Government spending

Before, during and after this month’s budget, Treasurer Joe Hockey sounded dire warnings about Australia’s “budget emergency”. Amidst this fear-mongering, it was a pleasant relief to come across a dissenting view. In a recent interview on 2SER Dr Stephanie Kelton (Department of Economics at the University of Missouri in Kansas City) argued that the government budget is very different from a household budget, however appealing that analogy might be. Governments like the Australian government, with its own free-floating currency can spend more than they take in taxation without worrying about running out of money. While the economy is weak, the government can comfortably run a deficit. The constraint to worry about is the risk of  inflation, which means curbing spending once the economy heats up.

I posted a link to Facebook, and immediately drew comment from a more conservatively libertarian-minded friend: “of course a deficit is a bad thing!”. Pressed for an explanation, he argued that government spending was inefficient and “crowded out” more productive private sector investment. This did not surprise me. Deep down, the primary concern of many fiscal conservatives is government spending itself, not a deficit. This is easy to test: ask them whether they would be happy to see the deficit closed by increased taxes rather than decreased spending. The answer is generally no, and helps explain why so many more traditional conservatives are horrified by the prospect of the Coalition’s planned tax on higher income earners….sorry, “deficit levy”.

From there, the debate deteriorated. North Korea was compared to South Korea as evidence of the proposition that government spending was harmful, while a left-leaning supporter asked whether this meant Somalia’s economy should be preferred to Sweden’s. Perhaps foolishly, I proffered a link to an academic paper (on the website of that bastion of left-wing thought, the St.Louis Fed) which presented a theoretical argument to the “crowding out” thesis. My sparring partner then rightly asked whether the thread was simply becoming a rehash of the decades old Keynes vs Hayek feud, a feud best illustrated by Planet Money’s inimitable music video.

Macroeconomic theory was never going to get us anywhere (as I should have known only too well). Instead, the answer lay in the data, with more sensible examples than North Korea and Somalia. Aiming to keep the process fair, avoiding the perils of mining data until I found an answer that suited me, here was my proposal:

I’m going to grab a broad cross-section of countries over a range of years and compare a measure of government expenditure (as % of GDP to be comparable across countries) to a measure of economic success (I’m thinking GDP per capita in constant prices).

If indeed government spending is inherently bad for an economy, we should see a negative correlation: more spending, weaker economy and vice versa. My own expectation was to see no real relationship at all. In a period of economic weakness, I do think that government spending can provide an important stimulus, but I do not think that overall government spending is inherently good or bad.

The chart below illustrates the relationship for 32 countries taken from the IMF’s data eLibrary. To eliminate short-term cyclical effects, government spending and GDP per capita (in US$ converted using purchasing power-parity) was averaged over the period 2002-2012.

Govt. Spending vs GDP

The countries in this IMF data set are all relatively wealthy, with stable political structures and institutions. All but one is classified as a “democracy” by the Polity Project (the exception is Singapore, which is classified as an “anocracy” due to an assessment of a high autocracy rating). This helps to eliminate more extreme structural variances between the countries in the study, providing a better test of the impact of government spending. Even so, there are two outliers in this data set. Luxembourg has by far the highest GDP per capita and Mexico quite low GDP per capita, with the lowest rate of government spending.

The chart below removes these outliers. There is no clear pattern to the data. There is no doubt that government spending can be well-directed or wasted, but for me this chart convincingly debunks a simple hypothesis that overall government spending is necessarily bad for the economy.

Government Spending vs GDP per capita

Now look for the cross (+) on the chart: it is Australia (IMF does not include data for New Zealand and we are the sole representative of Oceania). Despite Hockey’s concerns about a budget emergency, Australia is a wealthy country with a relatively low rate of government spending. Among these 30 countries, only Switzerland and South Korea spend less. These figures are long run averages, so perhaps the “age of entitlement” has pushed up spending in recent years? Hardly. Spending for 2012 was 35.7% compared to the 2002-2012 average of 35.3%. The shift in the balance of government spending from surplus to deficit is the result of declining taxation revenues rather than increased spending. Mining tax anyone?

Possibly Related Posts (automatically generated):

16 thoughts on “Government spending

  1. Zebra

    I am not surprised by this result. While it may be a bit of a simplification I imagine that a lot of discretionary Government spending goes to welfare (lower income people, middle class families and pensioners) and gets spent on stuff we all need to live whether that is food, flat screen tvs or nappies (depending on your current age the last one is variable).

    However if the government lowers taxes at the top end then the top earners spend their extra income on investments: pushing up house and share and bond prices. More money chasing the same investments, prices rise, yields fall, but exactly the same income is distributed amongst the same people who already own them. You might argue that this lowers the cost of capital and encourages new businesses but the primary driver of new business investment is confidence in consumer spending, not interest rates (except when they are very high which hasn’t happened for almost 20 years). Also if the Govt doesn’t run as much of a deficit then bond yields will also fall which will reduce the cost of capital so income for these people (the very rich and self-funded retirees) will increase but will they spend it on more…stuff?

    So if you really want to boost the economy then increase government spending on welfare and if you are going to have a tax decrease do it for the poor and middle class, not the rich.

  2. Stubborn Mule Post author

    @Zebra: I agree that in many ways the more interesting discussion is not about the aggregate spending (or revenue) position for the government, but the composition. I think that (almost) everyone can agree in principle that some forms of government spending are more efficient or of greater utility to the community than others. Rather than obsessing about slashing everything, we should focus on identifying those areas that provide the greatest benefit and supporting or increasing the spending there, while always challenging the less useful expenditure. Of course that is too nuanced and complicated for headlines and soundbites!

  3. Ken

    So if borrowing doesn’t hurt and it is so good, then why don’t we borrow more? Somehow Japan has got away with it, but that may well be because they have large businesses that have been happy to loan them the money. What will eventually wreck your figures is the drop in GDP of high debt countries as they find they can no longer borrow. GDP seems to be the most flawed measure of anything, but it doesn’t seem to stop people from using it to prove arguments. Anything that creates debt seems to produce higher GDP.

    My prediction is that in 20 years time public debt in Australia will be at least 60% GDP, Household 100%+, business 50% and we will be complaining that our economy ant function because nobody can borrow cheaply.

  4. Stubborn Mule Post author

    @Ken: I didn’t say borrowing is good. I said that government spending wasn’t bad and that in a country like Australia we shouldn’t worry about running a deficit during periods of relative weakness. Typically rising government debt is a symptom rather than a cause of problems. The high levels of government debt in Japan was the result of extended economic weakness, eroding taxation revenue. It was not the cause of the problem. Also, for countries like Australia, USA, Canada and UK with their own currencies which are free-floating, funding government debt is not a problem and rising government debt does not lead to increase funding costs for the government. Look at how low Japan’s interest rates have been for many years. Despite the loss of its AAA credit rating in August 2011, US treasury yields have been at the lowest levels in years.

    This comment is verging on becoming a post in its own right. Perhaps I should just point you to this post from 5 years ago.

  5. Ken

    It seems that we’ve decided to ignore the lessons of the GFC which was that while private debt was rising, everything looked fine, it was only when it started falling that things fell apart. We’ve now switched to driving economies with public debt and hoping that nothing will go wrong, and while other economies have got away with it, I wouldn’t like to count on it. At some stage it becomes obvious to anyone holding your currency that eventually you will start printing it. Some countries will get away with it, others won’t.

  6. Stubborn Mule Post author

    @Ken: there are some crucial lessons from the financial crisis, but worrying about Australia’s government deficit was not one of them. Changes in net government debt and private sector debt tend to go in opposite directions; the balancing item is the foreign sector. For both sectors to save on a net basis, the country has to run a current account surplus (quite unlikely for Australia!). The rapid rise in public sector debt in Australia was closely associated with the period of government surpluses. As I argued in the post linked to above, for Australia I worry far more about increases in private sector debt that public sector debt. At this point it’s important to emphasis the role of Australia’s sovereign, free-floating currency in arriving at that position. European sovereigns do not have sovereign control over the euro: that has been ceded to the EU. So, the position of European sovereigns is far closer to that of a corporation or household and very different from that of a country like Australia, the US or Japan.

  7. Stubborn Mule Post author

    @Tristam: sorry, your comment was held up in the moderation queue while I was having problems with the site. Published now!

    As for that link, I am actually surprised that the correlation is not higher! I deliberately chose to analyse outright levels of spending and wealth rather than changes. There are good reasons to expect to see correlations in short-term changes (say up to 3 years), but with causation running from GDP to spending not the other way around. In a developed economy like the US or Australia, much government spending acts as an “automatic stabiliser”, responding to changes in the economy. When the economy stalls, unemployment increases and transfer payments rise, while profits fall and tax revenues falls as a result. As the Cato chart shows, the relationship is very noisy, with other factors bumping things around too, which is why I based my analysis on a longer-term average of the stock to eliminate the short-term effect of automatic stabilisers, performing the comparison across countries rather than through time.

    P.S. I will change the label to libertarian!

  8. Tristam Smyth

    A few points:

    1) I didn’t see your power calculation? What was the 95% chance that you were able to show a 5% difference?

    2) I think an analysis based on change in GDP growth subsequent to a change in government spending would answer the question more accurately.

    3) In defence of the North Korea South Korea analogy:

    My premise was the history of the 20th century demonstrates the comprehensive success of market based economies over government run economies. There were a few points in time where this illustration works extremely well, when countries with similar people and resources are divided along a political line to a predominantly market based approach to a predominantly government run approach. In 1953 North and South Korea were divided along this way, 60 years later, the difference in the living standards is extreme. There were, of course other factors involved. Similar situations with similar results occurred in East and West Germany, Singapore and Malaysia, Hong Kong and China, Western Europe and Eastern Europe. There are probably other good South American and African examples as well.

  9. Zebra

    The problem with all these technical arguments about the appropriate level of government debt is that they are all based on classical economics. Whatever side you take. I think a lot of the problems of government debt are better explained by agency theory. Politicians are “agents” (independent units who really act in their own best interest whatever their job title says) and don’t necessarily act in the best interests of “shareholders” aka us. A classic example is where the CEO of a company may recommend accepting a merger or takeover offer, claiming it is in the best interest of shareholders whereas in reality the decision is made because it will lead to her or him being the CEO of a much larger company (with a bigger salary package); or receiving a generous termination payment.

    Similarly the PM or Treasurer may choose to spend more because it improves their image in the eyes of their constituency, or cut the budget and spend less, for the same reason. Canny shareholders recognise this and so try and align the CEOs interests with their own through bonuses based on share price or profits. So maybe we need to find a way to align politicians interests to our own. The problem is that whereas most shareholders all want the same 2 basic things (dividends plus share price growth) voters are a disparate bunch. The list of what we want from governments would probably run into a dozen or more pages. So I think we get the government we want. In which case the only really important thing is that (a) political parties tell the truth about their intentions and (b) we act in our long term rather than short term interest.

  10. Zebra

    I should also add if you are of a conservative or libertarian frame of mind to go to the “Ramp Up” website (obvs a lefty ABC disability website, defunded in the latest budget)

    I am an occasional commenter (I am disabled) challenging some of the professional disability advocates who contribute what I consider to be dubious economic claims about lack of government services and spending, particularly about the cost and benefits of the NDIS.

    But I suggest going to the comments sections and reading those from disabled people and their carers and then see what you think. It is quite appalling we live in Australia and there are people who live like this. But enough of me please go and read them and tell me what you think?

  11. Stubborn Mule Post author

    @Tristam: 1) power calculations are a bit subtle for regressions and I don’t think your question as it stands makes sense. 5% difference from what? In this scenario we’d be looking for a non-zero regression slope and the probability of a false negative (1 – power) will depend on what size effect is considered to be a true effect. The bigger the slope required the high the power as it is easier to detect. What I can give you (when I am back in front of the computer) is the p-value associated with the slope.

    2) Working with changes is still fiddly. National account statistics are crude beasts so you would want a reasonable lag to be sure you’re not blurring causality, but if you make the lag too long you’ll just pick up the reverse swing of the cycle. If we assumed A caused B and both were cyclical, a lagged regression may lead you to conclude that B caused A! No wonder, as noted in the world bank article in your other comment, the literature covering the study of the links between government spending and growth is vast yet inconclusive.

    That article is more cautious than the Cato piece about its correlation conclusions, but I do think there are further factors not addressed. For example, I used GDP per capita to adjust for population effects. I would like to re-do the World Bank numbers on that basis to There were, of course other factors involvedensure their figures are not skewed by the low populations growth of Western Europe. Also I remain unconvinced that the stage of development is adequately covered. There is plenty of evidence that productivity gains slow (Australia would never expect China level growth rates). Other studies, including the World Bank one, observe that wealthier countries tend to spend more on welfare (there is a hint of that in my chart with Mexico’s position) so what the World Banks sees across Europe could easily be correlation not causation.

    3) My initial argument, as Danny notes, was not that we should change our political regime, merely that, in a modern economy with strong institutions and a free-floating sovereign country (so not North Korea, not Zimbabwe and, in fact not even euro-zone countries) we should not worry about the level of government spending per se (unless the economy is overheating, so the spending could add to inflation). As James notes, we’d be better off thinking about where that spending should be directed.

  12. Stubborn Mule Post author

    @Tristam: statistical measures from a linear regression of GDP per capita versus government spending: the p-value of the slope of the regression is 0.39 (so there is really no reason here to reject the hypothesis of slope of zero). Eliminating Mexico and Luxembourg only serves to increase the p-value to 0.71. For what it is worth (very little!), the fitted slope is 0.2: a positive relationship, so higher government spending tends to be associated with higher wealth (but of course correlation is not causation, in either direction!). A 95% confidence interval around the slope is -0.3 to 0.8. Excluding Mexico and Luxembourg the fitted slope is 0.1 and the 95% confidence interval is -0.4 to 0.6. So, the statistics reinforce the conclusions drawn from inspection of the charts: there is no convincing relationship there.

  13. Stubborn Mule Post author

    @Zebra: read through some of Ramp Up and comments…a sampler should be compulsory reading for all policy-makers. Certainly the idea that the only thing stopping people getting work is the DSP is simplistic (I choose that word simply because it is polite).

  14. Zebra

    thanks Sean. It is very easy to challenge other peoples left/right wing views online but when you get the chance to read the genuine statements of disabled people and their carers it is quite moving (by which I mean it literally moves or changes your view). They are not the usual lefty commentariat on Twitter or CiF (Comment is Free, the Guardian comment website), they are not very articulate and they sometimes use bad grammar and spelling. Like that’s a bad thing.

Leave a Reply