Yves here. Some readers questioned the discussion of Bitcoin’s deflationary qualities in a recent post, so this discussion is a useful follow-on.
By Dan Kervick, who does research in decision theory and analytic metaphysics. Cross posted from New Economic Perspectives
I appeared today on The Attitude, broadcast by WNHN 94.7 in Concord, New Hampshire, to talk with host Arnie Arnesen about the Bitcoin phenomenon. The podcast of the second hour of the show can be accessed at the link below. My appearance occurs right at the beginning of the hour:
The purpose of our brief discussion was just to provide some general background information for Arnie’s listeners about Bitcoin, including what bitcoins are and why anyone would buy them or accept them in exchange for goods and services. We touched on several topics related to the Bitcoin phenomenon, but there is one very peculiar and puzzling feature of Bitcoin that we didn’t get to discuss and that seems especially important to me: the Bitcoin system has what appears to be a built-in deflationary architecture.
When the Federal Reserve System was created, it was charged with providing the US with an “elastic currency”. That means that the quantity of Fed-issued dollars in circulation is supposed to vary in response to the changing dynamics and needs of the real economy. The Fed is expected to monitor economic activity, and conduct a monetary policy that provides us with a stable but flexible medium of exchange.
Bitcoin, by contrast, is much more rigidly designed so that new bitcoins are introduced into the system at a mathematically predictable rate that is almost completely independent of any economic activity for which bitcoins might be used. New bitcoin production is supposed to take place at an exponentially decreasing rate so that production decreases by about 50% every four years. As a result, the number of bitcoins in existence will effectively flatten out at 21 million in about 2040 – if anybody is still using the Bitcoin system by then. But long before 2040 the rate of bitcoin growth will slow very dramatically.
The Bitcoin system therefore possesses a hard-coded and extremely rigid monetary policy determined by the software itself, software which lives on the computers of everyone who is participating in that system. Now, you could say this means Bitcoin’s monetary policy is decentralized. That’s certainly how Bitcoin enthusiasts tend to describe it. But another way of looking at it is that it is that Bitcoin’s monetary policy is highly centralized in the persons of the people who wrote the Bitcoin code, and who established a weirdly inflexible Bitcoin monetary policy regime in advance – determined for all time.
Now what does this mean for the future value of Bitcoin as a medium of exchange? That all depends on whether the Bitcoin economy – the universe of producers of goods and services who accept bitcoins in payment – continues to grow, or instead settles into a small and unchanging niche economy for a limited number of enthusiasts. But suppose as a thought experiment that the Bitcoin economy continues to grow, and that the volume of goods bought and sold with bitcoins continues to increase, as the rate of bitcoin creation first slows and then flattens. Then one of two extremes might occur: either (i) prices in bitcoins remain stable as the rate of bitcoin transactions increase, or (ii) the rate of transactions stays roughly the same, but bitcoin prices fall as the finite quantity of bitcoins is spread over more and more transactions. Since the pace of transactions depends on real-world constraints on production and consumption, the effect that is likely to be the dominant one is that prices will fall. In other words, there will be a deflationary spiral in the Bitcoin economy. This makes Bitcoin a poor long-term candidate for a stable, alternative medium of exchange.
Deflation might appear to be an attractive thing at first look. Wouldn’t it be nice for our money to appreciate in value as the prices for goods and services continually fall? But economists associate deflation with two negative phenomena: First, if prices are falling then the incentive to hoard the currency increases, since anybody who possesses that currency is seeing its value increase each day. Thus, the currency itself becomes an appreciating investment vehicle for its owner, so long as it isn’t spent. Hoarding by an individual agent is no big deal, but it is clearly bad news for the economy when hoarding is widespread, since if people stop buying things, then producers stop producing things and stop paying workers to produce things. That’s one reason why downturns are often associated with deflation, and growth is usually associated with modest inflation.
The other problem with deflation is that contracts and debts are usually fixed in nominal terms, and so deflation makes debt more onerous. Imagine an office worker, Sal, with a $50,000 annual salary and a $200,000 debt, such as a mortgage debt. Now suppose there is a general deflation, and both consumer prices and wages drop 20% over some period of time. Sal’s wages fall to $40,000. Sal’s ability to buy groceries is unaffected since grocery prices have also fallen by 20%, but the $200,000 debt is now worth five times Sal’s annual salary rather than four times the salary, and has become much more burdensome. If the deflation continues, Sal will be wiped out. But before that happens, Sal’s creditor makes out handsomely as the real value of Sal’s monthly payments increases.
Bitcoins are infinitely divisible, so while there is an ultimate cap on the quantity of bitcoins, there is no lower limit on Bitcoin denominations: there is no Bitcoin “penny” that can’t be subdivided further. So Bitcoin’s designers seem to have built these deflationary prospects into the system as a feature, not a bug. And here we must look at another curious feature of the Bitcoin system, the feature its developers decided to call “mining”. As we have noted, Bitcoin has a built-in mechanism for adding new bitcoins to the system at a decreasing geometric rate. But note that new bitcoins are not simply sprinkled evenly among all bitcoin users when they are added to the system. They are awarded to “miners” – in practice, people who have substantial computing power and computing speed at their disposal – in exchange for those miners using some of their computer power to win online races to authenticate new blocks of bitcoin transactions.
So you can see why you would very much like to be a miner in a thriving Bitcoin economy and why early adopters of Bitcoin are so fanatical about keeping the system going. Those who manage to accumulate bitcoins in the earlier stages when the pace of bitcoin creation is high, could profit handsomely when the deflationary phase kicks in. These miners would, if the world-conquering dreams of the Bitcoiners ever came to pass, be something like the descendants of medieval vassals who acquired some poor land from their lords in an early era when there was still much land to be claimed and settled, and who then became fabulously wealthy over time by hanging onto their holdings as the finite stock of land was all brought into private owner ship and production while the population continued to increase.
So it looks to me like the developers of Bitcoin were thinking like this: “Mining system + deflationary architecture = we’re rich!!!”
However, none of these get-rich dreams are likely to materialize. Since the system’s architecture is set in the unchanging software, and we are looking at only a few years rather than several centuries as the time frame for the rapid exponential decrease in new bitcoin additions, sophisticated investors can all see the pre-determined and rapidly approaching deflationary future of Bitcoin, and will price that into their decisions now. Their incentive is to buy in the immediate term, accumulate some bitcoins, watch closely as they appreciate, and then dump them on suckers at the first sign of the broad market realization of the inevitability of the deflationary stage. There might be a few volatile swings before the final collapse hits for good, but eventually the speculative demand for bitcoin will evaporate, there will be a massive selloff, the deflation will reverse into a brief hyperinflationary spasm … and then pfffft.