Why I don’t post on crypto

Tyler Cowen recently noted that monetary economists don’t have much useful to say about crypto:

But take [monetary economics] in general — has it had anything interesting to say about crypto developments? I don’t expect it to have predicted crypto, or its price, any more than I expect macroeconomists to have predicted recessions (see Scott Sumner on that one). But surely monetary theory should be able to help us better understand crypto? And its price.

Money is one of the most ambiguous terms in the English language. What does it mean to say, “Bill Gates has a lot of money”? That he has a fat wallet full of dollar bills? Or a lot of Microsoft stock? The field of “monetary economics” might better be described as the field of “medium of account economics”. Because crypto is not a medium of account, our models are not very useful in explaining changes in its value. if you want to learn about it, you can visit the latest posts at Funfair’s Blog.

Some people might say, “Wait a minute, crytpo has many money-like attributes. It is often used as a store of value. Like fiat currency, it has no intrinsic value. Occasionally, it is even used as a medium of exchange.” Yes, all that is true. And any true “theory of money” should have something useful to say about crypto.

But the field that economists call “monetary theory” is not a theory of money, it’s a theory of the medium of account. And crypto is not a medium of account. Sorry.

PS. What about stablecoins? Well, we do have models that explain changes in the value of stable coins. Those are the models that explain price inflation (not always very well, but no worse with stablecoins than with other media of exchange.)



20 Responses to “Why I don’t post on crypto”

  1. Gravatar of Kester Pembroke Kester Pembroke
    21. November 2021 at 13:46

    Sorry for irrelevant post but what is your response to this criticism of monetarism.

    The whole point of excluding ‘public deposits’ from the totals was because that doesn’t have to circulate. But apparently everything else does because ‘interest rates’.

    Yet in reality it doesn’t. There is no difference between a bill in the mint and one in a wallet. Both of them just haven’t been spent yet.

    They are trying to predict a flow level from a stock amount based upon a belief, and that is supposed to show something in the real economy. The evidence from reality is that it shows nothing. It’s a complete waste of time that has distracted people for 50 years.

    The national debt is the counterparty of the national financial assets. They are equal.

    In MMT we call it the ‘net financial assets of the private sector’. It’s a stock. You can tell this by the denomination. It’s in dollars, pounds, yen, etc.

    It’s nothing to do with money in circulation. That is a flow characteristic and depends on how much of the nation’s financial assets are changing hands at the moment. The denomination of that is different: dollars/month, pounds/week, yen/annum.

    The difference is the same as the difference between miles and miles per hour.

    The much beloved-by-monetarists ‘M’ series calculations try to take the entire national debt and work out what ‘money in circulation’ there is from that using various series. They are all wrong because they try to predict a flow level from the level of a stock.

    What they assume is that nobody holds money for its own sake. They only hold it because they are induced to do so by an interest rate. And that isn’t how people work.

    People hold money for status and insurance purposes, and the lack of spending that money represents is largely why there is unemployment.

    The calculations of the amount of national debt, currency quantities, bank deposits and the rest largely miss the point. It’s the flow that matters, not the stock.

  2. Gravatar of Matthias Matthias
    21. November 2021 at 15:08

    Scott, from what I can tell, if you want to read some economics applicable to crypto, George Selgin’s work on currencies and especially privately issued currencies is very apt. Especially the so called stable coins.

  3. Gravatar of Effem Effem
    21. November 2021 at 15:10

    From a strictly economics standpoint I understand where you are coming from. But economics cannot be separated from politics and the Fed should know this. Seeing a new asset gain $4t of value, largely driven by a narrative that central banks have abused their privileges, should at least get their attention. Further, letting inflation dramatically overshoot at the same time strikes me as a very bad idea.

    The Fed is but a few political surprises away from losing independence. I’d expect them to know this and take their public perception more seriously. At the very least they should get inflation under control and stop pretending they have any ability to forecast it.

  4. Gravatar of ssumner ssumner
    21. November 2021 at 16:51

    Kester, I did a critique of MMT a few months back in a two part series over at Econlib, if you are interested.

    Matthias, Yes, Selgin is very good on that topic, although I don’t believe he tries to explain changes in the price of cryptocurrencies. (Which is probably a good thing.)

    Effem, You said:

    “Seeing a new asset gain $4t of value, largely driven by a narrative that central banks have abused their privileges, should at least get their attention.”

    How do we know this is true? (Not saying it isn’t, I’m genuinely curious.)

    And yes, the Fed should stick to its 2% FAIT.

  5. Gravatar of anon anon
    21. November 2021 at 20:41

    This is off-topic, but I was looking at some papers on the minimum wage and I found something fascinating, and I was wondering your views on this issue.

    In 1968, the unemployment rate was around 3.8%, and the Black unemployment rate was around 6.6%, yet the minimum wage was around 70-80% of GDP per Capita around the, with the minimum wage being $1.60 and the GDP per capita being around $4696.

    Today, the minimum wage is much weaker, only around 26% of GDP per capita, and yet the differential between White and Black unemployment rates are much bigger than in 1968. Does this imply that the minimum wage doesn’t affect unemployment as much as other factors like Welfare and UI?

  6. Gravatar of Carl Carl
    21. November 2021 at 21:14

    I just read this article by George Selgin, https://www.capitalismmagazine.com/2021/11/what-money-is/, explaining why money is neither a store of value, nor a medium of account, but simply a medium of exchange. I have to confess, I’m not sure whether the two of you are in agreement or disagreement about the definition of money.

    For my part, I like thinking of the field as the study of the medium of account. It sounds like all you’re doing is studying how to convert imperial units into metric units, but then when I try to nail it down it ends up sounding more like the study of how to establish the relative values of millions of things which have different and fluctuating values for millions or billions of people while doing as little as possible to influence those values, which, of course, sounds impossible.

  7. Gravatar of David S David S
    22. November 2021 at 03:36

    I’ve noticed that Scott avoids posting on any type of volatile asset–except the Bucks.

    Based on the conditions set for crypto being worthy of commentary or analysis as “money” it’s going to be a long wait for it to be relevant to monetary policy. The Fed could create some sort of digital currency, but I doubt it would be based on current blockchain technology. I also doubt that will happen anytime soon, but it would be nice to see some improvements to the way electronic transfers are handled—which doesn’t do anything to change how our current monetary system is structured.

  8. Gravatar of Michael Rulle Michael Rulle
    22. November 2021 at 07:49

    I am trying to understand why, for example, the creator of Etherium believes it has utility relative to government fiat currency. I speak with crypto people frequently—and one thing they do not discuss is why it has marginal utility for society. Of course, it is possible an economy will grow around it—-where it is a medium of exchange. For example, there is reason companies cannot be entirely driven within an Etherium world. But I do not understand the appeal—and my suspicion is many who are involved do not either. They certainly are not good at explaining it.

    Reasons are given–“it disintermediates” and is cheaper. Maybe. But the overnight “repo” rate is 5-10%. I know many who are absolutely excited about it–but cannot explain why. Supposedly, the transaction capacity of blockchain per day is a fraction of banks.

    I think it is an interesting technology looking for an application—-and in fact “apps” are the driver of etherium—but so far they consist of buying, selling (of Etherium!) and lending and borrowing for businesses. I do not know the name of one business —but there must be some. Perhaps in countries with bad and weak banking systems is where it may have the most utility.

  9. Gravatar of Michael Rulle Michael Rulle
    22. November 2021 at 07:51

    PS–I need to re-read before hitting submit–wish you had an edit function–“there is no reason” (line 5-6 above) not “there is reason”

  10. Gravatar of Michael Rulle Michael Rulle
    22. November 2021 at 07:56

    The “market”, versus “commentators”, seem to like Powell. Good that he was reappointed. It would be amusing if Dems thought he was more likely to be inflationary—-he might be—because he is less likely to tighten too early. Really don’t care the reason—-just glad he was reappointed.

  11. Gravatar of Michael Rulle Michael Rulle
    22. November 2021 at 08:30

    Re: Market Cap of Cryptocurrencies

    https://coinmarketcap.com/1/ States total Market Cap is 2.6T. 70+ percent is Bitcoin and Etherium—of the 14,623 Cryptocurrencies trading on 431 exchanges. I do believe it is more likely their utility will exhibit themselves outside the first world—but that is a guess.

    Paul Tudor Jones likes Bitcoin over Gold—why not? But if that is all it is, it is not much.

  12. Gravatar of Michael Rulle Michael Rulle
    22. November 2021 at 08:50

    Re: market capitalization

    It is an interesting concept. It is not what can be exchanged into M0, M1, M2 or M3. For example, it is estimated that the value of all gold in the world including reserves is 100Trillion—–except if you try and sell it all—same with MSFT. Market cap is an equilibrium value where supply and demand meet at a given price measured in Money as a unit of account. I think.

  13. Gravatar of ssumner ssumner
    22. November 2021 at 09:27

    anon, You asked:

    “Does this imply that the minimum wage doesn’t affect unemployment as much as other factors like Welfare and UI?”

    The current federal minimum wage ($7.25/hour) is below the equilibrium wage, and hence has almost no impact on employment. So the answer is “yes”.

  14. Gravatar of ssumner ssumner
    22. November 2021 at 09:34

    Carl, George said:

    “Consider the case of Brazil in 1992. In that year alone prices expressed in Brazil’s official currency unit, the cruzeiro, rose more than tenfold. But rather than express prices in cruzeiros, which would have meant changing them daily, if not more than once a day, hotels, restaurants, and many other businesses switched to posting prices in dollars.”

    Here’s how I would respond to that. If you want to model the Brazilian price level in cruzeiros, i.e., the price level that rose tenfold, then cruzeiros are money. If you want to model the Brazilian price level in US dollars, then dollars are money.

  15. Gravatar of Carl Carl
    22. November 2021 at 10:16

    Can I take that a step further and say if I want to model the price level in gold, then gold is money? (I promise not to ask about crypto or other commodity money next.)

  16. Gravatar of ssumner ssumner
    22. November 2021 at 10:34

    Carl, Yes, I wrote a whole book with that theme, called “The Midas Paradox.”

  17. Gravatar of Carl Carl
    22. November 2021 at 11:05


  18. Gravatar of fnietom fnietom
    23. November 2021 at 02:40

    Given NGDPLT based on prediction markets is an incentive compatible system, did you consider a cryptocurrency may be the easiest way to make it a reality?

  19. Gravatar of Ryan T Ryan T
    23. November 2021 at 10:09

    @Michael Rulle

    The end game of crypto is complete decentralization of the internet (and possibly more).

    Trust for centralized systems is falling, you are at the whims of the owner and successful systems become too valuable to remain independent.

    Until now there was no alternative, but with crypto you can have what is effectively an ownerless protocol that is powered by people viewing or participating in it.

    Its not clear what those communities will look like. Its easy to imagine decentralized twitter, but we can probably do better. The point is that the community will be mathematically impervious to tampering, censorship, and government surveilance. There will be no server you can bomb or owner you can compromise.

    When such communities become a part of (or replace) the existing internet they will need to use a decentralized currency.

    Tyler is right when he says there are too many smart people who are obsessed with making this reality happen. And after Bitcoin most of them are rich.

    A future without crypto will probably end with every government achieving China-like ubiquitous surveilance. Whether or not you agree, for true crypto believers the stakes are borderline existential.

  20. Gravatar of ssumner ssumner
    23. November 2021 at 22:41

    fnietom, I’m not sure it makes sense to base NGDPLT on prediction markets. I’d use ordinary futures contracts.

Leave a Reply