Editor's Note: Below is a Hedgeye Guest Contributor research note written by Dr. Daniel Thornton. During his 33-year career at the St. Louis Fed, Thornton served as vice president and economic advisor. He currently runs D.L. Thornton Economics, an economic research consultancy.
A brief note on our contributor policy. While this column does not necessarily reflect the opinion of Hedgeye, suffice to say, more often than not we concur with our contributors. In the piece below, Thornton discusses how "Economists like to talk about aggregate demand and aggregate supply in spite of the fact that neither exist in any meaningful, real-world way."
I was surprised to find several economists talking about “the demand for money” in the context of determining the rate of inflation. They talk as if money demand is meaningful in the U.S. But it’s not. Indeed, I doubt that it is useful in any financially sophisticated economy. This Common Sense Economics explains what money demand once was and why it is no longer meaningful or useful.
What Money Is
Before we can talk about the demand for money, we must know what money is. So what’s money? Money is something that societies invented to facilitate trade. If everyone was self-sufficient and there was no exchange of goods and services among individuals (no trade), there would be no need for money—the word wouldn’t exist. Here is the basic idea of what money does: If you are good at producing corn and your next-door neighbor is good at producing pigs, you and your neighbor are likely to want to exchange corn and pork. No problem, you sit down over a cup of coffee and agree on how many bushels of corn will be exchanged for a certain quantity of pork—it’s easy! This is called barter.
But most trade occurs between strangers. Barter is not very efficient for this reason, and for a variety of others, see Money in a Theory of Exchange. Consequently, societies invented money. Money is just an asset that is used to facilitate trade—think gold. One simply exchanges a certain quantity of money for the things you want. The person selling you those things then uses the money to purchase the thing he wants and so on and so forth. Money is said to flow through the economy. But this characterization is misleading because money is being held by someone at every point in time: Money is an asset.
Like everything else, money has evolved over time. Initially, monies were commodities, like gold and silver. These commodities had intrinsic value, so people were happy to trade goods for them. Modern-day currencies—the dollar, yen, pound, euro, franc, etc.—have no intrinsic value; they are held solely because they can be used in exchange. These are called fiat monies because they are money simply because a particular sovereign government declares they are.
To understand the demand for money, it is useful to initially assume that money consists solely of currency and that all transactions are carried out with currency. So how much money do people demand? The answer is provided by simply adding up all the currency that people are holding. Of course, people give up money when purchases are made and get money when something is sold. Consequently, the amount of money one holds fluctuates over time.
Hence, it more reasonable to talk about the demand for money on average over an interval of time, say a month, rather than at a particular point in time. How much currency do people hold on average over a month? Of course, this will depend on the price of the things people buy and a variety of other factors. The level of prices is important because the higher the price level is, the less real goods a given amount of money will buy: If apples cost $1, $100 will buy 100 apples, but if they cost $2 (think Honey Crisp apples), $100 will only buy 50.
Consequently, economists talk about the demand for real money—the amount of money one wants to hold divided by the price level. The higher the price level, the more money a person would want to hold. In any event, if you knew how much currency (money) each person wants to hold on average during a month, you could simply add all the individual money demands up and find the total demand for money.
The government can print as much currency (money) as it wants. Therefore, the supply of money would be totally determined by the government. If the government prints more money than people demand, people will try to get rid of the excess money by spending it. The demand for goods and services will increase and, so too, will the prices of goods and services; the price level—the money-price of goods and services—will increase. The equilibrium price level will be determined by the quantity of currency (money) demanded relative to the quantity supplied—the price level will be determined by the supply of and the demand for money! If the government increases the amount of currency it supplies, the price level will rise. If it reduces the supply of currency, the price level will fall. This is the monetary theory of inflation: Colloquially, too much money chasing too few goods.
Things become a little more complicated if people use both checks and currency to make their purchases. Now the demand for money is the demand for checking account balances and currency. The major difference is that the central bank has less control over the supply of money.
Money Demand in Financially Sophisticated Economies
The big problem arises when people begin using methods for making transactions that cannot be measured effectively. For example, I make over 80% of my transactions using a credit card. In my case, this is not a big problem because I pay off my outstanding balance monthly by drawing on my checking account balance. So the only effect of making my purchases with a credit card is that my average checking account balance is higher than it would otherwise be—my demand for money is higher. If everyone did this, there would be no problem. The traditional M1 measure of money — total checkable deposits plus currency held by the on-bank public—would be useful.
A more difficult problem arises when banks pay a higher rate of interest on savings accounts than on checking accounts. This gives me an incentive to keep more funds in my savings account. I can make inter-month transfers from my savings to my checking account when I need more money. This is easy and relatively costless in the world of on-line banking. In this circumstance, I can significantly reduce my demand for checking account balances by transferring funds from my savings account when it is time to pay off my credit card balance. Hence, my demand for money would be lower because my average checking account balance would be lower; however, my average savings account balance would be higher.
You might be saying, “Well, just include savings balances in the definition of money.” Indeed, some economists have done this; this is the M2 definition of money—M1 plus savings account balances at depository institutions. One problem with doing this is that I may be holding some balances in my savings account for reasons other than making transactions—for saving. Hence, including them would overstate my demand for money. There is also the ancillary problem that the measure of money now includes something that cannot be directly used to make transactions; I must transfer funds from my savings account to my checking account before I can use them to make transactions, i.e., before they become money.
The above issues weaken the link between money and prices, i.e., the link between the demand for and the supply of money. But a much begger problem is caused by the widespread use of credit cards for making transactions. The problem is there is no way to measure the quantity of transactions services held by individuals in credit cards. Indeed, credit cards enable individuals to make lots of transactions even if they hold little or no money, M1 or M2. Hence, there is no way to determine the quantity of money demanded. There is no way to determine the supply of money either.
This is an enormous problem for the monetary theory of inflation. For example, the M1 measure of money has increased by 230 percent during the last 7.5 years, more than during the previous 30 years, yet inflation has been modest and is below the Fed’s 2 percent inflation objective.
Several economists have argued that money need not be measurable in order to have knowledge of the demand for money, or to conduct monetary policy in order to control inflation. They argue that it is possible to know whether the supply of money is too large or too small by observing aggregate demand. They suggest that a marked increase in aggregate demand signals that the supply of money is growing too fast—faster than the demand for money. A decrease in aggregate demand means the money supply isn’t growing fast enough—the supply of money is not keeping up with money demand. Their conclusion: Money doesn’t need to be quantifiable; the behavior of aggregate spending is all central bankers need in order to manage the money supply and, thereby, control inflation.
There are a number of reasons why this assertion is absurd. The most important is the FACT that economists and policymakers have no idea what aggregate demand is. I have noted elsewhere, Does Aggregate Supply Exist?, that economists like to talk about aggregate demand and aggregate supply in spite of the fact that neither exist in any meaningful, real-world way. Hence, it is impossible to know whether aggregate demand is increasing too fast or has increased too much, or whether it isn’t increasing fast enough.
In the model that most economists and policymakers use, real GDP is determined by the intersection of aggregate demand and aggregate supply at any point in time. But all economists and policymakers see is an estimate of the current level of real GDP. If real GDP is growing faster or slower, they don’t know whether it is due to factors affecting supply or demand. They simply assume that if the economy is growing slower than expected it’s due to weak demand; if it’s growing faster than expected it’s due to strong demand.
I don’t believe economists should talk about the demand for something that cannot be quantified. If it cannot be measured, it is difficult, if not impossible, to estimate the demand for it. Consequently, it cannot be used to conduct policy or do anything else. When the bulk of transactions were carried out by checks and currency, it made sense to talk about the demand for M1. As I noted above, there was an issue about how well the supply of M1 could be controlled, but one could make a reasonable attempt at identifying the demand for M1 for policy purposes. It is well known that the relationship between M1 and things economists and policymakers care about broke down in the early 1980s, see The Velocity Puzzle. Since then, economists and policymakers talk about money, but pay little or no attention to any measure of it.
It makes no sense to talk about the demand for money in financially sophisticated economies. The problem is that a large and increasing amount of transactions are being carried out by a variety of things that cannot be measured. If there was a strong, stable, and predictable relationship between the growth of GDP and inflation, one could make a stronger case that money was not important for controlling inflation, but that is not the case.
However, even if there was a strong, stable relationship between inflation and output growth, it is not clear how well policymakers could control inflation because it is not clear that monetary policy actions have a strong effect on either inflation or output growth, see Why the Fed's Zero Interest Rate Policy Failed. The real problem is that economists don’t have a theory of inflation that has any credible predictive power, see Monetary Policy and Inflation. In any event, it makes no sense to think that the demand for money is useful for policy purposes.
It is important to note that the above analysis in no way implies that money is not an important concept or that money is not important for the economy or transactions. The existence of currency, and immediate substitutes for it (checkable deposits), are used to make many transactions. More importantly, as I have noted elsewhere, Why Money Matters, the existence of money is essential for the widespread use of credit and, hence, for the ability to use credit to make day-to-day transactions. It’s also essential for final settlement; nearly all transactions are settled using currency or checking account balances.