Measuring Leviathan redux: Public Spending Myths (furthering the debate, I hope)


In a previous post – Measuring Leviathan: Big Government and the Myths of Public Spending – I tried to explain and explore some of the mythology that has grown up around public spending and – probably more importantly – put forward some ideas about how we ought to think about public spending. I used the past 50 years or so of UK public spending to illustrate my points. It especially showed some things that people generally find very surprising about the last Labour government.

An academic colleague posted a response, which I quote in full below. I have chosen to respond in full because this comment rather helpfully illustrates many of the problems I was trying to clarify. So apologies to my colleague if this seems like an ‘attack’ (or strictly speaking ‘counter-attack’) – it is meant in a constructive way and to further the debate. Anyone else is welcome to join in.

Here is the comment:

“Many things can and are claimed about the last Labour government but I think it is difficult to sustain the argument that its public spending was lower than any government since the 1960s. Figure 4 is not very accurate as it averages very different levels of spending within a government. What is clear from your figures is that in real terms spending under Labour was much higher and in some ways that is more illustrative than percentage of GDP which is dependent on the growth of the economy. Also as figure 3 shows, spending as a percentage of GDP was on a downward trend under the Conservatives but on an upward trend under Labour. So Labour’s finishing point in 2010 was much higher than the Conservatives in 1997. Labour shackled themselves to Conservative spending plans for two years in 1997 and so in the context of a growing economy spending was bound to go down as a percentage of GDP but after the first two years spending increased rapidly, and as you say, especially in health and education. Unlike previous Labour Governments, after 1997 New Labour were not constrained by the danger of a run on sterling.”

Let me deal with all these points, one at a time.

Many things can and are claimed about the last Labour government but I think it is difficult to sustain the argument that its public spending was lower than any government since the 1960s.

The crucial phrase here is “lower than any government since the 1960s”. The question is simply, by what measure? As I pointed out in my original blog post, if you measure public spending in either nominal or real terms then almost any government can be shown to “spend more” than the last one. My argument is that, despite it’s short-comings, comparing public spending with a measure of national wealth (I choose GDP) is the only sensible way of studying long-term and comparative trends in public spending. If there is an alternative – propose it, rather than just reject something that seems counter-intuitive.

‘Seems’ is the important word – there is a widespread belief, spread by New Labour themselves, their political opponents, and the media, that the last Labour government spent a lot – too much or the right amount, depending on who you are – on public services. But a belief, even a widely held one, is not necessarily reality. What my analysis showed was that – in historical terms New Labour, prior to the Global financial Crisis, did not spend vastly more on public services as a proportion of national wealth. That is what the numbers show, quite clearly, whether or not some people “believe” otherwise.

Figure 4 is not very accurate as it averages very different levels of spending within a government.

Firstly, Figure 4 is completely accurate. What you seem to be challenging is not its accuracy but its utility. Because it “averages very different levels of spending within a government”, you seem to be saying, it is not really very useful. Actually, what it averages is spending as a proportion of GDP, which is rather different, but let’s assume that was just a slip. What is the use of averaging public spending as a proportion of GDP for a whole government? Well, I think Figure 4 rather makes the point – it shows that, in one respect at least, some of the popular mythology about governments over the past 50 years is highly inaccurate. Labour and Conservative governments were not that much different, and when they were it was in often unexpected ways. The story is thus obviously far more complex than the simplistic Labour equals big spenders, Tories equals small spenders, myth. There may be very good, and very interesting explanations of why the numbers appear as they do – and that is the purpose of ‘good’ numbers in my view – to act as a ‘can-opener’ to asking more penetrating and interesting questions.

You are correct that average spending over a government can hide important facts. For example, a government that started out with very high spending and ended with very low could have the same average spending as a government that started out with ‘average’ spending and continued like that throughout its term in office. That is why we need to look more closely at what happens within a governments’ period of office – which I have done frequently elsewhere and will do more of in my book ‘Strategy in Government – from growth to austerity’ (out next year, hopefully).

But that doesn’t negate the utility of at least looking at governments’ spending as a whole during their term in office – and the best measure of that is their average spending as a proportion of GDP. Again, if there is a better one, please propose it.

What is clear from your figures is that in real terms spending under Labour was much higher and in some ways that is more illustrative than percentage of GDP which is dependent on the growth of the economy.

I’ve already answered this, but to reiterate: virtually any governments real terms spending will be higher, on average, than the last ones. I already made the point about GDP, growth and its effects on spending/GDP ratios in the original post, but in my view that doesn’t negate it’s utility as an analytic device.

Also as figure 3 shows, spending as a percentage of GDP was on a downward trend under the Conservatives but on an upward trend under Labour. So Labour’s finishing point in 2010 was much higher than the Conservatives in 1997.

I hate to point out the intellectual inconsistency of saying in the previous sentence that real-terms spending is a better measure than the spending/GDP ratio, and then relying precisely on the latter to make your very next point.

You also ignore your own (correct) point about the spending/GDP ratio being dependent on both spending and growth (or recession). New Labour’s higher spending/GDP figure in 2010 was due almost entirely to the GFC effects on the wider economy, not to any sudden increase in spending. True, Labour could have chosen to immediately slash spending in line with the recession in the private sector (i.e. by about 6% of GDP). I leave you to make your own estimates of the effect that would have had.

However, and here is my central point, Labour’s pre-GFC figure for spending/GDP is below the long-term average and, even while growth in the economy was continuing, had levelled off in 2004-07.

Labour shackled themselves to Conservative spending plans for two years in 1997 and so in the context of a growing economy spending was bound to go down as a percentage of GDP but after the first two years spending increased rapidly, and as you say, especially in health and education.

This just reiterates a point I have made frequently on Whitehall Watch, in repeated evidence to the Treasury Select Committee, and elsewhere. Again, I’ll be exploring this further in my new book.

Unlike previous Labour Governments, after 1997 New Labour were not constrained by the danger of a run on sterling.

True. But I’m not sure what you’re point is with regard the previous discussion?

I hope this rather long response serves its purpose: which is to try and explain, and explore, what sorts of measures we should be using in the debate on public spending. I realise I haven’t covered lots of issues (taxes, borrowing, etc) which I will do (and have done) elsewhere. My main mission is to try and change the terms of debate away from impressionistic, and often misleading, myths towards more clear analysis based on the actual facts about public spending in the UK and elsewhere.

Advertisements

About Colin Talbot

Professor of Government at the University of Manchester, England.
This entry was posted in Whitehall. Bookmark the permalink.

5 Responses to Measuring Leviathan redux: Public Spending Myths (furthering the debate, I hope)

  1. MikeF says:

    Colin,

    To a relative novice on such matters your deconstruction of your colleagues statement and the explanation of your own rationale and base of evidence used to support your own position is extremely helpful and, for me at least, provides clarity as to your opinion.

    Although at first surprised by your comment that Labour did nothing to disabuse the electorate that they were high spenders I can now understand why , as a democratically elected body, they chose to take this position

  2. Martin Smith says:

    Colin
    Thanks for your interesting and important response to my comments.
    I think that there are several issues with what you say.
    One of the major issues is that what data is used to some extent depends on the question that is asked. Spending as a percentage of GDP clearly answers some questions but also it can give a false impression because it is dependent, not only on what governments spend, but on the size of the economy. Average spending as a percentage of GDP is an artefact of the level of spending at the start of a government and the growth or decline of the economy and says little about government intentions. What is clear about the Conservative governments of 1979-97 was that the trend in spending was downward whilst for Labour the trend was upwards. Government spending as a percentage of GDP under Labour also started increasing long before the economic downturn. Whilst Government spending as a percentage of GDP has increased under the coalition, their intention is to cut spending and the growth is a consequence of the economy shrinking and demand for welfare spending increasing in an economic downturn.

    If you look at real terms spending, as you say, it tends to increase across governments. However, there is little doubt that under New Labour there was a significant upturn in real terms spending that started in 2000 as soon as the commitment to Conservative spending plans ended. Labour increased spending well above trend and indeed increased real terms spending by a 1/3 across their government. No previous government has increased spending to such an extent. Indeed, despite what you say about real term spending increasing government on government, your figure 2 shows, that from 2010 real term spending declined.
    You write: ‘What my analysis showed was that – in historical terms New Labour, prior to the Global financial Crisis, did not spend vastly more on public services as a proportion of national wealth’. But this is only partially true. Labour spent much more on health and education than previous administrations (both in real terms and as percentage of GDP). Indeed, Blair made a commitment to increase spending as a percentage of GDP on health. As the ISF points out: ‘Spending on public services has increased by an average of 4.4% a year in real terms under Labour, significantly faster than the 0.7% a year average seen under the Conservatives from 1979 to 1997. This is largely due to increases in spending on the NHS, education and transport.’ Again the IFS shows that under Labour, the UK went from having 22nd highest total government outlays in the OECD in 1997 6th in 2010. Again, in comparison to previous government and governments internationally, it is hard not to come to the conclusion that Labour did significantly increase spending. So, it may be possible to make some sort of argument that spending as ‘a proportion of national wealth’ did not increase but this does not, in my view, give a very accurate interpretation of the way Labour supported spending on public services. Again the figures show when Labour came into office overall spending was 39.9 per cent of GDP and when they left it was 48.1 per cent. Only some of this is the result of the economic downturn.
    It is right to contextualise these figures and develop different interpretations but I think it is hard not to conclude that Labour after 2000 made a step change in the amount spent on public services and indeed spent more than either previous Labour or Conservative governments.
    The data is actually open to different interpretations and it does depend how it is disaggregated. However, it is hard to question Labour’s spending on health and education.

  3. Colin Talbot says:

    Dear Martin,

    I am afraid your analysis does not stack up. You seem to be trying to make the myth that New Labour spent more stand up, which is simply not supported by the evidence.

    You say that “for Labour the trend was upwards. Government spending as a percentage of GDP under Labour also started increasing long before the economic downturn.” This is incorrect.

    The trend for New Labour went through four clearly discernible phases:

    I) Prudence (1997-99) when Labour continued with Tory spending plans (which even the then Tory Chancellor Ken Clarke has admitted he never intended to follow thru on). The result was the lowest level of spending as a percentage of GDP in the past 50 years.

    2) Expansion (2000-2004) when Labour did steeply increase public spending, especially on health and education. But, crucially, not as rapidly in other areas. (Indeed labour tried to expand so fast that at one point they had an £8bn underspend). But at the end of this 4-5 year period total public spending was still, as a proportion of GDP, less than the long-term average. Spending on health especially was much higher (around 8% of GDP, up from about 5%) – but TOTAL spending remained below the long-term average.

    3) Consolidation (2004-2007) when Labour eased off on the rate of expansion and spending actually marginally decreased as a proportion of GDP.

    What is important to note about these first three phases was that growth in the economy was pretty consistent – so changes to spend/GDP ratios are almost entirely due to spending decisions.

    4) Stimulus (2007-2010) – Labour actually continued with their less expansionary spending plans between 07 and 2010 – so why call this “stimulus”. Because after 2007 the economy tanked, but Labour continued to spend at the same rate as a policy decision – to try and off-set the recessionary effects of the GFC. (Alistair Darling’s account is a useful source for the “why” of spending in this period).

    The real story of New Labour is thus much more complex than a simplistic “New Labour spent lots” narrative. I should also just add that between 1997 and 2007 Labour never exceeded 40% GDP debt or 3% of GDP deficit, the rules it has set itself. Again, not what the “received wisdom” of current mythology says.

  4. Richard White says:

    Colin,

    I’m no expert in these matters but I know enough about statistics to be sure that you can make a case for just about any interpretation of raw data if you have a mind to. I’ve read your and Martin’s interpretation and I think Martin has the stronger argument.

    It seems to me that what is relevant is the trend in the statistics established during the tenure of a Government. Any analysis should take into account the time necessary for an incoming administration to have an effect – I guess 1 – 2 years. I think this would yield a very different picture from that which you represented in your original article.

    Your final point, about Labour never breaking their own fiscal rules only stands up because you have been selective with dates. I suggest that far from being mythology, “received wisdom” is actually a pretty good representation of Labour’s financial mismanagement whilst in power.

  5. Hi, the crucial figure here is government debt. There are some good graphics at http://pol-check.blogspot.co.uk/2011/05/worst-governments-since-first-world-war.html which give the actual public spending and debt figures.

    Does public spending aid an economy? If you look at the history of the UK economy you can see that increasing the rate of public spending growth is always a bad idea. This effect is called Sydenham’s Law – see http://pol-check.blogspot.co.uk/2013/01/sydenhams-law-of-public-expenditure-and.html

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s