This state of affairs has a very classic life cycle and related to other form of carry trades. This currency carry trade is also very deeply related to the volatility in currencies and other asset classes. The fact that currency volatility has remained incredibly low over the last decade has mitigated, so far, forced unwinds of leverage in this carry strategy. Can it last indefinitely?
One scenario in which currency volatility can rise sharply is a hiccup in risk assets that results in a rush into US fixed income assets for protection or due to an aggressive rate cut by the Fed, which would result in the narrowing of the interest rate differential between US rates and foreign rates (assuming that negative European rates don’t become more negative at the same speed) and a consequent fall in the forward exchange rate. This scenario would create a much lower yield pick-up when the currency hedges are rolled forward, resulting in possibly less demand for foreign bonds at their already low yields. The other scenario, which seems to be almost non-existent in the collective consciousness, is that European short term rates can rise a fair bit if there is a change in the ECBs philosophy towards quantitative easing and negative rates (note that Draghi is likely leaving in October), which would also possibly result in a collapse in the forward exchange rate and hence the yield pickup from hedging. Not a high probability forecast, but certainly a tail risk.
Let us apply the same analysis above first for a Japanese Yen based investor and then a Euro based investor. For a Yen based investor, the chart above shows that US treasuries are the worst yield of the bunch on a hedged basis (negative). One could argue that financial alchemy this time converts gold into dirt! But this leads to another question: should one prefer to buy US 10 year treasuries at a yield of 2.08% un-hedged, or Italy at 2.52% hedged? In the first case if we prefer to buy the un-hedged US Treasury, we are taking currency risk, but no credit risk. In the second case, we are taking no currency risk, but taking lots of credit risk (of Italy). In a late cycle environment, credit risk is often more dangerous than currency risk. For a Euro investor France provides almost no yield, and Italy provides the only other positive yield. The US is at the bottom again in terms of a negative hedged yield for a Euro investor across the whole yield curve. What is not shown in this chart but is a fact is that for a US investor, the longest duration German Bunds (actual yield of a measly 0.32% for 30 years) actually yield 3.33% currency hedged, and for a Euro investor, the 30 year US bond yields only -0.34% currency hedged. In other words, even as monetary policy remains easy, investors are being tempted by the yields in the longest, most duration sensitive part of the yield curve.
The point is this: Whenever there is a type of alchemy that (1) turns high into low and low into high, (2) uses a “yield curve” mismatch, (3) is exposed to shocks to volatility, (4) uses a derivative contract that needs to be rolled (5) depends on Central Bank policy for survival of its benefits, one should get cautious. In the case of the global government bond markets, the currency hedging tail is indeed wagging the bond dog. Just like yield curve inversions within one currency can cause mayhem to bond markets by upsetting the “carry” arbitrage between maturities, the sudden collapse in interest rate differentials, upward shocks to volatility or to forward exchange rates between two different currencies can create the same sort of chaos in currency markets.
It is said that “there are no bad bonds, only bad prices”, and the currency hedging markets are a good example of how “trading sardines” are being created in the global bond markets.