Top-Down Portfolio Construction vs. Selling Stocks That Drop

Includes: BAC, JPM, MYTAY, TAR
by: Roger Nusbaum

Sometimes I get emails from Motley Fool that promote articles on their site. I actually read one of the articles; it was called Why You Should Sell. The title hooked me in.

In the first part of the article they cite a study that concluded that people hold onto stocks that they no longer think are good buys. They call this 'realization utility'.

The article then goes on to say you should always sell when you have a better place to put your money -- and today, a host of superior companies are on sale.

There is something to this, but from where I sit the article is woefully incomplete and strikes me as reactive, not proactive. If you believe in top-down portfolio construction then you believe the stock chosen is the least important part of the process (not unimportant, but less important than defense, or not ahead of selection of sector, country, size, style, volatility, yield). In top-down, most stocks included in the portfolio offer several attributes, for example a large telecom stock from an emerging market that adds volatility. If a stock like that fits in, then you would seek out what you think is the best way to capture this.

Maybe you did this exercise three years ago and came up with Telefonica de Argentina (TAR). Then maybe after a couple of years you decided that Argentina was not a good place to stay, but you felt from the top down that you wanted to capture some emerging market exposure through the telecom sector. So you then needed to find a better proxy for emerging market telecom, and maybe that led you to Magyar Telekom (MTA), the Ma Bell of Hungary. If TAR was the best proxy but then MTA became a better proxy, then it would make sense to swap, regardless of when (remember, the assumption is you still want emerging market telecom).

I would caution against selling a stock simply because it is down; that is not enough information. In the last two years JPMorgan (NYSE:JPM) is down about 52% while the Financial Sector SPDR (NYSEARCA:XLF) is down about 75%. JPM appears to be one of the healthier banks and while in hindsight selling at $53 on May 9, 2007 would have been better than holding it, a case could be made for JPM being the healthiest proxy for US financials. Selling JPM now to go after some better mousetrap is probably a bad idea. Obviously if you had reason to think that from here JPM would stop being "the healthiest proxy," then you might want to sell.

This is subtle stuff. In sticking with financials, the best thing would have been the top-down decision to reduce the exposure when the yield curve started doing funky things. From the top down, reducing financial exposure two years ago becomes far more important than whether JPM is or is not a healthy proxy.

If you use ETFs and go narrower than SPY / EFA / IWM, chances are that whatever ETF you think is the best way to capture utilities still has the same attributes you thought it did when you bought. Swapping from one utilities ETF to another (XLU to PUI for example) is unlikely to change the portfolio.

This topic can get more granular with increasing or decreasing position size (as I believe in doing) and of course when you buy a stock, you may be wrong about it being the best proxy. Everyone gets this wrong at some frequency either because the conclusion was wrong or maybe because at some point something at the company changes. Bank of America (NYSE:BAC) was a great proxy for financials until, in my opinion, the Merrill Lynch merger. The company took action that changed the story.