What does it mean to say that something is inflationary?
Tyler Cowen has a new post looking at the question of whether cryptocurrencies are inflationary. I’d like to take a stab at this question from another perspective.
Let’s start with the question of whether cryptocurrencies like Bitcoin are “money”. If you view Bitcoin as money, then it has been hyperdeflationary. Prices of goods and services in terms of Bitcoin have plummeted at a phenomenal rate over the past decade.
But that’s clearly not what people mean when they ask whether Bitcoin is inflationary. They are thinking of inflation in US dollar terms. In that case, asking whether Bitcoin is inflationary is sort of like asking whether gold, copper, or shares of Tesla stock are inflationary. So are they?
In the most simple possible model, crypto doesn’t seem to be inflationary because it doesn’t directly affect the dollar money supply. But what if it somehow boosts velocity? Is that possible?
Crypto could theoretically cause market interest rates to rise, which would boost velocity. But if the Fed is targeting interest rates then this will not occur.
Cryptocurrency is new and uncharted but new rules are rising up to define acceptable practices. The murky is becoming clearer! Be wise, seek help from an expert like a cryptocurrency investment coach, and avoid legal troubles while making a little extra cash. It always pays to be a pioneer in a new land, so don’t hesitate to take the right steps—just take it with care.
Crypto might raise the natural interest rate to a level above the policy rate, and hence boost inflation. This is actually the best theoretical argument for crypto being inflationary, but it is still rather implausible.
First, why wouldn’t the Fed offset this effect by raising the policy rate? More importantly, why would crypto have any significant impact on the natural rate of interest? The total market value of cryptocurrencies is roughly $3 trillion, similar to Apple’s market value. People don’t typically ask if Apple stock is inflationary. They might ask if the stock market as a whole is inflationary, but not a single stock.
I suspect the confusion here comes from the fact that crypto is viewed as a form of “money” (it is often called “coins” or “currency”), and people wrongly think that monetary models of inflation are about money. They are not. Monetary models in macroeconomics are about the medium of account, the asset in terms of which goods and services are priced. Most cryptocurrencies are not media of account, and hence are not relevant to monetary models of the inflation, NGDP, etc. (And they aren’t even much of a medium of exchange.) As for stablecoins, I presume the Fed offsets their impact.
People often argue that monetary offset doesn’t work perfectly. That’s obviously true, but the implication of this fact is frequently misinterpreted. At times, the Fed is too expansionary (1970s, 2021) and at other times they are too contractionary (1930s, 2008-16). But ex ante we don’t know what sort of mistake they will make, and hence a lack of monetary offset is not an argument for either an inflationary or a deflationary effect.
Tyler spends a portion of his post discussing how crypto might affect aggregate demand, but I’m not persuaded by his reasoning. He doesn’t tell us why this effect would lead the Fed to make a mistake in one direction or another. The Fed knows about crypto. Perhaps they overestimate its inflationary impact or perhaps they underestimate its inflationary impact. I don’t know which is true, and I suspect that you don’t know either.
Perhaps Tyler is not thinking about monetary offset, rather he is viewing his analysis as a sort of input into Fed decision-making. Advice on how they might need to adjust policy from the baseline in response to crypto, in order to achieve effective monetary offset. But what is the baseline? There is no such thing as the Fed doing nothing. There is no baseline, except perhaps 2% inflation. But that merely assumes the answer! So what does it mean to talk about anything being inflationary? Holding what constant?
Tags:
25. January 2022 at 12:49
Is cryptocurrency inflationary or deflationary in a free banking regime?
25. January 2022 at 13:08
I think Tyler is great, in general, but when it comes to crypto he has made a variety of bad arguments over the years. For instance, I’ve heard or seen him make the argument a few times that if people want to hold x% of their portfolios in crypto, then it has some implication for the price of Bitcoin a few times.
25. January 2022 at 13:12
I view crypto as deflationary most likely. As real rates fall and inflation rises above-target it’s natural for people to seek out inflation hedges. Finding a hedge that doesn’t impact the CPI basket is probably the best case scenario for central banks.
As a thought experiment imagine all that money decided to invest in physical commodities as an inflation hedge instead. That could be highly inflationary. There are some signs this is occurring in small commodities (uranium most notably) but hardly on the level of crypto.
25. January 2022 at 14:12
Hi Scott,
I think what most of us in the laity mean when we describe some event or trend X as “inflationary” is that inflation is higher with X than it would have been without X. So, for example, if someone claims that a set of supply chain disruptions are inflationary, what they mean is that if we had not had the supply chain disruptions, inflation would have been lower than it was with them.
This view is obviously naive in not appreciating the role the Fed has in determining the inflation rate, and that the Fed can offset whatever “inflationary” impact a supply chain disruption in at least three ways:
1. having an automatic mechanism like an interest rate target that results in OMO’s that offset the “inflationary” effect of X;
2. making concrete changes in instruments like a fed funds rate target that offset the “inflationary” effect of X; or,
3. creating market expectations of on-target inflation (or NGDP growth) that alter the behavior of market participants so as to offset the “inflationary” effect of X.
It’s #3 that’s probably most important, and it’s the aspect of monetary policy that the laity finds hardest to understand or accept. They are always looking for “concrete steppes” that have a clear causal link to the price level instead of appreciating the way aggregate demand is linked to spending decisions that are themselves linked to expectations of future NGDP or inflation, expectations that can be altered without any change to the interest rate target or the size of the monetary base (“concrete steppes”).
$0.037 (adjusted for inflation)
-Ken
25. January 2022 at 17:50
I liked this post.
I’m confused though about this:
“I suspect the confusion here comes from the fact that crypto is viewed as a form of “money” (it is often called “coins” or “currency”), and people wrongly think that monetary models of inflation are about money. They are not. Monetary models in macroeconomics are about the medium of account, the asset in terms of which goods and services are priced.”
It reads to me like the concept of “money” is being distinguished from “medium of account”.
I thought they were the same thing.
What am I missing?
25. January 2022 at 18:06
John Hall wrote:
“I think Tyler is great, in general, but when it comes to crypto he has made a variety of bad arguments over the years. For instance, I’ve heard or seen him make the argument a few times that if people want to hold x% of their portfolios in crypto, then it has some implication for the price of Bitcoin a few times”
I am only a layman but I think it only looks like a bad argument if you’re working from a bad ‘equilibrium’ model that omits the activities in getting to any given equilibrium, which is itself never really fixed in reality, it’s always in flux.
The scenario of “people wanting to hold X% of their portfolio in crypto” doesn’t just happen instantly, there needs to be buying and selling of bitcoins for that X% to be reached from wherever it was prior. That path of changes in the supply and demand for bitcoins will necessarily affect the price.
25. January 2022 at 22:20
Ryan, Banking is not the issue, the issue is monetary policy. If by “free banking regime” you have in mind some sort of private monetary regime (such as a gold standard, or competitive private fiat currencies), that is a different question. I suspect that if the Bank of Canada decided to go out of business and completely deregulate banking then Canadians would adopt the US dollar. Would that be free banking?
If the Fed closed up shop I’m not certain what would happen.
I’d need to know what sort of monetary regime emerged under free banking to answer that question.
John, Isn’t the supply of Bitcoin pretty inelastic in the short run? So if the world as a whole decides it wants to hold 0.2% of its wealth as Bitcoin, rather than 0.1%, doesn’t that raise Bitcoin prices?
Effem, You seem to be ignoring monetary offset.
Ken, You said:
“I think what most of us in the laity mean when we describe some event or trend X as “inflationary” is that inflation is higher with X than it would have been without X. So, for example, if someone claims that a set of supply chain disruptions are inflationary, what they mean is that if we had not had the supply chain disruptions, inflation would have been lower than it was with them.”
That’s also my view. Exactly my view.
You said:
“This view is obviously naive in not appreciating the role the Fed has in determining the inflation rate, and that the Fed can offset whatever “inflationary” impact a supply chain disruption in at least three ways:”
That’s where we disagree. I think that your interpretation DOES take the Fed into account, or at least should in principle. Consider the analogy of the multiplier effect in the Keynesian fiscal model. When Keynesians say that a $100 increase in G causes GDP to go up by $250 dollars, they are assuming that the fiscal policy causes other actors in the economy to adjust their behavior. Thus when considering the effect of X you should consider how other actors, including the Fed, respond to X.
If you compare the inflation rate under X and not X, you must account for monetary policy under X and not X. Otherwise it’s impossible to make any comparison.
Farrell, My point is that calling something a “coin” or a “currency” leads people to view it as money. But it’s not a medium of account. If you want to say that money must be a medium of account, then you are correct. But most people seem to view cryptocurrencies as a form of money.
25. January 2022 at 23:04
The discussions about cryptocurrency, and the product itself, feel like they haven’t evolved much in 10 years. Contrast that with the changes we’ve seen with information technology, mRNA vaccines, and the Republican party. If we’re lucky, the James Webb telescope will even start working soon.
I don’t regard the proliferation of new digital currencies as a positive development. At least with commodity money there is a long history of agreement that certain rare metals have value, although I prefer a fairly stable fiat money system. Blockchain technology is interesting, but still remarkably cumbersome. I’m still annoyed by the chip debit card issued by my bank because it adds about a half a second to payment transactions at the supermarket.
26. January 2022 at 05:55
Scott,
There’s a bit of a bait-and-switch. People want to hold 0.2% of wealth in crypto, up from 0.1%, then that doesn’t mean they want hold 0.2% in bitcoin. Some of the increase might be from bitcoin, but not necessarily all. The increase in the share of wealth in crypto can be due to innovation in cryptocurrencies offered, rather than a shift in the demand for bitcoin. As a result, the elasticity of bitcoin isn’t the same as the elasticity of all crypto. For instance, they are only making a limited amount of additional bitcoins, but I can start bitcoin 2.0 that is exactly the same except with a different name.
26. January 2022 at 06:26
Scott, I don’t trust monetary offset as much as I used to with the current Fed at the helm. For all their “tough talk” inflation expectations are still considerably higher than they should be.
26. January 2022 at 07:17
I think that many people asking “is Bitcoin inflationary” are asking in the same sense as “Is the fact that most Turkish citizens have access to US Dollar bank accounts inflationary?”
If Turkish citizens did not have access to US Dollar accounts, then it would be easier for the central bank to do capital controls and prevent the Lira from spiraling downward.
Now, of course, if the Turkish central bank was actually doing a good job of inflation targeting this would be irrelevant. But given that the actual Turkish central bank is beholden to Erdogan’s love of “Islamic finance”, the US dollar is “inflationary” for Turkey.
Now, suppose that the president the USA started appointing people to the Federal Reserve who believed its mandate was not “price stability” but rather “combating climate change” or “opposing structural racism”. In such a world, Bitcoin might be “inflationary” because it limits the ability of the Federal Reserve to coerce savers.
26. January 2022 at 09:00
David, I’m sort of agnostic on the question of crypto’s value to society. There are big costs (electricity, use in extortion), and perhaps big benefits, which are hard to quantify, but certainly real.
John, I agree, but I still find the general claim to be plausible in a directional sense, even though it’s hard to estimate the size of the effect.
Effem, A year ago I was bombarded with commenters saying monetary offset was wrong because the Fed was allowing TIPS spreads to fall too low. Now the reverse. Isn’t it possible the Fed is making mistakes?
Logan, Yes, see my next post.
26. January 2022 at 11:07
Hi Scott, you wrote:
> That’s where we disagree. I think that your
> interpretation DOES take the Fed into account,
> or at least should in principle.
Nope, no disagreement. If you and I used the term “inflationary”, we would absolutely take the Fed into account.
We’d argue that under an omniscient fixed-rate-inflation-targeting central bank, nothing could ever be inflationary, because Offset. Under an actual CB, for some X to be inflationary, X must cause the CB to either raise or overshoot its inflation target. When you and I are talking, a supply chain disruption could even be deflationary, if we thought the CB would overreact.
I am confident this is not what the lay press or most non-economists mean when they use the word “inflationary”. They’re probably ignoring the CB, or if not, they’re imagining any CB response won’t matter for a long time (the “long and variable lags” mistake). It’s more “concrete steppes” thinking: “a supply chain disruption must be inflationary because the same amount of money will be chasing fewer goods,” ignoring the CB’s power to control AD.
26. January 2022 at 11:34
Scott, we have 7% inflation. This is not like a mild undershoot.
26. January 2022 at 17:30
Scott, also note that the argument is about the share of crypto assets in investors aggregate wealth, not the dollar value of crypto assets. It really only is a statement about prices of crypto assets minus non-crypto prices. If non-crypto prices fall but crypto prices fall by less, then the share can increase. However, crypto has been more volatile.
There are also other possibilities about risk preferences or expected volatility that could lead to movements in shares.
26. January 2022 at 21:45
“My point is that calling something a “coin” or a “currency” leads people to view it as money. But it’s not a medium of account. If you want to say that money must be a medium of account, then you are correct. But most people seem to view cryptocurrencies as a form of money.”
Ok thanks, that makes sense.