More bits on Bitcoin
Jean Tirole opines in the FT about the social costs of crypto-currencies, prompted no doubt by the continued surge in the relative price of Bitcoin, depicted below [y axis in £].
Tirole writes:
"Bitcoin’s social value is rather elusive to me. Consider seigniorage: an expansion in the money supply traditionally provides the government with extra resources. As it should, the proceeds of issuance should go to the community. In the case of bitcoin, the first minted coins went into private hands. Newly minted coins create the equivalent of a wasteful arm’s race. “Mining pools” compete to obtain bitcoins by investing in computing power and spending on electricity. There goes the seigniorage."
Bitcoin - and to emphasise, we are not just talking about Bitcoin, as there are now 100s of competitors - does have a questionable social value. But that it displaces seigniorage is not high on the list of its downsides. Governments down the ages have very often abused the power to raise finance by issuing money, and devising monetary and fiscal institutions that can prevent this has been a priority, and a qualified success.
Tirole continues:
"Bitcoin may be a libertarian dream, but it is a real headache for anyone who views public policy as a necessary complement to market economies. It is still too often used for tax evasion or money laundering. And how would central banks run countercyclical policies in a world of private cryptocurrencies?"
These are fair points. However: cash, one of the assets Bitcoin and similar seems to compete with, is also used for illicit and illegal activity. [Sufficiently so in India that it prompted the government to withdraw and redenominate 85% of the note issue]. That any technology can be used in a way that is detrimental to the public good is not enough for us to think of eliminating or curtailing it. The question is whether the costs outweigh the benefits.
The final remark the paragraph above, about the impossibility of running countercyclical policy with current crypto-currency protocols, also merits comment.
Would central banks lose control over monetary policy if something like Bitcoin took over? Roger Farmer tweeted the same thought:
In principle, central banks can retain control over the economy so long as they retain the ability to define the unit of account.
Imagine an economy that was just textile manufacturing. Central banks could adjust the definition of a metre to control the business cycle. Lengthening the definition of a metre would lower the real price of textiles [how much you get for a 'metre'] and, provided prices were sticky, and posted as currency units per metre, boost demand.
The notion of having a different unit of account from the medium of exhange surfaced in the context of solutions to the problem of the bound imposed on central bank rates by the property that cash returns zero interest. Buiter and Kimball are associated with the idea that the central bank might manage the unit of account so that the medium of exchange [cash] depreciates in value against it, thus yielding a negative interest rate, and permitting negative rates to emerge on market instruments.
So even if everyone shifted to Bitcoin, a central bank might still have its monetary policy lever. Personally I think Roger/Jean's concerns about a take-over are real.
Collective, private decisions to ditch the official medium of exchange have typically involved ditching the unit of account too.
In history this was because monetary policy wrecked the unit of account function and the store of value/medium of exchange function of money. We are contemplating here a world in which the unit of account function has not been wrecked by central banks. So it's conceivable that only the medium of exchange would shift. But the (recent) historical precedent that the two tend to go together, even the conceptual difficulty of disentangling them, makes a Bitcoin takeover that disempowers central banks at least as probable. (I say recent because if we go back to medieval times, say in continental Europe it was pretty common to observe units of account different from the multiple media of exchange circulating).
The aspect of crypto-currencies that concerns people, the protocol of essentially fixed supply, may be precisely what will limit their spread and preserve central bank leverage over their economy.
As David Andolfatto [or David Blockchain as he prefers now to be known] and others have pointed out, the fixed supply means that fluctuations in Bitcoin money demand are not accommodated and are felt in the price, making the price inherently more volatile [than that of central bank money].
The desirability of a currency protocol often - though not always - dictates the extent to which they are used in the future. Repeating an example already cited here, those countries that dollarised were the ones with the worst local protocols [for managing their own currency].
Moreover, just as the protocols that govern central bank-note issue have changed, mostly for the better [the recent Indian demonetization, and Venezuela two recent exceptions] so the protocols governing crypto-currencies might evolve for the better too. One reading of monetary history - not too Panglossian - is a slow process of discovering what works for the common good. Anecdotally, I know from interacting with some of them that cryptocurrency developers understand the fixed supply 'problem', and it is not beyond the bounds of possibility that a better algorithmic protocol, or even one with human committees, emerges.
Which leads us to remember that central banks - nothing but a committee driven money protocol defined by the inflation targets, interest rate setting procedures, etc - could step in and provide their own. Indeed, the Fed, the BoE, the Norges Bank and perhaps others have openly contemplated this idea. [Detail: they already do provide digital currency to financial intermediary counterparties. The question is whether they provide it to all of us.]