Are hawks and doves simply confused?

I say yes.

Commenter BC tried to explain why it might be rational for some people to be hawks and others end up being doves:

I think of the dove and hawk designations as denoting the bias or errors that one makes in implementing discretionary policy. Doves tend to underestimate the likelihood that low inflation is “transitory” and overestimate the likelihood that high inflation is transitory and thus tend to overestimate the amount of stimulus needed when inflation is low and underestimate the amount of contractionary policy needed when inflation is high. Vice versa for hawks. One can also think in terms of expected future inflation, which is unobservable. Doves’ inflation expectations are persistently lower than hawks’ expectations. Thus, under the same current conditions, doves tend to advocate more stimulus than hawks.

While I concede this explanation might be true, I believe it quite unlikely.  If the dispute were merely a technical disagreement about how to forecast inflation, then hawkishness and dovishness would be 100% uncorrelated with political ideology.  But that’s clearly not the case.  Being hawkish is strongly correlated with being right of center, and dovishness is correlated with being left of center.  That tells me that hawks and doves are simply confused.

Elsewhere I’ve argued that in a world of 2% inflation targets it no longer makes any sense to be a hawk or a dove.  At one time those two terms did have a coherent meaning; hawks preferred a lower inflation rate than doves.  There was no inflation target at that time.  So in 1976, hawks and doves might have disagreed about setting the fed funds target at 7%, even as they agreed that this setting would likely lead to 6.2% inflation.  I believe these two groups continue to exist because they wrongly think we still live in a world where this disagreement has meaning.

We can have a world were the inflation rate is always exactly 2%.  Or we can have a policy that tries to push inflation above 2% during some periods and below 2% during other periods (my preference).  Hawkishness and dovishness have absolutely no role to play in either of those worlds.  Some doves seem to think it’s always possible to have an inflation rate that’s higher than expected, that is, policy can be consistently “expansionary”.  Some hawks wrongly believe the opposite outcome is possible.  They are simply confused.

Hawks were right that policy was too expansionary during the Great Inflation.  But they were wrong about policy during the Great Recession.  It’s wrong to be a hawk for the simple reason that it’s wrong to always favor a more contractionary policy.  That’s like always favoring turning the steering wheel in one direction.

Doves were right that policy was too tight during the Great Recession, but they are in danger of overstaying their welcome and continuing to advocate monetary stimulus at all times.  It makes no sense to always favor an expansionary policy, as we now know that expansionary policies are stimulative only to the extent to which policy is more expansionary than expected.

People should not be either hawks or doves; they should favor easier or tighter money based on whether AD is too low or too high to hit the central bank’s target.

That does not mean we can’t have ideological debates about the proper target of monetary policy.  We can and should have those debates.  (I favor NGDPLT, at roughly 4%).  But if three of the four people in the car have voted to go to Las Vegas, the fact that the driver prefers Taos is completely irrelevant.  The driver needs to steer the car towards Las Vegas.  (The driver can demand the four go see Penn and Teller instead of Britney Spears, as compensation for losing out on Taos.)



26 Responses to “Are hawks and doves simply confused?”

  1. Gravatar of Effem Effem
    19. February 2018 at 13:20

    What do you call a world of below-2% inflation but in which decisions being made today are going to force central banks to accept higher-than-target inflation in the future?

  2. Gravatar of bill bill
    19. February 2018 at 14:03

    I think your description fits better if we had a 2% inflation level target. But with the current bygones be bygones approach, I think that there are hawks that would consistently undershoot 2% inflation and doves who would overshoot given the chance. I too favor 4% NGDPLT to either a 2% inflation target or 2% price level target.

  3. Gravatar of Max Max
    19. February 2018 at 14:56

    The answer to how one can be a hawk or dove in a 2% target regime is simple: by missing the target.

    The economy is not going to force a central bank to stick to its target. On the contrary, there will be a lot of pressure not to undo misses.

  4. Gravatar of Fred Fred
    19. February 2018 at 15:23

    I think, in a world where central banks actually target their forecasts, this description is 100% on point: if 2 percent is the goal, it should be the 1 or 2-year-ahead forecast. There isn’t really much ‘dovishness’ or ‘hawkishness’ built into such an equation other than forecast-model uncertainty, which admittedly could be shaped by similar biases (e.g., faith in the Phillips curve).

    I see a few potential issues, though:

    (a) Confidence bands or tolerance ranges. If 2 percent is the goal, it should lie in the center of the confidence band of a symmetric inflation target. This may not be the case for hawks: the distribution of outcomes from their model have be skewed toward undershooting, doves toward overshooting. This necessarily affects how they may vote on a policy, provided we can’t precisely predict the outcomes. (I agree markets do a better job at this.)

    (b) Risks that can’t be factored into modal forecasts, a similar concern to the above. The risk, for example, of a hard landing in China or spillovers from EME’s stalled rate hikes for quite a while, and the ‘doves’ were far more willing to account for–and sometimes exaggerate–these risks, whereas hawks would often undersell them. (In hindsight, I think they were probably right on this, though probably for the wrong reasons since hindsight, of course, is 20/20.)

    Another argument might be that hawks tend to favor rules, and should, in theory, treat a 2 percent target as a rule. Doves who favor ‘discretion’ may want an inflation target range of acceptable outcomes to expand the diameter of the dartboard to preempt criticism from consistently missing the bullseye.

  5. Gravatar of dtoh dtoh
    19. February 2018 at 15:30


    You said, “But if three of the four people in the car have voted to go to Las Vegas, the fact that the driver prefers Taos is completely irrelevant. The driver needs to steer the car towards Las Vegas. ”

    That would be nice, but I don’t think the Fed operates that way. What’s even worse is that the Fed has a legal obligation to drive toward Las Vegas, and yet they still try to drive to Taos.

  6. Gravatar of ssumner ssumner
    19. February 2018 at 15:48

    Everyone, I think people are too cynical here. Hawks and doves are not evil people, they both favor monetary stability. It does no good to cycle between 1% and 3% inflation, as the relative power and hawks and doves changes over time.

    If the hawks had gotten their way in 2010-11, then Fed policy would have ended up very bad and very unstable EVEN USING THE VALUE SYSTEM OF THE HAWKS.

  7. Gravatar of ssumner ssumner
    19. February 2018 at 15:51

    The hawks at the ECB wanted higher rates in 2011. They won. How’d that work out for the hawks? What are rates like in the eurozone today? Is that because the ECB had to respond to that 2011 mistake with sharp rate cuts, because the ECB’s inflation target is a very real and powerful long run constraint on their behavior?

  8. Gravatar of bill bill
    19. February 2018 at 16:06

    It’s that I think hawks will cycle between 1%-2% and doves cycle between 2% to 3%. When inflation is at 1.7%, hawks want to “play it safe” and raise rates pre-emptively. Doves want to “play it safe” and wait for inflation to actually hit the 2% before raising. Each just has a different view of what “safe” is. Neither is being evil. They each think their way is the best way to get to 2%.

  9. Gravatar of ssumner ssumner
    19. February 2018 at 16:31

    Bill, That would be evil. The Fed has a symmetrical target. Again, look at the ECB in 2011. How’d that work out for the hawks?

  10. Gravatar of ssumner ssumner
    19. February 2018 at 16:35

    The Fed has an agreed upon target of 2%. If you don’t accept that, you should get the hell off the FOMC, as all you’ll be doing is destabilizing the economy. These people are smart, they know that.

    Even doves like Yellen tried to keep inflation close to 2%—people are being too cyclical.

  11. Gravatar of B Cole B Cole
    19. February 2018 at 16:43

    I think Scott Sumner is right.

    That said, I favor NGDP LT at 5% to 6%.

    So maybe I am a dove-ette.

    I think the questions of monetary policy soon shift to instruments.

    If the US becomes like Japan then the question of QE comes to the fore.

    Also, forgotten today is that the Fed used to buy bonds directly from the Treasury when an auction came up short. This started in World War II and persisted into the 1970s and avoids all the claptrap of dealing with commercial banks, reserves, interest on excess reserves, and primary dealers.

    In the next recession there will much to ponder. Hawks and doves have a lot to fight about— against a backdrop of gigantic deficits and debts. Like Japan?

  12. Gravatar of Jeff Jeff
    19. February 2018 at 16:51

    I visited my brother who lives in Vegas a few years ago and he took us to see Penn and Teller’s show. It was fantastic!

  13. Gravatar of Benny Lava Benny Lava
    19. February 2018 at 17:03

    Simpler explanation: we’re all Keynesians now. Hawks wanted lower inflation and tighter money so that Democrats would lose the 2010 and 2012 elections. Based on the evaporation of “hawks” after the last election cycle vindicates this view. Even Yglesias had a write up about it. It is both fiscal and monetary. They aren’t stupid. They just want to win elections.

  14. Gravatar of bill bill
    19. February 2018 at 18:41

    2011 worked out poorly for all Euro countries.

  15. Gravatar of Philo Philo
    19. February 2018 at 21:04

    (1) If the Fed targets 2% inflation and you want a 1% inflation target, doesn’t that make you a hawk?

    (2) I disagree with this: “People should . . . favor easier or tighter money based on whether AD is too low or too high to hit the central bank’s target,” because people might reasonably object to the central bank’s target. You half-acknowledge this, but add: “[I]f three of the four people in the car have voted to go to Las Vegas, the fact that the driver prefers Taos is completely irrelevant. The driver needs to steer the car towards Las Vegas.” This is wrong; the driver *can* drive to Taos, and *maybe this would produce the better result*.

    (3) “[W]e now know that expansionary policies are stimulative only to the extent to which policy is more expansionary than expected.” I thought you were willing to identify stimulative policy with setting “the” interest rate below the natural rate, whether or not this was unexpected.

  16. Gravatar of Fred Fred
    19. February 2018 at 21:34


    (1) I don’t think this is necessarily the case–there’s some debate in the literature on what the ‘optimal’ inflation rate is, and some research comes in about 1.5% or somewhat lower. It’s possible to want to reduce this somewhat during ‘normal’ times but to approach the target in a ‘dovish’ way.

    (2) I don’t think this is quite the right way to read Sumner’s analogy. There isn’t really a ‘driver’ in this case unless you want to understate the strides toward democracy on the FOMC made by Bernanke and Yellen–who aren’t Alan Greenspan or Larry Summers! This is almost working for a university where you are paid to teach Macro 101, but due to opposition to Keynesian theory–say you’re a Neofisheran (yup, I won’t pass up the chance to take a shot at the Neofisherans)–you refuse to teach the class, but instead decide to teach French 101. The university, rightfully, would fire you. People on the Fed can ‘disagree’ with it–Evans and Bullard are sympathetic to NGDP targeting, for example–but, as long as they serve on the Fed, they’re bound by it.

    (3) I think Sumner does accept a negative real interest rate is ‘stimulative,’ but prefers NGDP expectations as an indicator of the policy stance. I struggled a bit while reading this line, as it seems to imply money neutrality, though I think his point here was that policy cannot be ‘stimulative’ forever because eventually prices adjust, so you can’t be ‘dovish’ forever.

  17. Gravatar of Fred Fred
    19. February 2018 at 21:36

    My apologies, I meant a ‘negative real interest gap.’ A negative real interest rate surely would be a terrible way to measure the stance of policy.

  18. Gravatar of Bob Bob
    20. February 2018 at 05:34

    It’s just political/tribal. Doves think placing priority on low unemployment is more virtuous. Hawks think that inflating away debt or papering over politicians’ irresponsibility is less virtuous.

  19. Gravatar of Fred Fred
    20. February 2018 at 06:30

    Bob, I think the best counterexample to this is Jim Bullard. He’s an intellectual heavyweight, for sure, who has been on almost every side of the hawk-dove spectrum–right now, he’s quite hard to pin even pin down, as I believe Sumner wrote about a while back when his most recent ‘new view of the economy’ paper was published last June. I don’t think this applies to him. I’d say the same is true for Loretta Mester. Some are probably influenced by their background, e.g., Esther George worked in markets I believe, as did Rich Fisher, and probably grew to despise inflation risk and thus they placed it at the top of their list of priorities.

    So, I agree with Sumner that this is probably cynical thinking–though I’m not quite sure if my original comment fell into the same bucket.

  20. Gravatar of ssumner ssumner
    20. February 2018 at 11:12

    Benny, I’m not talking about corrupt people, I know very well how they think. I’m talking about people who are persistently hawks or doves, regardless of who is in office.

    Bill, That’s right.

    Philo, Sure you can disagree, but once a decision has been made you need to accept it. If you start driving to Taos, then as soon as you get tired and switch drivers the car will start heading back to Vegas. That creates “driving cycles” analogous to business cycles. That’s bad for everyone.

    We have a system is place, that can only work effectively if everyone plays by the rules. If monetary policymakers are corrupt, the economy becomes more unstable.

    3. There are different ways to measure “expansionary”—relative to the policy target, or relative to the public’s expectations. Hopefully they coincide, but not necessarily, as you suggest.

  21. Gravatar of Philo Philo
    20. February 2018 at 20:38

    “[O]nce a decision has been made you need to accept it.” No, I don’t. I decide whether to accept it case by case.

  22. Gravatar of Jeff L Jeff L
    21. February 2018 at 02:23

    I get that we are focusing on the Fed missing its actual target, but it’s almost more notable that there has been a long time period with the five year forward inflation being well below 2%, when CPI usually trends above PCE.

    And then there are those who if they were trading the market, would bet on 5 year forward inflation (or nominal GDP in a better regime) moving in a particular direction. How can you not classify the people who would cause these prediction markets to clear by taking directional views as hawks or doves?

    In a perfect world with a perfect prediction markets that are only measuring the data we want to measure with national stats that will not get significant historical revisions we can say that the job of the Fed is to be the computer that transforms the signals into policy. We don’t yet live in that world.

    In our world we have a market for an inflation metric that isn’t exactly tied to the index the Fed is targeting. That market even needs to be adjusted for differing liquidity premiums.

    There is no forecast for what will happen to unemployment. So of the two variables they are targeting, they only have long-term predictions for two. This is easy if everything suggests going in the same direction (there has obviously been a strong case that the fed needed to be much more dovish) but that won’t always be the case. And even if they agree, they will differ on the extent of action needed. The most aggressive people arguing for stimulus will be doves, the more cautious members will be hawks.

    As long as central bankers exist in the messy real world, hawks and doves will continue to exist.

  23. Gravatar of ssumner ssumner
    21. February 2018 at 09:28

    Philo, In that case either you are corrupt, or you would never be appointed to the FOMC. If you told the truth, and told Congress you would not play by the rules, you’d never be confirmed.

    Jeff, You say the most aggressive people will be hawks, but does that apply when tight money is needed? When inflation has overshot?

    No, this is not about bad forecasts, hawks tend to be politically conservative and doves tend to be liberal. It has almost nothing to do with a specific mathematical model for forecasting inflation. Reality is not conservative or liberal, inflation moves according to fundamental macroeconomic principles. People living in the real world should be pragmatic, not hawkish or dovish.

    Yes, hawks and doves may continue to exist, because irrational people continue to exist.

  24. Gravatar of Jeff L Jeff L
    21. February 2018 at 14:10

    The people wanting to be more aggressive, who would be labeled as doves, would be those who believe the models saying unconventional policy is weak and more QE & other metrics is needed to achieve the desired impact. Those labeled as hawks would be those who believe studies saying that the expected impact of a smaller amount of unconventional action is very high.

    Another way there would be hawks and doves in a scenario where it is apparently obvious that more easing is needed would be if the doves think that there is no monetary policy impact that occurs at a lag while the hawks believe there is a lag.

    These measures are likely to be correlated with ideology, to the extent that confidence in the application of new ideas (openness to experience) is correlated to ideology.

    I’m starting to wonder if your above argument is a subset of the “rational people sharing the same values should be able to come to an agreement after enough discussion” theory. In that case it is a tautology that people who disagree about anything are irrational.

  25. Gravatar of Philo Philo
    23. February 2018 at 18:49

    Lying to Congress is not necessarily corrupt. They lie to us plenty.

  26. Gravatar of ssumner ssumner
    26. February 2018 at 07:43

    Jeff, That doesn’t explain hawks and doves when not at the zero bound.

Leave a Reply