3.2  What are the challenges in achieving a good fiscal structure?

The previous sections argue that the structure of fiscal policy can have an important role to play to enhance long-term economic growth. The challenge for governments and advisors is to turn these insights into practical recommendations. However, in reality there are a number of constraints on the ability of governments to realise the full potential of growth-enhancing expenditure and well designed taxation structures.

First, the information requirements on policy-makers to exploit the insights of endogenous growth models are demanding. Policy-makers must first identify information asymmetries or externalities and understand when the market is not able to correctly price these externalities. Even if policymakers can identify externalities and understand the role of the market in addressing these externalities, they must also understand how government can improve the marginal productivity of the private sector’s physical capital and labour, the level at which public provision will crowd out more efficient private activity and the appropriate design of governance structures and institutions to deliver public services. Similarly, in order for policy-makers to be able to calculate accurately the full economic cost of taxation they must be able to understand the private sector response to taxation initiatives, including the extent to which a tax will change agents’ behaviour.

These demanding information requirements therefore create the potential for “government failure”. For example, even in those areas of fiscal policy that, a priori, might be expected to have the greatest potential for growth, there is controversy surrounding the strength of empirical estimates. For example, Zagler and Durnecker (2003) conclude that even for estimates of the impact of expenditures on research and development the literature is divided. Similarly, there is a considerable variation in estimates of the effects of government infrastructure spending on private investment and economic growth. For example, Perotti (2005) is critical of estimates of production and cost function-based estimates of the social returns to public capital expenditure which underpin Aschauer’s results (Aschauer 1989, 2000). Perotti points out that these approaches have not satisfactorily controlled for joint endogeneity of public and private production inputs and for dynamic macroeconomic effects, and that vector autoregressive modelling approaches that endeavour to tackle these problems have tended to reveal much smaller effects of public capital spending on economic growth.

Further, in many cases there may be regulatory or institutional solutions that could more directly tackle the market impediment. Assessment of fiscal policy initiatives should therefore be made in the context of comparative institutional advantages and should be assessed against an analysis of how private sector responses, coping mechanisms and institutions that would evolve in the absence of policy.

Where it is determined that fiscal policy is the best response, proposals need to be assessed under a robust cost-benefit framework. However, information constraints imply that there are limits to the accuracy of cost-benefit analysis. Aside from the difficulties discussed in the preceding paragraphs, full cost-benefit analysis is difficult because the cost of financing is crucial to determining the impact of the policy. Theoretical and empirical estimates of the effects of government expenditures on growth will be biased unless the cost of financing is explicitly taken into account.[26]

In the strict tax-smoothing approach discussed by CS First Boston (1995), this financing problem is solved by the idea that government expenditure decisions (based on efficiency grounds) can in principle be separated from the taxation financing decisions which in turn are assumed to also be based on efficiency grounds. However, government expenditure or taxation initiatives may induce behavioural responses that result in offsetting effects on growth. In those circumstances the assumption of independent and non-distorting financing decisions may be invalid and the financing of government fiscal initiatives at the margin will not be independent of the fiscal initiatives. However, fiscal institutions are not designed to enable identification of the true marginal economic cost of financing a particular project.

Consistent with this argument, Gemmell, Kneller and Sanz (2007) find that in OECD countries, fiscal variables (taxes, expenditures and budget deficit ratios) are often stationary implying that growth-affecting fiscal initiatives are often reversed. This implies that when governments have increased productive expenditures with growth-enhancing consequences, they have simultaneously tended to finance them with increases in growth-inhibiting taxes.

Another implication of the previous discussion is that evaluation of proposals should not simply be limited to new fiscal proposals but should also be applied to the existing expenditure base in order to determine whether the funding of new initiatives can be more efficiently provided by substitution of prevailing programmes rather than new taxes.

Political economy literature also highlights constraints on the political process providing support for proposals based on a sound cost-benefit analysis, and growth enhancing initiatives. A growing body of literature has, for example, examined the interaction between income distribution, political behaviour and growth. Alesina and Rodrik (1994) and Persson and Tabellini (1994), for example, evaluate how democratic institutions deal with distribution issues and how those decisions can influence taxation structures and growth. Recent changes to New Zealand’s electoral system may have had implications for the way politics and fiscal decisions interact to influence growth. Wilkinson (2004), for example, argues that incentives for sound value-for-money assessments may be even more problematic today under New Zealand’s Mixed Member Proportional representation (MMP) parliamentary political system given each coalition member has an incentive to provide support to their constituents.