11 Mar 2009
What should we expect from equities? To answer this requires a long-term perspective. A week may be a long time in politics, but even a decade is too short to judge stock returns. Fortunately, the Credit Suisse Global Investment Returns Yearbook 2009 draws on 109 years of data for 17 countries that together represent some 90 per cent of world stock market value. Elroy Dimson, Paul Marsh and Mike Staunton provide an insight into some of the Yearbook’s highlights
The last decade has been the lost decade. The 21st century began with a savage bear market. By its nadir in March 2003, US stocks had fallen 45 per cent, UK and Japanese equities had halved, and German stocks had fallen by two-thirds. Markets then staged a remarkable recovery, only to plunge again late in 2007 into another epic bear market fuelled by the credit and banking crisis. Since 2000, the MSCI World index has lost a third of its value in real (inflation-adjusted) terms, while the major markets all gave negative real returns of an annualized -4 per cent to -6 per cent.
The demons of chance are meant to be more generous than this. Equity investors require a reward for risk. At the end of 1999, investors cannot have expected, let alone required, a negative risk premium from equities, otherwise they would simply have avoided them. Looking at the nine years that followed does not tell us that risk premiums have decreased, but just that investors were unlucky. Indeed, they received a savage reminder that the very nature of the risk for which they
sought a reward means that events can turn out badly, even
over multiple years.
In contrast, the 1990s was a golden age. Inflation fell from the high levels of the 1970s and late 1980s, lowering interest rates and bond yields. Meanwhile, expected profits growth accelerated. This led to strong performance from equities (except in Japan), bonds and even bills.
The 1990s contrast starkly with the opening years of the 21st century. Yet the 1990s are just as misleading. Golden ages, by definition, recur infrequently. To understand risk and return in the markets we need to examine much longer periods than one, or even two, decades. This is because stock markets are so volatile.
Long run returns and extreme periods
An initial sum of USD 1 invested in US equities in 1900 grew, with dividends reinvested, at an annualized rate of 9.2 per cent per year to become USD 14,276 by the end of 2008. Such is the power -- over 109 years -- of compound interest, "the most powerful force in the universe" (a phrase incorrectly attributed to Albert Einstein).
Since US consumer prices rose by almost 25-fold over this period, it is more helpful to compare returns in real terms. An initial investment of USD 1 would have grown in purchasing power by 582 times. The corresponding multiples for bonds and bills are 9.9 and 2.9 times the initial investment, respectively. These terminal real wealth figures correspond to annualized real returns of 6.0 per cent on equities, 2.1 per cent on bonds and 1.0 per cent on bills.
Events that were traumatic at the time now appear just as setbacks within a longer-term secular rise. Consider the extremes of stock market performance since 1900.
The two world wars were less damaging to world equities (real returns of -18 per cent and -12 per cent) than the peacetime bear markets (real returns of -44 per cent to -54 per cent). The worst bear market to date was the Wall Street Crash from 1929 to 1931, when the world index fell by 54 per cent in real, US dollar terms. However, this remains a close call. The peak to trough real return during the current banking/credit crash stands at -53 per cent. If the current remission falters and we hit new lows, it could yet become the worst bear market on record. In its short nine-year life, the 21st century already has the dubious honour of hosting two of the four worst bear markets in history.
Looking at real returns over four "golden ages", the 1990s was the most muted, with the world index showing a real return of 113 per cent. The world index rose by appreciably more during the 1980s (255 per cent in real terms) and the two post-world war recovery periods -- by 206 per cent in the decade after World War I and 516 per cent from 1949 to 1959.
Long run returns around the world
Until recently, most of the long-run evidence cited on historical asset returns drew almost exclusively on the US experience. This gives rise to a serious danger of "success" bias, since in the 20th century, the United States rapidly became the world's foremost political, military, and economic power. By focusing on the world's most successful economy, investors could gain a misleading impression of equity returns elsewhere, or of future equity returns for the USA itself.
We have looked at annualized real equity, bond and bill returns over the last 109 years. The real equity return was positive in every location, typically at a level of 3 -- 6 per cent. Equities were the best performing asset class everywhere. Furthermore, bonds beat bills everywhere except Germany. This overall pattern of equities beating bonds, and of bonds outperforming bills, is precisely as we would expect, since equities are riskier than bonds, while bonds are riskier than cash.
While in most countries bonds gave a positive real return, five countries experienced negative returns. The latter were also among the worst equity performers. Mostly, their poor performance dates back to the first half of the 20th century, and these were the countries that suffered most from the ravages of war and civil strife, and from periods of high or hyperinflation, typically associated with wars and their aftermath.
While US stocks performed well, the USA was not the top performer, nor were its returns especially high relative to the world averages. Many of the best performing equity markets over the last 109 years tended to be resource-rich and, quite often, New World countries.
The historical equity risk premium
Over the long run, investment in equities has proved rewarding, but has been accompanied by significant volatility. Investors dislike volatility and they will invest in equities only if they expect compensation for this risk. What we would really like to know is what risk premium investors require today, as this determines current valuations and future expected returns. Sadly, there is no reliable way of observing this, but what we can do is measure the risk premium that investors have obtained in the past.
We measure the historical equity premium by comparing past equity returns with the return on risk-free investments. Some people use treasury bills (very short-term, default-free, government securities) as the risk-free benchmark, while others use long-term government bonds. We prefer treasury bills, as bonds are subject to uncertainty about future inflation and real interest rates.
The annualized premium, relative to bills from 1900 to 2008, was 5.0 per cent for the USA, 3.7 per cent for the world ex-US and 4.2 per cent for the world. The annualized premium relative to bonds was 3.8 per cent for the USA and 3.4 per cent for the world.
Risk premium components
Is the historical equity premium a good guide to what investors expected and priced in beforehand as their required compensation for risk? Because equities are so volatile, we cannot be sure of this, even over periods as long as 109 years. Investors may have enjoyed more than their share of good luck, making the past too good to last. If so, the historical premium would reflect "the triumph of the optimists" -- the success of equity investors -- and overstate what we could expect in future.
An alternative approach is to delve deeper to infer what investors in each country were expecting, on average, in the past. We do this by decomposing the historical premium into three major components, namely, (i) the (geometric) mean dividend yield net of the real risk free rate, (ii) the annualized growth rate of real dividends, and (iii) the annualized change in the price/dividend ratio over time.
Of these three, the dividend yield has been the dominant factor historically. This may seem surprising, since day-to-day, investors seem focused on capital gains and stock price movements. Indeed, over a single year, equities are so volatile that most of an investor's return comes from capital gains or losses, with dividends adding a relatively modest amount.
The longer the investment horizon, the more important is dividend income. For the seriously long-term investor, the value of a portfolio corresponds closely to the present value of dividends. The present value of the (eventual) capital appreciation dwindles greatly in significance.
The other two major components of the equity premium are the growth rate of real dividends and the change in the price/dividend ratio. Real dividend growth has been lower than is often assumed. Real US dividends grew at an annualized rate of just 1.2 per cent. Most countries recorded real dividend growth of less than 1 per cent. Dividends and, probably, earnings have barely outpaced inflation. The final contributor to the equity risk premium is changes in valuation ratios, but the importance of this can also be overstated. Over the last 109 years, the price/dividend ratio of the world index grew by just 0.36 per cent per year.
The annualized historical risk premium relative to bills on a globally diversified equity portfolio (the world index) was 4.2 per cent. This comprises 3.2 per cent for the amount by which annual dividends exceeded the real risk free rate, 0.65 per cent per year from real dividend growth and 0.36 per cent per year from re-rating, i.e., an increase in the price to dividend ratio. Using this decomposition, we can now return to the question of whether 4.2 per cent was what investors required/expected in advance. Our analysis indicates that part of this amount arises from past good fortune and factors that are unlikely to recur.
Similarly, our analysis indicates that dividend growth turned out to be higher than expected. The 20th century opened with much promise, and only a pessimist would have believed that the next 50 years would involve widespread civil and international wars, the Wall Street Crash, Great Depression, episodes of hyperinflation, the spread of communism, and the start of the Cold War. During 1900--1949, the annualized real return on the world equity index was 3.5 per cent. By 1950, only the most rampant optimist would have dreamt that over the following half-century, the annualized real return would be 9.0 per cent. Yet the second half of the 20th century was a period when many events turned out better than expected. There was no third world war, the Cuban missile crisis was defused, the Berlin Wall fell, the Cold War ended, productivity and efficiency accelerated, technology progressed, and governance became stockholder driven. The 9.0 per cent annualized real return on world equities from 1950 to 1999 almost certainly exceeded expectations and more than compensated for the poor first half of the 20th century.
This type of reasoning coupled with more formal analysis leads us to conclude that the 4.2 per cent per year historical equity premium on the world index exceeded expectations, and was higher than the premium investors required in advance. After adjusting for non-repeatable factors, we infer that investors expect an annualized equity premium (relative to bills) of around 3 -- 3.5 per cent. This is below the long run historical premium and well below the premium in the second half of the 20th century. Many investment books still cite figures as high as 7 per cent, but investors who rely on such numbers are likely to be disappointed.
Nevertheless, even with a lower equity risk premium of 3.5 per cent per year, equity returns still compound rapidly. Equity investors can expect to be more than 40 per cent richer relative to investing in cash over a 10-year horizon, and twice as rich over 20 years. This represents a substantial premium that should encourage investors not to lose faith in equities.
However, while investors should keep faith with stocks, they should not harbour fantasies of an immediate return to either previous (and with hindsight, unrealistic) market levels, or to previous high rates of return. Markets are likely to take a long time to recover from the battering they have received during the credit and banking crisis.
In spite of this, we are confident that equity investors should continue to expect an appreciable long-run risk premium, albeit a somewhat smaller one than historically. We were spoiled by the high returns of the 1980s and 1990s, when equities seemed a sure fire route to getting rich quickly. Today, as we look ahead, while we should expect to enrich ourselves from equities, the process is likely to be one of getting rich more slowly. However, this does not mean getting steadily richer. Equity returns are far from steady-- they are very volatile. Markets will not get to their higher destination smoothly: returns could easily come in short bursts rather than gently over time. We need to take a long-term view, and be ready for the inevitable periodic setbacks, which can be severe, while recognizing that there are risks to being out of equities as well as in.
Elroy Dimson (firstname.lastname@example.org) is BGI Professor of Investment Management. Paul Marsh (email@example.com) is Emeritus Professor of Finance at London Business School. Mike Staunton is Director at London Share Price Database.