Take a look at the chart. It shows how much of our GDP has been spent by government under each of our last six Prime Ministers (counting the 1974-79 administration as one).
44 years ago when Harold Wilson first took the controls, government spent 38%. This year, it expects to spend 42% (£589bn), but the path from 1964 has been extraordinarily bumpy, and back in the mid-70s they spent 50% (we're using HM Treasury's figures on Total Managed Expenditure- TME).
Where will the line go next?
According to Labour's spending plans, over the next three years spending will increase by 2.0% pa in real terms, against GDP growth of 2.4% pa (see the Pre-Budget Report- Table B3). So the government's share is planned to fall.
But that is almost certainly too optimistic. We have before us an ordeal of grievous debt overhangs, crunching credit, sagging house prices, and ever greater competition from the Far East. Barely a day goes by without news of some fresh problem, with a return to seventies style stagflation (or at least stag-lite) a distinct possibility.
Mainstream GDP forecasters- always reluctant to depart from the pack- are rapidly cutting their numbers. The Economist forecast survey now says 1.7% for 2008, down from over 2% a couple of months back, and well below that HMT assumed average of 2.4% pa.
What's more, as the economy weakens, public spending increases, driven by social security payments. So the increase in government's share is doubly boosted, and as the chart shows, the overall effect can be huge. In the early 90's recession, the share jumped by nearly 5 percentage points in just two years.
So far from the government share decreasing as planned, our guess is that it's about to take another significant leg up. Tyler's fag packet says that with TME planned to rise by 2% pa over the next three years, just one year of zero GDP growth would immediately increase the GDP share from 42% to 43%, or even 44%.
Against that background, let's return to the Tories' fiscal policy, and Shadow Chief Secretary Hammond's FT interview last week where he described calls for lower public expenditure as "barmy" (blogged here). We were by no means alone in squawking, and later in the week he posted a clarification on ConservativeHome:
"A Conservative Government would share the proceeds of growth between the funding our public services need and the competitive lower taxes our economy demands. This means that over an economic cycle the economy will grow faster than the government, and government spending will fall as a percentage of GDP. Pursuing this approach over an economic cycle creates the headroom for sustainably lower taxes. And let no one be in any doubt that we are a party that believes in lower taxes."
Sounds simple, doesn't it. Especially since the whole policy is calibrated against the "trend" rate of GDP growth:
"Our policy of sharing the proceeds of growth is defined over the economic cycle... Over the course of an economic cycle a period of below trend growth is, by definition, followed by a period of above trend growth."
Unfortunately, as Philip Hammond surely knows, it won't be quite like that. For one thing, we can forget "by definition" growth. Hammond assumes 2.75% pa trend growth, but in a jobless joyless world, that will be but a distant dream: even Darling is now only assuming 2.5% pa (see here). And public expenditure will simply not stick to that 2% pa projection.
What would Hammond and his boss do then?
Borrow even more? But how would that square with their promised fiscal stability?
Put up taxes? That's what Thatcher did (in 1981), but he's just reaffirmed his belief in lower taxes, and told us that anyway, tax increases "would risk destabilising the economy at a moment of vulnerability" (which IIRC is precisely what those 364 eminent economists said to Thatch).
So what then?
Only two previous administrations managed to share the proceeds of growth. As the following chart shows, Thatcher reduced government's share by nearly 5 percentage points, and Major by about one per cent. All the others- including a Tory administration- increased spending faster than the growth in GDP. Wilson was worst, increasing government's take by over 4 percentage points, but Blair was not far behind.
So, given neither had a third fiscal rule, how did they do it?
Major did it by maintaining a firm grip on public spending for five years as the economy pulled out of recession. And for his pains he went down to the biggest Tory election defeat since 1407.
Thatcher did it by imposing an even more extended period of restraint: from 1979-80 to 1990-91 public expenditure increased by only 1.5% pa in real terms, although GDP was growing at 2.25% pa. And for her pains, she has gone down in the BBC Book of History's Most Evil People.
But Thatcher was Thatcher, and can we seriously imagine Cameron/Osborn/Hammond - miserably out of office for 13 years and facing a "challenging" economy - standing up for a repeat performance?
In his CH post, Hammond mentions the OECD work strongly endorsing the use of explicit expenditure targets- that third fiscal rule (see many previous BOM blogs eg here). He reckons their "sharing the proceeds" is just such a rule. But it isn't.
The rules which have shown to be most effective (such as the 1990 US Budget Enforcement Act) incorporate explicit and quantified upfront caps on spending. Sharing the proceeds is much vaguer - eg with his assumed 2.75% pa GDP growth it could mean real public expenditure growth anywhere between 0.1% to 6.5%, hardly an explicit quantified commitment. Plus, of course, just like Brown's Golden Rule, it leaves virtually unlimited wiggle room around defining the cycle.
There are always a million reasons for higher spending, and politicos always find it difficult to resist. Which is why they need explicit expenditure targets to anchor themselves: whatever the pain, a publicly announced spending target has to be given a high priority or credibility is shredded.
Sharing the proceeds of growth is not an anchor. And drifting politicos whose main proccupation will be getting relected in 2014 are already preparing the ground for wiggle.
Hammond may genuinely imagine his rule will deliver "headroom for sustainably lower taxes", but unless he is more explicit, it just ain't going to happen.
PS As this post may have suggested, Tyler is still a bit confused over what's actually being shared here. If GDP grows by 2.75%, we know public expenditure will grow by less than that. But is that less in percentage terms, or money terms? The percentage formulation- apparently espoused by Hammond- is very similar to the Reform Growth Rule, but it was rejected by Cameron during the leadership campaign. On the other hand, since public expenditure is "only" 42% of GDP the money formulation would allow up to 6.5% growth, as stated above. It's all rather vague.PPS Back in 2006 George explained it thus: "As the economy grows, and revenue flows into the exchequer, we will share the proceeds of that growth between improving public services and reducing borrowing or lowering taxes. Over an economic cycle, the state will consume a smaller share of national income." Sound reasonable, until you realise his two sentences suggest rather different things. On revenue growth, governments conventionally index tax allowances against RPI price inflation, but do not automatically adjust for the increase in real incomes (or indeed house prices). So as incomes and wealth grow, more and more people get dragged into higher tax bands. The tax burden increases even as tax rates remain unchanged. For Labour such fiscal drag has been a key tool of stealth taxation, and according to the IFS, by the end of their second term it was raising them an extra £22bn pa- more than 6 pence on the standard rate of income tax. So sharing the revenue growth from fiscal drag is most definitely not the same thing as cutting taxes. You could share the revenue growth and the tax burden could still increase. But at least his second sentence is clearer- "over an economic cycle, the state will consume a smaller share of national income." Which is also pretty well what Hammond said last week.