A simple way to stockpicking success

The Dogs of the Dow' is a well-known investment strategy that involves buying the most beaten-up stocks in the market, betting that they'll bounce back. Now, reports Bloomberg, a paper from David M Smith, associate professor at the State University of New York, suggests the strategy can be taken global you can earn superior long-term returns by buying the world's worst-performing stock-market indices, using low-cost exchange-traded funds (ETFs). So how does this Dogs of the World' strategy work?

The logic behind the original Dogs strategy is to buy the stocks in any given index the Dow Jones, or FTSE 100, say with the highest dividend yields. A high yield suggests the stock is out of favour with investors (because when prices fall, the yield rises, assuming that the dividend payout stays the same). Why buy such battered stocks? Because markets often overreact.

When investors get fearful, they sell first and think later. By buying these unpopular stocks, you are betting their prices will rebound, returning a quick profit. You get a decent dividend thrown in, assuming it isn't cut. The strategy has a reasonably good track record. But how does it work if applied to global markets?

Taking the dogs global

There are some big differences between the Dogs of the World' and the classic Dogs of the Dow' strategy. But both are based on the principle of mean reversion' the idea that the most beaten-up stocks and markets won't stay that way forever, but will return towards fair value'.

Smith's paper looks at 45 national stock markets, as measured by the Morgan Stanley Capital International (MSCI) equity indices database. The MSCI indices represent both developed and emerging markets, although Smith notes that mean reversion tends to be strongest in emerging markets. This is because developed markets are more efficient': data on under or over-valued assets is transmitted more quickly between investors, resulting in fewer price abnormalities and faster corrections.

For 18 of the markets, data are available back to 1970. The rest become eligible for the study from the year when an MSCI index was first published for them. As with the Dogs of the Dow strategy, to make it into the selection, an index has to have been among the five worst performers over the previous year. However, while the stock-picking strategy turns the stocks over fairly rapidly, Smith's version is longer-term.

He takes a five-year holding period and assumes an investor drips 20% of their total planned investment in at the start, and 20% a year over the following four years. As Smith notes, "If the investor commits a full allocation to the five worst-performing markets from the previous year, and holds for five years, the investor forfeits the chance to take advantage of potential mean reversion using markets that perform worst in year one, year two, year three and year four."

The mechanics

Here's how it works. Taking the previous year's performance up to 31 December, you allocate 20% of your portfolio equally to the five worst-performing markets and keep 80% in cash. At the end of year one, you find the five worst-performing markets for that year and allocate another 20%, keeping 60% in cash. This continues until you are fully invested, with 0% in cash, a point that is reached at the start of year five.

At the start of year six, you then take the very first 20% allocation, and reinvest it in year five's worst performers. "In this manner the portfolio is repopulated each year on a rolling basis," notes Smith.

But does it work?

It certainly does. Between 1971 and 2012, the Dogs of the World approach produced annual compound returns of 10.39%. That is nearly 1% higher than the 9.45% return on the MSCI benchmark a significant difference.

There is one caveat: the Dogs portfolio is more volatile than a portfolio made up solely of the benchmark (this is known as a higher standard deviation of returns' in other words, the portfolio moves up and down a lot more). However, there is also a measure called the Sharpe ratio, which compares risk-adjusted returns between portfolios. Using this measure, there is little difference between the Dogs and the benchmark return. To translate that into English, the Dogs generate sufficient returns to justify the extra risk.

Now, a critic would argue that having to make a rolling investment in as many as five poorly performing markets each year is quite a fiddle, not to mention quite nerve-wracking. However, Smith re-tested the strategy using just the single worst-performing market and concludes that "investing for four to five years would have provided high returns, albeit with elevated volatility levels".

Have a punt on the dogs

The good news is that there are now ETFs available for 41 of the 45 markets reviewed by Smith (of which 24 are developed and 21 are emerging). These come with a cost in the form of a small bid-to-offer spread (the gap between the buying and selling price) and an annual management fee. Yet an investor who followed his strategy, rather than just buying and holding a broader global index, would enjoy compounded annual returns 246 basis points (2.46%) higher than the index.

If you fancy giving it a shot, based on the first quarter 2013 MSCI performance, the latest Dogs are Spain (LSE: EWP) down 6.4%; Italy (NYSE: EWI) down 9.7%; Russia (NYSE: RSX) down 3.6%; China (NYSE: PEK) down 4.5%; and Korea (NYSE: EWY) down 4.6%.