Brad DeLong Convinces Me I’m Not Insane
Remember that famous psychology experiment where the unwitting subject is in a room with ten people, and then the administrators show two lines–one much shorter than the other? If the first nine people (who in reality are all confederates with the experimenters) say the shorter line is longer, then the last guy might go against his eyes and agree. But if just one person says the obvious, then the actual test subject is much more likely to say the truth too. In other words, people are afraid to be completely alone in their views, but if just one other person agrees, then they’ll speak what is staring them in the face.
(BTW I am probably taking serious liberties with the details of that experiment. See here for an actual summary.)
Anyway, Brad DeLong says the obvious regarding Scott Sumner. The context is Sumner criticizing modern Keynesians for thinking monetary policy was easy in 2008, since interest rates came down. Here’s DeLong:
Well, I would say that not just “modern Keynesians” but a lot of people believed that monetary policy was expansionary in 2008.
They believed so not just because (safe) nominal (and real) interest rates were falling, but because the money supply was expanding. Indeed, since 2007 the Federal Reserve has tripled the monetary base…If expanding the monetary base to three times its previous size is not “expansionary”, what could possibly be?
Then DeLong quotes some other guy (Matt Rognlie) who argued that we have contractionary Fed policy because real interest rates are too high. DeLong says:
“Wait a minute!” you say. “The ten-year nominal Treasury bond rate has fallen 325 basis points since 2007. The three-month Treasury bill rate has fallen by 500 basis points since 2007. And this is not a case in which apparently low nominal interest rates are really high real interest rates because of expected deflation: expected inflation has fallen by only about a third as much as nominal interest rates have fallen. The real interest rate on ten-year Treasury bonds has fallen from 2.5% per annum in 2007 to zero today…
“And Matt wants us to believe that this pushing-up of the present value of a real dollar of cash flow ten years in the future by more than one-quarter–this more than doubling of the present value of a real dollar of cash flow thirty years in the future–this tripling of the monetary base–is contractionary? What is going on here?!”
What Sumner (and Rognlie) should say, I think–in order to avoid confusing readers who try to wrap their minds around the idea that a large monetary expansion is contractionary–is that monetary policy was expansionary but the expansion was not large enough to cope with the macroeconomic problem.
But for some reason they don’t want to say this.
THANK YOU Professor DeLong. I can put away my bottle of crazy pills now.
P.S. One thing in fairness to Scott Sumner, he does have a point that “real interest rates” (as implied by TIPS yields) spiked in the fall of 2008 during the crisis. (Look at DeLong’s chart to see what I mean.) So if Scott wants to say, “The Fed tightened for a few months right then, and since has loosened but not enough,” then OK I won’t quibble.
Cue: MamMoth using an argument (that I predicted that he would use) supporting DeLong’s reasoning. Five, four, three, two…..
Not sure I follow you Joseph. In this post, I’m not being sarcastic. I’m agreeing with DeLong, that the Fed has clearly been engaged in unprecedentedly expansionary policy since 2008.
Sorry, I am referring to the quoted material. The “reasoning” is DeLong’s interpretation of the quoted material.
After looking back, I can see the confusion.
I doubt it. MMTers don’t believe monetary policy does anything.
That’s not necessarily true. They just argue that interest rates are complex variables, which come in with a lag, and which have distributional impacts on flows between different sectors of the economy, and between the non-government sector and government sector. To go someway to understanding the impact you would need to have a clearer picture of the liability structure of the economy. Furthermore they reject the notion that there is some simply linear relationship between investment decisions and interest rates.
When you say liability structure, you don’t just mean the current debts in the system, right? You would need to know about the possible debts which may be created (something like a demand “curve” for investment funds as a function of interest rates), right?
Sorry, I didn’t reply sooner. Also, looks like the other thread that we were in has closed 🙁
Well as I understand it:
“liability structure” is the term Minsky uses to refer to the financing arrangements within the economy. For Minsky, the key institution of a capitalist economy is money, and the primary source of money (credit) are banking institutions. Credit is provided by banks in financing arrangements. Banks issue credit to credit worthy borrowers (which is subjectively determined, and procyclical). So it refers to current debts in the system (depending on how the financial arrangements handle interest rate changes) but also new debt.
The way I see this is a process through time, which is bounded by institutional norms.
From my understanding: Post Keynesians begin with a world in which the future is uncertain, and they then begin thinking about how agents respond to this uncertainty, such as via institutions, rules of thumb, and expectations. The decision to investment is based upon an expectation that 1. the investment will succeed in some particular time frame, and 2. that the cash flow generated from the investment and from other sources will be sufficient to cover expenses. The role of interest rates is really of secondary importance, but I imagine that there would be some threshold which would make investment decisions more sensitive to interest rates.
Another complicating factor would be that investment decisions would be based upon expectations of future interest rates, and how they expect buyers to respond to the changes. If you imagine a good whose demand is particularly sensitive to interest rate changes, then investment decisions will be heavily weighted by the (expected) future path.
The latter is speculation on my part. I’d see it as an empirical question, and dependent on financing arrangements.
Sorry for the rant. I’ll be interested to read what you think.
Sumner explains here why he believes money policy was tight in 2008 despite some of the commonly indicators indicating otherwise
http://www.cato-unbound.org/2009/09/14/scott-sumner/the-real-problem-was-nominal/
you have to admit that the introduction of the feds IOR policy really messes with the analysis of the first round of QE, so i feel some of scotts frustration on this one. now the other rounds have clearly been expansionary, as we have been seeing for awhile in the cpi
Sometimes when I read the likes of Sumner, DeLong and Rognlie, I get the thought that I am living in a socialist economy with free range chicken as opposed to a cooped up chicken, which enables these interventionist economists to believe that they are something other than tools of the state, and doing something other than just providing the socialist planners with free information with which to integrate into their plans.
At any rate, Sumner and DeLong are espousing an a priori theory that is not capable of being refuted or verified on the basis of empirical data. They think it can, however. But no matter what happens, Sumner and DeLong are right. If the economy is performing well, then monetary policy is loose enough. If the economy is performing poorly, then monetary policy is not loose enough. No matter what the central bank does, Sumner can’t be wrong. In this respect, Sumner and DeLong are really pleading with the state to listen more to them rather than others. They’re like a free range chickens who feel so grateful about being free range as opposed to being cooped up that they want to advise the chicken farmers on how to best slaughter chickens, and are willing to pretend that their a priori theory is empirical like the state wants all economic theories to be before they will accept them, since they can’t experiment on the chickens with a priori theories.
I think Sumner especially is confused about how the market works. Sumner wants everyone to believe that the state can control individual subjective valuations. He thinks that with enough inflation, the state can control “NGDP,” and if they can control NGDP, they can control economic prosperity. The flaw there is easy to see. NGDP is a concept that includes an individual-based demand for money holding. As long as the Fed cannot control individual valuations, which is always, then they can’t control “NGDP.” It would be like expecting a particular volume of music to always lead to the same earplugs being used. Individuals do not, contrary to Sumner, respond the exact same way to inflation every time the Fed inflates.
What Sumner is really not saying, is that individuals are choosing to place “too high” a value on money for holding, thus nullifying the desired effects of inflation, which is of course “spending.” Doesn’t matter on what. If inflation can lead to a rise in spending money on weapons of war, or on centralization of healthcare, or on reducing the production of any other civilian good, which reduces prosperity, but nevertheless increases “NGDP,” then that’s good enough for Sumner.
Sumner and DeLong don’t understand that NGDP is not an object of individual action. Maybe Sumner can explain why an object that is not an object of individual action can possibly make individual’s lives better off, because it makes no sense to me. It would be like claiming that changing the number of stars in the sky will make the economy better off. NGDP is nothing but a statistical quantity abstracted from millions of individual choices. Yes, it is a singular statistic, and Sumner et al really believe that it can serve as a variable of control, and thus enable the chicken farmers to control what gives the variable meaning, as if controlling the effect can enable one to control the cause.
NGDP is the result of millions if not billions of individual choices and valuations. It has no existence apart from individual valuations. It is not a sum of money spent by an entity that “covers the costs” of “the economy” thus enabling “aggregate profitable investment” and “aggregate employment opportunities.” The chicken farmers who inflate cannot control individual valuations according to some constant causal operative process of “print and spend and make NGDP rise by 3% per year and they will come.”
This crude aggregate masks all the substantive process taking place in the micro-economy, which is run by individual valuations.
Sumner wants to blame the Fed for this because he’s too chicken to blame individuals choosing to hold more money. He’s too chicken to say this because he doesn’t want to appear as an ally to the Keynesian chickens on the other side of the farm. They’re not giving the chicken farmers good advice, you see. Sumner’s chicken gang should be the dominant one in the farm. So he’ll blame the farmer’s chicken feather plucking machine rather than the farmer’s degutting machine.
Sumner and DeLong should just admit that their theory about aggregate demand is a priori and not empirical. But my guess is that while they don’t want to be allies to each other, they certainly don’t want to be identified with the Austrian chickens, because the Austrian chickens aren’t grateful to the farmers in any way, because they know that there is no farmer behind the door at all, just a bunch of crazy chickens thinking they’re not chickens because they’re armed and have the ability to use force that is contrary to the health of chickens.