Investing in the right stock market sectors can make a big difference to your long-term returns. How big? Well, over the last 115 years, an investment in the tobacco sector would have returned more than 5,000 times as much money as an equivalent investment in the shipping industry, according to this year's Credit Suisse Global Investment Returns Yearbook.

Can investors predict the best performers in advance? Is there a way of profitably rotating in and out of different sectors? Should investors, mindful of an ever-changing world, prioritise new and growing sectors and steer clear of industries they believe to be going the way of the dodo?

Dramatic changes

Certainly, history shows that stock markets change dramatically over time. In 1900, railway companies accounted for almost two-thirds of the total US market, and for almost half of the UK market, compared to less than 1 per cent today. More than 80 per cent of the total US market was in industries that are now either small or extinct, according to the aforementioned Credit Suisse yearbook.

They have been replaced by industries that were either tiny or non-existent in 1900: oil and gas, technology, pharmaceuticals and biotechnology, telecoms, and media. Picking the wrong sectors can be costly. Since 1900, the average annual difference between the best- and worst-performing sectors has been more than 100 per cent. Two-thirds of sectors have underperformed the overall market over that time period.

New sectors

One might assume that the big winners are to be found in new and emerging sectors. However, it’s not always obvious as to who will be tomorrow’s big winners, which is why new technologies tend to be initially viewed with scepticism.

In 1825, for example, the Quarterly Review scoffed that there was nothing "more palpably absurd than the prospect of locomotives travelling twice as fast as stagecoaches". In 1909, Scientific American argued that the automobile "has practically reached the limit of its development". In 1977, Digital Equipment Corporation founder Ken Olsen scoffed that "there is no reason anyone would want a computer in their home".

More problematically for investors, scepticism tends to be superseded by exuberance, causing bubbles in emerging sectors. Dotcom investors know all about the dangers of overpaying for growth; the Nasdaq went on to lose 82 per cent of its value after the technology bubble burst in March 2000.

Golden oldies

Similarly, industries in secular decline can actually deliver the goods for investors. The Credit Suisse yearbook notes shares in railroads have actually outperformed airlines, road transport stocks and the overall US market since 1970.

The failure of Penn Central in 1970 – then the largest-ever US bankruptcy – led to dirt-cheap share prices, setting the scene for future outperformance, a cause that was facilitated further by industry rationalisation, productivity gains and deregulation.

Airlines have been the worst-performing sector over the same time period.

The same phenomenon can be seen in the ever-changing S&P 500 index. The index is rebalanced every year, with declining firms being replaced by “leading firms in leading industries”.

Since 1957, almost 1,000 stocks have been removed from the S&P 500; the same number of stocks has joined the index. According to author and Wharton finance professor Jeremy Siegel, if you bought the original 500 firms in 1957 and never sold any of them, you would have outperformed the actual index over the following half century. Clearly, it's perilous to both write off old fogies and to bet on up-and-coming players.

IPOs

Analysis of initial public offerings (IPOs) confirms this point. Buckets of studies, both in the US and UK, show that companies that float on the stock exchange tend to badly underperform in the coming years.

Excitement over future prospects for a particular sector – think internet stocks during the late 1990s, or social media stocks in recent years – results in prices being bid up excessively, setting the scene for years of poor returns.

In fact, it’s generally the case that the older the stock, the better. The Credit Suisse yearbook looked at the performance of different UK stocks over the last 35 years. Stocks that had been around for four to seven years outperformed those aged three or under.

Stocks aged between eight and 20 did better again, with the best results being those that had been around for 20 years or more. An annually rebalanced portfolio that invested in the oldest stocks would have returned almost three times as much money as one that focused on the youngest stocks.

Strategies

This has obvious implications for investors debating sector-rotation strategies. Investors tend to overvalue the new and undervalue the old; consequently, in both the US and the UK, cheap sectors have outperformed expensive growth sectors.

However, a momentum strategy – buying the previous year’s best-performing sectors – does even better than the value approach. Since 1900, US investors would have returned 870 times as much money by buying winning rather than losing industries. In the UK, results are similar.

Momentum and value may seem to be polar opposites, but this need not be the case. Indeed, some investors marry the two. Still, no approach works all the time; typically, such strategies fail in about one year in three. “In the case of momentum”, the yearbook notes, “the failures can be especially painful at market turning points”.

For many, fretting about sector valuation and momentum and the need to rebalance annually may seem like too much work, especially when index funds offer cheap, diversified portfolios.

However, sector rotation may appeal to patient long-term investors with a strong stomach, the yearbook suggests. At the very least, "the past success of these strategies may provide food for thought". Is rotation worth it? Diversification pays dividends Passive investors may feel sector rotation is not worth the hassle. However, they would do well to consider if their own portfolios are sufficiently diversified across industries.

The Iseq’s reliance on the banking industry meant investors were pummelled during the banking crash, with the index losing 81 per cent of its value. Today, it remains dangerously reliant on CRH, which accounts for a quarter of the overall value of the Irish market. Just three industries account for roughly 85 per cent of the index’s value.

The Irish case is quite extreme, but other countries also suffer from undiversified indices. According to the Global Investment Returns Yearbook, the three largest industries account for at least 40 per cent of country capitalisation in 42 out of 47 countries. In 15 countries, the three largest sectors accounted for more than 70 per cent of country capitalisation. Even in the most diversified countries – the US, UK, Japan, France – the three largest industries make up between 26 and 36 per cent of total market capitalisation.

Different countries are dominated by different sectors. Financial stocks account for a third of the market value of emerging markets. Technology accounts for 17 per cent of the US market, compared to just 1 per cent in the UK, which is dominated by oil and gas and mining stocks (the resource sector accounts for more than 25 per cent of the UK market). Germany and Japan are heavily weighted towards manufacturing industries, with little weightings in resources.

Most investors suffer from home bias, heavily tilting their portfolio towards their home market. The dangers of doing so are obvious from the above statistics: investors in most countries will end up with dangerously lopsided, undiversified portfolios. Clearly, investors need to diversify across different sectors in different countries.