Searching for seasonal inspiration

There are tell-tale market signals at this time of year … but it’s still a matter of timing.

Summary: In the United States, research has shown that the best period to be in shares is between November and April. A rally into Christmas, and through the New Year, is also helpful in predicting the direction of the market longer term. But the November-April buying system doesn’t appear to work as well in Australia.
Key take-out: The best way to minimise general market risk is to market time an exchange-traded fund using a trend-following system all-year round.
Key beneficiaries: General investors. Category: Strategy.

Jeffery Hirsch, editor-in-chief of the Stock Trader’s Almanac, contends that while stockmarkets can’t be predicted, they nevertheless exhibit well-established cyclical patterns. These can’t be dogmatically applied to timing the market, but they can prove a useful guide.

Here are three important takes from his book, Stock Market Cycles:

1. Sell in May and go away, come back on St Ledgers Day (late October)

The best six months of the year to be in shares are November to April and the worst are May to October.

According to Hirsch, a profitable switching strategy is to buy shares at the start of November, hold them until the end of April and then switch to bonds or money market funds or short stocks for the six months to October.

On Hirsh’s figures, if someone at the start of 1950 had invested $10,000 in a portfolio of stocks representing the Dow Jones Industrial Average (DJIA) and then followed this switching strategy their net worth would have grown to $674,073 by the end of 2012. By contrast, if they had followed the opposite switching strategy and punted on the worst six months of the year, their net worth 62 years later would have been minus $1,024.

Independent academic studies (see http://www.streetsmartreport.com/comm4) have found such a switching strategy works in most markets around the world.

2. Santa Clause rally

Hirsch found that over the last 60 years, when there was no Santa Claus rally (last five trading days of the year and first two days of New Year), there was a two-thirds chance the year ahead would be flat or negative.

3. As January goes, so goes the year

When the first five trading days of the New Year recorded a positive gain, then in 85% of cases the rest of the year was also positive. Also, for four years out of five since 1950, as January went so went the year. All down Januaries were followed by a new or continuing bear market, a correction or a flat year.

When Santa came to town, the first five days were up and the January barometer was positive, then 92% of the time the year boomed with an average gain of 17.5%.

When the share index’s low in the March quarter was above its low of the December quarter, the year ahead almost always rallied.

Hirsch’s risk-averse approach to shares

Hirsh concludes his cyclical analysis by reminding us that since the creation of exchange-traded funds in 1993 ordinary investors have been able to trade the S&P 500 index (or in Australia’s case the S&P/ASX 200 index) without resorting to expensive futures contracts, risky index options or complex derivative instruments.

He vows his six-months strategy of switching between a single equity ETF (such as SPDR S&P/ASX 200 Fund, Vanguard Australian Shares Index ETF, iShares MSCI Australia 200, or SPDR S&P/ASX 50 Fund in Australia) and a money market account will “steadily build wealth over time with half the risk (or less) of a buy and hold approach”. In other words, don’t risk your money on shares between May and October.

As a further precaution, during the best six months he advocates holding shares (via a single equity ETF) only when the stock index is trading above its trend line (as measured by its exponential moving average) and exhibits positive momentum (as gauged by its Moving Average Convergence-Divergence – MACD – indicator). These technical indicators are available on any charting platform. One example using a 50 day moving average and a longer term MACD histogram is shown below.

Does Hirsch’s six-month switching strategy work in Australia?

Selling at the start of May and buying at the start of November would have been a winning strategy in Australia in only six of the 14 half-yearly periods from November 1, 2006 to November 1, 2013. See the next chart where the vertical blue bars represent buy signals and the vertical red bars represent sell signals. Nevertheless, such a switching strategy would still have done better than the All Ordinaries index over this volatile period – a capital gain of 9.8% versus 4.3%.

By contrast, a simple trend-trading system based on a 50-day by 200-day moving average crossover would have won on five of its nine signals over this seven-year period and produced a total gain of 18.5%. Again, the vertical blue and red bars represent buy and sell signals respectively.

Also Hirsch’s evidence is purely US-based. One Australian study spanning 1985-2012 found that the best six months locally are December to May and the worst June to November (click here). Another local study found the best months between 1992 and 2011 were December and April and the worst were June and September (click here). See next chart.

Conclusion

In summary, Hirsch’s advice is:

  1. Avoid holding shares between May and October because that is when most corrections and crashes occur;
  2. If there’s a Santa Clause rally, a First Five Day rally and a January rally, then it’s normally safe to be in shares for the rest of the year; and
  3. As an extra safeguard, market time the best six months (November to April) by applying a trend-following system to a listed equity fund.

I agree with Hirsch that the best way to minimise specific stock risk is to hold a diverse portfolio of shares via a low-cost, highly liquid ETF (such as the SPDR S&P/ASX 200 Fund).

But where I part company with him is that I think the best way to minimise general market risk is to market time the ETF using a trend-following system all-year round.

In other words, buy the ETF when the market is trending up and sell it when the market is trending down. That produces better results than a crude six-month switching strategy based on a typical in-year pattern, because not every year is typical.


Percy Allan is a director of MarketTiming.com.au For a free three week trial of its newsletter and trend-trading strategies for listed ETF funds, see www.markettiming.com.au.