Media Briefings


  • Published Date: January 2011

Getting a fiscal stimulus package ‘wrong’ could mean that any short-run gains in economic growth are achieved at a cost of lower, perhaps substantially lower, growth over the longer run. That is one of the findings of research by Professors Norman Gemmell, Richard Kneller and Ismael Sanz, published in the February 2011 issue of the Economic Journal.

In what ways might a stimulus be wrong? The evidence suggests that what matters most for longer-run growth is not so much the overall level of taxes and public spending, but rather the structure of the tax system – for example, the rates of income tax versus VAT – and
the type of spending – for example, infrastructure versus benefit payments.

The recent global recession has renewed interest in questions about the relationship between fiscal policy and economic growth. Can fiscal policy affect the growth rate of output? If so, are these effects temporary (affecting only output levels over the longer run) or permanent (so that output growth rates are persistently higher)?

Most macroeconomists agree that cuts in taxes or increases in public spending will increase an economy’s growth rate over the ‘short run’. Where they disagree is on whether the effects over the ‘long run’ are zero, positive or negative – and on exactly how many years best describe these two periods.

The empirical evidence so far points strongly to a positive GDP stimulus over periods of one to five years from tax cuts or public expenditure increases. But beyond this horizon, the reliability of evidence has typically been clouded by limitations of the data, the testing methods or the theoretical foundations of the analysis.

Despite these difficulties, recent evidence has begun to point to positive long-run growth effects from changes in some taxes and public expenditures. That is, rather than (or as well as) the overall level of taxes and spending, it seems to be the structure of the tax system and
the type of spending that matter most for longer-run growth.

But two key issues remain unclear. First, how long it takes for these longer-run growth effects to occur – do they happen quickly or take many years to build up? Second, what is the net effect

on growth if both taxes and public spending both rise or fall, or if government borrowing has to rise to fund spending increases – is the net growth impact likely to be positive or negative?

This new study provides answers to both these questions for a panel of OECD countries since the 1970s, using methods that are better able to cope with some of the data and methodology limitations of previous work. Among the conclusions:

  • Changes in fiscal structure – between different forms of tax and/or types of expenditure
    – affect GDP growth rates over the long run.
  • This ‘long-run’ result is achieved quite quickly, typically within a few years following fiscal policy changes. Previous evidence of so-called ‘long-run’ growth effects of fiscal policy seem to have picked up both the confirmed ‘short-run effects’ and some evidence that these persist beyond a one to five year horizon.
  • OECD governments regularly change their tax and expenditure levels, in both upward and downward directions. So it is not surprising that higher or lower GDP growth rates are rarely persistent over many years.
  • Rather, it seems that the combination of regular reversals of fiscal policy and simultaneous changes to taxes, spending and public debt (which often have counteracting growth effects) ensures that observed growth stimuli are generally short- lived in OECD countries.
  • On the question of whether increasing both public spending (which enhance growth) and tax rates (which damage growth) would be beneficial in net terms for long-run growth, the answer is: ‘it depends on the type of tax/expenditure that is increased, but net effects are often quite small – either negative or positive’.
  •  For example, even if public spending is on infrastructure (which is often found to have a positive impact on GDP), if this is financed using highly distortionary (personal or corporate) income taxes, the net long-run impact is negligible.
  • But getting the stimulus package ‘wrong’ could mean that any short-run growth gains are achieved at a cost of lower, perhaps substantially lower, growth over the longer run.

The researchers acknowledge that little is known about the size of possible adverse longer-run growth effects from large and sustained fiscal deficits, following a major stimulus package. This is especially true where large inter-country fiscal transfers are involved, such as the recent
European Union bailout packages for Greece and Ireland in response to their huge fiscal deficits.


Notes for editors: ‘The Timing and Persistence of Fiscal Policy Impacts on Growth: Evidence from OECD Countries’ by Norman Gemmell, Richard Kneller and Ismael Sanz is published in the February 2011 issue of the Economic Journal.

Norman Gemmell is chief economist at the New Zealand Treasury. Richard Kneller is at the University of Nottingham. Ismael Sanz is at Universidad Rey Juan Carlos.

For further information: contact Norman Gemmell via email:; or Romesh Vaitilingam on +44-7768-661095 (email: