So, after a week's hiatus, mainly due to busyness but also due to paying too much attention to the election, I'm back with another academic article review :)
This time, we'll be taking a look at "Money is an experience good: Competition and trust in the private provision of money" (2012), by Ramon Marimon, Juan Pablo Nicolini, and Pedro Teles.
The paper is published in the Journal of Monetary Economics. JME, as its name suggests, is one of the top journals in the world about monetary economics, so this paper's got some serious cachet!

Review

This paper asks a question that should be of interest to any Bitcoiner: "Can privately issued currency be efficiently provided by a competitive market for currency?"
This question is of interest to Bitcoiners because Bitcoin is exactly that: privately issued currency competing in a marketplace of currencies (shitcoins, fiat, gift cards... these all count.)
This paper shows that competitive markets cannot generally provide currency efficiently unless the currency issuers can commit to a future monetary policy path (i.e. they can commit to no inflation, or a fixed and scheduled inflation of the money supply.)
Now, what does that freakin' sound like? It sounds like Bitcoin! Bitcoin is a privately issued currency with a fixed monetary policy path. As long as Bitcoin's commitment to its predefined path remains credible, Bitcoin allows for an efficient monetary equilibrium to emerge. (More on credible commitment later.)
The argument is basically that without a credible commitment mechanism, currency issuers will always have an incentive to inflate the money supply once some people in the economy are holding their currency. They have an incentive to do so in order to reduce their nominal liabilities. Even if an issuer has no plans to inflate the money supply, they cannot credibly signal this to the market. Thus, zero nominal rate (a condition for efficiency) cannot be sustained in a competitive equilibrium without credible commitment.
This brings us back to Bitcoin. How does Bitcoin credibly commit to its monetary policy path?
In theory, there is no credible commitment mechanism, even in Bitcoin, because the core developers may at some point in the future decide on a new monetary policy path, for example by issuing tail emissions.
In practice, however, Bitcoin's decentralization offers users some assurance of monetary policy stability. This leads us to the debate on ossification. As Jimmy Song has discussed, Bitcoin is not just a technology, it is money, and this paper confirms Jimmy's intuition that ossification is an important feature of high quality, reliable money.
Since I started reviewing academic economics papers related to Bitcoin, I've been pleasantly surprised at how often they confirm many of Bitcoiners' key intuitions, including Bitcoin's utility for international transfers, Bitcoin's disciplining effect on government deficits, and now the importance of Bitcoin's fixed supply schedule.
More and more, the evidence is piling up that Bitcoin is not just sound money in the minds of Bitcoiners, but it is sound money according to economic theory.

Do the authors understand Bitcoin?

Irrelevant, as the paper is not directly about Bitcoin.

Takeaways for Bitcoiners

  • Competitive currency markets are not efficient unless issuers can credibly commit to future monetary policy paths.
  • Bitcoin's decentralization offers some credible commitment about its future monetary policy path.
  • Ossification is an important feature of high quality money, and high quality (trustworthy) money is needed for efficient currency markets.

Selected quotes

(Some quotes modified for readability)
Can currency be efficiently provided by competitive markets? A traditional laissez-faire view--as has been expressed, for example, by Hayek--argues that competition drives the price of money (the nominal interest rate) to zero and private provision of currency is efficient.
We show that there is a major flaw in this competition argument, when applied to fiat money: If suppliers of currency cannot commit to their future actions, then competition loses its bite. Once agents hold a particular currency, there may be an incentive for the issuer to inflate the price of goods in terms of this currency, thereby reducing its outstanding liabilities.
Does competition still drive promised rates of return to the efficient level? Not if issuers of currencies are not trusted.
With commitment, currency competition achieves the efficient (Friedman rule) monetary equilibrium, as Hayek envisioned.
Without commitment, negative inflation cannot be sustained... (and) Hayek's conjecture that efficient monetary equilibria can be achieved through currency competition is not verified.
In one sense, money is also an experience good. The quality of money is the amount of goods that money can buy, which can only be observed ex post... The mechanism that can sustain high quality is trust, not competition.
Are the authors only talking about centralized currency issuers?
The competition point seems different if we're talking about something like private mints creating gold or silver coins, than if we're talking about private issuers of notes that are only redeemable with the issuer.
reply
No, they're talking about private issuers, creating fiat notes.
They do discuss the presence of a centralized issuer in one section. Unsurprisingly that leads to inefficient outcomes even with commitment.
reply
Are they all issuing fiat notes of the same currency?
What I meant by centralized issuers, was private issuers of notes that are only redeemable with them: i.e. BoA issues a 2 ounce gold note. No other bank is obligated to honor that redemption.
reply
I think that's the way it's modeled.
In the sense that, one issuer cannot offer to redeem the notes of another issue at a different price. Each issuer controls the price of their own notes.
reply
Ok. I believe all of these different institutions have existed, so it might be relevant to allow for them in a model of competing currency issuers.
The findings still seem fairly intuitive.
reply
Yes. I think the contribution was considering whether or not there are dynamic reputation strategies that could lead to an efficient outcome even without full commitment, and the answer is no. That part was a bit over my head though.
reply
It makes sense to me. Reputation offers a persistent opportunity for rent seeking. Eventually, someone will attempt to capitalize on that and they'll start outcompeting their peers, which triggers a race to the bottom.
That's why I wondered about the case of private issuers of actual precious metal coinage. Reputation plays a much smaller role there, although they still need people to believe their purity claims.
reply
Thanks for posting. I'm happy I got to read your post at this hour, since I'm suffering a bit of insomnia in a different time zone than I'm used to. This may be a basic question, but can any currency ever really have an enforceable commitment? As you point out, core developers can make unforeseen changes in the future?
Secondly, why can't commitment and trustworthiness simply be tossed into the mix when currencies compete? Lack of commitment is a disadvantage, but why does it completely preclude free market competition?
reply
can any currency ever really have an enforceable commitment?
Not sure. Bitcoin used clever design and cryptography to solve many things we didn't think possible before, so full commitment could be possible.
why can't commitment and trustworthiness simply be tossed into the mix when currencies compete?
I think that's one of the points of the paper. Can issuers simply compete on reputation?
One thing that economists like to talk about is "commitment technology". i.e. Is there some way for issuers to verifiably prove that they can commit to their promises. The paper's result is that without such a commitment technology, competition is not enough to keep the market efficient.
reply
How does a competing currency credibly demonstrate trustworthiness?
Reminds me of the used car or lemon problem
reply
Yes, that's the crux. Anyone can say they are trustworthy but how do you prove it?
reply
Hasn't Bitcoin already solved this problem then? I mean... 1) run a node 2) don't update it (generally speaking) 3) profit??? And by that I mean not change the monetary policy or inflation rate of the underlying asset (Bitcoin)?
reply
This is where we run into a Ship of Theseus type conundrum, which is the same as the hard fork problem.
If Bitcoin Core updates the monetary policy, but half the node-runners don't update, which network continues to be Bitcoin?
So you're right that the decentralization of Bitcoin (balance of powers between nodes and devs) kinda solves it, but it also kinda doesn't because of the possibility of hard forks.
reply
Anyone who wants to hard-fork and 'inflate' their Bitcoin by changing the monetary policy (for example by printing themselves a million Bitcoin) can go ahead right away and do so.
I won't be changing my node however. And there, in my opinion, is the solution. Plebs running nodes
reply
What if you're the only person who doesn't update your node, though? And everyone else does? I think that's the conundrum for Bitcoin with regard to fixed monetary policy.
reply
Updating a node... to increase the issuance rate devalues one's own holding, hard work, and savings. Nodes are the world's HOA of Bitcoin.
And HOAs if you have any experience with them are down-right militant. That's why it's important that people run nodes and Bitcoin generally speaking doesn't change (and it won't). Bitcoin needs time and education.