Factor investing in equity markets is one of the hot topics of these days. Many manufacturers of investment products offer exposure to small cap, momentum, minvol, value and other pure factors or factor blends. Many of them beating the cap-weighted index at relatively low cost.

I think of various reasons but I woud like to discuss: What are the reasons not everybody just invests in factor portfolios? What could be limitations, regulations, fears or other reasons not to invest in factor portfolios?

  • 3
    $\begingroup$ One limitation could be if too many people invested in the same factor portfolio that it would not behave in the desired way. I don't know of this happening just throwing a potential limitation out there... $\endgroup$ – amdopt May 8 '17 at 14:52
  • 1
    $\begingroup$ @amdopt yes, fear of crowding could be a limitation. I wonder if factors like momentum would even benefit from this ;) $\endgroup$ – Richard May 8 '17 at 15:00
  • $\begingroup$ Obviously some of them beat the index, but lots of others don't, and it's difficult to know in advance which ones. $\endgroup$ – jwg May 9 '17 at 14:18
  • $\begingroup$ @jwg in my view only 2 - 4 factors work and they have their pros and cons. I don't belief in too many factors myself. I wanted to leave the question more open. One could ask more directly about why one should not invest in my favourite factors (which I don't list) which are robust factors with big chances if outperformance. $\endgroup$ – Richard May 9 '17 at 14:25

This question goes to whether the historical returns to factors represent:

  1. Spurious results, overfitting, data mining...
  2. Mispricing
  3. Unexploitable effects
  4. Compensation for risk

Case 1: Spurious results etc...

If someone constructs a "stock tickers that begin with AAP or GOO" factor, the highly above average returns would almost certainly reflect a fishing expedition (or conditioning on future information) and would not be reproducible going forward.

Under a null of no above average returns, you're going to get portfolios that have above average historical returns with t-stats over 2. Beware.

For something like the Fama-French factor $\mathit{HML}$, this seems far less likely since it has continued to hold decades and decades after it was initially discovered. And if a factor works in other markets and asset classes, it may also increase confidence that the effect is really there. For example, Asness et. al. (2013) find value and momentum effects across a broad variety of asset classes.

Case 2: Mispricing

If a factor reflects mispricing by investors, some kind of psychological bias or error, there's the possibility that investors wisen up and the higher returns vanish!

For example, are we likely to see more egregious violations of the law of one price such as in the tech stock carveouts of Lamont and Thaler (2003)? Perhaps, but if investors get smarter, these types of anomalies should go away.

Case 3: Unexploitable effects

A related idea is that various anomalies can exist if they are unexploitable. There's a large literature on mispricing and short sale constraints.

A common question for higher turnover strategies (eg. momentum) is to what extent trading costs eat into estimated returns. For example, Novy-Marx and Velikov (2016) examine transaction costs and anomalies.

Another issue is what's the price you can actually trade at? In this blog post, Ernie Chan goes through some examples where a strategy appears to generate positive returns but actually doesn't!.

Case 4: Compensation for risk

If the factors represent compensation for risk, a risk that investors don't wish to hold, then there's a rational reason for the effect to continue. It doesn't violate economic laws of rationality for there to be positive insurance premiums and for the holders of unpleasant, aggregate risk to earn premiums for bearing that risk. The efficient market hypothesis of Eugene Fama does not imply that expected returns are constant.

This brings up a question of whether clients are capable of bearing a risk? When will they want cash? For example, a number of university endowments tried to follow the David Swensen, Yale Model and try to earn premiums for holding highly illiquid investments. When the 2008 financial crisis hit, all types of these illiquid investments in private equity, venture capital etc... became even more illiquid and ceased to pay dividends. Some major university endowments ended up issuing large bonds to raise cash...

If a university's plan in a financial crisis is to avoid cuts in programs by tapping an endowment, then investing a large portion of the endowment in illiquid securities may be problematic. If you go down to the level of tiny non-profits, their revenues (from contributions) can be highly correlated to the business cycle. Putting their cash in investments correlated with their revenue could put them at risk of disintegration in a crisis.

Declining above average returns...

Mclean and Pontiff (2016) find that many asset pricing anomalies in the literature decline in estimated magnitude post discovery and post publication. Do their results reflect investors learning (and reduced mispricing)? That some prior asset pricing anomalies were spurious or overstated? That compensation for risk has gone down? Some combination of the three?

A decline in the estimated magnitude of an effect in studies trying to replicate the results is a pervasive issue in science.


Asness, Clifford S., Tobias J. Moskowitz, and Lasse Heje Pedersen, 2013, "Value and Momentum Everywhere," Journal of Finance

Lamont, Owen A., and Richard H. Thaler, 2003, "Can the Market Add and Subtract? Mispricing in Tech Stock Carve‐outs," Journal of Political Economy

Mclean, David R. and Jeffrey Pontiff, 2016, "Does Academic Research Destroy Stock Return Predictability?" Journal of Finance

Novy-Marx, Robert, and Mihail Velikov, 2016, "A taxonomy of anomalies and their trading costs," Review of Financial Studies`


To add another perspective see this current and very relevant article with many unique and original insights (Kritzman is one of my favorite authors anyway):

Cocoma, Paula and Czasonis, Megan and Kritzman, Mark and Turkington, David, Facts About Factors (April 6, 2015). MIT Sloan Research Paper No. 5128-15. Available at SSRN: https://ssrn.com/abstract=2594485 or http://dx.doi.org/10.2139/ssrn.2594485


It has become fashionable to allocate portfolios to factors rather than to assets. The often stated motivation for this approach is that factors are less correlated with each other than assets; therefore, factors afford greater opportunity for diversification. This argument is specious, of course, because ultimately the portfolio must be invested in assets. It is, therefore, impossible to produce a better in-sample portfolio by describing the portfolio as a set of factors than assets. There are several potentially legitimate arguments, though, for favoring factor stratification over asset stratification. It could be that factors are easier to forecast than assets, because investors are better able to relate current information to future factor behavior than to future asset behavior. Unfortunately, we have no way of testing this conjecture generically. But there are several testable conjectures. Perhaps risk estimated from high-frequency returns predicts risk over longer horizons more reliably for factors than for assets. Or the statistical properties of large samples may predict the statistical properties of small samples more reliably for factors than for assets. Or, for the same sample size, the statistical properties of factors may be more stationary from one sample to the next than they are for assets. Finally, it may be that reducing the dimensionality of a large set of assets to a smaller set of factors reduces noise more effectively than reducing dimensionality to a smaller set of assets. We offer empirical evidence of the validity, or lack thereof, of these testable conjectures.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.