Can the Fed learn to speak a non-interest rate language?

I was reading a new book by Tim Congdon and came across this interesting quotation, discussing the flaw at the heart of New Keynesian economics:

In the New Keynesian schema, it [the interest rate] became in effect the only policy instrument, the factotum of macroeconomics.

Why did we have to end up with the worst possible policy instrument?  The only instrument that has a zero lower bound.  We could have chosen the monetary base, or the trade-weighted exchange rate, or the price on NGDP futures, or the TIPS spread, or the price of zinc.  But no, we had to pick nominal interest rates.

We ended up with a steering mechanism that locks up just when you most need it to work.  Even worse, central banks have so fallen in love with the mechanism that they can’t seem to shift to a different target.  Instead we end up with never-ending attempts to manipulate interest rates, even when short term rates have hit zero.  Promise to hold rates at zero for X number of years.  Or attempts to lower longer term rates.  Or to reduce interest rate risk spreads.

These proposals have a slightly pathetic quality, because (as Nick Rowe reminds us in this recent post) a policy that is expected to be successful will actually raise nominal rates.  So you have the Fed announcing that the goal of QE2 was to lower long term rates, and then when they start rising the Fed announces that the policy must be working.  It’s a wonder the Fed still has any credibility.  How’s this for communication?

WASHINGTON — The Federal Reserve made a rare promise on Tuesday to hold short-term interest rates near zero through at least the middle of 2013, in a sign that it has all but written off the chances of an expansion strong enough to drive up wages and prices. . . .

By its action, the Fed is declaring that it, too, sees little prospect of rapid growth and little risk of inflation. Its hope is that the showman’s gesture will spur investment and risk-taking by convincing markets that the cost of borrowing will not rise for at least two years.

The Fed’s statement, with its mix of grim tidings and welcome aid, contributed to wild market oscillations as investors struggled to make sense of the economy and the path ahead.  (emphasis added)

Well that will certainly whip up those animal spirits!

The zero rate bound doesn’t occur for variables like the monetary base.  Some Keynesians argue that this doesn’t matter; open market purchases become ineffective when rates hit zero, as one is merely swapping one asset for another.  But that’s not true, as a permanent increase in the base is inflationary.  And if the Fed had already been using the base, it would have been able to continue signaling future policy intentions as if nothing had happened.  In contrast, once rates hit zero the Fed can’t signal anything with changes in interest rates, because it can’t change interest rates.

Of course the New Keynesians also insist that interest rates are the transmission mechanism.  Not so.  When there’s a big apple crop, NGDP in apple terms soars.  No need to invoke interest rates.  Ditto for a big crop of Federal Reserve Notes.  And the mechanism that causes nominal shocks to have real effects is sticky wages and prices, not interest rates.  When I point out that rates hardly budged during the most expansionary monetary policy in US history, Keynesians start talking about rates falling relative to their Wicksellian equilibrium value.  Yes, but that’s pretty much true by definition, and true for any price.  If the Wicksellian equilibrium zinc price is the one consistent with 2% inflation, then the Fed can boost inflation above 2% if and only if it can raise zinc prices above their Wicksellian equilibrium.

The next meeting will be a big test for the Fed.  I don’t expect miracles, but I’d hope for at least some sign that they understand there’s nothing more they can do to generate recovery by fiddling with interest rates.  They need to indicate that they are at least attempting to communicate in some other language.

Most people seem to assume nothing major will be done until the three hawks leave in January.  In fact, Fed stimulus would be more credible if they could get at least one of the three to vote for it.  It seems to me that Kocherlakota offers the best hope.  At times he seems to indicate that he’s aware of the unemployment problem, but doesn’t like the lack of a nominal anchor in open-ended promises, such as two years of near-zero interest rates.  He might be willing to support stimulus, as long as there is an explicit promise to maintain prices or NGDP along a particular trajectory.  If I’m right, it’s quite possible that the fate of 100,000s of unemployed people might depend on what he decides.

It’s no way to run monetary policy.  We should have an explicit 5% NGDP target, and let the market set the money supply and interest rates.  But you go into recession-fighting with the Fed you have, not the Fed you wish you had.


Tags:

 
 
 

40 Responses to “Can the Fed learn to speak a non-interest rate language?”

  1. Gravatar of John John
    30. October 2011 at 19:12

    Scott,

    You might (not) like Krugman’s post

    http://krugman.blogs.nytimes.com/2011/10/30/a-volcker-moment-indeed-slightly-wonkish/

    What do you think of his claim that NGDP is just a way for the Fed to distract people from what fighting unemployment entails? Krugman compared it to the early 80s when the Fed used monetarist language to distract people from the recession they’d have to create when they needed to fight inflation.

    Since it’s me leaving this post, I just wanna say there’s no way that creating a lot of money for whatever purpose will leave us with a healthier economy that what we would have otherwise had 5-10 years down the road. History and (good) theory show us that rapid increases in money have bad consequences down the road.

    The people and the government have to trust free markets to grind out three yards and a cloud of dust instead of continual monetary Hail Marys. Let’s see real deregulation get the government off people’s backs so they’re allowed to start making stuff again.

  2. Gravatar of Steve Steve
    30. October 2011 at 19:33

    When the Fed announced QE it should have gone with smaller scale but “conditionally permanent” bond purchases. In other words, they should have eased from 0% interest rate to $25B bond purchases/month, indefinitely, given current conditions. If that weren’t enough, at subsequent meetings they could have eased to $50B/month or $75B/month.

    The mistake the Fed made was treating QE as a “temporary liquidity” tool aimed at lowering rates, rather than a potent monetary stimulus tool.

  3. Gravatar of K K
    30. October 2011 at 19:45

    Scott: “But that’s not true, as a permanent increase in the base is inflationary.”

    OK. But as we’ve discussed before, the current tripling of the monetary base is *not* permanent unless they plan to wait until NGDP triples before they raise rates again. And even then, as a clear signal it’s *utterly useless* because it assumes that velocity doesn’t change. If you want to say that the monetary base won’t be allowed to drop to the level of regulatory/liquidity demand until NGDP triples, *all* you have to say is that we won’t raise rates until NGDP triples. And it’s a lot clearer since it doesn’t make pointless, meaningless assumptions about velocity. The *only* relevant communication that is approximated by the base is the condition under which we will raise rates. The path of the base between now and the moment when we raise rates is literally, utterly irrelevant. So let’s stop wasting time pointlessly moving the base around and scaring the inflationistas. Just articulate the conditions under which we will raise the short rate *and* the dynamics of the short rate thereafter contingent on the state of the economy. That includes the basically useless information contained in the size of the base at the zero bound, and vastly more actually useful information. Let’s please stop this pointless (or rather counterproductive) treasury bond QE.

  4. Gravatar of K K
    30. October 2011 at 20:03

    So the problem is that you can do nothing with the base that you can’t do way better with the path of rates. The ZLB problem has literally nothing to do with the framing of monetary policy in terms of rates. The problem is that the existence of non-interest bearing currency provides a lower bound on the rate of return of savings which is superior return to investment under sufficient deflation. That has nothing to to with framing.

  5. Gravatar of Andy Harless Andy Harless
    30. October 2011 at 20:34

    Suppose apple orchards were obliged by law to spend their entire proceeds on other fruit, which the apple growers were prohibited from subsequently reselling except to buy back apples. In that case, a big apple crop wouldn’t necessarily produce much increase in NGDP in terms of apples (AGDP?). Neither would a big crop of federal reserve notes, if the growers thereof were obliged to spend the proceeds on other high quality dollar-denominated financial assets, as they in fact are. It’s likely, if the orchards had magic trees that could grow an arbitrarily large number of apples at no cost, that they could influence AGDP substantially, but there’s no telling how many apples they might have to grow to hit any given AGDP target.

    It’s also not clear to me that a permanent increase in the monetary base is always inflationary. Suppose that the market expects permanent deflation, such that interest rates are expected to be zero forever. In that case, an increase in the monetary base is really just trading one kind of bond for another. The market would be satiated with liquidity services (and expect to remain so), so making more such services available (or promising to make them available in the future) would have no effect on anything.

    That case may be unrealistic, but suppose the Fed were using the monetary base as its instrument. In that case any increase in the monetary base relative to expectations could be perceived as a signal that the Fed expects weaker economic conditions than previously thought. (“Economic conditions over the coming years are likely to warrant this increase in the monetary base.”) Then increases in the monetary base could result in lower NGDP expectations and thus might be deflationary rather than inflationary. And if the process continues, the endgame is my previous paragraph, in which increases in the monetary base have no effect.

    AFAICT the monetary base is just a potential sunspot, a Tinkerbell. I do imagine people would be more inclined to believe in this well-defined Tinkerbell, rather than the vague Tinkerbell that the Fed is trying to sell now. But expectations are still arbitrary in principle if money and bonds can become perfect substitutes. Of course it would be different if the Fed were allowed to buy real assets.

  6. Gravatar of Bob Murphy Bob Murphy
    30. October 2011 at 20:36

    Scott wrote:

    When there’s a big apple crop, NGDP in apple terms soars.

    I think I get why Scott used that quick example when talking about interest rates not being a necessary transmission mechanism. But strictly speaking, the Official Scott Sumner would never endorse such a nonsense statement, right? I mean, it’s not like a fixed growth in NGDP would constrain apple production. A big apple crop would be consistent with even constant NGDP, because apple prices could drop.

  7. Gravatar of Bob Murphy Bob Murphy
    30. October 2011 at 20:39

    UPDATE: Now that I’m re-reading it, I see that Scott means something like calculating NGDP in apple units, rather than dollar units. OK, but I still don’t see how a big apple crop ==> “NGDP in apple terms soars.” There was a big USD crop in the last few years–far far far above historical harvests–and NGDP in dollar terms didn’t soar.

    I realize every blogger probably does this, and I only pick on Scott because his views are so strange to me, but it seems his blog is chock full of such antinomies.

  8. Gravatar of Matt Rognlie Matt Rognlie
    30. October 2011 at 21:48

    I have to agree with K on the zero lower bound: the constraint isn’t as obvious when you move away from the language of interest rates, but it’s still there. Zero interest rates are just a sign that base money has become a perfect substitute for other forms of liquidity (e.g. T-bills), and that the asset swaps underlying conventional monetary policy are no longer meaningful. You can make the monetary base as large as you want, but as long as you’re using it to buy T-bills, you’re not going to get anywhere.

    You might say that we can get traction through unconventional operations—expanding the monetary base is no longer quite so meaningless if we’re buying long-dated bonds instead. But I don’t see why this is a rejection of the “interest rates” worldview taken broadly: after all, unconventional policy can be explicitly framed as an attempt to decrease long-term interest rates.

    You might also say that since the monetary base can still move at the zero lower bound (or whatever else you’d like to call the case of perfect substitutability), it’s better than interest rates, even if all this movement in the monetary base doesn’t have any direct effect—after all, it can still be used to communicate policy intentions. Maybe. But it’s just cheap talk, and a remarkably crude, ineffective form at that. There can never be any presumption that a base money injection by the Fed will be “permanent”. The demand for base money is simply too idiosyncratic and interest-inelastic, meaning that any kind of explicit commitment runs the risk of sparking serious instability down the road; and when commitment is impossible, changing the stock of base money today really is useless at the zero lower bound.

    Of course, you know all this, which is why you support an NGDP target rather than a base money target. I’m just pointing out that there is nothing uniquely awful about interest rates that is not also problematic (in fact, surely more so) for base money and various other indicators.

    The real question we should be asking about monetary “languages” is how they allow us to shape expectations. On this score, interest rates aren’t the worst “language”. You can, after all, express conditional commitments like those proposed by Charles Evans. This circumvents Nick Rowe’s objection that a sufficiently effective policy should raise equilibrium interest rates in the medium term. A conditional commitment explicitly acknowledges this, and is in effect a device for ruling out the “bad” equilibrium, by saying that no matter what people believe about the future they should consume and invest today: if the future is “bad” in the real GDP sense, then the path of interest rates will be extremely low, and if the future is “good”, then the case for consumption and investment is clear even if interest rates are a little higher.

    Now, I’m open to the possibility that an NGDP target is a more elegant and effective way of achieving this same end. But I think the debate isn’t so much “interest rates” vs. “NGDP” as it is “rules that fail to shape expectations in a sensible way” vs. “rules that do”.

    By the way, my guess is that even if the Fed implemented an NGDP level target, the nominal interest rate would still be used (and widely monitored) as the Fed’s intermediate target, in much the same way that central banks today with strict inflation targets nevertheless publish and debate the policy rate.

  9. Gravatar of JKH JKH
    31. October 2011 at 00:31

    MR,

    “Zero interest rates are just a sign that base money has become a perfect substitute for other forms of liquidity (e.g. T-bills), and that the asset swaps underlying conventional monetary policy are no longer meaningful.”

    But at positive interest rates, the central bank must pay interest on excess reserves to support the policy rate target (because, as you say, the demand for excess reserves is inelastic). And excess reserves are the relevant part of the monetary base, for purposes of exogenous influence on the size of the monetary base.

    So zero interest rates don’t differentiate that relevant part of the base from other forms of liquidity, in a way that is fundamentally different from the case where interest rates are positive. In both cases, short term liquidity is a near perfect match, and longer term liquidity differs by the yield curve.

    Similarly, commitments that increases in base money will be permanent are non-differentiating to the degree that interest rates must be paid on excess reserves when the policy rate turns back positive. A meaningful commitment to a permanent increase in base money suggests a commitment to permanently zero interest rates as well.

  10. Gravatar of Anonymous Anonymous
    31. October 2011 at 00:42

    Yikes…I think you’re placing too much hope in Kocherlakota tipping the scales here. Kocherlakota has made clear that he would only act as he did late last year if things did not get any better if one looked at the broad macroeconomic metrics. By his estimation, the fact that inflation is closer to target and growth is 2.5% qualifies as improvement (never mind the output gap). I have no hope that he will lead the charge to something closer to NGDP level targeting. He’s more reasonable than Fisher and Plosser, who seem all too keen to forget that it’s a DUAL mandate. Is that really saying much?
    On the other hand, if we do see movement from the Fed, it will likely come from the doves. I never understood why they didn’t go all out with aggression once the hawks played their last card. If Bernanke and the doves are willing to tolerate a three-vote dissent, they might as well make the shift in communications and regime change sooner rather than later.

  11. Gravatar of Nick Rowe Nick Rowe
    31. October 2011 at 02:06

    K and Matt are still thinking in interest rate language: “Scott has increased the monetary base, but he doesn’t really mean that. We know he’s going to take those words back again in future. But a commitment to keep interest rates too low for too long, now that’s the language we understand, because it’s meaningful.”

    Let me try an analogy. Take an inflation targeting central bank, in normal times (outside the ZLB). Suppose that when the Bank raises the interest rate, people interpret that to mean: “we are raising the interest rate 25bps for one week, then will lower it 50bp for one year after that”. If inflation were below target, that central bank would have to *raise* the interest rate, in order to loosen monetary policy.

    This reminds me of the old “grue/bleen paradox”.

    http://en.wikipedia.org/wiki/Grue_and_bleen

    http://plato.stanford.edu/entries/relativism/supplement4.html

    When it comes to talking about the monetary base, Scott is talking blue/green language, and K and Matt are interpreting him in grue/bleen language.

    When you change the language, the laws of inductive inference change too. Which matters, because it’s expectations of future monetary policy that matter.

    Another way of saying the same thing: what does it mean for the central bank to “do nothing” in future, as opposed to “doing something”?

    Roosevelt could raise the price of gold. He didn’t have to promise not to lower it back down again in future. People automatically assumed he would “do nothing” (with the price of gold) in future. That’s because in the olden days, they all spoke the gold price language.

  12. Gravatar of Nick Rowe Nick Rowe
    31. October 2011 at 02:12

    Hmm. My comment is stuck, maybe because I put 2 links in.

    Let me try without the links:

    K and Matt are still thinking in interest rate language: “Scott has increased the monetary base, but he doesn’t really mean that. We know he’s going to take those words back again in future. But a commitment to keep interest rates too low for too long, now that’s the language we understand, because it’s meaningful.”

    Let me try an analogy. Take an inflation targeting central bank, in normal times (outside the ZLB). Suppose that when the Bank raises the interest rate, people interpret that to mean: “we are raising the interest rate 25bps for one week, then will lower it 50bp for one year after that”. If inflation were below target, that central bank would have to *raise* the interest rate, in order to loosen monetary policy.

    This reminds me of the old “grue/bleen paradox”.

    Google “grue bleen” and read the wikipedia and Stanford entries.

    When it comes to talking about the monetary base, Scott is talking blue/green language, and K and Matt are interpreting him in grue/bleen language.

    When you change the language, the laws of inductive inference change too. Which matters, because it’s expectations of future monetary policy that matter.

    Another way of saying the same thing: what does it mean for the central bank to “do nothing” in future, as opposed to “doing something”?

    Roosevelt could raise the price of gold. He didn’t have to promise not to lower it back down again in future. People automatically assumed he would “do nothing” (with the price of gold) in future. That’s because in the olden days, they all spoke the gold price language.

  13. Gravatar of Nick Rowe Nick Rowe
    31. October 2011 at 02:20

    Another way to make the same point: if you change the strategy space in a game, it is well-understood that the Nash equilibrium will change. For example, the Cournot duopoly equilibrium is different from the Bertrand duopoly equilibrium, even if the two games have exactly the same demand functions and cost functions. You get a different price and quantity Nash equilibrium depending on whether the 2 firms are speaking Q language or P language.

  14. Gravatar of Bill Woolsey Bill Woolsey
    31. October 2011 at 03:37

    The Market Monetarist approach is for the central bank to purchase assets until nominal GDP is at target.

    Never in our thinking does it say, “only purchase T-bills.”

    Generally, there is no restriction on what assets are purchased, and one can imagine that the central bank would like to purchase short and safe assets. It can roll them over if needed and when it wants to sell them, there is little chance of capital loss and they will be paid off rapidly anyway.

    But, _it is automatic_! If the Fed has bought every T-bill in existence, and still nominal GDP is below target, _the rule_ is to purchase other assets that aren’t quite so attractive on that basis.

    If a central bank has legal restrictions on what it can buy, like the Fed does, then when it has bought everything it is legally entitled to buy, then there is a problem. But the Fed is about $8 trillion away from that point.

    By the way, even before the crisis the Fed’s balance sheet included government notes and bonds. It wasn’t solely T-bills.

    New Keynesians have a model that shows that a central bank can use open market operations in T-bills, ignore the quantity of money and focus solely on short term interest rates, and by creating expectations of the path of short term interest rates and inflation, keep inflation stable and close output gaps. Cool.

    But it is a mistake to assume that this is all that a central bank can do.

    I do agree that making commitments to permanent increases in the monetary base is irrelevant from a normative perspective.

  15. Gravatar of Dan Kervick Dan Kervick
    31. October 2011 at 04:07

    Isn’t it the case that there will always be some practical bound, no matter what kind of target is set?

    Suppose we were in a period of stagflation, with zero real GDP growth and inflation of 6% or 7%. Wouldn’t NGDP level targeting require tightening?

    The existence of a bound is inherent in the fact that most of the real economy is outside the reach of Fed, so there is always going to be some place where there is little more the Fed can do.

  16. Gravatar of dwb dwb
    31. October 2011 at 04:20

    The current Fed situation reminds me of the Monty Python skit from Monty Python and the holy grail (where the King fights the Black Knight). No! I can fight you with one arm! Come back I’ll bite your ankles!

  17. Gravatar of K K
    31. October 2011 at 05:11

    Matt: “By the way, my guess is that even if the Fed implemented an NGDP level target, the nominal interest rate would still be used (and widely monitored) as the Fed’s intermediate target…”

    I always took NGDP targeting to mean that we are switching the target from inflation to NGDP (level), but the policy instrument to achieve that is, under normal conditions, still the short rate. If we had NGDP futures then I guess the Fed could manipulate those and ignore the short rate. Or they could manipulate both. But surely they need an instrument that they can actively manipulate or they will only be yet another loud-mouthed think tank. The point of communications is to state what they’ll do (instruments) contingent on variables that define the state of the economy (targets).

    Nick: The trouble with blue/green is that it’s full of meaningless statements and almost meaningless statements. And it doesn’t have equivalents of some very meaningful statements in grue/bleen. I.e. in blue/green you can describe the path of the base while the short rate is at zero (meaningless). And you can describe the path after the short rate is no longer at zero (pretty well meaningless unless you know exactly how velocity will respond). When you speak grue/bleen you don’t make meaningless statements, and there are ways to say pretty well everything you can meaningfully say about monetary policy.

  18. Gravatar of ssumner ssumner
    31. October 2011 at 05:30

    Everyone, Thanks for all the great comments.

    John, You said;

    “Since it’s me leaving this post, I just wanna say there’s no way that creating a lot of money for whatever purpose will leave us with a healthier economy that what we would have otherwise had 5-10 years down the road. History and (good) theory show us that rapid increases in money have bad consequences down the road.”

    If the Fed does what I want the monetary base will be much smaller than if they continue the current policy, indeed less that half as big. You should welcome that result.

    Japanese history shows that big increases in the base don’t create inflation in the long run if accompanied by zero percent rates. What history did you have in mind? And how is it relevant to the US?

    Steve, Maybe, but $25 billion a month is way too low–at least $100 billion/month would have been appropriate.

    K, In a perfect world you might be right, but in the world we live in the Fed can’t communicate in the language of interest rates, but has some limited ability to communicate in the language of QE. The markets seems to have been able to read QE, but didn’t seem to get much out of the attempts to manipulate the yield curve. I think that’s partly because the Fed doesn’t understand that easier money means higher long term rates, and since the Fed is confused, the markets are also confused.

    Instead of promising to hold rates near zero until X happens, why not promise to push zinc prices higher until X happens?

    Andy, Yes, but we know that apple prices do change a lot, hence pears can’t be close substitutes. And we know that whenever rates are positive NGDP does change a lot when there are autonomous changes in cash. Hence T-bills aren’t close substitutes when deciding what to hold in your wallets (or even for bank reserves.)

    Andy, You said;

    “It’s also not clear to me that a permanent increase in the monetary base is always inflationary. Suppose that the market expects permanent deflation, such that interest rates are expected to be zero forever.”

    Yes, but then T-bonds and cash are identical, and hence the monetary base becomes T-bonds plus cash plus reserves. T-bonds are just high denomination cash. All debt is monetized. Now “monetary policy” must be unconventional (the purchase of foreign debt for instance), as there is no domestic debt to buy. It’s all base money.

    That would be a great “problem” to have, just ask the Italians.

    You said;

    “That case may be unrealistic, but suppose the Fed were using the monetary base as its instrument. In that case any increase in the monetary base relative to expectations could be perceived as a signal that the Fed expects weaker economic conditions than previously thought. (“Economic conditions over the coming years are likely to warrant this increase in the monetary base.”) Then increases in the monetary base could result in lower NGDP expectations and thus might be deflationary rather than inflationary. And if the process continues, the endgame is my previous paragraph, in which increases in the monetary base have no effect.”

    Yes, but can’t you say the same about rate cuts during normal times, they also might represent Fed fears of falling AD, and hence lead to falls in AD.

    I should say that this post is primarily intended to be pragmatic. I’m not trying to rule out all exotic equilibria on theoretical grounds (although there are ways to do that, as with a conditional promise to convert base money into a real asset under specified conditions) but rather I’m trying to explain why markets liked QE2 more than Op. Twist.

    In practice, the interest rate is far harder to communicate with than the monetary base. I should add that I obviously don’t favor targeting the base, but rather NGDP futures. But if we had to have a suboptimal policy instrument, I’d greatly prefer the base to interest rates. It’s way more robust.

    Bob, I meant that NGDP soars if the base is greatly increased when interest rates are positive, or expected to be positive.

    Matt, you said;

    “I have to agree with K on the zero lower bound: the constraint isn’t as obvious when you move away from the language of interest rates, but it’s still there. Zero interest rates are just a sign that base money has become a perfect substitute for other forms of liquidity (e.g. T-bills), and that the asset swaps underlying conventional monetary policy are no longer meaningful. You can make the monetary base as large as you want, but as long as you’re using it to buy T-bills, you’re not going to get anywhere.”

    I think this is wrong, because it ignores the way in which monetary policy communicates. Indeed it proves too much, as the same would be true when rates aren’t zero. Suppose I double the base today and announce that tomorrow I’ll cut the base in half. There will be no rise in prices, because the expected future base hasn’t risen. So when the Fed does OMOs in ordinary times there must be some expectation that it is signaling future policy, otherwise it wouldn’t “work.” The problem with interest rates is that once they hit zero, they can no longer be used as signals. The base still can. That’s why markets loved QE2.

    You said;

    “Maybe. But it’s just cheap talk, and a remarkably crude, ineffective form at that. There can never be any presumption that a base money injection by the Fed will be “permanent”. The demand for base money is simply too idiosyncratic and interest-inelastic, meaning that any kind of explicit commitment runs the risk of sparking serious instability down the road; and when commitment is impossible, changing the stock of base money today really is useless at the zero lower bound.”

    Now replace “base money injection” with “interest rate change,” and the exact same applies.

    I don’t agree that the Evans-type commitment is just as good, but that may be partly because the Fed can still buy non-traditional assets. An Evans commitment doesn’t necessarily prevent a Japanese scenario of 20 years of deflation (although I’d guess that in practice it probably would.) On the other hand NGDP futures targeting does prevent that scenario.

    Regarding your last point, I favor having the Fed use NGDP futures contracts as intermediate targets. In that case interest rates would be “just a price,” sort of how in the US exchange rates migrated from being the price of money in the 1930s, to just a price today.

    As I said above, this was meant to be a pragmatic post. I don’t deny that it is theoretically possible to communicate in the language of rates, but it seems that’s it’s really hard, because the Fed obviously doesn’t know how. Their promise to hold rates at zero for two years is exactly what that SHOULD NOT be doing, as it doesn’t communicate expansionary intentions at all.

    One final point about a MB instrument. Something like the McCallum rule for the base would have been massively more effective than the old gold standard. Now admittedly that’s a low bar, and we can do even better. But I’m not convinced it would have been a horrible policy. In contrast, a stable MB would have been almost as bad as the gold standard, and would have been a horrible policy.

    Anonymous, When I read Kocheralokota he seems far less of an inflation hawk than the others. Rather he wants an explicit rule to anchor the price level. So lets give it to him. In exchange, he might accept a couple years of “catch-up.”

    Thanks for those good analogies Nick. If only I had known when I was back in grad school that I needed game theory to understand macro.

    Bill, You said;

    “The Market Monetarist approach is for the central bank to purchase assets until nominal GDP is at target.
    Never in our thinking does it say, “only purchase T-bills.””

    Yes, and here’s why that’s important. When we move on to buying long term bonds, the meaning is clear—higher base equals a signal of monetary easing. But what do lower long term interest rates signal? I can’t tell.

    With a base instrument, we don’t give a damn whether buying long term bonds raises rates or lowers rates.

    I often hear that the Fed can only buy T-securities. The problem I have with that is that they actually buy all sorts of other stuff. My second question is this; can’t the Treasury give the Fed permission to buy forex? And isn’t the Treasury controlled by stimulus-loving Obama administration?

    Dan, Yes, if you had a 5% NGDP target it would require tightening. But so what?

    dwb, Nice analogy.

  19. Gravatar of ssumner ssumner
    31. October 2011 at 05:37

    K, I agree that if the Fed has some sort of explicit rule, like 2% inflation of 5% NGDP, level targeting, it might be possible to communicate with any price, including a path for the price of T-bills, but also the price of zinc.

    But what interests me more is which forms of communication work best with the current ambiguous and sub-optimal Fed policy. I still think QE communicates more effectively than fiddling with the yield curve, and my hunch is that equity markets agree with me.

    It’s easier to communicate by changing something right now, than by issuing promises to adhere to some sort of future path of policy, conditional on all sorts of factors. Maybe in theory there’s no difference, but in practice it seems there is.

  20. Gravatar of Dan Kervick Dan Kervick
    31. October 2011 at 05:58

    Dan, Yes, if you had a 5% NGDP target it would require tightening. But so what?

    Scott, the so what is that in an economy in recession the target would require reducing production, incomes and employment further, for the sake of bringing inflation down a few points. Now maybe Romer, Krugman and your other new NK liberal allies are cool with this commitment to future re-deployment of the Volckerian hammer as well. If so, I’d like to hear them acknowledge it at the outset.

  21. Gravatar of johnleemk johnleemk
    31. October 2011 at 06:25

    Dan,

    It would be very surprising and very contrary to textbook New Keynesian macro if Krugman or Romer endorse demand-side stimulus when inflation is high, even if output is stagnant or falling. When output is not growing in spite of high inflation, it’s most likely that the crisis is primarily supply-side, and demand stimulus, fiscal or monetary, would do nothing to fix that. (In other words, in a situation like this, John the Austrian would actually be right.)

    If inflation is galloping and output is stagnant, textbook macro still endorses cutting back on inflation. That’s the most fundamental first-cut analysis. There may be complicating factors indicating that the crisis is truly demand-side, not supply-side, but without any information besides the inflation rate and output trend, the recommendation of textbook macro is to tighten in that case.

  22. Gravatar of bill woolsey bill woolsey
    31. October 2011 at 07:28

    I think the Fed always implements monetary policy by changing the quantity of the base or else changing the demand for the base by changing the interest rate on reserves.

    This doesn’t mean that the Fed has to have some scheme for the base to be at some particular level at various future dates.

    And I have no problem with the Fed looking at interest rates and changing the base according to changes in interest rates.

    But I don’t think a complicated rule about what should happen to some particular short term interest rate is approprate either.

    Most importantly, when past experience regarding the relationship between interest rates and nominal GDP break down, including, for example, when expanding the monetary base would usually lower interest rates rapidly and then later expand nominal GDP, but the interest rate is zero, then there should be a seamless shift away from any concern about those interest rates.

    We will keep them at zero, and then, at some future time, when inflation or unemployment change, then according to the following functional relationship the interest rate will change a certain number of basis points–

    Well, it doesn’t seem to work very well, does it?

    The policy should be that future base money and interest rates will be at whatever level necessary for future expected nominal GDP to be on target.

    Now, no one cares about what base money will be other than its implications for nominal GDP (or the price level.) Interest rates, on the other hand, are prices that matter. Why should the Fed be trying to tell people what these prices will be? I am sure that some people, expecially commercial and investment banks appreciate having the Fed help stablize these prices. Just like farmers have supported price stabilization schemes too. But why shouldn’t market participants come up with estimates of what interest rates will be needed to keep future nominal GDP on target?

    I think the central banks respond to the special interest in finanical markets to stabilize short term rates. And so, their __real_ rule is keep short nominal interest rates stable subject to a contraint.

    Early on, that contraint was gold convertibility, and so, they would have to raise interest rates sometimes to offset a loss of gold reserves, usually through a trade deficit. Using regulation (like tariffs) to prevent the need to raise rates would be a possiblity.

    Of course, there really is no problem with building up gold reserves and so lowering rates when there is trade surplus is less necessary.

    After years, we now have an alternative contraint. Failing to raise nominal interest rates when expected inflation rises can create a hyperinflationary disaster. So, keep interest rates steady unless higher inflation expectations threaten disaster.

    Failing to lower rates can also lead to a disaster if expected deflation occurs. And so, keep interest rates steady unless a deflationary disaster threatens.

    And so, the major money center banks and investment banks can, as best possible, keep funding costs stable, and so borrow short and make commercial loans, international loans, underwrite securites, or whatever.

    Given this political logic, coming up with an economic rationale for why targeting short term interest rates is something that central bankers want. New Keynesians provide what they want.

    Compare to old monetarists and market monetarists. Rather than helping the central bankers get what they want, both are critics. The cental bank should just let short term interest rates be determined by market forces. For the old monetarists, keeping M2 on a stable growth path would keep nominal GDP on a more or less stable growth path and the inflation rate low and steady. Market monetarists are much better.

  23. Gravatar of Benjamin Cole Benjamin Cole
    31. October 2011 at 07:29

    Excellent post by Scott Sumner. Again.

  24. Gravatar of Dan Kervick Dan Kervick
    31. October 2011 at 07:36

    Johnlmeek, so in the textbooks they teach that if prices are rising in a recession due to supply side factors, as opposed to monetary factors, the smart play is to prioritize prices over production, and use monetary policy to reduce production further, so as to bring prices down?

  25. Gravatar of Doc Merlin Doc Merlin
    31. October 2011 at 07:41

    @ Dan Kervick

    It really doesn’t matter what the text books teach. People will learn to arbitrage away any positive effect of fed policy within a couple decades of the fed developing a new policy tool, then that tool will become useless. I don’t doubt that NGDP targeting will work, for a while, but then we will learn to arbitrage the benefits away.

  26. Gravatar of “Ben Volcker” and the monetary transmission mechanism « The Market Monetarist “Ben Volcker” and the monetary transmission mechanism « The Market Monetarist
    31. October 2011 at 07:44

    [...] Scott Sumner has a interesting comment on central banking “language” and “interest rates”. Share this:TwitterFacebookLike this:LikeBe the first to like this post. 11 Comments by Lars [...]

  27. Gravatar of Mike C Mike C
    31. October 2011 at 08:02

    I love this quote: “We ended up with a steering mechanism that locks up just when you most need it to work.”

    EVERYTHING Bernanke has been telling us (“hey we have more tools, i swear!”) seems to imply that this is the case.

  28. Gravatar of bill woolsey bill woolsey
    31. October 2011 at 09:37

    I find it puzzling that there could be any question about it.

    It really doesn’t matter exactly what the inflation rate or real growth might be.

    With growth path targeting, then the growth rate over the next year is the target for next year’s nominal GDP divided by its current value, all divided by the current value.

    The level of real growth and inflation right now don’t come into it.

    If things are going “well” and you are on the growth path now, then the growth rate of nominal GDP is rate of growth of the path.

    Inflation and real output growth then depend on market forces.

    So, 7% inflation and zero real growth. That does imply 7% nominal GDP growth. If you were on the target before, and the growth rate of the target is 5%, then you are now above target, and so the growth rate for nominal GDP would be less, approximately 3%.

    On the other hand, if you had been about 2% below target, then this 7% nominal growth returns to target, (again with 5% growth on target) and so, growth is 5% going forward.

    What nominal GDP growth path targeting does is make demand unit elastic on the whole. If prices rise 4% or 5% above trend, the real expenditures will fall 4% or 5% below trend, as will production. Given the usual assumptions, that would be 6% or 7% inflation and -1% or -2% real growth.

    With nominal GDP targeting, the prices of imported goods have no direct impact. Ignoring the domestic oil industry, a higher price for imported oil just makes the prices of imports rise. Now, if this causes a lower dollar exchange rate, so that foreigners demand more domestic products and there is a reduction in the demand for other imports, then this will indirectly raise the demand for domestic output (inculding any domestic value added to oil, like gasoline.) This will require monetary restraint to control demand.

    But to the degree that the oil sellers accumulate dollar assets (and so, prevent the dollar from falling,) then this net capital inflow requires lower interest rates and so, perhaps a monetary expansion. The demand for domestic products grows 5%. Nominal incomes grow 5%. And, of course, higher import prices mean real income grows less.

    Anyway, a more than trend incease in the price of some domestically produced good, will raise the price level a smaller amount (in proportion to its GDP share,) raising the price level and requiring real expenditure to rise less than trend in proportion. Bad harvest in Iowa raises the price of corn and reduces its quantity. If nominal expenditure on corn rises faster than trend, then nominal expenditure in the rest of the economy must grow less than trend. And so, that is contractionary in the rest of the economy. (It is really good, of course, for corn growers in Nebraska.)

  29. Gravatar of Link roundup « Negative Interest Link roundup « Negative Interest
    31. October 2011 at 09:40

    [...] Scott Sumner: Can the Fed learn to speak a non-interest rate language? Share this:TwitterFacebookLike this:LikeBe the first to like this [...]

  30. Gravatar of Anonymous Anonymous
    31. October 2011 at 10:40

    Kocherlakota is indeed more reasonable than the other two and I have a lot more respect for his views. Yet I still have little faith that he’ll accept some catch-up inflation unless conditions worsen on both fronts, unemployment and inflation.

    What do you think about the fact that Bernanke has been willing to tolerate a three-vote dissent for marginal policies such as MEP? While the Fed isn’t factional, Bernanke has no incentive to curry-favor any of the hawks now. If he’s comfortable with 7-3, he might as well do what all 7 basically view as necessary. If I’m channeling my revealed preference abilities, I’d say Dudley might be the member stopping Bernanke from getting more aggressive, given the exit strategy for the Open Market Desk.

  31. Gravatar of flow5 flow5
    31. October 2011 at 12:47

    Keynes’ liquidity preference curve is a false doctrine. But the monetary base is not a base for the expansion of new money & credit. What’s profound about targeting nominal gDp is that it eliminates interest rate targeting & replaces it with monetarist guidelines. And that’s obviously a surprise to everyone.

    E.g., Greenspan didn’t start “easing” on January 3, 2000, when the FFR was first lowered by 1/2, to 6%. Greenspan didn’t change from a “tight” monetary policy, to an “easier” monetary policy, until after 11 reductions in the FFR, ending just before the reduction on November 6, 2002 @ 1 & 1/4% (approximately coinciding with the bottom in equity prices).

    Greenspan then wildly reversed his “tight” money policy (at that point Greenspan was well behind the employment curve), and reverted to a very “easy” monetary policy — for 41 consecutive months (i.e., despite 17 raises in the FFR, -every single rate increase was “behind the curve”). I.e., Greenspan NEVER tightened monetary policy.

    I.e., Greenspan was responsible for both high employment (June 2003, @ 6.3%), and high inflation (rampant real-estate speculation, followed by widespread commodity speculation – peaking in July 2008 – Greenspan’s inflection point).

    And I haven’t talked about Bernanke (who should be fired). Greenspan’s failures are just one reason to target nominal gDp. It’s not the best target, but the only one dummies can hit.

  32. Gravatar of ssumner ssumner
    31. October 2011 at 13:15

    Dan, You do realize that the Fed can’t control the long run rate of real GDP growth, and that at best they can smooth out the business cycle. And that NGDP targeting is more countercyclical than inflation targeting. You seem to think the Fed should target RGDP–but that makes prices indeterminate.

    I’m quite sure Krugman and Romer understand all this.

    Bill, Yes, the Fed should not be setting interest rates.

    Thanks Ben and Mike.

    Anonymous, You may be right about Kocherlakota. Dudley used to work for G-S, let’s hope they set him straight. :)

    flow5, That’s right, rater cuts don’t equal easing.

  33. Gravatar of Max Max
    31. October 2011 at 13:30

    “The Market Monetarist approach is for the central bank to purchase assets until nominal GDP is at target.”

    If a central bank blindly buys stuff (other than treasuries), it can lose money. On the other hand, if it acts prudently, then it might not be able to hit its target.

    Central banks are supposed to be prudent, for good reasons.

  34. Gravatar of marcus nunes marcus nunes
    31. October 2011 at 13:49

    A story where interest rates “don´t make an appearance”:
    http://thefaintofheart.wordpress.com/2011/09/15/an-alternative-script-%E2%80%93-one-where-interest-rates-don%C2%B4t-make-an-%E2%80%9Cappearance%E2%80%9D/

  35. Gravatar of bill woolsey bill woolsey
    31. October 2011 at 13:56

    I don’t think central banks should try to maximize profit. If hitting the target for nominal GDP requires that the central bank lose money so be it.

  36. Gravatar of Becky Hargrove Becky Hargrove
    31. October 2011 at 16:45

    Someone said recently that you will need – at times – to get your points across simply and quickly. The first two paragraphs of this post have some potential sound bites that people will be able to carry with them and remember afterwards.

  37. Gravatar of Expectations Are Fundamental « Uneasy Money Expectations Are Fundamental « Uneasy Money
    31. October 2011 at 17:57

    [...] I read this morning on Scott Sumner’s blog the following quotation from a news item in the New York [...]

  38. Gravatar of K K
    31. October 2011 at 19:01

    Scott:

    “the Fed doesn’t understand that easier money means higher long term rates, and since the Fed is confused, the markets are also confused.”

    Yup. There is probably no other way to describe the thinking of a committee than “confused”. The Fed needs a term dictator like many other central banks.

    “Instead of promising to hold rates near zero until X happens, why not promise to push zinc prices higher until X happens?”

    Well, I don’t suppose you’d have to push the price up *that* high before the economy would find a substitute for all purposes and further buying would be economically irrelevant. But in general, I think you are correct that purchasing of high beta assets would eventually, inevitably work. But as I’ve said before, to me it smacks of industrial policy and I find policy subsidies to corporate interest extremely distasteful.

    I think there may be a way to circumvent the liquidity trap without the purchase capital assets though. if the Fed directly purchased NGDP futures I think that alone could control independently of sign of the expected change of NGDP. But it needs to have unlimited ability to use NGDP futures as an instrument, not as a target. Otherwise they will still have a major potential credibility problem at the ZLB.

    The way I see it, if they bought NGDP futures up to levels implying eg 10% growth (yup that means *nominal* growth as I have banished all references to real variables on this blog) then there’d be a large class of relative value players ready to invest in real assets and hedge their investments against 10% growth by selling the futures in delta. If 10% doesn’t give enough stimulus then 15% would surely do the trick. The point is, there’s no limit to how hard the Fed could push on the economy via arbitrage opportunities. And then the capital markets from the biggest hedge funds to a student worried about investing $200K in a college degree, will make the decisions about the allocation of capital but with insurance of a given level of growth guaranteed by the Fed.

    I think it could be improved even more with a developed NGDP futures options market. Let’s say your project needs at least 4% growth but would not benefit from significantly higher growth. If the Fed would sell cheap 4% NGDP puts, that’s all the hedge you could reasonably wish for in order to undertake the project.

    But I don’t believe that there’s anything transformative about “targeting”. If you want to transcend the liquidity trap you need the Fed to have an NGDP *instrument*.

  39. Gravatar of Britmouse Britmouse
    1. November 2011 at 03:14

    I think Lars Svensson also dislikes the obsession with interest rates, so some NKs are better than others! He said:

    [The] repo rate should not be a target variable for monetary policy, but merely an instrument without any inherent value.

    http://people.su.se/~leosven/papers/101124e.pdf

  40. Gravatar of ssumner ssumner
    4. November 2011 at 18:42

    Max, It can hit its target by buying Treasuries.

    Marcus, Thanks for the link.

    Bill, I agree.

    K, You said;

    “Well, I don’t suppose you’d have to push the price up *that* high before the economy would find a substitute for all purposes and further buying would be economically irrelevant.”

    You misunderstood me. I wasn’t contemplating the Fed buy zinc, but rather raise zinc prices by buying T-bonds. We saw how QE raised commodity prices.

    I proposed a NGDP futures instrument in a paper I published back in 1989–so I’m certainly on board for that approach.

    Britmouse, But he still focuses too much on rates as an intermediate target, in my view.

Leave a Reply