Ever wonder how many dollars there are? Where do they come from, where do they go? Why are they losing value so rapidly? A good friend has asked this question and found satisfactory answers elusive.
The question of “money supply” is one that economists have been wrangling with for a long time. You’d think there would be a simple, straightforward number, but it turns out there are multiple ways to define money, leading to names like M1, M2, MZM, etc. What’s more, it’s a moving target … the numbers are constantly changing.
For an official answer, a good place to start is with the agency that issues dollars, the United States Federal Reserve.
The absolute number of dollars, however, is less important than changes in the number. Kinda like if you live on the beach, you don’t care what sea level is but if it changes it’s a big deal.
Furthermore, there’s more than one way that dollars are created.
One is very simple: The Federal Reserve creates them from nothing in the act of buying Treasury securities. It takes the treasuries into its balance sheet as assets and virtually prints the money to pay for them as liabilities. The dollars the seller received didn’t exist before the transaction.
Another is when banks make loans. For this example let’s assume the banks maintain 10% (0.1) reserves. You deposit $1000 in the bank. The bank lends out $900 of it. So now someone else has $900 to spend, but you have not relinquished your claim on that money. (You can get it back on demand. Hence the name “demand deposits”.) So this has effectively increased the money supply by $900. What’s more, the $900 loaned can be deposited in the bank, upon which the bank can now lend $810. This can in turn be deposited in the bank, etcetera … with the result that when you tote everything up the original $1000 is now $10,000.
This is the so-called “money multiplier” effect of fractional reserve banking. The multiplier is the reciprocal of the reserve ratio. With this mechanism (unlike with direct Fed creation), the amount of money created is open ended, a function of both the reserve ratio and interest rates, as well as other financial factors. But it is for this reason that the number of dollars extant is not easily nailed down … they’re constantly being created and destroyed in the process of being lent by and repaid to banks.
Thanks for addressing what I contend is one of the most important questions in Finance and Financial markets.
Remember, the reserve requirement for banks is now zero.
And while you contend that the change is more important than the level, I wonder, how can you measure the change if you don’t know the level?
Yes … with a reserve requirement of zero the potential money is infinite. The way this happened is a bit tricky. In the mid-nineties the Greenspan Fed started allowing the banks to sweep their demand deposits (where there was a reserve requirement) into savings overnight (where there wasn’t). This way when the bank examiners reviewed the books at night the banks were found compliant. I think it no coincidence that this marked an inflection in stock prices that led to the bubble.
How can you know the change without knowing the level? Recall my example of the home on the beach. Ask the person that lives there, “How high is sea level?”. He would find it difficult to answer. He might locate an encyclopedia and tell you somewhere around 4000 miles from the center of the earth. But ask “How much has sea level changed since yesterday?”, and he could tell you quite precisely … in feet, maybe even inches.
Another example from the physical sciences: When temperature scales were first devised, it wasn’t understood exactly what was being measured. People just knew warmer from colder. So scientists arbitrarily set temperatures they would call “zero”, and allow negative readings below that level. The Celsius scale was originally based on zero being the freezing point of water. The Fahrenheit scale the freezing point of salt water. No attempt was made however to ascertain what temperature would correspond to no heat at all. So they settled for measuring temperature differences without attempting to state an absolute amount of temperature.
This is where the expression “absolute zero” came from. As science learned more about heat and temperature, it was able to identify a point at which there was no heat. This came to be known as absolute zero. Only then was it possible to give an absolute temperature, resulting in the creation of the Kelvin and Rankine scales.
Even today, most people only use the Celsius and Fahrenheit scales. These scales only state temperature differences – from their respective zeroes – because most of the time the absolute temperature is unimportant.
The relevance to this discussion is simple. Science was able to identify how to measure changes in how much heat there was long before they were able to measure the actual level of heat. Measuring change in a variable can indeed be easier than measuring its absolute level.
BTW the analogy between altitude and temperature is close. When we measure altitude we usually state it as a difference from sea level, as opposed to in absolute terms (distance from the center of the earth). We do the same with temperature, as a difference from a specified level, as opposed to in absolute terms (how far it is from no temperature at all). It’s quite similar with money as well … markets are generally much more interested in how money supply is changing than in its absolute level.
A person who lives by the sea can recognize a typical level of that sea, which is part of nature.
A person who lives in certain location will recognize climatic patterns, which is part of nature.
Within the man-made system of dollar supply, where that supply and it’s changes are “manufactured” by a select group of insiders, that “scarcely one man in a thousand (or was it a million) could understand”, I don’t have a clue what the level is, or the change, certainly not the change until after a significant delay, and having that relative knowledge after that delay does me no good because my savings and/or opportunities has already been fleeced by those who did have that knowledge.
Levels and Flows. Know that and your ability to navigate is tremendously improved. And the fact that those levels change is precisely why the information has to be updated as often as necessary.
The situation is also highly analogous to the absolute value of a dollar. It doesn’t matter what it is … as long as it doesn’t change nobody cares. A dollar could be arbitrarily set so it requires one tenth of one to buy a loaf of bread, or ten of them, and it’s not an issue. When that changes, however, we have inflation or deflation and it matters a lot because our whole economic system depends on having a reliable measure of value. You work for so many hours to receive so many dollars, and if the value of those dollars drops before you spend them you’ve been ripped off.
It applies even more so to the number of dollars, because we don’t even have a foolproof way to observe those numbers. We only observe the effects of changes in the number of dollars via changes in the market value of those dollars.
Is the market value of dollars only impacted by changes in the number of dollars? Also, to the extent changes in the number of dollars takes place, how long does it take for the market value of dollars to reflect that change?
(I believe you are familiar with the Cantillon effect, which says those close to money expansion benefit, while those farther away (ie common people and common savers) suffer.)
No, the market value of dollars is a function of both supply and demand for dollars. Like anything else. Of course, at a given level of demand, the supply is left as the variable. Most economists however fail to take demand into account, however, leaving demand the less acknowledged variable.
The change in market value due to changes in supply and demand is virtually instantaneous. It is however reflected first only in markets that trade in real time, tick by tick, such as bonds and stocks. The effects on consumer prices and wages in contrast can take years to respond … a key point in many Financology articles.
Indeed the production of new money results in a transfer of wealth to those near the beginning of the bucket brigade at the expense of those near the end.
Think of:
– All the Eurodollars. Jeff Snider, Brent Johnson, and others say that there are “uncounted” dollars abroad. Many of those went abroad in exchange for imported goods and services. Many dollars are sent abroad by foreigner’s working in this country. Jeff Snider, and others, say that dollars are lent out by financial institutions abroad beyond those dollars they hold, in effect creating new dollars. Occasional difficulty in foreign borrowers getting dollars to pay back the lenders, results in a dollar shortage/squeeze. This is where currency swaps are sometimes made between our FED and foreign central banks.
– How about those pallets containing billions of dollars that “disappeared” from Iranian airports.
– How about all the dollars held by foreign drug cartels.
– How about the dollars exchanged for gold and bills early in the 1900’s. (Dollars were exchanged for other assets.)
– How about all the dollars introduced via bad loans. The loans were later written off, but the dollars stayed in the system.
You just helped answer your own question, CB. These are just some of the reasons it’s impractical to get a count of the number of dollars in existence. Who knows how many are hiding underneath mattresses in Toledo? Or Shenzhen?
And as I pointed out earlier, they’re constantly being created and destroyed … even if you could get an exact count, it would be wrong in a fraction of a second.
But it is possible, in contrast, to observe the effects of changes in supply and demand for dollars via changes in the market value. And that of course is of direct concern to all holders and owers of dollars … directly or indirectly most of humanity. We do that here with the FDI – the Financology Dollar a index – most recently updated in The Big Takeaway.
There you can see graphically the effect of producing dollar supply far in excess of demand.
I would say it’s not so much of whether the Fed can take dollars out of circulation, but whether it will. When it buys Treasuries, it creates the dollars to buy them with de novo. When it sells them they are retired.
The complication arises in that this is far from the only way dollars are created and destroyed. Lending via ordinary banking transactions accounts for a huge amount of it … and there the Fed only targets the rate. The number is unlimited. At least with direct QE the number of dollars created is finite and controlled.
Interest rate manipulation is far more destructive.
Of course the buying and selling of assets other than pure monetary assets of gold and treasuries is fraught with peril too. If that’s all the Fed did the world would be a better place.
” Lending via ordinary banking transactions accounts for a huge amount of it … and there the Fed only targets the rate. The number is unlimited.”
this is technically true, the only limitation being the perceived quality and risk of the loan and the loan’s collateral. That is why I pointed out the great collateral that Treasuries are and the willingness of banks to loan against them as collateral. So, when the FED exchanges those balance sheet Treasuries for dollars they are doing something —– just not near as much as one might think. Unless those initially created dollars that were used to purchase those new Treasuries are retired AND those Treasuries are also retired, then additional purchasing power has been introduced into the system which has not gone away. First it was there with pure dollars. Then, after the exchange, it is there in the form of Treasuries, which are easily converted into dollars. So a bit more difficult to unwind monetized Treasury purchase than many think.
I don’t necessarily think I have persuaded you, and certainly not Wolf. That is why I would like to see this debated on an open forum. A lot of things were hashed out on the DR forum.
If I understand you correctly, John Hussman has made a similar point. Treasuries – securities issued by the US Treasury – and dollars – securities issued by the Federal Reserve – are fungible. Converting from one to the other by itself is just replacing one obligation with another.
Treas notes, Fed notes … what’s the dif?
The catch is … provided interest rates are very low. UST are yield bearing instruments … USD are not. This is one reason I maintain that interest rate policy is more destructive than quantitative policy.
Another relevant point is that ordinary people don’t run around with UST in their wallets to buy things with. They do with USD.
A more relevant point is that whatever the similarities and differences between USD and UST, the Fed actions of buying UST and selling UST are the inverse of each other. Dollars are created when the Fed buys UST and they’re destroyed when it sells UST. The amount of outstanding UST on the Treasury balance sheet remains the same regardless. The holders of the UST change from other entities to the Fed and back again.
So if the question is how many dollars are there, the answer is the number increases when the Fed buys stuff and decreases when it sells stuff.
No matter what the stuff is.