Decoupling

If we start to look at developed countries starting in the mid-1990s — so a 30-year historical window — other than a couple of outliers (Japan, Australia), business cycles and interest rate cycles were largely synchronised. (The U.K. gilt market was a “freakshow” due to liability matching and the BoE rate cycles were originally mild due to the high sensitivity of the housing market to floating rates, but there was still a decent amount of correlation in the belly of the curve.)

From a bond market analysis perspective, a mid-1990s cut off is quite often close to the beginning of detailed market data — which means that the “correlated era” corresponds to the entire data set. Fixed income derivatives were still growing, and deregulation of banking and fixed income trading was still a recent phenomenon. Meanwhile, data providers generally do not have detailed data for earlier eras.

Correlations in financial markets resulted from a few factors. Markets were deregulated and opened up to foreigners. Policymakers herded together, and interest rates peaked in the early 1980s, with high term risk premia. This meant that developed bond markets started at fundamentally cheap levels, creating a cushion for correlation even if the policy rates diverged slightly. Meanwhile, central bankers tended to follow the herd with regards to the policy rate, partly because of the belief in the importance for policy rate spreads for currencies.

(Conventional economics suggests that central banks have almost no margin for manoeuvre for the policy rate — sustained deviation from something like a Taylor Rule will cause the economy to spin into hyperinflation/hyperdeflation. However, if interest rate policy is a weak tool, then there is plenty of scope for a central bank to follow other central banks’ rates without anything interesting happening in the domestic economy.)

The economic cycles were correlated as they were hit by similar forces. The synchronisation started because countries blew up their domestic economies in a variety of ways in the early 1990s (Berlin Wall shock, North American regional real estate busts), and then neoliberal reforms led to sluggish growth in the 1990s. The following expansions were driven by “global” forces — tech investment in the 1990s leading to a tech bubble, then crazy financial practices in the 2000s. Then, austerity policies (of differing levels of severity) crippled growth in the 2010s. It is only a couple of years after the COVID shocks that policies are diverging.

Could This Time Be Different?

The usual drivers forcing coupling are not present in the case of a European recession. It seems unlikely that European consumer demand would fall enough to greatly effect the United States and other non-European developed economies. Foreign banks are unlikely to cripple themselves with bad loans to European entities. Unless there is something large hidden in the plumbing that I am unaware of (easily possible), there is unlikely to be a seizure of the global financial system.

Although this would allow short-term rates to diverge, it is entirely possible that long-term yields stick together. U.S. dollar yields have been discounting rate cuts for some time, and a recession in Europe would reinforce economic bearish vibes. We would need continued perkiness in U.S. inflation to break the linkage in longer-dated yields.