Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 28 May 2012

Does Monetary Policy make Austerity Irrelevant?

                There are three variations of this question that I have come across in recent posts and comments.
1) If monetary policy had targeted nominal GDP, it could have been successful whatever fiscal policy had been.
2) Even within the context of inflation targeting, Quantitative Easing (QE) allows monetary policy to overcome the problem of the zero lower bound (ZLB) for nominal interest rates. So inadequate demand can be put down to not enough QE.
3) Less austerity would have led to higher inflation, which given the recent behaviour of monetary policy makers  would have led to higher interest rates, leading demand back to where it now is.
On (1), I have said that the possibility of moving to a different monetary target has a lot to commend it, and that this should be discussed much more actively in the UK right now. However I do not think targeting nominal GDP means that the monetary authorities can achieve that target at all points in time. The main way I think it overcomes the zero lower bound (ZLB) for nominal interest rates is by promising to create higher inflation in the future, which is itself a cost. The more austerity reduces current demand, the more inflation we will have in the future to counteract it.
                Argument (2) in effect says that the ZLB does not matter: monetary policy just switches from one instrument to another. I think this is seriously wrong. Although recent studies suggest QE has some effect, everyone I talk to who is involved with monetary policy thinks the uncertainties and limitations of QE are of an order above those of conventional policy. If I had to choose between QE and fiscal policy as a way of regulating demand, I would choose fiscal policy.
                Argument (3) suggests that if we had not had austerity, inflation would have been higher, and monetary policymakers would have raised interest rates. There are two uncertainties here. First, we cannot be sure that if there had been less austerity, inflation would have been significantly higher. In the UK there is a very simple reason for this – part of the additional austerity was to raise VAT. But more generally, what we may be seeing today is that inflation is much less sensitive to the output gap when inflation is low.
Second, we cannot be sure how monetary policy makers would have reacted. As I have argued before, sensible monetary policy targets both inflation and the output gap, so when the latter is large and negative, you should be very tolerant of moderate and temporary excess inflation. You certainly do not appear to ignore the output gap. Unfortunately that is exactly what monetary policy makers did do in the Eurozone in 2011, and I really hope this decision is now regretted by the ECB. In the UK they nearly did the same, and I have discussed this episode at length. So I know what monetary policy should have done if we had had less austerity (not raise interest rates), but I do not know what they would have done.
                So I agree that, if demand was stronger because we had had less austerity and more fiscal stimulus, inflation might have been higher, and interest rates might have increased as a result. However not only am I uncertain about this, but the monetary policy response would have been a policy choice. In this situation, should I say that austerity does not matter? Should I instead blame the monetary authorities for something they might have done, had austerity not happened? This seems a bit odd to me.
                I’m also a bit concerned that all three arguments hark back to a place too many were at before (and even during) the recession, which was to believe macroeconomic stabilisation was only about monetary policy.  Now my own writing and research has been quite supportive of what I call the conventional assignment, under the right conditions. But those conditions do not apply at the ZLB, and they obviously do not apply to members of a currency union. There are also other circumstances where fiscal policy could in principle assist monetary policy in its stabilisation role. This tendency to discount the short term macroeconomic effects of fiscal policy is part of the reason we are in our current situation.


  1. Suppose potential output is growing at a constant 3% per year. Let's compare a 2% IT with a 5% NGDP level path target. With no shocks, both monetary policies have the same equilibrium.

    Now suppose there's an exogenous negative IS shock, that pushes the natural rate of interest below negative 2%. There's a problem. We hit the ZLB. Either we use fiscal policy to shift the IS curve vertically up, and raise the natural rate back up above negative 2%. Or we temporarily have to raise the inflation target or NGDP growth path. Or we suffer a recession because RGDP grows at less than 3%.

    I think the above paragraph is how you are looking at it.

    Now let's change the model. Suppose the IS curve slopes up, because (mpc+mpinvest)>1. Then maybe what caused low interest rates was not a downward shift of the IS curve caused by an exogenous shock, but a movement along the IS curve caused by tight monetary policy.

    Equivalently, assume a standard downward-sloping short run IS curve, but that a fall in expected future real income causes that IS curve to shift down/left. Again, tight monetary policy, understood as a fall in expected future NGDP (and hence RGDP), would cause the short run IS curve to shift left, and cause the low interest rates.

    In either case, a credible NGDP level path target would increase interest rates above the ZLB.

    The above three paragraphs are how Scott Sumner looks at it.

    1. You are right about how I see it. The policy raises both future output above the natural rate and future inflation above the normal target level to get current output/inflation close to the level that just releases the ZLB constraint. If current austerity means that the IS curve shifts further down/left, then you need more future inflation to do the same. But the rest is too subtle for me. Why should I assume that a fall in expected real incomes causes us to hit the ZLB now?

    2. Current consumption demand depends on expected future real income as well as on current real income.

      Current investment demand depends on expected future real income (expected future real demand) too. (There's little point in buying machines to increase capacity if the future demand won't be there to buy the extra output).

      So a fall in expected future real income would reduce current consumption plus investment demand (for given current real income and current real interest rates) and so shift the (short run) IS curve downwards, lowering the (short run) natural rate of interest. If the fall in expected future real income is big enough, that natural rate of interest drops below minus 2%, and the economy hits the ZLB.

    3. I understand that it could happen, but the natural story to tell about today is that it is the current value of variables that is kicking the IS curve left/down, rather than future expectations.

  2. Excellent analysis. If your economy is stuck in a ditch, both monetary policy and fiscal policy working together. It simply makes no sense for monetary policy to push harder and harder with QE or NGDP targeting when the Austerity fiscal policy keeps pointing the wheels into the ditch. Weak monetary QE policy is no match for Strong Austerity fiscal policy.

    Fiscal policy is critical to countercyclical efforts to end a recession. The danger of a downward deflationary spiral is in large part counteracted by automatic fiscal stabilizers. The recent shock simply overwhelmed both the automatic fiscal stabilizers and monetary policy. We need larger and better targeted fiscal stabilizers such as an infrastructure bank of projects to initiate when unemployment is too high. Austerity reduces the level of fiscal stabilization and makes it much more difficult for monetary policy to stabilize the economy.

    Monetary policy, no matter how good, cannot overcome bad and wrong-headed fiscal policy.

    jonny bakho

    1. Fiscal policy only has stimulatory effect if monetary policy lets it. It only has recessionary effects if monetary policy lets it. The monetary authority "moves last"; any level of fiscal stimulus can be negated by sufficiently tight monetary policy (see Japan), any level of fiscal contraction can be negated by sufficiently expansionary monetary policy. Hence the fiscal effect is only what is "left over" after the monetary authority has responded. The trouble in the UK is the BoE is nowhere explicit enough in promising to keep the level of transactions (ie. NGDP) up.

    2. Lorenzo - I wrote this post precisely because I see too much of this simplistic kind of analysis. Exactly how does monetary policy negate contractionary fiscal policy at the ZLB?

    3. Why are we at the zero lower bound? With borrowing at record low cost, why does business not borrow and invest?

      The answer has to do with demand and risk premium. When demand collapses, as it did in this recession, risk for return on investment skyrockets. Not only is new investment in production of goods and services unlikely to give returns, but much of the previous investment is now at risk because demand collapse has left overcapacity. The risk will remain high until demand for goods and services returns. Business will simply not invest in a sector that is currently over capacity because the risk for return on investment is too high. Business must be confident in future demand before investment will occur.

      Why did demand for goods and services collapse? It is not that there is nothing to be done or no demand for labor. High demand exists for infrastructure and its maintenance, workforce training &c. The demand is simply going unmet. The collapse occurred because people were over leveraged prior to the collapse and money that previously went to goods and services must now go to balance sheet repair. High unemployment has left many with reduced income. The unemployed must wait for demand to return and demand is waiting for the unemployed to return to work. This is the vicious circle that must be fixed.

      How to fix this? Demand must increase. The most direct way to increase demand is for BigG to borrow at very low rates and purchase goods and services that people need but can no longer afford. BigG can also purchase labor, provide job training and invest in the workforce in other ways. High unemployment means a drop in investment in the workforce, since workers develop many skills "On the Job". BigG can fill the gap.

      The indirect ways to stimulate demand are more complicated or politically more unacceptable. Monetary policy could set a negative interest rate or pay investors to expand goods and services to offset the risk premium. However, paying investors crosses the line between monetary and fiscal policy. At some point, BigG gives people money to invest and that is fiscal stimulus. It is not clear that more overcapacity is a good strategy. Monetary policy could try to force a higher inflation rate that would make lending at zero interest negative in real terms (inflation rate higher than interest rate).

      One problem is how high inflation must be to generate an upward wage-price spiral. If the necessary inflation rate is too high, other untoward consequences might occur. Another problem is the wage-price spiral. Monetary policy is very good at stomping wage-price spirals with higher interest rates. Interest rates have no higher bound. Monetary policy has few independent tools to inflate wages. Wage inflation requires a tight labor market which in a global economy doesn't exist, a regulatory “wage control” policy or BigG supplements to wages. Can monetary policy inflate wages at the ZLB? Wages could be inflated by increasing the minimum wage and COLA to retirees and beneficiaries. However, these steps are traditionally "fiscal policy". A higher inflation rate could force fiscal policy to increase benefits, but the Lords of Austerity might not play along, simply flatline benefits and reduce real spending. The higher inflation would then lead to lower demand, not the intended effect.

      Basically, there are no good ways for Monetary policy to stimulate demand or stimulate wage inflation at the zero lower bound because risk premiums can only be lowered by first increasing demand. The FIRST step to recovery is an increase in demand. The most direct way to increase demand is direct fiscal policy by BigG. Once demand is high enough, the wage-price spiral will move the economy away from the ZLB and monetary policy will only then be in a position of control.

      jonny bakho

  3. We can debate what the optimal response of monetary policy would be till the proverbial cows come home, but the fact is that the Bank of England has a fairly straightforward mandate. It is supposed to keep inflation close to 2% on the CPI measure, and subject to that to support output and employment. What that means in practice is that between 1%-3% inflation, the Bank can pretty much do what it likes, so it's free to decide not to respond to fiscal policy that shifts inflation within that range. Once inflation goes above 3% (or below 1%) it must explain itself in writing to the Chancellor. Thus, if inflation is above 3% the Bank is effectively failing in its primary task of controlling inflation, and it had better have a a good reason why. This is the situation the Bank has been in till recently, and the Chancellor has indulged the failure because of the state of the economy.

    The point is that with inflation above 3% the Bank's main priority has to be to reduce inflation. It can't simply decide to let inflation increase yet further, sticking two fingers up at the government (and the electorate it represents). If the Bank were wiling to allow further increases in inflation, it could have done more QE. Thus we have to assume inflation was at the upper limit of what policy-makers would accept, and that if fiscal stimulus increased inflation the Bank would have increased rates in response. Mervin King basically said as much when he endorsed austerity.

    The more interesting question is whether fiscal stimulus would have increased inflation significantly. There seems to be consensus that interest rate cuts and QE have increased inflation, but is this mainly due to the exchange rate? Is the transmission mechanism of fiscal policy such that it has more effect on output and less on inflation that monetary policy (at least at low interest rates)? If so, would tighter monetary policy combined with fiscal stimulus be a better policy mix?

    1. I do not think the mandate is quite as straight forward as you suggest. The Bank quite rightly has always stressed the lags between its actions and their effect, and so it sees itself trying to control future inflation - they often mention inflation two years ahead. Furthermore, if the output gap today is a good predictor of inflation in two years time, then in principle you could argue that the Bank's interpretation of their mandate is equivalent to minimising the kind of loss function that economists generally use for macro policy.
      The trouble has been that the link between the output gap and future inflation suggested above has not held up well for the last few years. I think, as a result, the Bank and MPC have grown reluctant to continue to use this argument, and have fallen back on looking at the current inflation numbers. So, for that reason, you may be right - but its still a policy choice, and hardly inevitable. As a result, austerity was a policy error.
      The point you make about the inflationary impact of fiscal and monetary policy is a good one. Just as different fiscal instruments have different demand impacts, they may also have different impacts on the demand/inflation trade-off. So, if inflation and monetary policy was really seen as a problem for relaxing austerity, a carefully designed fiscal relaxation might help. A cut in national insurance contributions for young and low-paid workers comes to mind.

    2. You certainly have a point about the time lags, and I think it's a real weakness of inflation-targeting because there's sense in which tomorrow never comes. So the Bank can never be held accountable for a failure to meet the target because it can always claim some unexpected shock raised inflation, but it's OK because they predict it will be on target next period. Then next period comes, and again inflation is too high, but there's no need to raise rates because the Bank confidently predicts inflation will return to target. Etc.

      In principle, this process could go on forever, but I do think in reality there would be political pressure on the bank to respond to increases in inflation beyond a certain point (even at the risk of causing an undershoot in the future) because otherwise it starts to lose credibility.

      Perhaps a price-level target would be more practical.

    3. It's not obvious to me that the MPC's reaction function has changed much over the years, as you suggest.

      They are repeating the same mistakes now that they made in Autumn 2008. Compare the August Inflation report with the May of 2008: the near-term CPI forecast has moved up, the 2-year forecast has drifted down, the median forecast well below target. And yet the MPC refused to ease. They refused to ease all the way to October 2008, a whole month after Lehman blew up.

      Today Spencer Dale - BoE chief economist - is all over the news showing off his hawkish credentials, actively calling for inflation to come down. The near-term forecasts moved up between Feb and May, and the medium-term forecast moved down. NGDP grew 0.2% in 2012 Q1, annualized rate.

      2008/2012 are good evidence that an inflation-targeting MPC has the capacity and willingness to sabotage any amount of deficit spending the Treasury might like to try.

    4. I'm maybe slightly less sure about what the MPC might do than you are, but as my post on inflation targeting not working showed, I do not disagree with your assessment of current monetary policy. The point this post tries to raise is what this then means about how you view the current government's fiscal austerity.

      One point is that in 2010 the government had no idea that inflation today would be such that the MPC might contemplate raising rates. It was just as likely that inflation would be low and the MPC was largely powerless because of the ZLB. So it was a reckless policy given uncertainty.

      A second point is this. Suppose Jonathan Portes suddenly became Chancellor, but was told by the MPC that if he relaxed austerity they might well raise interest rates. Should he, in those circumstance, scrap his plans. I would argue that he should not: call their bluff, and if they do raise rates then take them on.

      For both reasons, current monetary policy does not excuse the error of 2010 austerity.

    5. The fiscal tax/spending plans have been baked into to the MPC's forecasts and they have been setting policy accordingly to offset it. Mervyn King said exactly that in the May IR press conference (per my blog).

      Why would King have been calling for tight fiscal policy since 2009 if he thought he was powerless to offset it? Is he a sociopath? (Maybe let's not go there)

      Why would the government be uncertain about inflation undershooting due to weak AD in 2010, if they had just renewed the MPC's mandate to hit 2% inflation? Osborne clearly does not think the MPC is powerless, given his repeated "hints" about wanting more QE. Likewise, Darling takes credit in his bio for approving QE1 in 2009 as a way to avoid becoming like Japan. (Which we have by the metric of CPI inflation. UK CPI is up 11% since hitting the ZLB, and it's not all VAT; CPI-CT is up 6% since March '10.)

      On "taking on" the MPC - with deficit spending, really? Then truly
      we would become like Japan. Who would "win" if the MPC target 2% inflation, and the Treasury targets 4% with deficit spending? Is that a serious policy suggestion? Scott Sumner did a post on that.

      Why would any Chancellor even try? The Chancellor has the legal power to raise the inflation target to 4% on a whim - he just has to send a letter to the Bank, and they will jump to it. And he probably has room
      to require the Bank do NGDP level targeting, given the wording of the BoE Act.

      All he has to pay for is a stamp, not £30bn of new borrowing.

    6. I largely agree with Britmouse. It's a fair point that in 2010 it wasn't obvious that inflation would be above target today; most people probably expected the opposite. But I think at the time most people were quite bullish about the ability of QE to stimulate demand, and it's not obvious that they were wrong. When you say you would choose fiscal policy over QE, that's as much a political preference as anything. The Tories would not support higher spending, so it's a choice between QE and tax cuts. The tax cuts may or may not be effective depending on whether Ricardian equivalence holds, any incentive effects, and whether they are perceived as permanent, but they would certainly increase an already large deficit (at least initially). So relying on QE is a defensible choice.

      The idea of Chancellor Portas "taking on" the BoE for raising rates in the face of higher than expected inflation is amusing. Like "Look here, Mervin, inflation has been well above target for some time, your forecasts have consistently been wrong, and you seem incapable of keeping inflation below 3%. Now, what I want to know is: why on earth are you raising rates in response to a fiscal stimulus that is likely to exacerbate already high inflation?"

    7. I agree the government appeared to think that monetary policy was capable of offsetting what it was doing to AD with fiscal policy (see the end of my post on US/UK comparisons), and it was wrong - see the section of this post on QE. That is part of my point.

      Raising the inflation target to 4%, or NGDP targeting, might be a good idea, but it is not costless, and not obviously beneficial compared to less austerity - see the first section of this post.

      What interests me is why some people want to suggest that our current problems are all about monetary policy, and fiscal policy NEVER matters for stabilisation (if we had the right monetary policy). As I said in this post, this is a proposition that is right a lot of the time, but not all of the time. Do you think it is right all of the time?

  4. This comment has been removed by the author.

  5. Not just the government, but also the MPC, and I suppose the OBR. I would split your argument down:

    a) The government (and I guess the MPC) think that the CPI rate was (is) still a suitable proxy for judging whether AD growth is sufficient; i.e. that inflation targeting still works.

    b) The government thought monetary policy was capable of keeping the CPI rate on (or above!) target, regardless of the ZLB.

    I'd argue the empirical evidence suggests they were wrong about about (a) but correct about (b).

    With respect to fiscal/monetary stabilisation; is this merely the concern that the ZLB is a "special case"? I just don't see that, given the UK CPI outturns. Monetary policy done properly works by setting clear and credible nominal targets and threatening an unlimited expansion of the balance sheet until they are hit. That is how central banks move expectations, and the MPC did some of that in the dark days of early 2009. Right now all we have is Spencer Dale. :(

    I thought the Swiss experience of contracting their balance sheet after setting a credible nominal target was instructive; the ZLB didn't matter to the SNB.

    I don't understand the point about inflation being costly, but maybe you can whack me over the head with some papers/references. Threatening inflation is not costly. If we actually get high inflation; much faster demand growth results only in price increases and little output growth, that is just a supply-side problem, and the fiscal/monetary argument is moot. Or am I missing the point?

    1. Thanks, that is helpful. I certainly do not want to whack anyone over the head with academic papers, but maybe I will try and spell out my argument in a new post. (I think it is the same as, and comes from, Michael Woodford e.g. Gauti B. Eggertsson & Michael Woodford, 2003. "Optimal Monetary Policy in a Liquidity Trap," NBER Working Papers 9968 - but its not an easy read, for me included.) One of the good things I find about writing posts is that, just occasionally, you realise that what you thought was right is not so obviously right after all. Lets see - thanks for the comments.

  6. Interesting post. Regarding your response to point 1, I think it's a mistake to see monetary policy working by boosting inflation expectations, and (by implication) fiscal policy as boosting growth. Both boost expected future NGDP, with the P/Y split determined by the slope of the SRAS curve.

    I will concede one point, a fiscal stimulus consisting of lower VAT (or lower employer-side payroll tax in countries that have one) will be less inflationary in terms of MEASURED inflation. Oddly this will work by shifting both SRAS and AD to the right, assuming the central bank is an inflation targeter. That's probably one reason why the macro outcome looks worse since the coalition government took power.

    To me, that's a sort of back door monetary stimulus, caused by a quirk in the mandate we give central banks. We actually should be instructing them to target inflation net of VAT (or better yet an entirely different variable like NGDP.) But since we do instruct them to target total inflation, a VAT cut can boost employment in the short term. But I also suspect that Britain has some supply-side issues, as the fall and rise in VAT doesn't really explain the average inflation rate over the past 4 years, does it?

    In replying to Mainlymacro you say the government was not capable of offsetting the fiscal austerity. But that seems to conflict with everything I read. Isn't the BOE refraining from doing more because AD is where they want it to be? Obviously they'd prefer a better P/Y split, but as long as they have an inflation mandate they seem to feel more AD is not required, hence they refuse to take additional steps. It seems to me they lack a will to do more, not the ability to do more.

  7. In my previous comment I should have referred to your reply to econojon, not MainlyMacro.

  8. I recently came across your blog and have been reading along. I thought I would leave my first comment. I don’t know what to say except that I have enjoyed reading. Nice blog.The post is written in very a good manner and it entails many useful information for me. I am happy to find your distinguished way of writing the post. Now you make it easy for me to understand and implement the concept. payroll services oxford Thank you for the post.

  9. This comment has been removed by a blog administrator.


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.