Should we worry about deflation in a Bitcoin standard world ?

A very interesting debate occurs last week between George Selgin and Saifedean Ammous about the concrete management of a deflation economy in case Bitcoin becomes the world’s monetary standard. Juice sumed up the points made by the two sides and adds some context about fractional reserve banking. He also states in his own view how Nash can reconciliate Selgin and Ammous’ views.

This made an excellent excuse for me to read some of Selgin’s work, which I have been contemplating for some time now. As he mentionned himself his book Less than zero: The Case for a Falling Price Level in a Growing Economy as a must-read to understand what he means in his debate with Saifedean, I read it first to have a balanced view as I’m already more familiar with Saifedean’s arguments.

I tried to figure things out and see if we should really worry about this kind of deflation, and what we can do about it.


In Less than zero, published in 1997, G. Selgin doesn’t contemplate to make table rase, but considers that as Central Banks’ monetary policies shifted from highly inflationnist in the 70’s to targeting “zero” or near-zero inflation in the 90’s, the case for not only zero but negative inflation, or deflation, to follow productivity rate of growth, should be considered as a next step. He tries to make a pragmatic arguments and doesn’t assume fundamental changes to the way things work in the current system.

On the other hand, Saifedean in The Bitcoin Standard advocates a telluric shift in the functioning of the global monetary system. Instead of the current fiat system in which the US Dollar serves as the world reserve currency, he imagines a Bitcoin standard quite similar to the old Gold standard in some aspects, with bitcoins being the reserve currency used between Central Banks for settlement, and commercial banks’ issuance of tokens or other money-substitutes (ie. that have a fixed value in bitcoins and are maybe redeemable in it) for everyday transactions.

G. Selgin answered to a review of Saifedean’s book, which quoted him doubting that the fixed, “hard-cap” supply of bitcoins makes it suitable for a monetary standard. Let’s see what it means by looking directly at Less than zero.

Selgin’s argument for a productivity norm

Selgin makes very clear that his point is to consider the respective merits of two kind of inflation targeting that could be pursued by Central Banks :

  1. The “Zero inflation” target : Central Banks control monetary base (ie. quantity of monetary units in circulation) so that the aggregate of output prices remains constant.
  2. The “productivity norm” target : while Central Banks still counter variations in output prices provoked either by a change in the velocity of money (people spend more or less for various reason) or by a “supply shock” (unexpected variation in the supply of one or many materials that drives the price of productive input up), BUT not if output prices’ changes are a consequence of productivity gains.

To put things another way, there are 3 factors that can modify output prices according to Selgin :

  1. velocity of money
  2. supply of productive factors
  3. productivity

If aiming at zero inflation, a Central Bank needs to counteracts any change in these 3 factors. With a productivity norm, it only needs to care about the 2 first, and let the prices fluctuates freely with change in productivity. Except for catastrophic events like war, productivity is steadily rising every year, meaning that all things being equal prices should also steadily deflate.


Let’s suppose an economy that produces 100 different outputs. Due to increase of productivity, the supply for one of them (let’s call it “oil”) increase, driving its price down. What will happen ?

  1. If Central Banks target zero inflation, it would have to expand the monetary base in the proportion of the expected impact of the price of oil on the price index, stimulating aggregate demand (the global demand for any good), so that the price index stay constant. Oil price still falls, but less than it would have otherwise, while the price of the others 99 goods and every input involved in their production increase slightly.
  2. In the productivity norm scenario, as long as decrease in the price of oil is due to productivity improvement, it doesn’t need any intervention from the Central Bank. Oil price will decrease, and the whole price index will also slightly decrease.

What about wages ?

If we remember the original tweet, Selgin was concerned that workers might not readily accept that their nominal revenue shrink, even if their real revenue is increasing (nominal prices are falling faster than nominal revenue). If I understood it well, the main problem here is psychological, as people would notice a fall in their wages far before they notice any price’s fall. There is a very simple problem of perception here : maybe we are in fact all better off, but individuals don’t have a global understanding of what’s going on and might perceive that they are worst off because their nominal revenue is less than before.

In his book, Selgin indeed discuss at lenght this issue. He makes the proposal to target price index through nominal revenue, or wages. It’s not the place to discuss this at lenght, but what is relevant for our subject is indeed that his proposal should makes nominal wages stagnate or slightly increase, not deflating at the same rate with prices.

Free banking

Selgin thinks that his proposal of free banking will makes this scheme possible. The main caracteristics of the banking system he advocates are :

  • Central Banks are still around, with the mission to stabilise prices according to the “productivity norm”.
  • Commercial banks run on fractionary reserve, without any legal requirement.
  • There is a reserve currency, created by the Central Banks and used exclusively in interbanking settlement, and commercial banks create their own supply of currency to be lend to the public.

He argues that the Central Bank only needs to inflate (or contract) the monetary base of the reserve currency to adjust nominal revenue to the variation of productivity it anticipated. As far as I can tell, Saifedean would not disagree with most of what is proposed here, except for two things :

  • If the reserve currency is Bitcoin, then Central Banks can’t inflate it at will like says US Dollar.
  • Fractional reserve banking is fundamental for Selgin, while Saifedean, without absolutely ruling it out, still makes clear in his book that he doesn’t believe it would be possible on a large scale with a Bitcoin standard, due to the transparency of blockchain transactions that should allow almost anyone to easily control that a bank is solvent.

Fractional reserve banking

Juice saw clearly that the possibility and desirability of fractional reserve banking is what is at stake here, so he defines what he means by it in his article.

We discussed FRB on twitter sometimes ago, only to quickly find out that we don’t agree on it being a “normal” part of the bank business, but as always it’s more because that we weren’t talking about the same thing.

Quoting wikipedia, he makes the point that Fractional-reserve banking allows banks to act as financial intermediaries between borrowers and savers, and to provide longer-term loans to borrowers while providing immediate liquidity to depositors (providing the function of maturity transformation). He points out that being against FRB would then mean being against bank loan, which is correct given the definition above, but also obviously silly !

I think this definition is below the point, because it neglects a very important element : availability. Deposit is money immediately available to the account holder and (most of the time) doesn’t pay interest, but there are other kinds of accounts like time-deposit, that pay interests but are NOT available to the holder (or at least should not be available) any time he wants it.

You perfectly can lend money without FRB, as long as the deposit money pay interest to the holder and is not available to him before term. The point of FRB is that the same money is available to two or more people at the same time, effectively creating money out of thin air. I don’t agree that a deposit account is fundamentally different from a bailment from this point of view, I think you can’t argue that the depositor lend money to the banks by making a deposit.

Anyway, I side with de Soto in his Money, Bank Credit, and Economic Cycles on this matter. Maybe we could discuss it more at lenght another time.

Few answers and still many questions

How bitcoins will circulate between Central Banks in a Bitcoin Standard world ?

If I understood well the productivity norm proposed by Selgin, Central Banks should inflate monetary based when prices fall for other reasons than productivity improvement. We suppose that Bitcoin is the standard for at least a significant numbers of countries in the world : what will be the dynamic of exchange between them ? Is it possible that a country that experiences more deflation than the others for some reasons could lend or obtain in someways bitcoins from other countries’ Central Bank ?

The Gold Standard : wouldn’t workers just be fine in a deflationary environment ?

Selgin makes himself the point that the so-called Great Depression of the last quarter of the XIXth century was nothing like a depression for most people. Prices index lost about a third of its value, but it appears that nominal revenue was stable and even increasing, resulting in a huge improvement of the real revenue, and that most people certainly felt it was an ere of prosperity more than crisis.

Of course this experience doesn’t translate perfectly to a Bitcoin Standard, because as Selgin rightly said under the gold standard banks had some room to inflate the monetary base that would probably not exist in a Bitcoin Standard. But as the benefice of the Gold Standard were obvious in restrospect, it is very likely that the benefice from a Bitcoin Standard would be even more terrific, so can’t we just imagine that people could just cope with harsh deflation from time to time ?

Nash’s Ideal Money

Juice proposed that the concept of asymptotical Ideal Money of the late John Nash could be a way to resolve the issues we have envisioning a Bitcoin Standard. I won’t talk about it in lenght here as I don’t feel confortable with it yet, but I still want to ask an earnest question : if Bitcoin is to be a common inflation target for fiat currencies, effectively stabilising them relatively to each other, on what kind of market the price of bitcoins will be decided ? Or to put it in another way, what information the price of bitcoins will give us ? Who will trade bitcoins and what for in this world ?

I also think that Juice’s point about Selgin not considering the eventuality that the US Dollar won’t be the world reserve currency anymore is right, and I would even say that when talking about a possible Bitcoin Standard, you have to assume a lot of things would be different than today.

Empty your boat

To conclude on a lighter note, this debate made me think about a story :
If a man crosses a river
and an empty boat collides with his own skiff,
Even though he be bad tempered man
He will not become very angry.
But if he sees a man in the boat,
He will shout at him to steer clear.
If the shout is not heard, he will shout again, and yet again, and begin cursing.
And all because someone is in the boat.
Yet if the boat were empty,
He would not be shouting, and not angry.
If you can empty your own boat
Crossing the river of the world,
No one will oppose you,
No one will seek to harm you

Sorry if it sounds a bit cheesy, but I think it could be the solution at least to the perception problem with deflation. People can indeed bears a grudge about their revenue shrinking if they know that somewhere, someone took the decision to cut it. But what if no one is in command ? Could people really get mad at… Bitcoin ? Sounds like madness, isn’t it ?

Bitcoin is not a conventional human undertaking, like a company or an institution. I think it has demonstrated so far its ability to develop in its own way, regardless of what people think, say or even attempt to influence it. We still don’t have a good theoretical understanding of Bitcoin, yet we are forced to admit that… it just works so far.

My point is that it is more likely that we all will need to adapt to Bitcoin than the other way around. We don’t know exactly what’s happening, except that it is happening, and that we don’t have much control other it. You have to acknowledge that when thinking about Bitcoin, and you might get caught on details and miss the bigger picture.

Leave a Reply

Your email address will not be published.Required fields are marked *

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.