Winner of the New Statesman SPERI Prize in Political Economy 2016

Sunday, 21 June 2015

The Bank of England goes Underground

This is a short post to celebrate an important innovation at the Bank. They now have a blog, which not only has a great name (which those who have not been to London may need a subway map to understand), but looks like being an invaluable addition to the UK and economics blogging scene. As I suggested here, this is another in a long line of small innovations made possible by appointing Mark Carney as Governor. Those who experienced previous regimes can hardly believe it.

The blog promises a mix of posts in terms of both content and wonkishness, and to start we have an easily understandable discussion of the implications of driverless cars for the insurance industry, and a more technical piece on macro. The idea behind the macro post is however fairly simple. If interest rates cannot go below some lower bound, the distribution of forecast outcomes will be skewed. If bad things happen, things will be a lot worse than if good things happen. The novelty is to use what are called stochastic simulations of the Bank’s main macro model to quantify this (using, I have to add, a methodology proposed by one of my former Oxford PhD students – well done Tom). Here is a fan chart for inflation which I think is self-explanatory.


The blog does not discuss the policy implications, but they are pretty obvious. As Brad DeLong has recently pointed out (it’s a point that I and others have also made), with non-symmetrical outcomes like this, you should not choose the policy based on what is most likely to happen. Instead you bias your policy to shy away from the very bad outcomes. So in this case instead of aiming for 2% inflation as the most likely outcome, you aim for a policy where the most likely outcome is above 2%, to avoid a situation where the economy hits the lower bound for interest rates. To put it intuitively, when walking along a narrow path beside a cliff, it is natural and probably wise not to walk in exactly the middle of the path.   

27 comments:

  1. Yes of course:

    Inflation is the cure, not the disease. Ah well.

    ReplyDelete
    Replies
    1. Another 'Anonymous' clueless comment. Ah well.

      Delete
    2. Nothing clueless. Spot on. Just wrong to think that a higher inflation target would be damaging. It really is the cure is the cure to the "disease" of downwardly sticky wages.

      Delete
    3. A very generous interpretation of Anon' very limited and a million miles from "spot-on" comment, the content of which Anon doesn't even remotely allude to in their actual comment.

      Delete
    4. Simon21 June 2015 at 13:20

      Are you sure you've read SWL's post?

      "instead of aiming for 2% inflation as the most likely outcome, you aim for a policy where the most likely outcome is above 2%," in other words more Inflation is better than less Inflation. That might be acceptable if you had to fear harmful deflation, which is only the case of strong deflation expectations. To fear that in Britain's present situation is clearly - I repeat: clearly - absurd.


      Delete
    5. @Anon 14:23 - Actually, it's you who's asserting that inflation >2% is worse than inflation at 2%. Do you want to explain why that should be the case?

      Delete
    6. It is not absurd at all - its just a standard result when the uncertainty is asymmetric. What you might be able to argue from these simulation results is that the asymmetry is not that great, so the optimal policy might only involve a most likely outcome involving slightly above 2% (like 2.2%). It would be interesting to see if this intuition is correct.

      This argument would be stronger if the effective lower bound was 0% rather than 0.5%.

      The other thing you could argue, of course, is that the Bank has unconventional tools that could - at least partially - mitigate the ZLB problem, which these model simulations ignore.

      Delete
    7. Not the same anonymous.... that anonymous is not interested in understanding

      Here is the extent of his understanding

      Inflation bad... money worth less

      But you would probably make more money

      Inflation bad... money worth less

      But more people will get jobs the economy will expand...incomes will go up as much or more than inflation!

      Inflation bad... money worth less

      And so on

      Delete
    8. Anonymous21 June 2015 at 18:45

      Crystal clear. A cretin would understand that. But why aren't the Brazilians getting it?

      The Economist June 6th, 2015:

      "Employment and real incomes are contracting; interest and Inflation are rising."

      So much the worse for the facts?

      Delete
  2. As well as giving credit to Carney, I’d give credit (perhaps even more credit) to Bernanke who started blogging on Brookings Institution site a month or two ago.

    ReplyDelete
  3. Will there be any cranky Georgist MMT writers? No. I am the only one.
    God dangit.

    ReplyDelete
  4. Explain this!!!
    http://www.bankofengland.co.uk/publications/Documents/quarterlybulletin/2014/qb14q1prereleasemoneycreation.pdf

    ReplyDelete
    Replies
    1. Well known. What does it tell us about the advantages and disadvantages of inflation?

      Delete
    2. Well known and completely bogus and if this is the standard the BoE wants to maintain then they are intellectually bankrupt.

      Delete
    3. Nothing because it is completely bogus.

      Delete
    4. Tina, where is your evidence that this is 'completely bogus.' It accurately describes our monetary system. Loans create deposits. Are you saying the Bank of England are lying?
      You would have to believe in money multiplier fantasies to think any different.

      Delete
  5. Regarding the 'Inflation is bad' argument..

    If I am 21 with significant student debt that inflation (including nominal wage inflation) is good. It also means that employers are more likely to hire, which is better.

    If I am in my late 20s/early 30s with a large mortgage but minimal savings, then again moderate inflation is good as it reduces the nominal size of the mortgage.

    If I am an employer, then moderate nominal inflation is, again, good. Because wages are sticky downwards; with no inflation I cannot give anyone a pay rise (even if they are good) - and I can't cut the wages of a moderate employee who was taken on at too high a rate. And the debts of the company don't erode. Plus, I have to be very careful indeed about hiring in deflation, less so than with inflation.

    If I'm in my 40s or 50s, then although I should not have debts, I should have a pension fund. You might think this puts me in the 'deflation' camp.. but zero inflation to deflation means that the nominal returns on my pension fund will be small and disproportionately eaten by fees.

    As a retired person? If I have a big stash of cash under the bed then yes, deflation is good. But otherwise most pensioner benefits are index linked, and if my annuity is not, then it's unlikely that I'll notice - your need for disposable income rapidly drops once in to your 70s.

    So who is really hurt by moderate inflation, with 3-5-8% being 'moderate' in my book? People with large cash piles. Those who rely on rent for income might, if rents fail to keep up. People on fixed nominal incomes (are there any?) I'd really like to see an example of a major sector of the population or economy that really loses out.

    ReplyDelete
    Replies
    1. Andrew,

      You haven't thought this through, otherwise you would not have written "Inflation reduces the nominal size of the mortgage." You mean the real value.

      Anyway, I have Thomas Piketty on my side who says that inflation is not a solution because it is a tax on the poor. The reason is that pensions, welfare benefits and salaries (sticky, as you point out) are only adjusted with time lags whereas iInflation is usually continuous. And governments like to save when adjusting transfer payments (indecent, but that's reality).

      And if employers are more likely to hire in times of inflation, why are unemployment and (!) Inflation rising in Brazil?

      Delete
    2. OK, real value.

      Time lags become a problem with high levels of inflation, yes. But against that, debts don't adjust at all. So I'm not quite sure why it is a 'tax on the poor'. I actually think this is special pleading.

      And if incomes don't quite keep up that is offset by the real-debt-shrinkage effect. If you gave me the option of a 25% pay cut and my mortgage paid off right now, I'd take it.

      Brazil is not a developed country.. and I would say that past some limit inflation will cause more problems than it solves (Brazil has inflation just over 8%, which means that the 8% figure I pulled out of my backside as an upper limit must be correct. YMMV with that one.).



      Delete
    3. Andrew,

      I'll take Piketty's word against yours.

      " If you gave me the option of a 25% pay cut and my mortgage paid off right now, I'd take it." Not clear what you mean. Find me the bank that would acept that.

      Brazil is of course a far-off country of which we know little. And who can imagine that they can think straight? No wonder macroeconomics doesn't apply to them. You appear to be more Brazilian than I thought possible.


      Delete
    4. If inflation was reducing the cost of your mortgage, you would find that your interest rate was increasing to compensate.

      Delete
    5. What I mean is that if my mortgage vanished via inflation, I'd still be in a better position even if that same inflation has eroded my real-terms pay. Not quite sure why this is hard to understand.

      I'd also note that you haven't actually made an argument. Who really suffers under moderate inflation? Is is safer to take out a loan or employ someone when inflation is 5% or -4%?

      Delete
    6. "What I mean is that if my mortgage vanished via inflation, I'd still be in a better position even if that same inflation has eroded my real-terms pay. Not quite sure why this is hard to understand."
      Yep. The politicians have tricked people into supporting high land/house prices and low wages. Madness.

      Delete
    7. If your mortgage was going to 'vanish via inflation' you will find that your mortgage provider will charge an interest rate which takes that loss of value into account. Plus a bit more to provide a return on his investment. And if inflation is running high, it will have to be quite a bit more.
      Are you sure that your pay rises are going to keep up with your interest payments?

      Delete
  6. I didn’t see the blog as that much of a criticism of the official line given that they also said (but didn’t present nice charts to illustrate) that the negative inflation skews pretty much disappeared when they changed the lower bound constraint from 0.5% to 0.0%. That could well be more reaslistic assumption in light of the raft of central banks (ECB, SNB, Riksbank, DNB) who’ve taken rates below zero to seemingly no major negative effects on their financial systems thus far (although of course early days and the UK could be different). And it now looks like a much nearer-term issue than it was at the heat of the crisis (could get into retrospective criticism that “lucky to get away with it” I guess). And no discussion of how the possibilty of (extra) QE fits into the simulations. But overall good to see a bit more opennes at the old lady – although would be even nicer if things like “forecast challenges” got put out there! Mark Astley

    ReplyDelete
    Replies
    1. Actually I never meant to infer that it was a criticism of the official line, in part for the reason you give. I was just trying to make a general point about the implications of the ZLB.

      Delete
  7. The cliff metaphor is excellent. I agree it is the main reason the Bank should overshoot 2%. I identify it with OJ Blanchard even more than with you, Brad and Prof Krugman.

    There is also another less compelling reason why the Bank should overshoot 2% inflation by keeping the policy rate at zero. To the extent that market participants notice the pattern it will cause higher expected inflation if and when the policy rate is zero again. As noted by Krugman long ago, high expected inflation is exactly what the economy needs when at the ZLB.

    The costs of 3% rather than 2% inflation are universally estimated to be very low. The reason given for fearing and fighting any overshooting is that it reduces inflation fighting credibility. A reputation for causing inflation as the economy emerges from the liquidity trap is a good thing. The alleged main cost of over shooting is a benefit.

    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.