International asset allocation: how much is just right?
I’m a big believer in international diversification. The days when international stocks were considered “risky” should be long over. An investor who only invests in his or her home country is taking a much greater risk, as Japanese investors realized in 1989.
The case of Japan
The Japanese stock market is still down from its peak after 19 years. So what has happened to various Japanese portfolios during this time? Here are some details, thanks to bpp on the Bogleheads forum.
- 60% Japanese index, 40% bonds, rebalanced annually lost 7% from 1989 to 2007, with a trough of 38% in 2002. $1,000,000 invested in 1989 would be $930,000 in 2007.
- Without rebalancing, this 60/40 portfolio would be down 13% from 1989-2007, down 30% at its low point in 2002. $1,000,000 invested in 1989 would be $870,000 in 2007.
Now imagine that a Japanese investor was invested 50/50 in domestic (Japanese) and international (North America and Europe, mostly) stocks. How would this investor perform?
- 30% Japanese index, 30% international, 40% bonds, rebalanced annually: up 93% from 1989-2007. $1,000,000 invested in 1989 would be $1,930,000 in 2007.
That’s still pretty lousy performance, but people who invested gradually throughout the 80’s and 90’s would have done much better than those who invested a lump sum at the peak of the bubble.
And its peak, the Japanese market had an average P/E ratio of over 100. In hindsight, it should be clear that it was dramatically overvalued. Even at the peak of the internet bubble, the S&P 500 had P/E ratio of 46. It’s currently far lower than this. So barring a huge bubble, it seems unlikely that a developed nation’s market would follow Japan’s trajectory over the last 19 years. We’ll come more to this talk of high and low prices in a bit.
But this highlights the importance of international investing. Don’t risk everything on a single country’s health.
How to invest internationally
So how much should I devote to my home country, and how much internationally? There are several approaches.
1. Overweight domestic stocks. Most investors do this. The US market is maybe 40% of the global equities market (by capitalization), so a US investor who only devotes more than 40% of his equity allocation to the US is overweighting US stocks. There is a good reason to consider doing this, and it’s not that US stocks are better or less risky. It’s that US stocks are held in US currency, and unless you’re retiring to another country, you need to preserve your purchasing power in your home country, in US$. So if the dollar rises dramatically and you hold a lot of international stocks, or if the US sees higher inflation than the rest of the world, then a high international allocation will mean you have less spending power back home.
Of course, a healthy US bond allocation can help protect against this, as can domestic REITs or cash.
2. Keep it simple. Invest 50/50 in domestic and international stocks. This will generally overweight your country, though it isn’t too far from the US market capitalization of the last few decades. But it is simple and makes rebalancing easy, which is almost more important than finding the “perfect” allocation. And it is a nice balance between the previous approach (preserving your local purchasing power) and the fourth approach (avoiding bubbles - see below).
3. Follow the market capitalizations. As of this summer, the US stock market had about 41% of the global market. Canada was about 4%. Europe was about 29%, Pacific developed stocks were about 15%, and emerging markets were 11%.
Holding a portfolio that matched this basically means indexing the entire world. And in fact, there are new funds that try to do this (like VT). They don’t do it perfectly, and typically don’t include many small or mid-cap stocks. And the expense ratios are typically higher than just holding separate ETFs. But man, they make global investing easy!
The downside of this approach is that it doesn’t protect you from bubbles - if there is an emerging markets bubble, or another Japanese bubble, you’ll rise and fall with the markets. Of course, that is true of any cap-weighted index: a S&P 500 fund suffered the same problem with the bubble in tech stocks in a decade ago.
4. Follow fundamental weightings. There is a bit of controversy in the investment world today about fundamentally-weighted indexes, offered by companies like Powershares and DFA. These indexes don’t hold their stocks based on cap weighting, but based on fundamentals (like earnings, assets, etc.). This means that they are less prone to bubbles - they will own less of a high-priced stock than a cap-weighted index, and more of a low-priced stock.
These fundamentally-weighed indexes effectively give a value tilt to stocks, which should help their returns over the long run. But they are not without their critics. Amongst other things, whoever designs the fundamental weighting may be prone to performance chasing and backtesting, and these funds typically have higher expense ratios than cap-weighted funds.
A similar strategy could be used for international weightings, but with regular funds. This eliminates the second problem (higher expense ratios), and can be done in a relatively simple manner.
Here is what that would look like. These examples use P/E, but you could also use other measurements, like the price/book ratio used by Fama-French to quantify value.
- US - 41% market cap - 13.7 P/E - 37.8%
- Europe - 29% market cap - 10.82 P/E - 33.8%
- Pacific - 15% market cap - 13.41 P/E - 14.1%
- Canada - 4% market cap - 13.91 P/E - 3.6%
- Emerging - 11% market cap - 13.02 P/E - 10.7%
The biggest outlier here is Europe, with a relatively lower P/E ratio. This means that one overweights Europe slightly, while underweighting everything else a bit.
These numbers change frequently, and aren’t exact. It would be a lot of work to stick to this closely, so you’d probably be best off revising this infrequently, maybe on an annual basis, or perhaps even less frequently than that.
5. Overweight emerging stocks. These markets are likely to have higher volatility than developed markets, so if you can handle that and want to bet on higher return, you might want to overweight emerging market stocks.
And of course…
Remember that there is no perfect allocation. David Swensen recommends 30% US stocks, 15% developed, and 5% emerging. This overweights the US (60%) relative to international stocks, and overweights emerging stocks relative to other international stocks. Others recommend other things. You can’t know ahead of time what work best, but you can know that rebalancing to a reasonable plan is better than adjusting your allocation every few years.
So invest domestically, diversify internationally, and stay the course!
