January 13, 2022
Is it best to invest in US Stocks, International Stocks, or a mixture of the two?
As you can see from the table below, there were significant periods of time where the World Ex-USA Stocks outperformed the US Stocks. This occurred in the 1970’s, 1980’s, and in the 2000’s. (Data from 1969 – 2021 based on the MSCI World Ex USA and MSCI USA data but the table was created by Sanguine Sojourner)
So, how do we make sense of these historical returns in order to take a logical decision on our portfolios? Let’s first take a step back and discuss the goal of diversification and recent global changes in businesses worldwide.
The purpose of Diversification in an Investment Portfolio is to add assets that will allow you to maximize your returns while minimizing your risk. The goal is to gain the maximum returns possible for each unit of risk that you take. To do this, an investor will usually look to add assets that are not highly correlated.
In the last several decades there has been a major globalization trend. Many articles have been published indicating that the US Stock Market and Global Ex-US Stock Markets are now much more highly correlated than historically. I wanted to see if this was true so I created a chart showing the correlation between the US and Global Markets.
The chart includes data between 1969 – 2021 and is based on the MSCI World Ex USA and MSCI USA data. The Blue line shows the correlation between the markets based on all prior data up to that Date. The Orange line is based on the trailing 10 year data. You can see that the trailing 10 year data shows a much higher correlation between the Markets now (0.87 Correlation) vs. in the 1970’s (0.64 Correlation). This means that the US and Global Markets are now moving much more in step with one another.
The US Market data is composed of companies that are headquartered in the USA but many of these companies now have a large portion of their revenues that come from overseas.
The earnings insight report from FACTSET for Q4 2021 (upon which the chart above is based) indicates that S&P 500 companies have 59% of their Revenue coming from the USA but 41% is coming from Overseas. They further show that the US S&P 500 Technology Companies have over 50% of their revenues coming from Overseas.
The higher correlation between US and Global Markets makes sense given the large amount of revenue that US companies are generating from overseas.
Given this high correlation that now exists between the US and Global Markets and the high amount of overseas revenues of US Companies, is there any need to invest in overseas companies and take on the currency risk?
In Portfolio Visualizer, you can run a quick analysis of the Efficient Frontier for the US Stock Market vs. Global Ex-US Stock Market. The output is below:
The Efficient Frontier is a map of the best possible returns for a given level of risk. Each point on the graph above is based on a portfolio that includes X Percent US Stock Market and Y Percent Global Ex-US Stock Market. The point on the far left represents a Portfolio of 25% Global Ex-US Stocks/75% US Stocks and the point at the far right is 100% US Stocks. The Y Axis is the Expected Return of the Portfolio and the X Axis is the Risk which is measured by the Standard Deviation of the Returns. The graph includes data from 1986 to 2021 so 35 years of data. According to the graph, you could decrease your Risk (Volatility of Returns as measured by Std Deviation) by including Global Ex-US Stocks but you would also decrease your returns. Including more than 25% Global Ex-US Stocks would be inefficient meaning you would take more Risk but obtain lower returns. So, based on this analysis, you would want to hold between 75-100% US Stocks and no more than 25% Global Ex-US Stocks.
However . . . the Portfolio Visualizer analysis is only based on data going back to 1986 (35 years of data). I saw other graphs in various articles online that went back further but they were from a few years ago and were not updated. So, I ran my own analysis from 1970-2021 (51 years of data). Here is the graph
The chart includes data between 1970 – 2021 and is based on the MSCI World Ex USA and MSCI USA data. Based on this analysis, again a portfolio of 100% USA Stocks shows the highest returns. It is possible to decrease Risk (Volatility as measured by Std Deviation of Returns) by including up to 20% Global Ex USA Stocks but this also decreases returns. Adding more than 20% Global Ex-USA (International) Stocks is inefficient and will increase Risk while delivering Lower Returns.
Based on the Efficient Frontier analysis for US Stocks and Global Ex-US Stocks from 1970-2021, we should hold between 80%-100% US Stocks and between 0%-20% International Stocks.
This analysis generally assumes that neither the US or Global Ex-US Markets are vastly over or under priced. It assumes that the markets are efficient and prices accurately reflect the potential growth and future revenues and earnings of the companies. But what if the US Market was severely overpriced and the Global Ex-US Market was not? Or vice versa?
One metric to determine if Stocks are priced reasonably is the Cyclically Adjusted Price to Earnings Ratio (CAPE). This is defined as Price divided by the average of 10 years of earnings, adjusted for inflation.
The figure above is based on data from an article posted at FTSE Russell by Sandrine Soubeyran, Director Research & Analytics. As you can see, the US Market shows a current CAPE ratio over 30 whereas the 20 year average CAPE for the US Market is closer to 20. Most of the other International Markets are very close to their historical average. This likely means lower future returns for the US Market as it reverts to the mean.
The current valuation situation would support an argument that the US Market is overpriced and speculative and that lower future returns are likely.
So, when did the US Market start showing these high valuation levels? If we look at the CAPE10 PE Ratio’s for the S&P500 Index that are tracked and published regularly by multpl.com at this link we can generate the following graph below.
I added a trend line and marked 1996 as a Red Dot because valuations after that year seem to be significantly higher than historically. There is a spike during the Dot-Com Bubble in 2000 and the market corrected but rapidly came back to levels at or above 25. So it looks like the US Market has had fairly high valuations since that time. Obviously, this should be a warning that high valuations can remain for quite a long time. So, trying to time a crash is very difficult. Anyway, if we consider that the US Market has shown high valuations since 1996, what happens if we look for the Efficient Frontier between US and International stocks up to this point (So from 1970-1996)? This should show the Efficient Frontier between the US and International Markets when they are both at more normal valuation levels.
I ran the analysis and the result is shown in the graph above. Based on the Efficient Frontier analysis for US Stocks and Global Ex-US Stocks from 1970-1996, we would get the highest returns by holding 100% International Stocks. We could reduce our Risk/Volatility by holding up to 70% USA Stocks but this would also decrease our Returns. Holding more than 70% US Stocks would be inefficient as it would add more Risk/Volatility while also decreasing our returns.
Based on the Efficient Frontier analysis for US Stocks and Global Ex-US Stocks from 1970-1996, we would get the highest returns by holding 30%-100% International Stocks and between 0%-70% US Stocks.
So, I do not think the answer to the question of how much US vs. International Stock should be in your portfolio is a simple question to answer. However, there are a few things which seem clear based on the data we have reviewed:
- US and International Stocks are now closely correlated (0.87 Correlation) whereas this did not use to be the case. S&P500 Companies are showing 41% of their Revenues are generated overseas.
- US Stocks show high CAPE Ratios vs. historical levels (Feb 2022 Schiller CAPE for S&P500 of 36 vs. historical average of 16.92.) whereas International Stocks are more in line with historical levels. So, US Stocks may be overvalued.
- If we consider all data from 1970-2021 which INCLUDES the recent decades when US Stocks seem to show high valuations, the Efficient Frontier analysis shows we should hold between 80%-100% US Stocks and between 0%-20% International Stocks.
- If we consider only data from 1970-1996 which EXCLUDES the recent decades when US Stocks seem to show high valuations, the Efficient Frontier analysis shows we should hold between 30%-100% International Stocks and between 0%-70% US Stocks.
Therefore, to decide your portfolio allocation of US to International Stocks, I believe you have to decide if you believe the US Market is overvalued AND if you think those high valuation levels are unsustainable.
If you think they may be sustainable then perhaps you want to allocate as in #3 above.
If you think they are unsustainable then perhaps #4 is more appropriate for your allocation.
Disclaimer: I am just a guy that happened to retire early and mostly chases waves, travels, rock climbs, works out, and plays at investing. I am not a financial adviser and am only sharing my own personal analysis which you may or may not find beneficial. Always do your own analysis and consult a qualified professional financial adviser prior to investing.
