The Treasury

Global Navigation

Personal tools


Government and economic growth: Does size matter?

2.1  Balancing the costs and benefits of expenditure

2.1.1  Costs of financing expenditure

Perhaps the most important mechanism through which government expenditure impacts on economic performance is the costs of raising taxes to finance that expenditure. Taxes affect the decisions of households to save, supply labour and invest in human capital and of firms to produce, create jobs, invest and innovate, as well as the choice of savings channels and assets by investors (Johansson et al, 2008). By lowering the returns to earning income, taxes reduce incentives to work, save and invest, thereby “crowding out” or discouraging private sector activity (see Box 1 for more on the debt financing of expenditure).

The economic cost or ‘deadweight loss' created from these disincentives will vary significantly. It will depend upon the tax rate and the responsiveness of individuals or businesses to the tax (known as the elasticity). However, the deadweight loss from taxation rises more than proportionately with the tax rate and has an increasing additional or marginal cost (Barker et al, 2008). In other words, increasing a tax rate from 30% to 40% will undermine economic growth more than by increasing it from 20% to 30%.

As well as the level of taxes, the tax structure also matters. Setting the right mix is important, as the distortionary effects of collecting revenue from different sources can be very different (Johansson et al, 2008). Though all taxes have disincentive effects, taxes that reduce incentives to invest in human or physical capital and innovation are particularly damaging. Consequently, theory and evidence suggest that a shift from taxing incomes or profits to property or consumption can enhance growth (see, for example, Barrios and Schaechter, 2008 and Johansson et al, 2008). Consumption taxes may discourage work and investment in human capital but they appear to have a relatively minor impact on the long-run determinants of growth, such as investment, education or technical progress (Bassanini, 2001). Therefore, endogenous growth models tend to make a simplifying distinction between ‘distortionary' taxes that impact on investment decisions and ‘non-distortionary' taxes that have little impact on investment.

Box 1: Debt financing of expenditure

In addition to taxation, debt (or asset reduction) can be used to finance government expenditure. However, as government must eventually pay back debt, it simply shifts the financing of expenditure across time (Barker et al, 2008).

The notion of Ricardian Equivalence suggests that people will increase their savings in response to an increase in government expenditures financed by debt, in anticipation of the need to pay higher taxes in the future. In this case, financing expenditure with public debt will be perceived as equivalent to a rise in taxation and will have the same cost as taxation (Barker et al, 2008).

If households are not fully Ricardian, higher debt levels will themselves push up the interest and exchange rates facing households and businesses, particularly if expansionary fiscal policy drives tighter monetary policy. Furthermore, deficits perceived as unsustainable could push up the risk premium of borrowing and may discourage private investment through higher inflation or uncertainty around future government expenditure or taxation. High levels of government debt can also endanger macroeconomic stability if the risk of government default on high levels of domestic debt raises domestic financial system risks (Barker et al, 2008). These risks suggest the impact of deficit financing expenditure may be even greater than tax financing (this is supported by empirical evidence summarised in section 3.2.1).

A Treasury Working Paper (Labuschagne and Vowles, 2010) argues that New Zealand's relatively high interest rates do not currently reflect a risk premia imposed by international markets but instead reflect New Zealand's low rate of saving, which requires tighter monetary policy to maintain inflation within the official target range. Government expenditure is still “crowding out” the private sector through this mechanism - the resulting higher interest and exchange rates divert resources from the tradeable sector to finance government and household consumption (see also Treasury, 2010a).

The design of individual taxes also matters. In particular, a broad-base, low-rate principle has driven much of New Zealand's tax policy in recent years (see McLeod et al, 2001 or Tax Working Group, 2010). Broadening the base of taxes is likely to be a less growth-dampening way of increasing revenues than rate increases as it is less likely to change household or business decisions and discourages tax avoidance activities (Johansson et al, 2008). The administrative efficiency, simplicity, transparency and stability of revenue systems can also minimise the harmful effects of tax on growth by keeping the administrative burden on taxpayers and the public sector low (Barrios and Schaechter, 2008).

2.1.2  Contribution of government expenditure to economic growth

While financing expenditure carries costs to economic growth, some types of government expenditure are beneficial to economic performance. Some government expenditure is a prerequisite for a functioning market economy, such as a legal system to protect private property rights. Beyond this foundational level, expenditure initiatives may lift long-run growth rates by increasing investment in physical capital, knowledge, human capital, research and development or public infrastructure, particularly where market failures lead to under-investment by the private sector. For example:[1]

  • Physical capital. Government investment in physical capital could boost long-run economic growth if investment stimulates technological progress or if the productivity of businesses is boosted from others' investment or innovation (knowledge spillovers). Government can directly invest in physical capital or infrastructure or it can encourage private sector investment.
  • Human capital. Investments in human capital may have persistent impacts on economic growth if education enables ongoing innovation and advances in technological progress. Individuals and firms may under-invest in human capital from an economy-wide perspective as they will not factor in the positive flow-on effects to other workers and businesses from investment in education and training. This can be compounded by problems in accessing capital to finance investment in education, providing a rationale for government funding of education.
  • Technological development. Non-rivalry and less than full excludability of advances in knowledge mean that it is not desirable or possible for businesses to capture the full benefits from innovation. Consequently, the private sector may undertake less innovation than is socially optimal. This may justify some government involvement through direct provision and funding of research and development, as well as through indirect measures such as tax incentives and protection to intellectual property rights to encourage private-sector innovation.

However, even when market failures exist, the information constraints facing the public sector in designing appropriate policies mean that government interventions often have unintended consequences. It is also important to recognise that policymakers often focus on addressing market failures to boost overall living standards, rather than economic growth per se. Though economic growth is an important driver of living standards, they are not always synonymous and some policies may lower economic growth but be desirable because they boost other dimensions of living standards.[2] Furthermore, government expenditure typically extends beyond market failures to provide ‘merit' goods and services and income redistribution.

Given that these different types of expenditure are likely to have differential impacts on growth, economists have found it helpful to use a simplification that distinguishes between expenditures that are likely to boost private sector production or productivity (known as ‘productive' expenditure) and those that do not (labelled as 'unproductive' expenditure). In practice, it is not straightforward to make these distinctions as, for example, much public spending may have different impacts depending on how expenditure programmes are designed and delivered. Section 3.2.2 highlights that, though there are some broad themes, empirical studies as to what expenditures can be counted as ‘productive' are not conclusive.


  • [1]Examples drawn from (Bassanini et al, 2001).
  • [2]For example, the environment makes an important contribution to people's living standards. As businesses do not have an incentive to take into account the impact of their activities on the environment, policies that make them face these costs may lead to an improvement in overall living standards. However, it may also lead businesses to reduce their activities, at least in the short run, and may result in a reduction in economic growth, as measured by Gross Domestic Product (GDP).
Page top