The Distributional Effects of Expansionary Monetary Policy

“Inflation is always and everywhere a monetary phenomenon,” or so wrote Nobel laureate economist Milton Friedman. In simpler terms, if you create more money, prices go up, and that is the sole cause of inflation. Granted, there may be other causes of inflation, but the recent inflation we’ve had was preceded by a massive influx of newly created dollars from the Federal Reserve.

Do all prices increase at the same rate? Not at all. I am paying far less for gasoline than I was two and a half years ago, and gas prices are always going up and down. Food definitely costs quite a bit more than it did before COVID, but rotisserie chickens seem almost immune to price increases. Why do stores keep their rotisserie chicken prices low? Those chickens are a loss leader – you need to lure people into your store with something cheap so they will buy something pricier. There are reasons why prices do not all change at the same rate. Richard Cantillon, an economist from about three hundred years ago, noticed this, and he said that the prices that change first are for the goods purchased by the people who get the new money first. Cantillon effects are the eponymous term for prices that change at different rates.

You’ve probably heard about consumer prices and the consumer price index. That is basically a basket of goods, each with a price. When the price of the basket goes up, the consumer price index goes up, and that is inflation. Not everyone buys the same basket. Conveniently, the Bureau of Labor statistics produces a sub-index for prices of each category of goods in the basket, along with Consumer Expenditure Surveys that estimate typical expenditures by households in specific income ranges. By taking these sub-indices and the surveys, I created consumer price indices that are specific to these income ranges. Then, I estimated time series statistical models over a period of 1990 to 2022 to predict the effect of government spending and the Fed’s money creation on these income-specific price indices.

The results were not perfectly monotonic. Those poorest were harmed most by these policies, and the richest were harmed least. In the middle ranges, some were harmed and some weren’t. This is because of household characteristics. If two individuals earning incomes typical of the lowest quintile marry to become one household, suddenly the household income is higher, but consumption possibilities for the pair may not increase much. Likewise, middle income households often have more dependents. The observable pattern in the results is that, as consumption possibilities increase, the inflationary policies are less harmful. This requires more income per individual, not just more income per household. Intuitively, this makes sense. When newly-created dollars are available, they can be spent on necessities or luxuries, or saved. Those with more pressing needs will spend right away, and they will buy necessities. Cantillon effects kick in, and those are the goods that become more expensive the fastest. These results are consistent throughout the entire period from 1990 to 2022, although inflation was fairly low for most of that time until recently.

So, who was harmed the most by the recent inflation? Those who spent the new dollars on necessities, as opposed to those who were wealthy enough to save it or use it for luxuries. COVID stimulus checks may have been welcome respite to unemployed workers, but they contributed to inflation, combined with the Fed’s monetary policy. The inflation that ensued was worst for those who most needed the help. Herein lies an important lesson – policies can have ripple effects. They may harm those they are intended to help. If a policy will cause inflation, keep in mind that it will be worst for those least able to handle it.