Fiscal Policy and Growth with Complementarities and Constraints on Government

Governments are subject to a number of constraints that affect their ability to set fiscal policy optimally. This paper considers two. Firstly, governments are inevitably imperfectly informed about the production technology of firms and household preferences which both determine what level and what composition of public spending are optimal. Secondly, governments are constrained in their ability to change various elements of fiscal policy, due to, for example, quasi‐fixed expenditure items such as social welfare benefits linked to entitlement conditions and interest payments that depend on the stock of public debt accumulated in the past. Other factors that constrain governments in this respect include limited administrative capacity (governments are only able to concentrate on a limited number of issues at a time) and limited political capital required for fiscal changes. We refer to this type of constraint as budget rigidities. In this paper we derive the optimal level and the optimal composition of public spending in the situation in which governments are constrained in their ability to alter both and know little about the relative growth benefits of different public spending categories. As a theoretical framework, we use endogenous growth models with public finance in which fiscal policy affects the long‐run growth rate of the economy. Previous papers that have used this class of models typically ignore these constraints on governments, and assume that private and public inputs are substitutes. We generate a number of interesting results with respect to optimal fiscal policy in the presence of budget rigidities, informational limitations, or both in the realistic case when private and public inputs to private production that are provided by the government are complements. First, we show that the optimal level of productive public spending and the composition are interrelated: in particular, the optimal level of spending is higher when the composition is suboptimal, and the optimal share of public resources allocated to public investment may be very low when the level of spending is either too high or too low due to budget rigidities. This result contrasts with the common perception that public investment is the most important public spending category for long‐run growth. Secondly, we show that imperfect knowledge of the government is much more likely a constraining factor for fiscal policy with complementarity. The number of parameters that determine optimal fiscal policy increases compared to the case when public and private inputs are substitutes, and some model parameters which are commonly perceived not to impact on optimal fiscal policy in some cases are shown to play indeed a role. These results demonstrate that determining optimal fiscal policy even under growth maximization is highly complex in practice since for some of these parameters robust empirical estimates are not available and again contrast with the more simple but likely unrealistic case of substitutability. The third contribution is to analyze fiscal reform when governments are imperfectly informed and when budget rigidities imply that the government is only able to implement piecemeal fiscal policy changes which take the existing fiscal policy as its starting point. Given that the optimal fiscal policy parameter values are unknown, the optimal size and direction of fiscal policy parameter changes are both unclear. We then show that by limiting the magnitude of policy parameter changes, budget rigidities in fact reduce informational requirement so that in most situations, the design of optimal fiscal reform only requires information about the direction of change but not about its magnitude. We further stress the need for better information by showing that with complementarity, fiscal reforms are more likely to reduce rather than to augment long‐run growth.

Misch, Florian, Norman Gemmell and Richard Kneller (2011), Fiscal Policy and Growth with Complementarities and Constraints on Government, ZEW Discussion Paper No. 11-018, Mannheim.