Last-Minute Spending in Municipalities Facing Potential Amalgamation. Visible in the Budgets or Blurred by Budget Overruns?

Amalgamation of local governments alter the premises for economic decisions in the units involved. The impending merger shifts the responsibility for future economic commitments to the new, enlarged unit. Consequently, local decision-makers are given an incentive to overspend in order to accrue localized benefits at the expense of their amalgamation partners. Such effects have been analyzed in various empirical contexts, including amalgamation reforms in the Nordic and continental European countries, Japan, and the USA, as instances of the “common pool problem”. A key tenant is that distributive policies concentrates benefits in a specific geographic constituency and finances expenditures through generalized taxation. Increasing numbers of “districts” – in the present context, partners to the amalgamation project – increases the incentive for free-riding behavior, the so-called law of 1/n. We contribute to this literature by asking whether such “hoarding” behavior can occur even before actual amalgamations are decided, driven purely by territorial uncertainty and an anticipated threat of potential “extinction” And if so, whether such hoarding is primarily driven by budgeted overspending or by budget overruns. Based on data on the budgets and accounts of Norwegian local governments through the ongoing local government reform, we test a number of specific empirical assumptions on how contextual factors may lead to changes in hoarding behavior through an interim period, between the time when the reform was firstly announced and until final decision-making on amalgamations. Firstly, it will be in the interest of individual jurisdictions to monitor economic decisions of prospective amalgamation partners, in order to avoid an uneven outcome. Such partner monitoring can be expected to become increasingly acute through the duration of the interim period, because the exact identity of amalgamation partners may be unclear in the outset. However, monitoring costs can be assumed to vary between different types of economic decisions. Overspending decided in the annual budget may be more transparent and easier to monitor than decisions made through the year. Therefore, we expect the share of overspending decided in the annual budget to decrease as partner monitoring increases, whereas overspending decided through the year will increase correspondingly. Secondly, hoarding may be reflected in increased investments and/or by increased current expenditures accompanied by a weakening of the net surplus of running expenditures. When facing a potential extinction, on one hand the political incentives for last-minute investments are stronger than for last-minute increase of current expenditures. Investments on the other hand requires planning and may be more visible to potential amalgamation partners. The possibilities for hoarding are thus less extant for investments than for current expenditures. Accordingly, we expect the hoarding primarily to be reflected in a weakening of the net surplus of running expenditures in the interim period. Empirically, we apply a difference-in-difference logic and compare municipal fiscal policies before (2012-2013) and after (2014-2016) the reform process started across municipalities in different categories depending on variation in the likelihood of merger and on the other independent variables.