Many investors share two expectations. First, the expectation that stocks, over the long term, will outperform bonds. Second, the expectation that a portfolio with both stocks and bonds will create diversification benefits because of the negative correlation between their prices (if one goes up, the other goes down and vice versa). However, both of these expectations have become questionable, for reasons that we will explain.
To start with the second expectation, the year 2022 was a wake-up call, as the negative correlation between stocks and bonds came to a sudden stop. Following the war in Ukraine, higher inflation levels prompted central banks to raise interest rates and triggered a shift to a positive correlation between the two asset classes. Higher borrowing costs for companies negatively affected many stocks, while the value of existing bonds also decreased due to higher yields offered by new bonds. As a result, both stocks and bonds went down. As we have previously argued, it is uncertain what the correlation between stocks and bonds will look like in the coming years (because it depends on the level of inflation), but in most of history, the correlation has actually been positive:
Indeed, investors should take seriously the possibility that stocks and bonds will move in the same direction more often. See the following graph: it shows that even on a 1-year rolling basis, stocks and bonds have moved in the same direction 67% of the time (at minimum):
So what about the first expectation that stocks will outperform bonds? Here we should observe the findings by Edward F. McQuarrie, who recently published an interesting paper with some new, surprising data.
According to conventional wisdom, over a longer holding period, stocks are always better investments than bonds, albeit against a higher short-term risk level because of higher volatility or price swings. However, based on data from 1792 to the present, McQuarrie shows that for most of history, stocks and bonds have performed about the same. Only from 1942 to 1982 in the United States did stocks perform better than bonds for a prolonged period. In the rest of the world, the performance for stocks and government bonds from 1970 to 2019 was about the same: 5.1% versus 5.0% annualized.
To conclude, the conventional wisdom that says stocks will outperform bonds over the long term is based on a one-time 40-year period which occurred in only one country. It is clearly visible in the following graph:
The data forces us to rethink not only the history, but also the future of financial markets. McQuarrie says that many investors will dismiss his findings because he includes the 19th century, a period with “far different macroeconomic conditions”. However, the graph shows that recent times (post-1982) have seen the same pattern of bonds outperforming stocks. Indeed, we could currently be on another trajectory of the stock advantage slowly reversing into a deficit, driven by, for instance, a higher level of interest rates. Finally, as McQuarrie notes, even if the different phases of performance between stocks and bonds are randomness misperceived as patterns, it is still necessary to accept that stocks will not always beat bonds, no matter the holding period.
