This blog post is based on a working paper by Zsolt Darvas and Pia Hüttl:
Back in the 1960s, Valéry Giscard d’Estaing described as ‘exorbitant privilege’ the advantages that the United States enjoys on its foreign assets relative to its foreign liabilities. US investors earn more on their foreign assets abroad than foreigners earn on their US investments, resulting in a boost to annual investment income flow to the United States (Figure 1). And in several years, revaluation of US assets – due to stock-price increases, for example – came in higher than the revaluation of US liabilities.
We examined these US privileges in global comparison in a paper we just published with Pia Hüttl. In this blog post I focus on the yield (investment income flow) on foreign assets and liabilities. In a later post I’ll also look at revaluations.
In line with the literature, we find that the main reason for high yields on US net total assets is high yield on foreign direct investments (FDI) made by US investors abroad. For example, on average between 2000 and 2016, yield on US FDI abroad was 7.2%, while yield on German FDI abroad was much lower at 4.8%. Other continental European countries benefited from yields quite similar to German yields. Only a few other advanced countries, like Norway, Switzerland, Japan and the United Kingdom, had FDI yields comparable to the US.
What is the reason for the high US yields?
What is the reason for the high US yields? One answer could be risk; it is possible that US investors invest in riskier projects than, for example, German investors, and riskier investments should deliver (on average over a long time horizon) a higher yield.
Unfortunately, available data does not allow us the consideration of all aspects of risk. But we can control for an important risk factor: the country composition of foreign assets and liabilities. For example, FDI investment in Austria might be less risky than FDI investment in Thailand. Certainly, it is also possible that US investors invest in markedly different sectors of the Austrian economy, or if they invest in the sector of the Austrian economy, they might invest in companies within the same sector that have different risk profiles. While we cannot exclude this hypothesis, we believe that considering the country-composition of foreign investment already captures most of the risk factors.
We therefore calculate the average yield on FDI liabilities of 78 investment destination countries. For each country, we use weights which are proportional to FDI investment made by that country –for example, for the US we consider the country-composition of US FDI abroad. The results suggest that the US indeed invests in countries in which FDI yields are somewhat higher – but only somewhat. For example, between 2006 and 2016, the average FDI yield in countries in which the US invested was 5.9%, while the average yield in countries in which the Germans invested was 5.4%. Therefore, the geographical composition of FDI assets, or different riskiness of FDI investments, is only a small part of the story.
Much more important is the yield relative to average yield of the destination countries: US, and also British and Japanese investors, were able to outperform the average yield earned in the countries of their FDI destinations, while German and most other continental European investors earn just that average (Figure 2).
Therefore, one conclusion we draw is that risk likely explains only part of the large yields on US foreign assets. What explains the rest? We raise three possibilities.
Do investments in ‘tax optimisation’ countries distort FDI yields?
A recent study by Garcia-Bernardo and his co-authors used a numerical method to identify off-shore financial centres, which are frequently used for ‘tax optimisation’ purposes. We found that about 60% of US and 40% of UK FDI is invested in such countries, and Japanese investors also invested a surprisingly large share of Japan’s FDI investments in the Cayman Islands. In principle, this should not alter yields, given that we compare reported profit transfers (relative to FDI assets) and thereby undeclared income does not enter the statistics we use. However, when investment in ‘tax optimisation’ countries is so high, FDI yield and stock data might be measured imprecisely.
Does the treatment of intellectual property distort the statistics?
Some companies might establish the bulk of their intellectual property in their home country and have little physical investment in other countries, yet profit from these other countries might be related to their home-country intellectual property. Thereby, the ratio of profit to physical investment abroad can be large.
Could financial sophistication contribute to high yields on FDI assets?
Financial sophistication might help investors to better identify profitable investment opportunities and the US, the UK and Japan are financially quite sophisticated countries.
Further research should analyse the relevance of these and other possible reasons for the high FDI yields earned by US, UK and Japanese investors.