Today’s higher education institution business models are under increasing pressure in response to multiple factors, including major demographic shifts and enrollment declines, reduced state support, and increased efforts to provide more institutional aid to more students. As institutions seek ways to maintain long-term financial viability, more are exploring alternate budgeting models.
Auburn University, Auburn, Ala., began an exploratory process for a new resource allocation model in late 2011. The initial conversation was little more than a casual chat with a consulting firm to discuss options. In late 2012, the potential for adopting a new model gained momentum when Auburn’s provost at the time reached out to the CFO to collaborate on a project that would help revitalize a stagnant allocation process. After a lengthy implementation phase, which included due diligence by several constituency groups on campus, Auburn adopted its RCM model for full use in developing its FY17 budget.
This article details our process, including questions and concerns that emerged from various stakeholder groups, adjustments we made, and lessons learned that may prove useful for other institutions considering an RCM model.
When I began working at Auburn as a senior accountant in November 2005, it employed an incremental model for the largest portion of its overall budget. The most tenured members of senior administration could not remember a time when Auburn did not operate under this method, which involved pooling our largest sources of revenue (state appropriations and tuition) along with a few other smaller revenue sources into our base (or general fund) budget. These pooled resources were allocated to provide expense limits for units within the base budget. These funds represented about 65 percent of our overall budget, and it was revised annually based on expected changes in revenue and mandatory or strategic expense increases.
Despite the impression that Auburn operated strictly with an incremental budget, some components of our business practices did reflect key principles of an RCM-style model. For instance, we have always allowed funds to be carried over and allowed units that generated funds at the local level to retain those funds within the units. Because of these and some other practices, I assumed our transition to RCM might be less of a culture change.
For numerous reasons, at the outset the provost and deans were in favor of a new methodology for allocation. One reason was that we had become increasingly dependent on tuition in the aftermath of the Great Recession. In 2008, tuitionand fees made up approximately 44 percent of our unrestricted budget. By 2012, that had increased to 63 percent. In addition to changes in the mix of revenues, marketplace competition for nonresident students had also increased dramatically. Our enrollment model is built on a certain percentage of nonresident students and recruiting and retaining those students had become more difficult than ever before.
Commencement of Auburn’s RCM project—what we call our Strategic Budgeting Initiative—launched in early 2013 with the establishment of a steering committee to devise the guiding principles under which the model would operate. Our provost and CFO led the committee, with representatives from both the academic and administrative organizational structures. Because academic leadership wanted to know more about how the budgeting process would work, one of our initial guiding principles was to ensure that the model would be transparent. The second crucial principle, vetted through the steering committee, was to make sure we would build a model that was simple.
Plan Your Model Rollout
Implementation of the new model was broken into four main phases.
Phase 1: Assess the flow of funds. The first phase was to complete a funds flow assessment. To have any concept of where we wanted to go, we needed to fully understand how the funds were coming into the university and how those resources were being allocated to the different units on campus.
Given the magnitude that tuition and fees played in our financial model, it wasn’t surprising that the colleges were producing well over half of our revenue. What was a bit shocking was the level of subsidization needed in some of those disciplines that had a higher cost of instruction. In the initial funds flow assessment, the vast majority of our 12 degree-granting colleges or schools required some level of subsidy before factoring in university overhead costs.
Phase 2: Build the model. The second phase of implementation was the actual model build. The project consultant worked closely with the steering committee to identify which behaviors or functions we wanted to incentivize, levels of allocation at which those incentives might be created, grouping administrative/overhead units, and metrics by which the overhead units would be allocated to the revenue-generating units.
This phase was time-consuming because it required completing several iterations that would be acceptable to the steering committee and meet approval from the rest of our campus. It has always been central to Auburn’s brand to provide a positive undergraduate experience, so most of our model allocations were built around incentivizing instruction for undergraduate students by setting the allocation percentages for our undergraduate tuition pool at 70 percent to the college of instruction and 30 percent to the college of major/record.
Under the model, 70 percent of our state appropriations were allocated based on in-state student tuition, further incentivizing the instruction mission. The remainder of the state appropriations was allocated proportionate to the sponsored activity generated. So, while there were incentives for research productivity, they were overshadowed by the incentive to produce credit hours.
Our decision to group administrative units may be contrary to what one might expect. Instead of trying to group solely by the organizational chart of the institution, we tried to look at the function of the subunits within each major unit and determine the most appropriate home. We ultimately settled on six pools: academic and student services, administration, alumni affairs and development, facilities, sponsored research, and universitywide support. To illustrate that we were trying to place the subunits based on function and not by organizational chart, business and finance has units that fall under four of the six pools.
Finding the proper metric by which the administrative units would be allocated was also a challenge. We were up-front with everyone that there was no perfect variable we could use to allocate the costs of the central units, but we aimed to use the data point that made the most sense. For example, the variable chosen to allocate the academic and student services pool was “student credit hours instructed.”
One of the more interesting discussions might have been the variable that made the most logical sense: “square footage to allocate the facilities pool.” The point of contention was that our model did not account for quality or type of space (old vs. new, or lab vs. office vs. classroom). We ultimately convinced the detractors that the model allocation methodology was not intended to assess the quality of a space, rather the purpose was to serve as a mechanism by which to allocate the budget for facilities.
Phase 3: Build consensus. Our third phase of implementation was consensus-building. This is the area where we struggled most to gain traction. Most of our deans agreed that we needed to do something different since the allocation of resources was not reacting to current trends in enrollment and provided them no incentive to be strategic in their financial planning. Some were also fearful of how a deficit might impact their peers’ perceptions of them.
The greater campus community—specifically faculty—felt its opinions were not included for consideration and wanted a chance to vet the decisions made by the steering committee. The university senate asked for the opportunity to put together an ad hoc committee to study the model and offer input on changes.
After nine months, the university senate’s ad hoc committee delivered its recommendations. Chief among them was to advocate for a weighting structure for credit hours to quantify the cost of instruction. This was done in an effort to improve the large deficits of some of our higher cost of instruction colleges that had been modeled during the process. In the end, that recommendation was closely voted down by the senate.
The senate ultimately accepted the other recommendations of the ad hoc committee: (1) to provide for governance structures at the college level to determine no one department was negatively impacted by a dean’s decision; and (2) to ensure that the colleges and schools would have some security blanket from year-to-year with the amount of subvention provided. The administration agreed to review those on a case-by-case basis.
Not long after receiving the senate’s recommendations, we made the decision to move forward. While the review process of the university senate certainly slowed the momentum of the project, it was important to give this group an avenue to participate and provide the feedback it felt would make the model more acceptable to faculty. This process also demonstrated that shared governance remains an honored practice at Auburn and allowed each constituency to feel it can have a say in decisions that affect the entire campus.
Phase 4: Deploy the model. The fourth and final phase of implementation was deployment of the model. This phase took approximately 18 months. Part of this time was spent by our budget and planning services on learning the Excel workbook used to administer the model’s allocation methodology, along with renovation of our homegrown budget development tool. A significant amount of time was spent training the rest of campus on how the model worked and how to manage it with this tool.
Test Before Going Live
We actually spent FY16 building out reporting in what was considered our “shadow year”—a concept recommended by our consultant as a normal practice for such allocation redesigns. While this shadow year allowed us to test the numbers against the methodology one more time, it was more about trying to make sure that our office could manage the operational aspects.
For instance, we had to rebuild our homegrown budget development tool to accommodate the methodology. We also had to build quarterly reporting around the data variables used for allocation of the revenue and administrative costs.
During our shadow year, we were able to accomplish all these tasks, which ensured that we were capable of moving forward with the RCM model as adopted by the steering committee and the rest of the campus. No additional changes were made to the model because the numbers still showed that the methods were sound. During summer 2016, we built our first budget with the model in place for FY17.
The final construct of our model was to break allocations into four major categories: revenues (both direct and allocated); expenses (direct); central unit allocation (overhead allocated to revenue-generating units); and mission enhancement funds (taxes on academic revenues to cover subvention and strategic investments). We have now been live with our model for three full years and are currently operating in our fourth year.
As expected, there have been some bumps along the way.
Governance issues. As one example, organization of the governance committees was lacking during the first two years of being live with the model. Because all our university committees are a cross section of the campus, with the different constituency groups having representation and with annual rotations on and off, the education of the committees on the specifics of the allocation methodology has been the biggest challenge.
Each year, most of our initial meetings have been spent walking through what role the committee plays in the process and diving into model mechanics. In addition, the overall direction of the committees in the initial stages was lacking due to leadership transitions. Consequently, past practices of top-down approaches to budget decisions endured and the committees, feeling they weren’t adding value to the process, simply rubber-stamped what was presented.
The most recent committee process provided reason for optimism, specifically in the committee with oversight of the administrative unit budgets. With clear budget parameters and guidelines, the committee had a stake in the conversation and began asking difficult questions about proposed budgets and making recommendations about reductions. In some cases, it actually recommended units increase their proposals because it felt the budget was inadequate to support the university’s mission.
Unique operational areas. Another challenge that exists is how to operate with the two divisions that help support our land-grant mission. With so much interconnectedness through appointments at the Alabama Agricultural Experiment Station (AAES) and Alabama Cooperative Extension System (ACES), there are difficulties ensuring that funding is appropriately handled between all appointments.
Currently, we are still struggling to keep those stated assignments intact. Because the two units are viewed independently by our state and receive separate appropriations with very little other unrestricted revenue, inevitably more of the burden is shifted to Auburn’s main campus budget. This has a global impact on the other colleges and schools that do not have affiliations with AAES and ACES.
While maintaining autonomy for AAES and ACES, we have been trying to take a more holistic approach when we review the operations of our colleges and schools that do have affiliations with these units. We did not contemplate the impact these operations would have, and RCM implementation suffered in the first two years by not having a plan in place.
While this may be a problem specific to land-grant institutions, other schools may have similar considerations to make about other operational areas of the institution (e.g., auxiliaries, athletics, or other separately budgeted divisions) and how they influence decisions around allocating overhead.
Tweak as You Go
We are now coming up on our five-year review period, when we will unlock the model and begin testing different assumptions about allocation variables or percentages. When we started implementing our RCM model, we said that while we had to give it a chance to work, we would need to review it periodically to make sure we were incentivizing the right behaviors or to make changes in support of different strategic initiatives. Because we placed a heavy incentive on teaching credit hours, we are contemplating lowering our allocation of tuition to the college of instruction. We’re also considering revising the participation rate assessed on the revenues of the academic units. The initial rate was appropriate at the time, but we need to revisit it.
In that same vein, we’re also looking at which revenues should be assessed. Currently all unrestricted revenues (with the exception of finance and administration recovery) receive the charge. This includes all locally generated revenues such as fees collected for online tuition, sales of products, or clinical operations. If we were to shift course and assess those fees on allocated revenues only, it should result in an increased incentive for the colleges and schools to be more entrepreneurial. With an effort to improve our research profile, I would expect to see a greater emphasis placed on the research mission. We could accomplish this by increasing the amount of state appropriations allocated proportionate to sponsored activity.
Unfortunately, my hunch about a smooth transition to RCM was wrong. Even though there was consensus among academic leaders that a change was needed, there was also anxiety. Despite a desire by the majority of key stakeholders to attempt a more data-driven approach to resource allocation, it proved a challenge to change the mindset of those who saw the allocation process as an annual exercise instead of as a way to plan for the future.
In addition, because RCM requires a level of accountability that was largely nonexistent in our previous model, some are still trying to navigate those waters. So, one thing any institution must do before it entertains a change of this magnitude is to assess the readiness of the campus for the cultural change that will come and factor that into the implementation process.
Skill sets. You also must evaluate the skill sets of those who are in business management roles and work to complement their strengths with the expertise of other leaders. For the institution to be successful, you have to ensure that no one unit is lacking the talent to manage the financial portfolio of its area, and you have to be willing to coach them through the challenges they may face. We have made strides to hire capable personnel in the necessary areas, worked to provide them with the tools needed to be successful, and are continually evaluating ways to improve.
Technology tools. We also considered our readiness to implement a new budget system at the same time we were implementing the RCM model. In the end, we did not feel we could incorporate the additional change to the process without creating turmoil with our campus business personnel, so we postponed its consideration until our more recent implementation of Anaplan, a web-based enterprise platform for business planning.
As you consider undertaking a system conversion, think about what you ultimately want out of a software tool as you design the interface for your campus community (see sidebar, “Don’t Forget the Tech Supports”). Provide others the opportunity to participate in the RFP process and user acceptance testing and be sure to deliver training materials that can make things easier for them. Identify a project leader who has a skill set with a mix between technical and functional. I was fortunate to have an individual who met that criteria as well as an entire staff in the budget and planning services office that had varying skills to help make the implementation a success.
Collaboration. Finally, the most important factor for success is to create a collaborative environment with all stakeholders on campus. From the beginning, we had a broad-based group of campus leaders involved in the RCM implementation process, including some faculty on the steering committee.
However, as we progressed through the project it became evident that we should have opened the lines of communication with a broader group of the faculty constituency. The chief complaint from faculty leadership was a lack of understanding of the model concepts. Engagement with the group much earlier in the process might have allowed us to implement the model a year earlier. That said, the additional time was crucial for us to provide extra opportunities for our campus to develop a better understanding of the math behind the model.
Overall, Auburn’s RCM implementation was a success, despite a few bumps along the way. Our primary goal—to create transparency and accountability with our resource allocation methodology—was accomplished. Unit leaders have an understanding of how the revenues and administrative overhead are being allocated. As a result, deans are more actively engaged in recruiting students, decisions about unit-awarded tuition discounts, and strategic decisions around academic program creation, as well as monitoring expense trends.
The university community is also becoming more engaged in a connected financial planning process that integrates multiple functional areas in order to evaluate our ability to successfully fund long-term strategic objectives.
BRYAN ELMORE is assistant vice president, budgets and business operations, Auburn University, Auburn, Ala.