Deficits and Money


Alchemy or Banking?

A recent post of Paul Krugman brings to my attention a few paragraphs near the end of Jamie Galbraith’s testimony to the Deficit Commission which call out for a money view explication. To my mind, Krugman’s toy model does not really engage the issue, and Galbraith responds to that toy model rather than providing the needed explication. Both men are really more interested in deficits than in money, but not me, so here goes.

The title of Section 9 of Galbraith’s testimony gives the gist: “In Reality, the US Government Spends First & Borrows Later; Public Spending Creates a Demand for Treasuries in the Private Sector.” Krugman, perhaps sniffing MMT heresy, translates: “deficits are never a problem, as long as a country can issue its own currency.” But that’s not what Galbraith is saying, at least not in my reading.

Galbraith begins with the government making a payment by writing a check, for concreteness let us say a Social Security check. The immediate effect is to expand both sides of the balance sheet of the banking system by the same amount, additional reserves on the asset side and additional deposits on the liability side. On the Fed’s balance sheet, this same operation shows up as a debit from the Treasury account and a credit to the bank’s account.

Matters could well simply stop here. In Flow of Funds language, the Treasury has financed a cash outflow by dishoarding, by drawing down its money balances at the Fed.

But matters will likely not stop here. After the payment, the banking system has reserves that it didn’t have before, and they are paying only .25%. By hypothesis, banks would like to acquire an asset with a somewhat higher yield. That’s what Galbraith means when he says spending (by the government) creates a demand for Treasuries (by the private sector). But why Treasuries, as opposed to other private securities?

The key point—and here I aware of filling in a few steps that Galbraith jumps over—is that logically there are two different ways to satisfy the private sector demand. One is for asset prices to rise, and yields to fall, until the private sector decides that it is happy holding reserves that pay only .25%. The other way is for asset prices to stay the same and for the government to pay whatever interest is required to keep the private sector happy holding the incremental reserves.

In practice, if the Fed pegs the Fed Funds rate, then the latter road is the one chosen, more or less as a matter of course. Noting the downward pressure on the Fed Funds rate, the Fed sells a Treasury security, and its balance sheet contracts on both sides. On the other side of the transaction, the bank uses its low-yield reserves to buy the Treasury security. Galbraith likes to say it is just like moving money from a checking account to a saving account.

At any rate, the end result is that the private sector owns more Treasury debt than it did before, even though the Treasury has not issued any more debt. Where did it come from? The balance sheet analysis makes clear that the debt that the private sector now holds is debt that was formerly held by the Fed; it was the Fed’s asset-side counterpart to the liability-side deposit that the Treasury used to make the payment.

But what if the Treasury does not have a positive balance to begin with? No difference really. In this case, the Fed honoring the Treasury’s check just means granting an overdraft, which means that the Fed’s balance sheet expands on both sides by the same amount, additional reserves on the liability side and additional assets (the overdraft) on the asset side.

Personally I would be inclined to call this borrowing, but I can see why Galbraith might not; there is no increase in the Treasury’s indebtedness to the private sector, although there is an increase in the Fed’s liabilities. As in the previous case, the increase in the Treasury’s indebtedness to the private sector comes when the private sector trades in Fed debt (unwanted reserves) for Treasury debt that the Fed was already holding, checking to saving once again.

Note that so far I have not said anything at all about deficits. Everything I have said is entirely about the mechanics of payment and clearing, and it holds just as much when the government is running a budget surplus as when it is running a budget deficit.

The difference is that, when the government is running a surplus, it will be enjoying cash inflows more than sufficient to offset the cash outflows from its spending, so it will be regularly rebuilding its positive balances at the Fed or repaying its temporary overdraft. If the government is running a deficit, on the other hand, balances will not be rebuilt and overdrafts will not be repaid, so unless something else happens the Fed’s holding of Treasury bills will gradually be replaced by Treasury overdrafts.

This Treasury overdraft system is not so different from what happens in wartime, and here I think we get closer to the way Galbraith is thinking about the problem. (He is, as I have said, not as interested in the plumbing as I am.) He says: “There is never a shortfall of demand for Treasury bonds; Treasury auctions do not fail.” Quite so, but why so? Because the Fed stands in the wings as buyer of last resort, and as lender against Treasury collateral to banks and dealers who are buyers of first resort.

Instead of telling a story about the Fed pegging the Funds rate and accumulating Treasury overdrafts as the Treasury spends more than its tax receipts, we could tell a story about the Fed accumulating Treasury securities by setting the minimum price at auction. Same story really.

This is the money side of the deficit story. No alchemy, just banking.

Share your perspective