Wednesday, 4 April 2012

On successful fiscal consolidations

In a recent Vox piece, Alesina and Giavazzi argue that “adjustments achieved through spending cuts are less recessionary than those achieved through tax increases”. At first sight this seems to contradict basic macroeconomics. As I and others have pointed out on many occasions, the impact of cuts in government spending on goods and services are passed straight through to demand, while the income effect of temporary increases in tax will be smoothed by consumers. That is why balanced budget but temporary cuts in government spending are deflationary.
However, what we may have here is just another example of failing to condition on monetary policy. One of the most comprehensive studies of this issue, discussed by Alesina and Giavazzi, is contained in an IMF report, which uses a ‘narrative’ approach to identifying episodes of fiscal consolidation. (This approach was applied to monetary policy by Romer and Romer here: the detailed catalogue of fiscal events is in this IMF working paper. See Jeremie Cohen-Setton (Bruegel) for more on this.) As Alesina and Giavazzi are a little unfair in the way they characterise this report, I will quote extracts from its first four conclusions.

1)    “Fiscal consolidation typically has a contractionary effect on output. A fiscal consolidation equal to 1 percent of GDP typically reduces GDP by about 0.5 percent within two years and raises the unemployment rate by about 0.3 percentage point.”
2)    “Reductions in interest rates usually support output during episodes of fiscal consolidation”
3)    “A decline in the real value of the domestic currency typically plays an important cushioning role by spurring net exports and is usually due to nominal depreciation or currency devaluation.”
4)    “Fiscal contraction that relies on spending cuts tends to have smaller contractionary effects than tax-based adjustments. This is partly because central banks usually provide substantially more stimulus following a spending-based contraction than following a tax-based contraction. Monetary stimulus is particularly weak following indirect tax hikes (such as the value-added tax, VAT) that raise prices.”

The reaction of monetary policy is crucial here. As the report makes clear, if interest rates cannot fall to offset the impact of fiscal consolidation, or if currencies cannot depreciate because everyone is implementing austerity, the deflationary impact will be much greater.
            To quote Alesina and Giavazzi, the report’s authors “agree that spending-based adjustments are indeed those that work – but not because of their composition, rather because almost ‘by chance’ spending-based adjustments are accompanied by reductions in long-term interest rates, or a stabilisation of the exchange rate, the stock market, or all of the above.” That is unfair. As the quotes above show, and any reasonable reading of the whole report confirms, the impact of consolidation is directly linked to the way monetary policy works. Perhaps the crime committed by the IMF report is that it didn’t stress enough the effects of taxes on the confidence of entrepreneurs that Alesina and Giavazzi seem to think is central.
            Point (4) does indeed imply that cutting spending is less contractionary than raising taxes, but again the reaction of monetary policy is crucial. If, as is suggested, monetary policy does not reduce interest rates following tax increases because of the impact of taxes on prices, then it is monetary policy that is leading to the difference in the impact of spending and taxes.
            There is another interesting, if tentative, result from this analysis. Government spending here includes transfers as well as consumption and investment. The report finds that cutting transfers is mildly expansionary, while the costs of cutting consumption or investment are greater, although they do caution about small sample sizes. As cuts in transfers can be smoothed, this fits with basic theory. Alternatively, it may be that cutting transfers is signalling some kind of intent, which may in turn encourage the monetary authority to ease monetary policy more.
            The reason for stressing the role of monetary policy in all these findings should be obvious. At the zero lower bound, monetary policy cannot compensate in the normal way for the deflationary impact of fiscal consolidation. We cannot use evidence from the past when monetary policy was not so constrained to tell us what will happen today. This is well known for austerity in general, but it applies equally to the composition of fiscal consolidation.



  1. I think I'll have to disagree with you here.
    Optimal fiscal policy at the ZLB (as analysed by people like Correia, Nicolini and others-too lazy to attach exact papers, but google search should work)is to lower consumption and business investment taxes in order to mimic movements in the natural interest rate that monetary policy can no longer track. From this perspective, a consolidation based on cuts in government spending could be less damaging. As for the credibility issues mentionned in your other post on this, I think the most coherent existing analysis is in Corsetti et al's (2012) IMF working paper, which should be mandatory reading for all believers in the keynesian sticky price approach (especially for this guy Krugman).

    1. Thanks for this, as both references I had not seen. A quick look at the Corsetti paper does indeed suggest it is talking about the mechanisms I discussed in my earlier post (, and there are other papers along similar lines (e.g. by Leeper and co-authors). I'm not quite sure why Brad DeLong appears to argue differently (see his recent Vox piece).
      The Correia et al paper formalises and extends an idea I first wrote about some time ago (e.g. my paper in Oxford Review in 2000), which is how indirect taxes can mimic monetary policy. I'm not sure it supports using government spending to consolidate, however - maybe I should write something on this.
      Thanks again for the references.

  2. “At the zero lower bound, monetary policy cannot compensate in the normal way for the deflationary impact of fiscal consolidation.” I trust you are not suggesting that therefor monetary policy cannot compensate at all, professor. If so, I beg to differ.

    Monetary policy can perfectly well compensate in an “abnormal” or unconventional way. That is, at the zero bound, the QUANTITATIVE effects of monetary policy can be exploited. That has been done in the case of quantitative easing. QE could be used to do the “compensation”. But I don’t care for trickle down: I prefer trickle up.

    To put it bluntly, if the deflationary effect of consolidation is excessive, the authorities can just feed cash into household bank accounts till those accounts have swelled by enough to induce enough extra household spending to counteract the above deflationary effect.

    Keynes said, “Look after unemployment, and the budget looks after itself”. I don’t think he intended to summarise EXACTLY the above points in that phrase, but he was saying something pretty near.

  3. Thanks a lot for this. I have my macro final tomorrow and both the Vox and IMF reports were assigned readings. While reading the papers I also could not help but think that they were significantly downplaying the effects of monetary policy.


Originally I did not moderate comments. However spam comments get through and some contain links which then lead some security programmes to give warnings about the blog. I therefore now have to moderate all comments. That will unfortunately lead to a delay before your comment appears. Normally this should be no longer than a day, but just occasionally could be. Please do not be put off by this.