Exiting from Monetary Stimulus: A Better Plan for the Fed

Policy Innovation Memorandum No. 30

by Benn Steil, Sr. Fellow & Dir. of International Economics, Council on Foreign Relations

The U.S. Federal Reserve’s monetary stimulus efforts have an undesirable side effect that needs to be managed with great care: the Fed has amassed a huge stock of mortgage-backed securities (MBS) that it will eventually want to liquidate without damaging the nascent housing recovery. What is needed to accomplish this is a relatively simple but innovative scheme whereby the Fed can, in one transaction, transform its MBS holdings into an equivalent amount of U.S. Treasury securities. Such an arrangement would allow the Fed to use conventional means of raising interest rates when inflation threatens without the worrisome economic and political consequences of selling mortgage-backed securities.

The Problem

In a sustained, extraordinary policy undertaking to counter the enduring economic headwinds of the 2008 financial crisis, the Fed has pumped trillions of dollars of liquidity into the banking system over the past four and a half years. It has accomplished this by buying, with newly conjured dollars, a historically unprecedented amount and variety of securities. In the process, its balance sheet has ballooned from $900 billion to $3.1 trillion, and it is expected to expand further, to about $4 trillion, later this year.

At some point as the economy normalizes, most likely around 2015 according to the Fed’s current view, the Fed will wish to begin tightening monetary policy in order to prevent the exceptional level of liquidity in the banking system from feeding into inflation. It would normally do this by selling securities from its oversized balance sheet.

The big challenge the Fed will face in carrying this out, however, will be managing the economic and political impact of liquidating the huge stock of MBS it has amassed — currently amounting to nearly $1 trillion, and expected to reach $1.5 trillion by the end of the year. Doing so has the potential to drive up mortgage rates, depress house and real estate prices, and trigger intervention from policymakers reacting to aggrieved constituents.

Federal Reserve Board chairman Ben Bernanke appears to be very aware of such risks, and he has suggested that the Fed might, at least initially, use alternative means of tightening monetary conditions—not involving securities sales—when the time comes. Yet these means are likely to be less effective, as well as equally controversial.

Getting From Here to Normal

If the amount of MBS outstanding remains roughly flat over the coming year, and the Fed winds up swallowing $1.5 trillion of the total stock, it will own a massive 30 percent of the market. This is not the sort of market in which the Fed will wish to be selling securities when it judges the time right to tighten monetary conditions. Such sales have the potential to generate large and sudden falls in the price of MBS, making it costlier for banks to issue mortgages and therefore driving up their rates—rates that the Fed has worked hard to push down over the past four years.

Bernanke has been anxious to assure the markets that he has other tools in his chest for tightening monetary policy—two in particular. The first is to entice banks to move a portion of their excess monetary reserves held at the Fed to term deposits (similar to CDs), which would lock up that money for a fixed period. The second is to use “reverse repos,” in which the Fed continuously borrows money from the banks, using its Treasury securities as collateral.

Though repos are collateralized, and term deposits are not, the two tools are functionally identical. Both reduce the amount of funds banks have available to loan out, which serves the Fed’s purpose of restraining credit growth and inflation. But both also have a hidden catch: the Fed will lose control over interest rates.

If the Fed is unwilling to sell its mortgage securities in order to tighten policy, and if it is instead determined to drain a specific quantity of bank reserves through term deposits or the like, it will have to pay whatever rate the market demands. Simply put, the Fed must choose between managing the level of reserves and managing rates. It cannot do both.

Yet the Fed is unlikely to be willing to allow rates to rise as far and as fast as the market may demand in a term deposit auction. Indeed, the European Central Bank (ECB), which has carried out many such auctions since 2010, places caps on the rates it pays. In consequence, seven of its auctions have failed, meaning that the ECB failed to withdraw euros from the market despite its public pledges to do so. The Fed will lose credibility at a crucial time if it does the same.

Therefore, a conventional Fed approach to monetary tightening—selling securities from its balance sheet — is the better one, at least provided that the composition of its balance sheet does not force it to sell illiquid and sector-specific securities like MBS. The question then becomes how the Fed can normalize the composition of its balance sheet—that is, fill it completely with Treasury securities—before it needs to tighten monetary policy.

The Plan

The Fed should sell its MBS portfolio to the Treasury at face value in exchange for an actuarially equivalent amount of Treasury securities, newly issued for the purpose of facilitating the swap. The maturity of these new Treasury securities could be set either to match the expected maturity of the Fed’s MBS portfolio or to allow them to roll off at the same pace as the Fed expects it will wish to contract its balance sheet (each approach has its own technical merits). The transaction would be neutral for the size of the Fed’s balance sheet; only the composition would change.

There is a clear precedent for the Treasury doing this. The Housing and Economic Recovery Act of 2008 (HERA) gave the Treasury the authority to purchase MBS guaranteed by Fannie Mae and Freddie Mac. The Treasury started buying MBS in October 2008, stopping in December 2009 when the face value of its holdings reached $192 billion. The effect of these transactions was to transfer riskier securities from the private sector to the public sector.

In the case of the proposed Fed-Treasury securities swap, however, there is no such transfer of risk from the private sector to the public — one arm of government is merely swapping securities with another. The overall financial risk to the government as a whole remains unchanged.

Benefits Versus Costs

The benefits to the government, however, in terms of carrying out monetary policy effectively would be considerable. Instead of having to sell MBS on the market in order to soak up dollars and restrain credit growth and inflation, the Fed would be able to sell Treasury securities. The market for Treasury securities is the deepest and most liquid in the world, meaning that disruption to the market would be minimal while the impact on mortgage rates and house prices would be more moderate and less sudden than if the Fed was selling MBS. In consequence, there will also be less alarm in Congress over the possible negative economic side effects of the Fed tightening monetary policy.

Concern may be raised in Congress about the risk to the Treasury of absorbing the Fed’s MBS portfolio. Of course, the value of the securities can rise or fall as mortgage default rates and other factors change. The U.S. taxpayer will bear that risk for as long as any arm of the U.S. government holds them. Yet it is worth noting that if the Treasury buys them from the Fed at face value, it will immediately acquire a portfolio with unrealized gains of roughly $53 billion, according to the Fed’s most recently published estimate. And if the Treasury simply holds the MBS until they mature, it will avoid the losses the Fed would suffer if it were obliged to sell those same securities in an environment in which interest rates were rising (and bond prices, therefore, falling).

In short, the Fed swapping its mortgage-backed securities with the Treasury in return for Treasury securities is better for U.S. taxpayers. It preserves the value of the securities. More important, it is better for sustaining the recovery of the U.S. economy, as it affords the Fed an exit strategy from its long period of extraordinary monetary interventions that will be less disruptive to the mortgage and housing markets.

Valid concern may still be raised that the Treasury, in buying the Fed’s MBS portfolio, would be abetting the Fed in escaping responsibility for managing the necessary aftermath of its extraordinary market interventions. Bernanke has acknowledged that “the hurdle for using nontraditional policies should be higher than for traditional policies” because of the attendant costs, such as exposing the Fed to abnormally high financial risks. Yet if such costs can be off-loaded onto another arm of government, they may be insufficiently accounted for. HERA offers a template for future crisis interventions that addresses this concern—a template in which only the Treasury, and not the Fed, intervenes in markets other than those for its own securities.