The rise of 'smart beta' investing has led to explosive growth in research into what drives stockmarket returns. Smart beta means identifying factors that seem to explain the performance of different investments, such as value (cheap stocks outperform expensive ones on average) and size (small stocks outperform larger ones on average).
Literally hundreds of factors have now been tested: one recent study* compiled a list of 316 drawn from other research papers. Most of these turned out to be of minimal value when retested; only a few genuinely seemed to be associated with better returns.
Yet, rather strangely, one of the most obvious factors has barely been studied. For well over a hundred years, investors have classified stocks into groups basedon what industries they operate in. And almost everybody knows that the pattern of returns varies between industries: some sectors tend to be fairly steady (utilities, for example), some are volatile (mining), and some like to bankrupt investors on a regular basis (airlines).
Start your trial of MoneyWeek today
Get unparalleled financial insight, analysis and expert opinion you can profit from.
Start your trial
Sign up for MoneyWeek’s free twice-daily newsletter.
Join more than 165,000 subscribers and keep yourself informed with latest financial news, insights and expert analysis to help you understand what really matters when it comes to your finances.
Join more than 165,000 subscribers and keep yourself informed with latest financial news, insights and expert analysis to help you understand what really matters when it comes to your finances.
Rigorous research into which industries have done better over the long term has been in notably short supply. However, a recent study** by Elroy Dimson, Paul Marsh and Mike Staunton of London Business School has made an effort to plug that gap, by looking at returns for industries from 1900 to date for both the US and UK. The results are interesting.
Latest Videos FromMoneyWeek
Don't be worthy
The study's first finding is that things that are bad for you often deliver healthy profits just as cynical investors have long believed. The best-performing industry in the US over the 115 years was tobacco, returning an average of 14.6% per year. In the UK, the top performer was alcoholic beverages.
A similar pattern has existed in other stockmarkets around the world over a shorter period: "sin stocks" beat the overall market in 19 out of 21 countries in the period from 1970 to 2007.
By contrast, the worst performers were what are usually considered worthy industries not necessarily ethical as such, but the kind of sectors that make the economy function. The worst-performing American industry in the study was shipbuilding and shipping, while the weakest UK industry was engineering.
There are always a lot of caveats with this kind of study. The numbers reflect returns for industries that existed in 1900 and have continued to exist up until the present day. That's only a small number of sectors. In many cases, industries that are important today didn't exist 115 years ago: IT is the most obvious example, but there are many others.
Some industries may have vanished permanently due to changes in the economy, or temporarily due to political factors (there is no full US index for alcohol due to prohibition, or a UK one for sectors such as railways due to post-war nationalisation).
More subtly, the importance of certain industries in markets may have declined greatly: in 1900, railroads were 63% of the US stockmarket and about half of the UK one. Today they are less than 1% of the US and nonexistent in the UK. Despite that, returns on railroads in America have been relatively good over time, especially in the post-war period.
So there is clearly a great deal of hindsight bias in looking at long-term sector returns: it isn't necessarily clear at the time which industries are likely to grow or decline. However, some trends seem clear. First, investing in hot new industries often disappoints. That's partly because enthusiastic investors push up valuations to levels that future growth can't justify, creating a bubble, and partly because these breakthroughs may in turn be displaced by new ones.
Hence canals were the British boom of the early 1800s, peaking in value in 1824, before their position began to be eroded by the age of steam. Railways then evolved into a spectacular bubble that burst in 1844. There is certainly a warning here for investors today in industries such as IT.
One interesting consequence of this seems to be that the age of stocks is related to performance. Newer industries, especially those that are heavily hyped, will tend to be dominated by stocks that have listed more recently. Yet returns from stocks that have had their initial public offering (IPO) within the last three years are notably poor.
A strategy of investing in stocks of less than three years "seasoning" from 1980 to 2014 would have returned just £20 for every £1 invested at the start of 1980. Investing in stocks with four to seven years seasoning would have returned £33, those with 8-20 years seasoning £49 and those of over 20 years seasoning £61. Overall, old and tested beats new and exciting.
Avoid competition
So is investing in specific industries a sensible route to better returns? Not exactly. Investors certainly shouldn't buy into a sector just because it's performed well in the past: the economy changes and industries will go into decline. But it's worth considering why some sectors have consistently done well.
It's notable that those industries with the best performance tend to have barriers to entry that help to protect incumbents. These barriers may be unique physical assets: the main US railroad companies control physical infrastructure that is vital to moving goods around the country.
They could be valuable brands, especially in industries that produce consumer goods (drinks firms are an example of this). Conversely, those sectors that are readily open to new competition such as textiles manufacturing have generally done very poorly. If you can identify good firms in industries with high barriers and avoid paying too much for them you'll probably be onto a long-term winner.
*. ..and the Cross-Section of Expected Returns, Campbell Harvey, Yan Liu and Heqing Zhu (2014)
** Industries: Their rise and fall, Elroy Dimson, Paul Marsh and Mike Staunton, Credit Suisse Global Investment Returns Yearbook (2015)
