New Developments in Long-Term Asset Management

Second Annual Conference
May 19-20, 2017

Efficiently Inefficient Markets for Assets

and Asset Management

Nicolae B. G├órleanu of the University of California, Berkeley, and Lasse H. Pedersen of the Copenhagen Business School study the interaction between securities markets and markets for asset management, addressing questions such as which investors should be active and which passive, which managers are more likely to be successful, how asset management fees are set, and which markets are more likely to be efficient.   [Download the paper]

Other Conference Papers

Institutional Investors and Information Acquisition, Matthijs Breugem and Adrian Buss

The Relevance of Broker Networks for Information Diffusion in the Stock Market, Marco Di Maggio, Francesco A. Franzoni, Amir Kermani, and Carlo Sommavilla

Asset Insulators, Gabriel Chodorow-Reich, Andra C. Ghent, and Valentin Haddad

Do Institutional Incentives Distort Asset Prices? Anton Lines

Chasing Private Information, Marcin T. Kacperczyk and Emiliano Pagnotta

ETF Arbitrage under Liquidity Mismatch, Kevin Pan and Yao Zeng

Replicating Private Equity, Erik Stafford

< 2016 Conference Papers>

Their key observation is that identifying a good manager requires considerable resources, just as identifying good securities does. They argue that this is a highly relevant friction; for instance, most institutional investors have staff dedicated to continuously evaluating and interacting with potential asset managers. This departure from the literature results in natural justifications for a number of phenomena in the realm of asset management that otherwise are quite puzzling.

Consider the conventional view that securities markets are fully efficient, meaning that security prices reflect all available information. This implies that the markets for asset management are deeply inefficient: With fully efficient securities markets, asset managers can add no value, so the billions of dollars paid to managers are a complete waste. The researchers argue that a more plausible view is that both markets — for securities and for asset managers — display an equilibrium degree of inefficiency, or, in other words, are "efficiently inefficient." Thus, securities markets are inefficient enough that asset managers have an incentive to collect information about securities, but efficient enough that beating the market is difficult. Likewise, markets for asset management are inefficient enough that investors have an incentive to search for good managers, but efficient enough that this requires resources.

The researchers' model determines the equilibrium asset management fee, linking it to the degree of securities market inefficiency. Whereas the literature has traditionally argued about whether the market is efficient or not, these researchers argue instead for estimating the magnitude of inefficiency. For example, if the securities market were fully efficient, their model predicts that the asset management fee should be zero. Hence, the positive asset management fees observed suggest that investors believe that some managers can add value. More specifically, they estimate that the observed asset management fees of around 1 percent imply that securities markets must be approximately 6 percent inefficient.

The economic mechanism analyzed by the researchers has clear implications for who should be active and who benefits more from passive investing. Since the cost of learning about managers does not scale with an investor's capital, unlike the potential benefits of investing with an informed manager, the model implies that larger investors should be more inclined to incur the cost of finding and vetting informed active managers. For smaller investors, on the other hand, spending significant resources on locating a good manager is not economical, so they are better served by minimizing costs through passive investing.


The supposition that the set of investors consists of a group whose members follow this advice — search for a manager only if you are large enough — and a set of "noise allocators" who pick random managers yields several predictions for manager performance. For example, managers whose clientele consists of larger and more sophisticated investors should outperform the average. This is because larger investors are more likely to have vetted the manager to ensure that it has a process in place to analyze securities, manage risk, and so forth, whereas smaller investors are more likely to be noise allocators. The model's predictions are consistent with empirical studies. For instance, it has been documented that institutional investors select managers who outperform those of retail investors, and large institutional investors select managers who perform particularly well.

Finally, the model predicts that the good managers should outperform even after fees in order for searching investors to at least recoup the search cost. The model can therefore account for the recent findings that it is possible to identify managers that "beat the market" after fees, while the average active manager does not. Tellingly, outperformance is larger in markets where identifying a good manager is more difficult.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us