If Deflation Is Coming, Sell Your Gold

The debate over inflation versus deflation has been going on in hard money circles since about 1973. The debate has gone on within academic circles for well over a century. The economists are as confused as the general public, but they are confused in a far more sophisticated way. They turn confusion into a science.

I follow the Austrian School of economics on monetary theory. The most important study of the theory of money within the Austrian School camp was published in 1912, The Theory of Money and Credit, written by Ludwig von Mises. You can download it for free here.

More popular and more readable books have been written on this by his disciple, Murray Rothbard. His book, What Has Government Done to Our Money? is the clearest exposition ever written. You can download it here.

I have also written a short book on the topic, Mises on Money. Download it here.

Finally, there is my more detailed book, Honest Money.

Austrian School economists define inflation as follows: “an increase in the money supply.” All other schools of thought define inflation as follows: “an increase in our favorite price index.”

Austrian School economists define deflation as follows: “a decrease in the money supply.” All other schools of thought and define deflation as follows: “a decrease in our favorite price index.”


You don’t buy a price index. You buy a specific product or service. It does not matter what the price index says if you can’t afford to buy much.

What is a price index? It is a statistical construct that is created in terms of a theory of how an economy operates. It is supposed to measure increases and decreases in prices. It does this. The question is: Which prices? Retail? Wholesale? Both?

How many prices are there? Far more than the number of retail products and services. In a recent estimate of the number of products — not services (60% of the U.S. economy) — in the New York City area, one economist offered this number: at least 10 billion. (Eric Beinhocker, The Origin of Wealth, p. 9.) This does not include discounts, for which New York City is famous.

Now add Los Angeles, Chicago, Dallas, and Atlanta.

Get the picture?

Begin here: “A price index is a statistical illusion. Some illusions are more accurate than others. Nobody knows which ones.”

Economists, meaning a committee of economists, try to come to some agreement about which products and which services in an economy are representative of the largest number of people living in this economy. There is no real agreement here, but they have to come up with something. In addition to deciding which products and services are most representative, they also have to come up with a system of assigning relevance, or “weights.” For example, they have to decide which is more important for the supposedly typical resident: food or fuel. There is no agreement here, either. Economists do their best to examine statistics of how much money was spent last year on which items. Then they come to an agreement of sorts regarding the products and services to be used in the index, and how much importance to assign to each of them.

Then the economists turn their theory over to statisticians. Statisticians, hired by the government, create sophisticated statistical analyses based on statistical sampling. When they are done with this sampling (at government expense and compelled by law for those questioned to answer), they put together an index of prices for a particular time period.

The problem here should be obvious: the public changes its taste over time. A product or service that was widely used in one period of time falls in popularity. Similarly, some other product, possibly one that did not even exist 10 years ago, becomes wildly popular. Think about the iPod.

This means that the statisticians must get together with the economists and hash out the issue all over again in the future. We do not know how long this future ought to be. It probably should be five years. But who knows? The problem is, when the products in the “basket” change and the importance assigned to each product changes, how can you compare the price index of five years ago with today’s price index?

This debate has gone on for decades, and it will go on for as long as there are economists who want to study the relationship between practically anything, especially money, and price changes. A good book on this is by Oskar Morgenstern, a student of Mises but not a disciple: On the Accuracy of Economic Observations (1963). For a summary, read this.

Economists are probably less agreed on what constitutes money than they are about what constitutes the proper basket of goods to be used as representative of prices. So, I guarantee you, economists do not agree on these matters. At best, there is a kind of sullen acceptance of a particular index.

The most popular index of prices in the United States is the Consumer Price Index, or CPI. I prefer a rival index, the median CPI, published by the Federal Reserve Bank of Cleveland. I think it is more accurate, but I would not spend a lot of time trying to prove this. The important thing is this: you pick one index, and then you follow the trend for that index. The trend is what counts. There can be revising with an index every few years, some of which is legitimate, but the general trend will reflect changes in the economy. The trend will reflect how changes in the money supply change the prices of products and services that a large number of people purchase.


There can be price inflation without an increase in the money supply. This can happen during times of a disaster. For example, the price of goods rises during a hurricane, especially goods such as bottled water, gasoline, and canned goods. Similarly, in times of increasing productivity, it is possible to have falling prices. The money supply stays fairly constant, and the number of goods and services increases. This is a situation that we can call, “more goods chasing the same amount of money.”

Long-term price deflation would exist in a free-market economy that operated in terms of the legal requirement that all warehouse receipts for precious metals would be backed 100% by the metals promised by the warehouse receipts. This would be a system of 100% reserve banking. This is what Murray Rothbard favored. Prices might fall at 2% or 3% per annum. This would be good for the consumer. Falling prices are a good thing if this comes as a result of increased productivity. If you don’t think so, look at the price of electronic calculators or computers.

In almost every Western country in the years following World War II, prices have risen. In Western industrial nations, prices have risen at anywhere from 2% to 5% per annum. This is because the money supply has increased in these nations. A steady, long-term increase in the price level is always the result of an expansion of the monetary base by the central bank, which is spent into circulation and then multiplies through the fractional reserve banking system. No other explanation of long-term price inflation has ever found any theoretical or statistical validation. On this, virtually all economists are agreed. (This is rare.)

When an economist predicts inflation, he means something greater than the normal 3% per annum. Whenever you read a prediction saying “inflation ahead,” this prediction is saying nothing special. If somebody says “mass inflation ahead,” that is significant. If somebody says “hyperinflation ahead,” that is very significant. In every case in the last half-century, anybody who made this prediction turned out to be wrong. There is one prediction that has been even more wrong: the prediction of falling prices. The last time this happened was in the mid-1950s. It has not happened yet in the United States. The consumer price index rose by 0.1% in 2008. The median CPI rose by 2.9%.

The big debate today is over what constitutes money. If money is M-1, and M-1 rises, there has been an increase in the money supply. But if, through a refusal of banks to lend, the money multiplier falls, the two forces can offset each other.

The multiplier falls when one of two things happens: (1) the Federal Reserve increases the legal reserve requirement for banks; (2) the Federal Reserve pays interest on reserves held at the FED. This was never done in the United States until October 2008. It was not legally allowed to be done. The law was changed so that it would be legal to do this, beginning in October of 2011. Without warning, the Federal Reserve decided to jump the gun by three years. It began paying interest to banks for reserves in the fall of 2008. Excess reserves rose rapidly, in the range of $800 billion. This offset the effect of the increase in the money supply, as denominated by M-1. You can see the two graphs here.

This was a deliberate change in Federal Reserve policy. It was done specifically to neutralize the effect of the doubling of the monetary base by the Federal Reserve, beginning in September 2008.

The debate between those who predict price inflation and those who predict price deflation is operationally meaningless if we are only talking about 2% or 3% either way. Statistical sampling can produce an error that large. Also, the weighing of the particular price index could easily have an error range this large. Those who predict mass inflation or hyperinflation are predicting something significant. Anyone who predicts 2% or 3% price deflation is not predicting anything very significant. It can happen, although it hasn’t happened in over half a century. The debate isn’t about 2% or 3% either way. The debate is about whether or not we are entering into a fundamentally different period of time, in which the traditional 3% price inflation will be supplanted by something fundamentally different.

The case for the double-digit price deflation is so weak that no economist predicts it in public. I do not mean hardly any economists, or not too many economists; I mean no one academically trained with a Ph.D. in economics predicts price deflation comparable to that which took place from 1930 to 1933. Anyone who predicts that, and predicts that over several years, has identified himself as not being a trained economist. I’m not exaggerating; I really mean it. No such economist exists. If he does, he has hidden himself well.

There are not many economists who predict price inflation of 10% per annum or higher. There are a few, but they are not famous, and they are almost always Austrian school economists. Why do they predict this? Because they look at the expansion of the monetary base, and they know that the banks will eventually begin lending, which will then multiply the money supply according to standard monetary theory regarding fractional reserve banking. There is nothing unique about Austrian School economics with respect to the operation of fractional reserve banking. The main difference is this: those Austrian school economists who follow Rothbard (not all do) argue that fractional reserve banking is immoral. Fractional reserves are a form of theft through fraud. They also argue that fractional reserve banking leads to the boom bust cycle. This makes them unique. But there is nothing unique about the analysis of how the system operates, as presented in Rothbard’s book, The Mystery of Banking.


Within the hard money movement, there are a few people, following John Exter, who predict price deflation and a rise in the price of gold. I have believed for 35 years that this system cannot be defended. If there is severe price deflation, then the hardest money asset in the world is the currency that is being deflated. If there is a shrinking money supply, there will be a shrinking price level. Hard money in this situation is the national currency that is shrinking. Gold would fall under such a currency system. There is no such currency.

The fact that some asset prices fall in a time of price deflation only means that these prices were imputed prices, which means that the last transaction sets the prices. This is true of stock prices. This is true of commodity prices in a futures market. But when a particular commodity or product is bought on a regular basis, such as food in a supermarket, the possibility of error in forecasting falls remarkably. This is because people buy the products on a regular basis, and so those who imputed future prices based on the latest price probably will not make a serious mistake, either up or down. It will be confirmed in the market shortly.

Prices of homes can go up or down rapidly and shortly, but this is because people rarely sell their homes. The price of a house down the block is imputed to all the houses on the block. That can happen at any time.

The money supply does not shrink just because somebody’s house falls in price, because somebody else’s house down the block sold for 10% less this year than it did last year. The money supply has not changed at all. The money is in somebody’s bank account. If one person loses money in a transaction, somebody else will be able to buy whatever it was that fell in price. He will buy it because he has money in a bank account. When he writes a check to buy the item, that money is transferred to the seller’s bank account. The money supply has not changed.

It is possible that capital assets can fall in price because of leverage in the previous boom period. In other words, people made contracts that they cannot fulfill. When the banks do this, and they do not fulfill them, and they are allowed to go bankrupt, this does shrink the money supply. This happened in 1930—33. But the point is, the central bank does not allow banks to go bankrupt in this way any longer. The money supply generated by the banking system as a whole does not change because bank A goes bankrupt, and its assets are bought by bank B.

Whenever there is a threat that there will be a contraction of the money supply because banks default on all debt, the central bank intervenes. So does the Federal Deposit Insurance Corporation, which arranges that the depositors’ assets are transferred to another bank. The depositors are protected, and their deposits continue to remain the basis of an expansion of loans. One more time: the money supply does not shrink because a bank goes belly up. The central bank always intervenes to make certain that the money supply does not shrink.

It has not shrunk so far. Because the affected, operational supply of excess reserves has increased because the Federal Reserve System changed its policy and began paying interest on reserves does not change the fact that the monetary base has doubled. It does not change the fact that the money supply, whether M-1, M-2, or MZM, will rise rapidly when banks begin lending the money that presently is tied up at the Federal Reserve System. The banks will begin lending soon, because the Federal Reserve System is now paying 1/10 of 1% per annum on this money. The banks are losing money hand over fist because they have to pay depositors a rate of interest, and they are not lending the money at the FED at a higher rate of interest. So, they cannot make any profit on the difference between the interest rate to the borrower and the interest rate to the depositor.


I hope this clears up some of the debate between those who predict serious price inflation and those who predict serious price deflation. When I say serious, I mean at least 10% per annum either way. If somebody is predicting 2% price inflation, and someone else is predicting 2% price deflation, flip a coin.

If somebody is predicting serious price deflation or serious price inflation, make that person answer this question: “Under your scenario, what is the relationship among the monetary base, the money supply, and prices?” Everyone in this debate knows that the monetary base has doubled. An unresolved question is this: “Will the banks lend the money that is presently at the Federal Reserve in the form of excess reserves earning one-tenth of one percent?”

Then ask this: “When the banks pull their money out of the FED when the economy revives or else they are going bust because of interest paid to depositors, and when that money gets to the general public, what will be the result: price inflation or price deflation?” Make certain that these are the two issues that the person who predicts either serious price inflation or serious price deflation deals with in print. If you don’t understand his answer, you can be pretty sure that he doesn’t, either.

By the way, the deflationists have not had their prediction come true in a single year since 1967, when I started tracking this debate. They may go into rapturous joy if, in 2009, the CPI falls by 1%, if it does, which I doubt. “We were right!” But the dollar was worth over six times as much in 1967 when J. Irving Weiss (deflationist Martin Weiss’s father) first presented his deflationist prediction at the very first gold investment conference, sponsored by Harry Schultz, which I attended. The dollar was almost five times as high when John Exter pitched his theory in 1973. You can verify these numbers here. They use the CPI as the index.

When prices fall by 10% or more per annum for three consecutive years, I will be impressed. Even 5% will get my attention. Until then, let me say this: “The deflationists’ never-changing prediction has been wrong since 1967.”


What I am saying is this: prices are more likely to rise by 10% or more for three consecutive years than fall by 5% for three consecutive years. I will up the ante. They are more likely to rise by 20% than fall by 5% for three consecutive years.

For more evidence from a retired professor of finance, click here.

January 31, 2009

Gary North [send him mail] is the author of Mises on Money. Visit http://www.garynorth.com. He is also the author of a free 20-volume series, An Economic Commentary on the Bible.

Copyright © 2009 LewRockwell.com