These minutes reflect
discussion and debate at a meeting of a committee of the
Minutes
Senate Committee on Finance and Planning
Tuesday, November 30, 2004
2:30 – 4:15
238A Morrill Hall
Present:
Charles Campbell (chair), Kendal Beer, Rose Blixt, David
Chapman, Arthur Erdman, Daniel Feeney, Scott Fine, Steve Fitzgerald, Thomas Klein,
Joseph Konstan, Michael Korth, Cleon Melsa, Kathleen O'Brien, Richard
Pfutzenreuter, Terry Roe, Charles Speaks, Kate VandenBosch, Susan Van Voorhis, Warren
Warwick
Absent:
Calvin Alexander, Seth Haskell, Joshua Jacobsen,
Lincoln Kallsen, Ian McMillan, Diane Parker, Thomas Stinson, Alfred Sullivan,
Michael Volna
Guests:
Julie Tonneson (Office of Budget and Finance)
[In these minutes: the budget model]
The
Budget Model
Professor
Campbell convened the meeting at 2:30 and noted that the sole agenda item was
the University's budget model.
Professor
Feeney began by commenting that everyone on the Committee knows of the view of
the Academic Health Center Finance and Planning Committee, whose
recommendations were brought to this Committee earlier. He suggested there is an issue the Committee
should discuss: Does it want high or low
levels of incentives to generate and conserve revenue? If it favors high levels, it should gravitate
toward the model suggested by the AHC Finance and Planning Committee, which
essentially calls for almost full attribution of costs and revenues to the
colleges. Does the Committee favor
strong central management of resources or strong central oversight of resource
management? There is controversy on this
Committee about that issue, and some on the Committee may have seen a draft
summary of a meeting that took place on a Saturday morning earlier this month
that raised a number of questions about a model based on the AHC Finance and
Planning suggestion. There is also some
confusion about the budget model and the strategic planning process; the point
is, he said, that the budget model should support the strategic plan.
Professor
Speaks responded that he favored incentives to conserve and spend money wisely
and to generate the maximum amount of revenue.
His question is whether those incentives derive from a budget model or
from leadership by the President, Provost, and Senior Vice President for the
Health Sciences. The Saturday meeting,
he said, consisted of faculty governance leaders from three colleges and there
has NOT been any official summary prepared.
The meeting was a conversation to obtain views and reactions to the
budget model discussions and to help formulate and crystallize questions that
need to be addressed. He and Professor
Campbell prepared a set of questions from the meeting that they believe the
budget model working group should address.
Professor
Erdman said that with respect to the two options Professor Feeney outlined, at
one level it does not matter which budget model is selected; what is important
is that administrators can plan, will know the rules, and that the system is
transparent.
Ms.
Blixt recalled that Vice President Pfutzenreuter had said the working group
would present information to President Bruininks in mid-November. Was that done? What was the result? Ms. Tonneson said the President did review
the four options the working group was considering; the President concluded the
working group should not spend time pursing the model that would have
centralized revenues or develop a plan based on mission activities, but that it
should continue to examine the other two (make a few changes in the current
model or move toward a fully-allocated cost and revenue model).
Professor
Konstan said he had two reactions. At
this point, the position he advocates is probably lost, but he wished to be on
the record about his views. He said
Professor Feeney has one point correct:
Does the Committee favor a model that has more or less entrepreneurship. What has not been asked is whether the
incentives in the model will be for quality rather than money. There is a tradeoff, he said: Does an excellent college that has few
options for generating revenues have the opportunity to spend money?
Professor Konstan said he had prepared an amateur
Talmudic analysis of the budget proposals about why he does not believe in full
attribution of revenues and costs.
1. Either
the base conditions will be the same under the new model, or they will change (a
tautology)—either units will be made whole under a new model or they will not.
There will be a transition from the current model to the new
model. That transition will presumably involve assessing the costs and
allocating the revenues for each unit. In the unlikely situation that the
unit's newly allocated revenue matches its newly attributed costs, everything
is stable. In the more likely situation where these are in imbalance, the
base conditions could conceivably be equalized by increasing or reducing central
allocations of state funds. If a unit's net increase in "internal
revenue" (revenue minus attributes costs) is greater than its current
subsidy, then presumably that unit will see a net budget increase at the start
(it is unknown whether such cases exist, and cannot be known until the formulas
are established).
1a. If
the base conditions are the same, then by definition no unit is harmed or
helped. Central will retain the same amount of funding (including
recaptured cost money).
1b. If the
base conditions are not the same, there are by definition losers. Any
unit that wins (i.e., has a total increase in budget due to the model
change) means an offsetting loser either in another unit or in central
discretionary funds.
In either case, the amount of discretionary funding available to
the President to "steer the ship" is the same or less. The only
way to generate new funds for this person, apart from state allocations, would
be to increase the "compact tax" or some other fee that would hurt
each unit (perhaps unequally, perhaps equally).
2. Either legislative
allocations will keep up with cost increases, or they won't.
Increased costs are real--both within the units and in
central administration. The only question is where the funds come
from to cover those increases. They may come equally (or
disproportionately) from the state allocation, or they may come from
tuition and unit-generated revenue.
2a. If
the state keeps up with its fair share of the cost of running the University,
then the President will retain an ability to steer the ship through allocation
of state funds to different units. In this case, one could
argue that this budget model (full attribution of revenues and costs) will work
well. Indeed, though, we believe that any model proposed works well
when state allocations increase.
2b. If
the state continues to under-contribute, either through small increases or cuts
to the budget, then the President's ability to steer the ship is severely
compromised. Either the President must reallocate central funds to cover
central administrative costs (decreasing the amount available for investing and
steering), or the President must increase the proportional taxes on units (for
central administration/services) and will be under significant pressure to use
what remains of state funds to help units coping with increased taxes remain
whole. Note that the President has certain easy choices here—it is easy
to keep units that are financially "self-sufficient" afloat, since
they can increase revenue through tuition or ICR recovery. What is hard
is deciding to subsidize important units that are not financially
self-sufficient. As the state allocation shrinks, such units are harder
and harder to support under this model. This problem does not exist under
a shared-revenue model, since revenue is partially designated for
University-wide use.
3. Either
central service units are aware of unit contributions and can tailor their
services to match, or not.
A key question is whether the libraries, facilities management,
registrar, and other service units will be aware of how the cost allocation
formula leads to differential revenue from each unit.
3a. If
they are unaware or not allowed to tailor their services, then there will be a
continued model of encouraging units to consume as much as they can for the
measured cost. Units will continue to pressure the library for greater
acquisitions, subscriptions, and services. Research support services will
be used with abandon. No real conservation will occur.
3b. If
they are aware and can tailor their services, we will have a very different
model of operation. Units will be encouraged to conserve, or to increase
their allocation to the level of desired services. Questions such as
"should there be a special library for xxx" will be based on the
metric and cost allocation. Before long, I believe that units will be
smart enough to start negotiating at a more detailed level to match support
with cost allocation, and we will have to abandon the simple accounting
(headcount, square footage) and revert to either custom negotiations or more
complex measurements. Either one undermines the simplicity of the system.
Fundamentally, I still believe that moving to a "near-ETOB
(every tub on its own bottom)" model of running a university makes little
sense. I understand that certain units have been disadvantaged by the
current budget model, and indeed by current budgetary decisions. Others
would be disadvantaged by any new model. Except for providing
incentives for conservation, this is a zero-sum game. And thus, the
question is who is "steering" this game. If we believe that the
University's academic leadership—specifically the President—should steer, then
we need to give him the resources to do so, and acknowledge that a consequence
may be that certain units legitimately feel that they are being used to
subsidize others.
Mr. Klein said this is complex and
that whatever the institution adopts must be something that can be understood
throughout the organization and be communicated across the layers and
levels. The key question on what the
Committee believes should be adopted depends on what the most important one or
two behaviors one wants to provide incentives for. If the Committee believes it is a top
priority to allow the President to steer the ship, the model should provide
that incentive. If the top priority is
to distribute overhead expenses and identify who buys what from whom in order
to improve cost control, then cost control should be rewarded. When the incentives don’t match the goals one
gets into gaming the system. A question
that should be asked about the budget model is "what actions will lead to
the University being one of the top three public research universities in the
country?" If it is quality of the
faculty, then the model should provide incentives for the recruitment and
support of the best faculty.
Professor Konstan said he agreed
with the point about the faculty. He
said he worries that full attribution of costs and revenues would not allow the
President to steer the ship. Any unit
can figure out what model will be best for it and then argue that the model it
likes does what is best for the University.
He said he was not sure there was any objective way to measure what
budget model is best for the University.
Mr. Klein responded that there need not be agreement on that point and
this Committee will not reach agreement.
But those who must make the decision can gain insights from this
discussion; this is the consultative process and the administration can hear
views about what are the 1-2-3 most important factors to the success of the
University in carrying out its mission.
Professor Speaks said that Vice
President Pfutzenreuter had described models in place at the
Professor Roe said one needs to divide up the problem. Tenured and tenure-track faculty are a high
proportion of costs and are stable over the short run (2-3 years). So one must allocate money to cover operating
costs. Over the longer run, the question
is about the number of faculty who should be in this or that college. Then, the question of resource allocation can
be focused more easily into two parts, one is the short run (2-3 yrs) where
emphasis can be placed on the efficient use of resources not related to faculty
salary and mostly within colleges, the other in the longer run which relates to
the efficient allocation of faculty resources among colleges. That in turn raises the question of whether
funds for tenured and tenure-track faculty should be held centrally.
Professor Feeney said that from the
AHC Finance and Planning Committee perspective, the proposal is to allocate
costs and revenues and create incentives; state funds are to be used to
subsidize colleges that cannot make it on their own—but that is then a
conscious decision to subsidize. The AHC
is often the canary in the mine, he said; CLA's revenues are now more than 70%
from tuition and the old paradigm that state funds pay for the bulk of the
college is not valid any more. So it is
necessary to identify a way to allocate state funds. Many find the taxes objectionable because the
money goes to things they do not understand and they are not sure the money is
used for common goods. And to the
comment that there will be gaming of the system, everyone does that now.
One model that provides the
President discretion over the allocation of state funds, Professor Speaks
observed. Of the $540 million per year
in state funds, how much of that is really discretionary, if a large percentage
is committed to faculty and staff salaries?
Vice President Pfutzenreuter said he did not know, but the President
would have discretionary authority over the entire amount. If 85% of the non-sponsored CLA budget is for
salaries and fringe benefits, Professor Speaks inquired, in what sense does the
President have authority? He could
direct a college to raise tuition or cut expenses, for example, and not give
the state funds to the college, Mr. Pfutzenreuter explained. Professor Speaks said he found this point
very interesting. He said he has been
concerned that if all is going well and there are discretionary funds, the
President allocates them as he thinks best; if a unit loses some portion of its
funding, the President can direct it to make up the money from other
sources. The President can do that now,
Mr. Pfutzenreuter pointed out. So the
President can use his discretion to help a unit that lost revenue or take
revenues from a unit and tell it to raise tuition, Professor Speaks asked? He can indeed do that now, Mr. Pfutzenreuter
affirmed.
Does this mean each college could
charge different rates of tuition,
Professor Speaks said that
philosophically he could accept the proposition that the budget model should
lead to increased efficiencies in the Office of Budget and Finance and in the
colleges. He recalled that Provost
Sullivan has said the budget model must support the strategic positioning
process. He said he needed help
understanding which of the budget model options would do what Provost Sullivan
said is required. If Provost Sullivan is
correct, that the budget model must support and enhance the strategic plan and
academic mission, but if the University has not reached any conclusions about
the strategic plan, how can anyone select a budget model? Ms. Tonneson said that Senior Vice President
Cerra is working on a description of the link between the budget model and
strategic planning; he is trying to answer that question. Professor Speaks said he did not know what
the final strategic planning or positioning would be but he wagered that it
will look very much like what was presented to the Board of Regents in November
in terms of framing concepts and vision.
It would be reasonable to ask what effect the various budget models
would have on what was presented to the Board.
Vice President O'Brien commented that
strategic planning documents are out for comment and that a lot of people have
sent comments. The Regents will adopt a
resolution in March and the statement will likely have some reference to being
among the top three public research universities in the country. She said she believes the internal budget
model is a process that is informed by the institution's goals and objectives;
the model does not make decisions but it does structure the decisions. The internal budget model is only one
component of decision-making.
Mr. Pfutzenreuter said that if the
visions and aspirations are specific enough that they call for emphasizing
undergraduate education in CLA, for example, one can ask if the budget model
hurts or supports that vision. But if
the vision is to be among the top three publics in the country, it is difficult
to say if the budget model supports the vision.
One must look at the budget model as a process. Vice President O'Brien agreed. She said the budget model can provide
incentives for or discourage change. She
has been charged by the President and Board of Regents to instill a culture of
productivity, compliance, and quality improvement on the service side of the
institution. This includes transparency
in service and support unit activities as they address faculty, staff, and
student needs; the budget model can provide incentives for change.
This suggests the budget model
should have transparency as an attribute, Mr. Klein said, especially where the
costs of services are attributed to units. That will make the process easy to
understand and promote change.
If the University says it will be
one of the top three public institutions, Mr. Klein said he would argue that
there are two actions for which incentives must be provided. For support units, efficiency of services may
be the most important incentive. For academic
units, the caliber of the faculty is among the most important considerations,
so he would ask what model would provide the stronger opportunity to pay
faculty salaries and invest in things important for faculty success. Again, he said, there must be transparency in
the information so people can understand what is expected.
Professor Speaks said he was not
speaking in favor of or against any particular budget model, but is trying to
identify questions that need to be answered in order to further the academic
mission of the University. Those
questions must be answered before one can make a rational evaluation of budget
models, including questions about attributed costs and the bases of
evaluation. Until those issues are
wrestled with, and algorithms set, he said he could not figure out what the
best budget model would be. If there is
a way to evaluate the goals relative to the model, he would like to know about
it. Vice President Pfutzenreuter agreed
with Professor Speaks's point.
As soon as one knows the algorithms
the gaming will start, Professor Konstan predicted. With respect to accountability, that will
come down to who sets the budget. If a
unit head receives a bill for a service, that does not create any
accountability; if the unit head objects, he or she can complain to the
President or Vice President O'Brien—which he or she can do today as well. Or the system could be set up so the unit can
go to the market: If they are paying
Facilities Management or some other unit too much, they can contract with an
outside vendor. If they believe the
librarian is being paid too much, they would hire their own librarian. The dean, however, does not have the option
to withhold funds for services.
Professor Konstan also said that
time is an important issue. Whatever
reduces the amount of time that faculty, department heads, and deans must spend
thinking about the budget model will make this a better university. The central administration (and the
One factor that should be considered
in all the models, Professor Erdman said, is the burden on departments and
overhead. Some could create more, some
less, and this burden must be kept in mind.
His department, he said, has fewer people who are doing more work. If the system adds burdens to departments, it
could collapse from its own weight.
Ms. Van Voorhis asked when the
attributed costs would be developed. Mr.
Pfutzenreuter said that no one has decided on a budget model, and both of the
ones the President is interested in have attributed costs. They need to do more work on attributed costs
but have not yet started. Ms. Van
Voorhis commented that if she had to provide an accounting for all her costs,
she would need to hire an accountant; Mr. Pfutzenreuter said that she would not
bill colleges but that she would, as is true today, need to provide her budget
to the administration. How would a unit
like hers obtain money for something like new carpeting, she asked? By coming to the administration, just as they
do at present, Mr. Pfutzenreuter told her.
If salaries are about 80% of total
costs in academic units, Professor Roe said, they are fixed in the short
term. So one is talking about 20% of the
budget. One must think about management
and transaction costs as they pertain to this 20%. He said he liked the report from the AHC
Finance and Planning Committee and its thinking about incentives, but one must
bear in mind this only affects about 20% of the budget. Across the entire University, compensation
accounts for 60-65% of costs, Ms. Tonneson said, but agreed that it is about
80% in academic units. That is because
academic unit budgets do not include attributed costs, Vice President O'Brien
observed.
What incentives are there for the
other 80% in the long term to encourage behavior with respect to quality, Mr.
Klein asked? What provides incentives to
make the right decisions? That is the
point that Professor Speaks is pounding on, Professor Roe pointed out. The question has not been answered. The point is that there is not enough
information at the level of the dean to make decisions; this is about
allocating dollars across colleges. The
budget model does not do that, people do, Professor Korth maintained; the
budget model provides a structure. They
have already seen that central administration can manipulate the model to do
what it needs and wants to do—and it has made good and necessary decisions—but
the current model has flexibility in it.
What is missing is understanding, and it is not clear why that could not
be provided.
The budget model does not solve
everything, Professor Konstan said, but there are radical things one could
shift on the cost or revenue side. One
could decide that the budget follows faculty lines so units are not starved for
support if they do well. One could
decide the same thing about students.
Deans and the administration could decide on enrollment; given X number
of slots, "stuff" would follow to the college. Attribution does this to some extent, but not
on the faculty side; one could create a link there as well. To focus on quality rather than quantity
means there must be sufficient support assured.
Professor Erdman said he agreed with
Professor Korth that the system depends on the wisdom of the person making the
decision. Perhaps the model does not
matter as long as it is transparent—"and we can all go home." Professor Speaks said he understood that
Senior Vice President Cerra has said the current model must go. What led to that conclusion? Frustration with the increasing taxes and
fees, Mr. Pfutzenreuter said, and frustration about what happened to the money
and where it can go.
Mr. Pfutzenreuter agreed that the
timing of the strategic positioning discussions is frustrating. If they knew that the goal is to be one of
the top three public universities, they could then look to see, for example, if
the full attribution of debt and operating costs to research facilities would
have an impact on achieving that goal.
Has the University turned its back on the
Professor Campbell said he was
concerned about how the "test drive" will be conducted. Ms. Tonneson said the President has asked for
a timeline and consultation, which are in process. They are trying to determine, between the two
models still on the table, what items of analysis must be completed. The discussions have been kept at a high
level thus far so that units are not all asking "what happens to me?" But there will be a test drive, Professor
Campbell asked? It is their
responsibility to do so, Ms. Tonneson said.
This suggests the potential for a
tug of war between units so they can see if they will come out ahead or behind
financially in a new model, Mr. Klein said.
If all the units advocate their own positions, one can predict that
there will be a draw in the tug of war with no clear cut solution that all
units agree on. That means the decision should be at a level that is not tied
to any of the units—the decision on the budget model should be pushed up to a
small group of financial experts and senior University leaders who can select
the model they think is best for the overall University. He agreed with Professor Erdman: put your money on the one model we think is
best, live with it, and go home. People
can then grade the system chosen on its transparency; if it is transparent,
that will get at efficiencies and not take a lot of time to figure out.
Professor Speaks said that if, under
a new system, there are no winners and losers, and nothing is changed, this is
not worth the effort. There must be a
desirable outcome. Professor VandenBosch
agreed and said that is why strategic planning must come first so that people
know what will not be supported in the future.
The effect may not be immediate, but there will be an effect on
decisions about units. If the budget
model decision can be delayed, Professor Speaks commented, there is a chance
that it and the strategic planning process could be made to work together. Professor Korth, however, said that strategic
planning could take place separately from the budget model discussion; he said
he did not see the link between the two.
The budget model is flexible and will allow different strategic
positions, Ms. Tonneson said. The
communication from Senior Vice President Cerra will try to show what will
change year to year and what will not.
What if Professor Korth is right and
the problem is not the budget model but the difficulties of decision-making,
Mr. Klein inquired? The budget model can
perhaps enhance decision-making, but what if the problem is that the University
is not good at making decisions? Maybe
when it is are "stuck between two crummy financial choices," it finds
it hard to make a decision and take action.
Maybe that's the real issue? To
make a decision means there are instruments to manipulate, Professor Roe said,
and the instrument is typically money.
To make a decision means there must be a lever, and that is the link
between the budget model and strategic planning. What levers does central administration have,
Professor Campbell asked? State funds,
the compact tax (which can be adjusted year to year), and setting the central
budgets, Ms. Tonneson said.
What has not been brought up,
Professor Feeney said, is that while transparency is seen as desirable,
transparency needs to be defined. The
University must also address what the legislature thinks transparency is. He recalled a conversation between the
Faculty Consultative Committee and Charlie Weaver, Executive Director of the Minnesota
Business Partnership; they asked if there was a way to depoliticize the higher
education budget, and while there was no answer to the question, it was clear
that as funds decline, demands for accountability increase. There is thus an incentive for the University
to be transparent. The University must
get rid of across-the-board taxes, which are not doing the units any good. One can look at square footage, classroom
use, etc., but he said he did not believe that the information needed to make
decisions is available. The central
administration needs levers but so do the deans; they need to be able to make
decisions about the level of services, for example.
Professor Speaks said again he
doubted the final strategic planning documents would differ much from what the
Board of Regents saw in November.
Professor VandenBosch pointed out that the college documents are due on
December 1, which will provide an insight into what the colleges say about
mission and applying the criteria. There
is much that is simultaneous that should be serial, Professor Speaks concluded,
and the deans could do a better job if they knew the overall strategic
plan.
There has been talk about
predictability, Professor Campbell commented, a system that would allow
colleges to see three years out. The
hope is to make the RULES stable, Ms. Tonneson said, but they cannot provide
stability on revenues. But if there are
internal taxes, they could be lagged so that colleges could calculate
them. But the University will never know
about state funding, so it will not be able to make decisions about tuition, so
units can't know numbers into the future.
Professor Speaks said he agreed entirely, but he was also not so
confident that the rules could be fixed, either. If a unit looks out three years and makes
assumptions about tuition, enrollment, state funds, etc., and then there is a
catastrophe—some major revenue source declines significantly, for example—will
the rules then have to be modified? Ms.
Tonneson said she did not believe so; the discussion is about process rules
(allocation of tuition and bases of attribution of costs).
Professor VandenBosch pointed out
that with the budget models, one is talking about attributing costs and
revenues to colleges. When one talks
about incentives, one is talking about departments. The way colleges deliver money to departments
varies with the college, so incentives may have different effects in different
colleges. She said she had mixed
feelings: She did not want to see
micromanagement of the colleges, but the structure has to pass incentives
through the colleges to departments.
That comes back to decisions made by people, Professor Korth
maintained. A lot of what the Committee
has heard over the last several years about the desire for transparency is at
the college level—people do not know how deans distribute funds. The budget model will not change that. Ms. Tonneson said the working group has
talked about this issue but does not want the budget model to apply at the
department level. To make it do so would
add a lot of complexity. Colleges are
accountable for their behavior. If one
added comprehensibility and transparency as part of the rules, could the
administration intervene if it believed a college was not providing enough information
to departments, Mr. Klein asked? One
central function could be to play a role in making sure departments have the
information they need. Ms. Tonneson said
that it could intervene if it had to.
This is the nub of something
important, Professor Chapman suggested.
Central administration does not want to reach to departments, but the
success of the budget process depends on something not under central
control. If the deans and department
heads do not agree on the budget model, it will not matter what the
administration does. And it is in
departments that the money for teaching and research resides, Professor Roe
added.
Professor Feeney said this was also related to risk management
(e.g., not all faculty salaries are on O&M funds). Do they view the levers to manage risk as at
the presidential/budget and finance level or in the colleges? Ms. Tonneson said she assumed it would be in
the colleges, which is where the funds are.
But faculty are tenured at the University level, Professor Roe said, so
there is an imbalance. This will tend to induce deans to hold larger savings
accounts to cover risk, which are funds withdrawn from productivity. Ms. Tonneson said there has been discussion
on this point because it would be possible to have a smaller savings account if
the risk were managed centrally.
Professor Campbell said that
transparency requires knowledge, such as how much one is charged and how much
something costs. But that element of
transparency is not in the model because it would complicate it too much. Ms. Tonneson said she did not believe it
possible or workable to identify the cost for services for each unit, so they
use proxies. In some cases, they can get
close to actual costs; in other cases, not.
Mr. Klein said it is possible to set up an elaborate costing system down
to the lowest level, and that system becomes one of the core competencies of
the institution, but it is very burdensome.
No universities do anything like that, Ms. Tonneson commented.
Do they see attributed items as
possibly having an effect on behavior, Professor Feeney asked? Or will the charges be so amalgamated that
units cannot see individual costs (e.g., square footage for classrooms)? If the charges are amalgamated, there will be
no change and things will be just as they are now. Ms. Tonneson said they did not expect there
would be a lot of incentives to change
behavior except in big ticket areas. How
the system is set up will make a difference—and in some cases, the University
does not WANT units to use less of a service (for example, financial aid and
the libraries).
Professor Campbell said that the
discussion had run out, thanked everyone for their contributions, and adjourned
the meeting at 4:00.
--
Gary Engstrand