A new Fed-Treasury Accord?
A reporter asked me about a new Fed-Treasury Accord, referencing articles in Bloomberg and Forbes about Kevin Warsh’s comments. What do I think?
As a reminder, the first Fed-Treasury Accord in 1951 (see last post) reestablished Fed independence. From WWII to 1951, the Fed had committed to buy as many US Treasury bonds as necessary to keep the long term interest rate below 2.5%. It found itself buying a lot of bonds, which it felt was causing inflation. In the Accord, the Fed won the right to buy bonds only when it thought prudent, and to allow long-term interest rates to float upward.
That background is important. A lot of commentary accuses Warsh of becoming Treasury Secretary Bessent’s puppet at the Fed, and setting the Fed up to finance deficits without complaint. I think this is exactly the opposite of what is likely to transpire. Warsh has spent over a decade saying that the Fed needs to buy less government bonds, shrink its footprint on markets, let long-term interest rate go where they want to go, and do less overall. By doing less, in less political areas, the Fed can retain and strengthen its independence. The invocation of the Accord, in which the Fed won the right not to buy bonds tells you what’s up. Otherwise, it would be “repeal the accord.”
Of course the success of any new accord depends on what’s in it. But I see a few areas where a new accord would make a lot of sense and do a lot of good. Fiscal and monetary policy overlap. To the extent they can be separated, it’s a good idea to be clear who has responsibility for what. To the extent that they cannot be separated, it’s a good idea to have a clear channel of communication and cooperation.
The Fed and Treasury need to be clear that the Treasury is in charge of fiscal policy. Only by staying away from fiscal policy can the Fed hope to be independent. Independent agencies cannot be in charge of taxing and spending. That’s a job for politically accountable agencies, that face the voters periodically.
In practice what does that mean? First, Fed and Treasury need to agree who is in charge of the maturity structure of government debt. The Treasury balances issuing long term vs. short term debt. It’s just like choosing a fixed vs. floating mortgage when you buy a house. The long term debt is a little bit more expensive on average. But then, if interest rates rise, the payments stay the same and you’re not at risk of losing the house. Once the Treasury issues long-term debt, higher interest rates do not cause additional budget deficits, at least for the maturity of the debt.
In its various Quantitative Easing programs, the Fed bought long-term debt and gave out overnight debt (interest paying reserves) in return. That undoes some of the interest rate protection the Treasury bought. The wife went to the bank, and got the 30 year fixed. The husband went back to the bank and said no, no, swap that back to half floating. Interest rates went up, the wife thought “how wise I was.” The husband must admit sheepishly what happened and how their mortgage payments are going up. We can pretend only he lost money, but in the end they’re married. Likewise, the Fed and Treasury pretend to be separate but in the end, the Fed’s losses end up with the Treasury, and thus the taxpayer. Right now the US is paying a trillion dollars a year of interest costs on the debt. That’s not all the Fed’s fault. I’ve been arguing for a decade that the Treasury should have borrowed much longer to buy just this protection, and even issue perpetual debt. But the Fed did make that substantially worse by undoing a lot of the Treasury’s long term debt issues.
When I have asked in the past, Treasury officials said “we issue the debt, the Fed is independent.” The Fed of course denies any fiscal concerns. But that’s untenable. Someone needs to be in charge of the overall maturity structure and its risk-reward consequences for the taxpayer. In my view, this is a job for the Treasury. If the Fed wants to buy lots of Treasury debt in order to provide lots of bank reserves, then the Fed should buy only short-term bonds, or issue reserves via collateralized lending as the ECB does, without buying Treasury debt. If the Fed needs to buy long term debt temporarily for “market function” reasons, then it needs to swap out the interest rate risk with the Treasury, or exchange that promptly for short-term debt.
The treasury could help this a lot by issuing fixed value, floating rate, electronically transferable debt in the first place. (As well as perpetuities and fixed-for-floating swaps between the two.) If it’s right for the government overall to issue more short-term and overnight debt, why does the Treasury issue long term debt, then the Fed has to buy it up and issue short term debt in its place? The treasury doing so could free the Fed from a huge headache. Indeed, if the Treasury were to offer such debt that would free the Fed from more mountains of headaches, including the persistent demand for narrow banks, money market funds, segregated accounts, stable coins, and other private intermediaries who want to hold overnight debt, issue stable claims, offer transactions services, earn a small spread, and undercut the big banks’ deposit oligopoly.
In sum, Accord Item #1: Treasury is in charge of the maturity structure of US government debt.
Second, the Fed needs to get out of credit allocation. Buying mortgage backed securities to funnel low interest loans to homeowners rather than businesses is a political decision. Again, maybe ok if there is some short-run financial dislocation, but not for years and years. (I may disagree with a lot of the Fed’s financial policies, and its endless put option for markets, but that is the Fed’s job, and that’s our point here.) If that’s going to happen, then the Treasury should buy the mortgage backed securities directly. That’s already happening, with the recent $200b purchase. Again, in my view it’s a bad policy, but it is correctly made by the politically accountable branch of government, which is in charge of all the other subsidies and take from Peter to pay Paul policies of the government, and faces the taxpayer’s wrath.
Third, Fed and treasury need to prepare for the next crisis. This is the hard one, and goes right to the issues of 1951. In 2020, the Fed cried “dislocation” and promptly monetized $3 billion of new treasury debt, holding down long term interest rates just as if it were WWII again. That’s part (a small part in my view, but larger in may others’ view) of why we got a big inflation. Not funding deficits in a crisis is a much bigger deal than not funding deficits in every day affairs. A new crisis will come, will that happen again?
That’s just the tip of the iceberg. Then there is financial regulation, where the Fed steers lending here and there, protects big banks, bails out here and there in crises, and now offers explicit price support as for corporate bonds in 2020. These are political. The Fed usually offloads the explicit credit risk to the Treasury, but not the decisions.
In sum, the coverage of this event paints it as reversing the accord and making the Fed subservient to the Treasury. The Accord was the Fed’s singular achievement of being able to say “no.” It looks to me that Warsh commendably wants to say “no” more often, shrinking the Fed’s footprint, make the Treasury take the heat for its decisions, and thereby preserving Fed independence.
Your narrative implicitly treats government liabilities as the settlement asset whose valuation absorbs fiscal expectations when interest rates are pegged or otherwise non-informative. In that sense, inflation in 1951 looks less like a Phillips-curve or policy reaction story and more like a one-time revaluation of a fixed nominal settlement base under a change in expected future real claims. You don’t need a monetary policy rule to get the result, it falls straight out of a valuation identity once the settlement base is taken as exogenous. Under fixed supply and growing real activity, the default drift is deflation unless fiscal news intervenes (which seems consistent with the way output growth enters your 1951 interpretation)
If the Federal Reserve is paying more interest on reserves than it is receiving from its holdings of securities, then the deficit is recorded on the FR's books as "deferred expense" on the asset side of the balance sheet. The Treasury account at the Fed is not impacted by losses that the Fed is booking as a result of IOR > interest revenue, ceteris paribus.
At the close of each period, if the Fed's accumulated surplus is greater than a legislatively specified threshold value, then the Fed credits the Treasury's account by the difference between the accumulated surplus and the threshold value. If the Fed records a deficit in the period, the deficit is booked as an increase in the deferred expense account. The Treasury's account at the Fed is unaffected. Which is to say, that the Treasury does not absorb the Fed's losses from its IOR policy.
In terms of 2020-2021, the Fed's policy of buying municipal and state bonds addressed a market failure during that period. The monetized debt, if you will allow that term in this context, provided an orderly market for the issuance of munis and state bonds to avoid crippling the municipal and state funding during a unique event -- the pandemic shutdown that gripped the nation and prevented the flow of capital investment in the ordinary course of events. Liken it to a state of war in its impact at the state and municipal government operations.
As to Warsh, his views are all over the map. There is a level of incoherence in his views that defies definition of where his policy preferences will lie if and when his nomination for the chairmanship of the Federal Reserve System Board of Governors is confirmed by the Senate.