working Paper 24765, where an updated version may be available.
Lochstoer and Tetlock decompose the returns of five well-known anomalies into cash flow and discount rate news. Common patterns emerge across all factor portfolios and their mean-variance efficient combination. The main source of anomaly return variation is news about cash flows. Anomaly cash flow and discount rate components are strongly negatively correlated, and this negative correlation is driven by news about long-run cash flows. Interestingly, anomaly cash flow (discount rate) news is approximately uncorrelated with market cash flow (discount rate) news. These rich empirical patterns are useful for guiding specifications of asset pricing models and evaluating myriad theories of anomalies.
Dimmock, Wang, and Yang analyze a long-term investor's dynamic spending and asset allocation decisions by incorporating an illiquid alternative asset into an otherwise standard modern portfolio theory framework. The alternative asset has a lock-up period, but can be voluntarily liquidated or increased by paying a proportional cost prior to the lock-ups expiration. The investor benefits from liquidity diversification, which results from the alternative assets' staggered maturity dates. The quantitative results of the calibrated model match the spending and asset allocation decisions of university endowment funds, if the alternative asset earns an expected risk-adjusted net-of-fees return of 2-3% (with public equity as the benchmark).
This paper was distributed as Working Paper 25559, where an updated version may be available.
The anomalies literature is infested with widespread p-hacking. Hou, Xue, and Zhang replicate this literature by compiling a large data library with 447 anomalies. With microcaps alleviated via NYSE breakpoints and value-weighted returns, 286 anomalies (64%) including 95 out of 102 liquidity variables (93%) are insignificant at the 5% level. Imposing the t-cutoff of three raises the number of insignificance to 380 (85%). Even for the 161 significant anomalies, their magnitudes are often much lower than originally reported. Among the 161, the q-factor model leaves 115 alphas insignificant . In all, capital markets are more efficient than previously recognized.
This paper was distributed as Working Paper 23394, where an updated version may be available.
This is a test by Alex. We might add extra info here.