An Approach To Costing Core Government For First Nations
Prepared for Indian Affairs and
Northern Development Canada
Allan M. Maslove, Ph.D.
The purpose of this discussion paper is to assist Indian and Northern Affairs Canada (INAC) in its thinking about the costing of core governance in First Nations communities. There are two main aspects to this exercise that are addressed in the following discussion: 1) the appropriateness of the functions identified as core governance activities in FN communities and key characteristics affecting the costs of these functions; and 2) suggestions on a methodology to undertake a costing exercise for core governance.
The paper is organized into four main sections. The first section will briefly comment on two papers provided by INAC which essentially describe the Department's understanding of the core governance functions. The second main section will identify and briefly discuss parameters that influence the unit costs of providing these functions and factors that affect the level of spending. The third section suggests and discusses methodologies that might be employed in undertaking the costing exercise. A final section offers some concluding comments and recommendations.
INAC's identification of First Nations' core governance functions are developed in two internal papers authored by Brent Sprague. The first is "Costing the Administration of Government — a Research Review" (March 2007). This paper distinguishes among the terms "governance", "government" and "core governance". It then begins to identify the elements of cost for core functions. It concludes with an examination of the (very limited) literature on the cost of central government in small communities, whether Aboriginal or not. The second paper, "Identifying Core Governance Functions To Effectively Target Governance Support Funding" (January 2008) moves to a much more detailed identification of the functions of core governance in three areas: legislative (functions related to the political body), administrative (functions related to the administration), and judicial (functions related to the body that resolves disputes).
These papers, particularly the second, result in a very detailed and complete identification and classification of the range of activities that would constitute core government in a small FN community. To that extent it is easy to answer the first question identified in the Introduction in the affirmative; the appropriate functions for a "package" of core government activities have been identified. To implement the methodology I suggest in this paper a further level of disaggregation may be required; this will be discussed below.
A few comments on these papers will lead to the costing discussion in the following sections. The identification of the three functions and the three corresponding branches of core government, plus the detailed enumeration of the activities under each, suggest, at first glance, that one could simply attach a price to each element, add them up, and thus construct a reasonable budget for core governments. (To be clear, this is not a view that I am attributing to the author of the two papers.) But, core government (and government in general) is not the same as stamping out identical widgets; every case is somewhat different, at least at the margins. For example, the level and style of community engagement will vary from community to community. The same will be true of engagements with other governments (other First Nations, federal, provincial or municipal), and of feedback mechanisms that the government chooses to establish (ranging from informal constituency feedback to formal evaluations).
A valid distinction is made between governance and government. For purposes of costing however, this may not be a material distinction. If government, and especially core government, is the set of bodies and activities through which governance is established, advanced and maintained, it may not be practical or necessary to speak of cost of government as distinct from costs of governance.
The studies referenced in the two discussion papers (as well as other literature I consulted), tend to conclude/observe that costs are higher for smaller governments (usually municipalities) than for larger ones. The implication of viewing the data in this way is that in some sense, larger (municipal) governments are more efficient than smaller ones. I suggest there is another, more helpful, interpretation which I will discuss below.
In this section we discuss variables that could influence costs and in the next we develop a methodological approach to estimation. Before embarking on this discussion, however, it is useful to clarify one point. The question being addressed is the cost of providing core government functions; it is not about how these costs should be distributed. Consequently we will not get into discussions of how INAC grants should be structured, what proportion should FN governments be asked to contribute, nor issues of equalization in grant systems. Once it has a clear picture of core government costing, INAC may wish to consider the design of its grant programs accordingly, but the grant system is not an issue of core government costing per se, and therefore, it is beyond the scope of the present discussion.
A number of factors will affect the costs of core government, some of which are exogenous and some of which reflect choices of the FN government. Among the first group are geographic location, population size and the proportion of the population living off-reserve. The second group of factors mainly includes decisions made by the FN government having to do with the style and process of governance it desires to implement and maintain.
Geographic remoteness and northern locations increase cost structures facing FN governments. This relationship is strongest and most clearly seen in the services sectors of the government. For example, in remote northern regions construction costs for houses will be higher as will some operating costs (e.g., heating, electricity); similarly road construction and maintenance costs will be higher. In the performance of core government functions, remoteness as a cost driver may be less strong and less obvious but it is still likely to be the case. Some direct costs (e.g., office heating and electric bills) will increase, but these are relatively minor components of the core government budget. By far the largest line-items in core budgets will be salaries and associated human resource costs. These are often higher in remote areas (e.g., Northern premiums) but not necessarily uniformly so. For example, if the FN government draws its core government staff primarily from within the community, the remote/Northern premium may be smaller than if the government were to "import" staff from (southern) urban areas. Another cost that will clearly be higher for remote communities will be travel costs incurred by officials involved in core government functions (and indeed sectoral functions as well). On the other hand, FN communities that are in close proximity to large metropolitan areas may also find their human resource costs are pushed upwards because wages and salaries in the regional labour market are higher, and the FN government faces stiffer competition from alternative employers.
Core government activities are probably as close as one gets in public administration to the classic economic example of overhead or fixed costs. The establishment of a government requires some prerequisite functions (inputs) even for the smallest community population, and these costs are likely to remain fixed (with perhaps only minor increases) as the community's population grows, at least over some range. Significant increases in these costs are likely to come in discrete steps when population increases sufficiently to require more elaborate or more complex core government functions. This is what lies behind the earlier noted finding in the literature that per capita central government costs are lower for larger communities than for smaller communities.
The reduction in per capita costs as population grows reflects the spreading of these fixed costs across a larger base. It is not an indicator of the superiority of larger governments as it is sometimes interpreted in the literature. (The fixed input factor may, however, suggest opportunities for governments to seek alternative modes of provision in order to reduce their overhead costs. One alternative is to collaborate with other governments in the provision of these overhead functions in order to spread the fixed costs over a larger population base. One particular form of doing this is to purchase the activity in the market. Operating a payroll/benefits system would be an example.) A cost function that might accurately model this pattern is a fixed cost with very low (or zero) marginal cost with respect to population, punctuated by discrete increases as population achieves critical threshold sizes.
FN government functions increasingly involve community members who opt to live off-reserve. Probably the most notable example is the inclusion of off-reserve members in elections. Such arrangements likely increase the costs of core governance. That is, a First Nation community of a given population will likely find some of its core governance costs increase as a greater proportion of that population lives off-reserve.
Style and process of governance:
As already noted, core government activities are not produced in standard (identical) units. While the functions may be common across governments (e.g., legislative support, dispute resolution), the forms in which they are produced may vary reflecting differences in the cultures of the communities, their sizes, or other characteristics. Citizen engagement, for example, may vary from the very informal (the chief and council members talking to members on the street) to highly structured and formal (facilitated "town hall' meetings and engagement processes integrated with policy formation). Similarly, feedback mechanisms may vary from informal conversations with providers and consumers to formal evaluations of programs. Some governments may choose to provide their councils with little in the way of legislative support while others may have more extensive data collection and research services to underpin council deliberations. To cite a final example, governments choose different levels and forms of engagement with other governments whether Aboriginal, municipal, provincial or federal.
Each of these choices, and other similar ones, will impact the cost of providing core government. A question for INAC will be to decide what standards of activity should be recognized in the costing exercise as essential to core government and what will not be recognized and, in effect, treated as elective (i.e., non-essential) add-ons.
A Suggested Costing Methodology
The existing literature (limited as it is) has, for the most part, suggested two approaches to estimate the costs of government for FN communities, namely comparative analyses of (small) municipalities and other FN communities with self-government agreements in place (SGFN). Comparisons are constructed at a fairly aggregate level, using total expenditures or broad categories of expenditures.
The methodology suggested here incorporates these two approaches within a broader framework. It also develops estimates at a more micro (disaggregated) level. The framework is intended to conform to three criteria that the costing formula should respect. First, the formula should strike a balance between cost minimization and creating sufficient fiscal capacity for a First Nation government to meet its core needs. Second, while the costing formula is rooted in micro-level activities or tasks, recipients should have the discretion to allocate funds according to local priorities. Setting local priorities includes, among other things, determining which core government functions the community wishes to undertake, the methods it wishes to utilize, and the levels at which it wishes to undertake these activities. Third, while the formula may be somewhat complex, it should be transparent insofar as the techniques used to estimate each component are clear.
The methodology approaches the cost estimation task from four directions — pricing market-purchasable activities, examining municipal comparators, examining SGFN comparators, and costing a prototype model. The first of these approaches is categorically distinct from the other three in terms of the quality and reliability of the estimates generated. Estimates produced in this manner would be expected to be of relatively high quality because, in many cases, well organized competitive markets exist. If so, the observed market prices would indicate minimum but realistic costs for these services. Note that this approach does not imply that FN governments must buy these services rather than produce them internally; it only establishes reasonable values.
Core government functions (tasks in the Sprague 2008 terminology) are disaggregated into clearly definable categories or groupings of activities (e.g., payroll processing, drafting of laws/by-laws, IT support for a legislative body). The first step for each grouping is to consider whether these services are commonly available for purchase in a market. These "market-purchasable" activities include services/tasks provided by the legislative, administrative and dispute resolution (judicial) sectors of core governments that can be reasonably be considered as items that are available in open markets. Indeed, in many instances (whether Aboriginal or municipal) they are purchased. Examples include (but are certainly not limited to) information technology, some human resources tasks (such as payroll services, drafting job descriptions), auditing and evaluation, legal services (including the drafting of laws and by-laws), and research (including data assembly). For each of these activities or tasks, one might indicate the quantity that might reasonably be purchased (that is, the contract specifications) with a relatively high degree of clarity and specificity, and price these purchases in the relevant market.
Other activities/tasks, for various reasons are almost always internally produced. Acquiring these services through market transactions may not be feasible or desirable because of a number of possible factors: security/secrecy requirements; the difficulty or impossibility of writing contracts that anticipate contingencies or changing requirements; the difficulty in monitoring contract compliance; or simply the fundamental public nature of the activity (e.g., the function of the legislature itself cannot be contracted out though some legislative support tasks may be). In these cases the minimum reasonable costs are more difficult to estimate and, therefore, the three other estimating procedures would be utilized in order to capture a reasonable envelope of cost estimates: estimating the core government expenditures incurred by size-comparable municipalities, estimating core government expenditures of self-governing First Nations (SGFN), and developing hypothetical budgets for defined tasks for a prototype community (or a range of communities varying by size, location, and off-reserve population). As in the case of the market-based activities, a micro approach is suggested in which separate estimates are calculated for typical tasks or collections of similar tasks (such as administrative support for chief/mayor and council, intergovernmental affairs activities, operating a financial management system).
To elaborate, in the case of municipal governments,Ontario municipalities file Financial Information Returns in a standard format which provides this type of disaggregation. One could extract and analyse the data for only small municipalities (say in three categories – population of 500 or less, 501 – 1500 and 1501 – 3000). In line with the earlier discussion, one might also wish to classify municipalities by geographic location, paying special attention to remoteness and metropolitan proximity. One might undertake a similar analysis of First Nations under self government with attention to location and the proportion of the First Nation members not living in the community (a proxy for off-reserve). The remaining approach would be the hypothetical construction of core government activity profiles for a community (or communities) with defined characteristics (size, location, off-reserve population) and the determination of budgets for those prototypical services.
These estimates, arrived at through the three estimating approaches, would comprise an envelope of estimates of core government service costs not in the market-based category. Obviously the narrower the range of variance across the approaches, the more confidence one might have in their accuracy.
Following these approaches one would then have available 1) market-based cost estimates for a range of activities or tasks, and 2) an average or possibly a range of estimated costs for the remaining activities or tasks that would constitute a core government system for a FN community. The summation of these cost estimates would be the aggregate cost estimate of core government activities for a defined (by size, location, etc.) First Nation community. This approach is summarized in Chart 1.
The methodology suggested here would generate high-quality estimates of the costs of core government because 1) it builds the aggregate estimates from disaggregated groupings of activities; 2) it utilizes, where possible, market data that provide firm cost estimates for activities that are more clearly definable and thus amenable to contracting; and 3) it estimates other costs based on data from three independent approaches.
Initially this will be a more complex and costly methodology than is involved in the more aggregated examination of relevant comparators. However, maintaining the model over time should be relatively straightforward once the activity/task categories are defined and the data sources are established.
Ultimately, one must keep in mind that any exercise that attempts to estimate costs for abstract and non-standardized core government activities will involve ranges of judgment and uncertainty. Using market prices where possible will diminish but certainly not eliminate this uncertainty.
- Brown, Trevor L. & Matthew Potoski, "Transaction Costs and Institutional Explanations for Government Service Production Decisions," Journal of Public Administration Research and Theory, Vol. 13, no. 4, 2003.
- National Centre For First Nations Governance
- Ontario Ministry of Municipal Affairs, Financial Information Returns
- Sprague, Brent, "Costing the Administration of Government — A Research Review", INAC, March 2007.
- Sprague, Brent, "Identifying Core Governance Functions To Effectively Target Governance Support Funding", INAC, January 2008.
- Williamson, Oliver E., "Transaction-\kCost Economics: The Governance of Contractual Relations", Journal of Law and Economics, Vol. 22, no. 2, 1979.
Thank you for your feedback