Brian Fallow: Fiscal straitjackets always find a Houdini

A bill to impose a statutory limit on the rate at which government spending can increase - in effect freezing it in real, per capita terms - was introduced to Parliament last week.

The Spending Cap (People's Veto) Bill is an Act Party initiative; National at this stage is committed to supporting it only as far as select committee consideration.

Its advocates see it as a mechanism for progressively reducing government spending as a share of GDP, which they argue would boost economic growth.

Its opponents see it as way too restrictive, especially in perilous times - the fiscal equivalent of putting a straitjacket on a goal-keeper.

In a small, open economy liable to be sideswiped by international shocks, the Government needs to have maximum flexibility to respond.

Fiscal rules, of which spending limits are one variety, are not uncommon. An International Monetary Fund study in 2009 identified 23 countries with expenditure rules.

How effective they are is a moot point, however.

Finland, the Netherlands and Sweden appear to have had positive experiences, the Treasury says.

But also on the list is Japan, which since 1947 has had a rule that government expenditure should not exceed revenue.

"Since 1975, except during the period 1990-93, the Government has requested a waiver of this rule every year," the IMF notes. Japan's government debt is over 200 per cent of GDP.

It illustrates the general point that a fiscal rule which one Parliament tries to impose on its successors is no substitute for the actual political will to curb spending or to be reasonable in piling up obligations on future taxpayers.

The Treasury's regulatory impact statement on the bill notes that "the spending cap and the results of any referenda [to lift the cap] will bind the Government only to the extent that Parliament continues to agree to remain within these constraints. A future Parliament has the ability to amend the level and/or operation of the expenditure cap or to appeal it outright."

It also notes the proposed spending cap could have the perverse effect of encouraging future governments to bump the cap up again to meet their policy objectives through tax breaks instead.

That can have even greater deadweight costs and the fiscal effects can be just as bad.

Calls for some kind of constraint on government spending have intensified since the surge in expenditure during Labour's last term, 2005 to 2008.

Between 2004/05 and 2007/08 government revenue climbed by $11 billion or 21 per cent.

Over the same three years government spending rose $12 billion or 27 per cent.

The Treasury estimates that, had the proposed spending cap been in place over that period, by 2010 government spending would have been about $10 billion lower than it was.

More of that revenue could have been used for some mix of tax cuts, debt reduction or capital expenditure. "Depending on the mix of alternative choices, this could have meant less inflationary pressure in the economy and less pressure on monetary policy."

In a period when the economy is going strong and revenues exceed expectations, it is hard to tell how much of the extra revenue is cyclical froth and how much is structural beer.

The risk is that the structural part is overestimated, and if a spending cap is in place the surplus is used to fund tax cuts instead. The impact on public debt levels could be just as bad, and tax cuts are even harder to unwind than spending initiatives.

Colorado adopted a "taxpayer bill of rights" which imposed caps on both expenditure and revenue, and had a referendum mechanism to override them.

It succeeded in shrinking the size of the state government relative to its economy. However, the Treasury cites empirical analysis that it says indicates the framework did not have a positive effect on Colorado's economy, because the state reduced spending of relatively high value, including education and infrastructure.

The Spending Cap Bill, however, does include some carve-outs intended to mitigate the side-effects of such a measure.

It would apply only to operating spending, not capital expenditure. That is to guard against the risk that governments would scale back or defer potentially productive projects in preference to operating spending that might have a bigger political constituency.

Nor would it apply to the unemployment benefit. That is an item of government expenditure which rises during recessions and falls during period of economic expansion. Including it in the cap would make fiscal policy more procyclical.

Because the cap affects only the spending side of the ledger, another part of the "automatic stabilisers" that kick in during a recession - the hit to tax revenues - would be unaffected. Fiscal policy overall would still turn expansionary during a downturn, as it did during the recession of 2008-09.

A jump in spending necessitated by national emergencies like natural disasters would escape the cap.

So would the Government's interest bill, on the grounds that the existing Public Finance Act has "been shown to provide an effective limit to growth in finance costs".

Indeed, the existing framework, which enjoys cross-party support, was good enough to bring net government debt down from over 50 per cent of GDP in 1992 to below 10 per cent by 2006, where it remained until the global financial crisis hit.

Even with the lagged effects of the GFC and the Crown's share of the costs of the Canterbury earthquakes, net debt is forecast to peak at just under 30 per cent of GDP in three years, and the Government says it is committed to getting it back to 20 per cent of GDP by the early 2020s.

These are the sorts of numbers associated with emerging economies - these days the provident ones - rather than advanced economies. The IMF estimates net government debt among the G20 countries over the next five years will average 82 per cent.

"If it ain't broke, don't fix it" comes to mind.

The risk is we end up with a fiscal approach that says when times are good and the Government's books are in the black, it is time to cut taxes, because it clearly does not need the money.

But when times are hard and books are in the red, it is time to cut spending because of all that debt that is piling up.

Either way the verb is the same: cut.