Silly charts and bad economics
A short while ago, the esteemed Adam Smith Institute (ASI) produced this chart as part of a post from Sam Bowman:
I criticised the chart in this post on four grounds:
- the figures were not adjusted for inflation
- the figures were shown in Euros, which meant that the UK's spending in 2009 was overstated because of the devaluation of sterling
- the figures were not quoted in relation to the size of the countries' economies
- there was no allowance for cyclicality (automatic increase in government spending as benefits bills increase in economic downturns due to unemployment and wage cuts).
And I produced a lot of charts of my own showing that when the above are taken into account, the conclusions of Sam Bowman's post - that there hadn't yet been any serious spending cuts and there was far worse to come - were only partly justified.
So what did the ASI do? They issued the same chart AGAIN in a different post, by Vuk Vukovic. And he produced from it an even more mistaken analysis.
I pointed this out in a comment on the blog, and to my amusement the ASI then replaced the chart with this one:
Spot the difference? Yes - these are real rather than nominal figures. But the chart is are still in Euros and we don't know if allowance has been made for translation differences for the UK, there is still no attempt to relate the spending figures to economic output (GDP) and it still makes no allowance for cyclical factors. So this chart is no more meaningful than the previous one.
Showing absolute figures without reference to the size of the economy gives a completely misleading impression. To help make my point, here's a silly chart of my own:
This chart shows European government debt absolute figures, irrespective of GDP - in Euros, so no allowance for sterling depreciation. Who exactly has the biggest debt pile? Yup, that's right. Since 2009, Germany. And Greece has the lowest (of these countries, anyway). So if we take absolute figures only, ignoring the size of the economy - which is what the ASI does with their chart - then it should be German debt on which yields are heading for the moon. It should be Germany facing default and exit from the Eurozone. It should be Germany facing sanctions and fines. Shouldn't it?
But when you plot European debt in relation to GDP - as it is usually shown - the picture changes completely:
Germany's debt pile looks - er, quite large, at 82% of GDP, which is well above the Maastricht convergence criteria. In fact it looks about the same as France's and the UK's. Spain's debt/GDP is actually lower at the moment, though I reckon that will change radically when it is forced to bail out its banks and its regions, as will happen pretty soon. But Germany's debt certainly isn't the largest in the Eurozone when you compare it to GDP. That honour, surprise surprise, belongs to Greece.
In fact the debt to GDP chart is no more sensible than the absolute debt chart. Debt is accumulated deficits over years, whereas GDP is an annual figure; it could therefore be argued that quoting government debt in relation to GDP compares apples and oranges. It certainly tells you absolutely nothing about the ability of the assets of the country to support that level of debt - which is the country's solvency. What would arguably be better would be to map the COST of debt - interest payments and refinancing - in a given year against GDP. That at least would give some idea of the ability of the economy to service the debt. However, I digress.
To be fair to the ASI, their chart does show that governments in Europe generally are increasing their spending rather than reducing it - which is the point they were making. But it isn't that simple. My final objection to their graph is that it takes no account of cyclical factors. All of the countries in the ASI's graph are in recession at the moment: Greece's recession is the worst in recorded history. When people are losing their jobs and experiencing wage cuts - which is what happens in recessions - the benefits bill automatically increases, so Government spending increases as the economy contracts. So there may be real cuts in government spending in other areas, but the effect of these on the total Government spending bill may be wiped out by the automatic increase.
Attempting to reduce the benefits bill is completely counterproductive. Benefits soften the impact of unemployment and wage cuts, but they do not fully replace the lost income - nor should they, or there would be no incentive for people to seek work. People therefore suffer reduction in their real incomes, and that causes a real demand decrease in the economy despite the automatic increase in government spending. Without those benefits the demand reduction would be far greater. Therefore cyclical increase in government spending does not constitute economic stimulus in the sense of actively trying to create growth, whatever Vukovic may think. All it does is prevent demand falling off a cliff.
Vukovic argues that "entitlements" - which would include unemployment and in-work benefits - should be cut to "reduce dependency on the state". He suggests that the private sector cannot expand because the public sector is crowding it out. But unemployment in Spain is at 25% and youth unemployment is at an all-time high across Europe. There is clearly spare capacity - lots of it. It is NOT POSSIBLE for the public sector to be crowding out the private sector at the moment. What is actually happening is that the private sector is retrenching - it is hoarding cash, paying off debt and waiting for better times. There is no evidence that the private sector in any of these countries is yet ready to provide the jobs and wages that are needed to enable the benefits bill to reduce naturally. Cutting benefits would therefore cause real deprivation. It is a simply appalling idea that shows a comprehensive lack of understanding both of the situation in Europe and, frankly, of basic economics.
Edward Harrison produced a post earlier today in which he bewailed the fact that people don't seem to understand national accounting. Vukovic's post demonstrates this in spades. Government spending cuts DON'T "allow the private sector to expand" when the economy is operating at less than full capacity. Unless they are matched by equivalent tax cuts they force the private sector to bear more cost, which reduces its ability to save and to invest.
I am certainly not arguing for out-of-control government spending such as that demonstrated by Greece in the years prior to the financial crisis. Nor am I suggesting that structural reforms are not needed in many of these countries. But trying to cut the absolute amount of government spending in a recession will make the recession even worse, and attempting to reduce or dismantle unemployment and in-work benefits (automatic stabilisers) will cause real suffering. Even the IMF - hardly a supporter of profligate government spending - is now suggesting that the pace of fiscal consolidation should be gentler and automatic stabilisers should be allowed to do their job.
The ASI might as well have titled Vukovic's post "How to have an even longer and deeper recession in Europe". Because that's what he is proposing.
I criticised the chart in this post on four grounds:
- the figures were not adjusted for inflation
- the figures were shown in Euros, which meant that the UK's spending in 2009 was overstated because of the devaluation of sterling
- the figures were not quoted in relation to the size of the countries' economies
- there was no allowance for cyclicality (automatic increase in government spending as benefits bills increase in economic downturns due to unemployment and wage cuts).
And I produced a lot of charts of my own showing that when the above are taken into account, the conclusions of Sam Bowman's post - that there hadn't yet been any serious spending cuts and there was far worse to come - were only partly justified.
So what did the ASI do? They issued the same chart AGAIN in a different post, by Vuk Vukovic. And he produced from it an even more mistaken analysis.
I pointed this out in a comment on the blog, and to my amusement the ASI then replaced the chart with this one:
Spot the difference? Yes - these are real rather than nominal figures. But the chart is are still in Euros and we don't know if allowance has been made for translation differences for the UK, there is still no attempt to relate the spending figures to economic output (GDP) and it still makes no allowance for cyclical factors. So this chart is no more meaningful than the previous one.
Showing absolute figures without reference to the size of the economy gives a completely misleading impression. To help make my point, here's a silly chart of my own:
This chart shows European government debt absolute figures, irrespective of GDP - in Euros, so no allowance for sterling depreciation. Who exactly has the biggest debt pile? Yup, that's right. Since 2009, Germany. And Greece has the lowest (of these countries, anyway). So if we take absolute figures only, ignoring the size of the economy - which is what the ASI does with their chart - then it should be German debt on which yields are heading for the moon. It should be Germany facing default and exit from the Eurozone. It should be Germany facing sanctions and fines. Shouldn't it?
But when you plot European debt in relation to GDP - as it is usually shown - the picture changes completely:
Germany's debt pile looks - er, quite large, at 82% of GDP, which is well above the Maastricht convergence criteria. In fact it looks about the same as France's and the UK's. Spain's debt/GDP is actually lower at the moment, though I reckon that will change radically when it is forced to bail out its banks and its regions, as will happen pretty soon. But Germany's debt certainly isn't the largest in the Eurozone when you compare it to GDP. That honour, surprise surprise, belongs to Greece.
In fact the debt to GDP chart is no more sensible than the absolute debt chart. Debt is accumulated deficits over years, whereas GDP is an annual figure; it could therefore be argued that quoting government debt in relation to GDP compares apples and oranges. It certainly tells you absolutely nothing about the ability of the assets of the country to support that level of debt - which is the country's solvency. What would arguably be better would be to map the COST of debt - interest payments and refinancing - in a given year against GDP. That at least would give some idea of the ability of the economy to service the debt. However, I digress.
To be fair to the ASI, their chart does show that governments in Europe generally are increasing their spending rather than reducing it - which is the point they were making. But it isn't that simple. My final objection to their graph is that it takes no account of cyclical factors. All of the countries in the ASI's graph are in recession at the moment: Greece's recession is the worst in recorded history. When people are losing their jobs and experiencing wage cuts - which is what happens in recessions - the benefits bill automatically increases, so Government spending increases as the economy contracts. So there may be real cuts in government spending in other areas, but the effect of these on the total Government spending bill may be wiped out by the automatic increase.
Attempting to reduce the benefits bill is completely counterproductive. Benefits soften the impact of unemployment and wage cuts, but they do not fully replace the lost income - nor should they, or there would be no incentive for people to seek work. People therefore suffer reduction in their real incomes, and that causes a real demand decrease in the economy despite the automatic increase in government spending. Without those benefits the demand reduction would be far greater. Therefore cyclical increase in government spending does not constitute economic stimulus in the sense of actively trying to create growth, whatever Vukovic may think. All it does is prevent demand falling off a cliff.
Vukovic argues that "entitlements" - which would include unemployment and in-work benefits - should be cut to "reduce dependency on the state". He suggests that the private sector cannot expand because the public sector is crowding it out. But unemployment in Spain is at 25% and youth unemployment is at an all-time high across Europe. There is clearly spare capacity - lots of it. It is NOT POSSIBLE for the public sector to be crowding out the private sector at the moment. What is actually happening is that the private sector is retrenching - it is hoarding cash, paying off debt and waiting for better times. There is no evidence that the private sector in any of these countries is yet ready to provide the jobs and wages that are needed to enable the benefits bill to reduce naturally. Cutting benefits would therefore cause real deprivation. It is a simply appalling idea that shows a comprehensive lack of understanding both of the situation in Europe and, frankly, of basic economics.
Edward Harrison produced a post earlier today in which he bewailed the fact that people don't seem to understand national accounting. Vukovic's post demonstrates this in spades. Government spending cuts DON'T "allow the private sector to expand" when the economy is operating at less than full capacity. Unless they are matched by equivalent tax cuts they force the private sector to bear more cost, which reduces its ability to save and to invest.
I am certainly not arguing for out-of-control government spending such as that demonstrated by Greece in the years prior to the financial crisis. Nor am I suggesting that structural reforms are not needed in many of these countries. But trying to cut the absolute amount of government spending in a recession will make the recession even worse, and attempting to reduce or dismantle unemployment and in-work benefits (automatic stabilisers) will cause real suffering. Even the IMF - hardly a supporter of profligate government spending - is now suggesting that the pace of fiscal consolidation should be gentler and automatic stabilisers should be allowed to do their job.
The ASI might as well have titled Vukovic's post "How to have an even longer and deeper recession in Europe". Because that's what he is proposing.
Ability to service a debt has to be measured across the whole lifetime of the debt. If we're entering a sustained period of decline, our future ability to pay will be less One of the reasons we have the whole cult of growth is to shortcircuit such concerns, of course, as in the mythical rose-tinted tomorrow we'll always be better off. Which we won't be, thanks to Peak Everything.
ReplyDeleteWe could of course discount the cost of debt service over the lifetime of the debt to create a present value. But comparing that with GDP would be as meaningless as comparing the whole value of the debt with GDP as if it all has to be paid off this year. What we really should do is discount the future value of GDP over the lifetime of the debt as well - quite a feat, considering that some of our debt is PERPETUAL. We have no idea what our future growth rate will be and therefore whether the debt we have now can be serviced. You may be right that we are entering a period of sustained decline, or you may be wrong - history is not on your side, but I'm aware that Peak Everything might be a game changer. But frankly, attempting to forecast whether our present debt level is serviceable in the future is fantasy economics. The only indication we have as to the future serviceability of debt is current GDP and forecast trends - and as we all know, those are about as reliable as the long-range weather forecast.
DeleteFrances - you point out the obvious.
ReplyDeleteIt is not that the public sector is too big, particularly when a substantial portion of it is in benefits of some sort. This as you say prevents the economy falling off a cliff. It is that the private sector is too small.
But it is invidious to compare public and private sectors. They have completely different motives and functions.
The public sector is about service to the population and country. It should use the most efficient means of delivery, which sadly it doesn't in many cases, but it provides services like education and health care, sweeping the streets and maintaining law, order and defence. These are essentially people facing and that means it needs to employ people to deliver the service. These services are essential for the private sector to survive and few seem to recognise this, joining the 'private good, public bad' rants of certain people on the basis of thinking that even Winnie the Pooh could see through.
The private sector's function on the other hand is to make a profit. If it can make a bigger profit without employing more people, that is fine. If it can make a bigger profit by employing even fewer people, even better.
But there is another key difference between public and private which needs to be factored into any calculations.
If the public sector cuts back, who picks up the tab? As people move from employment to unemployment, the exchequer loses tax revenue and spends on benefits. So it makes very little sense to cut particularly low paid public sector jobs - all we get is no work for not much less money in total. The public sector is a service and as such, private-sector levels of remuneration are obscene where there is no stakeholder to sack or close the 'business'. People in the public sector used to consider it a privilege to serve the nation (or Her Maj). Now some of them just regard it as a gravy train to bolster their pensions. You can lose these high-paid public jobs with much more ease - they will move back to the private sector from whence they came. It is much more difficult for carers, librarians or whatever to find private sector work.
On the other hand, a private company making workers redundant have no more financial responsibility for them - apart perhaps from a paltry redundancy pay and some pension contributions. The private company can then go on and survive while the taxpayer picks up the benefit bill plus the loss in tax revenue. So the private company makes a profit only if the state picks up the tab or we have abject poverty all around.
I am not arguing particularly for a massive public sector but really that the government should recognise that the public sector is not a business even if it should use best business approaches.
The future must therefore be to promote the private sector by all means, which probably includes a dedicated SME bank along German lines, low taxation (without the scams), efficient government supporting business.
I saw very little of this in the budget. In fact it was a disgrace that extended far beyond the pasty tax and the child benefit fiascos.
In the EU, we have perhaps awarded ourselves a little too much comfort by trying to improve the quality of society with decent education and health care etc. I don't apologise for this - it is a good objective. The answer cannot be to become India, pushing people into poverty. India is trying to climb out of that cesspit just as Greece is falling in. Do we want to join it?