Winner of the New Statesman SPERI Prize in Political Economy 2016


Friday 1 November 2013

Something we can all agree on?

Well, all economists at least. Can we all agree that public spending on maintaining and modernising public infrastructure should be increased rather than reduced in a severe recession, whatever the concerns about debt? The spending I have in mind are repairing and refurbishing state owned schools, hospitals, roads - that kind of thing.

Now all those who believe in countercyclical fiscal policy at the Zero Lower Bound will of course agree. But I think those economists who remain worried about the level of government debt should also be able to agree. The reason is that concerns over debt are almost always concerns about long run sustainability. Yet by not fixing holes in the road, or leaks in the school roof, you do nothing to improve the long run sustainability of debt. Almost certainly you are just postponing something that will have to be done sometime, and the act of postponing probably makes the eventual expenditure higher. So postponing public infrastructure investment justs shifts around the timing of public spending, and could well increase its eventually size. It does nothing to tackle issues of debt sustainability.

To see the logic of this more formally, you just have to slightly adapt the analysis in DeLong and Summers (pdf). Their model focuses on hysteresis effects from lower GDP today in reducing output tomorrow, which reduces tax receipts tomorrow etc. But public investment today, in the form of improving existing public capital, will increase output tomorrow just because public capital enhances private output. This idea is pretty standard. For example, in the European Commission’s QUEST DSGE model that I highlighted recently, whereas the long run GDP impact of permanent fiscal consolidation is generally positive for most instruments (because permanently lower government consumption eventually reduces debt and distortionary taxes), the long run multiplier from reducing public investment increases over time, as Annex 1 shows. (By the way, if you read this post, do read Jan i’nt Veld’s response in comments here.) 

There are two other reasons why shifting government spending from the future to the present is clearly a good idea in a recession. First, undertaking public investment is likely to be cheaper in a recession. Firms that will do the repairs want jobs and labour is relatively cheap. Second, to the extent that aggregate demand is a problem, shifting the pattern of public spending this way (what I call ‘pure’ countercyclical fiscal policy) is obviously a good idea. This, after all, is essentially what monetary policy is trying to do for private spending.

I am encouraged to think that nearly all macroeconomists could sign up to this by Ken Rogoff, who while clearly being sympathetic to the attempts by some governments to go in for overall fiscal consolidation, does agree that public investment should not have been cut. Why do I think this is an important point to make? Because it just might stop governments doing the opposite. I have discussed UK public investment before (or the strange attempts to get the private sector to fund public investment), and yesterday Cardiff Garcia pointed out how much US non-defense spending on structures has fallen in the last few years.


This reflects a natural tendency of politicians to do the opposite of what is sensible when they feel they have to cut back on spending. Reducing consumption spending is nearly always politically painful, but it is easier to get away with delaying repairs to (or modernisation of) schools, hospitals or highways. It was partly for this reason that Gordon Brown exempted investment spending from his first fiscal rule in 1998, even though the macroeconomic justification for doing so was debatable (see my discussion here). If all economists could sign up to saying that cutting public investment in a recession is a bad idea, whatever the concerns about debt, then just maybe politicians might think twice about doing this.  

16 comments:

  1. In a nutshell the problems of the Western world. State/GDP of close to or even over 50%, but not being able to finance the basic duties of the state and create the pre conditions to make some bucks in the future.
    In other words entitlements eat up the major parts of the budget and seem to be the hardest to cut next to that.

    ReplyDelete
  2. I don’t agree with the idea that infrastructure spending should be increased all that much in a recession: reason is that it takes TIME to get most infrastructure projects going (sometimes years), and by that time the recession may be over. I.e. it’s far better to increase spending over ALL AREAS of public spending. And even better, increase PRIVATE spending as well.

    But obviously that’s not to rule out increased infrastructure spending in the LONG TERM.

    “the long run GDP impact of permanent fiscal consolidation is generally positive”. Don’t agree: it’s impossible to generalise. E.G. if the private sector WANTS TO HOLD more government debt (even at a near zero rate) than previously, then it’s just got to be allowed to hold it. If we don’t allow that, then the private sector will try to save, and we get paradox of thrift unemployment. (Government debt and monetary base are referred to collectively as “private sector net financial assets” by MMTers, and if the private sector wants more PSNFA, it’s just got to be allowed to have it.)

    “permanently lower government consumption eventually reduces debt and distortionary taxes”. The question as to what public spending should be as a proportion of GDP is a PURELY POLITICAL decision. Personally, I’m a bit right of centre and don’t want any big increase in public spending. But the left wing of the Labour Party would want a big increase in public spending, and there is absolutely no way I can prove them wrong. Likewise, they can’t prove me wrong.

    Next, I don’t agree that taxes are inherently “distortionary”. Obviously they CAN BE: e.g. to take an extreme example, a big increase in VAT on blue jeans, but not on any other type of clothing would be distortionary. But increasing tax on the income of all income groups by the same percentage involves very little distortion. Likewise, increasing VAT on all or nearly all products is non-distortionary.

    As for Ken Rogoff’s idea (put in an FT article) that we should invest in what he called “high return” public sector investments in a recession, that’s nonsense because if an investment (public or private sector) is “high return” it should be made ANYWAY – recession or no recession.

    ReplyDelete
    Replies
    1. Time is not relevant here. The point is that one person's expenditure is another person's income. It's called macroeconomics.

      Delete
    2. Obviously ignoring time, i.e. throwing money at infrastructure in an inefficient manner will raise aggregate demand. But it’s not a bad idea to get VALUE FOR MONEY from public spending is it? Keynes was right when he said it would be possible to escape a recession by paying people to spend all day digging holes in the ground and filling them up. But I can think of better ways of escaping recessions, as could Keynes.

      Delete
    3. Value for money also irrelevant (in capitals or not). The post takes as given the level of infrastructure. It may be good or bad, the point is that replacement investment should occur in times of recession. I think you're groping for other things to object to, rather than what Simon is actually saying.

      Delete
    4. The idea that infrastructure should be replaced REGARDLESS of whether it is “good or bad” is raving bonkers. Beeching claimed (rightly or wrongly) that various railway lines were an economic basket case. Assuming he was right, and assuming there was a recession at that time, would you have ploughed millions into upgrading those lines, knowing full well that it was all money down the drain?

      Personally I’d have spent the money on something that made better economic sense (which would have had much the same aggregate demand increasing effect pound for pound as money spent on defunct railway lines).

      Next, simply restating the idea that “replacement investment should occur in times of recession” does not in any way undermine my point that some infrastructure projects take time to get going. Moreover, infrastructure projects require specialised skills and machinery. The idea that the two latter will necessarily be readily available, given a big increase in infrastructure spending, is just pie in the sky.

      And finally, I’m not “groping for things to object to” in Simon’s post. I’ve a hundred better things to do. Quite the contrary: I’ve pointed to a major flaw in the ever popular idea that we should have big increases in infrastructure spending come a recession. People were saying that twenty years ago and fifty years ago. It’s a hoary old myth which unfortunately is very difficult to bury.

      Delete
    5. You're still fighting a different battle. Whether you're fixing the school roof (worthy?), or digging a big hole and filling it in again (unworthy?) it is always better to do this activity during a recession than in a boom. The issue of worthiness of course is important, but it's separate from the issue of timeliness.

      Scarcity of skills and machinery are in fact a further reason for undertaking replacement investment in a recession than in a boom - and certainly not a reason against it. Fixing the school roof, or mending the potholes in the inner city ring road require labour, which is abundant during recession. Your concern about timing here actually reinforces the case for procyclical spending, rather than undermining it.

      You're making a category error in thinking that Simon's post was a proposal to dig holes, or indeed spend on any 'basket case'.

      Delete
    6. Sorry - that should of course have been "anti-cyclical spending", rather than "pro-cyclical spending".

      Delete
  3. If you do not cut infrastructure, what do you cut? Health? Education? A tax on American owned investment bank bonuses, foreign owned London mansions and buy to let properties would be a start (and remove many other economic distortions and social inequities at the same time) but would it be enough?

    ReplyDelete
  4. Yeah. If they were being honest about their motives, that would probably work. But I think it's abundantly clear by now through their actions that what they're really interested in is dismantling the social safety net. And showing that the government can work successfully by supporting infrastructure would work against that goal.

    ReplyDelete
  5. The entire argument that increasing public expenditure during recessions is a good thing in the long-run is dependent on the presumption that we have competent, intelligent, informed politicians in charge - politicians who will pay down debt when the good times return. Unfortunately, experience teaches us that we do not have such politicians.

    ReplyDelete
    Replies
    1. There is solution to that problem (advocated by Prof. Richard Werner, Positive Money, and others) which is to have the amount of debt determined PURELY by a committee of economists, while strictly political decisions, like what proportion of GDP is allocated to public spending and how that is split between government departments, is left with politicians and the electorate.

      We’re sort of half way to latter system already in that decisions on stimulus are taken largely by committees of economists, e.g. the BoE MPC, and the sundry fiscal responsibility committees that are springing up round the world.

      Delete
    2. It is not only politicians, it is the people that inform them. Blair and Clinton were effective politicians at getting things through - but they were dependent on the policy advice they got. GIve them the right policy, and they could have sold that . But those were Washington Consensus days - free labour flows, capital flows, goods flows, light regulation on industry and labour are unambiguously good - "we can show you the maths, (oh, and of course, the model) to prove it". But the evidence?

      Delete
  6. Actually Simon, since spending on private capital enhances the productivity of the private sector, its multiplier could be higher under flexible prices (or under zero lower bound equilibrium where aggregate supply is steeper than aggregate demand in inflation/output space) than in the more standard sticky price zero lower bound equiblium with a steeper aggregate demand curve (since spending on productivity enhancing capital acts like a positive supply shock). But I suppose there are disagreements on the marginal return to public investment beyond a certain level (which could be hard to resolve econometrically at least given the strong priors of different political camps on the desirability of higher government spending outsize the zero lower bound). Baxter and King, 1993,AER, is a good reference on this I believe. The ECB NAWM modeling team (Coenen and co.) also has a good paper more recent paper evaluating the fiscal multiplier under various assumptions on productive government investment.

    ReplyDelete
  7. This comment has been removed by the author.

    ReplyDelete
  8. It is incomplete to apply macro-economic principles to state expenditure without considering base-line spending. Even with no infrastructural investment, at least 40% of The state's budget is utilised to keep the wheels of government turning in the form of pay, pensions, etc., etc. Parkinson's Law says that the number of staff increases to fit the time available. The less governments do, the more staff they need. Expect to see debt/GDP ratios increase, even in the absence of expenditure on infrastructure.

    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. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.