Consider the following two claims:
1. Given that NGDP quickly returned to the pre-Covid trend line, the recent bout of high inflation was caused by supply problems.
2. Given the recent supply problems, the recent bout of inflation was caused by the decision to bring NGDP quickly back to the previous trend line.
If you see the world the way I do, it’s almost physically painful to read the media as they debate the cause of higher inflation. It makes my hair hurt.
What caused the recent bout of inflation? What do you mean by “cause”?
Here’s what I said in my recent book, entitled The Money Illusion:
Similarly, a government policy could be said to have caused the Great Recession if—in a very plausible alternative policy setting—the exact same policy tool that was already being used would have avoided the recession. As an analogy, a bus driver might be said to have caused the bus to hit a tree if a different position of the steering wheel would have avoided the accident. We don’t think of bus drivers as “solving accident problems,” though we hope they will not cause accidents.
Because I prefer to think of causation in terms of policy counterfactuals, I use the optimal policy as a benchmark. I believe that it was appropriate to quickly bring NGDP back to the previous trend line, and as a result I might say that supply problems caused the recent bout of higher inflation. I also expect NGDP to overshoot the trend line over the next 12 months. If this does happen, then I will view an excessively expansionary monetary policy as causing the excessive inflation that is likely to occur over the next 12 months.
On the other hand, if someone says that the previous 12 months of inflation was caused by an expansionary monetary policy, I certainly would not claim they were wrong; I’d simply assume they favored a less expansionary monetary policy. Or perhaps they define “causation” differently from the way I do.
The world is full of terms that have never been clearly defined. Truth, reality, science, monetary policy, inflation, and causation are six examples of fuzzy concepts. Show me an unambiguous definition of any of them.
READER COMMENTS
Andrew_FL
Nov 15 2021 at 5:41pm
Friedman and Schwartz blamed the Fed for the Depression not by comparing Fed policy to some hypothetical alternative policy, but because, as I understood them, in their judgement the pre-Fed institutional arrangement would have handled the crisis in a manner which would have prevented a degeneration into such an extreme monetary contraction.
I think you are correct in one important respect: statements about the relative stance of monetary policy are not so much statements of fact as articulations of one’s implicit policy norm. For my own part, however, any implicit policy norm that isn’t “strict laissez faire in money and banking” has to justify itself on other grounds, and neither you or any other advocate of state (or other coercive) intervention has done so.
rsm
Nov 15 2021 at 9:32pm
Why is inflation a problem if my income keeps up? Why not keep real purchasing power stable instead of nominal prices?
Andrew_FL
Nov 15 2021 at 10:26pm
I’m only going to say this once, rsm, but your reply to me has nothing to do with my comment, and this is not the first time.
Scott Sumner
Nov 16 2021 at 1:15am
“Friedman and Schwartz blamed the Fed for the Depression not by comparing Fed policy to some hypothetical alternative policy, but because, as I understood them, in their judgement the pre-Fed institutional arrangement would have handled the crisis in a manner which would have prevented a degeneration into such an extreme monetary contraction.”
But the pre-1913 system is a “hypothetical alternative policy”, as is the actual hypothetical policy that they advocated.
Mark Z
Nov 16 2021 at 4:54pm
If George Selgin and and Larry White (I can’t find the full text free anywhere) are correct, a free market in money would tend to stabilize nominal income, and the fed targeting NGDP is the best approximation of laissez faire in currency achievable in the status quo. There’s no reason to assume that the laissez faire policy norm is the money supply being fixed (or price level being fixed).
Andrew_FL
Nov 17 2021 at 2:18am
You are correct about what laissez faire in banking would look like, however the argument that it’s just as good if the Central Bank apes this result in a cargo cult manner is one for one identical to the argument of the market socialists that planning absolutely could work, the planner just needs to play market.
Mark Z
Nov 18 2021 at 1:57am
The contention is actually that a planner that tries to emulate the market does better than one that doesn’t. If the government nationalized the auto industry and sold cars at cost, and then there were a dramatic increase in demand for cars, factories running out, used cars being sold on the black market at higher than the official price, what would a good economist advise the state car company to do if asked, and with it understood that ‘privatize cars’ is not an answer they’ll accept? Advising them to make more cars seems like good advice. ‘Keep producing the same number of cars forever because any change in output is automatically distortionary’ doesn’t.
Hayek and others argue that the market makes use of diffuse information inaccessible to the state to allocate resources better than the state could. You seem to be taking the more extreme position that no information is available to the state that it can use in economic planning, and thus all production policies are equally ignorant and bad. Otherwise, it makes perfect sense to support some preferred rule or policy if one takes as given that the state’s control over the industry will continue. This is why even free banking economists, like Selgin I believe, favor NGDP targeting as long as the Fed is in control of money.
Andrew_FL
Nov 18 2021 at 7:28pm
No, a good economist would tell them that the answer is to privatize the car industry, period. If they don’t like that answer, the duty of the good economist is to repeat it. You repeat it even if they throw you in the gulag or put you on the scaffold. If you give any other answer, you are not a good economist.
Roger Sparks
Nov 15 2021 at 6:28pm
Isn’t this more of a common sense thing? After all, we had a covid led customer chill, a CB sponsored money supply reverse-chill, and multiple government sponsored direct payments to all hands. Now common sense brings me the the realization that we had three economic events, each with different effect on the supply chain.
As time passes, we are beginning to see the combined impacts on the supply chain, part of which is labor. We can already see that more money placed into the economy than what-was-earned-by-suppliers has stimulated demand greatly. We can also see that covid has caused some cases of decreased production. The combination of reduced supply and increased demand has led to a paradox of blanks in store shelves all-the-while shiploads of supplies lay awaiting. Likely the blanks do not well correspond to awaits, but there has to be some correspondence.
I am not sure how we get a good NGDP read under these circumstances. I think any number we get is severely perturbed. I think I would go even farther to say that it is illogical to think that the recovery in NGDP is caused by more rapid product sales; it is far more logical to think that the recovery in NGDP is inflation driven and therefore very harmful to identifiable economic participants, all-the-while being helpful to some.
Thanks for the thought provoking blog postings
Roger Sparks
Nov 15 2021 at 6:45pm
Yes, I am seeing that (reduced supplies and blank shelves and covid chill) * (higher prices) = higher NGDP.
MarkLouis
Nov 15 2021 at 7:47pm
Makes sense and I very much appreciate your consistency. I predict that as NGDP overshoots in the coming months, many of the self-proclaimed NGDP targeting proponents will find new reasons that the Fed should not tighten. That does not bode well for our ability to view economics as a science and/or trust that the Fed could ever be counted on to stick to such a framework. We shall see.
James
Nov 16 2021 at 12:11am
“Truth, reality, science, monetary policy, inflation, and causation are six examples of fuzzy concepts. Show me an unambiguous definition of any of them.”
Please consider that maybe you have already been shown unambiguous definitions for all of these and simply failed to recognize them. What is the least ambiguous definition of causation that you have seen and what prevents you from calling it unambiguous?
Scott Sumner
Nov 16 2021 at 1:17am
I haven’t seen a convincing definition of any of those 6 concepts.
James
Nov 16 2021 at 4:49pm
I never asked if you found some definition to be convincing.
Please consider: Maybe you have seen unambiguous definitions for every word you list but those definitions appeared ambiguous due to uncareful reading.
Scott Sumner
Nov 16 2021 at 6:27pm
Instead of playing word games, why not provide me with some clear definitions?
James
Nov 17 2021 at 8:53pm
You accuse me of playing word games while you fail to read, let alone answer, my question. I repeat: What is the least ambiguous definition you have seen for causation and what prevents you from calling it unambiguous?
I want to be clear right now that I am only responding to your claim that causation and a handful of other words have (past tense) never been clearly defined.
Everett
Nov 18 2021 at 12:24pm
This goes to the argument about the definition of “inflation” below. This is why we use modifiers to words such as “proximal cause” or “a cause of”.
I think the issue with “cause and effect” may be the unstated assumptions of blame and an expectation of a single “cause” to an event.
Anything that is foreseeable as well as controllable or alterable is a cause when it comes to discussing policy. Therefore gravity is not (yet!) a cause, but any proximal actions are a cause.
To refer to jj’s analogy about the bus driver and the deer, is the driver a cause of the drink spill? Sure, but the driver certainly isn’t to blame. And had the driver not swerved the drink likely would have spilled some other way anyway. The policy should be that the driver swerves, and possibly that fences are erected along the highway, and maybe construct overpasses or underpasses for the deer to cross, or reconsider entirely having a high-speed road in this location.
jj
Nov 16 2021 at 10:20am
I think I have a slight improvement on the bus driver / tree analogy.
Say a rabid deer jumps in front of the bus. The bus driver to expertly swerve and avoid the collision, but a passenger spills their drink on their lap. What caused the spillage? The passenger’s poor reaction time, the bus driver, the deer, the rabies? The lack of a fence along the highway?
MarkLouis
Nov 16 2021 at 2:23pm
Scott, I’d like to request a post which I think would be helpful for many. I’d like to hear your interpretation of what the FAIT framework entails. The Fed has left some details fuzzy on purpose but they’ve at least commented in speeches on most aspects of the framework.
I’ve gathered: 1) 2% PCE price level target long-term, 2) temporary deviations allowed but made up for such that we re-converge with 2% path, 3) lookback period uncertain but most recommend 4-6 years.
Some open questions: 1) lookback period, 2) are permanent deviations from the price path tolerated; if so is this asymmetric?, 3) how long is reasonable to expect re-convergence with 2% path, 4) are downward divergences ever tolerated?, 5) are negative and positive supply shocks treated the same way?
I know you prefer NGDPLT but it’s going to be hard to reach a new framework if we can’t even understand the current one. I see a ton of confusion out there even among monetary economists. Maybe you can clear it up or at least provide a starting point for the discussion.
MarkLouis
Nov 16 2021 at 2:30pm
I can even recommend a title:
What is FAIT and is the Fed adhering to it?
Scott Sumner
Nov 16 2021 at 6:31pm
I will do more such posts in the future. Obviously the policy is a bit vague, and so there’s always going to be an element of interpretation. In my view, the clock starts at the end of 2019, which means we are now above trend. The “flexible” part suggests the Fed will not slavishly use the policy, but rather use it as a rough guidepost. Employment also matters.
MarkLouis
Nov 17 2021 at 7:00am
Agree there is definitely some ambiguity by design but many aspects of FAIT have been commented on in speeches etc. so we should be able to get a basic idea of what was intended.
Also agree that employment is part of the mandate and it’d be great to understand what drives the framework there: when is monetary policy most/least effective in boosting employment and how does the Fed discern that?
MarkLouis
Nov 17 2021 at 8:18am
Another confusing example you could shed some light on. Many Fed members commented around the rollout of FAIT that a 4-6 year window was most appropriate. Yet, I see many monetary economists suggesting that we ignore 5-year inflation expectations and focus on 5y5y instead. I think I understand their theory, 5 years is temporary I suppose but how can that be reconciled with FAIT. I’ve never seen so much confusion around how the Fed works.
Mark Brophy
Nov 16 2021 at 7:09pm
Inflation should be clearly defined as an increase in the money supply but many people define it as an increase in consume prices without regard to an increase in assets. An increase in money supply always results in “asset inflation” or “consumer inflation”.
rsm
Nov 16 2021 at 9:15pm
《it is not clear what is gained by controlling the price level. If business cycles are caused by real factors rather than by things that are affected by the rate of inflation, then many of the reasons for controlling inflation vanish.》
https://onlinelibrary.wiley.com/doi/full/10.1111/j.1540-6261.1986.tb04513.x
(Fischer Black, “Noise”)
Jon Murphy
Nov 17 2021 at 8:17am
Inflation is defined as a rise in the general price level. If only consumer prices are rising or asset prices are rising, then there is no inflation. We need a way to differentiate between relative price changes and price level changes. Inflation is that way.
In other words, simply redefining inflation (especially in a way that may confuse cause and effect) doesn’t change the underlying situation
Scott Sumner
Nov 17 2021 at 11:16am
Mark, You said:
“Inflation should be clearly defined as an increase in the money supply ”
No, that’s a terrible definition. It would just add more confusion. The term currently applies to a rise in prices, so why make this more complicated? Then we’d need an entirely new term for a rise in prices.
Michael
Nov 17 2021 at 6:41am
Scott, here is a hypothetical for you. Let’s assume:
You are correct and NGDP is going to overshoot the trend.
The Fed, using its current approach and tools, wants to tighten.
What should the Fed do/not do? Immediate rate hike? Change its forecasts/statements and rely on Mr. Norris? How quickly should they be aiming to get back to trend? How to avoid too much of a response leading to undershooting?
What is the right way to think about employment – still far below pre-pandemic levels – right now?
Now is an interesting moment for the market monetarists after more than a decade of being (wrongly) criticized as just wanting loose money all the time.
Scott Sumner
Nov 17 2021 at 11:18am
Obviously forward guidance is the most important factor, but I’d adjust the policy instruments (QE and interest rates) enough to bring down TIPS spreads.
MarkLouis
Nov 17 2021 at 2:50pm
Hard to fathom why they wouldn’t do something along the lines of: begin to shrink the balance sheet and revise the $amount weekly until 10y inflation expectations are ~2%. They may not even need to accelerate rate hikes.
Feels like they don’t fully trust their tool kit – like any tightening might crash financial assets. Hard to explain their behavior otherwise.
derek
Nov 17 2021 at 9:34am
I think a big part of why the media’s discussion is so illogical is that some of the fiscal/”demand” stimulus created supply chain effects. A major example here would be the lengthy stay for unemployment boosters, massive cash payments and various other stimulus, as this has allowed (and is still allowing) some workers to remain out of the labor force. So even if it doesn’t make that much sense to think of this stimulus as “causing” inflation from a monetary perspective, I think that it very much did from a logistics perspective. Consider also the way that the booming housing/remodelling market has allowed blue collar workers to double-dip on unemployment and informal, unreported contracting. One reason the media’s discussion of inflation and supply chains and stimulus is so maddening may be an unwillingness or inability to grapple with issues like this.
MarkLouis
Nov 17 2021 at 2:54pm
Not directly related to your post but another question that I feel like economists talk past each other on. I hear frequently that we need to “run hot” to tighten the labor market in an effort to get real wages up. But this feels circular to me: stimulus-> lower real wages (via sticky wages) -> more demand for labor……but then how do we make the jump to higher real wages?
Scott Sumner
Nov 17 2021 at 7:52pm
Yes, the “run the economy hot to get higher real wages” is a silly argument, as you say. Wages lag behind aggregate demand.
Yaakov
Nov 17 2021 at 5:47pm
If I understand you correctly you are saying that given the circumsteances the Fed purposely generated the inflation of the less 12 months and was correct in doing so.
But is that really the case? Did the Fed really see the correct policy the way you see it and therefore did what you thnk they should have done? Or did the Fed have some other policy in mind which just happened to match your favored policy?
If inflation continues in the next 12 months, will that be a sign that the Fed’s policy is wrong? Or does the reason the Fed did what it did not matter as long as they did the right thing?
Scott Sumner
Nov 17 2021 at 7:54pm
Yaakov, I’d put it this way. The Fed was right to quickly bring NGDP back to the trend line, even at the cost of higher inflation. If NGDP continues rising fast, then Fed policy will be too expansionary. The Fed is perfectly capable of following a good policy with a bad one.
Everett
Nov 18 2021 at 12:09pm
This and the comments has really got me thinking, so I have a slightly off-topic question: Has anyone been looking into the effect of the market creation of over $2 trillion in new currency (cryptocurrencies) on inflation?
Comments are closed.