So the customary 3-month to 10-year term-spread on US Treasuries has now remained inverted for over a month since the 23rd May; it’s well recorded that this negative figure can signal a higher likelihood of a US recession, but typically at least 12-24 months down the line. Of course, there was also a negative dip in March but, with the change in Fed rates trajectory, equity markets have been buoyed and deflationary and recessionary concerns put aside (for the near-term at least) with the S&P 500 pushing all-time highs just as US Treasury yields have fallen dramatically throughout May and June.
But considering the real cost-spreads from a number of major participants in the 3-month funding and 10-year financing paints a more relevant picture of where (or more precisely when) we stand in the evolution of the US Treasury yield curve and perhaps the business and industrial production cycles. For repurchasing agreements (repos), banks funding at LIBOR and foreign investors hedging currency (notably EUR and JPY) around the 3-month mark to facilitate longer-dated Treasury purchases, have all had to accommodate for costs much higher than the 3-month T-Bill yields which is one of the reasons that foreign buyers of Treasuries have tapered-off over the past year putting pressure on primary dealers and consequently restricting the Fed’s capacity to raise rates (as absorbing Treasuries puts strain on the High Quality Liquid Asset (HQLA) repo market with large potential overdrafts on settlement days causing bouts of unsettling volatility in the Effective Fed Funds Rate (EFFR)).
Factoring for 3-month repo, LIBOR, EUR and JPY hedging actual-cost term-spreads, for these participants in the Treasury market, have been negative since Mar-19, Dec-18, Oct-18 and Sep-18 respectively. Since the limitations of Basel III - with banks no longer having near-limitless capital to arbitrage money markets – this seems a better modern equivalent measure for an inverted curve and suggests that we have been in such an environment since the start of this year. Again, this still puts us many months ahead of a recessionary signal, but suggests we may at least be 5-months closer to such an event than many might think.
Another recent turn in markets is that the cost of hedging for Japanese investors has fallen dramatically in recent months which may draw investors back to the Treasury market, reversing the ~20% long-term fall in Japanese holdings of US Treasury since 2015 (from ~$1.25tn to ~$1tn). However, although this may help lessen the pressures on the EFFR noted above, if this is driven from a growing expectation of yen strength versus the dollar (which our NFA analysis and historical flows suggest is correlated with risk-off trends and recession expectations) it is certainly little cause for optimism.