To cut or not to cut?

Suppose you owned a premiership football club that was losing money and not winning many matches. You would face the following dilemma: should you spend more money on players, in the hope of improving the team’s performances, and therefore revenue, or should you save money by selling players but risk being relegated, which, if it happened, would mean that your club would lose out on huge amounts of money from lucrative TV deals?

Because your club is losing money, the current situation is unsustainable, so you are more or less forced to go one way or the other: either you try to spend your way to success or you play it safe. If you try the first strategy and it fails, then you end up in a much worse position than before — you have spent a lot of money and are still losing money. If you try the second strategy, then you will lose your best players, so that even if the club becomes financially sound it will probably face an extended period of not doing very well on the pitch.

In many countries at the moment, of which the UK is unfortunately one, there is a debate rather like this about the national finances. In the UK the national debt stands at somewhere between 900 billion and a trillion pounds. Those are the US billion and trillion, so I’m talking about 10^{12} pounds or so. Here are various facts to help put this figure in perspective.

1. It works out at about £16,000 for every person in the country.

2. Five years ago it was about half as much.

3. A graph of the national debt as a percentage of GDP looks something like this:

National debt as percentage of GDP

Note that it has sometimes been incredibly high — in the last century that was because of the cost of world wars.

4. The deficit (the rate of increase of the national debt per year) is running at around 140 billion.

So what do we do about this situation? There are two arguments, very similar to the arguments I discussed in connection with the hypothetical football club (but not entirely hypothetical — there have been actual examples, such as Leeds United, where attempting the spend-yourself-out-of-trouble strategy has had disastrous consequences). The first argument is often put forward by Ed Miliband and Ed Balls (leader of the opposition and shadow chancellor, respectively). They say that the best policy is to grow the economy and use the proceeds to reduce the deficit and eventually to start paying off the debt. The second is put forward by the government, who argue for reducing the deficit directly by means of deep spending cuts.

Who is right? More importantly, how do we decide? I don’t have good answers to these questions, but that is not my purpose in writing this post. Rather, I want to express my irritation at anybody who claims that the answer is obvious, which includes all politicians I have heard speaking about the subject.

To give a flavour of this, let me explain why it is “obvious” that spending cuts are a bad idea. By this I mean not that they cause a lot of misery — that is genuinely obvious — but that they are actually counterproductive, a claim that both Ed Miliband and Ed Balls have made.

As a preliminary, I shall need to dig out from the back of my brain some half-remembered ideas from an economics A’level I took about thirty years ago, which involved reading and understanding one textbook: Positive Economics by Richard Lipsey. He defined something called the circular flow of income, the rough idea of which is that we make money by providing goods and services and then go out and spend that money on other goods and services, and the money itself goes round and round. This can clearly be modelled by a flow on a graph (that varies over time), so the word “flow” isn’t too bad. (Lipsey tried to keep the maths to a minimum, so didn’t actually say this.) He then went on to talk about injections and withdrawals, which a mathematician might prefer to call sources and sinks. An example of a withdrawal is income tax, and an example of an injection is government spending. Another example of a withdrawal is personal saving, and another example of an injection is investment. And yet another withdrawal/injection pair is importing/exporting.

Keeping things qualitative, injections tend to cause the economy to expand and withdrawals make it contract. However, it isn’t quite as simple as that. Suppose, for example, that you inject money into the economy by printing it and using the printed money to finance a cut in income tax. Then a lot of people will have more money, so will be inclined to spend more. If production can increase then all is well and good, but if it can’t keep pace with the increased demand, then instead prices go up and the effect is inflation.

In general, injections create pressure in the direction of growth and inflation, whereas withdrawals create pressure in the direction of contraction and unemployment (the explanation for which is simple — if the economy is producing less then you need fewer people to do the producing).

Although I do not need to discuss it in the subsequent discussion, I should also mention monetary policy. Increasing the money supply has a growth/inflationary effect, as I have just said. Increasing interest rates has a shrinking effect, since the amount of money a company has to make to make more than the interest on a bank loan (or, if it already has the capital, to make more money than it would make simply by putting the capital in the bank and drawing the interest) goes up, making it harder to be profitable. And if your currency is strong, that has a shrinking effect because it tends to increase imports and decrease exports, thus increasing withdrawals and decreasing injections. And one final highly relevant effect: borrowing money is an injection, and paying it back is a withdrawal.

Obviously, there is much more to say than that little sketch, but it is enough for my purposes here. The argument given by the Eds is that spending cuts cause the economy to contract (or grow more slowly), that this causes unemployment to rise, and that this increases government spending because of the increased welfare bill, as well as decreasing the tax take. By contrast, if you increase public spending, then the effect is to make the economy grow, which increases the tax take and decreases the welfare bill, allowing the government to reduce the deficit.

On the face of it, this might seem a no-brainer: either you go for painful cuts that make the problem worse, or you go on a lovely spending spree that magically cures it. But actually it is a no-brainer in a different sense — it derives its plausibility from our temptation not to use our brains. The reason is that this argument is correct only if the amount of money spent stimulating the economy causes it to grow enough to increase revenue by enough to pay off what has been spent (and more). This might be the case, but there would have to be a significant gearing effect. In other words, what the Eds say is qualitatively correct (that is, when they say that A decreases if B increases, they are correct), but what matters is more quantitative: if the numbers don’t work out, then it’s more like the football team that splashes out on players who then don’t improve the results by enough to get the team out of financial trouble — which gets it into much worse trouble.

Here is another silly argument against spending cuts that one often hears. Every time new figures come out that show that the economy is shrinking, or growing disappointingly slowly, you can guarantee that there will be a Labour politician on the radio claiming that this shows that the policy of cuts is not working. This is silly because, as I’ve already discussed, it is just very basic economics that cutting spending tends to slow growth. And it’s not just Labour politicians who are silly in this way: I’ve also heard coalition politicians attempting to defend cuts by denying that they are a threat to growth, or even somehow saying that they are designed to create sustainable growth. (It’s just conceivable that something like that could be true. For instance, I have sometimes seen suggestions that shrinking and regrowing an economy can cause weaker and less efficient businesses to go under, leaving the economy leaner and fitter. But any effect there might be along these lines will be a long-term one.) So the entire argument is happening in the wrong place, so to speak. The true argument should not be about whether or not spending cuts suppress growth — of course they do — but about whether or not this unfortunate suppression of growth is actually necessary.

What about the argument that spending cuts reduce the government’s tax revenues and increase its welfare spending? Again, this is true, and again the conclusion one draws from it depends critically on those nasty little things that politicians don’t like to discuss publicly: numbers. Suppose, for instance, that a further spending cut of £1 billion per year will have knock-on effects that reduce tax revenues by £200m and increase the welfare bill by £150m. (I have no idea whether these numbers are even remotely plausible but I’m making an abstract point here.) Then the net effect is not counterproductive, since there is a total reduction to the deficit of £650m — it’s just that that reduction is less than the £1 billion. What that boils down to is the depressing fact, if you do decide to go for the cutting option, that in order to reduce the deficit by £1 billion you have to make cuts of more than £1 billion. Of course, it is logically possible that the knock-on effects of cuts might be so severe that they actually cancel out the cuts completely (which seems to be what certain Labour politicians are claiming), but to demonstrate that would be far more difficult than it is to use qualitative arguments about the direction of monotonicity of certain functions.

So far I’ve mainly argued against some common anti-cuts arguments, but the pro-cut arguments are not exactly watertight either, especially because of another phenomenon that politicians often (but not always) like to ignore: there is a continuum of possibilities. [To give another example, I always felt that the most powerful argument against going to war in Iraq was not that it was clearly unnecessary but that it was clearly unnecessary then, an argument that Robin Cook tried and failed to make the main one. In that case the continuum arose because we could decide not just on the yes/no question but also on the timing.] It isn’t just a question of whether to cut, but of how much to cut. This is a continuum that isn’t completely ignored, since Labour is not actually saying that it wouldn’t cut at all — just that it would cut less. But the debate does often seem to slip and become discretized into a yes/no question.

But if we stick with the how-much question, then the wisdom of what the Conservatives are doing becomes completely non-obvious. I do not see any compelling argument against the view that the country would be better off with a less severe package of cuts, and I also do not see any compelling argument that it would be worse off. Even at the current levels it seems that we will be running a deficit for several more years, so the plan is presumably to wait until the global economy improves (if it does, which is another major uncertainty — some have suggested that the banking crisis was a game-changer that has interrupted the normal economic cycle), at which point we can go into surplus. If that is the plan, how do we decide what the optimal deficit is for now? Should we go for the minimum level of cuts that will be sufficiently convincing to the international money markets that we can keep the interest rates on government debt at their current low levels?

I don’t know. But as I said early in this post, that’s the point: it is very hard indeed to know. My gut tells me that growing our way out of trouble is too risky a strategy — I don’t want the UK to be another Leeds United — but unlike our dear prime minister I don’t trust my gut if I don’t have the arguments to back it up. And not even my gut tells me how deep the cuts should be if we do have them. (There is also the question of which cuts to make if you do decide to cut by a certain amount, but that question is orthogonal to the one I am discussing here.)

I also do not believe anybody else who says that they have a clear (and correct, with all the numbers included) argument in favour of one particular level of cuts. But we have to make decisions, so what do we do? The answer seems to be to decide on ideological grounds (if you’re left wing then you want to cut less and if you’re right wing then you want to cut more) and then to go off and find some professional economists who can back up your point of view, which is of course easy because there will be highly reputable economists on both sides. Is there any other way of operating? Probably not, but in my dreams I like to imagine politicians actually admitting the truth, which is that the current circumstances are full of huge uncertainties, and the result of those uncertainties is that they are forced to take hugely risky decisions that might turn out to be disastrously wrong. (Just to be clear, I’m claiming here that any response to this situation is hugely risky and potentially disastrously wrong. It might seem that cutting is the safer option, but if the price is a significant contraction of the economy that wasn’t actually necessary, then I call that disastrously wrong.)

PS I haven’t suddenly decided to turn this into a political blog. In fact, I started writing this post long before the AV referendum but wasn’t happy with it. I plan to get back to some more purely mathematical posts before long.

29 Responses to “To cut or not to cut?”

  1. F Pait Says:

    During normal times the compromise “how much one should inject X how much one should withdraw” does not have a simple answer. It is a numerical and political compromise. Different parties have different priorities.

    These are not normal times in the UK, or in the US. People and companies are so willing to save, and unwilling to invest, that they lend money to the government at essentially zero interest. The pro cut argument has no merit whatsoever. More cuts increase unemployment, and increase the deficit because tax revenue goes down and welfare disbursements go up. Right now the Conservatives are wrong, and Keynes is right. Period.

    • Jonathan Phillips Says:

      Keynes was writing at a time of deflation, when money unspent increased in value from day to day and individuals and companies had every incentive to save – government spending thus took the place of private spending in promoting economic growth.

      “During World War I the British pound sterling was removed from the gold standard. The motivation for this policy change was to finance World War I; one of the results was inflation, and a rise in the gold price, along with the corresponding drop in international exchange rates for the pound. When the pound was returned to the gold standard after the war it was done on the basis of the pre-war gold price, which, since it was higher than equivalent price in gold, required prices to fall to realign with the higher target value of the pound.
      The UK experienced deflation of approx 10% in 1921, 14% in 1922, and 3 to 5% in the early 1930s.” (Wikipedia)

      It was also a time of largely closed economies, when protectionism was rife – and public spending stayed at home and didn’t leak out. The UK had a massive manufacturing industry, well able to meet all domestic needs.

      We really could do with an economist of Keynes’s ability to advise us today – but I suspect that his advice would not be the same as it was in the ’20s and ’30s.

  2. Jonathan Phillips Says:

    The government borrows to maintain spending and thus (it hopes) to promote economic activity and increase its revenue by more than enough to service the increasing debt. This is the same as a small firm borrowing from the bank in order e.g. to buy capital equipment which will bring in enough extra revenue to pay the interest on the loan and eventually to pay off the loan itself. The government, however, doesn’t have to repay the loan (at least not in full) but can continue refinancing – as long as the increased economic activity and tax revenues will pay the interest.

    But it’s all a question of degrees and amounts. Spending cuts can be perverse, leading to both decreases in tax revenues and increases in welfare payments – and to a widening of the deficit.

    And spending increases don’t necessarily promote economic growth – in which case tax revenues don’t keep pace, the deficit worsens, etc. Why don’t they promote growth? If the UK is hungry for imports (not just of cars and computers but also of capital equipment – we don’t even build our own trains any more), then increased government spending will tend to promote economic activity abroad rather than at home – and a fall in the £ against the $, the € and the yuan. A falling £ doesn’t help though – if all we produce is crap we won’t be able to persuade foreigners to buy it however cheap it is (this was very much the position in the 1970s, when we still had significant manufacturing industry).

    Either way inflation looms – and we have to balance the need to keep interest rates high (to hold down inflation) against the need to keep them low (to help promote capital investment). A problem here is that the low interest rates needed by industry also encourage irresponsible consumer spending and rising house prices. Credit controls and the like might help, but that sort of thing is unfashionable.

    The answers aren’t simple. Sadly the politicians are.

    My own (cynical?) view is that the Tories want to cut services not just (or even chiefly) to cut the deficit but rather to demolish the apparatus of the state, so that even a genuinely left-wing government (were we ever to get one) would have nothing to work with.

  3. Joseph Says:

    One thing your post has not talked about is the distribution of capital. In a free market, consumers decide where they spend their money, and this allows the producers (and their suppliers, and their supplier’s suppliers and so on) which are most favoured to grow and invest. This is reflected in a real way with the purchase of land, buildings, machinery as well as the training of staff. The point to take from this is that the size of the economy has limited importance as a metric for how well it is serving the population. Bubbles, like in housing, mean that one part of the economy has been overgrown in relation to what is needed. (This is malinvestment.)

    Government spending is a demand upon the productive capability of the country. It uses resources, mainly labour, which could be used elsewhere. The government wage bill is part of the GDP (if you view that as the total of all wages paid in the country), so naturally cutting that makes it fall. Government employees go on to spend their money, so of course there is a multiplier effect so that the places where they spent their money will also lose out. This does not tell us whether the government spending has been beneficial or efficient. An analogy would be a widget factory churning out widgets which are not bought, but are immediately thrown away. Everyone in the country pays a subscription to the factory, but get nothing in return. This factory consumes real resources, and is a burden on the economy, but produces nothing useful. If it were to be shut down, its employees would lose their jobs and GDP would decrease. It could even harm other sectors which were especially used in the manufacture of widgets – if it contains microchips, microchip manufacturers would be hurt. The point of this illustration is not that the whole of government spending is pointless, but that part of it which is useless is not good just because it boosts total spending in the economy.

    You mentioned that cuts could lead to increased welfare benefits. If government employees lose their jobs, and go on to welfare, obviously the welfare payments will be less than their old wages, so there is a saving.

    • Doormat Says:

      Your example factory makes perfect sense, but it is a fiction. Suppose that people enjoyed watching the factory and particularly like the widgets being thrown away. Would the factory be “useless” then? If yes, then why is Opera (or Popular music, or any particular art-form you personally dislike) not useless. What if your factory produced plastic containers which are filled with a liquid available very cheaply from “taps”; people buy the bottles for their alleged convenience, and then throw them away. Is that “useless”. I view the bottled water industry as such, but clearly lots of people don’t. If I went “for a drive” on a Sunday afternoon (I actually “went for a walk”) would the use of petrol involved be the same as throwing away the widgets?

      In short, your example is fine. But then you write “…that part of it which is useless is not good…” How am I meant to decide which part of government spending is useless? I’m pretty sure the government does not, in actuality, fund factories which produce widgets which are immediately landfilled. I’m sure that there are corner cases where pretty much everyone would agree on the “uselessness” of the task; but I contend that these would be a fraction of a percent of government spending. Beyond that you *are* making a value judgement, and you *are* in the realm of politics not economics, which was pretty much the point Tim’s original post was making, I think.

      The general view of market economies, I thought, was that if someone was willing to pay for it, then the widget wasn’t useless. (And I’ve seen people, claiming to be economists, make exactly this point when e.g. arguing against environmentalist who would agree that “going for a drive” is useless). Once you start down the road of saying that some or other economic activity is useless, then I actually have a hard time seeing where you’d stop…

    • Richard Baron Says:

      One way to tell useful from useless activity is to ask whether people are paying for it, or would pay for it, out of their own pockets, entirely voluntarily. The fact that people freely buy things they plainly do not need shows that in a certain sense, those things are not useless.

      The difficulty with state spending is that we often cannot work out what people would freely buy. We cannot go by what they say, because saying and real spending are different things. Even if we could go by what people said, people’s incomes and needs vary too much to work out what would be a reasonable amount to spend on education or on healthcare (although we might ask what someone on mean, or median, income might spend). And some services are such that we would all automatically share in the benefit even if they were provided privately. Policing is an example.

  4. Robert Smart Says:

    We are going through a period where the inevitable decline in production of a crucial finite resource, oil, means that expanding economies is going to be difficult until we change infrastructure to electricity (and/or start producing a lot of synthetic liquid fuel). Yet conventional economists are oblivious to this. Krugman said recently that the oil crunch wouldn’t be a big problem since we only spend a small fraction (a few %) of income on oil derived stuff. This idea that money goes to oil or money goes to workers (etc) is wrong. The money flows through workers and on to other things. The money flows through oil purchases and on to oil rigs and oil workers etc. So a real understanding involves fluid flows, which we know has nasty discontinuities. So I’m convinced we desperately need a better understanding of economics to enable us to make sensible plans for all sorts of reasons: peak oil, climate change prevention, and the points raised here. So I’d encourage people to get involved in that. A Nobel Prize in Economics awaits :-). One place to start such a project might be John Baez’s Azimuth project, where a recent contributor claimed to find the Navier-Stokes equation pop up from noncommutative geometry. I’m not qualified to comment on that maths.

    • Observer Says:

      Politicians specialize in getting elected, so they are likely to have little clue about anything else, e.g., not so long ago UK and Switzerland were selling gold at ~$250 per ounce to buy bonds. Also, some problems are not fixed at their roots. E.g., the US switched much of their GDP growth to financial services due to many other industries closing or moving their operations out, leading to the current disaster. However, this is exacerbated by some currencies being pegged to the dollar, e.g., Japan’s very strong yen seems to result in their exports being quite smaller than Germany’s.

  5. Richard Baron Says:

    May I add some complexities to the problem?

    It is possible that a tough attitude that aims to eliminate the structural deficit within a few years will encourage private investment and business confidence, helping to counteract the adverse effects of a squeeze. That argument does get used. But putting a number on any such effect is of course impossible, either before or after the event.

    The question of how much to cut in total, and the question of what to cut, are not quite orthogonal. It may be that certain cuts would do less damage than others, for example by not cutting infrastructure projects that would in due course help business. A decision on the total may therefore be influenced by decisions on where to cut.

    In these discussions, we have to be very aware of what we regard as zero. Does a reduction in planned growth in spending count as a cut (taking the originally projected levels to be neither growth nor cut), or should we describe such a reduction as slower growth in spending (taking a continuation of current spending to be neither growth nor cut)? Of course this is only semantics, and we should ignore the rhetoric and get down to the numbers. But lots of other people in the debate will stick to the rhetoric, and one needs to be aware of that and explain to them why the discussion should be conducted in terms of the numbers. (Sticking to the numbers also saves us from getting bogged down in silly arguments about whether cutting the deficit means a cash cut, a real-terms cut or a cut as a percentage of GDP, although the argument about how our goal should be phrased, cash, real-terms or GDP, is not silly at all.)

    A more significant point about zero arises out of the objective of getting the structural deficit down to zero. If one does that, and then alternates deficits and surpluses with the cycle, then so long as we have real economic growth, the ratio of debt to GDP will gradually fall. How important is that goal? Should we allow ourselves to maintain something like the current ratio of debt to GDP? That would allow us to run small structural deficits for ever. And as Sellar and Yeatman said in 1066 And All That, “The National Debt is a very Good Thing and it would be dangerous to pay it off, for fear of Political Economy”.

    Finally, here are the Keynes v Hayek raps, for those who want to know what economics is really about.

    • Richard Baron Says:

      Whoops, my links to the raps have come out as black boxes that look like failed attempts to embed the clips. If they are black boxes for others too, then just go to Youtube and type Keynes Hayek Rap in the search box.

  6. Cardster Says:

    See “Zombie Economics (How dead ideas still walk amongst us)” by John Quiggan for other ideas that should have been killed off.

    There is irrefutable evidence that the UK government’s policies will make things worse rather than better for the average person; see the IMF intervention in any country. Also, such policies make the recession longer and deeper (as per the Republican and Conservative party policies during the Great Depression).

  7. Uwe Stroinski Says:

    Contrary to a football club a state owes the money to its people (with Greece and Portugal being obvious exceptions). If the state spends this money to the benefit of these people there shouldn’t be a big problem. If the state however spends the money on relatively expensive services and products of just a few of these people, then borrowing it back can lead to an arbitrage for them. Detecting and eliminating such behavior is hard and governments currently don’t receive a premium for doing it.

    Concerning mathematics. If I remember correctly, some 10 years ago you wrote a paper about the future of mathematics, automated proof assistants, automated provers aso. Do you plan to return to that line of thought?

  8. Spencer Says:

    I literally love this post. Very, very well said.

    In particular I for a long time have felt a certain “irritation at anybody who claims that the answer is obvious” to this issue and many other problems\issues discussed in the public domain, as it were.

  9. John Sidles Says:

    Let me agree with many that this is an outstanding post … and confess along with many to a certain “irritation at anybody who claims that the answer is obvious”.

    Our family’s experience has been largely conditioned by our sons adventures in the third world, both as a teacher in the outer islands of Micronesia, and as a US Marine in Iraq and Afghanistan.

    The challenges associated to these two circumstances were remarkably similar, and boiled down to two intimately united challenges: jobs and justice. What’s irritating about discussions that center upon economic policy is that these policies address job-creation only marginally, and justice not-at-all.

    For our family, what made the stresses associated to these enterprises bearable was their top-to-bottom narrative integrity: what I saw with my own eyes visiting my son in Micronesia, and what young Marines told us about the war, accorded closely with the sobering narratives of leaders in both the Peace Corps and the US Military.

    In recent decades, these institutions have given a central place to narrative integrity, and with regard to the practical realities of fostering justice and jobs, works that contain considerable food for thought include Neil Sheehan’s A Fiery Peace in a Cold War, H. R. McMaster’s Dereliction of Duty: Lyndon Johnson, Robert McNamara, the Joint Chiefs of Staff, and the Lies that Led to Vietnam, and WIlliam Slim’s Defeat into Victory (all of which appear on the USMC Commandant’s required reading list).

    For example, anyone who reads issues of Scientific American from the 1950s will be struck by the economic vitality evidenced by page-after-page of job advertisements. And this economic vitality was very obviously not founded upon the work of economists or politicians; rather it was the purposeful objective of mathematicians, scientists, and engineers … as is vividly described in Sheehan’s volume above.

    Obviously, everyone loves jobs and justice … mathematicians, scientists, and engineers just as much as politicians and economists. The point of this post, therefore, is that historically, mathematicians, scientists, and engineers have led the struggle for justice and jobs … not been passive spectators.

    And surely, the opportunities and resources available to the present generation of mathematicians, scientists, and engineers are considerably greater than those of any previous generation. So how sensible is it, to wait upon politicians and economists? It seems to me, that we need not wait, and should not wait.

  10. Nets Says:


    I like your post very much and think it addresses an important public issue that mathematicians should become involved in. We should become involved because what lies at the heart of the matter is rigor. There are clearly serious foundational issues in macroeconomics.

    In the US, the currently popular style of macroeconomics is “rational
    expectations.” This assumes that the world involves some randomness
    but that every agent in the world understands completely its stochastic
    properties and is trying to maximize the expectation of a utility function.
    [By the way, Gil Kalai and others have a lot to say about why maximization of expected utility might not be a good description of
    what human beings do. It boils down to the question: “does utility
    have units?” Von Neumann and Morgenstern said “yes”.] But rational expectations has problems that go far beyond reliance on expected utility maximization.

    I should back up and say that the rational expectations approach was an attempt to correct problems with the rigor of Keynes. Keynes’ model was
    basically static, and the followers of Lucas, who established rational
    expectations, wanted to take into account effects like people anticipating
    eventual raises in taxes that are likely to be associated with running
    large deficits in the present. In rational expectations, an optimization
    problem is solved iteratively to determine the outcome. This seems
    appealing as it suggests that there might be a way of choosing the right
    quantity of cuts.

    But this all has some weaknesses. Nobody really knows the utility functions of agents nor the properties of the production technology
    of the economy. Indeed even the existence theory starts to become
    problematic when you have more than one agent in the economy.

    In my opinion, the most problematic aspect of this kind of economic modelling is the choice of the random elements, the shocks. Every
    agent in the economy just naturally knows the stochastic properties
    of the world, but economists don’t. They just make something up.
    As an example, there is a respectable economics paper which investigates the question “Why was the economy of the US so good
    during the period in which Alan Greenspan was Fed chair?” The
    paper begins by proposing this might be caused in some way
    by stochastic volatility, rules this hypothesis out, and concludes that
    the real reason must be that during the Greenspan era, all the
    shocks were positive. A major open problem is that the rational
    expectations point of view doesn’t account for all the volatility
    in financial markets. (I.e. the volatility of earnings doesn’t seem to
    account for the volatility of share prices.) If I am permitted to say
    that anything seems obvious, it seems obvious that the reason
    for this is that the volatility is caused by the market participants
    themselves applying randomized strategies against one another.

    With all these difficulties, it is not surprising that economists do not
    take their own models too seriously as descriptions of reality, They
    are just models. If your model has nonlinear elements which you would
    like to study by linearizing, then the linearization is also a model. This
    is convenient because it obviates the need to estimate errors. Everyone
    can have a model, these models have different outcomes, and no one
    can say which really holds because systematic experiments are impossible.

    Given this state of affairs, one should be thinking not about the immediate question: how big should the cuts be, but instead about
    the more fundamental question: how should economics be done?

    At least, that’s my worthless opinion.


    • gowers Says:

      Thanks for this very interesting comment. It raises one of my favourite philosophical problems (which doesn’t seem to be as prominent in academic philosophy as it ought to be). In brief, the problem is how one should behave in the face of extreme uncertainty. To give an obvious and pressing example, what is the correct response to the threat of climate change? If we could say things like, “If you do A then the probability that B happens will be p,” then we could try to decide what we thought about the utilities of various As and Bs and come to a decision based on our attitude to risk etc. etc. Even this wouldn’t be a straightforward problem by any means, but it becomes a whole lot less straightforward when you don’t even know the values of p. Indeed, we often don’t seem to know even roughly the values of p. And yet we have to make policy decisions with very serious consequences. Bayesian methods probably help a bit, but the level of uncertainty is so high that I don’t think they solve the problem.

      What is the problem exactly? I don’t have a precise formulation, but a rough one is to determine what a perfectly rational person would do under such circumstances. Of course, it’s far from clear that the concept of perfect rationality makes sense under such circumstances, but perhaps the answer would be that there isn’t a unique rational course of action. The challenge, however, is to find a convincing model that can be analysed mathematically.

      Another example is if you have to decide whether to undergo a risky medical treatment. Even if you are given statistics, it is highly likely that they will be averages over some fairly general population and that you yourself will have properties that would change the probabilities if you conditioned on them. So although you may be given some probability to think about, you won’t actually be confident that it is the “real” probability. (By “the real” probability I mean something like that you line up a large number of parallel universes and have the operation in each one, and then see in which proportion of those universes you survive.)

      Another example that occurred in my lifetime was the beef/CJD scare in Britain. It was genuinely unclear whether it was rational to continue to eat beef (I did) in the light of the possibility of contracting CJD from it. Not even the order of magnitude of the probability of getting CJD was known. Luckily it turned out to be very small indeed.

    • John Sidles Says:

      It is interesting that John von Neumann made much the same points as the above posts by Nets and Gowers in an essay Can We Survive Technology? (1955) as follows:

      What safeguard remains? Apparently only day-to-day—or perhaps year-to-year—opportunistic measures, a long sequence of small, correct decisions. And this is not surprising. After all, the crisis is due to the rapidity of progress, to the probable further acceleration thereof, and to the reaching of certain critical relationships. Specifically, the effects that we are now beginning to produce are of the same order of magnitude as “the great globe itself.” … The most hopeful answer is that the human species has been subjected to similar tests before, and seems to have a congenital ability to come through, after varying amounts of trouble.

      To summarize and pull together these various points of view—extending from von Neumann and Turing’s generation to the present generation of mathematicians like Thurston and Gowers (and many others)—I have essayed a tribute both to mathematics in general and to Alan Turing in particular, titled Why does the world need mathematicians? and posted it to Dick Lipton’s weblog under the recent topic “Happy Anniversary to Turing’s Paper.”

  11. Ilan Says:

    There once was a man called Gowers
    with stout mathematical powers.
    “Support a big spending cut”,
    thus commanded his gut;
    yet his brain kept resisting for hours

  12. Antione Says:

    how about a balance between the two cut as many players you deem over paid and a liabilty as much as possible. Then spend to get some freah blood in there that may earn their money.

  13. proaonuiq Says:

    ¿To cut or not to cut?

    When lacking of good theoretical grounds, the best we can do in social sciences is to take decisions based on international comparisons. Coincidentally I´ve found a recent very interesting french report, which might help the discussion.

    It can be uploaded in PDF format at this webpage :

    at the end of the page, link “Telecharger le tableau de bord de l´emploi publique”. The data comes from OECD.

    I was looking for an international comparison of the average salaries in the the public service comparing to average salaries in the private sector. The most approximate and updated statistic I´ve found is in this repport. Now I understand why in my country, 90% of young people (according to polls) want to be civil servant. My recomendation to my country at present (this is also valid for Greece) is to increase public service employees while keeping or reducing the public servant salary mass. That is more cuts to present civil servants for paying new hirings.

    As the UK situation, my opinion, if I may, is that the public sector is still quite high by international standards. It is more close to the nordic model than to the neoliberal model, similar to the french´s situation. I do not know what to say about the nordic model. I do not find Norway as an example to follow, but Sweden seems to particularly doing well in innovation and knowledge-based industry and taking advantage of the economies of scale the globalizing trend offers. An interesting pre-crisis post about the nordic model here: In so small countries a few succesfull multinationals can make great impact in the economy. Bigger countries need more than a few multinationals…On the other hand it seems that nordic countries are losing positions in terms of GDP per capita PPP since 1970.

  14. Douglas Wade Says:

    It seems pretty obvious that one should try to control spending, and cut those things that do not particularly stimulate the economy, whilst at the same time avoid killing the goose that lays the golden egg. In fact, one must feed the goose as best one can afford.
    A sensible footie club would look to see where relatively painless savings can be made, negotiate hard on players’ wages whilst telling them they are marvellous and can beat anybody, and see if there is any money down the back of the sofa to bring in a decent centre forward for a few crucial matches.
    This is not a matter immeasurable uncertainties versus the necessity of action. This is common sense house-keeping, Tim.
    Anyway, I hope this finds you well. Do you see much of Monty & Imre these days?

    Douglas Wade

  15. mathinenglish Says:

    There always a fine line between overspending and over cutting.
    That’s where a good manager comes into play.
    I think that the answer is somewhat in the middle and also culturally biased. Different countries have different business ethics and rules. Some clubs rely heavily on state and government soft loans, where other clubs, from other countries, have to abide to a different set of rules.

  16. L. Says:

    “Suppose you owned a premiership football club that was losing money and not winning many matches. You would face the following dilemma: should you spend more money on players, in the hope of improving the team’s performances, and therefore revenue, or should you save money by selling players but risk being relegated, which, if it happened, would mean that your club would lose out on huge amounts of money from lucrative TV deals?”
    Are you talking about River Plate (Argentina)’s recent situation???? They chose second option…

  17. Octavio Says:

    Your crystal-clear explanation of macroeconomics is very enlightening. It replaces a lot of books on the subject.

    And I think that is why the contribution of mathematicians is so important in this regard.

    Thank you very much!

  18. Erik LaLone Says:

    What about dead-weight loss?

    The answer is obviously neither spending nor cutting.

    The government should try to smooth spending over time, to create a stable atmosphere for business.

    Trying to spend or cut your way out is like chasing the market you might catch it once in a while but in the long-run you’ll probably lose out to a dollar-cost averaging strategy.

  19. MonkeeRench Says:

    Nets said: “Why was the economy of the US so good
    during the period in which Alan Greenspan was Fed chair?” … stochastic volatility, rules this hypothesis out, and concludes that the real reason must be that during the Greenspan era, all the shocks were positive.”

    Hardly all — the worst shock was the 1998 failure, collapse, and unwinding of Nobelists’ Long Term Capital Management, whose loud forecast of the collapse ten years later was widely ignored under corrupt regulation and foundationless financial mathematics.

    Surely there is a Silent Majority of mathematicians who KNOW the basis of current contingency claims financial instruments (e.g. high risk derivatives, credit default swaps, etc) is fatally flawed and that their continued acceptance presages further disastrous shocks (“Black Swans”). Where is the OUTRAGE of professional mathematicians who wink and nod away such travesty while the economic structures crumble as institutions and governments just helplessly struggle with what they can’t possibly understand?

  20. Neal Says:

    The word “structural” doesn’t appear anywhere in your analysis and I can’t work out why. It’s the fact that we have a STRUCTURAL deficit – i.e. we’d still be running a deficit even if the economy was in a period of strong growth – that is very rationale for the cuts in the first place.

    I actually think you are being a little unfair to the politicians (of all mainstream hues). It’s true that we do get some buffoonery from the back-bench rent-a-quotes, but the front benches are pretty much singing from the same hymn sheet. They all believe that cuts are necessary and that all believe must be implemented in a way does as little damage as possible to an already fragile economy. They just disagree about how fast and how far-reaching the cuts should be. It’s actually the public and not the politicians that are most guilty of “discretizing” the debate into a yes/no question.

    The one thing that ought to be “obvious” in this debate is whilst it might be possible to stimulate growth through spending in the short term, it’s simply not a sustainable strategy in the longer term. The plain fact of the matter is that the public sector doesn’t generate a single penny of wealth. Real growth, therefore, has to ultimately be underpinned by growth in the private sector.

    • gowers Says:

      Thanks for the extra technical term — the fact that it wasn’t in the post is a reflection of my ignorance.

      I’m not sure I agree that the front benches are singing from the same hymn sheet. Osborne, for instance, says that if we followed the suggestions of Ed Balls then our credit rating would worsen, with disastrous consequences for interest rates. Meanwhile, Ed Balls says that if we continue with the policies of George Osborne, then growth will be stifled to the point where the cuts are counterproductive. So although in a sense they just differ about the pace of cuts, it shades into a more qualitative debate.

      To put that slightly differently, to listen to Ed Balls speak, you could quite reasonably assume that he thinks the following. We are in a recession, and in recessions you should borrow to stimulate growth. If we do that, then in due course the world economy will improve and we won’t have crippled ours. Yes, our national debt will be higher, but once things have got better we’ll be in a much better position to start paying it off, needing much less savage cuts to do so. The potential problem with that argument (which I may not have put optimally), it seems to me, is that if the numbers don’t quite work out then it leaves us in very big trouble.

  21. Neal Says:

    “Thanks for the extra technical term — the fact that it wasn’t in the post is a reflection of my ignorance.”

    There are some that don’t accept the very concept. I half wondered whether you counted yourself among them. They argue that there are simply large deficits and slightly less large deficits. I can seen their point – it’s very difficult to calculate how much of a deficit is should be attributed to the economic “weather” and how much is down to excessive spend. To do so requires that we come up with an imaginary norm for growth against which we can compare the real-world position.

    I was probably overstating the case when I said that mainstream politicians were all singing from the same hymn sheet. But I do think they appreciate – better than the public at least – that we cannot simply go on borrowing at the rate we have been. Otherwise – sooner or later – we’ll start to look like a credit risk.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: