Bitcoin and the Myth of Hyperdeflation

Economists are up in arms about Bitcoin — but do their concerns hold water?

Previous: “Why Bitcoin Shouldn’t Die in a Fire”

One of the most exciting effects of the rise of Bitcoin has been its role in inspiring popular interest in economic issues, especially those related to the theory of money. This is a very good thing, if only because economic education in the US has been in a sad state for quite some time, much to the detriment of policy. Bitcoin’s architecture runs directly afoul of some deeply cherished sacred cows of modern Keynesian economic theory, and the result has been a lively (public) debate on the role of money for the first time in decades. I dearly hope it will continue.

One can hardly turn on the news these days without hearing Keynesians speaking direly about Bitcoin. Keynesian objections to Bitcoin revolve around two basic complaints. The first, regarding the ‘intrinsic value’ theory of money, was addressed in my last piece. This time, we’re going to be talking about the second Keynesian objection to Bitcoins, which has to do with its deflationary nature.

Modern Keynesian theory holds that deflation (an increase in the real value of a unit of currency) is very destructive, more so than inflation (a decrease in the real value of the same unit). The reasoning here is that, if a currency is deflating, there’s an incentive for consumers to hoard it (as opposed to spending or investing it). Then, because people are spending less money, the prices of goods drop as businesses try to find the new price at which people are willing to buy. This produces more deflation, and the process spins out of control until the entire economy implodes on itself. This can be understood by analogy to hyperinflationary spirals, in which the sudden increase in the quantity of currency (and its corresponding drop in value) causes people to lose faith in it and try to sell it, which further reduces the value until the currency’s buying power drops to zero.

Such is the fear of this sort of collapse that it remains one of the fundamental theoretical justifications for the central banking system (such as the Federal Reserve of the United States). The logic goes as follows: if deflation is dangerous, and the economy insists on continuing to grow, then a fixed currency supply is a recipe for disaster. Instead, a level of inflation above and beyond economic growth must be maintained. By maintaining a central bank, currency can be printed to outpace economic growth and ensure that inflation continues uninterrupted. The other theoretical justification for the central banks is, of course, that they can print enormous amounts of money in the event of an economic collapse to stimulate aggregate demand (total demand for goods and services). In Keynesian thinking, stimulating aggregate demand by increasing credit or the money supply is the way to kick-start an ailing economy.

All of this means that Keynesians’ predictions for Bitcoin economics are more than a little apocalyptic. There are, after all, a finite number of bitcoins that will ever exist (after the mining eventually tapers off decades from now, as per the protocol). The US economy, in the meantime, continues to experience a rate of real GDP growth of about 4.5% per year. According to standard Keynesian analysis, therefore, the Bitcoin economy will inevitably experience out of control deflation before collapsing. Paul Krugman (perhaps the foremost Keynesian in the United States) notes in a New York Times Op Ed that Bitcoin’s price is currently in a meteoric rise (as Keynes’ model predicts), and declares the entire Bitcoin experiment a bust on that grounds.

Sadly, Krugman is making a serious mistake here, if not being outright dishonest. In order for Bitcoin’s rise in price to be the result of a fixed currency supply, its value would have to be (in fact) fixed. In reality, the number of bitcoins in circulation is increasing as they are mined (and will continue to rise at a slowing rate for the next century). The instability in the currency supply is not the result of the fixed nature of the currency, but the huge public interest. Bitcoin, due to its limited market penetration and the glut of amateur speculators, is in a uniquely vulnerable position for the time being. As the Bitcoin market is colonized by more experienced traders, the currency should stabilize (allowing more market penetration and resolving these growing pains). The real properties of Bitcoin’s protocol won’t become clear for years yet.

Even when Bitcoin’s growth rate falls below that of the GDP, though, there are compelling reasons to believe that hyperdeflation is not inevitable. The Keynesian model discusses a feedback loop, and likely a real one — where it goes wrong is in assuming that that feedback loop will inevitably go to infinity. Some iterating functions converge, and some diverge, and the Keynesian model does not provide substantial evidence of the latter. There are, after all, only so far that prices can fall (at a given level of deflation) before it ceases to be profitable to participate in a market. That side of the feedback loop has hard brakes built in. On the other side, the degree of hoarding you expect to see is contingent on the rate of deflation. For a mere few percent annual value increase, we already know that most consumers won’t defer purchases very much (judging by the deplorable state of American savings). If people were inclined to hoard a currency for a modest 4.5% annual interest, they’d already be hording it in their savings accounts. Doubtless, there is a rate of deflation that will cause economic collapse, but there’s no reason to think that it’s as low as the deflation we expect to see long term in the Bitcoin economy.

More so, though, the idea that people won’t invest in a deflationary economy is absurd. Imagine that you own one Bitcoin, and you know at the end of the year that it’ll have 4.5% more buying power. Imagine that you’re presented with the opportunity to invest your bitcoin at 5% annual interest rate. If you choose to hoard the Bitcoin, you’ll get a 4.5% increase in buying power at the end of the year. If you invest it, you’ll get a 9.7% increase in buying power. The supposed incentive not to invest is nonexistant. The truth is, of course, that the bulk all financial transactions are not made up of luxury purchases that can be deferred indefinitely, but investments and necessities like rent and food and insurance, none of which should be much impacted by deflation.

Beyond poking holes in the model, though, the evidence is also against the Keynesians on this one. Notably, deflationary spirals, the great evil used to justify the central banks, does not appear to have ever actually happened. Inflationary spirals, in contrast, happen all the time (a notable example being in Germany after World War I, which produced the famous photos of housewives using money as tinder). The same sort of dramatic hyperdeflation is nowhere in evidence. You might expect this sort of economic collapse to have occurred during the industrial revolution (a time of high economic growth on a fixed, gold economy), but it didn’t. You might also expect it to have happened during various economic booms prior to the switch to fiat currency- but it didn’t. In fact, in the entire history of the gold and silver economy, there are periods of high deflation, but no associated economic collapse.

Ben Bernanke (the chairman of the Federal Reserve) has been known to claim that the Great Depression was the result of a deflationary spiral — a grim example, if he were right. However, even this doesn’t hold up to scrutiny. The only school of economics to predict the great crash before it occurred was the Austrian school, a school which notably does not believe in deflationary spirals. Their justification for their prediction was the inflation of credit which occurred in the decade prior. If the value of an economic model is measured by its predictive ability, then the Keynesian model of deflation begins to look tenuous in the extreme.

That isn’t to say that there are no consequences to deflation. Deflation makes the real value of debts rise, even as the cost of goods fall. This is problematic when your annual salary is less than your total debt, because the value of your debt increases faster than your salary: for a debt sufficiently larger than your annual income, you are essentially experience an automatic 4.5% annual increase of your debt, even assuming a nominal zero percent interest rate — which is problematic, considering the fraction of Americans who are underwater on their mortgage or student loans. That said, in a deflationary economy, the incentives to go into debt are greatly reduced, so, if Bitcoins become a dominant medium of exchange, it’s unlikely that this state of affairs would continue to hold. It’s important to weigh these potential risks against the benefits that Bitcoin offers (portability, reliability, and decentralization).

It’s inevitable that Bitcoin will experience significant deflation in the next few decades due to the nature of its protocol, and it’s likely that this deflation will eventually cause some problems for the Bitcoin economy — but the apocalyptic predictions of mainstream Keynesians are totally unjustifiable. The theoretical basis is thin, and the experimental basis is nonexistant. The sky is not falling, Chicken Little is just a Keynesian. There is absolutely no reason to believe that a low rate of deflation, in a healthy economy, is any worse than the low rate of inflation endorsed by the Fed. Bitcoin economies will likely be at least as stable as historic gold economies, and, if Keynesians would like to claim otherwise, the burden of proof rests with them.

To read more articles like this, follow Andre on Twitter: @AndreTI