Calomiris on Fed policy

Charles Calomiris suggests that the Fed will probably need to raise rates this year:

Furthermore, because the fall in energy prices is a positive supply shock, nominal GDP growth will not be reduced by the decline in energy prices; indeed, it is likely to accelerate going forward. That means real GDP growth will accelerate alongside nominal GDP growth. From the perspective of forward-looking inflation targeting, the Fed rightly understands that it needs to maintain its plan to begin to remove accommodation this year.

Overall it’s a thoughtful article, but I can’t help thinking that something is missing. First let’s review the difference between inflation and NGDP targeting.  If we use the equation of exchange:


You can think of the two regimes as offsetting velocity shocks, but responding differently to supply shocks.  Under strict inflation targeting, you shoot for 2% inflation no matter what.  Under flexible inflation targeting (such as the Fed’s dual mandate, or NGDP targeting), the Fed lets inflation rise and RGDP fall during a negative supply shock.  Calomiris correctly points out that in the near term it might make sense for the Fed to let inflation fall below 2%, and RGDP to accelerate, as long as NGDP is well behaved.

So far so good.  So what’s my nagging worry here?  My concern is that the Fed doesn’t seem to have the right REGIME in place.  So even if they handle the current oil price decline correctly, I’m not confident they will avoid the horrendous mistakes made in 2008-09, in the next recession.  To understand those mistakes let’s review three options:

1.  Stable inflation

2.  Countercyclical inflation (this is what the Congress, Calomiris and I all seem to prefer.)

3.  Procyclical inflation

You might be thinking; “Wait a minute, I don’t recall Congress asking for countercyclical inflation.”  Not explicitly, but it’s implicit in the dual mandate. Obviously the dual mandate rules out option #1, which ignores jobs.  Any loss function that worries about both inflation and jobs, will bias policy toward more stimulus when unemployment is high, and less stimulus when unemployment is low.  This means the Fed will face a “trade-off” and will tolerate slightly above target inflation when unemployment is higher than desired, and slightly lower than target inflation when unemployment is low.  Inflation should be countercyclical.

Let’s contrast this approach with its exact opposite.  Suppose Congress had instructed the Fed to target inflation at 2%, but, “while you are at it, try to be as cruel to the jobless as possible.  Create as much jobs market instability as possible, consistent with 2% inflation over time.”

Under that mandate (let’s call it the “cruel mandate”) the Fed would do a tight money policy when unemployment is above target.  Yes, they’d miss their inflation target on the low side, but that loss would be offset by the “benefit” that they’d receive from screwing the workers.  Under that sadistic policy they’d run a procyclical inflation rate, just the opposite of what they are currently supposed to do.

Now of course this is exactly what the Fed did in 2009, they ran inflation well below target during a period of 10% unemployment.  Defenders of the Fed will claim that this was unintentional.  I agree.  But it was also due to a flawed inflation targeting IT regime, which biases you toward procyclical inflation, especially at the zero bound.  And my fear is that that regime still is in place.

Let’s suppose the next 5 years are pretty good, and then we have another recession.   If you look at the next 5 years in isolation, the period would be a “boom” in a relative sense.  In that case Calomiris is quite right that the Fed should run inflation a bit below target right now.  But this policy only makes sense if it is offset by above target inflation during the high unemployment periods before and after the boom.  In fact, we did the opposite during 2009-13, and I fear we will again do the opposite in the next recession. I fear that instead of inflation rising in the next recession, (which would make Calomiris’s current recommendation appropriate, the Fed will let inflation fall, which will put us right back at the zero bound, and retrospectively make Calomiris’s proposed policy a mistake.

Robert Lucas emphasized that you need to think in terms of policy regimes, not day-to-day decisions.  I don’t care very much whether the Fed raises rates by a quarter point this year.  I care a lot whether they successfully make inflation rise during the next recession, or disastrously allow it to fall.

So what will it be; the pre-2008 dual mandate or a continuation of the post-2008 cruel mandate?

BTW, there are several options that could make the cruel mandate outcome less likely. One option is to raise the inflation target to 3% or 4%, to minimize the zero bound problem. A better option is NGDP level targeting, which also reduces the zero bound problem, at a lower inflation rate.  Best of all is NGDP futures targeting.  No zero bound problem.

HT  Ramesh Ponnuru



22 Responses to “Calomiris on Fed policy”

  1. Gravatar of Don Geddis Don Geddis
    29. January 2015 at 20:45

    “Dual mandate” vs. “cruel mandate”. That’s gold, right there! Nice turn of phrase.

  2. Gravatar of Lorenzo from Oz Lorenzo from Oz
    29. January 2015 at 21:34

    Once you understand–in my case, due to reading this blog–that a shift in interest rates is a very different thing when done by the RBA compared to done by the Fed because they have different policy regimes even though they have the same dual mandate, so much makes so much more sense.

    But it is so hard for the folk of the concrete steppes to see that.

  3. Gravatar of Ray Lopez Ray Lopez
    29. January 2015 at 22:15

    Sumner makes two fundamental mistakes. Three actually.

    1. He assumes Charles Calomiris, a giant economist from Columbia University (he stands over 24 hands high and weighs 30 stones), does not know what he is talking about. So only Sumner knows anything?

    2. You might be thinking; “Wait a minute, I don’t recall Congress asking for countercyclical inflation.” Not explicitly, but it’s implicit in the dual mandate. – wrongly assumes the Phillips Curve is still viable (‘a little inflation is the tradeoff for less unemployment’ – However, to give Sumner some credit, Congress itself believes in the Phillips curve, having been indoctrinated by Fed economists over the years. So indeed the Fed has to dance to this silly tune.

    3. Uses the Lucas “rational expectations” fudge fairy to support his claims. What is this fudge fairy? It’s not a chocolate eating Peter Pan, but rather it’s the following ‘logic’: if policy X does not give expected outcome Y, then simply claim that the policy makers were not credible in executing the policy X in the eyes of the public, hence the public did not act as expected to give outcome Y. This is metaphysics. If the policy does not work, you can safely claim the policymakers were not credible. But you cannot test this hypothesis, making in unscientific.

  4. Gravatar of Rajat Rajat
    30. January 2015 at 03:23

    I don’t think Congress expects the dual mandate to produce counter-cyclical inflation and I’m not sure Congress sees the dual mandate as relevant to supply shocks.

    The average (Congress)person, economics commentator and central banker thinks of monetary policy as having long lags, which make it very difficult to achieve 2% inflation and the natural rate of unemployment simultaneously. Hence, most people interpret the dual mandate as allowing for high inflation when unemployment is low and vice versa. In other words, the dual mandate is a way of providing the Fed with some margin for policy error by saying that missing either the inflation or unemployment target is not all bad because society receives some value from the under/overshoot. Pro-cyclical inflation if you like.

    Counter-cyclical inflation only makes sense when you think of supply shocks and I don’t think most people see supply shocks as relevant to the dual mandate. This is because a supply shock is either very obvious and can be safely ignored (the Australian RBA for instance tries to ‘look through’ obvious supply shocks), or else are fiendishly difficult to identify in real time (eg the UK negative productivity shock). In the latter case, the central bank will usually be loath to accept the high inflation especially when unemployment is high because it will look like the bank is missing both targets and because it is a nominal variable, everyone agrees (?) the inflation target is more important to a central bank.

    Re 2009, the Fed’s defenders would say the miss was unintentional – they were trying really hard but monetary policy was struggling to gain traction. Policy was highly stimulatory, don’t you know?

    The same problem could arise under NGDP targeting, even NGDPLT, for so long as low interest rates and QE are seen as reflective of very easy policy and raise the risk of future dangerous ‘imbalances’.

  5. Gravatar of Rajat Rajat
    30. January 2015 at 03:37

    Ergo, I don’t think Calomiris is making a point relating to the dual mandate. What he’s saying is that the positive supply shock will have only a temporary effect on inflation – ie the Fed should ‘look through’ it. Rather, the positive supply shock will mask a brewing pick-up in underlying inflation, so it is important for the Fed to get moving soon because due to effectiveness lags, a rise in the Fed funds rate mid-year will take some time to affect future inflation. He is not saying the Fed should try to undershoot its inflation target; he’s saying they need to move soon to prevent inflation overshooting later on.

  6. Gravatar of Nick Nick
    30. January 2015 at 03:59

    Bravo on ‘giant’ calomiris. That’s good stuff. But what’s his reach?

  7. Gravatar of foosion foosion
    30. January 2015 at 04:00

    I don’t care very much whether the Fed raises rates by a quarter point this year. I care a lot whether they successfully make inflation rise during the next recession, or disastrously allow it to fall.

    I worry that raising a quarter point in the middle of this year, including the signal it sends, could help move the economy towards recession.

    Given the current state of the economy, a rate hike, even a small one, could “disastrously allow it to fall”.

  8. Gravatar of TravisV TravisV
    30. January 2015 at 04:53

    Yglesias: “The average American household was poorer in 2013 than it was in 1983”

  9. Gravatar of Jeff Jeff
    30. January 2015 at 05:13

    Ray, you either don’t understand what “rational expectations” means, or you’re being dishonest.

    Abraham Lincoln anticipated rational expectation when he said “You can fool some of the people all of the time, and all of the people some of the time, but you can’t fool all of the people all of the time.” If you think about how you would use that insight as part of your economic modeling, what you’ll come up with is rational expectations or something very close to it. The way Lucas uses the term, it is a restriction imposed on the economist constructing an economic model in which expectations play a role. All it says is that those expectations should be consistent with what the model itself predicts.

    If you think your model is correct then you think it is capturing something important about how the world works. Otherwise you wouldn’t have bothered. If people aren’t stupid, then as Old Abe says, they will figure out how the world works and expect the same outcomes that your correct model predicts. That is what we mean by “rational expectations”.

    Lucas and others in various places use this idea to bash policies that can only succeed if people are too stupid to understand what the policymakers are up to. If you see some regularity in economic data that makes you think people are that dumb, and you move to exploit their stupidity, it is likely to come back and bite you. That’s my colloquial version of the Lucas Critique.

    The idea that some policies have to be credible to be effective has nothing to do with rational expectations. Take the Swiss peg as an example. If the Swiss central bank were perfectly credible, then all it would have to do is announce the peg, and the Swiss franc would immediately drop to Euro 1.20, and the bank wouldn’t have had to do anything else. But in the real world, no central bank is that credible, so the Swiss did have to spend a lot of francs buying Euro-denominated assets to keep the franc from appreciating above the peg. As it turned out, those who found the Swiss central bank not credible were right. But I don’t see why you think this has anything to do with rational expectations, which is, I repeat, a modeling strategy.

  10. Gravatar of Don Don
    30. January 2015 at 05:37

    I’d like to question the word “unintentional” in the statement “Defenders of the Fed will claim that this was unintentional. I agree.” That statement implies that the Fed was confused about their abilities to measure or control inflation. But we know that creating inflation is always doable and measurement is pretty good. Thus it must have been intentional. I assume that they have their motivations as there are always winners and losers.

  11. Gravatar of Brian Donohue Brian Donohue
    30. January 2015 at 05:51


    The funny thing is, understanding rational expectations is the key to understanding why the Phillips curve trade-off breaks down. Loopy Lopez strikes again!

    @Travis, I wonder what including the value of Social Security (and other income streams like pensions for our public-sector friends) in the very low interest rate environment of today versus 1983 looks like? Aren’t these claims assets?

  12. Gravatar of Sam Sam
    30. January 2015 at 06:39


    Off topic, the 2014Q4 Hypermind NGDP market outcome was determined a little more than an hour ago. (2.544% SAAR of NGDP growth, by my calculation, from BEA’s NIPA table 1.1.5.) The market (~4.35%) was off by even more than the Bloomberg consensus (4.2%)!

  13. Gravatar of Ray Lopez Ray Lopez
    30. January 2015 at 06:42

    @Jeff–thanks, that was a good informal definition of the Lucas Critique. A better one is here:

    But my criticism stands. If your model X does not give expected result Y (or that result breaks down) you can claim that the model does not account for the Lucas “deep parameters” and the parameters of the model is not structural, i.e. not policy-invariant. But this, unfortunately, is also metaphysics.

    In fact, economics does not have any ‘deep parameters’: this is a mistaken Holy Grail that economists use to justify their existence. The same policies can give radically different outcomes. My favorite passage was from a commentator on this blog the other day, I repeat it below*. The key point is that ‘things got better’ regardless of policy. Judge for yourself the implications for every crackpot today who advocates an economic ‘silver bullet’.

    @Brian Donohue – insults aside, I am pleased we both seem to agree the Phillips Curve does not work. That’s progress.

    * From:
    “In the United States, voters replaced Republicans with Democrats in 1932 and the economy improved. In Britain and Australia, voters replaced Labor governments with conservatives and the economy improved. In Sweden, voters replaced Conservatives with Liberals, then with Social Democrats, and the economy improved. In the Canadian agricultural province of Saskatchewan, voters replaced Conservatives with Socialists and the economy improved. In the adjacent agricultural province of Alberta, voters replaced a socialist party with a right-leaning party created from scratch by a charismatic radio preacher peddling a flighty share-the-wealth scheme, and the economy improved. In Weimar Germany, where economic distress was deeper and longer lasting, voters rejected all of the mainstream parties, the Nazis seized power, and the economy improved. In every case, the party that happened to be in power when the Depression eased went on to dominate politics for a decade or more thereafter. It seems far-fetched to imagine that all these contradictory shifts represented well-considered ideological conversions. A more parsimonious interpretation is that voters simply””and simple-mindedly””rewarded whoever happened to be in power when things got better.”

  14. Gravatar of ThomasH ThomasH
    30. January 2015 at 06:42

    Perhaps one advantage of NGDPL targeting over IT is that as normally defined, the price level includes volatile tradeable goods. This not not a big problem for a large economy like the US if the Fed sticks to a “core” inflation concept, but is for a small economy. There IT really needs to stick to untradeable goods and to let the exchange rate adjust.

  15. Gravatar of Anthony McNease Anthony McNease
    30. January 2015 at 07:35

    Did anybody else here James Bullard’s interview on Bloomberg this morning? Depressing. When asked about raising rates with already very low interest rates, strong dollar, weak global demand he said that our economy is heating up fast and if we didn’t raise rates soon that it “might be too late.” Too late for what he never really said. When asked about the lack of wage inflation he dismissed it as irrelevant. There’s little correlation between wage inflation and real inflation, and that it was a lagging indicator not a leading one. When asked about the danger of raising rates too soon, causing growth to slow down too much and then having to lower rates again in response he basically said that was no big deal. Policies change all the time.

    How did this guy get to be a Fed president?

  16. Gravatar of TravisV TravisV
    30. January 2015 at 07:37

    “Fed’s Bulllard Says Zero Rates No Longer Appropriate for U.S.”

  17. Gravatar of ssumner ssumner
    30. January 2015 at 08:11

    Thanks Don.

    Lorenzo, Good point.

    Ray, You said:

    “He assumes Charles Calomiris, a giant economist from Columbia University (he stands over 24 hands high and weighs 30 stones), does not know what he is talking about. So only Sumner knows anything?”

    After I said:

    “Overall it’s a thoughtful article, but I can’t help thinking that something is missing.”

    And as for your “explanation” of rational expectations, . . . words fail me.

    And congrats on becoming the laughing stock of the Econlog world. We are all enjoying your zany humor.

    Rajat, You said:

    “Policy was highly stimulatory, don’t you know?”

    I presume you are joking here.

    And I don’t follow the rest of your argument. A dual mandate differs from a pure IT regime only when there are supply shocks. They are identical with demand shocks occur. And procyclical inflation is clearly inferior to a 100% pure inflation target, it’s not even debatable, unless I’m missing something. Procyclical inflation makes fluctuations in unemployment even greater than if you have constant inflation.

    foosion, That’s possible, but it would not be the quarter point rise, it would be the policy that follows it.

    Don, Maybe a bit of each. Perhaps they were worried that if they did what it takes to prevent deflation, they would have overshot to high inflation.

    Thanks Sam, That will happen a lot. Congratulations on whoever won.

  18. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    30. January 2015 at 08:19

    First, Calomiris isn’t a monetary economist, he’s a banking guy. And a very good one, if you haven’t already read Fragile By Design, you should;

    However, I always hate discussions of MP based around interest rates. To be fair to Calomiris, that appears to be the way the game is played right now–the way Tom Brady is supposed to play with an overinflated football. But, I’ve heard Calomiris, on a radio interview, speak well of NGDP targeting, so maybe Scott should open a line of communication with him.

  19. Gravatar of foosion foosion
    30. January 2015 at 08:23

    Scott, regarding raising rates this year, I’m concerned about two things, first that anyone who thinks it makes sense to raise rates now is likely to be following a policy that bodes ill for the future (overly hawkish, not well grounded in reality, not likely to react appropriately to the next recession, etc.). See, for example, Bullard. Second, the market reaction and its effect on the real economy.

  20. Gravatar of benjamin cole benjamin cole
    30. January 2015 at 13:27

    Excellent blogging.

  21. Gravatar of Rajat Rajat
    30. January 2015 at 15:55

    Scott, I agree that a dual mandate / NGDP targeting differs from a pure IT regime only when there are supply shocks. I was just referring to how I think most people understand the dual mandate. I think it’s instructive that Lorenzo (second comment) says that the Fed and the RBA have the same dual mandate, when the RBA’s target is 2-3% inflation over the cycle – a flexible inflation target. Yet most people in Australia construe the RBA’s target as allowing downside misses when demand is weak / UnN is high and upside misses when demand is high / UnN is low; most ignore supply shocks in assessing the RBA’s performance against its target. In other words, the flexible inflation target provides freedom to err. I suspect the average Fed-watcher likewise sees the dual mandate as permitting a degree of lag-induced error.

  22. Gravatar of ssumner ssumner
    30. January 2015 at 20:13

    Patrick, Yes, as I said I find his views to generally be quite reasonable.

    foosion, Those are valid concerns.

    Rajat, It’s hard to believe any “Fed watcher” is that clueless, but I suppose anything is possible.

    If it’s just freedom to err, obviously there’d be no point in a dual mandate, you’d just give a single inflation mandate, with a range (Like Australia.)

    It’s really sad that there are people out there who think it’s “appropriate” for the rate of inflation to decline during recessions.

Leave a Reply