Winner of the New Statesman SPERI Prize in Political Economy 2016

Saturday 27 October 2012

The UK and Austerity: some facts

There has been some recent debate about whether UK austerity is responsible for the poor performance of the UK economy (which remains poor, despite the Q3 growth numbers). The debate could be summarised thus:

Austerity Critics (e.g. Paul Krugman): “The UK has gone for strong austerity, and since it did so GDP has stagnated – told you so.”

Austerity Apologists (e.g. Tyler Cowen): “But if you ignore tax increases, and public investment, actually there has not been much austerity. The weakness of the UK economy reflects other factors.”

So, for those who are confused, here are some facts.[1]

Austerity can involve tax increases, cuts in transfers and spending cuts, so it is natural to look at the overall deficit. As I have suggested before, the best figure for the direction and impact of policy is the cyclically adjusted primary deficit. Both IMF and OECD estimates show substantial austerity.

UK Cyclically adjusted primary deficit


However any deficit figure may not be a good guide to the aggregate demand impact of policy, because many tax and transfer changes will have smaller multipliers than changes in spending on goods and services. As I noted here, the IMF suggest that UK austerity over the full 2009-2013 period is relatively focused on government spending rather than taxes. However what about 2011 and 2012 in particular? The table below looks at the government spending on consumption and investment numbers that go into the national accounts.

Growth Rates in UK Government Spending on Goods and Services, and Employment
Government consumption


OBR March

Government investment

OBR March
Government employment

OBR March[4]

The OBR’s forecast is quite old, but the new one will not come out until 5th December. I’ve included the OECD’s June numbers because their 2012 figure is a clear outlier compared to other forecasts, and because many people (myself included) use these forecasts. If they are right it will make a significant difference, but for the rest of this post I’ll assume they are not, and use the OBR numbers.

Government investment makes up only about a tenth of government spending on goods and services (G for short), so putting the two together the OBR numbers suggest a fall in G of 1% in 2011, about flat in 2012 and a fall of almost 1.5% in 2013. So, the ‘contribution to growth’ of G to GDP is to reduce it by a bit more than a quarter of one percent in 2011 and 2013, with no impact in 2012. (OBR forecast (pdf), Table 3.4.)

However this ‘contribution to growth’ number in effect compares what actually happens to a counterfactual in which there was no growth in G. A better counterfactual is if G had increased by, say, 2% a year in each year, which would be a kind of neutral, average sort of figure.[6] On that basis cuts in G directly reduce total GDP by about three quarters of a percent in 2011 and 2013, and by half of a percent in 2012. If the multiplier was only one, these are still big numbers, but if it was two they become really large. It would mean that the UK economy might have grown by over 2% in 2011 rather than by less than 1%, without allowing anything for the impact of higher VAT.

So whichever way you look at it, it seems that austerity has had a major impact on UK growth. Of course other things have been important too, but I’m not sure anyone has actually claimed that austerity is the only thing holding back the UK economy.

But in one important sense this is all beside the point. Those who criticise austerity only require three things to be true:

(1) Austerity is real, rather than something imaginary. The figures above make that clear.
(2) Multipliers at the Zero Lower Bound are significant.
(3) If we had not had austerity, monetary policy would not have offset stronger growth.[5]

This is why the multiplier debate is so important. There is a lovely non-sequitur in Chris Giles FT piece on this recently, which Jonathan and Richard Portes have already commented on. Chris correctly notes that theory suggests that multipliers are larger if interest rates are stuck at zero, and then says “so theory tells us very little about the likely effect of fiscal policy on economic growth”. As I have argued many times, theory is pretty clear that multipliers on G will be greater than one in current circumstances, and could be a lot greater than one. And as Paul Krugman quite rightly keeps reminding us, it is theory that has stood up pretty well since the crisis.  

[1] My thanks to Jonathan Portes for some discussion on this issue, but I alone am responsible for what appears here.
[2] Source: IMF Fiscal Monitor October 2012 Statistical Table 2.
[3] Source: OECD Economic Outlook June 2012.
[4] Final quarter of financial year, so 2011 figure is actually 2012Q1/2011Q1.
[5] Unfortunately we cannot be certain that (3) is true, as I have discussed before. However we cannot be certain the other way either, which is sufficient in my view to continue to criticise UK austerity. 
[6] Postscript. We are interested in why the UK economy did not grow by, at the very least, 2% pa. So the natural counterfactual is where G grew in line with GDP at 2%. It is nonsense to say that G did not contribute to zero growth because it also did not grow. 


  1. The multiplier is dynamic, not static. We know many of the factors that reduce the multiplier and factors that increase the multiplier.

    The first step in correcting policy is the admission that multipliers are indeed dynamic and not a constant as some models state as a simplification.

    The second step is more robust models to estimate the multiplier or range of the multiplier.

    The third step is to replace constant multipliers with dynamic multipliers in economic models.

    THis should force economists to validate the assumptions behind the multiplier calculation. This is the way to stop the denial that multipliers can be quite large by policy elites who take the multiplier estimates from authority without regard to no longer valid assumptions.
    -jonny bakho

  2. 'Economic Growth' cannot happen forever, it's a physical and ecological impossibility

    Any economic system which requires perpetual 'growth' to stay 'healthy' is fundamentally broken.

    That's the issue we need to deal with.

    1. Completely wrong. Physical growth and ecological destruction cannot happen forever in a finite world - yes.

      But economic growth - as in nominal GDP growth - is just an increase in numbers in computers. And it is important that those numbers increase exponentially forever.

      Economic growth can reflect population growth and increased use of environmental resources. But a society with zero population growth that is environmentally sustainable must also have economic growth to accomodate new types of products and a little inflation.

      If you want a new iphone, new solar panels, a job and a 2% payrise each year, you want exponential economic growth forever.

      If you don't want any new kinds of products ever or any inflation ever and you are happy to have a higher chance of being unemployed, then keep campaigning for zero nominal GDP growth.

  3. if one sees society as a contest, existing power structure vs citizens, "austerity" is quite good for maintaining the status quol

  4. The argument seems heavily dependent on the counter factual which is correct. But the counter factual of 2% growth is an assumption. Can u defend the assumption a bit? Why would govt expenditure grow when population is static and GDP is flat? I assume you are using real and not nominal numbers in the above post.

    1. I'm glad to raised this. I don't think its an assumption, but what is appropriate given the question. The implicit question is 'why hasn't GDP grown by at least trend', which in this case would be at least 2%. If GDP was growing at trend, we would expect G to grow at the same rate.

  5. Assumption "(3) If we had not had austerity, monetary policy would not have offset stronger growth"

    If it had, then you could blame the Bank of England, right?


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.