James A. Moffett Lecture

April 29, 2010 - James A. Moffett Lecture

Center for Human Values, Princeton University

"Leading Universities in the 21st Century: Chances and Challenges"

At an international meeting earlier this year, thirty presidents of major universities from Africa, Asia, Europe, Latin America and the United States gathered to discuss (among other issues) how universities in the developed world could help those in the developing world build their teaching and research capacity. (Many African universities, for example, have such a small pool of PhDs to draw upon that most of their faculty cannot teach advanced courses or engage in original or innovative research.) After we debated a list of capacity-building means, one American university president suggested that we preface any list with a statement that the goal of building the world’s capacity for higher education is integral to our institutional mission.

After the meeting adjourned, another American university president quietly commented to me: Are we not already sufficiently inundated with so many demands on our universities’ time and resources without having to build university capacity all over the world as yet another part of our mission, as worthy as this goal might be? This essay tries to come to defensible, action-guiding terms with the real tension reflected in this exchange.

As a political philosopher and university president, I am intrigued by how many hard choices among worthy ends universities face and how no common theories or philosophies can adequately deal with these hard choices. Hard choices confronting individuals and institutions are commonly characterized as follows: So many worthy goals, so little time, and too few resources to pursue all ends to their maximum. For universities, these important considerations overlook one other critical factor: so many powerful constituencies, including faculty, students and parents, alumni, public officials and the general public.

How should university leaders think and act on this problem of hard choices? I offer only my own perspective, one that I suspect is neither uniquely right nor wildly idiosyncratic.

First I suggest why it is hard for university leaders to set priorities that entail change.

Second, I consider ways in which we can and cannot effectively defend such university priorities.

Third, I offer a paradigmatic example of defending a high priority goal, namely the direction of more financial aid toward low-income and middle-income students.

I conclude by applying my approach to university priority setting to my colleague’s concern about expanding the American university’s 21st century mission to include yet another goal.

I focus on hard choices that require change, since change comes hard to universities. I focus on the most selective private American research and teaching universities, since I know them the best, but much that I say applies to other selective private and public American colleges and universities as well.

Universities must change in order to thrive, and our universities have changed dramatically over time, for example, in the students we admit, the faculty we hire and the subjects we teach. Yet we are structured in a way that makes us slow to change relative to many other institutions. A trustee is said to have told Duke’s former president, Nannerl Keohane, that the moment he hears the world is coming to an end, he will take up residence at Duke, because everything takes a year longer to happen there.

Some comparisons bring even our glacial-speed change to a stop. You may have heard (as I did from a past president of the University of Chicago) that leading a university is like being a cemetery groundskeeper: you preside over a lot of people but you can’t get them to do very much.

When the President of the University of California system, Mark Yudof, recently invoked the cadaver comparison in a New York Times Magazine article, some UC faculty members took offense. The times are so tough and tense in the UC system that even humor may unfortunately (but understandably) backfire.

Clark Kerr, the first Chancellor of Berkeley who later became President of the California system, set a standard for frank and often-humorous expression, famously noting that the three major administrative problems a president faces on campus are sex for the students, athletics for the alumni and parking for the faculty. In his Godkind lectures at Harvard, published in 1964 as The Uses of the University, Kerr confronted the hard choices among worthy university ends with this memorable summation that caused quite a stir:

A university anywhere can aim no higher than to be as British as possible for the sake of the undergraduates, as German as possible for the sake of the graduates and research personnel, as American as possible for the sake of the public at large – and as confused as possible for the sake of the preservation of the whole uneasy balance [p. 14].

Less famous than the idea of the American multiversity being a confusing muddle of un-prioritized ends was the frank advice that Kerr offered his fellow presidents after delivering the Godkind lectures. Reflecting on the firestorm of controversy that ensued, Kerr asked himself: “Would I have chosen to give these lectures when I did knowing what I know now?” Kerr’s reply: “The answer is absolutely ‘no.’ [S]hould [I] have given these lectures as an active president regardless of the date, and the answer is ‘almost certainly not.’” He concluded “that discretion is the better part of valor—a rule that most presidents do not have to be told to obey” [p. 160].

A history professor at UC Berkeley called Kerr’s lectures the “least discreet” ever delivered by an American university president. Kerr was not trying to earn this distinction; neither am I. My goal this afternoon is edification, not indiscretion.

Universities are less like organic communities, as Kerr observed, and far more like small cities [p. 31]. As important, and something Kerr neglected to emphasize: to earn our title (since we don’t have a right to exist), universities must be dedicated above all to high quality, high level teaching and research. Even if we are like small cities, we cannot serve every purpose of a city for our many constituents. Yet to say that we must above all undertake high quality, high level research and teaching is sorely inadequate to guide university leadership in identifying, articulating, and pursuing specific goals.

Packed into the most common characterization of our mission (teaching, research and for some of us clinical practices) are a seemingly infinite number of specific goals. Considering only the pursuit of research, we can arrive at an incalculable variety of fascinating intellectual questions and goals to pursue. Then add the goal of developing the intellectual and leadership potential of our students. Then add the goal of engaging with select local, national, and global communities. Then add the goal of publicly demonstrating the practical value of our research, teaching, and clinical practices.

In principle, admitting the best students, hiring the best faculty, delivering the best teaching, research, and clinical practices, and educationally engaging with local and global communities are all compatible goals that enhance a university’s basic mission over time. The fact that these multiple goals do not conflict in theory (save for our scarce resources) renders our need to make hard choices less obvious.

No general theory exists to choose among these desirable goals that are compatible in principle but conflicting in practice due to limited time and resources. I first began to fully appreciate the gravity of this problem when I became Dean of the Faculty at Princeton in 1995. The most frequent demand of a Dean of the Faculty is to give departments more faculty lines than her budget permits. The answer I found to be most effective after I had tried all the others: Okay, you want me to authorize you to hire more faculty members than I have the budget to provide? Who do you want me to fire?

It should be obvious that no university can afford to hire every star faculty member in every interesting and important field. But it often isn’t. While our universities all teach advanced mathematics, we are far more adept at adding than subtracting.

In the wake of the Great Recession, scarcity of resources has become more obvious than it was in the 1990s, but -- now and then -- scarce resources are actually not a sufficient answer to a request for more; my answer in the form of a rhetorical question should not have been persuasive in itself. Yes, resources for hiring are limited. But this response rarely, if ever, provides a fully adequate answer to the question of why any particular department-- Chemistry or Comparative Literature, Molecular Biology or Middle Eastern Studies, Physics or Politics--should be prevented from adding great faculty, and no other department asked to shrink.

It is the responsibility of university leadership substantively to defend a set of spending and saving priorities. Across-the-board freezes (or cuts) to all departments make both the weak and the strong weaker. This is a bad recipe for maximizing a university’s strength in hard times; just as giving across the board increases is a bad recipe in good times.

For every complex problem, H. L. Mencken said, there is an answer that is clear, simple and wrong. That’s what I think of simple recipes for setting university priorities: across the board increases and decreases, growth by grass-roots evolution or factional mediation, and top-down planning without buy-in.

So, is there a good recipe? Just as universities have a plurality of compatible purposes, I, too, draw upon a plurality of compatible principles to help decide how to distribute resources, without either applying a comprehensive theory to guide my decisions or adopting the simple approach of “muddling through.”

The best recipes in my repertoire consist of multiple ingredients that are affordable and combinable in ways that can satisfy diverse and discerning palettes. My recipe for allocating resources among academic divisions and departments includes the following ingredients:

  1. create comparable strength among the divisions that are essential to the university’s teaching and research capacity, namely humanities, social sciences, science, engineering
  2. enable the strongest departments within each division to get stronger
  3. ensure that weak departments essential to the University’s educational mission do not fall below a respectable level
  4. shrink weak departments that are not educationally essential
  5. get buy-in for your approach or alternatives that you are persuaded to defend by deliberating with a wide range of faculty, staff, fellow deans, and wise friends

This complex recipe presupposes a set of higher priorities that has been articulated to and embraced by the university community. These priorities ideally will be strategic to the university’s sense of mission and purpose, and will play to a university’s strengths and its shared sense of institutional self. At the same time they must resonate beyond the university with more widely shared and publicly defensible purposes of higher education.

Taking Penn’s set of priorities as a case in point: increasing access, integrating knowledge, and engaging locally and globally aim to satisfy these standards. Increasing access is essential to our educational mission. Integrating knowledge resonates with our sense of self as a university that applies interdisciplinary knowledge to address complex practical problems. Engaging locally and globally plays to Penn’s particular strengths. All also resonate with publicly defensible purposes of higher education.

Since no objective theory requires these or any other comprehensive set of university priorities, it falls to university leadership to articulate and defend the university’s priorities in a way that enhances our quality, reputation, and sense of institutional self over time.

Articulating and defending a set of (ideally inspiring) university priorities may seem easy until it comes to supporting them with scare resources. The single greatest challenge, and chance, created by achieving a broad consensus on university priorities is the ability of university leadership to raise and spend obviously limited resources in pursuit of the university’s previously acknowledged highest priorities. Let me offer an example of how this works in a way that neither resigns one to muddling through nor ties one to a comprehensive theory.

Increasing access is one of Penn’s highest priorities. And like building the best faculty in high priority fields, financial aid based on need is extremely expensive. A high degree of ongoing mindfulness is needed to pursue each wisely.

Since there are more highly qualified students with impressive non-academic qualifications from high-income American families than the most selective universities can collectively admit, why should socioeconomic diversity -- added to academic qualifications -- be seen as important, especially when resources are scarce?

Diversity is not an educational end-in-itself. But when accompanied by convincing academic qualifications, socioeconomic diversity (like racial, religious, and ethnic diversity) serves as a means to three extremely important ends of higher education: first, equalizing opportunity; second, educating leaders for all sectors of society; and third, enriching the educational experience of all students on campus.

To the extent that we are truly committed to the three important ends of higher education that can be served by socioeconomic diversity, we need to push for more progress in recruiting highly qualified students not only from low-income but also from middle-income families, who are the most underrepresented on our campuses when academic qualifications are taken into account.

My concern for increasing access began with a focus on increasing the representation of low-income students on our campus, and this remains an important priority. To understand why I think increasing access for middle-income students should be an equally high priority, some simple data and analysis are essential. I began by asking what percent of students from the top income quintile--the richest 20 percent--of American families with college bound children are on our campuses?

A staggering 57 percent of our enrollments come from the richest 20 percent of American families with college-bound children. This comparison overstates the over-representation in one important way. Since we rightly admit only students who are academically highly qualified, we need to ask: what percent of students who are highly qualified (with combined SATs over 1200 and high grades) come from the richest quintile? The answer is that 35.8 percent of all highly qualified seniors come from the top quintile.

Nearly 36 percent of the highly qualified seniors come from the top income group while 57 percent of our students come from this group. The richest 20 percent of American families are overrepresented on our campuses by a margin of 21.2 percent [i].

James A. Moffett Lecture image

This overrepresentation of the top quintile on our campuses squeezes out students from all of other income groups. That’s headline number one: The Big Squeeze. I had expected as much, but I did not know that the margin of over-representation was so high as to leave every other income quintile -- 80 percent of the American families with college bound children--underrepresented.

As much as I expected headline number one, I did not anticipate headline number two: The greatest underrepresentation by far on our campuses is not in either of the two lowest-income groups but in the middle-income group. Since these data have been analyzed, this headline--the even bigger Middle Income Squeeze--has not gained as much attention as it deserves.

Let me share these data with you. The lowest income quintile among American college-bound families accounts for 7.5 percent of all highly qualified seniors, and these students are underrepresented on selective private university campuses by .6 percent. The second lowest income quintile accounts for 10.9 percent of highly qualified seniors; and these students are underrepresented on our campuses by 3.7 percent. Together, these students represent American college-bound families earning under about $41,000, and they are underrepresented by 4.3 percent.

Moffett Lecture image

Now consider the middle 20 percent of graduating seniors by income, who come from families earning about $41,000 to $61,000. The middle-income quintile accounts for fully 20.3 percent of all highly qualified high school graduates, but they account for only 11.9 percent of our undergraduate enrollment. They are underrepresented by 8.4 percent, almost twice the under-representation of the two lowest quintiles combined.

Students in the second highest income group, from families with incomes between about $62,000 and $94,000, constitute 25.5 percent of highly qualified students, and our campuses enroll 17.1 percent of this group. They are squeezed by the same 8.4 percent margin as the middle-income group.

Moffett Lecture image

Here’s another view that makes the middle income squeeze even more striking.

 Moffett Lecture image

Numbers only tell half of the story. In making the case as clearly as I can for increasing access for middle- and low-income students, I enhance statistics with narratives that reveal the qualitative loss of diversity on our campuses. The middle-income students at Penn whose peers are underrepresented include: a first-generation college student who is a gifted writer and the daughter of a New Hampshire auto mechanic; a campus leader who is the son of a truck driver from Texas, and an ambitious young man who is pursuing both a doctorate in philosophy and a law degree and is the son of a grocery store clerk.

Excellence and equity tell us that we should not let these talented students fall on the losing side of an arbitrary line dividing those deemed deserving of avid recruiting attention from those who are not.

Yet recruiting more middle-income students is costly and takes a high degree of discretion to pursue wisely. [ii] Middle-income students tend to think of our universities as unaffordable. They do not realize that only families for whom it is affordable pay our sticker price.

Dramatic changes are needed to help shift these perceptions. In order to afford a financial aid policy that substitutes grants for loans for all undergraduates with demonstrated need, Penn made financial aid one of the three highest priorities in our $3.5 billion campaign. Our Trustees agreed to increase the payout on the financial aid portion of our endowment to help bridge the financial gap until the Campaign is complete. Now all Penn undergraduates -- not just our low-income financial aid students -- receive all grant, no loan packages. Since 2004, we increased our annual undergraduate financial aid budget by 78 percent, from $83.5 to $149 million dollars. [iii]

Increasing access for middle- and low-income students is one of the most daunting and worthy moral challenges facing the most selective American universities today. For all but the handful of most highly endowed universities to pursue this goal consistently with balancing our budgets, need-based financial aid must be one of our two or three highest priorities.

I said earlier that I know of no good general theory that would enable us to choose among the many desirable goals that a university can pursue. How, then can I make the case for need-based financial aid being one of the highest priorities of Penn and (by implication) similarly situated universities? The best that any leader can do is to make the case for prioritizing some ends over others by offering clear reasons and inspiring examples that illustrate those reasons, and -- as essentially -- by eagerly engaging with individuals who have a sincere interest in the institution and are willing to consider the best reasons for and against compelling institutional priorities.

Let’s now consider the earlier question: Is building higher education capacity in the developing world integral to our university’s essential mission of higher education?

Political theorists may view this tension -- between devoting the university’s resources to improving one’s own faculty and student body, on the one hand, and devoting resources to improving higher education in less developed countries, on the other -- as analogous to the widely debated dilemma of special obligations to the citizens of one’s own country versus general obligations to individuals wherever they happen to reside. Analogous, perhaps, but the debate in political philosophy goes on without reaching a set of compelling action-guiding conclusions.

Universities, however, must make (or else they must abdicate) action-guiding decisions as we speak about whether and where to help build higher educational capacity in developing countries.

My response again draws upon a plurality of principles that are compatible with a wide range of competing theories. The most relevant guiding principle here is: Ought implies can. In fact, American universities cannot do as much as many of us would willingly and affordably do, were it possible, to build the capacity of a sister university in the developing world while maintaining our own ability to support high level, high quality research and teaching for the students and faculty on our campuses.

Moreover, the leaders of many African universities tell us in no uncertain terms that they would rather build their capacity internally to their continent, by drawing upon the help of their African counterparts, than by enlisting the direct aid of American or European universities. They welcome true partnerships, like those supported by the Gates and Merck Foundations in Botswana, which have helped the Batswana both to build their first medical school and to combat the AIDS epidemic.

American universities can (and increasingly do) find mutually beneficial and affordable ways to help build capacity for higher education in developing countries. These ways include supporting faculty collaborations and student exchanges, research and teaching internships, and open curriculum courseware. We also admit more students than ever before from the developing world, and we equip them as best we can to give back to their country of origin. Selectively undertaken, these programs enhance our teaching and research quality, while also broadening our missions.

In these and other ways, we can accept the goal of building the world’s capacity for higher education as part of our institutional mission. We can satisfy this goal, without mission overload, not by one-way aid to universities in the developing world, but rather by carefully chosen collaborations and partnerships that mutually advance our teaching and research capacities.

So many worthy goals, so little time, too few resources to pursue all ends to the maximum, and so many powerful constituencies. I offer no single or simple recipe for how every university leader should think and act on the problem of hard choices. Instead, I draw upon a plurality of principles, which will yield different conclusions in differing university contexts. I distinguish between those strategic goals (such as increasing access) that the core educational mission of our universities make morally salient if not obligatory, and those strategic goals (such as integrating knowledge and engaging locally and globally) that play to the particular strengths of a particular university.

American teaching and research universities have long been viewed as engines of individual empowerment, economic development, creativity, and innovation. The demand for what we provide has never been higher.

At the same time, trust in how we are governed has never been lower. Some of the most searing critiques of the modern American university take off from Clark Kerr’s image of an institution whose basic instinct is preservation rather than adaptation to purposive change, and whose governance is mediating among factions rather than articulating institutional priorities.

I do not know whether it is imprudent, as Kerr warned, to be frank about how we set our priorities, but I think it morally right that we do so. I admit to having no surefire theory to guide my decisions. What I bring to the table is a repertoire of recipes and an eagerness to deliberate, frankly and open-mindedly as often as possible with a dash of humor. My most fervent hope is to inspire the broadest range of constituencies to join together in improving the capacity of American universities to contribute as much as we can to our communities, our society, and the world.

Amy Gutmann, Ph.D.


[i] The most complete recent data we have come from a 2003 survey of highly selective colleges and universities. More recent data have yet to be fully compiled and analyzed, but what is available reinforces the conclusions highlighted here from the 2003 data.

[ii] For example, family income is at best a rough surrogate for the kind of diversity that furthers the meaningful goals of diversifying our campuses. Being first in one’s family to attend college and from a blue-collar family may be more relevant than low- or middle-income to our goals of increasing equality of opportunity, educating leaders for all sectors of society, and enriching the educational atmosphere on campus.

[iii] At the same time, we have raised $407 million dollars in faculty support, another of our highest priorities, with $117.5 million committed to 21 new Penn Integrates Knowledge university professorships, who are distinguished by holding joint appointments in two schools.