Sunday, April 10, 2022

Fed psychology updates

Updates and rumination on my last few posts, why has the Fed responded so slowly to inflation. (Last post

1. Forward guidance? 

For the last several years, the Fed has placed more and more weight on "forward guidance." This is the theory that by promising to keep interest rates low in the future, even after the time to do so will have passed, the Fed can stimulate immediately. That is especially useful at the zero bound, and it is an important and explicit part of the Fed's new (well, pre-covid!) strategy. 

I and others were critical. Who will believe that the Fed, ex-post, will do what is not right at the time? I complained, will the Fed ever say to Congress, "yes, we should be raising rates, but we promised to keep them low 3 years ago when we were fighting deflation, so we have to keep that promise now. Sorry, inflation is going to have to rip a little stronger." 

Well, that seems to be exactly what the Fed is doing. Surely some thought of "we promised to keep rates low, now we'd better do it or people will never believe our promises" might be what's going on. I would be curious from Fed insiders if this is part of the discussion. 

I initially discounted this theory because then the Fed would be talking loudly about it. If you want to buy credibility, then say that's what you're doing. But in retrospect, saying loudly you're going to screw up inflation because you promised to do so isn't the wisest thing to do, so perhaps the Fed is making good on its promises, and not saying so loudly, to square that circle.  

 2. Fisherianism? 

I titled my first post, "is the Fed Fisherian?" A commenter asked whether in my view the Fed really thinks that raising rates to 4% rather than 2% would raise inflation.  Probably not. The model I used represents the Fed's forecasts; and the Fisherian prediction is a consequence of that model. But that doesn't mean the Fed believes all parts of that model 

The outlook is a judgmental forecast, informed by in-sample experience. None of us have much experience with the long run -- what happens, on average, if the Fed raises rates to 4% and leaves them there for a decade not responding to anything? 

On the other hand... A big piece of evidence for the new (rather than old, spirals) Keynesian model is the very stable behavior of inflation at the zero bound. And that is experience with interest rates stuck for a long time. There was no spiral, and inflation was quite stable. It's not hard to extrapolate that a constant 2% rate for a decade would produce the same inflation just 2% higher. 

3. Real rates? 

What is the core belief differentiating the Fed view from the traditional view? I think it is the effect of real raters on inflation. In the Fed's outlook, there is still a substantial period in which inflation is higher than nominal rates, so the real rate is negative. In the Fed's outlook, this period of negative real rates does not constitute additional monetary stimulus that sends inflation higher. The "temporary" shocks fade away, despite the negative real rates. In the traditional view, this period of negative real rates itself creates additional stimulus, and additional shock pushing inflation higher. That, not fisherianism, or the nature of expectations, may be the core of the debate. 


  1. “In the Fed's outlook, this period of negative real rates does not constitute additional monetary stimulus that sends inflation higher.”

    In the model,
    (1) x(t) = xᵉ(t) - σ · [ i(t) - r - πᵉ(t) ],
    (2) π(t) = β · πᵉ(t) + κ · x(t).
    … with xᵉ(t) = 0, Ɐ t, and σ = 1, κ = 0.5, β = 0.99, and, r = 0.5 , the FOMC must choose i(t) given expectations πᵉ(t). In the equation set (1) and (2), the manipulated variable is u(t) = i(t) – r , when r is fixed or constant.

    If r > 0, then u(t) and, consequently, i(t), must be greater than it would otherwise be than if r < 0 or r = 0, to achieve the same control outcome, ceteris paribus. Thus, the FOMC cannot look on r as being neutral following a shock to equilibrium, if (1) holds.

    However, if the value of r is a consequence of monetary coercion -- interest rates imposed by the central bank that serve only to suppress rates of return on financial assets (MBS, etc.) for the convenience of government fiscal balances, or to otherwise induce a monetary shock to raise the rate of inflation in order to lower real bond values -- then negative r is a consequence not of the natural or neutral real rate of interest that governs the efficient allocation of scarce capital, but as a consequence of a planned and cynical effort on the part of the central bank and the fiscal authority conspiring together to advance a predetermined political outcome. The shock, if it comes, is an inflationary spiral that upsets the assumptions of the status quo ex ante. We seem to be in that situation at present.

    Does r < 0 then not constitute 'monetary stimulus'? Or, is it a consequence of monetary stimulus itself? From eqn. (1), r < 0, raises x(t) Ɐ t, ceteris paribus. If we take the FOMC's explanations at face value, then the purpose of sticking at the lower zero bound was purposeful to stimulate output and lower unemployment -- the manipulated variable, u(t) < 0, raises x(t), and lowers unemployment. From eqn. (2) a higher x(t) raises the rate of inflation, π(t), ceteris paribus—we must therefore conclude that a negative real rate of interest, r < 0, is conducive to a higher rate of inflation.

  2. Perhaps it would be helpful if the Fed experimented a bit with policy? I discuss this in more detail here:

  3. Your observation on realized real rates is good. You earlier suggested the Fed does not believe in a strong short term negative effect, but the shift of their rate curve vs. the NK model may be informative that they believe in a stronger one than you give them credit for. It is remarkable how close you get to modeling their beliefs (Taylor would be happy about that at least), why not push it a little further to parameterize those key differentiating factors and try to crack their implied forward model?

  4. "In the traditional view, this period of negative real rates itself creates additional stimulus, and additional shock pushing inflation higher." It's worth restating this as when you subsidize something you get more of it. To keep the interest rate below the real rate of return you have to subsidize borrowing. So an analogous question is do loan subsidies increase prices? For some the the answer depends on whether we're talking about university tuition, but for the rest of us it's the subsidies, stupid.

  5. This comment has been removed by the author.

  6. How do you distinguish between inflation (declining value of money units) and price increases resulting from rebalancing supply and demand?

  7. The Fed failed because it looked at people not working because of Covid policies and foolishly thought that it needed to juice the economy to correct that problem. But people weren't choosing to stay away from the workplace because of a lack of jobs. The Fed made this very clear in Powell comments.


Comments are welcome. Keep it short, polite, and on topic.

Thanks to a few abusers I am now moderating comments. I welcome thoughtful disagreement. I will block comments with insulting or abusive language. I'm also blocking totally inane comments. Try to make some sense. I am much more likely to allow critical comments if you have the honesty and courage to use your real name.