Why Bitcoin isn’t always deflationary

Late last Friday FT Alphaville (ie, this reporter) took part in an excessively long Clubhouse session with bitcoin blogger and maths lover Allen Farrington, which touched on many different aspects of the world’s most famous orange-tinted novel financial system.

While many aspects of the conversation will be old ground to readers of this blog, there was one discussion point that hit new territory. It pertained to whether bitcoin is truly a deflationary currency or not. Our take was that bitcoin may not be as immune to inflation as many believe. (But also that that’s not necessarily a bad thing.)

You can catch up on the broad brush of the discussion via this incredibly detailed write up by Farrington, which comes with the added bonus of a lot of maths being applied to the ideas.

But for those who don’t feel like digesting an 11 minute read with equations, the gist of the point is this: it’s often forgotten in the bitcoin world that it’s not just quantity of money that influences inflation, but velocity too. And in the bitcoin economy there’s a lot of pent up velocity to contend with, not least because of all those wallets bearing huge sums of value that never transact (like Satoshi’s own stash).

In the conventional “fractional reserve” fiat world, velocity is a function of monetary multiples, which are basically a function of how quickly the commercial banking sector can reuse or re-lend the underlying reserves in the system. This in turn influences the money pyramid and the overall money supply. As the grand experiment with QE has shown, the broader money supply is not just a function of core central bank liabilities, but the wider system’s capacity to reuse those reserves and turn them into private liabilities.

In a fiat-based fractional reserve system you cannot escape the money multiplier effect. If you keep your wealth in the banking system, somebody somewhere is putting it to work by re-lending it.

This, however, is not true of bitcoin. In a hyper-bitcoinised world you have the ability to opt out of the perpetual re-lending loop. While you can if you wish, engage in having your savings re-lent to earn a bit of extra return (say to those who want to use the loans as a shorting mechanism), at the end of the day it’s entirely voluntary. The underlying system, at least in theory, is fully reserved.

Those who believe a full reserve system is a better and more stable system will argue that’s a good thing. And maybe it is, or isn’t. That’s beyond the scope of this post.

Our point is simply this: just because a system is fully reserved and all lending is done with the explicit conscious consent of the wealth holder doesn’t mean it is resistant to inflation.

Bitcoin looks disinflationary right now. But this is because those who choose to opt out of putting their money to work effectively reduce the velocity of their money to zero. The delightful thing about bitcoin is that thanks to the openness of the bitcoin ledger we can see quite clearly how much cash never transacts in this way. For example, according to a report by Crystal Blockchain, as of July 2020, there were over 10m bitcoins (worth $85bn at the time) accumulated in dormant bitcoin addresses with no outgoing transactions at all in 2019. More recent data from bitinfocharts suggests some 17.8m bitcoins out of roughly 18.75m outstanding are currently classified as residing in dormant addresses.

So what does this mean for inflation?

What we posited was that a structure in which lending cannot be done without explicit consent transfers “fractional reserve” risk away from banks over to the managers of real economy inventories.

This is because ensuring there are always enough goods and services in the economy to meet bitcoin’s potential redemption capacity, were all coins to suddenly be mobilised, is both impractical and excessively wasteful. The very point of just-in-time inventory management is that residual or excessive stock should not be kept on standby. Shopkeepers determine how much inventory they should hold based on historic demand trends, not how much money there is in circulation or potentially could be.

It stands to reason that in a bitcoin economy inventory managers would, on a macro level, end up under-reserving physical goods relative to the potential monetary capacity to redeem such goods. Given how much bitcoin sits immobile in wallets that almost never transact, not doing so would be competitively inefficient and commercial madness.

But doing so would also undoubtedly heighten the system’s fragility to external supply shocks, choke points or other crises that have the capacity to motivate bitcoin savers to suddenly drawdown on previously immobile bitcoin reserves and cause an effective run on inventories. Previously non-circulating money could very suddenly start circulating quickly.

The only corrective mechanism in such a scenario would be for the price of goods and services in bitcoin terms to go up sharply to counter the unexpected demand, manifesting in a sudden and destabilising inflation.

Farrington counters quite reasonably that in a bitcoin world there would at least be a cap to the burst of inflation. Effects would be transitory until new inventory came online or until new excess wealth had been created, and thus immobilised. In theory, the price signals would be super strong and stimulate supply. The inflation might also enhance social mobility, as old wealth would be drawn down and transferred to those with the capacity to create the new wealth that the system needs.

The question is, is this really all that different to the current system? And how capped is this sort of inflation really?

If bitcoin transactions were, in theory, to achieve maximum velocity across all wallets, only supply shocks could induce broad-brush inflations. Additional demand, say because of population growth, could never be catered to with money creation, meaning the consequences would always be deflationary. Which is to say any emerging disequilibrium could only be cleared by making some portion of the economy poorer and the other part richer by incentivising other parts of the economy to save more.

In such a world, it would be very unlikely that the bitcoin economy could stave off other parallel money systems or credit creation from coming to the rescue of that money shortage. (Especially if the system was to remain a voluntary non-coercive, non-fiat system.) And then monetary-induced inflation threats would return to the table.

Chances are whether it’s a bitcoin world or a fiat world, money will remain endogenous, expanding or contracting in conjunction with the capacity of the underlying economy to provide its citizens with the stuff it needs. That’s not to say bitcoin doesn’t have uniquely useful properties. This reporter is definitely coming round to the idea that resistance to censorship is an important quality for a monetary system to offer. As is, we think, the freedom to opt into alternative harder, stateless currency systems if your own one hits the inflationary fan.

But whether bitcoin’s best attribute is its inflation resistance in a broader hyper-bitcoinised context, we’re not so sure about. Inflation is not necessarily removed from such a world, just reframed.