In other words, the US behaves like a bank. The US also acts as a venture capitalist (Gourinchas and Rey 2005) to the extent that it borrows from foreigners in the form of debt and, in turn, acquires foreign equity assets. As a result, the US benefits from a so-called “return discount” and a “composition effect”. The return effect is reflected in the fact that the US generates higher returns on its assets than it pays out on its liabilities. The composition effect reflects the fact that the US tends to borrow short and to lend long.
By contrast, the emerging markets incur financial losses as a result of what might be termed the “return premium”: the rate of return on their liabilities is greater than the rate of return on their assets. While the US benefits from the composition effect, the EMs can be said to suffer from an “adverse composition effect”: they borrow long (mainly in the form of FDI) and to the extent that they accumulate foreign claims these are heavily concentrated in short-term, generally low-yielding debt (aka FX reserves).
Put differently, the US is “long equity” and “short debt”. The EMs are generally short equity and most of them are also short debt. Only China and Russia are net long debt (that is, they are net creditors). This seems to account for the somewhat surprising fact that the US generates positive investment income in spite of being an international debtor, while both China and Russia generate negative investment income on their positive net international investment position (IIP). The composition of EM international balance sheets translates into them paying more on their liabilities than they earn on their assets.
Two other countries that stand out in this respect are Korea and Turkey, where the rate of return on foreign assets exceeds the rate of return on liabilities. In the case of Korea, this seems to be in part attributable to the low share of equity claims in total foreign liabilities as well as a relatively low share of FX reserves in total foreign assets. (Technically, Korea is not an emerging but a newly-industrialised economy.) It is admittedly more of a puzzle why Turkey should generate a higher return on its assets than its liabilities given the composition of its international balance sheet. In part this may be attributable to the relatively low share of FDI in total liabilities.
In other words, the EMs are willing to hold low-yielding US (and other advanced-economy) assets that offer significant liquidity and low risk, while the US (and other advanced economies) are more heavily invested in typically higher-risk, historically generally higher-yielding equity assets, including FDI. This is why the international financial position of the EMs contrasts sharply with that of the major advanced economies (US, Japan, Germany or G3).
The G3 are all long equity and – with the exception of the US – they are all long debt. Once more, the US is the exception among the G3 due to its short debt position, while China and Russia are the exception in the EM-10 due to their long debt position. Admittedly, this picture would change somewhat if the UK, France and Italy, all of which are net international debtors, were included in the advanced-economy sample.
Moreover, unlike in the EMs, a large share of G3 liabilities is denominated in local currency (LCY), while their assets contain a significant foreign-currency component on account of their long foreign equity position. (With G3 FDI primarily concentrated in other advanced economies, the currency valuation and composition effects will be less pronounced than for EMs). The large share of equity claims on the asset side of their balance sheet as well as the generally lower return on their debt liabilities translate into a greater profitability of the G3’s IIP.
Thus the G3 tend to earn higher returns on their foreign assets than they pay out on their foreign liabilities, in aggregate. The EMs pay out more on their liabilities than they earn on their assets. This is due to both the return and composition effects. Until the EMs restructure their international balance sheet significantly, this situation is unlikely to change even once G3 interest rates start to rise. This restructuring may be driven by the private sector as international financial integration proceeds, including private EM investors increasing their share of FDI and foreign portfolio claims, and EM public sectors (incl. central banks) shifting out of low-risk sovereign debt securities into higher-risk equity assets.
The EM with the largest absolute amount of FX reserves is, of course, China. At the end of Q1, Chinese FX reserves amounted to USD 4 tr. Given a relatively closed capital account, very low foreign debt and a favourable current account position, USD 1-2 tr of FX reserves would be more than sufficient in terms of liquidity to insulate the Chinese economy from shocks. If, hypothetically but unrealistically, China were to shift USD 2-3 tr out of highly-rated US government bonds into world equity markets, this would have a tangible impact. Any such impact would naturally depend on the time horizon of such adjustment. World equity market capitalisation is USD 64 tr. Hypothetically, USD 2-3 tr would allow China to acquire more than 1/10 of all listed US stocks and/ or all German stocks – assuming the market prices did not move against the buyer. Bottom line: Greater diversification might not only raise the return of EM-held foreign assets. It might also help support selected developed markets assets such as equities, while at the margin it would potentially be less good news for US Treasuries, of course.