Bitcoin is not deflationary

I want to set the record straight because I hear this a lot. Bitcoin is not inherently deflationary, nor is it inherently inflationary. Most of the time this argument is attributable to a fundamental misunderstanding about what “inflation” means in economics.

I’ll take the definition from The Economist’s online glossary:

Rising PRICES, across the board. Inflation means less bang for your buck, as it erodes the purchasing power of a unit of currency. Inflation usually refers to CONSUMER PRICES, but it can also be applied to other prices (wholesale goods, WAGES, ASSETS, and so on). It is usually expressed as an annual percentage rate of change on an INDEX NUMBER.

I like this definition, in part, because the formatting of the glossary conveniently uppercased the word “PRICES” in the exact location where some would have mistakenly written “money supply.” An increase in money supply is not the same as inflation, though it is related in ways I discuss below. If you already know this, you can save your attention and skip to the conclusion.

If you are used to referring to increases in money supply as inflation, do not feel too bad because it, unfortunately, is an overloaded term. The Wikipedia article on inflation notes,

The term “inflation” originally referred to increases in the amount of money in circulation. However, most economists today use the term “inflation” to refer to a rise in the price level. An increase in the money supply may be called monetary inflation, to distinguish it from rising prices, which may also for clarity be called “price inflation”.

I encourage you to keep the semantic difference in mind, because as I am about to explain, the difference is subtle and meaningful.

Money supply and price levels

I am not an economist, nor am I qualified to give an analysis of what causes inflation, but I’ve compiled what I hope is some fairly uncontroversial information. Inflation is a persistent change in the purchasing power of a currency, and it is often measured using consumer price indices (CPI). A CPI tracks the price of a basket of consumer goods and services over time. In the United States, the Bureau of Labor Statistics (BLS) aggregates a set of them called the U.S. CPI. Inflation is often calculated as the year-over-year change in the CPI. This is a graph of the CPI and the inflation rate of the US dollar over the past 60 years:

Courtesy of https://data.bls.gov

As you can see, and likely already knew, the US dollar has devalued quite a lot over this period. Using the inflation calculator provided by the BLS, you calculate that a $4 carton of milk today might have cost you only $2.62 just 20 years ago. This has happened by design. Most economists agree that low, positive rates of inflation are beneficial to the economy, and so the Federal Reserve uses its powers to target such an inflation rate.

How does it do this? Well, the prices of goods and services are a function of supply and demand. Consequently, the overall price level of an economy is a function of aggregate supply and aggregate demand of goods and services purchasable with a particular currency. By increasing the total supply of money in the economy, the Federal Reserve can raise aggregate demand. This is generally explained as having more dollars chasing the same amount of goods and services. As a result of heightened demand, the CPI increases and the dollar devalues. If the Fed were to remove liquidity (read: dollars) from the economy, the opposite would happen.

While money supply is certainly one factor in the overall price level, it is not the only one. The reason it gets so much attention is because it is the easiest for the government to control, but the economy’s overall willingness to spend is equally important. Within the quantity theory of money, this is captured by the term velocity. As can be seen in the chart above, during the Great Recession in 2008 the inflation rate actually dropped below 0%. This was not the desired outcome of any particular Fed policy; it was just the market’s natural behavior.

The supply of Bitcoin

One of Bitcoin’s core guarantees is that the money issuance schedule is perfectly predictable. From the beginning of the network 50 BTC is issued every 10 minutes, with the rate halving every four years or so. The chart below shows the total amount issued over time along with the rate of change. Note that the chart labels refer to monetary inflation, not price inflation.

Bitcoin supply schedule (courtesy of http://bashco.github.io/)

This rule is explicitly coded into the protocol and will never change. What is not under the control of the protocol is people’s demand for Bitcoin. Knowing what we now do about price indices, the price level of goods in a BTC-denominated market can be estimated as the CPI divided by the BTC/USD exchange rate. Here is a graph of the adjusted CPI and inflation rate in Bitcoin since 2013:

US consumer price index and inflation rate denominated in BTC since 2013

Overall, yes, Bitcoin has on average experienced deflation. But that was not the case in 2014, when the price dropped 57% in dollar terms. Another thing to note about this calculation is that the inflation rate is dominated by changes in the Bitcoin price, not changes in the US CPI. This basically means that if you claim that Bitcoin’s inflation rate is perfectly predictable, then you can predict the price for years to come (and if so, should definitely give me a heads up before the next rally).

Why does this matter?

This discussion usually arises when discussing the monetary policy of crypto assets like Bitcoin and comparing to token projects with more sophisticated ones like MakerDAO or Basecoin. I am not arguing here that Bitcoin’s policy is good or bad, just that is important to use the correct terminology when discussing the economics of cryptocurrencies. There is a large body of economic research into the causes and effects of inflation and when designing or evaluating the issuance conditions of new tokens, everyone should be on the same page. Confusing money supply schedule with inflation rate is like using the word “force” to refer to “acceleration” in physics: they are related but the distinction is very important.