Printer Friendly

Responsibility center budgeting and management "lite" in university finance: why is RCB/RCM never fully deployed?

Despite its promise of revenue generation, cost reduction, and a host of other benefits, what is it about RCB/RCM that leads universities to deploy it only partially?


Starting in the early 1970s, a number of large research-intensive universities in North America began to experiment with an organizational and budgetary concept that came to be generically called responsibility center budgeting/responsibility center management or simply RCB/RCM. The attraction of RCB/RCM is multifaceted: by introducing a market model, it promises to promote generation of revenue, reduction of cost, enhancement of planning and budgeting through decentralization, and relocation of decision making as close as possible to those organizational levels with the most relevant expertise and the highest stake in the outcome (Hearn et al. 2006; Priest et al. 2002; Whalen 1991). The principal objectives of RCB/RCM are to link the total costs of programs with the revenue they generate, relocate responsibility for planning and budgeting, and in turn improve strategic performance in the allocation and generation of resources and the delivery of services (Strauss and Curry 2002).

Three decades later between 60 and 70 major universities in the United States and Canada (Ziskin 2014) and a few in Europe (Cekic 2010; Clark 1998, 2004; Oduoza 2009) follow the practice, albeit using several different but generically similar names, for example, "incentive-based budgeting," "value-centered management," and even "every tub on its own bottom." In general, private universities implemented the system before public universities (Brinkman and Morgan 1997). However, this rapidly changed; currently, more than half of the North American public universities in the top 50 of the Times Higher Education World University Rankings (Reuters-Thomson 2016) league table have adopted some form of RCB/RCM.

A study conducted by the University Leadership Council of universities in the United States and Canada that had installed RCB/RCM indicated that none had implemented it fully (Meshreky 2008). All were using it in hybrid combinations with other traditional budget models. Previous studies (Hearn et al. 2006; Lang 2002; Priest et al. 2002; West et al. 1997) anticipated this might be the case.


Sometimes as little as 10 percent to as much as 50 percent of an institution's budget is allocated on the basis of RCB/RCM. First, why is this occurring? Is there something fundamental in the theory of RCB/RCM that needs further refinement, or are there simply a few practical kinks that need to be worked out? Second, is the effectiveness of RCB/RCM limited in some way by the manner in which income flows to the university? In other words, does "resource dependence" (Birnbaum 1983; Leslie, Oaxaca, and Rhoades 2002) make a difference? Third, as universities step back from the full deployment of RCB/RCM, what complementary methods of planning and budgeting do they use? Fourth, is the partial deployment of RCB/RCM explained as much by limitations on the revenue side as on the cost side, or is the explanation to be found principally on one side or the other? Finally, is there a match (or mismatch) between the sources of revenue and the internal allocation of revenue under RCB/RCM?

Sometimes as little as 10 percent to as much as 50 percent of an institution's budget is allocated on the basis of RCB/RCM.

Data for this study were gathered in two distinct phases. The first involved documentary analysis of RCB/RCM models at 23 universities followed up by interview-based case studies of five public research universities in four different jurisdictions: the University of Lethbridge, the University of Michigan, the University of New Hampshire, Queen's University, and the University of Toronto. These institutions also differ in the extent to which they are "public." Three--the University of Michigan, the University of Toronto, and the University of New Hampshire--fit the definition of "public in name only": each receives less than half of its revenue from the government. The University of Lethbridge is on the other end of the "public" spectrum, with two-thirds or more of its annual revenue coming from government sources. Queen's University can be considered to be in between the "public in name only" and public categories.

The documentary analysis phase was undertaken to understand why each institution selected RCB/RCM. The contextual information gathered during this phase was subsequently applied during the interview phase. Institutional documents, including business plans, planning and budget committee reports, strategic plans, minutes of meetings, and fact books, were analyzed. In addition to institutional documents, reports from external debt rating agencies and national, provincial, and state faculty and university associations were reviewed along with state, provincial, and federal higher education finance reports. Comparable documents from several other universities that had installed RCB/RCM were also examined.

The second phase of research consisted of in-person and semi-structured interviews conducted in 2013-2014. In total, 62 individuals in senior leadership, administrative, and faculty positions were interviewed, including presidents, vice presidents, provosts, vice provosts, deans, associate deans, chairs, chief administrative officers, chief budget officers, and professors. These individuals were selected to participate based on their direct working knowledge of the particular RCB/RCM model in use at their respective university. Each in-depth interview lasted between 60 and 90 minutes and employed semi-structured and open-ended questions to allow informants to answer from their own frame of reference. Open-ended interview questions focused on the use of RCB/RCM by the participant's institution, the rationale for its selection, and the social, political, and economic context in which the institution operates.


As the research progressed from phase to phase it became apparent that the broad research questions outlined at the outset would be more clearly answered by breaking them down into a series of more specific queries that reflected empirical evidence. The specific questions are as follows:


The University of Toronto and Queen's University deployed performance-based budgeting prior to the installation of RCB/RCM. In the case of Toronto, the movement was evolutionary: limited versions had been in place for more than a decade beforehand. In the case of Queen's, which was informed by the prior experience of Toronto, the movement can be described as arrested evolution: implementation from the start included full application of RCB/RCM with later partial devolution. In both cases, the prior performance-based model continued to be used in conjunction with RCB/RCM, mainly to allocate "held back" strategic funds and set expense budgets for centrally shared services.

The experience of the University of Michigan was similar to the experiences of Toronto and Queen's: from performance-based budgeting to RCB/RCM for academic divisions and a few service units coupled with performance-based budgeting for centrally shared services. But at Michigan there was an intermediate step. Michigan initially planned an almost complete application of RCB/RCM under which budgets for service units would be decentralized and set through market competition (Courant and Knepp 2002). The plan was rescinded as several large central services--mainly in finance and human resources--were repatriated.

In all three cases--Toronto, Queen's, and Michigan--the percentage of annual expense not allocated by RCB/RCM was between 35 and 40 percent, and the alternative model used in conjunction with RCB/RCM was performance based.

At the University of New Hampshire incremental budgeting preceded the introduction of RCB/RCM. Initially, RCB/RCM was deployed in its nearly "pure" form in order to improve fiscal management, make informed decisions about budget allocations in support of the university's mission, and ameliorate suspicion and mistrust that had built up around the former process. That, however, was relatively short lived as a program planning and budgeting model was introduced to work alongside RCB/RCM, the complexity of which made it no more transparent and trusted than the prior model.

At the University of Lethbridge RCB/RCM replaced a system of formula budgeting. Upon its introduction, RCB/RCM was intended to determine between 60 and 70 percent of operating expense with the balance determined by incremental budgeting based on agreements that were put in place when RCB/RCM was installed. That arrangement has remained in place. There was no reversion to the prior model.

It appears, then, that three of the five case study universities never moved fully away from the budget models that preceded the introduction of RCB/RCM. Two did. One returned to the status quo ante by relying less on RCB/RCM than planned and coupling it with a different conventional model. And one planned from the start to couple RCB/RCM with an alternative model.


Although the track record of RCB/RCM as a means of reducing costs is not promising, all the universities in this study cited cost reduction as a reason for adopting it. However, of the 62 participants interviewed, only 22 (35 percent) reported having actually realized reduced costs. In terms of generating revenue, the track record of RCB/RCM was much better at all the case study universities. Eight-five percent of participants interviewed reported gains in revenue. Other studies have drawn similar conclusions (Curry, Laws, and Strauss 2013; Lang 2002; Massy 1996, 2003). Clearly, the track record is very lopsided. To understand this phenomenon, care was taken with certain definitions: unit costs versus net costs. In all five case studies RCB/RCM often forced balanced budgets. Thus, in terms of net costs, it would appear that RCB/RCM has a relatively beneficial effect. The appearance, however, is misleading; in all of the case studies, budgets were balanced predominantly by increasing revenue instead of by reducing unit costs.

In time, most of the universities studied moved to redefine the functional differences between direct costs, indirect costs, overhead costs, and infrastructure costs. In terms of RCB/RCM, direct costs are those that would have been located in academic unit budgets prior to the implementation of RCB/RCM, for example, faculty salaries, non-capital equipment, and the administrative expense of deans' offices.

The differences between indirect costs and overhead costs affect the degree of market behavior that RCB/RCM intends to promote. One could say that indirect costs are the same as direct costs but are located in other non-academic budgets. An example of this is the cost of removing hazardous wastes, which is a service that academic units may purchase from the university at large or from licensed private firms. Thus, there can be a market for indirect costs: academic units can control cost by adjusting the volume of service, seeking a cost advantage, or seeking a benefit advantage. In other words, insofar as academic units as "purchasers" are concerned, indirect costs are not fixed; they are variable in terms of the outputs that each unit seeks.

One could also say the same about overhead costs: they are the same as direct costs but are located in other non-academic budgets. But the analogy is incomplete because, unlike indirect costs, there is virtually no market for overhead costs. They are fixed costs that remain the same regardless of output. Academic units can adjust their volume in order to control the costs attributed to them under RCB/RCM protocols, but cannot seek unit cost advantage.

One university studied described the difference in terms of two metaphorical "cost" pizza pies. Each pie displayed the total costs of centrally provided shared services. The slices of each pie represented the costs attributed to individual academic units. For the indirect cost version of the pie, academic units had an incentive to reduce the size of the pie (unit costs) from 18 inches to 12 inches, add extra cheese, remove anchovies, and reduce the size of their individual slices. For the overhead version of the pie, academic units could not change the size of the pizza or any of the toppings, but they could change their proportional slices of the pie (volume) in a zero-sum game. This is not hypothetical. A research project still underway at the University of Toronto has shown that deans and campus principals are acutely aware of the difference between the two and are far more responsive to the incentive inherent in the indirect version (Morrin 2017).

Another university described the difference in terms of monopoly behavior. The provision of central services was not monopolistic in the case of indirect costs because there was at least the possibility of competition; however, for overhead costs there was not. Monopolies drive costs up as they possess price-setting power and offer few incentives for improving cost effectiveness. This inherently undercuts the theoretical cost cutting that RCB/RCM promises to deliver. Under the University of Toronto's RCB/RCM protocol, costs are organized into 12 "cost driver" categories. Of the 12 categories, only two allowed for any negotiation of price at the user level. Those two account for only eight percent of the university's spending on centrally shared services. For the remaining 10, there were no real markets.

What, then, is partial in the deployment of RCB/RCM? Planning and budgeting for centrally provided services in four of the five universities studied was outside the RCB/RCM protocol and followed other budget models, most often performance-based budgeting. This was more pronounced in those universities that attributed more of the costs of shared services to overhead than to indirect costs. One of the participating universities--the University of Michigan--significantly shifted the balance from indirect cost to overhead after several years of experience with its version of RCB/RCM, known then as value-center management (VCM). Currently, Michigan is "recentralizing" several services under an "administrative services transformation plan," thus increasing overhead costs and commensurately reducing indirect costs for which units are responsible. To cover the increased central costs, holdbacks and taxes are assessed on academic units. The funds thus recovered flow to the provost and increase the funding available each year for discretionary allocation.

One of the universities that was examined as part of the document analysis--York University--came relatively late to RCB/RCM and, from the start, systematically reinforced monopoly power by disallowing academic units to opt out of any centrally provided service (York University 2012). The observations from Michigan and York hint at another reason for the partial implementation of RCB/RCM: the deliberate creation of a market based on the indirect costs of centrally provided services can threaten collectively bargained labor agreements.

Infrastructure costs are "partial" in a different respect. Some universities include capital costs in RCB/RCM while others do not. For example, under RCB/RCM at the University of New Hampshire, the School of Business saved its operating surpluses over a number of years and solicited matching donations for a new building. At the University of Toronto, there is an operating budget to which RCB/RCM applies, and there is a capital budget to which it does not. Here "partial" has a somewhat different meaning: the university holds back a portion of revenue from academic units to create a "university fund" from which allocations may be made to either the operating budget or the capital budget.

At the University of Michigan a similar process takes place under a "general fund supplement" (GFS), which represents the additional support provided to a unit by the provost beyond the net of the unit's revenues and costs. The RCB/RCM model at Michigan was designed so that most units would require ongoing supplementation rather than functioning as "tubs on their own bottoms." This element of the budget model gives the provost strategic leverage in determining the budget and influence over the activities undertaken by units. Interestingly, the GFS takes into account both the historical measure of a unit's pre-RCB/RCM funding and the post-RCB/RCM growth (or decline) of funding.


In the case of some universities in this study, the answer is yes and involves internal and external factors. Three of the five universities studied have medical schools and teaching hospitals. To some extent the funding formulas used by their respective governments applied to medical programs in the same way they applied to all academic programs, in which case the RCB/RCM protocols worked. But not all government funding for the medical schools flowed through the allocation formulas. A large portion flowed through various "alternative funding plans," which can be roughly compared to funding medical schools as health maintenance organizations (HMOs). In practical terms, this funding bypassed RCB/RCM, which was therefore partial. In these cases, the universities served as de facto bankers with a fiduciary obligation to flow the funds through to the medical schools and teaching hospitals.

Funding was also partial in another respect. In two of the three universities in this study that have medical schools, relatively large numbers of clinical faculty in the teaching hospitals were included in collective bargaining for all university faculty. Salary and benefit costs were determined university-wide with ability to pay reckoned in terms of increases in government funding. Only by coincidence did the increases in alternative funding plan subsidies match the increases in centrally negotiated salaries and benefits. When they did not, the universities had to take revenue "off the top"--that is, outside the RCB/RCM budget model--to meet obligatory costs that the alternative funding plans did not cover.

The market effect of RCB/RCM on the generation of revenue is mainly found in three areas: enrollment volume, enrollment price (tuition fees), and contract research. There is a further effect on revenue that cannot be categorized as market behavior. All the universities studied reported an increased involvement by deans and faculty in fund-raising and in turn an increase in funding from private sources. The attribution of this revenue is simple so long as donors restrict their gifts to specific initiatives or programs. If not, allocation problems arise, leading to three "mismatch" problems. The first and second mismatches are matters of implicit incentive: deans and faculty, first, are not only encouraged to participate more in fund-raising--which would be explicit--but, second, are also encouraged implicitly to persuade donors to restrict their gifts to particular purposes within their own programs and perforce not to those of their fellow deans. One result of this behavior is a loss of financial liquidity that may, in turn, adversely affect the cost of borrowing. Bond rating agencies are beginning to significantly discount restricted income as collateral (Dominion Bond Rating Service 2009; Standard and Poor's Ratings Services 2012). Under RCB/RCM this makes no difference to academic units because the institution-wide cost of borrowing is invisible to them. This leads to what in practical effect is a fiduciary tragedy of the commons. Two of the universities in this study had become aware of this effect of RCB/RCM and were funding debt from revenue "off the top" before allocation to academic units.

Only unrestricted gifts act with certainty as a means of reducing net cost. Deans can direct unrestricted gifts to high-priority initiatives, which often include reducing deficits. Restricted gifts, however, depending on the donors' restrictions, may not serve to reduce costs at all. They may make new initiatives in effect free, and they may make it possible to move existing expenses "off budget," as in the case of endowed chairs, but in neither case are costs actually reduced. In other words, in terms of matches and mismatches, it is the donor and not the university that is determining the match and in effect setting institutional priorities. Recognizing this, two of the universities in the study introduced strict regulations for fund-raising that overrode the discretion of academic units under RCB/RCM with regard to restricted gifts for endowed chairs and scholarships.

The third fund-raising mismatch relates to the capacity to raise funds. Some programs--health sciences, for example--are relatively easy to raise funds for. Others--education and journalism, for example--are not (Rae 2005). Although under RCB/RCM the cost of fund-raising is charged to academic units, the final effect may be partial depending on the metrics of attribution, particularly with regard to the practice of inducing private gifts by matching them with institutional funds. Both Ontario universities in the study have done this on a large scale. If the terms of the match were one to one, RCB/RCM could be said to be fully deployed. But if the match were, say, two to one for those academic units where it was difficult to raise funds and one to two for those academic units where it was easier to raise funds, RCB/RCM would not be fully deployed.


The answer is yes, for reasons having to do with the basic structure of enrollment-driven funding formulas. Funding formulas are often more than abstract algorithms or political exercises in visible fairness and impartiality. They can have policy consequences, even if sometimes unintended (Darling et al. 1989). Formulas are based inherently on averages and on programs instead of institutions (Lang 2005). The ineluctable result, to take an example from the Ontario universities in this study, is that an undergraduate student in forestry at university X generates the same amount of funding as an undergraduate student in forestry at university Y. But the forestry program at university X may rely on growth chambers as teaching laboratories while the program at university Y relies on outdoor arboretums. Both, in terms of the quality of instruction, may be equally legitimate, but the costs associated with each program may vary significantly and be above or below the funding formula average. A strict application of RCB/RCM would force coercive isomorphism: one program, to balance its budget, would have to adopt the cost structure of the other. By holding some revenue back from RCB/RCM, a university can counteract the effects of isomorphism in academic programming.

Another explanation for the partial deployment of RCB/RCM can be found in the effects of the funding formulas themselves. Almost from the inception of RCB/RCM there have been concerns about its effects on interdisciplinary programs and service teaching. At three of the participating universities these concerns have become reality. Interdisciplinary activities under RCB/RCM are asymmetrical in the sense that some participants see RCB/RCM as a system that promises that actual costs will be recovered, while others see it as promising full attribution of generated revenue. This is not a problem if the programs in question are treated identically under their respective funding formulas. If they are not, however, RCB/RCM becomes an endemic impediment instead of a boon (Kirp and Roberts 2002; Massy 2003). To avoid this scenario, and in order for resource dependence to be neutralized, universities deploy RCB/RCM partially by attributing revenue as if their programs were funded equally, even when they were not.


* Holdbacks. Holdbacks can be found in some of the earliest applications of RCB/RCM. What this study indicates is that holdbacks take a variety of forms. From the cases examined, a small lexicon can be devised: some holdbacks are transitional and planned while others are transitional and coincidental. Some are strategic while others are pragmatic. Some are permanent and others temporary.

Transitional holdbacks are temporary and contingent. Their purpose is to finance a bridge between a prior budget model and RCB/RCM for academic units whose attributed revenue falls short of their gross expenses. The university thus acts as an underwriter of a budgeting and planning exercise for which the affected academic units are responsible. An example from this study is the University of Toronto's "university fund," the size of which was determined primarily by the projected costs of transition from the previous budget model to RCB/RCM over nine years. The University of Michigan and the University of New Hampshire have similar arrangements for holding academic units harmless under RCB/RCM by permanently supplementing the income of those who were sustainable before the introduction of RCB/RCM but whose budgets would otherwise be in deficit afterward.

Universities must seek to control their debt. To understand this, consider the alternative. RCB/RCM typically allows academic units to carry forward surpluses and deficits. This is a major incentive. It discourages the "spend it or lose it" behavior that is characteristic of most other forms of budgeting. It decentralizes multiyear planning and budgeting so that, in theory, academic units will be enabled to balance their budgets strategically instead of opportunistically. However, in the absence of underwriting by holdback, negative carry-forwards will add to the institution's debt.

Comparing the Ontario universities elucidates further the transitional holdback and explains the difference between planned and coincidental transitional holdbacks. In the University of Toronto's initial iteration of RCB/RCM, large-scale deficits at the unit level were not expected. The assumption was that the carryforward policy would be sufficient to manage local shortfalls. It was not until the university had decided to go ahead with full-scale RCB/RCM that the frequency and scale of local deficits were fully understood. The result was a holdback that was more coincidental than planned. On the other hand, York University and McMaster University, both of which began to investigate RCB/RCM about five years after the University of Toronto's implementation, instituted planned transitional holdbacks (McMaster 2007; York University 2012). At Queen's University, the approach was between planned and coincidental. The coincidence was a large debt that arose from recent capital projects that were pan-institutional. This led to a debate, instigated mainly from the bottom up by deans, about whether or not the debt should be serviced through RCB/RCM and, if so, how. The solution was to fund the debt centrally before the installation of RCB/RCM. In other words, the debt was held back.

Other universities have permanent holdbacks. Some of these are corrective and some are strategic. Corrective holdbacks are usually connected to resource dependence. To take an example from this study, a university recognizes the averaging effects of an enrollment-driven funding formula (Lang 2005), concludes that some academic units "cannot cut their suits to fit the cloth," and uses a holdback to correct for the anomalous effect of the formula. This might seem to be a transitional holdback, but it is fundamentally different because it assumes that some academic units will never be able to balance their budgets under RCB/RCM. The corrective holdback, then, is as fixed and permanent as a public subsidy via a funding formula would have been. In other words, the unit deficits are not moving targets. Once the subsidy from the holdback is in place, the RCB/RCM protocols apply fully.

Strategic and permanent holdbacks may also be deployed to address problems of resource dependence and create budget room for new initiatives. In this sense, the holdback is a strategic means of combating coercive isomorphism and maintaining institutional mission. One such example is that of the general service fund at the University of Michigan, under which most units are not intended to function as "tubs on their own bottoms" but instead receive supplements from the provost through the holdback fund (Hanlon and Schweitzer 2008).

Holdbacks may also be transitional when their purpose is to provide seed money for initiatives that in time will become self-funding as required by normal RCB/RCM protocols. In this case, the normal RCB/RCM protocols apply from the start on the cost side: all costs are attributed to the initiative. The holdback is used to substitute partially for income.

* Exemptions. Holdbacks are not the only reason for the partial deployment of RCB/RCM. Another frequent rationale for partial deployment is exemption. This occurs mainly on the cost side, when certain categories of expenses are not managed under an RCB/RCM regime. As shared services move away from being classified as indirect costs to overhead costs they also move away from the market signals that, in the theory of RCB/RCM, should act to incentivize efficiency and responsiveness by academic units as purchasers. The movement away from market signals can take two forms. One, as discussed, is the attribution of more costs as overhead, which limits academic units as users of services to controlling only volume instead of volume and price. The other form, which can be observed at the University of Michigan, involves the repatriation of some central services that had been decentralized under RCB/RCM and the simplification of cost attribution to flat university-wide rates.

Exemption on the revenue side occurs when a major source of revenue cannot be accurately and reliably attributed to academic units through normal RCB/RCM protocols. An alternative funding plan for a medical school is an example. Performance funding, which is sometimes called incentive funding, may have a similar effect depending on how it is structured (Dougherty and Reddy 2013; Lang 2014; Ziskin 2014). Performance funding rarely if ever operates at the level of academic units. A university may gain funding if, to use a common example, its rate of graduation rises. The rise, however, might be a net result of increases in some faculties, decreases in some faculties, and no changes in other faculties. The "pooling" in practical effect neutralizes RCB/RCM.

* Facilities and space. What about the cost of space? Is it a de facto exemption or not? Can there be real markets for space? In much of the RCB/RCM research literature--and in some of the universities in this study--the reduction of costs related to space is often cited as evidence of success. However, the issue of space allocation and cost is more complicated than it appears. It has three parts: space utilization, space brokerage, and capital budgeting. Space utilization can lead to a reduction of costs at the level of the academic unit. Through RCB/RCM, faculties become aware of the costs of the space they occupy, are charged for those costs, and have a strong incentive to occupy less space. Depending on how the costs of space are attributed, the incentive may be about volume and price or, more usually in the universities in this study, about volume only. As a faculty reduces its space inventory, its costs go down and the consequent savings can be used either to balance its budget or to fund enhancements. That is what RCB/RCM is supposed to do.

But the reduction in space utilization costs at the faculty level does not necessarily result in the reduction of costs at the institutional level. Why not? In terms of the theoretical foundation of RCB/RCM, the answer is that there may be no market demand for the space that a given academic unit has jettisoned. Every university in the study reported this problem. RCB/RCM thus demands a space brokerage or "bank," the function of which is to simulate a market by in effect "reselling" space that faculties return to the institutional space inventory. Sometimes this works for certain categories of space; typical examples are offices and small classrooms. In the case of special purpose space, inflexibility and lack of demand creates a de facto exemption because true markets for such space do not exist.

These problems may be mitigated in two ways. If some programs are in leased space, they may in practical effect "purchase" lower cost surplus space within the university's inventory through what amounts to an internal space brokerage. The cost of the space brokerage itself must be absorbed by the university at large as an overhead cost.

The other mitigation depends on the status of the capital budget. If the budget is real--that is, more than a schedule of projects to be undertaken--the capital budget functioning as a de facto market may in effect "purchase" existing space from the operating budget through the brokerage at a cost below that of new space. In this there is a real savings to the university's bottom line. This happened to some degree at the University of Toronto.

Absent mitigation, what most of the universities in this study do is factor the cost of unoccupied surplus space into the operating cost of all space and then pass that cost back to academic units either as an indirect cost or overhead cost. In other words, to use the University of Toronto as an example, the occupancy cost per net assignable square meter of space includes the cost of some space that an academic unit does not actually occupy, i.e., space that has been given up to the university-wide inventory of surplus space. Because the stock of nominally surplus space is much less than the stock of assigned space, there is still a savings to the academic unit, but it is not as large as it would have been if there were a real market for space. Thus, space is sometimes exempt and sometimes not. It is a test of some of the theoretical presumptions that RCB/RCM makes about the efficacy of market behavior as an incentive to reduce cost inventory. In a for-profit firm, for example, holding space centrally as an asset that no one else wants would be described as a liquidity problem.

* Strategic exemptions. Some exemptions are matters of strategic choice. Campus security is an example; there are no practical reasons on the RCB side of RCB/RCM for the costs of campus security not to be allocated to academic units as indirect costs. In other words, market behavior could have invited faculties to decide how much and what kind of security they want. This, however, poses a moral hazard: there is no incentive for individual faculties to guard against the risks faced by faculty, staff, and students wherever they may be on campus. In terms of complexity, it may also invite adverse selection because, even absent moral hazard, individual faculties may not have sufficient information about the security needs of other faculties to make informed decisions (Spence 2001).

Another example is the unified school of graduate studies; all of the five case study universities had such a school. Like campus security, there were no practical reasons for not attributing the cost of a unified graduate school as an indirect cost, in which case a faculty could decide to take full responsibility for its graduate programs or rely on a university-wide graduate school as a sort of "back room." Here the risk is adverse selection: some faculties under the financial imperatives of RCB/RCM may compromise quality at the expense of the reputation of other faculties and the university overall. This is the sort of problem that Kirp and Roberts (2002) identified at the University of Virginia. For example, a unified school of graduate studies may be less efficient than a decentralized arrangement, but for strategic reasons, a university--such as all those in this study--may exempt it from full application of RCB/RCM.

Finally, there are external exemptions. It is easy to understand why, for example, the costs of presidential offices and governing bodies are attributed to academic units as overhead costs: a university can have only one president and one governing body. The RCB side of RCB/RCM handles this well. There are, however, three major categories of expense that have to be exempt from the full deployment of RCB/RCM for reasons beyond the control of the university: debt service, insurance, and endowment management. Most major universities use debt to an extent that bond ratings are necessary. Those agencies and the lenders whom they inform do not and will not disassemble debt. Creditworthiness and the cost of borrowing are determined at the institutional level. Similarly, insurers will only assess risk institution-wide. Endowments are essentially perpetual contracts between donors and universities. Even when endowments are "restricted" to uses specified by the donor, fiduciary responsibility always resides with the institution. One could say that these categories of expense are something like holdbacks for shared services. They are in the sense that each is always funded "off the top," but they are not in the sense that the institution has no discretion whatsoever to decentralize their management or delegate responsibility for it. In the University of Toronto's early RCB/RCM period, these expenses were set aside in a category called "contractual obligations and policy commitments" that was funded before any strategic budget modeling was undertaken.



There are conflicting views in the literature about how size and complexity affect and are affected by RCB/RCM. One school of thought is that decentralization as a collateral effect of RCB/RCM is advantageous predominantly for large, complex universities (DeHayes et al. 1994; Hearn et al. 2006; Lang 1999; Stocum and Rooney 1997; Weir 1984; West et al. 1997; Whalen 1991). DeHayes et al. (1994) in particular found support for this position, noting that during the implementation of RCB/RCM at Indiana University the smaller regional campuses were excluded from the initial trial specifically because of their size. (However, at the University of Toronto, all campuses, including the two smaller campuses, were included.)

Conversely, West et al. (1997) found that the use of RCB/RCM was not restricted to those universities with budgets in excess of $550 million, and Stocum and Rooney (1997) argued that both small and large universities can use the concept of decentralization effectively. This, however, assumes that RCB/RCM is the only means by which a university can decentralize and that decentralization is the primary reason for a university to have selected RCB/RCM in the first place. For some of the universities in this study--the University of Toronto, for example--decentralization was one objective in installing RCB/RCM, while for others--for example, Queen's University--it was a by-product that was seen to be possibly as much a problem as a boon.

These views might not be as counterpoised as they seem. Institutional size and complexity are not the same. Decentralization offers little or no advantage in a university that is relatively homogeneous, regardless of size. A large university may have fewer faculties than a small university and therefore be less complex. Its organizational pyramid may be relatively steep. However, achievement of economy of scale may offer little or no advantage in a university below a certain size, regardless of complexity. Here, in considering some of the theoretical foundations of RCB/RCM, we should recall that in terms of economy of scale there is a Goldilocks effect: the size of a university may be too large, too small, or "just right" (Eastman and Lang 2001; Patterson 1999; Schumacher 1983; Sears 1983; Toutkoushian 2009). In terms of budgetary incentives, RCB/RCM has proven to be effective in promoting enrollment growth. If growth increases size to an optimal or "just right" level, unit costs should fall. But because RCB/RCM inherently rewards and promotes growth, it is not effective in reducing size to an optimal level or, in turn, reducing unit costs. Instead, it can get in the way of "rightsizing" or optimizing, as seems to have been the experience so far at the University of Toronto.


A number of the concepts promoted under RCB/RCM, such as continual improvement (Lang 1999); the establishment of a congruence of visions, values, and goals university wide (Stocum and Rooney 1997); and the application of subsidiarity to improve decision making, can be applied in many, if not all, universities. In other words, there are benefits. However, costs can offset those benefits. In each of the universities in this study, RCB/RCM was selected over other models as a device for improving the allocation and deployment of resources to realize efficiencies. In simple cost/benefit terms, the theoretical assumption is that RCB/RCM will produce either a cost advantage--the same benefits at lower cost--or a quality advantage--greater benefits at the same cost. In terms of return on investment, the assumption is that RCB/RCM will at least pay for itself, either by reducing cost or increasing revenue. These assumptions may or may not be true.

The full deployment of RCB/RCM is costly. Whether a university is large or small, complex or homogeneous, RCB/RCM requires sophisticated and expensive information systems. For example, at the University of Toronto, it was difficult, without manually rewriting programs, to link the financial information system, the human resources/payroll information system, and the student records system. This was a serious impediment to attributing the revenue and costs of graduate instruction since some students were also employed as research assistants and as sessional instructors. There are, of course, several reasons for a university to make major investments in information systems, but absent those investments, RCB/RCM's benefits will either not be delivered or will be delivered at high cost, which in turn makes the rate of return on investment lower. This might explain the apparent preference in cost attribution for overhead over indirect costs: calculating university-wide overhead charges places fewer demands on information systems.

At the University of Michigan, after the initial installation of its value-centered management model, a cost problem was recognized on the RCM side that materially affected the cost/benefit and return on investment equations. The problem was an increased and unanticipated need for greater financial expertise to manage RCB/RCM at the faculty level. Deans had to assume responsibilities normally associated with CEOs. This necessitated the upgrading of position descriptions and, in turn, higher levels of compensation (Hanlon and Schweitzer 2008), which resulted in less favorable cost/benefit and return on investment equations. In Michigan's case, some aspects of decentralization were reversed by recentralizing several financially related services and instituting a flat-rate overhead charge.


RCB/RCM, in theory, depends on internal markets to generate revenue and reduce costs. In practice, however, there is little evidence that universities deploying RCB/RCM have made unconditional efforts to create and stimulate internal markets or, when they have, that these markets have produced the desired results. The market incentive has two integral parts. The first--the RCB part--is the attribution of all costs--direct, indirect, overhead, and sometimes capital infrastructure--and all revenue to all faculties and campuses. This enables each faculty or campus to know its net revenue/expense position. Massy (2016) described this as being "market smart and margin conscious." The second involves allowing faculties and campuses to keep the revenue they generate and retain the savings they realize when they reduce costs. They may in turn reinvest the net gains in programs and services or set them against shortfalls between income and expense. All five of the participating universities devised cost accounting protocols for attributing revenue and indirect and overhead costs. Two--Michigan and Toronto--had sophisticated attribution models in place beforehand. Thus, the groundwork for entrepreneurial market behavior was successfully laid.

The track record of RCB/RCM as a means of generating external revenue is good. Faculties and campuses have responded by becoming more involved in fund-raising, adapting programs to markets, competing for research funding, and optimizing enrollment-driven revenue by paying attention to economies of scale and scope. Internally, the track record of market incentives is, at best, mixed. Few intended results have been realized, and some unintended and undesired results have arisen.

On the cost reduction side, it has proven difficult to create or at least simulate markets in which academic units can reduce the unit costs of centrally provided services either by using fewer services (volume) or by seeking cost advantages from the service providers (prices). In theory, market behavior should reduce unit costs, either by lowering internal prices or by allowing academic units to seek cost and quality advantages elsewhere. Practice has not followed theory. There are either no or very few real markets internally. Where there are markets, they are often monopolistic. There are some identifiable reasons for this. The reason most frequently observed in the five case study universities is that the costs of centrally provided services were attributed in ways that limited the academic units' ability to reduce them except by adjusting their volume. One university sought to mitigate this by creating a "shared services scoresheet" in which deans, as service users, could indicate the priority they attached to each service, the level of service they required, and the quality of service they needed. Otherwise, there were few real opportunities to seek price advantages. In some of the cases studied, collectively bargained agreements protected centrally provided services to such an extent that market behavior was trumped by de facto monopolies. In other cases, universities, in order to curb moral hazard and adverse selection, chose not to allow academic units to "shop for" or decline certain thus "exempted" services. This pattern of enforcing self-provision of centrally shared services that could otherwise be purchased elsewhere at lower cost is sometimes called "small numbers contracting" (Williamson 2009). This adds transaction costs; thus, costs of centrally shared services may not be reduced and may actually be increased.

On the revenue generation side, it has not been difficult to create internal markets. Instead, it has been difficult to control them. Just as the costs of centrally provided services are charged to faculty and campus budgets as users, faculties using the information produced by RCB/RCM can charge other units for "service teaching" they provide to them. Like centrally provided services, the providers of service teaching have no incentive to reduce costs as long as they can recover them from other faculties. The providing faculties rely on the cost recovery protocols of RCB/RCM while the using faculties rely on the revenue attribution protocols. In one of the universities in the study--Toronto--it took two years under RCB/RCM to resolve a service teaching dispute between two major faculties.

Internal markets have also been shown to promote predatory competition for students, the enrollment of whom may raise faculty revenue but not improve the university's bottom line.

RCB/RCM, because of its inherent decentralizing effect, expands local autonomy to the point where it can become a counterproductive centrifugal force. That is why Kirp and Roberts (2002) talked about the University of Virginia "breaking up" under an RCB/RCM regime and why Burke (2007) referred to "fragmentation." When that happens under RCB/RCM, a university needs a means of curbing the potential excesses of unregulated market behavior. To understand this let us imagine a modified version of the well-known Boston Consulting Group "stars, cash cows, dogs, and questions" strategic matrix and amend it slightly to include "stars, dogs, props, and problems" as classifications of academic units within a university. One faculty may, under RCB/RCM, choose to withdraw from service teaching--a "prop" to another faculty--in order to protect its "star" programs. If it were to do that, "star" programs in the other faculty may be put in jeopardy, either in terms of quality or net cost or both. The other faculty in response may have to spend more, thus raising its unit costs per student in order to protect quality.

To address the undesired consequences of internal market behavior, some universities in the study installed de facto academic "fair trade commissions" alongside RCB/RCM to regulate issues such as the repatriation of service courses and the intra-institutional competition for students. The commissions also prevent "price gouging" for intraprogram service teaching and research services. Specifically, the University of New Hampshire and the University of Lethbridge reported implementing similar processes to encourage service teaching, discourage "poaching" of students, and curb duplication of course offerings. Thus, in practice, contrary to theory, RCB/RCM may be ineffective at creating viable internal markets and in certain circumstances may disable them.


In terms of performance, the theoretical promise of RCB/RCM matches actual practice better on the revenue side than the cost side. This appears to be a key factor in the partial deployment of RCB/RCM. In this context, partial deployment may take two forms. One is defined by scope: RCB/RCM is not applied to all categories of a university's revenue and expense. The other is defined by the extent of its application to those categories that are within its scope. In both, partial deployment may be a matter of strategic choice, feasibility, necessity, or default. It is as if RCB/RCM were bipolar. Its theory with regard to revenue is based mainly on quasi-market incentives. But what seems to be an almost universal practice of holding back some revenue and limiting market competition for central services indicates that the incentives are not as strong as theory presupposes. With regard to cost, the theory is based on an expectation that academic units, acting in response to market incentives, will improve efficiency by seeking cost advantages, benefit advantages, or both. But by attributing more costs as overhead (fixed) than indirect (variable), by exempting large categories of expense from nominal RCB/RCM markets, and by regulating entrepreneurial behavior, practice blunts the impact of theory as it applies to unit costs.

It is as if RCB/RCM were bipolar. Its theory with regard to revenue is based mainly on quasi-market incentives. But what seems to be an almost universal practice of holding back some revenue and limiting market competition for central services indicates that the incentives are not as strong as theory presupposes.

This suggests a particular area in which theory may be usefully refined. Cost, in most of the previous literature about RCB/RCM, is construed as net cost. The case studies show that when RCB/RCM has worked in terms of cost it has done so almost exclusively in terms of volume: fewer services are provided or used but with no change in nominal price as a reflection of unit costs. The refinement would be to subdivide the RCB/RCM cost equation into two distinct parts: net cost and gross or unit costs. A similar refinement might apply to revenue where the subdivision would be between pre- and post-holdback revenue.

A broader refinement could be to think of RCB/RCM in terms of contract theory. In the first instance, RCB/RCM is a means of changing relationships between the university as a principal and its faculties as agents with the objective being the optimization of allocations of net revenue and expense. RCB/RCM is a means of solving the problem of asymmetrical information by sending "signals," as defined by Spence (2001), that will change the behavior of faculties as purchasers of centrally provided services. The theoretical nub of the problem may be that because holdbacks, exemptions, and other practices render RCB/RCM "partial," signaling cannot fully correct the problem of asymmetrical information. Faculties as purchasers do not really know the costs of centrally provided services. They know those costs only as "prices" that are attributed to them. Nor do they have as much information about how amounts held back are determined and allocated as they do about their RCB/RCM income and expense. Thus, the result is two "contracts." In no university in this study was there any recognition of a need to correct these asymmetries. Because there was no recognition, no effort in the form of signals was made to mitigate the inherent collision between the two, except through after-the-fact review processes. This may be a manifestation of a point made by Massy (2007) about the lack of priority that RCB/RCM in practice assigns to the quality of academic programs. The review is in practical effect an alternative "signal" from the university as principal to its agent--a faculty, school, or campus--about quality.

With or without refinement, it is improbable that RCB/RCM will ever be fully deployed. The impediments that stand in the way of functioning internal markets are immovable, mainly for pragmatic reasons. No matter how sophisticated models for attributing costs become, they will be expressed as average costs. Deans, campus principals, and department chairs, however, think in terms of marginal costs. That they do so should not be surprising because the attribution of revenue under RCB/RCM either actually is or seems to be marginal and for that reason acts as a powerful incentive. This explains the empirical finding that the track record of RCB/RCM is much better in generating revenue than in reducing costs and that even when RCB/RCM reduces net costs it almost always does so on the revenue side.


Birnbaum, R. 1983. Maintaining Diversity in Higher Education. San Francisco: Jossey-Bass.

Brinkman, P. T., and A. W. Morgan. 1997. Changing Fiscal Strategies for Planning. In Planning and Management for a Changing Environment: A Handbook on Redesigning Postsecondary Institutions, ed. M. W. Peterson, D. D. Dill, L. A. Mets, and Associates, 288-306. San Francisco: Jossey-Bass.

Burke, J. C. 2007. The Fragmented University. In Fixing the Fragmented University: Decentralization with Direction, ed. J. C. Burke, 1-27. Bolton, MA: Anker.

Cekic, O. 2010. Budgeting and Organizational Culture: Exploring Responsibility Center Management (RCM) and Cultural Change at a Public University. Saabrucken, Germany: Lambert Academic Publishing.

Clark, B. R. 1998. Creating Entrepreneurial Universities: Organizational Pathways of Transformation. Oxford: Pergamon.

--. 2004. Sustaining Change in Universities: Continuities in Case Studies and Concepts. New York: McGraw-Hill.

Courant, P. N., and M. Knepp. 2002. Budgeting with the UB Model at the University of Michigan. Accessed July 5, 2017:

Curry, J. R., A. L. Laws, and J. C. Strauss. 2013. Responsibility Center Management: A Guide to Balancing Academic Entrepreneurship with Fiscal Responsibility. Washington, DC: National Association of College and University Business Officers.

Darling, A. L., M. D. England, D. W. Lang, and R. Lopers-Sweetman. 1989. Autonomy and Control: A University Funding Formula as an Instrument of Public Policy. Higher Education 18 (5): 559-83.

DeHayes, D., R. Dick, L. Finkel, J. Froehlich, D. Gaither, R. Keck, R. Koleski, R. Martin, R. Pfile, D. Robbins, P. Rooney, and M. Rosentraub. 1994. Responsibility Center Management: An Assessment of the Last Five Years: 1989-1994. Review committee paper, Indiana University-Purdue University Indianapolis.

Dominion Bond Rating Service. 2009. Rating Report: McMaster University. March 18. Accessed July 5, 2017:

Dougherty, K. J., and V. Reddy. 2013. Performance Funding in Higher Education: What Are the Mechanisms? What Are the Impacts? San Francisco: Jossey-Bass.

Eastman, J., and D. Lang. 2001. Mergers in Higher Education: Lessons from Theory and Experience. Toronto: University of Toronto Press.

Hanlon, P. J., and G. L. Schweitzer. 2008. Budgeting with the UB Model at the University of Michigan. Updated version. Accessed July 5, 2017:

Hearn, J. C., D. R. Lewis, L. Kallsen, J. M. Holdsworth, and L. M. Jones. 2006. "Incentives for Managed Growth": A Case Study of Incentives-Based Planning and Budgeting in a Large Public Research University. Journal of Higher Education 77 (2): 286-316.

Kirp, D. L., and P. S. Roberts. 2002. Mr. Jefferson's University Breaks Up. Public Interest (Summer): 70-84.

Lang, D. W. 1999. Responsibility Centre Budgeting and Responsibility Centre Management in Theory and Practice. Higher Education Management 11 (3): 81-111.

--. 2002. Responsibility Center Budgeting and Management at the University of Toronto. In Incentive-Based Budgeting Systems in Public Universities, ed. D. M. Priest, W. E. Becker, D. Hossler, and E. P. St. John, 109-36. Northhampton, MA: Edward Elgar.

--. 2005. Formulaic Approaches to the Funding of Colleges and Universities. In International Handbook of Educational Policy, ed. N. Bascia, A. Cumming, A. Datnow, and K. Leithwood, 371-91. Manchester, UK: Springer.

--. 2014. Clark's Triangle, Fiscal Incentives, and a New Relationship between the State and Universities. Comparative & International Higher Education 6 (Fall): 54-63.

Leslie, L. L., R. L. Oaxaca, and G. Rhoades. 2002. Revenue Flux and University Behavior. In Incentive-Based Budgeting Systems in Public Universities, ed. D. M. Priest, W. E. Becker, D. Hossler, and E. P. St. John, 55-92. Northhampton, MA: Edward Elgar.

Massy, W. F. 1996. Resource Allocation in Higher Education. Ann Arbor, MI: University of Michigan Press.

--. 2003. Honoring the Trust: Quality and Cost Containment in Higher Education. Williston, VT: Anker.

--. 2007. Using the Budget to Fight Fragmentation and Improve Quality. In Fixing the Fragmented University: Decentralization with Direction, ed. J. C. Burke, 122-44. Bolton, MA: Anker.

--. 2016. Reengineering the University: How to Be Mission Centered, Market Smart, and Margin Conscious. Baltimore: Johns Hopkins University Press.

McMaster, M. 2007. Partnerships between Administrative and Academic Managers: How Deans and Faculty Managers Work Together. Association for Tertiary Education Management. Accessed September 13, 2013.

Meshreky, M. 2008. Beyond RCM--Innovative University Budget Models. Custom research brief, University Leadership Council, Education Advisory Board.

Morrin, A. 2017. Decentralized Budgeting and Its Impact on the Quality of Decision Making at a Large Research Intensive University. Doctoral thesis, University of Toronto.

Oduoza, C. F. 2009. Reflections on Costing, Pricing and Income Measurement at UK Higher Education Institutions. Journal of Higher Education Policy and Management 31 (2): 133-47.

Patterson, G. 1999. Findings on Economies of Scale in Higher Education: Implications for Strategies of Merger and Alliance. Paper presented at the EAIR 21st Annual Forum, Lund University, Sweden.

Priest, D. M., W. E. Becker, D. Hossler, and E. P. St. John, eds. 2002. Incentive-Based Budgeting Systems in Public Universities. Northhampton, MA: Edward Elgar.

Rae, B. 2005. Ontario: A Leader in Learning. Toronto: Ministry of Training, Colleges and Universities.

Reuters-Thomson. 2016. Times Higher Education World University Rankings. Accessed July 14, 2017:!/page/0/length/25/sort_by/rank/sort_order/asc/cols/scores.

Schumacher, C. 1983. The Problem of Scale in Higher Education. In Economies of Scale in Higher Education, ed. S. Goodlad. Guildford, UK: Society for Research in Higher Education.

Sears, K. 1983. Economies of Scale in Higher Education. In Economies of Scale in Higher Education, ed. S. Goodlad. Guildford, UK: Society for Research in Higher Education.

Spence, A. M. 2001. Signaling in Retrospect and the Informational Structure of Markets. Nobel Prize lecture, Stockholm, Sweden, December 8.

Standard and Poor's Ratings Services. 2012. RatingsDirect: Queen's University. December 19. Accessed July 5, 2017:

Stocum, D. L., and P. M. Rooney. 1997. Responding to Resource Constraints: A Departmentally Based System of Responsibility Center Management. Change: The Magazine of Higher Learning 29 (5): 50-57.

Strauss, J. C., and J. R. Curry. 2002. Responsibility Center Management: Lessons from 25 Years of Decentralized Management. Washington, DC: National Association of College and University Business Officers.

Toutkoushian, R. 2009. An Economist's Perspective on the Privatization of Public Higher Education. In Privatizing the Public University: Perspectives from Across the Academy, ed. C. C. Morphew and P. D. Eckel, 60-87. Baltimore: Johns Hopkins University Press.

Weir, M. 1984. Efficiency Measurement in Government. Bureaucrat 13 (2): 38-42.

West, J. A., V. Seidita, J. DiMattia, and E. L. Whalen. 1997. RCM as a Catalyst: Study Examines Use of Responsibility Center Management on Campus. Business Officer (August): 24-28.

Whalen, E. L. 1991. Responsibility Center Budgeting: An Approach to Decentralized Management for Institutions of Higher Education. Bloomington: Indiana University Press.

Williamson, O. E. 2009. Transaction Cost Economics: The Natural Progression. Nobel Prize lecture, Stockholm, Sweden, December 8.

York University. 2012. Report on the Working Group on Budget Modeling Stage 1. May 9. Accessed July 5, 2017:

Ziskin, M. 2014. HEQCO Project on Outcomes-based Funding of Higher Education. Draft report, Indiana University Project on Academic Success. March.

by Darren Deering and Daniel W. Lang


DARREN DEERING is currently a director of administration at the University of Toronto and has held various senior leadership roles in industry and higher education. He received his Ph.D. from the University of Toronto in 2015. His research interests focus on budgeting, management, governance, and organizational design in higher education institutions.

DANIEL W. LANG, now emeritus professor in theory and policy studies at the University of Toronto, was senior policy advisor to the president; vice provost, planning and budget; and vice president, computing and communications. His principal areas of interest are institutional planning and management, finance, accountability, and quality assurance.
COPYRIGHT 2017 Society for College and University Planning
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2017 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Deering, Darren; Lang, Daniel W.
Publication:Planning for Higher Education
Article Type:Report
Geographic Code:1CANA
Date:Apr 1, 2017
Previous Article:Fiscal structure and public college and university credit ratings: an exploratory analysis.
Next Article:Beyond the headlines: the mechanics of a tuition reset.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters