Winner of the New Statesman SPERI Prize in Political Economy 2016


Saturday 3 January 2015

In defence of NGDP targets

For macroeconomists

Tony Yates had recently written a couple of posts (here, and here, but see also the discussion with Andy Harless on the second) slamming the idea of NGDP targets. (From now on I assume this refers to targeting the level of NGDP.) Now you might think that NGDP targets do not need any support from lukewarm advocates like me, given all the supporters in the econ blogging world. That would be wrong, because - as Tony rightly says - most advocates of NGDP targets tend to argue in a model free way. Both he and I want to stay close to the academic literature, at least as a starting point.

I think Tony is wrong when he says that “the case for levels based targets – including NGDP levels targets – is, both practically and analytically, extremely weak”. In making such a claim, Tony should be very worried that one of the supporters of NGDP targets is Michael Woodford, who literally wrote the book on modern monetary theory.

He rightly focuses on the big plus for any levels based target, which is that it can mimic the optimal but time inconsistent policy. After a negative shock to inflation, the central bank commits to future above target inflation which reduces the impact of the negative shock. This is useful in any situation, but particularly at the Zero Lower Bound. He has two arguments against this: one analytical and one practical.

The analytical argument is that in more complex models discretionary policy can yield multiple equilibria. I do not see why that is relevant. We are comparing policy that follows a rule (NGDP targets) with an optimal policy that assumes commitment.

His practical argument against NGDP is that it would be difficult to communicate. I think that kind of follows from its time inconsistent nature. So is Tony saying that the central bank can never follow a time inconsistent optimal policy? The whole point (in my view) of NGDP targets is that they try and directly address the communication problem. The welfare gains from following the optimal time inconsistent policy are large (see my discussion of Iván Werning’s paper for example), so to wave those away as ‘difficult to communicate’ seems - if I may say so - terribly old school central banking.

In a similar manner, Tony argues - again correctly - that these commitment gains depend on forward looking expectations. Under adaptive expectations, levels targets are costly. However central banks, alongside the academic literature, do routinely assume rational expectations in most of the stuff they do. Furthermore, a key argument for inflation targets is that agents are forward looking, and it is an argument that central banks frequently invoke. We need some consistency here.

One final argument Tony uses that really should not be made is that simple models show that inflation variability is about twenty times more important than output variability in assessing welfare, and therefore NGDP targets give too great a weight to output. (The link is a bit more complicated than that - I tried to sketch the maths here.) While these models may be a useful approximation to how the economy works (which is what they were designed to do), they are not a useful approximation to weighing the welfare costs of inflation variability against output variability (which they were never designed for). We know that from survey data and empirical work that looks at the cost of unemployment: I argue this case in detail here. These considerations suggest to me that targeting NGDP will give too little weight to output relative to inflation following shocks.

Having said all this, it is great that Tony is opening up the discussion on the correct level, so we can get away from what often seems like faith based arguments for NGDP targets. I think the framework that he seems to have in mind is also the correct one: the ultimate policy target would be inflation (and the output gap: I would want a dual mandate), and NGDP would be an intermediate target to achieving welfare maximising paths. So I hope this discussion continues. My one last plea is that arguments make clear whether a NGDP targeting regime is being compared to some form of optimal policy, or policy as currently practiced: as I suggest here these are (unfortunately) different things. 


22 comments:

  1. "Under adaptive expectations, level targets are costly," but I think even under purely adaptive expectations, they could be worth the cost in that they prevent the central bank from going off the rails as many did in the 1970's. A level target provides a built-in punishment mechanism for a central bank that fails to hit its target. Looking back at the 1970's, central banks were repeatedly too slow to revise their forecasts, and by the time they finally got it, we were in a big mess. Level targeting would have avoided that. I also think that, even under adaptive expectations, level targeting would help with the zero bound, in that it would light a fire under central banks, push them to use more aggressive experimental policies (at least it would have in the US, not sure about UK & EZ). It seems clear in the retrospect of the past six years that central bank policies (including those of the Fed), as aggressive as they may have seemed to many, have been much too timid.

    ReplyDelete
    Replies
    1. On a tangent, I was surprised when Tony Yates claimed in the comments that the Fed is symmetrical in response from above and below the 2% target. Tim Duy and David Beckworth convincingly argue for an existing bias to hit the target from below, which makes it more of a ceiling than a central tendency.

      Delete
  2. Good post.

    I think that both Keynesians and Market Monetarists agree that current policy is not optimal. We just argue about what is optimal.

    You say I don't read your posts, but I do my best. I'd still like to know how you address
    1. The "single monetary and fiscal authority question", a key part of the monetary offset critique of fiscal policy.
    2. The effectiveness of G question. Put another way, isn't there an upper level of G/GDP that becomes sub-optimal?

    3. Do you agree with Market Monetarists that obsessive IT was the prime cause of the Global Financial Crisis? (I used to blame the banks, but have since changed my mind. They did some poor banking, but the generalised nature of the crisis required a more macro/monetary explanation.)

    ReplyDelete
    Replies
    1. On 1. I have read you wishing to be a benevolent dictator, but jokes aside, what is your real answer to this problem when almost the whole world splits monetary from fiscal authority?

      Delete
  3. Real GDP is 1%, price inflation is 2%, and productivity growth is 2% with a 5% NGDP target.

    Do you want to raise price inflation to 4% or higher?

    ReplyDelete
    Replies
    1. You seem to be assuming that the economy is at its LRAS curve and for some reason won't move along the SRAS curve to the right beyond the LRAS curve. Ok, let's grant those assumptions: what are the benefits of AD expansion under such circumstances?

      Central bank credibility. Let's say that the economy is in the NRU range in your scenario, which it presumably is if output is at the LRAS curve, and therefore expected inflation = actual inflation. However, this would not be the case if the central bank's NGDP target was taken seriously by the public, so the central bank must lack credibility. It is a good thing for central banks to have credibility, and therefore it is a good thing to increase NGDP and thereby raise inflation.

      Of course, the example tells us nothing about whether or not NGDP targeting is pro-inflation or anti-inflation, because make the same assumptions about an economy with 7% RGDP growth and 0% inflation: NGDP targeting would imply deflation of -2%.

      Delete
    2. W. Peden said: "You seem to be assuming that the economy is at its LRAS curve"

      I don't think I am. All of the NGDP targeting I have seen would say raise price inflation. What if price inflation goes up and real GDP goes down?

      Delete
    3. Why does RGDP go down when NGDP goes up?

      Delete
    4. Raise prices by 5%. Wages do not keep up. Most entities buy less. RGDP goes to 0% from 1% (NGDP is 5% or close). Firms react to quantities not prices. They do not expand or could even contract supply. With 2% productivity growth, hours worked falls, most likely from fewer workers.

      Delete
    5. "Raise prices by 5%."

      Make that raise prices to 4% growth.

      "(NGDP is 5% or close)."

      Make that (NGDP is 4% or close).

      Now raise prices from 4% growth to 5% growth. RGDP stays at 0%.

      Delete
  4. I'm no economist but this debate is really interesting to me as I have been following Sumner and Rowe's blogs and it was one of their weak point that all their arguments for NGDPLT seems to apply sometimes better to simply higher inflation targets. It is one aspect that seems central to their view yet they are not that convincing about it.

    From what I understand, NGDPLT is optimal in some ways in that if you want inflation that is on average the lowest possible but yet high when it needs to be, NGDPLT seems get you somewhere close to that. Also NGDPLT tends to make past aggregate contracts add up. You are less likely to have liquidity issues if the economy doesn't need to suddenly pay a larger % of GDP to honor past nominal liabilities.

    Do these benefits offset the potential loss in price stability? I cannot tell. It seems that we should first concentrate on implementing higher targets and then maybe optimize the nature of the targets. Alternatively, going to PLT might provide marginally more stimulation when needed without having to raise the actual target rate of increase which might make this change easier to swallow politically.

    ReplyDelete
    Replies
    1. Writing this made me think of a few more benefits. In a NGDP targeted world, contracts are a claim on a fixed proportion of the economy instead of on a fixed amount of real value. This means that if RGDP is unexpectedly 5% higher you get or pay 5% more real value, if it is lesser you get less. This automatic sharing of the economic pie is a huge stabiliser but I also wonder if it couldn't help with the problem of inequality. Both on the upside where everybody's real pay is automatically higher instead of a few well positioned capitalists capturing all the gains and on the downside where everybody gets a real paycut instead of a few people getting canned and businesses going bankrupt. While it protects capitalists from default on the downside, it may hurt the less competitive ones on the upside where they can't afford the automatic real pay raise (because price rises are not large enough to cover their employees negotiated contracts) . But then, a period a real growth is a better time to lose your job since better paying ones are probably available and social nets are probably better funded than in recessions.

      Delete
    2. "From what I understand, NGDPLT is optimal in some ways in that if you want inflation that is on average the lowest possible but yet high when it needs to be, NGDPLT seems get you somewhere close to that"

      Yes, and the same for low-inflation. Consider the case of a positive supply shock, e.g. a big fall in the oil price (unrealistic, I know, but let's use our imaginations) which reduces inflation to well below target or even into deflation. Under such circumstances, one arguably wants inflation to be low, rather than stoke up a boom as we had in the UK under such circumsntaces in 1986-1988 when inflation was steady but NGDP growth accelerated from about 7.5% to about 11.5%. It is hard for an inflation-targeting central bank to keep inflation well below target in response to supply-side disinflation, as flexible inflation-targeting requires, and the timing may be very difficult and discretionary.

      Also, from what I've read of the recent formal literature on the relationship between macroeconomic stability and financial stability (the most recent paper I read was by Patrick Minford et al on the monetary base and financial stability) it seems that it is income stability rather than price stability that really matters. This is not surprising, in that most financial contracts like mortgages and corporate bonds are not index linked, and they are based on the contracting parties' estimates of the debtor's nominal income. If I lend you £20 with a promise to pay me back tomorrow, that's based on an assumption of your ability to obtain that £20 in the next >24 hours AND inflation. And the average flow of funds to fulfill financial contracts corresponds closely NGDP growth.

      Delete
  5. Shouldn't our target be something related to structural constraints, like matching wage inflation to productivity growth? We're already targeting inflation based on a number someone pulled out of a hat. Can we be certain NGDP targets won't suffer a similar problem?

    ReplyDelete
    Replies
    1. The NGDP growth target is the long run trend, in order to promote nominal stability. Nominal instability with sticky wages risks mass unemployment - standard macro. The long run trend in inflation has not much importance.

      Another key part of the Market Monetarist critique is that inflation is poorly understood and in any case very difficult to measure, just glance through the studies of the ONS or the US BLS on hedonic adjustments and you will be surprised by the low quality and non-standardisation of methods. Inflation is subject to more common and bigger revisions than NGDP.

      Delete
  6. even with high levels of indexation to past inflation (up to 65%!) PLT performs better than IT: https://ideas.repec.org/p/bdr/borrec/783.html

    ReplyDelete
  7. It's indeed unfortunate that textbook NK models imply that the output gap is of second-order importance in the welfare function.

    I would have thought that the reverse was more likely. Generalised inflation - at least when anticipated - in the form of all prices and wages rising at the same rate doesn't have obvious detrimental consequences for welfare. (Though indeed there would likely be some redistributive consequences e.g. That debtors may benefit whilst creditors lose.) On the other hand the economic pain of unemployment is absolutely devastating.

    Even if designed for other purposes it's difficult to take seriously models which have such obviously wrong properties.

    ReplyDelete
    Replies
    1. It's especially odd in that New Keynesians were pioneers on research on hysteresis! If it is true that potential real output is significantly determined by the past size of the output gap, then the output gap is very important.

      Delete
    2. Indeed, I suspect if mainstream NK macro was a chess player, Magnus, it would have long resigned by now.

      Delete
  8. Mmmm, Keynes showed we have to invest more of that preexisting wealth into the current economy so we could increase income and production and employment and gdp

    Why after 80 years is this still a question

    ReplyDelete
  9. simple models show that inflation variability is about twenty times more important than output variability in assessing welfare
    One hard rule we should commit to is that if a model contradicts obvious empiric facts then we should abandon the model, not the facts.

    And it is an obvious empiric fact that variations in labour demand have MUCH larger direct welfare consequences than variations in the price level - just ask the man on the Clapham omnibus whether he'd prefer a doubling of the unemployment rate or a doubling of the inflation rate. Like all that RBC rubbish, people who push the line that stable money is more important than stable real incomes usually just reveal their own vested interests (yes, you may argue that stable money is a MEANS to stable real incomes, but I don't think history teaches that to be true and anyway its a different argument from the one these models embody).

    I suspect the root cause of these simple models being so far astray is their representative agent basis - where's the heterogeneity?

    ReplyDelete

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.