Here I offer some thoughts on Bloomberg’s account that the Fed has made inquiries with its dealers about the feasibility of a significant increase in the Fed’s reverse repo operations.
First, a little background. The traditional tool of monetary policy is an open market purchase, in which the Fed purchased U.S. Treasury securities that had previously been held by someone in the private sector. The Fed would pay for those securities by crediting deposits in an account that the selling bank had with the Federal Reserve. These reserve deposits of banks represent claims that the bank could use, if it wished, to withdraw green currency from the Federal Reserve. The volume of reserve deposits historically was extremely important in determining the interest rate at which banks would lend the deposits to one another overnight. The traditional understanding of monetary policy was that the Fed would use open market purchases to achieve its desired objectives for the overnight interest rate and the money supply.
If the Fed wished to implement a purely temporary increase in the volume of reserve deposits, historically the tool of choice was a repurchase agreement, in which the Fed would buy a particular security with a promise to return it at a specified future date. The purchase was again implemented by creation of reserve deposits, and when the security was returned, those deposits came back into the Fed.
The Fed began to see a new potential use for these repos after the initial banking difficulties in August 2007. Although repos were traditionally used as a device for temporarily injecting reserves, their structure amounts to a collateralized loan from the Fed to the counterparty. The Fed’s objective subsequent to August 2007 was to increase the volume of its lending and support the market for certain securities that it could accept as collateral for repos. Thus the Fed utilized an expansion of repurchase agreements as one of the initial tools for responding to the crisis, simply rolling them over to create a de facto expanded lending facility.
The graph below tracks the various assets held by the Federal Reserve since the beginning of 2007. The height of the graph represents the total asset holdings at the end of each week, with the colors indicating the contribution of each category. Repos are represented by the turquoise band. This traditionally had been small and highly variable, but grew significantly in the early phases of the financial crisis. Later the Fed came to use direct loans through its Term Auction Facility in preference to repos. Since January, the Fed has been directly buying up mortgage backed securities and agency debt in the way it used to purchase Treasury securities.
A separate question is what happens to all the reserve deposits created through this process. The Fed has never wanted to see the huge volume of reserves it created end up as currency held by the public, for fear this would be inflationary. It has relied on several devices to keep that from happening. One was use of the Treasury’s account with the Fed, another traditional feature of Fed operations that ballooned as it became adapted to new purposes. The Fed asked the Treasury to borrow funds that it simply left in deposit in its account with the Fed. These idle reserves held by the Treasury absorbed some of the vast increase in new reserves created by the Fed.
A more important tool was that the Fed started paying interest on reserves in October 2008, and by November had increased that rate to the target fed funds rate itself. This created a very strong incentive for banks to simply hold reserves idle at the end of each day rather than lend them out on the overnight fed funds market. In effect, by paying interest on reserves, the Fed is borrowing directly from banks and using the proceeds for the various asset expansions detailed above.
The graph below shows the Fed’s liabilities at the end of each week. The height of the graph is, by definition, exactly equal to the height of the previous graph at every date. The first graph tracks what assets the Fed acquired with its operations, while the second graph shows where the funds it created ended up. The surge in the Treasury account (in yellow) and excess reserves of member banks (green) explain why the huge expansion in the Fed’s balance sheet has not translated so far into a massive increase in the quantity of currency held by the public (blue).
The question under discussion at the moment is the extent to which the Fed could continue to rely on these two devices– Treasury borrowing on its behalf and banks’ willingness to simply hold the ballooning reserves– to contain the monetary consequences of its expansion. The traditional political gamesmanship over the debt ceiling could well induce the Treasury to want to discontinue its facilitation of the expansion of the Fed’s balance sheet, in which case the Fed must either reduce some of its lending or count on banks to hold even more excess reserves. Some in the Fed are assuming that they could always ensure the latter outcome, if needed, by raising the interest rate the Fed pays on reserves. But clearly the Fed has no desire at the moment to raise interest rates, so it’s difficult for me to imagine them taking that step any time soon.
Where else could the Fed get the funds? Fed Chairman Ben Bernanke described his contingency thinking last July:
the Federal Reserve could drain bank reserves and reduce the excess liquidity at other institutions by arranging large-scale reverse repurchase agreements with financial market participants, including banks, government-sponsored enterprises and other institutions. Reverse repurchase agreements involve the sale by the Fed of securities from its portfolio with an agreement to buy the securities back at a slightly higher price at a later date.
Just as the Fed converted the use of repos, which had historically been used on a small scale to temporarily add reserves, into a much larger operation with which it could lend broadly on a long-term basis, it is now contemplating using the reverse repo, which had historically been used on a small scale to temporarily drain reserves, into a much larger operation with which it could borrow broadly on a long-term basis. Thus we saw the following report from Bloomberg last week:
The Federal Reserve has started talks with bond dealers about withdrawing the unprecedented amount of cash injected into the financial system the last two years, according to people with knowledge of the discussions.
Central bank officials are discussing plans to use so-called reverse repurchase agreements to drain some of the $1 trillion they pumped into the economy, said the people, who declined to be identified because the talks are private. That’s where the Fed sells securities to its 18 primary dealers for a specific period, temporarily decreasing the amount of money available in the banking system.
There’s no sense that policy makers intend to withdraw funds anytime soon, said the people. The central bank’s challenge is to decrease the cash without stunting the economy’s recovery and before it sparks inflation. Fed Chairman Ben S. Bernanke said in a July Wall Street Journal opinion article that reverse repos are one tool to accomplish that goal without raising interest rates.
“One thing the Fed has to figure out is if they can launch pilot programs without spooking the market and creating the perception that they are about to tighten,” said Louis Crandall, chief economist at Wrightson ICAP LLC, a Jersey City, New Jersey-based research firm that specializes in government finance. “They are discussing things like accounting issues, and updating the governing documents to the volume of reverse repos the dealer community could absorb.”
Is this a feasible interim plan for handling the liability side without increasing either the money supply or interest rates? In a mechanical sense I believe the answer is yes. But the nature of inflationary pressures that we should be watching at the moment would arise from a depreciation of the dollar relative to other currencies and increase in the dollar price of internationally traded commodities. A modest move toward a weaker dollar and slightly higher inflation would be welcome. But the concern in my mind is whether a flight from the dollar could become more precipitous and destabilizing. It may not be the most likely scenario, but it is one for which I hope there has been some contingency planning.
And if the Treasury and the Fed think they could prevent that simply by borrowing even more without raising interest rates, they are mistaken.