Investment Think Tanks

Historical sector returns and the future of investing

Analysing equity market returns over a very long period – 1900 to 2014 – reveals a dramatic transformation in the dominance of certain sectors. Elroy Dimson, chair of the Centre for Endowment Asset Management at Cambridge Judge Business School, and emeritus professor of finance at London Business School, outlined the lessons investors can take from long term financial asset returns at the Conexus Financial Investor Think Tank in London last week.

In his presentation “Investing in financial assets for the long term” Dimson analyses sectors and returns over more than 100 years using a database of market returns from 23 countries from the period 1900 to 2014 that he co-developed with Paul Marsh and Mike Staunton.
The database includes 21 unbroken histories – with China and Russia having broken histories – that looks at stocks, bonds, bills, inflation, currency, and GDP over 115 years.
Analysing the data reveals a “great transformation” in the dominance of industries, or sectors. For example in 1900 in the US around two thirds of the index was in railroads, but come 2015, the industries that occupied 80 per cent of the weight of the index in 1900 either don’t exist or are very small. In modern-day analysis the index is made up of banks, health, tech and other industrials.
“We can learn something from these long-term changes,” Dimson says. “In 1900 about 80 per cent of the weight was in sectors that don’t exist or are very small today, and conversely the industries that dominate today didn’t exist a century ago.”
In the UK it is the same pattern, with 65 per cent of the index made up of industries that don’t exist in 2015.
“Long term sector changes are important,” Dimson says, “particularly when you look at performance.”
From the period 1900 to 2014 the top performing sector in the US was tobacco.
“$1 invested in 1900 would be worth $6.3 million in 2010,” Dimson says. “In the UK alcohol was the biggest winner, with £1 invested in 1900 giving you $243,152 in 2010.
“When you are projecting into the future, this history tells you there are only a few sectors that survived the time period that outperformed the market.”
Dimson points out that one of the risks of looking at history like this is it leaves out industries like wagon makers, canal boats, steam locomotives or candle makers that were not going to survive. It is also important to consider that the start date of the period being assessed will affect the result.
“If you start at 1930 then history is very different,” he says.
What is clear is there is a “technology effect” in markets over time.
“200 years ago a disruptive technology was the use of water canals, where you could move 60 times the amount of goods compared to wagons. But if you were an investor in canals the returns weren’t very good in the 19th century. Investors lost out.”
Over time, this theme of disruptive technologies not benefiting investors plays out.
In fact analysis shows that the creators, owners and innovators of new technology, along with society, are winners, but investors are often losers.
Dimson quotes from Alasdair Nairn’s book, “Engines that Move Markets: Technology Investing from Railroads to the Internet and Beyond”, and concludes that there are a number of timeless lessons from technology that still apply today:
-new technologies generate bubbles
-there needs to be a sustainable competitive advantage
-and the greatest beneficiaries of new technologies are the insiders including the innovators and founders of the businesses, but also consumers and societies – but not necessarily the investors.
One way of demonstrating this is to look at the returns of companies from the time of their initial public offer (IPO).
Looking at market adjusted returns for IPOs in the US reveals that on the first day they trade at +17.9 per cent and for the next three years trade at -18.6 per cent.
Similarly in the UK on the first day, IPOs trade at +8.5 per cent, for the next two years trade at -9.4 per cent and for the next five years trade at -31.6 per cent.
Further, the returns improve with the years of seasoning since IPOs.
Dimson, Marsh and Staunton looked over the period 1980-2014, and £1 invested in a strategy that holds companies that are more than 20 years old since the IPO of the company would generate £61.
If the strategy was to hold companies which at the start of each calendar year had been listed for 8-20 years since their IPO, they would have generated £49. In contrast £1 invested in a strategy with companies that had 4-7 years since the IPO would return £33, and £1 invested in companies with less than three years since the IPO would generate £20.
“The older the company being purchased the better the performance. The long-term record is very striking,” Dimson says. “In general IPOs create new industries which on average underperform in their early years.”
In a bid to assess the performance of new technologies over the long term, and in fact whether backing new technology is a recipe for good performance, the analysis shows that railroads were the only transportation industry to outperform the market in the period 1900-2014.
“$1 invested in railroads in 1900 generated $62,019 against the market of $39,134, compared to roads that produced $10,436 and air that produced $7,194,” he says.
Dimson goes on to say that weighting of industries by country in the FTSE International shows there are many countries that are dominated by a couple of industries. In fact there are many countries that have less than 50 per cent in three industries.
“To diversify across industries, investors need to invest globally,” he says.
Taking sector analysis and applying it to responsible investing, Dimson shows the outperformance of “sin” industries since 1900.
In the US, tobacco beat the market with a return of 14.6 per cent versus 9.6 per cent, and in the UK tobacco beat the market by a similar margin.
While tobacco was the best performer in the US, in the UK alcohol has been the best performer since 1900.
Similarly taking the FTSE index, it shows the return on markets ranked by their corruption tendency showed that 21st Century returns had been highest in the most corrupt countries.
While there are a number of strategies for implementing responsible investing – including exiting companies and engaging with companies – Dimson argues that exiting companies may depress stock prices and raise the expected returns from vice stocks.
He says engaging with companies improves corporate behaviour which tends to be followed by share price gains. And says a strategy may be to buy the shares of irresponsible companies, clean up the business and then move on to the next “clean-up” opportunity.

Return to event coverage

Join the discussion