I'm off on a little camping trip, so I'm proud to allow Larry Russell, a fellow blogger, to contribute. This is a long article (but with lots of great information), so it will be split into two parts. Part 2 will publish early next week.
Introduction to evaluating historical financial asset class returns with and without inflation
Very often, long-term historical U.S. asset class returns for the primary stock, bond, and cash financial investment asset classes are presented as:
- graphics with cumulative long-term returns,
- simple historical averages, or
- dense tables of numbers
These approaches can be less than satisfying, because the relative performance of asset classes has varied over time depending upon the financial, economic, monetary, political, and inflationary situation within major sub-periods.
In this article, first I first present 85-year asset class cumulative returns graphics to illustrate this problem. Then, I break down this longer period into five major sub-periods that illustrate how asset class performance has varied depending upon the conditions of that period.
We begin by looking at cumulative U.S. stock, bond, and cash asset class returns from the beginning of 1928 to the end of 2013 with and without inflation. Then, because cumulative stock returns completely overwhelm those of other asset classes graphically, we will look at just U.S. bond and cash returns from 1928 to 2013 with and without inflation.
Later, we will break down 1928 to 2013 into five distinct periods:
- Stock market crash, Great Depression, and World War II (1929 through 1946)
- Post-World War II economic expansion (1947 through 1968)
- Escalation of inflation and increasing economic stagnation (1969 through 1981)
- Stock and bond bull markets as inflation recedes, including the Dot Com mania (1982 through 1999)
- Dot Com crash and recovery; then Credit Crisis, Great Recession, and recovery (2000 through 2013)
Major economic, political, and securities market events influenced my choice of 1947, 1968, 1982, and 2000 as the breakpoints in splitting 1928 to 2013 into five periods. No attempt was made to make these five periods be of equal duration. The focus was on selecting distinct periods that could be described by the major economic, political, and securities market events that characterized them.
For the period represented by all graphics, it is assumed that one dollar was invested in each of the U.S. cash, bond, and stock asset classes at the beginning of the first year. No additional funds were added thereafter. Any change in asset value over time represents the total annually compounded gross return (pre-tax without investment expenses) for that asset class. Annual asset class returns include changes in asset market value plus reinvested interest and dividends.
For some of these breakpoints, perhaps one could argue that they should have been made a year or more earlier or later. Being overly concerned about a precise break-point year tends to miss the point. By breaking this 85-year period into five shorter periods, we can more effectively see graphically how the stock, bond, and cash asset classes performed in relative terms. Should you prefer a different year as a break point, simply view the adjacent graphics, note value changes for nearby years, and then make mental adjustments to a particular graphic.
This article includes a total of 14 historical U.S. asset class returns graphics. Seven graphics are provided in terms of inflationary dollars and a corresponding seven in real dollars. For each period discussed, the first graphic presents asset class values using inflationary dollars. Then, the following paired graphic provides the same information in real dollar terms adjusted for inflation. Real dollars represent constant purchasing power over time. (Note that on the screen, the background color of all inflationary graphics is lighter blue, and the background color of all real dollar graphics is darker blue.)
Concerning the U.S. annual data for stock (S&P Index), bond (10 year Treasury Bonds), cash (3 month Treasury Bills), and inflation (U.S. urban Consumer Price Index), several sources were consulted including the Bureau of Labor Statistics, Standard and Poors, Damodaran Online (Dr. Aswath Damodaran, Professor of Finance, New York University Stern School), and Online Data (Dr. Robert J. Shiller, Professor of Economics and Finance, Yale University).
Primary U.S. asset class returns from 1928 to 2013 using inflationary dollars
Over very long periods and especially during the second half of the 20th century, cumulative stock returns have been so high in relative terms that they have dwarfed bond and cash returns. Yet few investors have the luxury of holding only stock investments for decades hoping to capture higher stock returns. Even committed buy-and-hold investors with a high tolerance for investment risk sometimes need to draw cash down their investments to fund unplanned negative cash flow during their working careers.
Note also that academic research has demonstrated that even highly risk tolerant investors tend to benefit from holding at least some minority cash and bond asset positions over time. This is especially true, when there is uncertainty about when an investor might need to draw down some of their equity assets to fund negative cash flow. Unplanned cash flow draw-downs tend to be correlated with the loss of income associated economic downturns, and these downturns tend to be highly correlated with stock market sell-offs.
This 1928 to 2013 inflationary dollar graphic illustrates how stock asset class returns have dwarfed cash and bond returns. One dollar invested in stocks at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor of over 2,500 measured in inflationary dollars.
Cumulative stock returns have been so high in relative terms that this graphic makes it appear that cash and bonds have barely appreciated over this long period. Furthermore, relatively high asset appreciation in recent decades makes it appear that stock investment returns only began to diverge in the mid-1970s to leave cash and bonds in the dust. Later graphics will demonstrate that this was not actually the case.
Primary U.S. asset class returns from 1928 to 2013 using real dollars
This graphic presents the same cumulative long-term cash, bond, and stock returns, but the data have been adjusted for inflation. One dollar invested in stocks at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor of about 190 measured in real, constant purchasing power dollars.
In addition, this real dollar chart now makes it seem that stock returns began to diverge from cash and bond returns in the mid-1950s rather than in the mid-1970s, which was suggested by the earlier inflationary dollar graphic.
Primary U.S. asset class returns without stocks from 1928 to 2013 using inflationary dollars
From the two charts above, clearly we can learn something about cumulative long term U.S. stock returns, but nothing about cash and bond returns. Since stock returns have been so overwhelming in relative terms, one approach is simply to exclude stock returns if one wishes to understand better relative cumulative cash and bond returns over this 85 year period graphically.
The graphic below indicates that one dollar invested in 10 year U.S. Treasury bonds at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor of about 64 measured in inflationary dollars. For cash returns one dollar invested in 3 month U.S. Treasury bills at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor of about 20 measured in inflationary dollars. Note that over this period urban CPI inflation eroded the value of a dollar by a factor of about 13 times.
Primary U.S. asset class returns without stocks from 1928 to 2013 using real dollars
This chart presents the same cumulative gross long-term cash and bond returns, but the data have been adjusted for inflation. As noted above, inflation eroded purchasing power over this period by a factor of about 13.
This graphic indicates that one dollar invested in 10 year U.S. Treasury bonds at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor somewhat above 4.5, when measured in real dollars. For cash returns, one dollar invested in 3 month U.S. Treasury bills at the beginning of 1928 through the end of 2013 had a cumulative gross total return factor of about 1.5 measured in real dollars.
Long-term investment returns broken down into five sub-periods
The remainder of this article will break down cumulative 1928 to 2013 returns into the five distinct periods listed above in the introduction. For each of these periods, you will find separate graphics of cumulative inflationary dollars and real dollars for the cash, bond, and stock asset classes.
Again, keep in mind that it is assumed that one dollar is invested into each of the three major cash, bond, and stock asset classes at the beginning of each of these five sub-periods. Total gross returns are presented, including asset market value changes plus reinvested interest and dividend payouts. Returns are gross and do not include any taxes or investment expenses.
Stock market crash, Great Depression, and World War II from 1929 through 1946 using inflationary dollars
While the full period includes 1928, this graphic starts with January 1929 values. Doing so emphasizes the behavior of asset classes including and after the 1929 stock market crash, but without the effects of the significant run-up in stock asset values during the 1920s.
One of the first things that this chart illustrates is that stock prices declined in a continuous train wreck until 1932. So much historical memory has focused on 1929 and Black Friday that some people do not realize that the years following were much worse in terms of year-over-year returns.
In the first half of 1929, stock prices had continued to run up, so the full year effect of the 1929 crash was that 1928 year-end to 1929 year-end stock prices prices fell “only” by 8.3%. In contrast, year-over-year stock prices collapsed by an additional 25.1% during 1930, then 45.8% more in 1931, and by 8.6% more during 1932. Measured by calendar years, the cumulative collapse was 65% for 1929 through 1932.
This inflationary dollar graphic shows this 65% value for one dollar invested in stocks at the beginning of 1929. Because it uses annual figures, this graphic tends to mask the somewhat greater peak to trough decline that occurred in stock prices measured on a daily basis. This is not very important, because our purpose is to understand relative behaviors between the asset classes from the 1929 stock market crash, through the Great Depression, and through World War II.
Clearly, an investor with a portfolio much more heavily invested in bonds and cash did much better throughout most of this period until after WWII, when equity prices recovered and almost closed the valuation gap with bonds. However, when the stock to bond gap was at its greatest during the Depression at the end of 1932, in inflationary dollars the initial bond dollar was worth $1.15, while the initial stock dollar was worth only $.35 for a difference of 80 cents. A second peak inflationary dollar valuation gap occurred in 1941, when the original bond dollar was worth $1.59, while the original stock dollar was worth only $.65 for a difference of 94 cents.
In addition to the more well known advantage of bonds and cash versus stocks over this period, there is a lesser know story that this chart tells us. From 1929 through 1932, the U.S. economy suffered cumulative deflation of almost 25%. If you ever wonder why central banks during the more recent Credit Crisis and Great Recession expressed so much concern about deflation and took such extraordinary measures to prevent deflation, the reasons can be found in the 1929 to 1932 period.
Laissez faire national fiscal and monetary policies during 1929 to 1932 period were almost completely hands-off, and securities markets, the economy, and consumer prices were left to find their own equilibrium points. Unfortunately, those were very low equilibrium points that induced a severe and sustained depression.
Economic demand plummeted, because the one-quarter of the working population that was unemployed could afford to buy very little. Moreover, the remainder of the population who still had jobs and/or possessed financial assets did not consume as they might normally. Deflation in the Depression was so pernicious, because declining prices provided an incentive to delay consumption. Particularly during the first five years of the Depression, the cost of products and services declined steadily. Delaying consumption is hardly a recipe for a growing economy, and the 1929 to 1932 period is an important case study for anyone who wishes to understand the deleterious effects of deflation.
Stock market crash, Great Depression, and World War II from 1929 through 1946 using real dollars
The 1929 through 1946 real dollar graphic below adjusts cash, bond, and stock values for changes in the urban CPI over this period. This chart displays asset values in terms of constant purchasing power, and thus the CPI line is displayed as a constant $1 over this period.
Note that the real dollar charts below for each of these five major sub-periods all draw a constant CPI, wherein the CPI is, in effect, adjusted by itself. Holding CPI at a constant $1 value is intended to remind the reader that all asset values on these real dollar charts are inflation adjusted and reflect constant purchasing power over time.
During the worst of the financial collapse from 1929 through 1933, it is notable that a flight to asset safety caused the real dollar value of both cash and bonds to escalate by 50% to about $1.50 for each dollar invested at the beginning of 1929 through 1933. Also, in 1931, stock values had begun to stabilize in real dollar terms and recover unevenly in value between 1932 and 1936, reflecting the fragile financial recovery that began with the New Deal era.
In 1934, cash and bond asset values began to diverge, as the value of holding cash declined over the remainder of the period, while bond values moved upward to peak at about $2 in real dollar terms in 1940, just before U.S. entry into WWII. In contrast, real dollar stock asset values experienced another significant set-back in 1937, as stimulus measures were reduced.
Stepped up economic activity related to war production caused stock asset values to rise throughout the war and almost close the gap with bonds by 1945. However, economic adjustments after the end of the war caused another decline in stock prices, as inflation rose by 8.1% in 1946, 14.1% in 1947, and 8.1% in 1948.
Lawrence J. Russell wrote this article. He is a Pasadena, California fee-only family financial planning consultant and registered investment adviser. He is also the developer of the VeriPlan personal financial planning software.