Winner of the New Statesman SPERI Prize in Political Economy 2016

Thursday 22 November 2012

Offsetting private sector financial balances with fiscal policy

Although I used this post from Chris Dillow as an excuse to talk about a particular problem with looking at ex post estimates of cyclically adjusted budget deficits, I did not address the main issue of his post, which is how the government budget deficit should respond to private sector financial imbalances. As Rich Clayton points out in comments, the issue raised by Chris is important, and it has been raised by others [1], but I’m not sure there are any simple answers.

Suppose the private sector runs a surplus or deficit, which has nothing to do with the business cycle (i.e. it runs a cyclically adjusted surplus or deficit). Should the public sector attempt to match this by running a deficit or surplus (so the national surplus or deficit is much smaller), or should it focus on doing the right thing in relation to the stock of government debt. If the government has a fixed debt to GDP target, for example, should it try to keep to that target, even though the private sector is running a surplus or deficit? This is equivalent to the question of whether the government should worry about current account surpluses or deficits that have nothing to do with the business cycle.[2][3]

There are examples where it would not be appropriate for fiscal policy to respond to private sector surpluses or deficits, which come from what is sometimes called the intertemporal approach to the current account. Suppose, for example, that the economy discovers a finite natural resource, which will temporarily boost output and incomes. Before the resource is extracted, the private sector (and therefore the current account) is likely to go into deficit because of any investment required for extraction, and because consumers will anticipate the income gains and borrow to smooth consumption. When the resource is extracted, the private sector should be in surplus as it builds up assets to prepare for when the resource runs out. In both cases it would make no sense for the government to try and offset these private sector financial imbalances, because they represent the optimal response to the resource discovery.

Now consider the same example, but while the resource is being extracted the private sector does not run a surplus. Perhaps it wrongly believes the resource will last forever, or it is just very myopic. In this case you could argue that the public sector could compensate for the private sector’s ignorance or short-sightedness, and run surpluses on their behalf by raising taxes (or not cutting taxes with the revenue raised by taxing the resource industry).[4] You could think North Sea oil and contrast Norway and the UK here. You could also think this involved the government being rather paternalistic, or alternatively wanting to pass on some of the benefits of the resource to future generations.

Coming more up to date, consider this example. The financial sector for some reason becomes highly risk averse, making it very difficult for firms and consumers to borrow. Monetary policy succeeds in preventing any downturn in economic activity (by, for example, encouraging savers to save less), but the overall scale of credit constraints in the economy remains elevated. Fiscal policy could relax these constraints by cutting taxes. These tax cuts might have little impact on savers if they were fairly Ricardian, but they would help those who were credit constrained spend more. If we viewed the risk aversion by the financial sector as a distortion, then fiscal policy would be reducing the distortion, and thereby increasing social welfare.

Now think about a situation where there is a housing bubble. Despite the best advice of economists and politicians, the private sector borrows to buy real estate because they believe prices can only go up. There is only modest inflationary pressure (house prices are not in the CPI), and the monetary authorities successfully deal with it with small increases in interest rates, but these do not prick the bubble. There are no macroprudential controls. Should the fiscal authority run surpluses in this case? There is a large literature on whether monetary policy should deviate from the inflation target to try and prick the bubble (I discussed a recent contribution here), but it generally takes fiscal policy as given. Should it be?

There is a general consensus that in stabilising the business cycle, monetary policy should take the lead, if it is available (we are not in a monetary union) and unconstrained (we are not at the zero lower bound). We can think of aggregate excess demand as arising because sticky prices mean real interest rates are inappropriate, and monetary policy is the obvious instrument to fix that. But if the distortion is not sticky prices, but irrational exuberance interacting with some financial market friction, does monetary policy have any superiority over fiscal instruments? In particular, if the problem is largely confined to the housing market, would some change in housing taxation not get closer to undoing the distortion than a generalised increase in interest rates. (A change in some macroprudential control might do better still.)

There are even clearer theoretical examples where fiscal policy is the appropriate tool, although they are of less obvious practical relevance. A shift in the monopoly markup of prices over costs (a ‘cost push’ shock) can be exactly offset by changes in tax rates or subsidies (zero additional social costs), whereas monetary policy can only balance additional social costs between inflation and output. Of course varying some particular fiscal instrument is not the same as changing the budget deficit. However if the distortion was temporary (e.g. a housing bubble), and some tax was changed temporarily to counteract it, then there are good tax smoothing reasons why you would not want to change other taxes (or raise spending) to offset the impact on the public accounts.

So there seem to be examples of where it might make sense to run fiscal deficits (surpluses) to partially offset private sector surpluses (deficits), but also examples where it does not. To sort out which is which we need to ask what the distortion or friction is (if any) that is giving rise to the private sector imbalance, and whether there are appropriate fiscal instruments that act on similar margins that could offset the distortion. In other words, we need to think about the microeconomics behind the macro.

[1] In both cases in response to the large corporate sector surplus that has emerged in the UK since around 2002.
[2] If the private sector runs a surplus, but the government is in balance, then the current account must be in surplus by identity.
[3] I must stress that no business cycle issues are involved. Forget the zero lower bound, and assume the central bank successfully moves interest rates such that the output gap is zero, and inflation is on target. This will be the case whatever fiscal policy does, so there is no traditional business cycle motive for any particular fiscal policy stance.
[4] Short sighted consumers will not be Ricardian.


  1. First and foremost, the Fed should have intervened to prick the housing bubble. Interest rate hikes may have been too blunt of a tool, but there were other options. Specifically, if Greenspan had directed the public's attention to the bubble through public speeches and through the research coming out of the Fed during the 2002-05 period, then the bubble would have deflated itself -- MP is all about expectations. In this scenario, where the Fed is so ignorant to an $8 trillion bubble, the collapse of which has the potential to ruin tens of millions of lives, the fiscal authorities should probably step in.

  2. “If the government has a fixed debt to GDP target…” Only a lunatic government would aim for such a target. As Keynes said, “Look after unemployment, and the budget looks after itself”. I.e. look after unemployment, and you’ll find the debt (and/or monetary base) expands some years and contracts in others. That’s just numbers coming out in the wash. Ignore them.

    Next, I like the way Prof Simon says that “There is a general consensus that in stabilising the business cycle, monetary policy should take the lead…” and then cites an example of where that policy falls flat on its face (an asset price bubble scenario).

    If there’s a housing bubble, then it seems obvious to me that the authorities should try to ameliorate it with raised interest rates, while using fiscal to maintain demand at the full employment level.

    So to answer Prof Simon’s question, “does monetary policy have any superiority over fiscal instruments?”, my answer is “no”. In fact I set out numerous reasons why monetary policy is ******* (insert expletive of your choice). See:

    Will someone please set out a similarly brief and cogent reasons as to why monetary is better than fiscal?

  3. There is also the problem of mistimed fiscal action. I don't know about the UK, but in the US fiscal actions have a habit of kicking-in (or fading out) at just the wrong time. If the private sector surplus/deficit is expected to be transient, then a fiscal response might be mistimed unless we assume prescient politicians ( the US half of them believe Adam and Eve rode around on the backs of dinosaurs). This is one answer to commenter Ralph Musgrave's question regarding why monetary policy is generally better than fiscal policy.

    But even if we ignore the fiscal timing issue, what about the second order effects? Let's take the natural resources example in which the private sector runs a temporary deficit while setting up the extraction infrastructure, thereby prompting the fiscal authorities to run a temporary government surplus. What should the government do with these surplus revenues? If the government pays down existing debt, then presumably this will have an effect on the interest rate prompting a monetary response.

    1. "If the private sector surplus/deficit is expected to be transient, then a fiscal response might be mistimed unless we assume prescient politicians"

      Or correctly designed and powerful automatic stabilisers.

      "If the government pays down existing debt, then presumably this will have an effect on the interest rate prompting a monetary response."

      No it doesn't affect the interest rate. The interest rate is always exactly where the central bank wants it to be. They set the rate.

      And let's put to bed once and for all the myth that the central bank is in anyway independent of government.

    2. "Or correctly designed and powerful automatic stabilisers."

      Okay, but I believe Prof. Wren-Lewis was talking about cases in which the business cycle was not a factor, so in those cases automatic stabilizers would not be particularly relevant.

      No it doesn't affect the interest rate. The interest rate is always exactly where the central bank wants it to be. They set the rate.

      Right, but you seem to have missed the part where I said "...prompting a monetary response." Again, my reading of Wren-Lewis' post is that he is mainly focused on the fiscal response when there is no cyclically adjusted surplus or deficit. And that's fine. But I don't think you can just talk about the fiscal side because the monetary authority will almost always respond in some way.


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.