More on the possibility and risks of a recession

So both the (inverted) yield curve and the Sahm rule indicate a recession. This together with two months of slower employment creation, and the slightly higher unemployment rate, has many wondering whether the economy will crash soon. I discussed before -- a while ago, before the pandemic recession, that had nothing to do with the yield curve -- why an inverted yield curve doesn't necessarily mean a forthcoming recession. The Sahm rule, like the inverted yield curve has an impressive track record. It suggests that if three month moving average of the rate of unemployment rises 0.5 or more above the minimum of the same averages for the previous twelve months a recession is under way. Figure below, although scale doesn't help, shows that we are at 0.57 for last August [ominous music here].

This is essentially an a-theoretical measure, contrary to the yield curve which could have different explanations, including the Wicksellian one used by the mainstream. It expresses basically a trend. Unemployment rates go up in recessions, and after a while going up, you're basically in one. The two episodes in which it fails, as far I can tell, were in 1959 and 2003. No clear reason for why, as opposed to housing market measures that failed in 1951, 1967, and perhaps now (if you believe me), because of two wars (Korea and Vietnam) and fiscal packages (Bidenomics).

In a few weeks now, the Fed is very likely, almost certainly really, going to reduce interest rates. I don't expect that to stimulate the economy much, and it certainly will not create any danger about an inflationary resurgence. Also, as I noted on Marketplace a week or so ago, there are some positive signs about the economy. Real wages at the bottom are growing, and consumption went up. And no, you should not be concerned with the low savings rate. I was a little less sanguine that I sound in that short soundbite, but overall I think that's correct.

Sure enough a recession could certainly imply a return of Trump, and Trumponomics. I doubt that it would cause inflation and that his election would deepen the recession, as some had argued. Not because tax cuts for the wealthy would stimulate the economy. But the truth is that Republicans in power don't care about the deficit or debt. They care about cutting social benefits, and about facilitating the lucrative relations between the corporate sector and government. What Jamie Galbraith called the Predator State. Trumponomics shares with Reaganomics, and other GOP supply side voodoo economics notions, a persistent characteristic. It is always against unions and higher wages, and always for lower taxes for the wealthy.

Dems are less consistent, but certainly less keen on both (and Kamala seems to have accepted Bidenomics and the pro-union agenda). The problem with that is that it has opened the door for right wing populism. The differences with previous versions of conservative economics are subtle, and in some sense rhetorical, since they do very little for working people. On right wing populism economics and their connection to the working class, and the problematic relation of Dems with the working class, I suggest the recent piece by Kim Phillips-Fein on the London Review of Books. She says:

"In 1968 George Wallace talked to a working class that was afraid of dispossession. Trump speaks to workers too, but more directly uses the language of money and corporate success; his appeal derives from identification with the boss -- reflecting the extent to which he seeks to win the allegiance of small business owners. Just as American political institutions have been hollowed out since the 1960s, so has the country's political economy, in ways that have helped to increase Trumps' appeal."

The self-made man myth, an American neologism (Henry Clay, if I'm not wrong), is incredibly corrosive. I hope I'm right and we can avoid a recession, and Trump.

The inverted yield curve and the recession

The inverted yield curve, as it is well-known, indicates a forthcoming recession. I used it last year to suggest that the recession was not in the near horizon. The conventional explanation follows Wicksellian ideas (see this old post). In the Wicksellian story, one can think of the 10 year bond rate as a proxy for the natural rate of interest, and the Fed Funds for the monetary or banking rate. Hence, whenever the short-term rate (Fed Funds) is above the long-term one, it would be reasonable to assume that borrowing short-term is a bad idea, there is not enough borrowing, and investment falls short of savings. Lower investment would be the cause of the recession, and of deflationary forces.

Graph below show the difference between the 10-year bond rate and the Fed Funds, which I have noted I prefer to the more common 10-2 spread, since the Fed Funds is more clearly a policy variable, dependent on decisions of the FOMC.

And the yield curve has turned negative, which does indicate (look at the past in the graph) a high likelihood of a recession. But I remain skeptical, even though according to the BEA GDP growth slowed down a bit in the second quarter, and the trade deficit fell, due to lower imports (that decreased more than exports), both signs of a slowing economy.

First let me explain that you don't need to believe that the inverted yield curve would cause the recession because the monetary rate is above the natural rate of interest. You may very well think that there is nothing special or natural about the long-term rate. Post Keynesians often think in terms of uncertainty, and the role of expectations. Note that the fears of a recession in this case are related to a collapse of investment (this is the view of certain posties, for example, John Harvey here, that provides always reasonable and clear analysis; he claims to be skeptical about the yield curve).

In the Wicksellian story the high short-term rates (in comparison to the natural) discourage investment too. Here the idea of the marginal productivity of capital plays a central role, while posties would suggest that expectations are more important. But the mechanism is the same. Higher interest rates would lead, along an investment curve that is negatively sloped with respect to interest, to lower levels of investment. You could park your money in short-term securities, and avoid investment.*

But there is no reason to take a marginalist or Wicksellian version of the story. While for WIcksell the natural rate is not a policy variable, that might not be the case with the long-term rate (the 10 year government bonds). In fact, both rates can be influenced by the central bank, and the Fed has a history of switching to longer term securities after a crisis. After the 2008 Global Recession, the Fed increased its holdings of long-term government bonds, maintaining a low interest rate for government debt, and also bought significant amounts of Mortgage Based Securities. Buying long-term bonds pushes their price up, and reduces its remuneration. So lower tong-term rates have been a result of policy decision to some extent.

When the Fed decided to reduce its holdings of long-term securities it basically announced that the interest rate on the long-term bonds would go up. But last year the Fed announced that the program would slowdown and eventually end by this summer. So it basically reversed its previous policy stance, at the same time that it was increasing the short-term rate, presumably because the economy was beyond full employment (or the natural rate of unemployment, if you believe in Friedman's Wicksellian story; for the Fed description of its policies go here). And changes in interest rates and in the structure of rates should have significant effects on the balance sheets of agents spending, and affect the level of activity.

However, I wouldn't expect investment to be the key variable affected by higher short-term interest rates. Indebted agents would cut spending immediately if higher interest rates put pressure on their budgets, either because they have to pay higher interest or because they can borrow at less favorable terms. And sure lower consumption would then impact investment. Firms seeing that consumption is not too strong, would curtail investment, following the so-called accelerator. So I can live with the story that an inverted yield curve, because of a significant and fast increase of short-term rates, can lead to a recession.

And that might happen sooner than I think. But I'm still unsure about the reasons for expecting that immediately. Note that trade is more often than not blamed for the coming recession. See, for example, Greg Ip, from the Wall Street Journal, here, suggesting that trade and not the Fed would be the one blamed for the recession in the future. But as I noted before, I would expect the impact of tariffs to be stronger in China than in the US, and to be more on prices than on quantities. So expect more inflation, and some disruption of the production chains, but not a recession. And the government budget deal seems to inject some additional fiscal stimulus. Perhaps not enough, and perhaps other forces would be sufficient to throw the economy into a recession. But I still think the case for an immediate recession is still not a slam dunk. The slow recovery might continue for a while. But certainly things look worse now than when I wrote last year.

* Yes, that is open to the capital debates critique.

Bill Gross and the Yield Curve

Tyler Durden at ZeroHedge, and others, are discussing Bill Gross's recent rant on his monthly letter to investors about the yield curve and the possibility of a Trump recession. Bill Gross sees in the decline of the 10-year bond rate since the early 1980s a secular (like Summers and his secular stagnation, it seems everything is secular now) trend, and concludes that the long term rate cannot go above 2.6% or so. In his words:

"So for 10-year Treasuries, a multiple of influences obscure a rational conclusion that yields must inevitably move higher during Trump's first year in office. When the fundamentals are confusing, however, technical indicators may come to the rescue and it's there where a super three decade downward sloping trend line for 10-year yields could be critical. Shown in the chart below, it's obvious to most observers that 10-year yields have been moving downward since their secular peak in the early 1980s, and at a rather linear rate. 30 basis point declines on average for the past 30 years have lowered the 10-year from 10% in 1987 to the current 2.40%... And this is my only forecast for the 10-year in 2017. If 2.60% is broken on the upside – if yields move higher than 2.60% – a secular bear bond market has begun."

The fear is, of course, that if the Fed continues to tighten monetary policy, then the yield curve (the difference between the long term rate and the base rate) would be inverted (see figure) and a recession would follow.

There is nothing secular about the falling rate long or short term rate, however, which Gross analyzes in nominal terms. The declining tendency is just the result of the lower rates of inflation, what Bernanke called the Great Moderation, even if the causes are not the ones he suggested. Sure enough, in real terms, the 10-year treasuries are also down, but I cannot see a trend.

What I suspect is going on in the graph above are three different phases, in which the long term rate has fluctuated around different levels. During the Golden Age (50s through 70s) the rate was relatively low, a result of tight regulation, and extensive capital controls, which allowed low rates at home. After the Great Inflation, with negative rates, and the Carter/Reagan/Clinton deregulation rates were considerably higher. What we observe in the 2000s is that after several bubble-led booms, nominal rates have been forced to the floor. But there is no secular trend associated to this, it is the result of policy choices given the macro and regulatory regime we have.

It's is perfectly possible, if Trump really promotes a fiscal expansion, that the base rate would rise, and the long rate too, keeping a positive yield curve, and a higher average real 10-year bond rate. Above 2.6% for sure. Not saying it will happen. Just that it is plausible, and there is no secular trend that precludes it.

On the possibility of a recession, again

So the yield curve is really flat, not inverted, but really flat, and that has many  (or here) afraid of an impending recession. The fear is basically associated to the inverted yield curve (see below: when the blue line is above the red and green lines, there is an inverted yield curve, with a high short rate and lower longer rates, signaling a recession) which is really flat, and the danger that the Fed will rise the rate in the next meeting in a few weeks.

Yield Curve (click to enlarge)

Blanchard, cited in one of the WSJ pieces above, thinks that the Fed might be forced to hike interest rates, since the economy, presumably is close to the natural rate, that is, to full employment and inflation is a danger. Note that Blanchard is suggesting that the 2% target should be upheld, even though he was against this when he was at the IMF.

I think this is misguided on many levels. On the theoretical front the reliance on the natural rate is problematic, of course. The notion that we are close to full employment is doubtful to say the least, given that participation rate has not recovered much at all since the last recession. Real wages have not grown that much either, and with lower commodity prices the risk of higher inflation (if one is concerned not just with core inflation) is not particularly high. The FOMC should leave the short run rate unchanged. But that is not enough. The Eccles mantra should be repeated more often. Monetary policy is like pushing on a string, in a situation like this one. We need fiscal stimulus.

Yield curves and recessions

Let me get back to my discussion of the neo-Wicksellian macro model. One important feature of the model, is that it suggests that the central bank controls the rate of interest, and it should try to flatten the yield curve. That is, the bank rate (short run) should adjust to the natural rate (long run). In many respects this is the general rule behind all conventional stories about central banking. The Taylor Rule or the New Keynesian ideas behind Clarida, Galí and Gertler are basically a variation of Wicksell's story.

One thing that is also important about Wicksell's rule is that if the yield curve is negatively sloped (the bank rate is higher than the natural rate) then a recession (deflationary forces) are to be expected. The graph below uses the Fed Funds for the bank rate and the 10 year Treasury bond rate for the natural rate. In between the gray lines the official NBER recessions are shown.

As it can be seen, after the red line (Fed Funds) moves above the blue line (Treasury bonds rate) a recession always follows. There are many problems with the Wicksellian model, not the least the assumption of a natural rate of interest, that was severely criticized by Keynes in the General Theory. But the empirical notion that an inverted yield curve forecasts a recession seems to survive any possible theoretical critique. More on the critique will be left for other posts.