Supply Shocks
David Wessel Hmm. Another question was what your view is about, what role did fiscal policy played in causing the inflation that you've been working so hard to restrain? How big an issue was fiscal policy and how big an issue is fiscal policy now, as you try and calibrate the right pace of monetary easing?
Christopher Waller Well, just from a, it's just a simple macroeconomics point of view. If you're going to increase the spending in the debt [sic?] by $6 trillion in a matter of two years, and then say that has no effect on demand, that seems just impossible to me. It isn't the only thing that contributed to the inflation, but it certainly has had to have had an impact. The reason I say that is, you know, people have been talking a lot about, oh, all the last six months shows this was all supply, all supply, all supply. Well, if these are temporary supply shocks, when they unwind, the price level should go back down to where it was. It's not.
Go to Fred. Pull up CPI. Take the log. Look at that thing. The level of inflation [sic?] is permanently higher. That doesn't happen with supply shocks. That comes from demand. And this was a permanent increase in demand and permanent increase in debt. So I think there clearly was in fact a fairly...
I am just going to go after the part of the text that I highlighted. Firstly, the “level of inflation being permanently higher” is either a mis-statement or not carefully thought through. So I will not sweat his exact wording and try to guess what he meant. (Why is his wording a problem? The inflation rate went up, and has since gone down. Although it might still be higher than previous averages, that would imply that one could claim that “inflation is permanently higher” every single time inflation rose in the 1950-2024 period. If he meant that “the price level is permanently higher,” well, that is how a 2% inflation target is actually supposed to work.)
My concern is the Economics 101ism of the analysis, which is all over the discussion of this topic. In the world of Economics 101, everything in the economy is the result of two lines intersecting. All discussions revolve around what happens when one — and only one — of those lines are allowed to move. (This can often devolve into a debate whether one of the lines is horizontal or vertical, which is Peak Economics 101.) In this case, there has to be either a supply shock, or a demand shock. It is forbidden that “demand” and “supply” change at the same time.
Although I am mere blogger with almost no academic training in economics even I was able to notice that the pandemic lockdown period saw the simultaneous impositions on the ability to produce goods (“supply shock”) as well as fiscal transfers designed to avoid Consumer Cash Flow Armageddon (“demand shock”). This was then followed the Russian invasion of Ukraine (supply shock), etc. Deaths, early retirements, and the shutdown of worker migration also created a supply shock in the labour market. Although I saw a lot of commentary putting weight on supply side factors for the inflation, I cannot recall any serious commentator saying that nothing happened on the demand side.
Meanwhile, the idea that the price level/inflation must go back to its previous level after the shock passes is remarkably static thinking. The supply shock also hit the labour market, and wages broadly rose. Empirically, higher nominal wage incomes implies higher nominal household expenditure — sustaining the higher price level. This is not like The Great Canadian Cauliflower Panic Of 2016 where there is a shock to a single commodity that has extremely limited effect on other prices; the whole structure saw price increases. The higher wages allows the economy to function with higher prices across the board, which is why we have aggregate wage and price levels in economic models in the first place.
Inflation is only supposed to head back down after the various markets sorted themselves out. Given the empirically demonstrated persistence of inflation, a model that predicts that inflation exhibits immediate steps up and down in response to “shocks” can easily be rejected.
Comments like this are why I think that the whole “supply versus demand” debate needs to be put out of its misery.
Side Rants
This episode also provides an interesting view into inflation psychology. The entire neoclassical enterprise is premised on everybody internalising the central bank target, and doing sophisticated calculations related to it. But back in the real world, I keep having conversations with/reading texts by people who are mad that prices are not going back to where they were in 2019. Given that Canadian inflation had been rising at an average close to the target 2% per year (and more if you believe the inflation truthers!) between 1990-2020, if people were really aware of what was happening to inflation, why would they ever expect prices to return to their previous level?
As a final aside, the somewhat garbled comment on debt could be turned into a little MMT rant. Deficits are outcomes of economic developments, and the exact monetary amounts are somewhat arbitrary in the sense that what constitutes a “stimulative” deficit depends on a lot of factors. (Some might object to the use of dollar amounts instead of a more sensible scaling versus GDP. I think the use of dollar amounts should be avoided in written communications about aggregate fiscal policy, but using them in verbal remarks is going to be more natural for someone who stares at economic data all day.)
Reserves Comments
Another comment by Waller caught my eye.
Christopher Waller Yeah. I mean, I made an argument for probably ten years. There's no economic theory that tells you how big a Central Bank's balance sheet should be. I know of no theory that tells you. You have Switzerland where, it's basically 100% of GDP or some number like that. So there's no real theory. And from a point of view of the reserves, I love a floor system because, as Milton Friedman once said, you want to put enough liquidity in the system that you satiate the system. So there's no scarcity or shortage.
I am in the process of going after another dubious theory of Milton Friedman (should be published next week), so I have no choice but to comment.
To the contrary of Waller’s assertion, there is an economic theory about the size of the central bank balance sheet, or at least part of it. The central bank’s balance sheet is equal to notes and coin holdings plus reserves held at the central bank. Notes and coins is literally a money demand function found in any mainstream monetary textbook (although one might debate how accurate the models are).
As for reserves, they are arbitrary. A banking system can function perfectly well with exactly zero reserves held overnight at the central bank. Reserve holdings are either a tax on banks (required reserves) or are pushed in by central bankers to replace short-dated Treasury securities (which they are economically equivalent to from the perspective of the private sector). (Admittedly, there are markets where banks hold reserves based on convention. Since conventions are arbitrary, the reserves required to meet them are arbitrary.)
If I have $100 in my bank account and my idiot bank insists that I hold that $100 “in reserve,” I cannot use that $100 to pay any incoming expenses. I have a stranded, illiquid asset that pays whatever interest rate my bank decides to offer on it. In the case of bank reserves, replace “idiot bank” with “idiot central bank” and you immediately see the issue with reserves that Economics 101 textbooks ignore.
If banks truly want to hold reserves at the central bank and not lend in wholesale money markets (including the interbank market), this is not a “money demand” issue. It is a signal that the central bank once again failed its core responsibility to properly regulate the banking system. Letting the banks bypass wholesale lending markets by using the central bank as an intermediary is just putting band-aid on a major wound. How many band-aids you apply is a secondary concern relative to dealing with the real problems.
Appendix: My Turn to Strawman!
Having just objected to a strawman argument about supply shocks, I will then turn around and hit my own strawman. To be clear, this is not based on the Waller interview. Rather than create a new article, I will just throw this little rant (based on some recent comments I saw elsewhere) into this appendix.
I have seen comments to the effect that the rise in nominal GDP tells us that we would have an inflation problem.
Any time someone uses nominal GDP in such a context, please return to the following (approximate) identity. (You could use logarithms to be exact.)
Nominal GDP growth rate = (deflator inflation rate) + (real GDP growth rate).
Look at that equation. Ask yourself this: under what circumstances will the deflator inflation rate go off in a completely direction than nominal GDP growth (ignoring very short period rates of change, where it does)? How often do we see this medium-term divergence in practice?
After that exercise, one can then question why someone would suggest using nominal GDP trajectory as a cause of inflation?
No comments:
Post a Comment
Note: Posts are manually moderated, with a varying delay. Some disappear.
The comment section here is largely dead. My Substack or Twitter are better places to have a conversation.
Given that this is largely a backup way to reach me, I am going to reject posts that annoy me. Please post lengthy essays elsewhere.