PWL Capital has been producing its own projections of future stock and bond returns for the last five years. These projections are a key input in preparing financial plans for our clients. The bedrock of these projections is the estimated rate of inflation. The reason for this is simple: investments really start creating wealth once they have produced a return in excess of inflation. This is why, in economic parlance, net-of-inflation investment returns are labelled “real returns.”
For the last three years, PWL’s inflation estimate has been running below the 2% target of both the Bank of Canada and the U.S. Federal Reserve. Furthermore, our estimates have remained below those of the IQPF and FPSC.1 Table 1 below documents these differences.
| 2014 | 2015 | 2016 | 2017 | 2018 | |
|---|---|---|---|---|---|
| PWL Capital2 | 2.0% | 2.0% | 1.8% | 1.6% | 1.6% |
| IQPF/FPSC | 2.0% | 2.0% | 2.1% | 2.0% | 2.0% |
| Cibles Banque du Canada/Réserve fédérale É.-U. | 2.0% | 2.0% | 2.0% | 2.0% | 2.0% |
Because of the above differences and other historical indicators, some observers have expressed concern that PWL’s inflation estimates might be underestimating reality. This article provides a historical background of inflation in Canada and the developed world, and explains PWL’s methodology for estimating long-term future inflation.
According to Dimson, Marsh and Staunton,3 the rate of inflation between 1900 and 2000 was 3.1% in Canada and 3.2% in the United States, which is almost double PWL’s projection of 1.6%, made at the beginning of 2018. Furthermore, Canadian and U.S. inflation has exceeded 3% on several occasions since the 1930s. Inflation rates on a decade-by-decade basis are documented in Table 2 below. The decades with an inflation rate in excess of 3% are shown in red.
| Canada | US | |
|---|---|---|
| 1930–39 | -1.76% | -2.05% |
| 1940–49 | 4.71% | 5.41% |
| 1950–59 | 2.29% | 2.20% |
| 1960–69 | 2.56% | 2.52% |
| 1970–79 | 7.55% | 7.37% |
| 1980–89 | 6.28% | 5.09% |
| 1990–99 | 2.10% | 2.93% |
| 2000–09 | 2.05% | 2.52% |
| 2010–18 | 1.76% | 1.78% |
Another concern expressed is that inflation rates have been materially higher than 2% in all developed countries except Switzerland in the 20th century, as outlined in column 1 of Table 3 below. The number for Germany is, of course, biased by a long period of hyperinflation from 1910 to 1930. However, as illustrated in column 2, inflation rates have been muted since 2001.
| 1900–2000 | 2001–2017 | |
| Australia | 4.1% | 2.5% |
| Belgium | 5.5% | 1.9% |
| Canada | 3.1% | 1.8% |
| Denmark | 4.1% | 1.6% |
| France | 7.9% | 1.4% |
| Germany | 34.6% | 1.4% |
| Ireland | 4.5% | 1.6% |
| Italy | 9.1% | 1.7% |
| Japan | 7.6% | 0.2% |
| Netherlands | 3.0% | 1.8% |
| Spain | 6.1% | 2.1% |
| Sweden | 3.7% | 1.3% |
| Switzerland | 2.2% | 0.4% |
| United Kingdom | 4.1% | 2.8% |
| United States | 3.2% | 2.1% |
PWL believes market signals provide the best estimate of the value of securities. For example, an investor deciding to dispose of a stockholding will seek to sell it at the highest possible price in order to maximize wealth. On the other side of the trade, the potential buyer aims to achieve a trade at the lowest possible price, also in the pursuit of maximum wealth. The trade occurs at a price on which both market participants agree voluntarily, despite their opposing interests (the seller wants the maximum price, the buyer wants the minimum price). It is the intersection of these opposing interests, plus the competition among a large number of buyers and sellers, that results in the equilibrium price, or the notion that the market price reflects the true value of a security.
While the stock market attracts much of the public’s attention, the market for government bonds is also extremely active and competitive. In Canada, the federal government issues two different types of bonds: “conventional” bonds and “real return” bonds (or RRBs). While the face value of conventional bonds is fixed and will remain the same until maturity, the face value of RRBs is indexed to the Canadian Consumer Price Index every month. The yield to maturity of these two instruments fluctuates with the changing prices throughout the trading session. As a general rule, conventional bonds tend to have higher yields than RRBs to compensate investors for their lack of inflation protection. For example, if you buy a one-year conventional bond at a yield of 4.5% and consumer prices rise 2% over the year, you’re going to end up with a 2.5% real return on your investment. If at the same time, your friend purchases a one-year RRB at a yield of 2.5%, they will get a 2.5% real return on investment regardless of future inflation, because the face value (or the amount that will be reimbursed by the issuer at maturity) will be increased by the rate of inflation. Since the only difference between conventional bonds and RRBs is inflation indexation, this means that their difference in yield provides a great (possibly the best) indication of what market participants believe future inflation will be.
PWL defines itself as a wealth management firm that provides services to long-term investors. Consequently, we are interested in making projections over a very long horizon. Therefore, we measure the expected long-run inflation by the difference in yield between conventional bonds and RRBs with a 30-year maturity:
Projected inflation = Yield30-Year Conventional Bonds – Yield30-Year Real Return Bonds
Furthermore, in order to reduce the ups and downs in our estimate, we use the 24-month average as our base assumption for financial planning purposes. Chart 1 below depicts inflation projections for Canada and the U.S. (The U.S. government also issues both conventional and inflation-indexed bonds.)
Although higher inflation was documented worldwide during the 20th century, we prefer to base our inflation projections on forward-looking indicators. In our judgement, the difference in yield to maturity between long-term conventional and real-return bonds provides a very accurate estimate of the market’s inflation expectations. We believe the intersecting interests of the multitude of bond investors provide an unbiased estimate of the value of these two types of bonds at any given time. If the market collectively expected inflation to be much higher than our estimate, then RRBs would be a bargain, and investors would bid up their price, resulting in a wider yield spread between the RRBs and conventional bonds. The opposite would be true if inflation estimates were much lower than our projection. We believe that the collective knowledge and profit motivation of participants in the government bond market provide the most accurate signal available about long-term future inflation.
