After 30 Years, the Bond Market Vigilantes Are Back!

The end of the long era of "secular stagnation"? Or have bonds now bwwn oversold?...

Now comes this morning’s employment report. It is a very good one for the employment-production side of the economy:

Bureau of Labor Statistics: Employment Situation Summary: ‘Total nonfarm payroll employment rose by 336,000 in September, and the unemployment rate was unchanged at 3.8 percent.… The unemployment rate held at 3.8 percent in September, and the number of unemployed persons was essentially unchanged at 6.4 million…. Both the labor force participation rate, at 62.8 percent, and the employment-population ratio, at 60.4 percent, were unchanged over the month…. The change in total nonfarm payroll employment for July was revised up by 79,000, from +157,000 to +236,000, and the change for August was revised up by 40,000, from +187,000 to +227,000…

455,000 more seasonally adjusted payroll workers in September than we, yesterday, had thought we had in August. About three times as much of a shift in our assessment of the payroll-employment side of the economy as we had expected. This is yet another brick in the structure that has been building month-to-month late spring:

• Inflation news has been significantly better than expected: inflation has declined more rapidly conditional on their not being a recession than pretty much anybody had imagined.

• Production and employment news has also been significantly better than expected: production and employment have grown more rapidly conditional on inflation’s having rapidly declined than pretty much anybody had imagined.

The first of these should have led people to reduce their estimates of r*: the long-term real interest rate at which the economy is in balance. Why? Because when the actual real interest rate is below r* there is upward pressure on inflation, and there has been none such. The second of these should have led people to raise their estimates of r*: the long-term real interest rate at which the economy is in balance. Why? Because when the actual real interest rate is above r*, there is downward pressure on production and employment, and there has been none such.

The two factors should have roughly cancelled out.

Yet they have not. As of yesterday’s close, 10-Year Treasury Bond interest rates—both nominal and real—were a full percentage point higher than they were in the spring. And Treasury interest rates are jumping now.

Tim Duy three days ago, giving advice wise ex post:

Tim Duy: October 3, 2023: ‘I am certainly not going to advise anyone to try and catch the falling knife that is the current US Treasury market. This move has been so quick and for many unexpected that we are entering a zone where things can break, and that will catch the Fed’s attention. At this time, however, the Fed likely isn’t ready to take action that would turn the tide on yields. The proximate cause for today’s selloff was the better-than-expected JOLTS report…. I think the reaction to the JOLTS report reflects the fragility in the bond market more than any new insights into the labor market…. Although markets have priced in a roughly 30% chance of a November hike, such a move was already very unlikely before the Treasury rout…. There just isn’t a compelling case to rush to hike rates at the next meeting after a string of soft inflation numbers that is likely to continue with the September CPI release next week…

And so Tim Duy says that deciding when it is time to go in and bet that U.S. Treasuries have been oversold and the next move in interest rates is likely to be down—that that is like trying to catch a falling knife. That is certainly an arresting image.

What is going on? This: Economists are not revising their estimates of r*—the “neutral” real interest rate at which the economy is in balance. But the implicit expectations in the bond market generated by traders’ expectations and trading patterns are certainly being revised. We cannot yet know whether this is the end of the long era of “secular stagnation”—that Ken Rogoff’s seeing it as a “debt supercycle” was a better take. But very possibly: