Winner of the New Statesman SPERI Prize in Political Economy 2016

Wednesday 9 September 2015

More (dark) thoughts on interest rates

The following has numbers for the UK, but the logic if not the numbers also apply to the US: see Mark Thoma here.

Imagine the following lottery. If you win, you receive a total of £5000 over the next few years. The cost of a ticket? The risk that inflation will be 0.5% higher than it would otherwise have been for a couple of years, where inflation includes the rate that wages increase.

To enter the lottery in the UK you need to cut interest rates. This lottery is just another way of describing the key argument I made in Monday’s Independent article (now complete with chart).

Of course you want to know the odds of getting the prize. The odds come in the form of a puzzle. What are the chances that the economy has, over the last ten years, permanently lost 15% of its normal ability to produce goods and services. Something that has probably never happened to the UK before. [1] Those are the chances of you NOT winning.

We can of course discuss those numbers. But in the UK that discussion appears largely absent. Instead all the talk is about interest rate increases. We seem to have collectively written off 15% of national GDP with just a shrug. ‘Oh that must be supply and there is nothing conventional macro policy can do’ is the general view. That view may be right, but it is important enough that this should be the centre of the national debate. Instead we talk about the need to normalise interest rates, as if the real economy was doing just fine.

Time for a DeLong type lament. If you had told me ten years ago that a decade hence UK output per head would be 15% below the 1955-2008 trend, inflation was zero and yet people would be talking about raising interest rates I would have said you were mad. If you had said that at a time when interest rates and real wages are at record lows the government was proposing to not invest for the future because that was the best way to prepare for the next crisis I would have said you knew nothing about business and economics. If you had said that just years after a huge financial crisis, followed by a host of financial scandals, the City regulator would be sacked because the Chancellor wanted less tough regulation I would have said you were thinking about some corrupt state and not the UK. If this is all a bad dream, what will it take to wake people up?

[1] Economies do appear to suffer some permanent loss to potential output after financial crises: there is a handy summary of studies here (table 4.1) - HT Andrew Goodwin   


  1. People go mad together, but they wake up one by one. Neo to Post-Keynesianism.

  2. " Economies do appear to suffer some permanent loss to potential output after financial crises"

    Why would this be a mystery? Think of all the businesses that went bust during the crisis and its aftermath - never to return again.


  3. Reality is worse than your nightmare. You might also have mentioned that Her Majesty's Most Loyal Opposition would have effectively endorsed the government's austerity position.

    The people who are supporting Jeremy Corbyn in droves can see what is going on.

    It would help the waking-up process if more people in prominent positions also publicly endorsed him...

  4. I like the BBC News 24 man Ben Thompson, who has spent many years being (easily) seconded to the City to watch the screen to see if interest rates will rise, and then being disappointed when it hasn't happened.

    Yet he keeps coming back for more, dressed always like a US television evangelical in that finely pressed suit.

  5. 15% of GDP is (obviously) huge. But I'm curious why you think it's evidence of deficient demand, rather than supply. As an undergraduate I was taught that IS-LM applied in the short run, when prices were sticky (or fixed) but in the long run, output was supply determined (the AS curve was vertical). This result is also present in New Keynesian models - the output gap is the different between the flex- and sticky-price equilibrium, so once prices have had a chance to reset (which after 7+ years, it seems likely that they have) then the output gap is zero, absent new shocks.

    I suspect that this logic is why the Bank of England's own estimates of the output gap are now very small - i.e. they take most of the deviation from this trend-line as supply, rather than demand-driven. I suspect that this is directly driving the current discussion.

    I take your point about the asymmetry of payoffs in case of raising rates too early. But how do you explain such persistent deficient demand in theory? Is there a new friction that you think can/could explain it in a NK framework? Until such a friction is incorporated into models, I suspect that the discussion at the Bank elsewhere will neglect persistent demand deficiency.

    1. Hysteresis.

      Noone is claiming that the current output gap is 15%. Just as you are not (I hope) claiming that the natural level of output that we have now is exactly the same as that we would have had regardless of the demand-side policies. Because, unfortunately, that is the implication of much of undergraduate-level macroeconomics.

      Deficient demand, by construction, is very directly visible in austerity plus now meaningful expectations of rate rises, together with low private-sector confidence.

    2. blenheim: your error, which is very common, reflects how the subject is taught. Just how are prices moving supposed to restore demand to pre-recession levels. The way it works in NK and trad Keynesian models is via monetary policy, but that cannot happen when we are at the ZLB and we have inflation targets. More here

      So if monetary policy cannot lead to a recovery in demand, why does the recovery not happen anyway, as for example consumers finish rebuilding their asset/liability positions? The answer to that is demand depends on expectations of future income, so if those expectations become pessimistic then there is no reason to expect demand to adjust to supply. And if consumers are following the prevailing wisdom for the UK, they will be pessimistic

    3. Thanks a lot, Professor. I suppose that I was thinking that in the limit, as the Calvo parameter goes to zero (or as time goes to infinity) the NK model becomes like an RBC model. Again, maybe that's because I think of NK = RBC+Calvo+Monopolistic Competition: as you point out in the link that's the way it's taught and maybe that's not the right way to think about it. But if you put a central bank into an RBC model it would be powerless, as money is neutral. So I'm a bit puzzled why in the NK model, after a long enough time, the same isn't true. I need to think about this some more.

      As an alternative, if by "deficient demand" we mean high demand for money, then I suppose that prolonged low inflation would lead to higher real money balances, which could increase demand. Though this might not be quantitatively important.

    4. Good grief I had no idea that undergraduate economics has regressed so far.

      So "in the long run, output was supply determined (the AS curve was vertical)". Yes - in the model. But as a certain gentleman whose name is apparently not now known in undergraduate courses once said "In the long run we are all dead". And even if you think that an exaggeration, in the short to medium run, while output is somehow returning to capacity, the unemployed people (and machines and capital) are not producing.

      I remember discussing this issue in the context of regional economics with a very bright new recruit who pointed out that any government intervention must by definition be inferior to the results of the market. I suggested he visit Portugal (crunch place at the time) and review the results of the market there.

      But my point is really that any response to a crisis that starts "In the long run ..." or "By definition ..." is very unlikely to have any relevance to the world we live in.

    5. blenheim: to add to what Simon said: If monetary policy is "bad enough" in a NK model, then, as prices get more and more flexible, it does not approach RBC in the limit. Instead it approaches either explosion or implosion. To see (roughly) why this is so, start with an ISLM model, then ask what happens if the central bank holds the nominal interest rate fixed when shocks hit (horizontal LM, and so vertical AD curve).

    6. Thanks a lot Nick. Though the explosiveness that you describe (in inflation, presumably?) does make me nervous about how far you could push a NK model.

      @David: The point is that the UK is now close to pre-crisis unemployment rates. This is why the Bank of England and others are skeptical that there's much spare capacity left in the economy, (I believe they think that output is less than 1% below potential, whereas SWL notes that actual output/head is 15% below the pre-crisis trend). I'm not sure what your Portugal anecdote has to do with it.

      @Magnus: Aside from the weak performance of output and productivity what's your evidence for hysteresis? In the 1980s, there was a plausible channel to do with "excluded workers" - the long term unemployed being shut out of the labour market. That isn't true now, at least in the UK. I'm not saying I discount it completely, but I just don't see the channel through which it could occur. Delong and Summers simply assume that hysteresis occurs, without either explaining why (e.g. discouraged workers) or providing any evidence.

    7. Well weak output and zero productivity growth are indeed symptomatic. The other smoking gun is the historically low investment rate that has prevailed since 2008. There are a few possibilities here - no need to rely on discouraged workers. Firms have greater incentive to innovate when labour is expensive and even though unemployment has not been so high the labour income share has been falling. Public sector cuts hardly put upward pressure on wages. Investment falls and the productivity potential of the economy falls. Moreover private and public investment are likely to be complementary so absence of the latter entails reductions in the former.

  6. Is there an argument to signal and/or carry out interest rate rises in order to dampen the housing market?

    1. It seems to me obvious that if one sector of the economy (housing) is out of sync with overall economic conditions, any response is much more likely to be successful if addressed to that sector rather than tightening the entire economy.

      In the old days I believe the Bank (or possibly Treasury) imposed quantitative limits on the amount that could be lent to the housing sector. These would probably be impracticable today but the housing market is basically so regulated (Stamp Duty, land registration) that it should not be beyond the wit of man, or even civil servants and economists, to work out a financial deterrence to 'irrational exuberance' in the sector. Whether the Daily Mail would allow it, is of course another question!

    2. Land Value Tax and targeted fiscal policy needed.

    3. Relaxing planning restrictions in/around London would

      (1) Boost the UK's labour productivity as more people would be able to take advantage of London's high wages
      (2) Transfer wealth from older, richer property owners to younger poorer graduates/professionals

      Foreign cities remain pleasant with far higher population densities. I'm surprised at the Left's inability to advocate such a beneficial progressive policy.

  7. This comment has been removed by the author.

  8. I understand the thrust of the piece and am sympathetic. But the nature of the bet does not appear well specified. Maybe, I am missing something.

  9. SW-L;
    " If you had told me ten years ago ... I would have said you were thinking about some corrupt state and not the UK."

    Ah! Glad you have finally come around; this whole scenario has nothing to do with economics, technocratic steering of the economy or any sound, reasonable and rational economic management. It's ideological. Them v Us. And we're losing.

    "what will it take to wake people up?"
    As many 'awake' people as possible going door-to-door until the sleepers get up. It's already happening; feel free to join in.

  10. Great post. "Normalizing interest rates" is one of the most frustrating ideas in modern economic policy. Higher interest rates are an intermediary target that seem to have become a final target in the minds of some.



    Trading binary options is the new trend among investors - especially those who are new to the industry and want to see fast gains. However, it's not easy to navigate among all those brokers, programs and binary options software options that have flooded the internet. What's even worse, many of them are simply scams that are created to "free" you from the burden of your money.

  12. Oh, man. Secular stagnation. The new normal. A kink in the potential output chart. Any coincidence with the property bubble is irrelevant. These are the tenets of our faith.

    Only apostates say, as Delong, "We have deep problems that are the result of the failure to spur a strong recovery."

    Only communists spend recklessly to replace lost demand.

  13. I'm sure I've pointed you to this already and been ignored? But in case not: bailoutswindle dot com

    Force banks to provide honestly priced finance to SMEs, kickstarting the economy by making the bank-levy the cost to the state of providing unemployment benefits; and linking bankers' top-rate of income tax to ten times the percentage rate of unemployment.
    Since big business is bypassing bank finance (according to M King - remember him?) then the only way banks can influence employment is be lending to SMEs.
    Please tell me why this wouldn't happen (you'd be the first economist to do that) if you say it wouldn't.

  14. To be honest i do not think the odds that the UK lost 15% of its productive capacity compared to its trend aren't that low. It happens elsewhere as well, look at Italy, Brazil or Russia. I think it could also be instructive to look at the way growth is calculated to explain the recent slowdown. If you have a bubble inflating and popping the losses following the bubble popping will be weighted higher than the gains before because gdp calculation is price weighted and price of the big inflated sectors were highest before the bubble popped.
    I have also made this point on my blog a couple of years ago, it is in German though so you most likely need to use the translation function:

  15. Nice post. Everything goes really crazy after the financial crunch. Let's hope for the better

  16. What needs to happen for interest rates to fall?


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.