Introduction

This User Guide provides information and explanations that will be of assistance to researchers interested in using the data on Canadian provincial government expenditures and revenues we have published here.

Our goal is to provide data on provincial government finances for every province spanning the period 1965/66 to the current fiscal year. As of this date, we have completed the collection of data from 1965/66 to the current fiscal year for six provinces, namely, British Columbia, Alberta, Saskatchewan, Manitoba, Ontario and New Brunswick. For two of the remaining provinces (Nova Scotia and PEI) our data begins with fiscal year 1980/81. The data for Newfoundland and Labrador is currently under construction. We are working to complete the database and will report progress with future editions of this User Guide.

In a research paper published in 2016 (Canadian Provincial Government Budget Data, 1980/81 to 2013/14 ), we discussed the need to gather data on provincial government finances because Statistics Canada terminated what were once long time series on provincial finances. With data spanning the period 1988/89 to 2008/09, Statistics Canada terminated the publication of its Financial Management System (FMS) data series describing details of provincial government finances. Some of those earlier time series are still available in hardcopy in a publication titled Public Finance Historical Data 1965/66-1991-92. A copy of this publication can be downloaded from this site. With luck, one may also find these data – extended to the end of fiscal year 1996/97 – by exploring old versions of Statistics Canada’s electronic data bases. However, finding those data is getting harder to do. We got lucky once and so downloaded those data to this site.

At the time we started our work the only other long time series on provincial finances still being updated was published in the Fiscal Reference Tables produced by the federal Department of Finance. These data are still being updated. In our 2016 paper we discussed problems with those data and why we believed our data represented an improvement.

Since we published our 2016 article, Statistics Canada has provided new versions of provincial and territorial government finances in Table 10-10-0017-01 (Canadian Government Finance Statistics for the Provincial and Territorial Governments) and Table 36-10-0450-01 (Revenue, Expenditure and Budgetary Balance – General Governments, Provincial and Territorial Accounts). Unfortunately, these new presentations of provincial government finances have several limitations relevant for researchers.

First, the new data presentations have only been extended backward to 2007. This leaves a 21-year gap between what is available in Public Finance Historical Data 1965/66-1991-92 and what is available from Statistics Canada now. If one is fortunate in being able to access old versions of Statistics Canada’s electronic data bases, data from 1965/66-1996/97 becomes available shrinking the gap to 10 years. A significant gap nonetheless remains. Relying solely on data currently being updated by Statistics Canada means researchers interested in understanding long-term trends or observing budget impacts of events such as the 1980-81 recession or the impacts on the provinces of federal fiscal retrenchments in the late 1990s are out of luck.

Second, the data reported in Statistics Canada Table 36-10-0450-01 (Revenue, Expenditure and Budgetary Balance – General Governments, Provincial and Territorial Accounts) is presented on a calendar rather than fiscal year basis making it challenging to compare to other sources of provincial finances such as public accounts, provincial budgets, or the Fiscal Reference Tables.

Third, these data define expenditures in ways that are not easily translated into measures of spending on public goods such as health care, education, and social assistance; descriptions of spending familiar to voters and so important for understanding what drives political outcomes. Simply put, the descriptions of revenue and spending categories do not align with how governments report their spending to voters nor how voters understand and evaluate their government’s choices with respect to these amounts.

Since the publication of our article in 2016, a very good source of data on provincial government revenues has been made available by Finances of the Nation . Many of the data series published by FON are available back to 1965-66 and so check the box for needing to provide researchers with long time series. Importantly, these data are based on those produced by Statistics Canada and so are presented using a common accounting definition. This makes it possible to compare the size of revenue categories across provinces. Finally, the data reports revenues in categories familiar to taxpayers and voters, categories such as personal income tax and property tax revenues.

A weakness of the FON data is the lack of information on spending categories. Only total program expenditures and debt charges are available in the FON dataset. Missing is information on spending allocated to health care, education, or any other key spending programs offered by provincial governments. This is true as well for the data reported in the Fiscal Reference Tables. Thus, neither of these sources of data is useful for describing provincial government expenditure choices, how those choices have changed over time, or the sensitivity of those choices to economic or political events. For researchers interested in how revenue shortfalls or booms, how debt accumulation or relief, and how political variables might influence decisions with respect to the provision of health care, social assistance, or education services, a different source of information is required.

The data we have compiled includes information on provincial government spending on the ‘big three’ spending categories of Health, Education, and Social Services. Users will observe significant changes in absolute and relative levels of spending on these categories over time, the type of variation attractive to anyone interested in testing for the existence of political business cycles and partisan influences on spending.

The data we have compiled, and are in the process of compiling, provides researchers with long time series. The shortest of our data series currently begin with fiscal year 1980/81 and the longest begins with fiscal year 1965/66. Our long-term goal is to try to report data for all provinces back to fiscal year 1965/66.