Winner of the New Statesman SPERI Prize in Political Economy 2016

Monday 6 May 2013

More on Naive Fiscal Cynicism

Paul Krugman is absolutely right that one of the rationales for the IMF and others framing fiscal policy in terms of the ‘speed of consolidation’ is a belief that left to themselves politicians will always and everywhere let debt rise. As he points out, the facts for the US tell a rather different story. Here I want to add a couple of additional points by looking at experience outside the US.

As should be well known by now, UK government debt was over a 100% of GDP between the wars, but declined rapidly and consistently to 50% from the second war to the mid 1970s. (See here, or this useful UK site.) The chart below, based on OBR data, takes up the story since then.

UK Debt to GDP ratio (%)

So perhaps the simplest description is that debt to GDP began to level off at around 40% of GDP (which was the target from 1998 to 2008), until the Great Recession hit. There is a lot of interesting detail here, but that would distract from the main point, which is that the UK is another clear counterexample to the idea that governments always let their debt rise.

However deficit bias is not a figment of international policymakers imagination. As I note here with Lars Calmfors, and others have noted before us, government debt in the OECD area as a whole almost doubled (40% to 75%) between the mid-70s and the mid-90s. This deficit bias came not from the UK, and not much from the US, but from Europe and Japan. (To see data on individual countries or country groups, see this nice IMF resource. [1])

So the only reasonable conclusion is that deficit bias is a problem, but not one that afflicts every government at all times. As Lars and I document, there have been various studies that have attempted to draw lessons from this diversity of experience: for example coalition governments may be more prone to deficit bias, but institutional set-ups where the finance ministry is strong less prone. My own support for fiscal councils partly stems from a desire to look for an institutional mechanism to help counter deficit bias. There is no magic bullet here, and institutional solutions will differ depending on national constitutional characteristics.

If this last point seems obviously reasonable, then lets change the dimension from across countries to across time and states of the world. While we may observe a tendency towards deficit bias in normal times, in times like now when government debt is high we seem to be observing a quite different bias - a bias towards austerity. The apparent consensus of 2008/9 that we needed fiscal stimulus looks like the aberration rather than the rule. An international organisation wanting to push sound economics needs to be doing more than arguing for a slower speed of consolidation. By advocating fiscal consolidation when the opposite is required you risk discrediting your advice at other times.

Here is another analogy. Doctors quite rightly encourage us to take more exercise. That is because we have a tendency to sit around too much, but this bias is not universal across people types or ages. Yet when we catch flu, the doctor prescribes rest, not exercise. Indeed, a good doctor may well be specific about the length of rest required, because they know too many patients may think they are better before they have fully recovered. If a doctor told us that while we had flu we should cut down from 30 minutes exercise to 15 or 10, we might begin to doubt their credentials.  

[1] To get the debt series, click on the blue subheading ‘Real GDP growth’.


  1. 1. There is probably a large difference in attitude in this respect between Northern (incl UK) and Southern European governments.
    I would say in general the Northern ones can be trusted with the debt (they might even be overcautious.
    The South as we see now simply looks the opposite. This is no anti-austerity it is mainly keeping things rolling that somewhere in the next couple of years has to be abolished as it is unaffordable.
    If you want a proper stimulus. You use the limited money that you have for your new businessplan and not to keep your cousin a few year more in a nice paying government job.
    Or summarised this attitude doesnot promise much good for debt reduction or structural reforms.

    2. Basically for the largest disaster area, the EZ. Places that have room to do some stimulus donot. And ones that have no room simply donot get their financial house under control and come up with a business model as a country that can replace the one we had and which doesnot look sustainable anymore (and in anyway).

    3. For the US we have the additional problem that Obama's deficit went for a large part in arrangements that are very difficult to reverse. The Obamacare sort. Obama for a large part said stimulus but actually meant transfers. We simply have to see how that will play out. But the combination of both a blocking majority for cutting transfers (and a few other things such as bombing 3rd world nations) as well as against structurally increase taxes doesnot make me very hopeful. This simply looks more like a structural overspending than a stimulus.
    Merkel is doing partly the same. Starting all sort of transferlike programms which will be hard to reverse (but will likely get her reelected).

    4. More for Europe than the US I hardly see structural measures that will make the aging issue financially sustainable. Reason herefor is clear (tough sell electorally). However with no funding to plug the aging hole it is hard to see where they get the money from to lower debt.
    Or in other words if aging costs and debtreduction/reversal of stimulus are going to bite each other and as it looks now they will I know where I will put my money on. You can see even in Holland when things end up as cuts for larger groups, it is still a very hard sell.
    Remember growth in income did the job earlier. If you take the cost of aging into consideration there will be no growth in real spendable income at least.

    5. Post-WW2 was different from now. Especially if you look at growth percentages (and inflation id). Made it much easier to work debt away (btw).

    6. Just an idea why not have a list of stimulus projects.
    - So permits can be arranged beforehand (and are not granted when the next crisis is already over.
    - Avoid that they go into new transfers (as these will be hard to reverse). If you like your welfarestate you have to assure that people who are stuctural payers get their unemployment benefit (without a huge cut). No better way to dump solidarity in the bin than do it otherwise.
    - Have some time to do all kinds of studies (so you can from another economic pov take the most efficient ones. A lot of the projects simply looked total crap.

    7.So in a nutshell. Agree with the historic evidence. However: there are some reasons why this time we should worry a bit more for the future. And the risk of non-compliance are likely mainly depending on social economic culture (if I can summarise it that way).

    8. Annex. It looks like the transfer-stimulus measures (very social of course) are also very ineffective. In a crisis like this one (mainly a longer term one). It ends up as consumption without getting the economy really started. Can fill up a short term gap (or avoid the bottom falling out), but imho not suited for long term crisis. However depending on what your exact definition of stimulus is most looks to be in that form.

  2. Why doesn't the IMF then fulminate against countries in which their politicians (in the UK and US, for example) consistently say that lowering the higher rate of income tax brings in more tax revenue, an argument which is frequently based on a change having taken place over one or two years?

    The IMF knows this is too short a period to make such an analysis.

  3. Can someone give a good reason by it is bad for the private sector to hold net financial assets in the form of government debt? Surely the value of these assets should rise at the same rate as private sector debt, since the former underpins the latter. It is when the two get out of synch that you risk a financial crisis.

    1. I don’t see that the two necessarily need be related, i.e. I don’t see why public debt underpins private debt.

      Public debt and monetary base simply satisfy the private sector’s desire to save: i.e. they avoid paradox of thrift unemployment. If the private sector went into a fit of “unthriftiness”, government debt could be reduced to zero, and there’d be no problems in consequence.

    2. But the monetary base, including government debt, is what banks use to support their fragile balance sheets. They need safe assets to optimise their capital adequacy ratio and use them as collateral for borrowing. Reduce these safe assets and banks will approach insolvency. Is it a coincidence that the US government has run multiyear surpluses 6 times in the past 200 years and in every case this ended with financial collapse and depressions, the only 6 depressions in US economic history? Why do no macro economists ever address this point?

    3. Commercial banks don’t need monetary base to shore up their balance sheets. As long as a bank gets decent quality collateral off those it lends to, the bank won’t have a problem.

      The big attraction of monetary base is that it’s a form of money widely accepted between commercial banks as a means of settling up at the end of each day. But if there were no base, commercial banks would still manage to settle up. They just let inter-bank debts stay in place for a few days or weeks, and if a bank looked like becoming seriously in debt to other banks, the latter would demand real assets (shares, property, etc) at an agreed value in settlement. And in fact commercial banks already by pass the settling up facility offered by central banks in that commercial banks often lend to each other.

      Re surpluses, I fully agree that surpluses tend to be followed by recessions. But I put that down to the paradox of thrift point I mentioned above. That is, if the government / central bank machine confiscates private sector assets, the private sector will try to save so as to build up its stock of assets again. The exception to that comes when the private sector is in irrational exuberance mode: there, there is a good case for confiscation so as to cool things down.

      Re your question “Why do no macro economists ever address this point?”, advocates of Modern Monetary Theory have pointed out time and again that surpluses tend to be followed by recessions, though I’m sure non MMTers have pointed to that “surplus” phenomenon as well.

    4. Ralph, banks do need monetary base. They need it to make position. Other assets are generally not liquid enough, and if they are liquid enough in a boom, they won't be in a recession when the value of those assets (like mortgages) falls. Other banks will then stop accepting those assets. That is how you get a credit crunch and ultimately a run on the banks. If banks didn't use gilts they would have to use more cash.

      I would also point out that persistent government surpluses are generally not deliberate policy actions but are the result of booms in the economy (e.g. USA in 1920s and Spain in 2007). It is the booms that often cause depressions, not the government surpluses (which are merely a by-product).

      The real policy failure is a macroeconomic one. It is the failure of economists to come up with sufficient macroeconomic tools and rules that direct the flow of capital smoothly around the economy and prevent excessive build-ups in areas such as speculative assets and the savings of the wealthy. That means not just concentrating on inflation and GDP growth as indicators of economic performance, but also managing real growth via capital infrastructure and optimising capital investment, reducing the Gini coefficients and unemployment, reducing regional variations in asset prices and the internal balance of trade, setting limits on house-price inflation and stock market booms, to name but a few. That means that economists have to reject much of the neoclassical synthesis and its collective belief in the efficient market hypothesis and the perfect allocation of capital, and instead embrace interventionism at an aggregate level, particularly during booms as well as recessions.

    5. Cantab83,

      I quite agree that IF a bank’s assets look like falling below the value of its liabilities, its creditors will want ready cash (i.e. monetary base) to settle any debts. But given the assumption I made above, namely that a bank gets “decent quality collateral” in exchange for the loans it makes, then the bank’s creditors will be happy with a promise to pay as opposed to actual payment with monetary base.

      Re surpluses, you make some good points, I think.

      I agree with much of your third paragraph, except that I think your plea for “interventionism” calls for a very high level of wisdom and expertise from the “intervenors”. That level of wisdom just ain’t there and never will be: e.g. I don’t see central bank officials spotting house price bubbles.

      Rather than interventionism I much prefer the “rules” to which you refer. And the set of rules that make up full reserve banking are just brilliant, here I think. First, and as regards bubbles, commercial banks just can’t lend money into existence willy nilly as they were doing prior to the crunch. Full reserve bars that activity.

      Second, sudden bank failures are impossible under full reserve, so that does a lot to moderate the credit crunches to which you refer.

    6. Ralph, I hope you aren't advocating the the abolition of fractional reserve banking.

      As for house price bubbles, actually they are pretty easy to spot and just as easy to stop. All you need is for government spending to increase supply, and mortgage regulation to control demand. See here! As for the target, well a median price/median household income ratio of about 4 would seem reasonable. Any more and disposable income falls too far, as do household consumption, savings ratios and growth.

  4. The Fabian Society has recently published two proposals aimed at some form of fiscal consolidation or discipline.

    The first was a rainy-day fund analogous to a sovereign wealth fund that could be used to fund bailouts in the case of financial crises without the need to either (a) print money or (b) rely on the bond market. Unfortunately this proposal is uncosted in the sense that it fails to define how much a government should squirrel away each year, and how that amount should be linked to the various macroeconomic indicators.

    The second was a new golden rule for borrowing that took account of private sector surpluses as well as government cyclical deficits. The reasoning being that the current recession and preceding great moderation coincided with a persistent cash surplus in the corporate sector. Yet the policy document failed to set out how a government could go about addressing this problem.

    It seems to me that both these ideas suffer from the same deficiency that Simon's fiscal councils do. All are good in theory, but all are useless in practice unless you can specify the exact circumstances when each would be used. None of these ideas is quantified, or causally related to the state of the wider economy.

    I do find all this rather reminiscent of Simon's open question from September last year over whether macroeconomics is science or engineering. These policies suggest it is currently neither, but that it should be both. I would also posit that a physicist or engineer would quantify and solve these policy problems rather than just floating vague ideas, and then sitting back and saying "Job done". The job has barely been started.

  5. Great Article and comments.

    I hope you get more readers downunder as they will be better educated.

    Keep it up please

  6. Its fabulous :)


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.