Mark Twain is frequently credited with the quote "History doesn't repeat itself, but it does rhyme." To that end, we thought that it would be important to look back at history before drawing conclusions about the future. In terms of investment returns, some of the first accurate calculations of investment returns over long investment horizons was performed by Roger Ibbotson and Rex Sinquefield. In their original research, completed in 1976 and updated in 2002, they focused on two key derived values. The first, was to look at all returns in excess of inflation. Part of the return available to investors is compensation for the erosion in purchasing power from the effects of inflation. These derived 'real returns' were expected to be more stable over time. The second, was to look at stock returns in excess of government bonds. The underlying concept is that investors should be compensated for taking on risk and that risk is relative to the risk free asset, a government issued bond. This excess return is termed the equity risk premium. The following table shows the results:
On average, over the very long term, government bonds have delivered close to 3% more than inflation and stocks have delivered over 6% more than government bonds. One of the strengths and drawbacks of these types of historical averages is that it tends to smooth the results of very different economic environments. Though it's true that over the long run these types of historic averages tend to hold, as John Maynard Keynes once wrote "The long run is a misleading guide to current affairs. In the long run we are all dead."
Another drawback to consider is that most of these longer term studies have focused on the U.S. market. This creates a bias by focusing on the market that delivered the best returns over that time period. By looking at the fifteen largest stock markets in the world in 1900, a different picture emerges. Four of the fifteen experienced a total loss of capital (China, Russia, Argentina and Egypt) and two other markets came close (Japan and Germany (twice)). By just looking at the best market over the past century, we may be inflating the return potential for the average market.
So let's instead look at Canadian data for long term returns and, in particular, lets look at how returns vary depending on the economic environment. To achieve this we will split the Canadian stock market history into three time periods: rising inflation (1969-1979), slowing inflation (1980-1998) and stable inflation (1956-1969 and 1999-2012). The results are as follows:
Clearly, the type of economic environment expected going forward would be important in determining reasonable expectations for returns. Of the three environments, we believe the third environment, stable inflation, is most likely. There has been concern voiced over the non-traditional measures being taken by certain central banks in an attempt to ensure there is sufficient liquidity in the banking system. Some economists worry that the increased money supply will inevitably lead to higher inflation, perhaps soon. We think this is unlikely in the near term as growth is constrained right now. We are still seeing significant capacity in economies such as the U.S. where unemployment remains stubbornly high and economic activity is not strong enough to cause wage price pressure to build. With this capacity in place, inflationary pressures should not build significantly. Taken all together, we expect inflation to be reasonably stable in the 2% range over the next several years. This should temper the extent to which bond yields rise in the near term.
Finally, once you have identified the economic environment, current valuations are equally important. For instance, in a stable inflation environment, the starting period government bond yield is indicative of future returns.
With long term government bond yields in the 2.5% range, it is pretty clear bond returns will be compressed. The most reasonable expectation is that bond returns will be roughly in-line with yields, also in the 2-3% range. These are much lower than bond returns over the past 10 or 20 years of close to 9%, and much lower than many investors are expecting.
So what does this mean for stock returns? If we look at the average equity risk premium of 4.4%, in a stable inflation environment, a reasonable expectation for stocks over the next several years would be in the 7% range (2.5% + 4.4%). However, future stock returns are also reliant on starting earnings yield (Earnings divided by Stock Price). When this number is high, returns tend to be higher than when the earnings yield is lower.
With stock market earnings yield in the 7% range, valuations are pretty reasonable compared to history. Looking purely at historic returns during time periods with similar valuations, we find that returns have been in the 10-11% range. Now keep in mind that these returns were delivered when interest rates were higher, so it is fair to assume that they represent an upper end of the range when it comes to expected returns. In addition, the expectation of more moderate economic growth going forward suggests returns slightly below this historical range for equities, perhaps in the 8% range for the next several years.
If we bring all the various components together, we can arrive at a reasonable set of forecasts for inflation, bond and stock returns given current valuations and the expected economic environment of stable inflation.
So what does all this mean for the next several years of investing? First off, these return expectations are somewhat lower than those many investors have grown accustomed to.
• Bond returns over the past 10-20 years have been elevated. Investors should use current yields as a better gauge for returns.
• In the current environment, stock valuations are reasonable and should be able to generate a reasonable premium to bonds. However, returns will likely be lower than historical averages.
Are bonds a reasonable investment over the next few years?
Even though the expected bond returns on government bonds are only between 2% and 3% over the next few years, they can still play a useful role in a balanced portfolio. Equities and other "risk assets" like corporate bonds and even provincial bonds tend to decline in price at the same time, as was seen in the financial crisis in 2008. The only true source of diversification at that time was government bonds which rose in price. While not expected, one can never rule out things getting worse, or simply not improving. Should this occur, government bonds will hold their value and provide investors with liquidity necessary to take advantage of the opportunities created when "risk assets" go on sale. The appropriate allocation to government bonds in balanced mandates can vary client-to-client but their merit, for many, still exist.