Miles Kimball and Brad DeLong Discuss Wallace Neutrality and Principles of Macroeconomics Textbooks
Let me provide a glossary for this discussion, in order of appearance terms that might be less familiar to some readers. But having written what is below, it is clear to me that it is more than a glossary. It has my most careful explanation yet in this blog of many key concepts. So it is worth reading even if you know what all of the terms mean.
IS-LM: The standard model of macroeconomics as taught in most undergraduate macroeconomics textbooks that many economists believe is a reasonable description of what causes business cycles despite the fact that it is hard work to get IS-LM-like results from economic models at the research frontier. IS stands for “combinations of output and interest rates where investment equals saving.” LM stands for “combinations of output and interest rates where liquidity demand for money L equals supply of money M."
grok: to understand. Here is the wikipedia definition of grok. Brad’s use of grok from Robert Heinlein’s book Stranger in a Strange Land references our shared interest in science fiction. Brad’s posting of this on his blog gives some evidence of his interest in science fiction.
ZLB: The "zero lower bound” on the nominal interest rate. That is, the fact that interest rates can’t go significantly below zero. (“Nominal” just means interest rates the way non-economists talk about them, as contrasted with the “real” or “inflation-adjusted” interest rates beloved of economists.) Why can’t the interest rate go much below zero? Because anyone can earn an interest rate of zero by burying cash in the back yard or hiding it in their mattress, so no one is willing to buy a bond that pays much below zero. People might buy a bond that pays a little below zero simply because it is more convenient than burying currency in the backyard, but they won’t accept an interest rate much below zero. The zero lower bound on the nominal interest rate is a problem for monetary policy since the Fed usually stimulates the economy by lowering short-term interest rates, which have recently been very close to the lower bound of zero.
AD: Aggregate demand, or the amount of spending people will do at a given price level. In models where prices adjust instantly to any change in the macroeconomic situation, aggregate demand is not very important; it only affects the price level and not much else. But in models that have sticky prices (any form of slow adjustment of the price level to new situations) aggregate demand governs the size of GDP in the period of time before prices adjust. Background: both Brad and I believe that in the real world, prices adjust only slowly to changes in the macroeconomic situation.
Keynesian model: In this context a component of IS-LM: what happens when not only the price level, but also interest rates are held constant. The reason that the zero lower bound (ZLB) is relevant here is that being pressed down against the floor of zero by the Fed is a reason why the interest rate might in fact stay constant in the face of a wide range of changes in the macroeconomic situation. In other contexts “the Keynesian model” might be a synonym for “IS-LM.
20-yr T < 0: Brad is saying that the inflation-adjusted (real) interest rate on 20-year Treasury bonds is now less than zero. In other words, if you buy 20-year Treasury bonds right now, what you earn will probably fall behind inflation. There is an ongoing discussion in the blogosphere about what the low interest rates on long-term Treasury bonds means for economic policy. See my post "What to Do When the World Desperately Wants to Lend Us Money.” In his tweet, Brad is simply pointing out how close to zero a wide range of interest rates paid by the U.S. government are right now. The implication is that the zero lower bound is starting to look relevant for long-term interest rates paid by the U.S. government as well as short-term interest rates paid by the U.S. government.
Wallace neutrality: In reference to either an economic model or to reality, Wallace neutrality is the statement that monetary policy can affect the economy in an important way if it the monetary policy action changes the path of current and future short-term interest rates paid by the government. Although Ben Bernanke does not believe that Wallace neutrality applies to the real world, the influence of Wallace neutrality thinking on the Fed is clear from the emphasis the Fed has put on telling the world what it is going to do with interest rates in the future. When the economy is at the zero lower bound, Wallace neutrality in the real world would mean that the only way the Fed could stimulate the economy now would be by promising to overheat the economy in the future, as I discuss in my post “Should the Fed Promise to Do the Wrong Thing in the Future to Have the Right Effect Now?” I have a series of other posts also discussing Wallace neutrality. In fact, essentially all of my posts listed under Monetary Policy in the June+ 2012 Table of Contents are about Wallace neutrality.
Noise traders: People who not only fail to make the best possible investment decisions for themselves but alsomake their mistakes in big enough herds that they move markets, changing the prices of stocks, bonds and other assets. (Note: Investors can make mistakes in herds for many reasons, not just what is called “herding behavior” in the economics literature.) The name “noise traders” suggest herds of investors who move markets by their actions, but herds of investors can also move markets by inaction in the fact of government actions. In models, Wallace neutrality arises from investors responding in highly intelligent ways to actions of the government that cancel out the effects of the government’s actions.
OLG: Having to do with “overlapping-generations” models, in which investors regularly die or otherwise leave the market and are replaced by other investors. Since I think investors stick around in asset markets for many years, and it requires fast turnover of investors to get much short-run action from overlapping generations models, this is not where I would turn to for a plausible model of departures from Wallace neutrality. Some sort of failure to optimize on the part of investors in herds as in the definition of “Noise traders” above is much more promising. The one other promising micro-foundation (story about what individual households and firms are doing) for departures from Wallace neutrality that I can think of is institutional structures that make firms behave differently than an optimizing investors would–for example, if assets rated AAA are required to meet certain regulatory or legal requirements.
Ricardian neutrality: See my post “Wallace Neutrality and Ricardian Neutrality” and the wikipedia article on “Ricardian Equivalence.” "Ricardian equivalence" is a synonym for “Ricardian neutrality." Economists use these phrases interchangeably and about equally often. The substance of Ricardian neutrality is actually not crucial to our discussion. We are looking at the status of Ricardian neutrality from a philosophy of science perspective and comparing it to the status of Wallace neutrality from a philosophy of science perspective.
Baseline modeling status: Both Ricardian neutrality and Wallace neutrality reflect how the simplest economics models behave within the category of "optimizing models.” Optimizing models are models that have in them very intelligent agents in them who do what is best for themselves. Both Brad and I agree that simple optimizing models should be the starting point for thinking about how the world works. Saying a model has baseline modeling status is saying that it should be the starting point for thinking about how the world works. The discussion is then about what might plausibly make things behave differently in the real world from that theoretical starting point. In other words, sometimes the starting point is a good place to stop, given the value of simplicity to human understanding; but sometimes the starting point is a bad place to stop because it misses an aspect of reality that is too important. In the case of balance sheet monetary policy, it is not enough to say that given what central banks have done in the past, Wallace neutrality is a good approximation, it is important to know whether Wallace neutrality is a good approximation over the full range of things that central banks could do. See my post “Future Heroes of Humanity and Heroes of Japan” for the kind of dramatic monetary policy move I have in mind when I say this.
Home bias: The well-documented tendency of investors to have a larger fraction of their assets in the stocks, bonds and other assets of their home country than would be warranted if they were trying to minimize the risk they face in order to get a certain level of expected returns. In other words, there is evidence that individual investors act as if they have a prejudice against foreign assets.
Effect of foreign assets purchases counts: I am saying that if the Fed bought Japanese or German bonds, for example, it would have an effect on exchange rates and the net exports of the U.S. and other countries, even when U.S. short-term interest rates are essentially at zero (the ZLB). Simple optimizing models imply that real (inflation-adjusted) exchange rates and net exports would be unaffected by the Fed buying Japanese or German bonds when short-term interest rates in the U.S. are already essentially zero (an example of Wallace neutrality), but I think many economists would predict that large purchases of Japanese or German bonds by the Fed would have an important effect. I have promised a future post on “International Finance: A Primer” explaining all of this better. (Based on sad experience, I have stopped promising specific dates for future posts, unless they actually exist in finished form in my queue.)
Long term T-bonds not powerful: Departures from Wallace neutrality should be especially small for long-term Treasury bonds for two reasons: (a) their interest rates are already fairly close to zero, so it is not that easy to bring them down further, and (b) there are many very smart traders in government bond markets with a lot of money behind them. What departures from Wallace neutrality there are for purchases of long-term government bonds probably arise more from institutional structures than from failures to optimize. In addition to likely having only a small departure from Wallace neutrality, buying long-term Treasury bonds is not ideal for balance sheet monetary policy for reasons forcefully argued by Larry Summers and discussed in my post “What to Do When the World Desperately Wants to Lend Us Money” There, within the Fed’s comfort zone, I recommend that the Fed buy mortgage backed securities. (I will discuss the issues with having the Fed buy foreign assets in a future post. Also see what Joseph Gagnon has to say about this.) Alternatively, the Fed could look for authority to buy a wider range of assets, either in interpretation of existing law or by asking Congress for authority to buy a wider range of assets, as it successfully asked for the authority to pay interest on reserves.
Everything the BOJ can do: My post “What to Do When the World Desperately Wants to Lend Us Money” discusses this.
Empower the Fed: Give the Fed legal authority to buy as wide a range of assets as the Bank of Japan is allowed to.