3. 3-Year EPS Growth

Description of Factor:

The 3 year estimated EPS growth rate is the consensus analyst estimate of the average annual growth rate in earnings per share over the next three years. The rationale behind this variable is the expectation that stocks that have high estimated growth rates will outperform stocks with low estimated growth rates. In contrast to the 1 Year EPS Growth factor, we estimated that a three year growth rate would be a long term, more sustainable indicator of a stock’s future performance. From an investment strategy standpoint, this would imply a growth focused strategy rather than a value focused strategy.

Analysis

Equal Weighted: As shown in the graph below, the quintiles have a relatively flat distribution of annualized average returns. Quintile 1 generates the highest return of 18.67% and quintile 4 the lowest with 15.62%. The return of quintile 5 is only marginally higher than that of quintile 4 with 15.98%. The standard deviation for quintile 1 is 17.63% while it numbers 12.54% for quintile 5. The average excess return of quintile 1 is (0.25)% and (3.08)% for quintile 5. The beta for quintile 1 is 1.13 with an R² of 70%, while the beta for quintile 5 is 0.81 with an R² of 71%.

Based on the distribution of the annualized average returns among quintiles, a long-short strategy is plausible. However, the spread of annualized average returns between quintile 1 and 5 is small, less than 300 basis points. This alone makes this strategy challenging. Additionally, the annualized average return for the highest returning quintile, quintile 1, is below the annualized average return of the market. This means that on average, a long-short strategy using the 3 year EPS growth rate would under perform the market. Also, quintile 1 yields the highest return only six out of the sixteen years surveyed. Quintile 5 is the worst performing portfolio in only four out of the sixteen years, while it is the best performing portfolio three years. Similarly, quintile 1 is the worst performing portfolio in four years. Hence, implementing a long-short strategy would be disastrous since your portfolios do not perform consistently. The monthly turnover ratios are relatively low, 12.52% for quintile 1 and 12.43% for quintile 5.

Value Weighted: Although the distribution of the annualized average returns does not follow a step function either, it is better than that of the equal weighted portfolios in as far as that the spread between returns of quintile 1 and quintile 5 is greater. The annualized average return for quintile 1 is the highest with 20.23% and the return for quintile 5 is the lowest with 13.49%. The standard deviation for quintile 1 is 16.26% while it numbers 12.41% for quintile 5. The average excess return of quintile 1 is 1.09% and (5.20)% for quintile 5. The beta for quintile 1 is 1.14, while the beta for quintile 5 is 0.80.

Compared to the equal weighted portfolio, the value weighted portfolio is better suited for the implementation of a long-short strategy. However, the excess return over the market, although positive for quintile 1, is too low with only 1.09% above the market to compensate for the risk taken. A closer inspection of the actual rankings of the portfolios illustrates this fact. Quintile 1 yields the highest return in only four of the sixteen years sampled while it generates the lowest return in six of the years. Quintile 5 is the worst performing portfolio in five years, but it is the best performing portfolio in four out of the sixteen years.

Conclusion:

From our analysis above, we conclude that the 3 Year EPS Growth factor is not suited to implements a long-short strategy. The distribution of annualized average returns among the five quintiles is sufficient from a directional standpoint to be feasible. However, the spread of these returns between the highest and lowest ranking quintiles is too low to base a strategy upon. In addition, the lack of consistency in the rankings of the quintiles further diminishes the factor’s predictive power.