Does International Investment Reduce Risk?

Deep Risk

Usually, this headline’s question is answered by statistics. The calculation typically compares the standard deviation of a domestic portfolio with that of a diversified investment, which includes securities of other countries.

All of that is fine; I have no quarrels with such exercises. However, short-term volatility is only one way to measure investment danger. Another is the potential long-term damage to a portfolio’s purchasing power: “deep risk,” as my friend Bill Bernstein writes. That is the topic for today’s column.

While Bernstein customarily measures deep risk over several decades, I will do so over 10-year increments. This will provide more independent periods without changing the general findings. Otherwise, my approach matches his. Unlike traditional investment calculations, which measure nominal total returns, deep risk assesses after-inflation performance. After all, if prices have tripled, doubling one’s money means financial failure rather than success.

The Data Source

My numbers come from three academics: Elroy Dimson, Paul Marsh, and Mike Staunton. They consist of the annual after-inflation returns from 1901, computed in US dollar terms, for four assets: 1) US bonds, 2) US stocks, 3) the rest of the world’s bonds, and 4) the rest of the world’s stocks. The trio supply the results for many single countries, which are wonderfully interesting—who wouldn’t want to know that from 1902 through 1913, New Zealand stocks provided 12 straight years of real gains?—but beside this article’s point.

I split the study into two segments: 1) periods that began before 1950, and 2) periods after 1950. The cause for the division should be apparent. The first half of the past century featured two world wars and a global depression. More recently, there have been no world wars nor such a steep economic downturn. One would therefore expect the sharpest declines to have occurred during the first epoch, obscuring the results for the second era.

Back in the Day

Let’s see the results for the earlier period. The chart below shows the lowest 10-year real outcomes for each asset class, through the starting date of 1949. The totals are cumulative.

The worst result for US bonds should not come as a shock. Wartime breeds inflation, and inflation destroys the value of fixed-income investments. Over the four years from 1916 through 1919, consumer prices rose by 83%. The interest payments from US bonds were backed by the faith and fidelity of the federal government. What those payments could buy, however, was not so much!

(Bondholders fared somewhat better through World War II but nevertheless ended 1951 with only 72 cents on their inflation-adjusted dollars.)

US equities, however, lost less value than is commonly advertised. We are accustomed to reading about the devastation of US stocks during the Great Depression when they fell almost 90%. Well, … yes and no. Stocks did nosedive from late 1929 through early 1933, but their 10-year returns were not so terrible, thanks to a strong subsequent rally. Also, consumer prices fell during the Great Depression, which helped equities maintain some of their buying power.

In contrast, international investments were an unmitigated disaster. Foreign bonds were particularly awful. Not only did they surrender more than 80% of their real value during their worst decade, but that date exactly overlapped with US bonds’ lowest showing, which also ended in 1920. Invest only in domestic bonds, lose 43% for the decade. Protect against deep risk by stashing half the portfolio overseas, lose 62% instead. Just great.

Foreign equities were not much better. On the bright side, their steepest loss was not quite as severe as with bonds, and it did not occur during the Great Depression. That said, a 66% real decline is nonetheless awful, as was the timing, because that downturn also occurred during the 10 years that concluded in 1920! For the entire century, the worst decade for US bonds, world stocks, and world bonds coincided. So much for the good old days.

Modern Times

Here are the more recent results, applying the same methodology.

The US results may come as a surprise. Their absolute losses match conventional expectations, as even those who did not live through the 1970s recall the era’s inflation, which sunk bond prices. And although one is not accustomed to seeing large cumulative 10-year losses for stocks, that result does make sense given how persistent inflation can erode purchasing power.

The remarkable aspect of the recent worst-case scenarios for US assets lies instead with their relative levels. The lowest post-1950 performances almost match those from the black-and-white generations.

The overseas results stand in stark contrast to the domestic outcomes. Whereas leaving the US once meant courting danger, international diversification has powerfully earned its keep during recent decades, with both bonds and stocks posting mild worst-case performances. Since 1950, their deep risk has looked nothing like that of the distant past.

The reason for their improvement is straightforward. Back in the day, Europe accounted for almost all foreign investment. When it became embroiled in continentwide wars, there was little escape, not for bondholders beset by inflation and sovereign defaults nor equity shareholders facing corporate ruin. Today, not only are there no such wars, but the geography is far more varied. Currently, for example, none of the world’s four largest stock markets are in Europe, as they consist of the US, China, Japan, and India.

Summary

International portfolio diversification is a relatively new concept, spurred by technological advancements that both improved global communications and simplified the process of overseas investing. As it turns out, the delay was fortuitously timed, as early US investors profited by staying home.

Since then, though, overseas diversification has delivered on its promise. As the managers of target-date funds can ruefully attest, investing in foreign securities does not necessarily increase a portfolio’s returns, but without question it reduces deep risk. That is very much a meaningful benefit.