We examine over the last decade the relative importance of country and sector effects in explaining the cross-sectional variation in returns of stocks from developed markets. We use a methodology similar to Heston and Rouwenhorst, and document that sector effects have dominated country effects since 1998. Sector effects peaked in 2001, and have since decreased. Excluding the United States or the TMT (technology, media and telecommunications) sectors induces an increase in country effects, and a decrease in sector effects, but does not change our general conclusion. The currency effects make the country effects more pronounced, and do not materially modify the sector effects. Since 1998, sector diversification would have been much more beneficial for global equity portfolio management than country diversification.