Aqr Case

Name: Jerome Liao

AQR Momentum Case

What is the evidence that this will be successful? Are the steps that the fund group are taking enough to make this a success? What are the reasons this might not be successful?

The first argument for the success of momentum investing is that momentum investors experienced above average returns because of the undiversifiable risk they exposed themselves. This was supported by the fact that momentum stocks showed a strong tendency to move together, therefore, momentum profits can only be achieved when one is exposed to the possibility of suffering significant losses.

A second reason why momentum is successful might be due to irrational behavior and investor biases as reflected in prices. For example, overreaction by investors to news. A company making a positive announcement that leads to a price increase results in more investors buying the company’s stock. This is driven by investors acting irrationally by assuming a momentum of good news repeating itself. Another behavioral theory explains momentum as an outcome of slow reaction to news. This theory believes that information will slowly leak into prices causing a slow growth effect over time.

The first academic paper that proved the success of the momentum strategy was published by Narasimhan Jegadeesh and Sheridan Titman (JT). JT used a prior period to define winner stocks and loser stocks, then created a portfolio that go long on winner stocks and short on loser stocks. Momentum returns were calculated as the returns to the portfolio that long decile 10 stocks (winners) and short decile 1 stocks (losers). Stocks were re-ranked each month and new portfolio was formed.

After JT published their findings in 1993, many believed that the momentum strategy will soon be ineffective as the market rushes to earn profits by exploiting it. However, Kenneth French maintained a data library online that included a data on momentum-based factor, which economist Eugene Fama called UMD (Up minus Down), momentum was derived using the highest 30th percentile of past return stocks (UP) minus the lowest 30th percentile of past returns stocks (DOWN). The average returns from the UMD data showed that UMD returns were slightly larger post-1992 period, which contradicted investors’ belief that momentum will fail once the strategy became known to the public.