What Causes Inflation?

Inflation remains one of the great enigmas of modern economics. In this week’s post, we will examine a simple theory of the price level, “Ratio Theory of the Price Level”, and a basic model that can be used for thinking about short-term movements in the price level “The Goods-Money Framework”. We will then use these ideas to examine some of the traditional explanations for inflation.

Despite extensive academic studies and seemingly endless debate, a quick keyword search on “what causes inflation?” will reveal a jungle of different ideas and opinions regarding the true drivers of inflation.

The traditional view, taught at high schools and colleges, is that inflation can be driven by “demand pull” or “cost push” factors. In essence, this is a macroeconomic extension of the basic microeconomic tenet that the price of a good can rise either because there is more demand for that good (“demand pull”) or because there is reduced supply for that good (“cost push”).

The “demand pull/cost push” model represents an “old Keynesian” view of how the world works: if aggregate demand increases, then real output should increase and the price level will rise, particularly if the economy is operating near full capacity.

Mainstream economists recognize that this simple aggregate supply and demand view of the world often fails to predict episodes of high inflation. Therefore, this basic Keynesian model has been fudged by the addition of something called “inflation expectations”. This “New Keynesian” model states that inflation is caused by either (1) too much demand, or (2) expectations of future inflation. The problem with the “inflation expectations” term in the New Keynesian models is that no one seems to have a good sense of what determines “inflation expectations”.

The issue is made more complicated by the fact that most economists recognize that, over the long term, the monetary base plays an important role in the determination of the price level. Even senior central bankers tie themselves in knots trying to explain how to reconcile the New Keynesian model of the world with the simple, empirical fact that money matters. Mervyn King, former Governor of the Bank of England, discusses this problem in his article “No money, no inflation – the role of money in the economy” (2002) in which he concludes that “the absence of money in the standard models which economists use will cause problems in the future”. Frankly, I couldn’t agree more.

If even central bankers can’t reconcile the competing views of what drives inflation, then this suggests that something is wrong with the underlying models. The view of The Money Enigma is that both Keynesian and Monetarist models fail to provide satisfactory models for the determination of the price level because they both start from the wrong fundamentals.

In order to build useful models of the price level, we need to go back and challenge the basics of current microeconomic theory. In particular, we need to develop a more comprehensive answer to the question “how is a price determined?”

It is the view of The Enigma Series that the current presentation of microeconomic price determination, namely the traditional supply and demand chart with price on the y-axis, presents a one-sided and very limited view of the price determination process.

The view of The Enigma Series is that every price is a function of two sets of supply and demand. More specifically, the price of one good (the primary good) in terms of another good (the measurement good) is a function of both supply and demand for the primary good and supply and demand for the measurement good.

A few weeks ago, we discussed how prices are determined in a genuine barter economy (an economy in which there is no commonly accepted medium of exchange). We asked the question “what determines the price of apples in banana terms? Is it supply and demand for apples? Or is it supply and demand for bananas?” The answer is both.

Every price is a relative expression of two market values. The market value of apples is determined by supply and demand for apples. The market value of bananas is determined by supply and demand for bananas. The price of apples, in banana terms, is a relative expression (a ratio) of these two market values. Therefore, the price of apples, in banana terms, is a function of two sets of supply and demand (see diagram below).

Price Determination

This principle can be extended to the determination of “money prices”. The price of a good, in money terms, is a relative expression of the market value of the good and the market value of money. For example, if the market value of one apple is three times the market value of one US Dollar, then the price of an apple, in US Dollar terms, is three dollars.

In general terms, the dollar price of an apple can rise for one of two reasons: either the market value of an apple rises (for example, there is a supply shortage), or the market value of the dollar falls.

The diagram below illustrates how the price of apples, in money terms, is determined by two sets of supply and demand: supply and demand for apples, and supply and demand for money. The key in this diagram is the y-axis unit of measurement: a “standard unit” of market value. Instead of using price, a relative measure of market value, on the y-axis, the diagrams above and below use an absolute measure of market value of the y-axis. The standard unit of market value is  a theoretical and invariable measure of the property of market value, just as “inches” are an invariable measure of the property of length.

Price Determined by Two Sets Supply and Demand

The key point that readers should take away from the above diagram is that every “money price” in our economy is a ratio. More specifically, the price of a good in money terms is a ratio of the market value of the good (the numerator) divided by the market value of money (the denominator).

The Inflation Enigma, the second paper in The Enigma Series, extends this simple microeconomic concept (every price is a relative expression of two market values) to a macroeconomic level. If the market value of money is the denominator of every money price in the economy, then the price level can be stated as a ratio of two market values: the “general value level”, a hypothetical measure of the absolute market value of the basket of goods/services, and the market value of money. This is called the Ratio Theory of the Price Level.

Ratio Theory of the Price Level

Ratio Theory of the Price Level states that the price level is a function of the value of goods relative to the value of money. If the value of goods rises relative to the value of money, then the price level rises (inflation). If the value of goods falls relative to the value of money, the price level falls (deflation).

We can best illustrate Ratio Theory with a simple macroeconomic framework called “The Goods-Money Framework” (see diagram below). The Goods-Money Framework is broken into a left side and right side. On the left side, aggregate supply and demand for goods/services determines real output (x-axis) and the market value of goods (y-axis), as measured in absolute terms. On the right side, supply and demand for money (the monetary base) determines the market value of money (again, market value is measured in absolute terms).

Goods Money Framework

The price level is a ratio of two macroeconomic equilibrium: the market value of goods, as determined on the left side of the model, divided by the market value of money, as determined on the right side of the model. Now, let’s get back to our original question.

What causes inflation?

The left side of the Goods-Money Framework provides some distinctly Keynesian answers to this question. All else equal, the price level will rise if the market value of goods (the “general value level”) rises. In a stylized sense, this can occur either because the aggregate demand curve shifts to the right (“demand pull”) or because the aggregate supply curve shifts to the left (“cost push”).

The right side of the Goods-Money Framework provides a somewhat more Monetarist perspective on the issue. All else equal, the price level will rise is the market value of money falls. In very simple terms, this can occur either because the supply of money (the monetary base) increases or because the demand for money falls.

In practice, both the left side and right side of the model are both moving at the same time. For example, deflationary forces that are acting on the left side of the model, (for example, “globalization” of the labor force), might be offset by inflationary forces on the right side (for example, aggressive monetary easing), leading to a net result where the price level changes little. [Note: if both the numerator and denominator fall by roughly the same percentage, then there is no change to the ratio itself].

While interpreting the left side of the framework is relatively straightforward, the right side of the framework is extremely complex. The main problem is that “supply and demand” is, in practice, a poor short-term model for the determination of the market value of money.

We know from recent experience that a large increase in the monetary base can have little short-term impact on the market value of money (and hence, little short-term impact on the price level). The reason for this is that money is a proportional claim on the future output of society. More importantly, money is a long-duration asset. The market value of money depends far more upon expectations of future levels of the monetary base than it does on the current levels of the monetary base.

The Velocity Enigma, the final paper in The Enigma Series, develops a valuation model for money that demonstrates that the current market value of money depends upon long-term expectations. More specifically, the current market value of money is highly dependent upon the expected long-term path of the “real output/base money” ratio.

As you can see, there is no simple answer to the question “what causes inflation?”

The traditional Keynesian view provides a very limited perspective on the issue. Importantly, the notion that inflation can only occur if the economy is overheating (the economy can only experience inflation if there is “too much demand”) is nonsense.

The price level depends upon a complex set of expectations. Most notably, the expected long-term path of the “real output/base money” ratio is the key determinant of the market value of money. In turn, the market value of money is the denominator of every “money price” in the economy.

This model provides a sensible explanation for how inflation can occur in a weak economic environment. If aggregate demand is weak, then this will place downward pressure on the market value of goods on the left side of our model. However, if confidence in the economic future of the country falters, then this can easily lead to a decline in the market value of money that overwhelms the fall in the market value of goods. In the context of Ratio Theory, as the denominator (the market value of money) falls more rapidly than the numerator (the market value of goods), the price level rises.

In summary, we can say that, in the short-term, the drivers of the price level are complex. Aggregate demand and supply matter, but expectations of the future path of the “real output/base money” ratio are critical.

While economists may disagree on the short-term drivers of the price level, there is at least a broader consensus on what drives the price level over the long-term: money. More specifically, the ratio of base money to real output is the key driver of the price level as measured from point to point over very long periods of time.

The Enigma Series provides a common sense explanation for this phenomenon.

Money is a special-form equity instrument of society that represents a proportional claim on the future output of society. The value of a proportional claim on the output of society will rise as real output rises and fall as the monetary base increases (i.e. as the number of claims against that output increases).

Therefore, if over a period of many years, the monetary base has grown at a much faster rate than real output (as it has in the United States over the past 80 years), we should expect the market value of money to have fallen significantly and, all else equal, the price level should have risen significantly.

Over long periods of time, it is this ratio of money/output that drives the price level. The question today is how long can the monetary base in the United States can remain at these extended levels without triggering a significant decline in the market value of money and reigniting inflation. Too many commentators who are worried about deflation are, at least implicitly, focused only on the left side of our model above. The key to inflation remains the right side of the model, the market value of money.