Factor Alpha & International Investing: Part 2 (of 4)

Categories International, Investing, Momentum, OSAM Research, Value

(This is the second in a series of four posts that examine the efficacy of factor-based investing in the International market.)

Passive or Active?

Weighting & Concentration

Factor selection is a key determinant of portfolio performance. The final part of portfolio construction that determines how much the portfolio differs from the benchmark is how stocks are weighted. The benchmark simply market cap weights the stocks in the universe giving the highest weighting to the largest companies. We would like to determine the effect concentration and market cap weighting have on return and risk-adjusted return. To do this we start with the constituents of the MSCI EAFE Index and create portfolios based on value as tested in the previous section.

We create two versions of the strategy:

  • Version 1 sorts all stocks in the MSCI EAFE on each rebalance date by valuation and builds portfolios from 100 to all of the stocks in the benchmark. (so the 100-stock version would be the 100 cheapest stocks on that date, and so on). Positions are equally-weighted (e.g., one percent each in the 100-stock portfolio).
  • Version 2 takes the same portfolios with the same stocks (from 100 to all of the benchmarks holdings) but weights the positions according to market cap. This method can create very top heavy weightings in the more concentrated portfolios (e.g., Toyota at 12 percent of the most recent 100-stock portfolio).

The graphs below illustrate the impact of concentration and cap-weighting on both forward one-year excess returns and active share against the MSCI EAFE index. As concentration decreases so does active share and excess return. Also, market-cap weighted portfolios offer significantly less return than equal-weighted. Even in the case where both portfolios own all of the stocks in the benchmark, the effect of equal-weighting versus the benchmark adds 2.3% to excess return. The ability for active managers to be different from their benchmark is critical to achieving success.

Average Excess Return (%)
2016-09-22_01

Active Share (%)
2016-09-22_02

Alpha vs. Capacity

In order to deliver strong investment results to their clients, asset managers must shift from the frictionless world of research to a real world setting. Another important aspect of capturing alpha in the international market is managing market impact. Using the ITG cost estimates we are able to get a sense of the estimated market impact when trading a portfolio. Trading commissions are a real cost but our focus here is on market impact, which matters more for large asset managers. When a money manager trades billions of dollars this can have the effect of moving the price of the stocks you are buying or selling—you pay a higher price when buying and receive a lower price when selling than if you were trading a much smaller $1 million account.

To show the impact of asset levels on market impact, we expand our universe to all stocks trading on developed international markets with a market capitalization greater than $200 million, which will include small and mid-cap companies where market impact costs are greater. Similar, to the previous analysis we build portfolio based on valuation ranging from 100 to all stocks in the opportunity set. Positions are rebalanced on a rolling annual basis, meaning the holding period for each position is at least one year.

Value Portfolios:

Equal-weighted
2016-09-22_03

Cap-weighted
2016-09-22_04

The market impact estimates are based on the 5-year period from 2010 to 2015. ITG cost curves are based on actual market conditions over this time frame. This table illustrates the paradigm of alpha against scale. We’ve shaded boxes denoting the points where impact (annualized) exceeds one percent. Over the various concentration levels, the average forward 1-year outperformance of the equal-weighted versus cap-weighted portfolios is approximately 2.5% (see “Average Excess Return” above). The concentrated portfolios, while providing greater alpha, cannot accommodate the scale that smart beta or other large asset managers are looking for. For example, at an asset level of $5.0 billion the 200 stock equal-weighted portfolio has a market impact cost of 1.10% versus 0.41% for the market cap weighted version. The key to achieving excess return is to invest in concentrated equal-weighted portfolios that select stocks based on proven themes—the caveat of course is you must be willing to accept lower asset levels.

Go to Part 3