Last week, Ben Bernanke tested the waters for “tapering,” or cutting back on the rate at which he carries out new asset purchases, and everything is going poorly. As James Bullard, the president of the Federal Reserve Bank of St. Louis, argued in discussing his dovish dissent with Wonkblog, “This was tighter policy. It’s all about tighter policy. You can communicate it one way or another way, but the markets are saying that they’re pulling up the probability we’re going to withdraw from the QE program sooner than they expected, and that’s having a big influence.”
But if you really believe in the expectations channel of monetary policy, can this even matter? Let’s use this to revisit an obscure monetary beef from fall 2012.
Cardiff Garcia had a recent post discussing the fragile alliance between fiscalists and monetarists at the zero lower bound. But one angle he missed was the disagreements between monetarists, or more generally those who believe that the Federal Reserve has a lot of “ammo” at the zero lower bound, over what really matters and how.
For instance, David Becksworth writes, “What is puzzling to me is how anyone could look at the outcome of this experiment and claim the Fed’s large scale asset programs (LSAPs) are not helpful.” But one of the most important and influential supporters of expansionary monetary policy, the one who probably helped put the Federal Reserve on its bold course in late 2012, thinks exactly this. And that person is the economist Michael Woodford.
To recap, the Fed took two major steps in 2012. First, it used a communication strategy to say that it would keep interest rates low until certain economic states were hit, such as unemployment hitting 6.5 percent or inflation hitting 2.5 percent. This was the Evans Rule, which used what is called the expectations channel. Second, the Fed started purchasing $85 billion a month in assets until this goal was hit. This was QE3, which used what is called the portfolio channel.
In his major September 2012 paper, Woodford argued that the latter step, the $85 billion in purchases every month, doesn’t even matter, because “‘portfolio-balance effects’ do not exist in a modern, general-equilibrium theory of asset prices.” At best, such QE-related purchases “can be helpful as ways of changing expectations about future policy — essentially, as a type of signalling that can usefully supplement purely verbal forms of forward guidance.” (He even calls the idea that purchases matter “1950s-vintage,” which is as cutting as you can get as a macroeconomist.)
To put it a different way, the Fed’s use of the portfolio channel only matters to the extent that the Fed isn’t being clear in its written statements about future interest rate policy and other means of setting expectations.
Woodford specifically called out research by the Peterson Institute for International Economics’ (and friend of the blog) Joseph Gagnon. Contra Woodford, Gagnon et al concluded in their research, “[QE] purchases led to economically meaningful and long-lasting reductions in longer-term interest rates on a range of securities [that] reflect lower risk premiums, including term premiums, rather than lower expectations of future short-term interest rates.” (Woodford thinks that expectations of future short-term interest rates are the only thing at play here.)
Woodford’s analysis of this research immediately came under attack in the blogosphere. James Hamilton at Econobrowser noted that “Gagnon, et. al.’s finding has also been confirmed by a number of other researchers using very different data sets and methods.” Gagnon himself responded here, defending the research and noting that Woodford’s theoretical assumptions “are violated in clear and obvious ways in the real world.“ (If you are interested in the nitty-gritty of the research agenda, it is worth following the links.)
For our purposes, what can the taper explain for us? Remember, in Woodford’s world of strict expectations, QE purchases don’t matter. Since the purchases don’t matter, moving future purchases up or down at the margins, keeping the expected future path of short-term interest rates constant, shouldn’t matter either. Raising the $85 billion to $100 billion wouldn’t help, and lowering it to $70 billion wouldn’t hurt, unless Bernanke also moved the expectations of future policy.
The taper was a test case for this theory. Bernanke meant to keep expectations of future short-term interest rates the same (there was no change to when interest rates would rise) while reducing the flow of QE purchases. But from our first readings, it has been accepted by the market as a major tightening of policy. This strikes me as a major victory for Gagnon and a loss for the strongest versions of the expectations channel.
Of course, the taper could be a signal that Bernanke has lost his coalition or is otherwise going soft on expansionary policy. If that’s the case, then according to the stronger version of the expectations theory, QE3 should never have been started, because it adds no value and is just another thing that could go wrong. Bernanke should just have focused on crafting a more articulate press release instead. This doesn’t seem the right lesson when a body of research argues purchases are making a difference.
An objective bystander would say that if the taper is being read as tightening even though future expectations language is the same, it means that we should be throwing everything we have at the problem because everything is in play. That includes fiscal policy. As Woodford writes, “[t]he most obvious source of a boost to current aggregate demand that would not depend solely on expectational channels is ﬁscal stimulus.” We should be expanding, rather than contracting, the portfolio channel, while also avoiding the sequester and extending the payroll tax cut. Arguments, like Woodford’s, about the supremacy of any one approach tend to get knocked down by all the other concerns.