AlphaLetters Weekly Paper Review #4 & #5

Quant working paper reviews from the guys at AlphaLetters (readers of the blog should recognize the first one!):

Article 3.
Category: Strategy, payout yield
Title: On the Importance of Measuring Payout Yield: Implications for Empirical Asset Pricing
Author: Jacob Boudoukh, Roni Michaely, Matthew Richardson, Michael R. Roberts
Source: SSRNworking paper

This paper tries to improve the traditional dividend payout yield by using total payout ratio (dividends plus repurchases) and net payout ratio (dividends plus repurchases minus issuances). It shows that these two new ratios have significant predictability of stock returns.

A portfolio of long(short) high(low) net payout ratio shows an annual return of 4.4%, higher than the strategies based on payout portfolio(3.3%), and dividend yield(2.1%).

AlphaLetters comments:
1. Why important
The total payout measure seems to be a logical candidate to improve the payout factor. After all, dividend and repurchases are both ways to return cash to investors, and the importance of repurchases has gone up dramatically these years. Before the recent IRS (US tax authority) change on dividends, from a tax perspective, firms should prefer to use repurchases as opposed to dividend, which are generally taxed at a higher rate.

2. Data
Data (84/07-03/12) are from CRSP and Compustat. Financial stocks are excluded.

3. Discussion
How would the recent change on dividend tax rate (“Jobs and Growth Tax Relief Reconciliation Act of 2003”) change the picture? A study of the recent performances of these two payout ratios should be very interesting.

Can we make money based on this new factor? What’s the correlation with a classic p/b value strategy? A look at figure 3 in the paper tells us that the correlation is certainly not negligible. The 4.4% annual profit seems not an entirely free lunch as it’s just raw return without adjusting for Fama -French factors.


Article 4.
Category: Strategy, short-term event
Title: Releases of Previously Published Information Move Aggregate
Author: Thomas Gilberty, Shimon Koganz, Lars Lochstoerx, Ataman
Source: CMU working paper

Replicate the U.S. Leading Economic Index (LEI) based on publicly available data and procedure. Long or short the stock market index depending on the change of the LEI.

Note that this is a very short-term, minute level strategy.

AlphaLetters comments:
1. Why important
We find this paper very interesting because it apparently shows that the stock market is not efficient, and stock investors collectively do not understand and process every piece of information as they become available. Consequently, an index that is based on “old” information is perceived as new information and moves market.

The authors show that change in the LEI index is positively associated with realized contemporaneous market returns, higher volatility and volume. The fact that post-announcement prices tend to revert shows that some traders are already taking advantage of this phenomenon.

2. Data
The LEI data is from the Conference Board. The S&P500 future price is from The crosssectional analysis uses individual stock transactions data from the TAQ database.

3. Discussions
We doubt that the capacity of this strategy is limited, given its very short-term nature and the fact that the LEI index is available for limited times each year.

The authors claim that stocks with higher sensitivity to macro economic factors respond less to the release of the LEI. This maybe because this is a short-term strategy, and the sensitivity to macro factors is measured based on long term basis. The seemingly discrepancy here should not be very surprising.