In this paper, panel regression analysis is used to examine the fiscal impact of urban growth on per-capita or per-household expenditures of providing municipal services in the province of Ontario, Canada. Three variables are used to measure growth: households, population and assessment. Using a panel data set for 68 municipalities, we find that for the most part, urban growth has no effect on per-capita or per-household expenditures. The policy implications of these results are discussed.
Keywords: Urban economics, municipal finance, taxes, urban growth, municipal expenditures
Dans ce papier nous utilisons les techniques de regressions avec donnees de panel afin d'analyser l'impact fiscal de la croissance urbaine sur les depenses, par personne ou par famille, attribuees a la provision de services municipaux dans la province de l'Ontario au Canada. Nous utilisons trois variables afin de mesurer la croissance: les families, la population et revaluation. Avec les donnees de panel sur 68 municipalites, nous trouvons, en grande partie, que la croissance urbaine n'a pas d'effet sur les depenses. Les implications sur les politiques sont aussi discutees.
Mots cles: Economie urbaine, finance municipale, taxe, expansion
Local municipal officials, be they elected or staff, tend to be growth-oriented based on the belief that urban growth results in incremental revenues that exceed incremental costs that then allows the municipality to either decrease taxes or to make investments in new capital projects. A common refrain by elected officials is that in the past, because of growth, taxes were kept low and now, if there is little growth, taxes will have to increase. For example, with the 2009 recession, some of Ontario's fastest growing municipalities such as York Region were finding the debt cost to finance the past growth to be problematic and asked the Province to raise its borrowing limit in order to provide the infrastructure needed for the development. Critics of growth, such as Tim Gray of Environmental Defence, cite such examples as evidence of the non-sustainability of growth. However, other fast-growing regions, such as Mississauga, did not face the same problems because they required developers to pay the infrastructure costs up front. The higher costs therefore were due to the method of financing and not growth itself (1)
The above-mentioned municipalities are somewhat typical of the approach used by various municipalities to finance capital infrastructure. Where a capital asset has a long lifespan, debt financing is commonly used to apportion costs to future generations who are the beneficiaries of the capital asset. However, in cases where benefits accrue to new developments, be they residential or non-residential, development charges and front end cost-sharing agreements are commonly used. In some cases, municipalities use a mix of revenue sources such as a self-imposed percentage of the capital cost financed through operating costs. The approach taken is often dependent on whether the development is an infill or on the urban periphery. When there are positive spillovers to adjacent communities, municipalities will cost share with the neighbouring municipality or apply for funding from upper-level governments. It should be noted that other variables, such as infilling versus green development and housing density, also affect the fiscal impact of growth.
Regardless of the method of financing capital assets, municipalities must also consider their aging infrastructure, especially given the shortfall in maintaining that infrastructure. The Federation of Canadian Municipalities identified the municipal infrastructure deficit as $123 billion in 2006 and predicted the shortfall to exponentially increase if not properly managed to close to $2 trillion in 2067 (Mirza 2007). The estimates are consistent with estimates by other groups such as the Canada West Foundation and the National Research Council of Canada.
Although there has been much research on the effect of high density versus urban sprawl development on the cost of providing municipal services, the literature on the fiscal impact of overall urban growth is limited. (2) In a non-empirical paper, Kushner (1992) demonstrated that the effect of urban growth in the short-run was dependent on whether the municipality had excess capacity and in the long-run on the municipality's size relative to optimal size. Thus, a municipality could very well have 'tax positive' growth, for which incremental tax revenues exceed growth costs in the short-run but 'tax negative' growth in the long-run, if that municipality grows beyond optimal size.
The question of optimal size is of course an empirical question. Canadian studies such as Desbiens (1996), Sancton (1996), Kushner, Masse, Peters and Soroka (1996), Vojnovic (1997), Slack (2002) and Kushner and Siegel (2005) found that larger cities tend not to have lower per-capita expenditures indicating that economies of scale are not important in the delivery of municipal services. Holcombe and Williams' (2009) study of U.S. municipalities, likewise concluded that municipal expenditures are characterized by constant returns to scale and that amalgamations or secessions would have very little impact on per-capita expenditures. Derksen's (1988) study of municipal amalgamations in European countries and Boyne's (1992) study of amalgamations in the U.S. came to the same conclusion. These results are consistent with studies estimating scale effects for individual services such as environmental services (Jerrett, Eyles and Dufournaud 2002), police and fire services (Found 2012; McDavid 2002; Southwick 2005), general government (Southwick 2012) and garbage collection (Hamilton 2013; Slack and Bird 2013) that also found little evidence of economies of scale. (3)
In terms of growth studies, the research is rather limited. Ladd (1994), using data for 248 large U.S. counties, found that fast-growing areas not only experienced higher tax burdens but may have also experienced reductions in service quality. Danielson and Wolpert (1992), in their examination of 365 municipalities in northern New Jersey, also found that population growth was detrimental, as did studies of municipalities in the Greater Chicago Metropolitan Area by the DuPage County Development Department (1991) and by Oakland and Testa (1995). In a limited impact study, Glaeser and Gyourko (2005) examined the effect of urban decline on housing markets and levels of income concluding that the urban decline does not mirror urban growth. In their examination of the impact of annexation on spending, Edwards and Xiao (2009) conclude that if annexation is accompanied by higher density, then per-capita spending is reduced and, conversely, is increased if density is reduced. Their results may explain the inconsistent results from various studies.
Although there is a paucity of empirical research on the effect of overall growth on expenditures, there is a substantial amount of research on the impact of specific developments on expenditures, normally referred to as fiscal impact analysis. In general, these case studies find that the revenues from residential developments do not cover their costs to the municipalities. Burchell, Listokin and Dolphin (1994), in their review of the fiscal impact analysis studies, report a hierarchy where industrial and commercial development are most advantageous and single family housing the most disadvantageous.
As indicated above, with the exception of studies on urban sprawl and specific developments, there is very little empirical work on the overall impact that urban growth has on taxes or on the cost of providing local services. The purpose of this paper is to fill this void by measuring the impact of urban growth on per-capita or per-household expenditures for providing local services for municipalities in the Province of Ontario. To measure growth, we use three variables: household growth, population growth and assessment growth. To the best of our knowledge, our study is the first to do so. The results should be of assistance to municipal officials in determining their approach to urban growth.
Our sample consists of municipalities in Ontario and uses data for two years, 1996 and 2006. The data were obtained from the Ontario Ministry of Municipal Affairs' Municipal Financial Information Returns (FIRs). The financial information from the municipalities contained in the FIRs must correspond directly to both the financial records and the audited financial statements of the municipality. Similarly, the assessment data in the FIRs is required to match exactly with the assessment data received from the Municipal Property Assessment Corporation (MPAC). A ten-year period was selected to allow for lags between revenues and the provision of additional services. Ladd (1994) and Oakland and Testa (1995) used a seven-year and a ten-year period respectively. (4)
Comparing taxes between municipalities is complicated by variations within and between the various assessment classes of property such as single-residential, multi-residential, commercial and industrial. Even within a class such as single-residential, there is significant variance between the types of households, ranging from a small bungalow to the so-called three-story 'monster' home. In the past, comparisons were usually made in terms of a standard three-bedroom bungalow but the three-bedroom bungalow is not as commonplace now as it has been in the past. Furthermore, the Province does not make a distinction between the...