Key to our perceptions of the economy and stock, bond and commodity markets is the relationship between interest rates and the business cycle. Has the economy bottomed? Or is it likely to fall deeper into recession? If it does, how long will the recession last?
The chart above compares the *change* over 12 months in the Fed's target Fed Funds rate with the year-on-year change in my monthly GDP proxy. Because a rise in interest rates leads to a fall in economic activity, the change in the Fed's target Fed Funds rate is plotted inverted. That is, rising interest rates are shown as negative. For example, the rise in the Fed Funds rate over the last year, from 0% to 4.3% is plotted as a decline. Doing it this way makes it easier to compare changes in interest rates with economic activity. Also, because interest rates affect the economy with a lag, the red line (the change in fed funds rate, inverted) has been shifted sideways (lagged) by 18 months. This gives you an implicit forecast of the likely change in economic activity over the next 18 months.
It's a little more complicated than that. First of all, the lag isn't fixed. As you can see, from 85 to 94, the lag was closer to 2 years. On the other hand, it seems to be much longer with the GFC in 2008. But over the last 15 years, 18 months has been about right.
Second, the amplitude of the declines in economic activity varies. However, the two occasions with the biggest variation were with the GFC (2008) and the Covid Crash (2020). The GFC was made much worse/deeper by the US mortgage crisis, after banks foolishly lent billions to borrowers who were unable to pay. Debt defaults and bank crashes are often a consequence of central bank tightening and subsequent recessions, but this time round it was much worse.
The Covid Crash caused a very deep but very short-lived recession. It wasn't caused by monetary or fiscal policy.
Take those two out, and the fit is much better.
Based on the relationship over the last 40 years, we would expect a recession as deep as the GFC over the course of 2023 and H1 2024. And that's before any debt defaults and bank failures. And before we take into account any fiscal tightening as the big covid deficits get scaled back.
OK, so could the relationship be about to break down? Why would it? Well, perhaps with Covid, supply bottlenecks, and war, these old (and logical relationships) might not work, for now. I'm not quite sure why this should happen, but it exists as a possibility. The world has turned topsy-turvy over the last 3 years.
A better argument is that in real terms, i.e., after inflation is removed, the Fed Funds rate is still negative. In other words, the monetary tightening by the Fed is much less than it appears on the face of it. The problem with that hypothesis is two-fold. First, inflation expectations haven't lifted anywhere near as much as headline inflation, and it's inflation expectations which matter. More important, the direct resultant of rising interest rates, money supply, is falling in real and nominal terms. And actual prices matter for real money supply.
At our Christmas function in 2007, I told clients that there was a serious risk of a deep downturn in 2008. Mortgage default rates had reached record highs during an upturn. If there was a downturn, default rates were likely to double, and that would take down the banking system. When the first payrolls data came in on January 7th 2008, I recalled our dealer from his holidays and I went into the office (I was also supposedly also on holiday) and we sold half the liquid shares, leaving small caps for later. The consensus at the time was that there would be no recession and that the payrolls data were a 'blip'. That consensus was completely wrong.
Once again, the consensus is clear: the US will experience a soft landing at worst. Stock markets are rallying on the back of this.
I suspect the consensus is wrong again.
Of course, the consensus may be right, and I may be wrong. We shall see.
No comments:
Post a Comment