Global exposure can be a great thing, but in the context of a diversified portfolio of stocks, too much of a good thing can increase risk.
There could be a couple of different ways to interpret the comment but the way I read it, the author is lumping together all foreign exposure as if it was one thing which is the wrong way to look at it. As we've talked about countless times each country has its own attributes as an investment destination. The attributes of a given country might be similar to the attributes of some other countries but it will not have the same attributes as every other investment destination.
Which countries having banking crises and which do not? Which countries have resources in the ground that the world must have and which do not? Which countries have problems with aging populations and which do not?
There are more differences of course and obviously the answers to every question are not all that cut and dried so for anyone inclined to spend the time it would be easy to construct a portfolio of different countries that takes in various types of fundamental attributes. The importance here is that countries with different attributes are likely to be at different points in their respective economic cycles which gives them a chance to be at different points in their respective stock market cycles which in turn creates the opportunity for lower long term portfolio volatility.
The best example of this is probably the fact that Brazil and Norway kept going up until June 2008 after the US peaked in October 2007. That example also makes a point about realistic expectations for this type of diversification. In the face of a short term market calamity it is not realistic to expect some market to be immune but a country that is fundamentally healthier has a decent chance of outperforming over longer periods of time. There were dozens of countries that did just that in the last decade and that will repeat again in the new decade (assuming the US turns out to be the laggard I think it will).
Contrary to the quote above I think an allocation to various countries with different fundamental attributes from each other and different from the US makes a portfolio less risky not more risky.
Generally we target 2-6% in any single country although China is more like 1% these days by virtue of its weighting in a couple of thematic ETFs we use for most clients. A 6% weighting would equate to two stocks with each one targeted at 3%.