Annex 3: Expenditure Need

What is expenditure need?

The Constitution refers to “making Equalization payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation.”

Some interpret this objective to mean taking into account not only the differences in fiscal capacity (revenue-raising ability) of provincial governments, but also their expenditure need. Two jurisdictions with very different program needs for their population (e.g., different health care, education, and social services pressures) may not be able to provide comparable levels of public services if their Equalization entitlements only take into account the differences in their ability to raise revenues.

The concept of expenditure need encompasses two notions. First is the relative demand for public services. A province with a larger proportion of elderly people requiring expensive health care, with a higher share of low-income residents needing social services, or with a higher ratio of school-age children has a greater need for public services. This is referred to by experts as the workload facing their public services. It is usually measured by the socio-demographic and economic characteristics of a province’s population.

The second dimension of expenditure need is the financial cost of providing a given set of services. Some provinces face higher wages, property costs and material expenses. These can be due to different wage rates in the public sector, higher costs of serving remote or sparsely populated areas, diseconomies of scale where fixed costs must be spread among a smaller population, or even to differences in the efficiency and effectiveness of government service delivery.

The complexity of measuring expenditure need

The most rigorous way of incorporating expenditure need in Equalization, as developed in Australia, is to construct the expenditure equivalent of the Representative Tax System (RTS). A Representative Expenditure System (RES) measures, for each spending area of a provincial government, the per capita expenditures it would have to make in order to provide typical provincial services of national average quality to its residents, assuming a national average rate of efficiency.

Like the RTS, an RES requires constructing and costing a standard for each spending area – a typical basket of public programs and services, representative of the average preferences and practices of all provinces. Such a basket would be dominated by the weight of the larger provinces, Ontario and Québec. Workload and costs in each sector would then be measured to arrive at an estimate of what each provincial government would need to spend to provide the standardized basket of services. Depending on whether that amount exceeds or falls short of the national average, the province is said to have an expenditure disadvantage or advantage. For a similar fiscal capacity, a jurisdiction with an expenditure disadvantage would receive higher Equalization entitlements. Similarly, a jurisdiction with an expenditure advantage would find its Equalization entitlement reduced.

For example, the measurement of expenditure need for child care, elementary, and secondary education would require first constructing a basket of education services representative of the average across Canada (teacher-pupil ratios, availability of special education services, etc.). One would then measure the cost of providing this basket of services across different provinces, taking into account workload differences (e.g., proportion of school-age children in the population, children requiring special education, children with English or French as a second language, etc.) and differences in the cost of providing that standardized basket (e.g., wage and salary conditions, real estate prices, and material costs).

All these steps require numerous judgement calls and adjustments in order to fairly measure the expenditure need for each province. For example, if a province had more rural or remote areas, adjustments would have to be made for class size and the additional cost of transportation. If one province had been more generous in its wage settlements with teachers’ unions, adjustments would have to be made to determine the wage costs to the province had it paid national average wages, adjusted for differences in the cost of living and supply and demand conditions.

Policy neutrality

Not only would an RES be extremely complex to construct, involving many judgements about how to appropriately measure expenditure need, but it would also be very difficult to design a system that would maintain policy neutrality.

Most experts agree that, on grounds of policy neutrality, differences in expenditure need should be equalized only to the extent that they are beyond the control of governments. Otherwise, Equalization distorts incentives for sound public management. That is why, for example, the RTS does not recognize debt or debt service levels as a cost of providing public services. To do so would provide an incentive for provinces to incur higher levels of debt. Similarly, when considering the possibility of designing an RES, it would be necessary to protect against compensating a government that has conceded higher salaries to its public sector, or that adopts less efficient modes of delivery for public services, compared with neighbouring jurisdictions.

The inability in practice to easily sort out what governments can and cannot control creates much concern about moral hazard and perverse incentives. How can one ever be sure how much of a province’s expenditure disadvantage with respect to health care or secondary education is due to workload or costs beyond its control?

Intrusiveness into provincial decision-making

The need to disentangle the sources of expenditure disadvantages requires extensive data and answers to many detailed questions from provincial departments administering public services and programs. Moreover, this data must be comparable. And it must be constructed on the basis of an agreed-upon standardized basket of services.

Beyond the extensive data demands, there are real issues about intrusiveness into provincial decision-making once this data enters the public domain. Comparisons would inevitably be made between actual provincial expenditures and the representative expenditure standard, putting pressure on governments that provide less than the standardized basket of services, or that provide services at a higher cost. Some governments would object to the use of a representative basket heavily weighted by the preferences of Canada’s largest provinces. After all, the purpose of federalism is to allow regional differences in policy preferences to be reflected in the services offered by provincial governments.

The gain in public accountability would have to be weighed against concerns about federal intrusion into areas of provincial responsibility. More importantly, it would have to be weighed against the materiality of the contribution of an RES to Equalization.

Materiality

Ultimately, the case for incorporating expenditure need into Equalization must rest on a presumption that it would make a significant difference to the level and allocation of Equalization payments. If it does not, there is no point in devoting public funds to the expensive conceptual and data investments required.

The Panel is not aware of any comprehensive attempt to measure expenditure need in Canada. It has heard many conjectures, based on fragmentary evidence, relating to a few key spending sectors (e.g., health, education, social services). It has noted the work of Anwar Shah suggesting that had expenditure need applied in 1991–92, it would have lowered total Equalization payments overall, reduced the share of all receiving provinces except Québec, and raised Québec’s share.¹ Other efforts have suggested that non-receiving provinces, which are often heavily urbanised, would have a much greater expenditure need (relating to both workload and cost factors) than some receiving provinces. There have also been suggestions that expenditure need disparities in Canada are likely much smaller than those arising from fiscal capacity, but this has not been supported by detailed investigation.

The Panel’s assessment

On balance, the Panel considers that the case for incorporating expenditure need into Equalization has not been made. There is no conclusive evidence that it would have a material effect on the size and allocation of Equalization payments. It would be premature to recommend a comprehensive effort at constructing an RES, given the conceptual and data difficulties involved and the issues it would raise with respect to jurisdictional responsibilities.

One of the principal concerns with introducing expenditure need assessment into the Equalization program is that it must be done on a comprehensive basis. Recognizing expenditure need may be easier and less controversial to do with specific transfer programs such as the Canada Health Transfer or the Canada
Social Transfer.

Table of Contents	Next