Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 17 July 2017

The OBR’s risk assessment lacks context

In its recent report on fiscal risks, the OBR talks a lot about all the shocks that could make the government debt to GDP ratio rise again. It then says the following:
“None of this should be taken as a recommendation to refrain from particular spending increases or tax cuts, or to avoid particular fiscal risks – that would lie beyond our remit. And there are those who believe fiscal policy is still too tight, given the pace of economic growth and the looseness of monetary policy. But ….”

Should I be grateful for the second sentence, being one of ‘those’ who think that way?
I think the reverse is true. The OBR has played the tune the government wanted, but it is the wrong tune, and this now mature and independent organisation is capable of much better than this. I will first deal with a particular issue to do with monetary policy, and then talk more generally about the concept of ‘fiscal risks’.

Our macroeconomic institutional architecture is based around what I have called the consensus view about macroeconomic policy. This consensus involves what economists call an assignment. The stabilisation of output and inflation is assigned to independent central banks operating monetary policy. Fiscal policy should be confined to managing the government’s deficit and debt, and to help it with that task there should be a combination of fiscal rules and independent fiscal institutions (aka fiscal councils, like the OBR).

In a consensus assignment world, the job of a fiscal council is to stop deficit bias: the tendency clearly observable in some countries before the global financial crisis for deficits to creep up over time. In particular, when all is going well and the deficit appears not to be an issue, it is their job to tell the government to 'fix the roof while the sun shines’.

As I and others have noted many times, this consensus assignment has an Achilles Heel, which is that nominal interest rates cannot go below a number close to zero, the so-called Zero Lower Bound (ZLB). In the absence of some mechanism to allow interest rates to become significantly negative, that ZLB problem means that sometimes fiscal policy makers have to help monetary policy in its stabilisation role. The simple consensus assignment breaks down.

Although most academic macroeconomists recognise that, our institutions find it hard to do so. Monetary policymakers in the UK and Eurozone find it very difficult to say that they have lost their main instrument, and that therefore they can no longer reliably do their job. It seems that our fiscal council, the OBR, has similar problems.

We are currently at the ZLB. The most immediate risk we therefore face is that we are hit by a negative shock and monetary policy is unable to respond effectively. Hence the quote from their document above. But as far as I can see that is it. In their section in the Executive Summary on the risk due to a recession I would have thought the ZLB problem was worth at least mentioning, but it does not appear. Indeed I’m not sure the term ZLB or liquidity trap appear anywhere in the document.

I’m sure the OBR would in defence say two things: assessments of fiscal risks generally look at risks to fiscal sustainability not macroeconomic stabilisation, and their remit precludes them from talking about alternative fiscal policy paths. This is all true. The Treasury wanted a report that would enable them to say we must continue with austerity because of all the risks identified by the OBR. The Treasury also wrote the OBR’s current remit. 

But the OBR is supposed to be independent. Just because the government tries to pretend that there is no Achilles Heel to the consensus assignment, that does not mean it has to go along with that act. In particular, it will (I hope) have noted that the main opposition - which came close to defeating the current government - has a fiscal rule that explicitly says that fiscal policy needs to switch from stabilising debt to stabilising the economy when interest rates are at their lower bound (like now). In this context, I think something beyond a single sentence alluding to the ZLB would have been appropriate.

Tony Yates said similar things yesterday. He also made another important point: a key role of government in many areas is to be a risk absorber. It assumes risks, because it is beneficial to take risks away from individuals or individual generations and spread them more widely, and often the state is the only institution that can do this. In addition, its deficit and debt should be a macroeconomic shock absorber. Given all that, why exactly should we be concerned if various shocks increase government debt? That is what is supposed to happen!

To put the term risk and attach it to some level of debt or deficit, giving us ‘fiscal risks’, is questionable. It is a bit like saying their is a risk that your central heating will come on if it gets cold: that is not a risk, but why it is there. The OBR would no doubt respond that the government has a mandate in terms of a deficit or debt target, and it has been asked to look at risks that this may not be met. But that should not stop an independent OBR from asking more fundamental questions.

Implicit in the idea of ‘fiscal risks’ is either a belief that there is an optimal level of debt which is below current levels, or a view that there is some level of debt so high that markets would start worrying about the government choosing to default. If we are worried about a debt burden on future generations, does it make sense to put all that burden on a current, already disadvantaged, younger working generation? Unless these key issues are addressed, all the risk assessment the report undertakes is meaningless, or worse still just provides support for the government’s misguided policy. It is time the OBR stopped being constrained by its remit, and started providing the public with a useful framework in which to think about ‘fiscal risks’.


  1. The “consensus assignment” involves increasing fiscal stimulus when interest rates are near zero (which in turn tends to raise interest rates) so that come the next recession, the central bank can implement stimulus by cutting interest rates. In short, mortgagors have to pay artificially elevated interest on their debts for years on end, so that central banks can implement their pet form of stimulus, instead of fiscal doing the stimulus (which as pointed out above works perfectly well). Sounds eminently logical….:-)

    Or have I missed something?

  2. It seems to me that there is no clear separation between fiscal risks to sustainability and those of macroeconomic stabilisation; the one is to some extent the obverse of the other.

    Clearly with rates at the ZLB the use of fiscal policy may be advantageous but there may be an issue with sustainability. The actual record, which you allude to, is that governments do not fix the roof while the sun shines and the debt burden gets ratcheted up over time until at some point it becomes unsustainable and will have a real impact on services which may be affected by the need to pay interest on legacy debts.

    You are right that the OBR should address these issues but will it?

  3. Paul Krugman blog, October 7, 2013, Credibility All The Way Down

    "3. It also seems to me that Rogoff has shifted his own argument. In the original piece, it was very straightforward: foreign investors would lose confidence, sending interest rates soaring, which would cause economic contraction. I think, though it’s hard to tell, that he has now dropped that story, replacing it with a fairly amorphous assertion that simple models miss the true nature of risk. OK, I guess, but I do think it’s important to note that Rogoff’s original assertion was that it was all very simple.

    4. Finally, on Wren-Lewis’s point more than mine: Wren-Lewis pointed out that a country like Britain could easily respond to a run on its bonds via quantitative easing. Rogoff says no, because this would undermine its credibility. I’m being a bit flip here, but as I read this we start with a credibility argument; when there turns out to be an easy way to deal with that problem, it’s rejected because that would undermine credibility. I keep looking for something fundamental, but right now it looks like turtles all the way down."

  4. "If we are worried about a debt burden on future generations, does it make sense to put all that burden on a current, already disadvantaged, younger working generation? "

    Hi professor, thank you for your blog posts. I learn so much and look forward to reading your posts.

    Your comment quoted above reminds me of a lecture Nick Stern gave in relation to climate change policy - that the discount rate is an ethical concept, we can't read it off capital markets. I wonder whether the OBR has thought about the rationale for the discount rate it uses.

    Thanks again for your reasoned and public advocacy for good public policy.



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.