by Charles Schwartz, Professor Emeritus, University of California, Berkeley                                    September 30, 2007

>>This series is available on the internet at

WHERE THE MONEY GOES – Lesson #2 (continued)

Here we present and analyze more detailed data about the excessive administrative bureaucracy throughout the University. We are now able to better locate and assess the areas of employment which have grown enormously without justification; and we estimate this bureaucratic  wastage amounts to $600 million per year.  Some theoretical discussion is presented; and we also relate this study to an external review of UC administration that The Regents have recently initiated.

The Old Data on UC Employment

     I will start by repeating the data presented in the previous paper of this series (Part 12, issued in May).  Table 1, below, shows comparative numbers of UC employees in various job classifications over the ten-year interval 1996-2006. This data shows that Management grew at the highest pace (+118%), far exceeding the overall employment growth (+31%); we also noticed another large category of non-academic staff (Fiscal, Management and Staff Services – coded F) which also grew alarmingly fast (+98%). It was noted that this data had been submitted to top UC officials more than once, with a request that they offer some justifications for these apparent cases of extreme bureaucratic bloat; but no answers have been given.

     Table 2, below, shows those most salient growth statistics detailed campus-by-campus. We see that there is considerable variation; but the overall pattern is clearly that in every location the ten-year growth in those two categories of administration has far outstripped the overall growth in employment. The most extreme campus is my own Berkeley – with the smallest growth in overall employment but the largest growth in management. Also, perhaps surprisingly, the excess of management growth over all employee growth appears to be the least at the Office of the President. We shall find this information useful in our later discussions.

TABLE 1.  Total Employment Numbers at UC:  Management Grows the Most
Personnel Classifications
Oct 1996
Oct 2006

SMG & MSP (Management) [M]
 +118% <<<

Academic Staff:

  Academic Administrators [S]

  Regular Teaching Faculty - Ladder Ranks [0]
  Regular Teaching Faculty - Acting Ranks [1]

  Lecturers [2]
  Other Teaching Faculty [3]
  Student Assistants [4]
  Research [5]
  Librarian [6]

  Cooperative Extension [7]

  University Extension [8]

  Other Academic Personnal [9]

Sub-Total  Academic Staff

Professional and Support Staff:

  Clerical & Allied Services [B]
  Communications-Arts & Graphics [D]
  Architecture, Engineering & Applied Svc [E]

  Fiscal, Management & Staff Svc [F]
  +98% <<<
  Food & Linen Services [C]
  Health Care & Allied Services [H]
  Maintenance, Fabrication & Operations [G]
  Protective Services [J]

  Sciences, Laboratory & Allied Services [I]
  Student Services [A]
  Other [Z]

Sub-Total PSS

TOTAL Employment

Student Enrollment

Numbers above count FTE (Full-Time Equivalent) positions, excluding DOE laboratories.
SMG = Senior Management Group
MSP = Management and Senior Professionals
The numbers and letters put in brackets [..] are the CTO/OSC identification codes.

Table 2. Ten Year Growth in Selected Employment Categories – by Campus
UC Campus
% Change in
Total Employees
% Change in
% Change in Staff
Category "F"
Office of the President
Los Angeles
San Diego
San Francisco
Santa Barbara
Santa Cruz
Total University

Source: Personnel-Related FTE Tables, for October 1996 and October 2006, at

The New Data on UC Employment

     Resulting from a request submitted through the Office of Public Records, I have obtained much more detailed data on UC’s overall employment statistics: Excel files listing some 2000 job descriptions, disaggregating the two dozen categories shown in Table 1, provide FTEs and some other interesting sorts of data. I am grateful to the staff at UCOP who assisted me in shaping my request for this information and for gathering it from the official electronic warehouse. [This was not free. There was a charge of $552 for the programming work needed to extract this data; and this was paid for out of my campus research funds.]

     In what follows I shall provide results of a very limited study of this new data, following the line of analysis started above.  First, we look at the first level of sub-categories for the Management positions.  Table 3. shows the relevant information for the four major components of code M jobs.

Table 3. First Details of the SMG & MSP (Management = Code M) Positions
 FTE @ 1996
 FTE @ 2006
 % Change
M05 Executive Program
M10 Managers
M20 Coaches & Related Professionals
[A]-[J] Senior Professionals

     The sub-category M05, the Executive Program, (also called the Senior Management Group or SMG) has job titles that include President of the University, Vice Presidents, Chancellors, Vice Chancellors, Deans and Directors. These are the positions that were at the heart of the recent scandals over Executive Compensation. It is a rather small group, comprising a mere 4% the whole M category of jobs, and, notably, it has hardly changed in size over this past decade.  We shall refer to this fact later on, in our discussion of theory of bureaucracy.

     We see that the greatest weight of the oversized growth in Management is attributable to the subgroups M10 and Senior Professionals. [I do not comment on the M20 situation.]  I will return to this after we have examined the category F jobs.

     For the job category coded F (Fiscal, Management and Staff Services) the new data shows seven different sub-categories; for simplicity, I show only the four largest components (which cover about 95% of this whole category) in Table 4.

Table 4. First Details of the Staff - Code F - Positions
 FTE @ 1996
 FTE @ 2006
 % Change
F10 Computer Operations
F15 Computer Programming & Analysis
F20 Admin, Budget/Pers Analysis
F30 Management Services
     Total (All Code F)

     To get some understanding of what these jobs are all about, let me first look into the largest component, F20. The official Job Specification for these positions can be found on the UCOP web site, as follows.

Administrative Analysts conduct or supervise responsible and complex administrative analysis requiring a knowledge of University administrative organization, policies, procedures, and practices; and perform other related duties as required.  Incumbents make analytic studies for campus or University-wide administrative officers; study existing and proposed administrative organizational structure, policies, and procedures; plan details of administrative studies; determine and locate sources for collecting information and data; review, analyze, and summarize reports of administrative officers, committees, and agencies; prepare directives, regulations, and other instructions for issuance to subordinate administrative units; provide consultative service in administrative management to departmental administrators; develop and recommend new administrative organizational structure, policies, and procedures; and establish and maintain contact with officials in the University, government, and industry for the collection and exchange of information.

     That seems to be an excellent definition for what we call a bureaucrat. (I hope readers will understand that I use that word, “bureaucrat,” in the descriptive rather than pejorative sense.) Furthermore, the Minimum Qualifications for such a job are stated to be:

Graduation from college with a major in business administration, economics, statistics, educational administration, political science, or an allied field, and […] years of experience in administrative analysis or operations research; or an equivalent combination of education and experience.

     All this does seem to support my original suspicion that the growth in job category F was tied to the growth in the job category M (Management), since these positions (F20) seem to be the top level of support staff for the University’s administrative officials. Bureaucratic accretion is the name given to the process whereby administrators proliferate themselves and expand their dominions. The growth rate of this F20 group is huge, comparable to that of the Management group.

     A contrasting component here is F30; these are mostly the Management Service Officers, who are stationed in each academic department. This group has grown very little and so I shall exclude it from my calculation of excessive administration.

     One also sees, in Table 4, that a fair portion of the F jobs – F10 and F15 - are in specifically computer-related activities. So the rapid growth in this sub-category may be attributable to the universal growth of computer usage in any large organization, and not something to be included in a study that focuses on administrative accretion.  Upon closer inspection, I find that a certain portion of the A-J Senior Professionals in the Management category discussed above also are identifiable as F15 personnel; I shall subtract them out, along with the F30 component, and arrive at the results shown in Table 5 for Adjusted Management Data.

Table 5. Adjusted UC Management Data - Seen as Excessive Over 10 Years
 FTE @ 1996
 FTE @ 2006
 % Change
Code F *
* The components F10, F15, and F30 have been removed, as has been the SMG.

Estimation of Bureaucratic Excess and Annual Cost

     We shall use the rule that any 10-year management growth in excess of the overall employment growth (31%) is excessive. This is a generous prescription given the principle of “economy of scale” – which implies that higher levels of management should not grow as fast as the total employment base. This principle is well borne out in the observation that the Senior Management Group has not grown at all over these ten years. This rule leads us to the following numbers for excess FTEs.
MSP* excess     =   5842 – 1.31 x 2637 = 2388
Code F* excess = 11439 – 1.31 x 5177 = 4657
If we had data going back farther than ten years, we would undoubtedly see larger numbers for the administrative excess; but we shall stay with this data for now.

     To get a dollar figure, we need to know how much those positions pay, on average, per year. For the MSP grades, the midpoint of the annual salary range is about $120,000 and for the PSS grades it is about $68,000. The resulting total cost of this excess in administration thus amounts to  $603 million per year.  This is the same as my earlier estimate (in Part 12) but here, based on the new data, it represents a much sharper focus in locating the administrative bloat.

     I do not claim it is proven that all of that $600 Million is wasted but, given the data presented here, I do challenge UC officials to demonstrate that it is not.

Discussion A – Theory of Bureaucratic Accretion

     I have no particular expertise in this subject, so I rely upon the most well-known authority. From his obituary in The New York Times, March 12, 1993:

   C. Northcote Parkinson, 83, Dies; Writer With a Wry View of Labor
     … British historian and writer who propounded the notion that  "work expands so as to fill the time available for its completion" …
     He argued that administrators and executives tend to make work for each other, and that because executives prefer to have subordinates rather than rivals, they create and perpetuate bureaucracies in which power is defined by the number of subordinates.
     A committee, he said, "grows organically, flourishes and blossoms, sunlit on top and shady beneath, until it dies, scattering the seeds from which other committees will spring."
     No matter how much work is actually getting accomplished, Mr. Parkinson wrote, the number of workers in an organization would relentlessly expand at a rate that he calculated, perhaps tongue in cheek, between 5.7 percent and 6.56 percent a year. …

     It is remarkable how well the data described above for UC fits in with this general theory.  The fact that the top level of UC administration (SMG) has not grown in size, fits perfectly with Parkinson.  The fact that the next two levels down, in the administrative hierarchy, have grown inordinately large does fit well with Parkinson. (In Table 5 we noted a 121% growth in ten years, which corresponds to an average rate of increase of 8.3% per year; this is a bit larger than Parkinson’s cheeky estimate of around 6%.)

Discussion B – The Regents’ Effort at Restructuring the UC Administration

     There is a very interesting new development called the  “UC Restructuring Initiatives,” aimed at the top level administrative operations of the university and it has been initiated by the people at the very top, specifically, the new Chairman of the Board of Regents, Richard Blum.  A big management consulting firm, Monitor Group, has been hired (at a cost of $7 million) to review UC administration, focusing on the Office of the President (UCOP), and to propose improvements in operations.

From the first official announcement, April 24, 2007:
“With this effort we have a great opportunity, for the first time in nearly 50 years, to take a fundamental look at the University’s administrative structure and align it with the modern needs and expectations of the University,” Blum said. “Not only will this effort result in a clearer and more effective administrative organization, it will achieve new efficiencies and cost savings to make the best possible use of the University’s resources.”

In an accompanying Q&A statement I found the following incongruity:
Q. What will be the impact on employees?
A. This effort is not aimed at cutting jobs – it is aimed at creating an organizational structure in which roles and relationships are better defined than they have been in the past, and where administrative processes across the system are more efficient and effective.

     My leading question is, How much of this is for public relations and how much is a serious attempt to improve administrative efficiency in this large organization?  After the first open discussion of this plan was presented at the May meeting of the Board, I sent an email to the official from Monitor who was leading this project, pointing to the data on UC administration I had collected in Part 12 of this series and offering to meet and provide further useful information.  There was no response at all.

     When another report was presented at the September meeting of the Regents, we heard more about what had been done and what could be anticipated.   According to Monitor Group, extensive interviews with administrative and faculty leaders on the several campuses revealed “widespread dissatisfaction” with the performance at the systemwide headquarters. New procedures would have to be created to provide satisfactory service and restore confidence; and Provost Hume was prepared with a set of initiatives to achieve those efficiencies. Executive Vice President Lapp offered the estimate that those reforms could lead to a savings of $20 – $25 million in next year’s operations and a similar amount the following year; but she also acknowledged that there would be substantial costs after that, around $100 million, to provide new IT systems that would be needed.

     There was some healthy skepticism heard.  One regent asked whether the bad reports from the campus’ leadership might reflect a contest over turf: less control from UCOP would mean greater autonomy and greater power on the campuses. (”What a way to control your own destiny!  What a way to control your own budget!”)

     How does all this relate to my analysis given above?  I do not necessarily see a conflict here. Rather, it appears to me that these are two distinct approaches and they are complementary; both of them ought to be pursued. If the primary object is to save money, then clearly my analysis points to a much higher priority target. But I do not denigrate the idea that a more effective operating system at the top is possible and desirable.  Frankly, I believe the choices of what to emphasize are political. That is not a bad thing; it just needs to be acknowledged.

     What do I mean by “political”? The UC Office of the President has really been disgraced and weakened by the recent scandal over executive compensation; and President Dynes’ role in that mess was inescapably awful. So it is easiest to pick on that single target, UCOP, and that is what Regent Blum has done.  My studies, however, point to the several campuses as the major seat of bureaucratic bloat and waste; and that will be a much more difficult set of stables to clean because of the solidly entrenched interests there.  This whole thing is very dynamic right now; and I would argue foremost for the maximum of openness.

A By-Product:  The Measure of Unionization

     One additional set of information in the new data set I have described above is the indication of which UC employment positions are represented by a collective bargaining agent, i.e., are unionized. Table 6, below, summarizes that data.

Table 6. Percent Unionization (FTE) by UC Employment Categories – October 2006
Job Classification
% Unionized

Job Classification
% Unionized
SMG & MSP (Management)

  Prof. & Support Staff:
  Academic Staff:

[A] Student Services
[S] Acad. Admin.

[B] Clerical
[0] Regular Faculty

[C] Food & Linen Svcs
[1] Acting Faculty

[D] Comm-Arts,Graphics
[2] Lecturers

[E] Arch & Engr
[3] Other Teaching Faculty

[F] Fiscal, Management
[4] Student Assistants

[G] Maint, Fabr & Ops
[5] Research

[H] Health Care
[6] Librarian

[I] Sciences, Labs
[7] Coop. Ext.

[J] Protective Svcs
[8] Univ. Ext.

Total Employment

     It is no surprise to see little unionization on the academic staff side (the faculty have the Academic Senate to represent their interests); by contrast, most of the professional and support staff are heavily unionized. It is interesting that the staff category F, which has been a focus of the study above, shows the smallest percentage of unionization on that side of Table 6. (In fact, when the computer-related portions of F are removed, the unionization there drops to zero.)