An earlier post on how the Fed changes the size of its balance sheet prompted several questions from readers about the Federal Reserve’s accounting of asset purchases and the payment of principal by the Treasury on Treasury securities owned by the Fed. In this post, we provide a more detailed explanation of the accounting rules that govern these transactions.
Overview of the Federal Reserve System
It is useful to start with an overview of the Federal Reserve System. The System is decentralized and consists of three key entities: (1) The Board of Governors of the Federal Reserve System (the Fed’s central governing board); (2) twelve Federal Reserve Banks (through which the System operates); and (3) the Federal Open Market Committee (the FOMC, which sets U.S. monetary policy and consists of members of the Board of Governors and some Reserve Bank presidents).
As the operating arms of the System, the Reserve Banks hold the System’s assets, liabilities, and capital. Accordingly, the Federal Reserve’s balance sheet actually represents the accounts and results of operations of the Reserve Banks—their combined and individual statements of condition are reported weekly in tables 5 and 6 in the H.4.1 statistical release and in annual audited financial statements of the Federal Reserve System.
Federal Reserve Purchases of Treasury Securities
Reserve Banks are authorized to purchase Treasury securities in the open market (the secondary market) under the terms of Section 14 of the Federal Reserve Act and at the direction of the FOMC. The FOMC has selected the Federal Reserve Bank of New York to execute open market operations. The securities acquired are then allocated across the Reserve Banks.
When a Reserve Bank purchases a Treasury security, it purchases an asset, typically from a bank or broker-dealer. It credits reserves (a liability of the Fed) to the reserve account of the seller (or the seller’s bank). The bank or broker-dealer may sell its own securities or may act as an agent on behalf of a client. As a holder of a Treasury security, a Reserve Bank has no special rights relative to other Treasury-holding entities. The Treasury security that the Fed has purchased is not “paid in full.” It remains an asset on the Fed’s balance sheet until the security matures or is redeemed by the Treasury in accordance with the terms of issuance.
The Relationship between the Fed and the Treasury
As the nation’s central bank, the Fed plays a number of important public policy roles, and monetary policy does indeed have fiscal implications. In trying to understand the effects of the Fed’s actions on public finances and debt, it can be convenient, in some cases, to think about a consolidated public sector balance sheet that sums together the respective assets and liabilities held by the Fed and the federal government. Nevertheless, as noted in an answer to a comment on our earlier post, the Federal Reserve Banks are independent entities, with their own balance sheets, separate from that of the Treasury. There is no authority to consolidate Reserve Bank and Treasury balance sheets.
One might wonder whether the Fed could coordinate with the Treasury and agree to an accounting offset, so that every time the Fed buys a Treasury security, the security is considered “paid in full.” In fact, neither the Board of Governors nor any Reserve Bank is authorized under the Federal Reserve Act to fund the repayment or retirement of a Treasury security; only the Treasury can do so, and it can do so only under the terms under which it issued the security.
Even if it could coordinate with the Treasury and agree to an accounting offset every time it buys a Treasury security, the Fed may prefer to hold on to these assets for future use. Indeed, just like any other entity with Treasury holdings, a Reserve Bank may resell or lend Treasury securities—as the New York Fed does—subject to the Section 14 open market limitation and FOMC directions. But again, a Reserve Bank does not have the authority to fund the payment or retirement of Treasury securities, which are obligations of the U.S. government.
What Happens When a Treasury Security on the Fed’s Balance Sheet Matures?
As discussed in the previous post, when a Treasury security on the Fed’s balance sheet comes due, the Fed can roll over the security or let it mature. A rollover is used if the Fed wants to keep the amount of Treasury securities it holds constant. Reserve Banks do not have the authority to purchase securities directly from the Treasury Department, but they may exchange maturing securities for newly issued securities (so that, on net, no new money is paid to the Treasury). This is how things are done under the current rollover program.
What if the Fed lets the security mature? As is standard in accounting, the payment of principal is not recorded as income (although the payment of interest is). When a Treasury security is paid off, the payment from the Treasury’s General Account at the Fed simply offsets the value of the maturing Treasury security. If the Treasury security was held by an individual, then a new asset—cash—would replace the old asset—the Treasury security—on the asset side of the individual’s balance sheet. This replacement of one asset by another does not represent income.
In the case of the Fed, things are slightly different because a Fed liability, balances in the Treasury General Account, is used to pay off the Treasury security; as a result, both the asset and the liability sides of the Fed’s balance sheet decrease. But, for the same reason stated above, this does not constitute income.
To Sum Up
The way the Fed accounts for changes to its balance sheet depends in part on the structure of the Federal Reserve System and the fact that the Fed is not part of the Treasury. In this post, we have tried to provide more detail on the accounting rules governing transactions that can affect the size of the Fed’s balance sheet.
The views expressed in this post are those of the authors and do not necessarily reflect the position of the Federal Reserve Bank of New York or the Federal Reserve System. Any errors or omissions are the responsibility of the authors.
Deborah Leonard is a vice president in the Federal Reserve Bank of New York’s Markets Group.
Antoine Martin is a senior vice president in the Bank’s Research and Statistics Group.
Jennifer Wolgemuth is a vice president in the Bank’s Legal Group.
How to cite this blog post:
Deborah Leonard, Antoine Martin, and Jennifer Wolgemuth, “A Closer Look at the Fed’s Balance Sheet Accounting,” Federal Reserve Bank of New York Liberty Street Economics (blog), August 4, 2017, http://libertystreeteconomics.newyorkfed.org/2017/08/a-closer-look-at-the-feds-balance-sheet-accounting.html.
DO: In the pre-crisis monetary policy implementation regime, the Fed achieved its target federal funds rate by actively managing the supply of reserves in the banking system. To do this, the New York Fed’s trading desk bought and sold Treasury securities in the open market to bring the supply of reserve balances in line with the estimated quantity of reserves demanded at the FOMC’s target rate. In calibrating these operations, it was necessary for the Desk to account for the net effects of movements in so-called “autonomous factors”—factors that affect the supply of reserves but are largely outside the direct influence of Fed policymakers or the Desk’s operations. Movements in and out of the Treasury general account (TGA) was one autonomous factor among many. (Changes in demand for U.S. dollar paper currency was another.) So while the actions of the Treasury could, combined with other factors, contribute to the need for an open market operation, the Federal Reserve did not react to specific Treasury actions. In the current policy implementation regime, where the Fed has an abundant level of reserves and achieves interest rate control through IOER and an overnight reverse repo facility, changes in autonomous factors do not affect achievement of the FOMC’s fed funds target range. More generally, the TGA balance does not impact (i) the FOMC’s setting of the monetary policy target, for example, a certain level of reserves or a particular interest rate or rate range, (ii) the FOMC’s monetary policy regime, for example, a corridor or a floor-type system, or (iii) the objectives of monetary policy, which are set by the Federal Reserve Act as “maximum employment, stable prices, and moderate long-term interest rates.”
Hi Liberty Street. Nice blog post. But I would really like to see one where the interactions between the Fed’s operations and the Treasury’s operations are explored. As I understand it, particularly in the pre-IOER world, the interactions could be quite important. Since spending by the Treasury adds to bank reserves, doesn’t this necessitate that the Fed take action to drain those reserves in order to maintain the Fed Funds rate? Similarly, if the Treasury sells a security to the public, this reduces the amount of bank reserves, so doesn’t that necessitate the Fed buy a security in order to offset the drain to ensure that the Fed Funds rate wouldn’t rise? In that case, isn’t the net result that the Treasury selling securities causes the Fed, at least temporarily, to move that same quantity of securities onto its balance sheet? I would really love to see you folks write a blog on these issues. Thanks!
JF: Thank you for your comment. We understood the first part of your comment to be about the Federal Reserve’s remittances to the Treasury and why payment of principal does not represent income. As noted in the answer to another comment, the Fed is required to remit its earnings to the Treasury after providing for the Fed’s cost of operations, payment of dividends, and reservation of any amount necessary to maintain Reserve Bank capital at no more than $10 billion (per Section 7(a) of the Federal Reserve Act: https://www.federalreserve.gov/aboutthefed/section7.htm). That principal repayments don’t contribute to income is standard accounting practice. Think about it this way: If Alex lends $100 to Chris and Chris pays the money back, the inflow of cash that Alex received isn’t “income.” It made him whole again, but did not increase his overall wealth. However, if Alex charged Chris interest on the loan, any interest payments Chris made would represent income, to compensate Alex for his service and the time value of his money. We understood the second part of your comment to be related to the relationship between the Fed and the Treasury. As noted in our post, the Federal Reserve Banks are independent of the Treasury This arrangement reflects a broad consensus among policymakers and academics worldwide regarding the importance of separating monetary policy decisions from political influence (see, for example, https://www.federalreserve.gov/faqs/why-is-it-important-to-separate-federal-reserve-monetary-policy-decisions-from-political-influence.htm). Relatedly, there is a separation between the objectives and parties responsible for the development and execution of fiscal policy and those of monetary policy (see, for example, https://www.federalreserve.gov/faqs/money_12855.htm). The accounting treatment of the Fed’s maturing Treasury securities doesn’t affect this separation. Indeed, since repayment of principal is not income, it has no bearing on the income of the Fed, the income of the Treasury, or the government’s overall need to borrow over time.
Doesnt the Treasury borrow money so it has enough in the TGA to offset the principal amounts being redeemed by the Fed? Why would Treasury agree to this type of offset? Wouldn’t it be smarter for Treasury to withhold their agreement until the Fed agrees that principal payemts are remitted too, and use offset accounting so that they do not need to borrow in the first place (in close technical terms, the transaction is stated as payment to the Fed in redemption by the Treasury who then remits the excess monies right back to Treasury, allowing both parties to offset without having to borrow in the first place – this means Treasury redeems, not the Fed, while the Fed just follows the surplus remittance rule). What legal authority is cited as the basis for the Fed to assert that principal redemption amounts are not subject to the rule of remittance? And if these sums must be remitted, what are the coordination agreements needed with Treasury (both sides need to agee). Please look at the optics here. In the first place, the bonds were bought by the Fed with made-up money, so to speak, not with money ‘earned’ in the economy. But somehow this novel position allows the Fed to force the public via its government to borrow anew to eliminate a position that was made up in the first place. I am sorry, this just does not seem right. I want Fed leadership. In my view it should tell the Trump Administration and Congress to raise taxes in order to remove excess reserves from the economy, if that is its goal, at which point it can use the offset approach they are thinking of now (where they are of the opinion that cash in the TGA is erased by redemption mechanisms). It just does not seem right to compel the public to borrow anew unless the Fed is agreeing that fiscal affairs should avoid taxation and use borrowing instead at a time when there is no economic exigency; again, this just does not seem right. Thanks for considering these questions and points.
AC: Thank you for your comment. The Fed is required to remit its earnings to the U.S. Treasury after providing for the cost of operations, payment of dividends, and reservation of any amount necessary to maintain Reserve Bank capital at no more than $10 billion. Interest income earned from coupons paid on the Fed’s holdings of Treasury securities is part of the Fed’s earnings. Like the Fed, most central banks around the world are subject to rules that specify who has a claim on their profit. In many cases, central bank profits are remitted to the local Treasury.
Is it true that coupons paid on Treasury securities held by Federal Reserve banks (as well as any profits earned on them) trickle back down to the US Treasury? This essentially constitutes free financing to an extent, even if they still have to come up with the principal payment?