Friday, 30 April 2010

Flawed Analysis From the IFS

by Michael Burke

The Institute for Fiscal Studies (IFS) has produced an analysis of the three main parties’ budget plans. The IFS has a reputation for voluminous and detailed analysis. The latest analysis runs to seven documents, but it is deeply flawed.

The essential points of the IFS analysis are to highlight the level of cuts required by the three main parties in order to meet their own targets and timetables for budget deficit-reduction. According to the IFS, the Tories would need to make cuts in public spending of £63.7 billion by 2014/15, (and so far have only identified 17.7 per cent of these). Labour would need cuts of £50.8 billion (of which they have currently identified 13.3 per cent) and the LibDems would require £46.5 billion of cuts (and have specified 27 per cent of those). The differences arise mainly from timing as well as the proportion of tax increases adopted. At the upper end, the Tory cuts are equivalent to the entire current budget for children, schools and families in the current financial year, in addition to the current budgets for international development, transport and most of the work and pensions budget too.

No Mandate

The unspecified spending cuts have led to the charge that the parties are being dishonest with the electorate, a charge which was widely take up by the media after the publication of the IFS report. This is so, even though the Labour and LibDem plans are estimated to be the tightest since the IMF was called in to Britain in the late 1970s, while Tory plans are reckoned to be the worst ever, certainly since WWII.

Clearly, there is no mandate for the type of swingeing cuts outlined, and cuts themselves remain deeply unpopular. In particular, public opinion is firmly against cuts to the schools’ or training budgets, health and pensions. It is also against increasing student fees or interest payments, as well as removing winter fuel payments and free TV licenses for the over 75s. Apart from cutting the BBC license fee, one of the other most popular cuts would be on scrapping Trident.

No Logic

While there is no support and no mandate for the cuts programmes, there is no logic to them either. The IFS actually does a great disservice to the level of the debate about the economy and government finances. This is because, simply accepting the Treasury economic forecasts and casting them in stone, IFS entirely ignores the economic impact of the cuts. Therefore it entirely misses the consequences of those cuts on the issue it raises above all else, namely the public sector finances.

In effect, the IFS approach mistakes action for effect. Therefore, every cut becomes a saving. On the IFS website, like that of the Financial Times, it is possible to construct your own package of spending cuts, and tot up the totals - a macabre computer game for neo-Thatcherites.

But for serious policy discussion it is vital to gauge the effect of every action on the economy and therefore on government finances. It is possible to do that from the Office of National Statistics’ Input-Output Tables, previously highlighted on SEB. These show the multiplier effects of government spending on other sectors. The multiplier for the category of government spending on ‘Health, Education and Social Work’ is 1.854. That is, every £1bn withdrawn from government spending in this area will lead to a decline in total output alone of £1.854bn. That is, lower government spending on health will lead to lower private sector output of medicines, health equipment and machinery and so on. This does not take account of the additional effect of lower spending by health workers laid off or whose pay has been cut, as the ONS struggles to compute that, for technical reasons. On the basis of output alone, the effect of cuts spelt out by the IFS would actually lead to decreased economic activity of £118bn in the case of the Tories, down to £86bn in the case of the LibDems. These are equivalent to 9.3% of GDP and 6.8% respectively, based on 2009 levels. This is more severe than the current recession that Britain is only now slowly emerging from.

As it was put in the New York Times on similar effects on other countries: “Officials from Standard & Poor’s said the main reason for downgrading the debt of Greece and Portugal was the prospect that forced austerity packages would be an even bigger drag on economic growth. It is the most vicious of circles: stagnating economies are forced to cut back more, which reduces their ability to generate revenue and thus pay off their debts."

That this much lower output has its own negative impact on government finances can clearly be seen with the UK also. This arises from two areas; lower taxes from lower levels of activity and increased government spending in other areas, mainly unemployment and welfare payments. SEB has previously shown how the ‘Treasury Ready-Reckoner’, the official Treasury analysis of the relationship between government finances and the economy, estimates this impact.

In effect 75 per cent of all changes in output are reproduced as changes in government finances in the second year. 50 per cent arises from changes to taxation revenues and the remainder from changes in government spending. Therefore we can calculate the actual change in government finances arising from these cuts.



In all cases the deficit actually worsens under the impact of the cuts programmes, with the deterioration in government finances ranging from around £12bn to £25bn.

Nor is it the case that a greater reliance on tax increases from Labour and the LibDems automatically alleviates the problem. Only transfers to the poor from the rich have the effect of boosting total demand, as the poor spend a greater proportion of their incomes. In this regard, the IFS argues that only Labour’s tax plans marginally favour the poor over the rich, whereas LibDem tax plans favour those on middle incomes, to the detriment of both the rich and the poor. Within a decimation of public services, pay and jobs, Labour’s plans are fractionally less harmful to the poor than those of the LibDems, while the Tories would need to be the worst government in post-WWII to achieve their plans. In reality, the effect of the cuts programmes are much worse even than the IFS estimates.

1 comment:

Rav said...

This is rather unfair to the IFS, which does not recommend any particular level of spending cuts in this paper, but simply tests how the parties have laid out their own deficit reduction plans.

Your analysis of the multiplier effect of Government spending is highly simplistic. Why is all the spending at stake assessed at the single multiplier rate for health and education spending? Much of the cuts, all parties agree, need to be elsewhere. Furthermore, the multiplier is an extremely broad average, masking government spending which is highly stimulative and plenty which is not. It is at least theoretically possible that the new Government will be able to "trim some fat" without taking this amount of money out of the system.

The more fundamental point you miss, however, is that the multipliers you describe only function in the context of a generally healthy fiscal environment. Yes, Greece's austerity measures will harm its economic growth; and yet, nobody believes they aren't necessary. Why? Because unless a country has the luxury of being without debt - no country does - they must above all retain the faith of the bond markets. When you're building a small business, you might feel you could expand more quickly with more funding, but that doesn't mean you can simply go over your overdraft limit.

The fact that the public are opposed to many specific spending cuts (although highly in favour of deficit reduction in general) is entirely irrelevant to an economic discussion. I see you make no effort to calculate the cost to the economy of scrapping Trident.

It is a valuable point to remind people that Government spending is stimulative and not, as many seem to think, a drag on the economy; and that, therefore, spending cuts will have a growth-reducing effect. But unfortunately you undermine the value of that point with selective statistics, irrelevant sermonizing and flawed analysis.