No. of Recommendations: 12
50 stocks is a lot to manage though.
Yes and no. These days with brokers that allow uploading a spreadsheet list of tickers and one-button rebalance, it's not really so bad.
And the turnover is not very large...about 2.5 picks out of 50 are swapped in the average quarter. Since things would only rarely fall of the list due to a big drop in ROE or a cut dividend, I presume a lot of the turnover comes from selling things that have gained in price (reducing the prospective yield) to buy something else that is [still] cheap relative to the coupon.
Are the results much different if you cut to 20 stocks, drop when out of 50?
The returns are fine in backtest. About the same total return, though more jagged of course.
The problem is that you'll start to get a lot more randomness...potential hits from the inevitable bad picks. I wouldn't put 5% of my portfolio into a single quant pick, just live with a bit of typing once in a while. I think a thing like this, designed to create a portfolio you could actually live from, needs a bit less risk. I wouldn't go under 40. (so maybe top 40, hold-till-drop at rank 70)
Since I don't really buy new US stocks any more, I am working up an international version, starting from a broad list of candidates chosen with the FT Global Equity screener. It has both ROE and 5-year-average ROE, so I plan to use the average of the two. I have no way of backtesting the result, but I don't expect it to be too bad other than the fact that ex-US dividends are more variable and more cyclical than US dividends. The hard part has been making the distinction between where a stock is listed (the country filter at FT) and the actual domicile of the company, which determines things like the withholding tax rate on dividends. So I'm able to sort on the after-tax yield. Most folks live in countries with double-tax treaties, so the withheld tax shows up as a credit on your local tax return, but for me it's a dead loss.
My first cut gave these results based on Friday's data:
Average before-withholding-tax yield of 6.37%/year
Average after-withholding-tax yield of 5.21%/year
Median 5-year growth in EPS: 15.3%/year
Median ROE, five year average: 29.5%
Median market cap: 3.2 billion euros
Average earnings yield: 10.1% (equating to a P/E of 9.9).
If that were a single company, it would seem to be pretty attractive investment.
The eligible countries I picked were Europe (but not some eastern Europe), Canada, Australia, New Zealand and Japan. The big "gotcha" is the enormous number of firms from all over the world that have a secondary listing in Germany, which have to be checked and eliminated individually. The more subtle things are dual share classes, so I'm simply picking the one with the most trading volume. Once the master table is set up, redoing the ranking is pretty easy.
Jim
