Tuesday, 20 October 2009

How much does the State spend as a percentage of GDP?

Slí Eile: Earlier this month, at a conference of the National Economic and Social Council, Joseph Stiglitz warned against a single-minded focus on reducing public liabilities at the cost of public assets. There is an assumption abroad (in Ireland at least) that we are ‘living beyond our means’ and ‘unless we cut public spending the consequences will be terrible’. I would be surprised if a majority of persons who think much about these issues, and follow them in the media and in casual conversation, would disagree with the line that says ‘you can’t continue borrowing that amount – €400 m a week or €21bn a year (and rising)’ ‘We are borrowing to pay for 2008 public services with 2004 taxes – 5 into 3 won’t go’ ‘you can’t raise taxes much any more – businesses and hard-earning tax payers are being hammered’ ‘you will scare foreign investors and market sentiment’ ‘we have to take the pain and do our bit to get the country moving’.

None of these popular sayings have anything directly to do with recovery or job creation. Rather, they concern one issue only – ‘the country is bankrupt – we have to sort out our public finances’. By implication this equates to saying ‘we first have to sort out public finances (and banking) BEFORE we can deal with the job crisis. Indeed, many commentators will add ‘by cutting public spending and keeping a lid on taxes we can help the traded sectors of the economy to price themselves back into world markets – the main basis on which Ireland inc can grow again’.

But, will ‘adjusting’ for €4bn in the coming budget achieve its desired goal? Michael Taft has already demonstrated that public deflation will not achieve the single-minded aim of reducing public borrowing by anything other than a very modest amount. I will not repeat the analysis which can be found here.

In short, cutting public spending – especially when targeted on low to mid-income public servants (believe it or not, there are some) and social welfare recipients – will further reduce tax receipts (other things constant) and add to the volume of social welfare payments (other things constant). The fall in retail sales and tax receipts commented on this blog site is connected to the strongly deflationary impulse of the December 2008 and April 2009 budgets. The extent of the impact is unknown and would be interesting to model, empirically, if the latest data and will were there to do it.

One element missing from the debate is a consideration of the following question:
During these traumatic and uncertain times what level of public services is desirable and attainable on various assumptions and scenarios?

So, are we spending too much on public assets?

Using the latest National Accounts data for 2007 (yes) from the CSO National Income and Expenditure 2008, total spending (current and capital) by Government (central and local) was €73.8bn in 2007. This represented about 38.9% of GDP in that year – the height of the boom. Unfortunately, for some reason, it is very difficult to get a comparable estimate of total public spending in 2008 and projected for 2009. The April Supplementary Budget contained a Macroeconomic and Fiscal Framework. Table 7 in that document sets out total projected current and capital spending for 2009 (but not earlier years) and subsequent years. The total of gross current and gross capital spending (before deduction of receipts in each case) comes to €73.6bn in 2009. This latter figure is unlikely to be directly comparable with the National Accounts figure for 2007.

Taking total projected gross spending in 2009 and dividing by GDP in 2009 (using the latest ESRI forecast), you get an estimate of just 45% of GDP. Something akin to this figure has been used by economists such as Colm McCarthy. For example, An Bord Snip Nua reported (page 2) that:

In 2009, gross voted current spending (not including Central Fund expenditure such as debt service costs) will absorb 39.3% of likely GNP, the highest figure since 1983. Total Government spending (the current voted spending figure, plus debt servicing and other Central Fund expenditure, as well as Exchequer capital which includes the payment to the NPRF) will absorb 51.1% of GNP, the highest figure since 1987.

This claim has gone unchallenged and, in the absence of an explanation about how Bord Snip arrived at this estimate, we have to caution against any conclusion that public spending has jumped from about 39% of GDP in 2007 to 51% in 2009. Included in the 51% figure is the pre-payment of some €1.5bn to the National Pension Reserve Fund.

In the absence of reliable and up to date estimates of national and public accounts, the simplest approach is to add up receipts (which are easier to quantify) and then add an estimate for total borrowing in 2009. Social Justice Ireland have estimated that total income including taxes, social insurance, local government receipts will come to €47bn in 2009. If this comes to pass, it would represent 28.7% of GDP (SJI are using a higher GDP estimate to arrive at a slightly lower estimate of 27.4%)

If total borrowing comes to about €22bn in 2009 – as many commentators now expect - then total spending in 2009 is likely to be somewhere in the region of €69bn (current and capital, local and central). That would give a total public spend of very roughly 42% of GDP. I will settle for that estimate in the absence of any better data (readers may point to better estimates somewhere).

42% - sounds like a lot. It is certainly the case that some areas of public spending continue to grow – higher social welfare costs associated with growing unemployment, the cost of the pay increases across the public sector in September 2008, the impact of rising population on health and education services etc. Set against a collapse in GDP (and a spectacular one for any industrialised country since the 1930s), a significant rise in public spending as a percentage of GDP is not unexpected.

How do we close the gap between 42% and 29% (for total receipts of various kinds)? Bearing in mind that GDP will contract by a further 3% at least in 2010 (and arising from this more unemployed and more social welfare payments at current rates of payment), the challenge for any administration is to significantly raise receipts without further deflating the economy, and continue to borrow to cover those components of the deficit that correspond to (i) capital investment (for which money is usually borrowed anyway, even in good times – roughly €10bn per April figures), (ii) current spending and tax shortfalls associated with the downturn (welfare payments and reduced tax receipts) – estimates vary but could be €6bn. The remaining gap could be in the order of €6-7bn – the structural element. I think that there is a case for continuing to run a deficit of about €15bn per annum – or about 10% of GDP in 2010 - while re-targetting public spending on job-creating and job-retaining projects and maintaining the real values of social welfare payments (any short-term gains made as a result of modest HCIP falls in 2008-09 are more than cancelled out by reductions in wider social provision, plus likely impending price increases in 2011 as prices recover). The ‘structural gap’ of some €7bn needs to be tackled through a combination of measures:

Capital taxes
Income tax base widening
Tax-relief reductions (especially the most inequitable)
Local taxes
Income tax surcharges on very high incomes
Corporate tax increases to 15% and pegged for 5 years.

Yet again, we hear the often repeated claim that the yawning public sector deficit has nothing to do with the banks and NAMA. For one thing, we would not be as much in this mess were it not for the banks (thanks for the apologies emanating from Kenmare). In the second place, the recapitalisation cost of the banks is exacting its own ‘opportunity cost’ on public finances. Don’t take my word for it. The Department of Finance has revealed that:

At end-September 2009, the Exchequer deficit is €20,158 million, compared to €9,404 million at end-September last year. The year-on-year deterioration in the deficit of some €10.8 billion is primarily explained by a decline in tax receipts of €4.8 billion, the €4 billion payment to Anglo Irish Bank and €1.7 billion in respect of the frontloading of the annual contribution to the National Pensions Reserve Fund (NPRF).

A parting thought - now, suppose that the world is headed towards a double-dip recession with a downward shock to world trade and sovereign defaults spreading across Europe – what would that mean for Irish exports, for Irish tax receipts from VAT and income – and for public spending on education, health and social welfare? However unlikely such a doomsday scenario looks, it is worth bringing together a number of ‘what if’ scenarios ranging from rapid economic recovery from 2010 onwards to double-dip world slump in 2010 (and L-shaped recession for Ireland which I think is the more likely as our key trading partners pull out of recession next year and we are left to clean up on debt, NAMA and the prolonged deflationary impact of Budgets 2008-2010 and beyond).

It is not a pretty sight and nobody knows for sure what is going to happen next. Not good for confidence, trust, and job-creating investment and consumption.

We need a change of direction politically. All this deflation stuff is not good.


Proposition Joe said...

Here's the oft-glossed-over sting in any scenario that prefers tax increases over spending cuts. Its not a nice thought, but there's no realistic scenario for building up income tax revenues that doesn't involve low and average earners paying susbtantially more than they do currently. Whatever about the feasability of current spending levels, we absolutely cannot maintain that spending within a tax régime where half the workforce pay no income tax at all, and a mere 11% pay at the top rate (as will be the case in 2009 according to recent radio interview with the Taoiseach).

However there are two big blockers standing in the way of doing anything about this. One, it would be highly unpalable, maybe even impossible, from a political standpoint to back away from the article of faith that its fundamentally a good thing to take lower earners out of the tax net. Two, reducing disposal income at the lower deciles would also be highly deflationary, for exactly the reason spelled out many times on this blog (i.e. lower earners tend spend all their income by necessity).

Aidan said...

I think you aren't comparing like with like when you question how expenditure could have gone from 39% to 51% in a couple of years. You already said yourself that the 51% was of GNP, whereas the 39% was on GDP. Given the disparity between the two for the Irish economy, 39% of GDP isn't that different to 51% of GNP.

Nat O`Connor said...

@ Proposition Joe

I agree that, over time, there is a need to move towards more people paying more tax. I also agree that it will be politically unpalatable.

Two ways in which this goal can be approached are (1) to reduce some major costs that absorb a disproportionate amount of household's income, and (2) to increase transparency about where taxation goes.

Re (1), at the moment housing costs seem to be particularly high. For example, we saw a huge increase in the gap between building costs for new housing and asking prices. Housing doesn't lend itself to working well as a free market, so a strong state role is inevitable. And the state is in a position to do more to regulate and limit housing costs. That would free up more income that can be taxed in exchange for better public services.

Re (2) we have traditionally shied away from hypothecation of tax revenue. There is an argument for more clearly linking various taxes and charges to the public services they provide (from waste and water at local authority level, to health and education at national level). Increased transparency would also help calculate the benefits of taxation, as has been done in Canada, where estimates of the value of public services show that households benefit from services worth tens of thousands of Canadian dollars annually.

If we had (1) and (2) in place, we would be in a better position to challenge the ideology of low taxes. Most people don't buy the cheapest goods and services in the economy; in a similar way, we can arrive at a situation where the public can see a clear relationship between cost and the quality of services they receive; and yes, there's probably a lot of waste and inefficiency that transparency would flush out at the same time.

Michael Taft said...

Part of the problem in assessing a year-on-year comparison for Government expenditure is that different agencies use different measurements. For instance, one has to go into the public estimates for each year because the headline tables produced in budget tables use net revenue and net expenditure (that is, excluding the social insurance fund - even though the fund represents both revenue and expenditure). The CSO National Account Classification includes redemption of securities which can distort the year-to-year figures if a security falls due in one year. For instance, in 2002 total expenditure was €54.5 billion but the following year it fell to €47.8 billion. The difference was not budget cuts but rather the NTMA redeeming debt.

A short-cut to all this is look at the EU Commission's Annex which compiles revenue and expenditure for all countries. This has the advantage of being consistent across the board and leaving out redemption of securities.

It shows that between 2002 and 2006 Irish expenditure remained constant as a % of GDP: approximately 33.7% (GNP - 39.5%). Things started going south as the recession set in:

2007: GDP (35.7%) / GNP (42%)
2008: GDP (41%) / GNP (48.2%)
2009: GDP (45.8) / GNP (55%)

There are three factors at work: expenditure is rising almost solely due to uemployment and related payments; collapsing output; and GNP is being hit worse than GDP.

John Fitzgerald, down in Kenmare, stated his medium-term preference of expenditure being 45% of GNP. This would represent a considerable increase over the pre-recession trend but this would also be taking into account a continuing high level of unemployment. However, this would still leave us behind the EU and Eurozone average expenditure levels pre-recession.

An Saoi said...

Propositon Joe - There are a number of points you raised, which I will try and deal with, avoiding technical tax speak. There are a massive number of tax "expenditures", i.e. special tax reliefs and deductions granted, which have been abused widely, for example film relief where a bank (Anglo Irish in the past) & a film company "borrowed" your income for a sum of money and you got tax relief. Some of these tax expenditures around property bear a large part of the blame for the mess we are in. But on your other point, yes certainly some of the 50% of those working and not paying tax should pay at least something

1) I must admit that I feel many of them should make a contribution, however small towards the cost of the State. This will not raise much in tax, but there is as much an issue of solidarity involved as anything else. Many are receiving other State payments, e.g. a child "working" in a family business. Should the Child benefit not be deducted from his tax credits? Should the same apply to the student receiving a grant?

2) There are literally thousands of millions in unused useless capital allowances being carried forward, arising from the various property based schemes introduced by FF over the past 30 years. It is time for a "use it or lose it" clause to be introduced in the next Finance Bill clearing out all these reliefs not used before 31/12/09. Normal losses against a trade or profession would continue, but would be ringfenced against that trade or profession. This has been proposed by Joan Burton and the Minister has introduced the idea in relation to the bank loans transferred to NAMA. Normal capital allces for the purchase of assets for the purpose of a real trade would continue.

3)Get rid of all interest relief against Case V (rents). No interest deductions against passive income.

4) Get rid of mortgage interest relief against principal private residences.

5)I am opposed to the further increase in the CT rate suggested above. The rate for manufacturing companies is increasing from 10% to 12.5% wef 01/07/2010. That is, as a US Tax Manager said, a 25% increase.

6) Let us cut the CAT threshold further and increase the rate to 30%.

7) Repeal much of Charlie McCreevey's pension rules which turned Small Self Administered Pension Funds into tax releived investment funds, nothing to do with pensions.

The above changes would increase tax yields by at least €4,000M next year and have little real effect on spending. They would act as a capital tax on much of the gains that have accrued to the friends of FF involved in property speculation with borrowed money.

Slí Eile said...

Gosh at this rate - at this rate (55%) the 'noughties' will be socialist instead of the 1970s. But, these estimates must surely include NPRF payments and other non-voted spending some of which goes to Anglo-Irish. More detail needed.

Michael Taft said...

Sli Eile - I doubt that the data used by the EU Commission includes NPRF and Anglo-Irish since they're compilation is based on criteria for the excessive deficit procedure. However, their data are estimates and I should have noted that in my comment. I was only using it to find an internally consistent of data to compare with previous years. In addition, they are exceedingly pessimistic regarding Ireland's deficit, predicting it would exceed -15% next year. I thought that overly pessimistic until the ESRI stated that this year, the deficit will fall to -13%.

Taking the April budget's estimates, total voted expenditure will reach 48% of GNP. This does not include the NPRF funding (it represents two years) and bank capitalisation. You're right, Sli Eile, we need to construct a definitive set of Irish data that isolates particular expenditure (NPRF, redemption of securities and, now, bank capitalisation).