JACKSON HOLE, Wyo. - The Federal Reserve has "undercut" the hoped-for benefits of its large-scale asset purchases (LSAPs) and created market volatility with "imprecise" communication of its intentions regarding those purchases, an economist addressing the Kansas City Federal Reserve Bank's annual symposium charged Friday.
The Fed, which is now buying $85 billion a month of longer term Treasury and mortgage-backed securities, would do better to more clearly define the parameters for phasing out this "quantitative easing," argues Arvind Krishnamurthy, a Northwestern University professor.
"It is imperative that central banks outline a framework for the use of LSAPs," he declares. "Without such a framework, investors do not know the conditions under which LSAPs will occur or will be unwound, which undercut the efficacy of policy targeted at long-term asset values."
Contending that purchases of longer-term Treasuries have little benefit, he urges the Fed, when the time comes, to first cease buying Treasury bonds and then sell down its Treasury portfolio. The next step in their recommended "exit" sequence would be for the Fed to sell its higher-coupon MBS.
Finally, "the Fed should cease its purchases of current-coupon MBS as this tool is currently the most beneficial source of economic stimulus," writes Krishnamurthy, a fellow of the National Bureau of Economic Research. His co-author is fellow Northwestern professor Annette Vissing-Jorgensen.
However, "such exit steps cannot be implemented in a situation where investors face uncertainty over the Fed's LSAP policy rule," the economists warn. "Currently, with the Fed's discretion strategy, any exit step will be taken by investors as a signal of policy-maker preferences, which can then have wider consequences."
Fed Chair Ben Bernanke, among others, has outlined a broad, conditional strategy for unwinding QE3, but has stressed there is no "pre-set" or "prescheduled" Fed strategy for unwinding and ultimately ending asset purchases. The paper contends this open-ended and flexible or, as the paper terms it, "discretionary" approach has significant pitfalls.
Bernanke, who is not at Jackson Hole this year, said as recently as July 18 that the Fed could start reducing its bond buying "later this year" if the economy performs as projected and end purchases by mid-2014 if the unemployment rate drops to around 7% and inflation has moved back toward the Fed's 2% target.
This is far too vague and, hence, counterproductive, according to Krishnamurthy. He concedes the Fed's desire for "flexibility" in the face of economic uncertainty, but warns this comes at the cost of "policy uncertainty" that negatively impacts financial markets and in turn the economy.
"Without minimizing these (flexibility) considerations on the part of the Fed, it is also important to note that policy uncertainty has an effect on investors and through investors' behavior on the effect of LSAPs on the economy," he observes. "An investor in the MBS market, or a bank originating a loan today, needs to forecast the time and state dependence of the Fed's LSAP policy in deciding how to invest today."
"What is the dependence of the Fed's purchases on the unemployment rate?" he asks. "If the unemployment rate falls below 7%, will the Fed only cease its purchases or will it sell off its portfolio?"
Krishnamurthy warns, "Any uncertainty over the Fed's future policy will carry a risk premium today, causing today's MBS prices to fall."
"It is therefore theoretically possible that the Fed's non-committal policy undercuts the benefits of LSAP policies," he adds.
And there's "another drawback" to the Fed's "imprecise communication," the Northwestern professor maintains. "It restricts the Fed's ability to fine-tune an exit."
His preferred sequence of exit steps is selling Treasury bonds, then selling higher coupon MBS, then ceasing purchase of current-coupon MBS.
"But given investors' limited knowledge of the Fed's intentions over LSAP policy, any of these actions will be taken by investors as a signal regarding the Fed's preferences, which will then have widespread consequences," he says.
"By being imprecise in the state-dependence of LSAP policy, the Fed has left it to investors to form expectations over the future of Krishnamurthy charges. "In turn, this has led to market volatility."
As proof of his allegations, he cites the sharp market fluctuations that occurred when Bernanke made his initial "tapering" declaration following the June 19 Federal Open Market Committee meeting. He notes that the anticipated timing of hikes in the federal funds rate in the futures market was moved up by four months and that bond yields spiked.
Because of the Fed's "lack of clarity on the state-contingency of its LSAP policy," talk of reducing asset purchases affects market expectations about more conventional monetary tightening, Krishnamurthy says.
"Investors only understand that LSAPs are a tool to be used when the zero-lower-bound is binding," he says. "Thus when the Fed communicates that it plans on not using LSAPs, investors assume that the zero-lower-bound will not be binding and that rate hikes will follow."
The FOMC provides numerical "forward guidance" on the future path of the funds rate, reiterating July 31 that it will not raise the rate at least until the unemployment rate falls to 6.5% provided forecasted inflation does not exceed 2.5%.
Some Fed policymakers favor providing similar forward guidance on asset purchases and the Fed balance sheet, but no consensus has been reached on how to do that. Some, such as Philadelphia Fed President Charles Plosser, advocate a more rule-based approach, while others like Fed Gov. Jeremy Stein prefer more discretion and flexibility.
The paper also contests the Fed's belief that its purchases of Treasury bonds and MBS both have widespread effects on all interest rates.
"The portfolio balance channel of QE works largely through narrow channels that affect the prices of purchased assets, with spillovers depending on particulars of the assets and economic conditions," he argues. "It does not, as the Fed proposes, work through broad channels such as affecting the term premium on all long-term bonds."
"We find little evidence of a broad channel through which purchases of long duration assets, both MBS and long-term Treasury bonds, reduce a duration risk premium (term premium) on all long-term fixed income assets," write Krishnamurthy and his co-author Vissing-Jorgensen. "While the Fed has often alluded to this channel in discussing the beneficial effects of QE, the empirical evidence is more consistent with narrow channels where asset purchases principally affect the price of the asset that is purchased."
They take a particularly dim view of Treasury bond purchases, saying they have had "limited spillover effects for private sector bond yields and thus limited economic benefits.
At least since the first round of quantitative easing during the financial crisis, the dominant channel in which MBS purchases have worked is what Krishnamurthy calls "the scarcity channel."
The Fed has been buying a large proportion of newly issued or "current coupon bonds," and Krishnamurthy says this "has led to a scarcity premium on the current coupon MBS, driving spreads on MBS relative to Treasury yields below zero."
"The scarcity of the current coupon MBS generates incentives for banks to originate more loans and relieve the shortage of the current coupon MBS," he explains. "The lowering of secondary market MBS rates through both capital constraints and scarcity channels likely have had beneficial macroeconomic effects."
When the Fed stops buying or sells Treasuries, it "will have minimal effects," he contends. "While it will raise the rates on long-term Treasury bonds and affect financing conditions for the U.S. government, it will have limited negative consequences to private borrowers."
By contrast, a cessation of MBS purchases could have much larger impact, both through the "scarcity channel" and also through the "capital constraint channel."
In the first channel, "if markets anticipate that the Fed is likely to cease its purchases, yields on current coupon MBS will rise immediately and independently of what the Fed does with its held portfolio," he writes.
"Dynamically, if the Fed tapers its purchases of MBS, the scarcity premium will gradually diminish further," says Krishnamurthy. "After the Fed ceases its purchases, this diminished scarcity premium will fully disappear as new loans are originated."
Under the capital constraints channel, "news about either a cessation of purchases or sales of the Fed's portfolio will increase MBS yields. The effects will be immediate and persistent," he goes on. "Significantly, this channel implies spillovers to other mortgage securities with prepayment risk, such as jumbo mortgage rates, whereas the scarcity channel implies effects only on the agency mortgage rate."
Given the relatively much larger effect of ceasing MBS purchases, he argues that the Fed should delay doing so - a conclusion which the FOMC majority has already reached.
Market News International is a real-time global news service for fixed-income and foreign exchange market professionals. See www.marketnews.com.