From inception of a monetary economy with a government-issued currency, it is clear that government spending must come before tax payments or purchases of government debt. The order of requirements is basically: (i) government defines its monetary unit of account; (ii) government imposes taxes and other obligations that can only finally be settled in that currency; (iii) government spends (or lends) its currency into existence; (iv) non-government can now obtain the currency and, among other things, pay its taxes and purchase government debt. It is clear that government spending must logically come before tax payments or purchases of government debt because non-government must be able to get hold of the currency before it can do these things
For those of us trying to convince others of this elementary point, a difficulty is that once the economy is up and running, and government has already spent more into the economy than it has taxed out, it can arrange procedures so as to make it appear as if tax payments come first. I am referring, here, to the constraints governments voluntarily impose on their own actions.
I’ll take the US as an example, partly because of its prominence in the world, partly because it has an especially convoluted set of procedures in place, and partly because there are extremely helpful treatments of the US case in a paper by Scott Fullwiler and post by L. Randall Wray. Similar, if less convoluted, procedures also apply in many other countries. I have previously put up a fairly long post pertaining to the US case, which closely follows the links just provided. Here, I will just focus on a couple of aspects that illustrate the challenges before us.
First consider government spending in relation to taxes.
In the US, government has voluntarily imposed upon itself the rule that the Treasury has to have the necessary balance in its Treasury account with the Fed before it can spend. The Treasury has ‘tax and loan’ accounts that are held at various banks. These hold sums resulting from tax payments and treasury purchases. Prior to the government spending, the Treasury makes calls on these tax and loan accounts. The tax and loan accounts are debited and the Treasury’s Account at the Fed is credited. This can make it appear as if tax payments are recycled rather than new government money being created when the government spends. Here, ‘government money’ is defined as currency (notes and coins) plus reserves. It is sometimes referred to as ‘high-powered money’ or by other names.
The fact that the Treasury makes calls on its tax and loan accounts prior to spending can be understandably confusing to people who are aware of this procedural reality. This requirement seems to fly in the face of the logical point that government spending is prior to tax payments. But it is only an appearance, not a reality.
One way we can see this is by noting that when the Treasury makes its calls on tax and loan accounts, not only are deposits in these accounts debited but reserve balances held at the Fed by the banks in question are also debited. Therefore, government money (in the form of reserve balances) is destroyed.
The Treasury’s account is credited, but this does not prevent the destruction of government money caused by the deletion of reserves, because the balance in the Treasury account is not government money. We can see that it is not government money because the balance in the Treasury account, which is an asset of the Treasury, is at the same time a liability of the Fed. So for the government as a whole, it nets to zero, and so cannot constitute government money. Government money is a liability of the government. But here, government money that used to exist no longer does.
In this procedure, the Treasury account is really little more than a record of what Treasury is now permitted to spend in keeping with the government’s voluntarily self-imposed constraint that the Treasury have a sufficient balance in its account prior to spending. When the Treasury actually spends, bank deposits are credited and reserve accounts also credited. New government money is created in the form of reserve balances.
So it is true to say that government spending creates new government money, whereas taxation destroys it.
Since the destruction of something cannot logically occur prior to its creation, clearly government spending (which creates government money) logically precedes tax payments (which destroy government money).
But it perhaps is unsurprising that people aware of these operational constraints might be led into thinking that tax payments come first.
Now consider government debt issuance.
In addition to the tax-and-loan-account “fun and games” described above, a Treasury auction, in which the US Government supposedly “borrows” its own money from the private sector, is initiated — in something of a circus act — by the Fed orchestrating a reserve add as the first leg of a repurchase agreement. This is an agreement in which primary dealers agree to exchange old treasuries for newly issued reserves and undertake to participate in the reverse operation once the government spending is about to be executed.
Functionally, the repurchase agreement enables the Fed to maintain control of the federal funds rate. In the first leg, it negates the effect of the Treasury auction on reserves, since the latter on its own depletes reserves. Then, in the second leg, the Fed has to negate the effect of the government spending, which will add reserves.
In effect, the Fed issues new government money (reserves). This money is made available to the private sector (with collateral required in return). The Treasury can then appear to “borrow” from the private sector. But this just causes a depletion of the government money issued in the first leg of the repurchase agreement.
Again, it is perhaps not a surprise that the operations cause as much confusion as they are possibly intended to cause.