• Does money have any role in the determination of inflation? Does printing too much money cause inflation? And what does it mean to say “too much money”? “Too much” relative to what?

• Milton Friedman once famously observed, “Inflation is always and everywhere a monetary phenomenon”. Less well known is his qualification to this statement. Friedman’s full observation was “Inflation is always and everywhere a monetary phenomenon in the sense that it is and can be produced only by a more rapid increase in the quantity of money than in output.” (Friedman, “Money Mischief”, page 49)

• Friedman’s qualification helps us narrow down our original question to the following, “Does too much money, relative to output, cause inflation?” Alternatively, we could ask, “Does growth in the monetary base that is significantly in excess of growth in real output cause inflation?”

• Empirical evidence strongly suggests that, when measured over long periods of time, growth in the monetary base that is significantly in excess of growth in real output does lead to a concomitant rise in the price level. In this sense, Friedman’s observation appears to be correct.

• However, over short periods of time, this relationship does not appear to hold. For example, over the past six or seven years, the monetary base of the United States has quadrupled, while real output has grown only modestly: yet the price level has barely moved higher.

• So, why is this the case? Why does the quantity theory of money work over long periods of time, but not over short periods of time?

• The view of The Money Enigma is that, over short periods of time, the primary driver of inflation is not the change in the current ratio of “base money/real output”, but the change in the expected 20-30 year future ratio of “base money/real output”. In other words, it is not the current level of growth in money relative to output that matters, but rather the expected long-term future growth of money relative to output that matters.

• In this sense, it is not “too much money” that causes inflation, but the expectation of “too much money” being created over the next 20 years that matters.

• Why might market participants suddenly expect a society to create too much money (relative to output) in the future? There are many possible reasons, but obvious reasons might include war, a secular decline in productivity, economic mismanagement, or just the sudden realization that a country has been living way beyond it means.

Inflation and the value of money

In the academic world, fashions come and go. In the late 1960s and early 1970s, there was much discussion about the role of money in the determination of inflation, a discussion that was led by the great minds of that time including Milton Friedman, Anna Schwartz and Philip Cagan.

Fast forward nearly fifty years, and it is not fashionable to discuss the role of money in the determination of the price level. Indeed, there is a view among many economists (notably, New Keynesian economists) that money is almost irrelevant to the discussion and that the size of the monetary base is only important in so far as it influences interest rates.

This is despite the fact that one of the strongest empirical relationships in economics remains the long-term correlation between the price level and the ratio of base money to real output. As Friedman once put it, inflation “…is and can be produced only by a more rapid increase in the quantity of money than in output.” This sentiment is clearly supported by the long-term data.

So, why is there such a disconnect? Why do academic economists largely dismiss the important role that the money/output ratio has in the determination of inflation?

First, the correlation between the price level and the money/output ratio breaks down in the short term. Although the quantity theory of money is valid over very long periods of time, it doesn’t work over short periods of time. Therefore, most economists feel comfortable ignoring quantity theory in their short-term forecasting of inflation.

Second, mainstream economics does not recognize the important role that the value of money plays in the determination of the price level. According to the orthodox view, the “value of money” has nothing to do with price determination!

Indeed, mainstream economics does not officially recognise the “market value of money” as a variable in any of its equations. Mainstream economics does not recognise the “market value of money” as a relevant economic variable because the view of mainstream (Keynesian) economics is that supply and demand for money determines the interest rate.

The view of The Money Enigma is that supply and demand for money (the monetary base) determines the market value of money (not the interest rate). In turn, the market value of money is the denominator of every money price in the economy and, therefore, is the denominator of the price level.

This general issue was discussed in a recent post titled “The Interest Rate is Not the Price of Money“. But rather than dwell on Keynesian theory, let’s briefly discuss how prices are determined at both a micro and macro level and then use this to continue with our discussion of the relationship between money and inflation.

What is a “price” and how is a price determined?

Price determination is a subject that we have discussed extensively over the past few months, so I don’t want to dwell on it in this post. We will discuss the basic principles today, but if you want more detail you should read the following posts:

“Every Price is a Function of Two Sets of Supply and Demand” (01/20/15)

“The Measurement of Market Value: Absolute, Relative and Real” (04/21/15)

“A New Economic Theory of Price Determination” (04/28/15)

In simple terms, the view of The Money Enigma is that every price is a relative expression of the market value of two goods.

Consider a simple exchange of two goods. Both goods must possess the property of market value in order for them to be exchanged. The price of the exchange simply measures the market value of one good in terms of another: the market value of a “primary good” in terms of the market value of a “measurement good”.

In our modern money-based economy, the measurement good most commonly used is money. The price of a good, in money terms, simply reflects the market value of the good relative to the market value of money. For example, if the price of a banana is $2, then we can say that the market value of one banana is twice the market value of one dollar.

The key point is that the price of a banana, in money terms, is a relative measure of value: it is determined not just by the market value of the banana (which can rise and fall) but also the market value of money (which can also rise and fall). If the market value of money falls, then, all else remaining equal, the price of the banana in money terms will rise.

In this sense, the market value of money is the denominator of every “money price” in the economy. In mathematical terms, the price of a good, in money terms, is a ratio of the market value of the good divided by the market value of money.

The trick to expressing this in terms of mathematical formula is recognizing that market value can be measured in the absolute. Just as we can measure any physical property in the absolute by using a “standard unit” of measurement for that property, so market value can be measured in the absolute by using a “standard unit” of measurement for market value.

The height of a tree can be measured using a standard unit for the measurement of height, namely “inches”. An “inch” is an invariable measure of the property of height. In economics, we can create an invariable measure of market value: a standard unit” for the measurement of market value that possesses the property of market value and is invariable in this property. Since no good exists that is invariable in market value, we need to create a theoretical measure, called “units of economic value”.

Once we have a standard unit for the measurement of market value, we can measure the market value of both goods being exchanged in terms of this standard unit. In other words, we can measure the market value of both goods in absolute terms.

This raises an obvious question: how is the market value of a good determined? In this respect, we can adapt an old paradigm: the market value of a good is determined by supply and demand for that good. If we plot supply and demand for each good in terms of our standard unit of market value, then we can see that the price of the primary good (good A) in terms of the measurement good (good B) is a function of two sets of supply and demand.

Supply and demand for the primary good (good A) determines the market value of the primary good. Supply and demand for the measurement good (good B) determines the market value of the measurement good. The price of the primary good in terms of the measurement good (the price of A in B terms) is the ratio of the market value of the primary good divided by the market value of the measurement good. This is a universal theory of price determination that we can apply to the determination of barter prices (good/good prices), money prices (good/money prices) and foreign exchange rates (money/money prices).

In a money-based economy, the price of a good, in money terms, is determined by the ratio of the market value of the good divided by the market value of money.

We can extend this microeconomic principle to a macroeconomic description of price level determination. If the market value of money is the denominator of every “money price” in our economy, then the market value of money is the denominator of the price level. (Remember, the price level is nothing more than a hypothetical measure of overall money prices for the set of goods and services that comprise the “basket of goods”.)

Once again, if we measure the overall market value of goods and services in terms of a standard unit of market value and denote this as V_G, and we measure the market value of money in terms of the same standard unit and denote this as V_M, then the price level is simply a ratio of V_G and V_M.

Ratio Theory of the Price Level simply states that the price level depends upon both the overall market value of the basket of goods and services and the market value of money. If the market value of the basket of goods and services is relatively stable over time, then the price level will be primarily determined by the direction of the market value of money. If the market value of money falls significantly over time, then the price level will rise significantly over that same period of time.

The determination of the market value of money

The view of The Money Enigma is that the market value of money is the denominator of every “money price” in the economy: the price of a good, in money terms, depends upon both the market value of the good and the market value of money.

If this theory is correct, then the market value of money plays a critical role in the determination of the price level and inflation. So, what determines the market value of money?

We have already hinted at part of the answer: supply and demand.

If every price is a function of two sets of supply and demand, then every money price must be a function of two sets of supply and demand. More specifically, the price of good A, in money terms, depends upon both supply and demand for good A and supply and demand for money (the monetary base).

As discussed in the introduction, the view of The Money Enigma is that supply and demand for money (the monetary base) determines the market value of money, not the interest rate. (The interest rate is determined by supply and demand for loanable funds).

While this might be an interesting first step, it really doesn’t tell us much about the factors that influence the value of money. In order to understand the specific factors that determine the market value of money, we need to develop a deeper understanding of what money is.

Over the past two weeks, we have discussed the nature of fiat money at length. In the first post, “The Evolution of Money: Why Does Fiat Money Have Value?” we traced the evolution of money from “commodity money” to “representative money” and finally to “fiat money”.

In that post, it was argued that representative money derives its value from an explicit contract: representative money is just a piece of paper that promises the holder of that piece of paper a real asset (normally, gold or silver) when that piece of paper is presented to its issuer.

When the gold/silver convertibility feature was removed, i.e. when the explicit contract was rendered null and void and the representative money became fiat money, the explicit contract was replaced by an implied-in-fact contract. In this way, fiat money derives its value contractually. Every asset derives its value from its physical properties (it is a real asset) or from its contractual properties (it is a financial instrument). Fiat money is not a real asset. Therefore, fiat money must be a financial instrument that derives its value from its contractual features.

The nature of the implied contract that governs fiat money was explored in a second post, “What Factors Influence the Value of Fiat Money?” While it is difficult to speculate on the exact nature of the implied contract, we can leverage finance theory to guide us in the right direction.

The view of The Money Enigma is that fiat money is a special-form, long-duration equity instrument issued by society. More specifically, fiat money represents a proportional claim on the future output of society.

And this brings us to the crux of the issue: what determines the value of fiat money and, consequently, the level of money prices in the economy?

If money is a proportional claim on the future output of society, then its value depends, at least primarily, upon future expectations of (1) real output, and (2) the size of the monetary base. Moreover, if money is a long-duration asset, then its value depends upon expectations regarding the long-term (20-30 year) path of these two important variables (real output and base money).

If the market suddenly decides that long-term (20 year) real output growth will be higher than previously anticipated, then the value of a proportional claim on that future output should rise (the market value of money should rise). All else remaining equal, the value of money will rise and the price level will fall.

In this example, there has been no change in current levels of real output or the monetary base, yet the price level has fallen. Why? The price level falls because it depends on the market value of money (the market value of money is the denominator of the price as per “Ratio Theory”). In turn, the market value of money depends upon long-term expectations. Current conditions really only matter to the market value of money to the degree that they impact expectations of long-term conditions. This is true of the value of any long-duration asset: current conditions are only important to the value of a long-duration asset in so far as they impact long-term expectations.

Now, let’s consider what happens if the market suddenly decides that the long-term growth rate of the monetary base will be much higher than previously anticipated. If money is a proportional claim on output, then more claims at some future point will mean that every claim is entitled to a smaller proportionate share of output at that future point. If the market decides that the future value of money will be lower, then this will have an immediate negative impact on the current value of money. Why? In simple terms, the market value of money depends upon a chain of future expectations regarding the future value of money.

Admittedly, this is a complicated concept and one that is explored in much greater detail in The Velocity Enigma, the third and final paper of The Enigma Series.

The key point that I wish to highlight is that, if proportional claim theory is correct, then the current market value of money depends upon the expected long-term path of both real output and the monetary base. Furthermore, since the market value of money is the denominator of the price level, the price level itself also depends upon the expected long-term path of both real output and the monetary base.

Bearing this in my mind, let’s return to our original question.

Does “too much money” cause inflation?

There can be little doubt that, over long periods of time (30 years+), growth in the monetary base that is greatly in excess of growth in real output will lead to a rise in the price level. There is strong empirical support for this observation.

This observation sits neatly with the theory that money is a proportional claim on the output of society. Over long periods of time, if the number of claims on output grows at a substantially faster rate than output, then the value of each claim should fall. In other words, if the monetary base grows at a substantially faster rate than output, then the market value of money should fall and the price level should rise (the market value of money is the denominator of the price level).

On the other hand, there is also compelling evidence to indicate that, over short periods of time, a dramatic increase in the monetary base can have little to no impact on the market value of money, even if the increase in the monetary base dwarfs any increase in real output during that same period of time.

This phenomenon has always been harder for economists to explain, but it can be explained by the theory that money is a special-form equity instrument and a long-duration, proportional claim on the future output of society.

If money is a long-duration, proportional claim on output, then the value of money will only be sensitive to changes in current levels of real output and the monetary base to the degree that changes in current levels impact expectations regarding the long-term path of both real output and the monetary base.

We can use a simple analogy from finance: the value of shares. The value of a share of common stock depends on the expected future cash flows that will accrue to the holder of that share. More specifically, a company’s stock price depends little on current earnings or current shares outstanding. Rather, the stock price is determined by expected long-term earnings per share. Therefore, it is the long-term path of both net earnings and shares outstanding that matter to the current value of a share of common stock.

Similarly, money is a long-duration asset and its value is primarily driven by expectations of the long-term real output/base money ratio, not by the current real output/base money ratio.

This has one important implication regarding market perception of monetary policy and its impact on inflation. If the market believes that a sudden rise in the monetary base is only “temporary” (it will be reversed in the next few years), then such an increase in the monetary base should have little to no impact on the value of money and, therefore, little to no impact on the price level.

However, if the market believes that a sudden rise in the monetary base is more “permanent” in nature, then that increase in the monetary base should lead to a fall in the value of money and a rise in the price level.

The view of The Money Enigma is that the dramatic increase in the monetary base in the United States has had little impact on the market value of the US Dollar (and little impact on the price level) because market participants believe that the increase is “temporary” in nature.

In slightly more sophisticated terms, the extraordinary actions of the Fed have not changed the market’s view regarding the long-term (20-30 year) path of the real output/base money ratio. Most market participants remain optimistic that real output will grow at solid rates for he next 30 years, even while the monetary base is reduced or at least capped at current levels. This has put a floor under the value of the US Dollar and a lid on the price level.

However, what happens if market expectations change? What will happen if the market becomes more pessimistic regarding the long-term economic prospects of the United States?

If market participants begin to believe that the Fed is unwilling or unable to reduce the monetary base, then this shift in expectations will begin to put downward pressure on the value of money and upward pressure on the price level. This fall in the value of money will be compounded if the market becomes more pessimistic about the long-term rate of real output growth in the United States. If this were to occur, then a return to double-digit levels of inflation is quite possible.