Duped Again: The Fed, Assets And Losses

In Wednesday’s WSJ, journalists who follow the Fed reported on a new Fed working paper that provided detailed simulations of how the exit strategy that the FOMC laid out in June 2012 would affect the size of Federal Reserve remittances to the Treasury. The focus of the article was on how potential capital losses might affect the flow of so-called “profits” back to the Treasury from interest received on the Fed’s SOMA (System Open Market Account) portfolio.

We have written extensively in the past about why it is misleading to view Fed remittances as “profits” and why it is totally inappropriate for the Treasury to treat those remittances as income for budget purposes. Briefly, when the Fed is correctly viewed as part of the government rather than a private-sector entity, payments of interest to the Fed are nothing but an intragovernmental transfer of funds. The Fed extracts its operating costs, required dividend payments and contributions to capital, and remits the remainder back to Treasury. The net effect of these intragovernmental funds transfers is always a net payment by the Treasury to the Fed to cover its operations. There is no profit made nor is it appropriate to refer to the transfer from the Fed to the Treasury as “profit,” as is done in the WSJ article. Equally misleading is the practice of scoring Fed remittances as income for budget purposes, which is a bit of pure accounting gimmickry. The Fed working paper avoids the use of the word profit but does refer to the remittances as net income, which might mislead those not familiar with the true nature of the funds transfers into thinking that the Fed made a profit.

As for the simulations presented in the paper, the authors are meticulous both about describing the Federal Reserve’s balance sheet and detailing the assumptions they make to generate their alternative simulations. The relevance of the simulations is, however, questionable for two reasons.

First, their estimates of losses on the Fed’s portfolio as interest rates rise are based upon what are likely to be unrealistically optimistic assumptions about how market participants, and hence interest rates, will respond to the Fed’s exit. Specifically, the losses experienced on asset sales would be quite different if market participants were to respond immediately to a shift in FOMC policy and interest rates and if the term structure were to shift abruptly upward to what might be the new equilibrium, long-run expected interest rates. The simulations assume, for example, that it would take 7 or more years for the 10-year rate to move from its current level to about 4.9%. It is totally unrealistic to assume that investors will stand by and tolerate a serial decline in the value of their portfolios without running for the exit as quickly as possible. Second, tracking the flow of remittances back to the Treasury is important only to the Treasury and its budget scoring, and is unrelated to FOMC policy.

At Cumberland we have been focusing on the Fed’s balance sheet for more than four years. But rather than focusing on the flow of net interest payments and losses that might accrue from asset sales, we have instead used estimates of the durations of assets in the Fed’s portfolio and what parallel shifts in the yield curve might mean for the economic value of the Fed’s net worth. Those estimates currently indicate that an upward shift of only 33 basis points would be sufficient to push the market value of the Fed’s capital account to zero. In short, the Fed’s balance sheet would be worth substantially less and in a much shorter period of time than the simulations posed by Fed staff suggest. The realization of those losses is solely the discretion of the Fed.

The Fed working paper also provides a useful discussion of the accounting of assets and how losses would be handled. It outlines three valuation methods: face, amortized, and market value. Typically, the Fed reports asset values at face, and thus if interest rates were to increase and the Fed sold assets, it would have to book a loss. If losses exceeded interest payments from the Treasury, then remittances would cease; but instead of booking the additional losses against the Fed’s capital account, a deferred-asset account would be created. That accounting gimmick creates a claim on future Treasury interest payments that would be used to write down the value of the deferred-asset account when the payments are received. The working paper claims this meets GAAP accounting standards and parallels how corporations treat tax loss carry-forwards, for example. Ultimately, however, it is nothing but a way to avoid recognizing a loss against capital that would reveal that the Fed might be insolvent.

It can be argued that the question of whether the Fed is insolvent is meaningless as long as the country is solvent and functioning, but a broke Fed would still look bad in the eyes of the rest of the world.

BY Bob Eisenbeis