A Special Report from the Office of MIT President Charles M. Vest • June 1998

President's Office | Reengineering at MIT | Tech Talk | MIT News Office | MIT


Ringing in the Changes

An interview with Senior Vice President Bill Dickson

With four decades at MIT -- and nearly 20 years as senior vice president -- William R. Dickson has supervised a tripling of MIT's physical size, and has managed the day-to-day operations of a university which, for many people, symbolizes the technological and information revolutions. Now preparing for retirement at the end of June, Dickson is still pushing hard to insure that MIT lives up to its national reputation for sound management, and that the innovations developed in the Institute's labs and research centers find their way into its facilities and administrative systems.

Back in 1993, MIT announced that the Institute needed to close a $10 million budget gap. Five years later, how are we doing?

In fiscal '97, the actual difference between income and expenses was somewhere around four and a half million dollars, and this year we project a difference of around $14 million. So I think on the whole we've held our ground, with smaller gaps in some years and larger ones in others, but all within a modest range.

Where do you think some of the important changes have occurred during that time?

We thought that, by operating on several major administrative areas, we could save the institution around $40 million annually. At that time, I think that the overall budget was around $1.2 billion. First, you have to factor out the research work both at Lincoln Labs and at the Institute, so you quickly get down to the point that, when we were talking about reducing expenses, we were really operating on around $400 million, not the $1.2 billion.

You said you identified a few key areas for change. What were those?

One of them was what has grown to be Management Reporting -- in other words, the structure within MIT for tracking and managing the business end of the Institute both centrally and in the departments. Another was the facilities area broadly defined, but particularly the operation of the physical plant. Another was the purchasing of -- and paying for -- goods and services.

So how did MIT do in pursuing the $40 million annual target?

Well, I think people in the community would remind you that -- over, say, the past five years -- they actually underwent three years of budget cuts from year to year of somewhere around two percent. What we haven't done is make such dramatic changes that we have been able to end the systemic imbalance. So, we've held our own -- but we haven't entirely made up for the fact that research funding has been more or less flat -- not only in terms of total dollars, but also in terms of what those dollars can be used to pay for.

This is the so-called "cost shifting" issue?

Correct, and this is one thing that affects our budget by a fair amount each year, and that continues to go on.

Another one of the costs we need to prepare for is that we will have to spend much more money on maintaining our buildings. Over the past 30 years, we were blessed to have many new buildings, but when a new building gets to be more than 25 years old, the mechanical and electrical systems become antiquated. I can remember when all these buildings were state-of-the-art, but now, of course, I know they need a lot more upkeep and rehab work.

So how would you rate the changes to date in terms of successes and stumbles?

Well, in anything we've done we've had some stumbles, because when you make change, you're bound to have ups and downs in the process. I can't think of anything that we have done that I would classify as a disaster for MIT. I think most of the things that we have done, in the end, have been beneficial, and I think the bulk of our constituents either would agree or would be neutral. Let me give you some examples.

One of the things that I had started studying before we started reengineering was the mail system at MIT (see related story). From a cost and service perspective, I think that, on balance, the new mail system is a success. What would you hear if you asked the community? Well, most would say it's not bad -- but I remember when changes were first initiated. We heard plenty of complaints. You had to go down and pick up your mail, and the sorters screwed up the delivery, and people didn't like the fact that we didn't switch stuff all at once, but only brought on mail centers as they became available and equipped. So that's one example.

Switching to the Student Services Center, you can see for yourself. I think that the students and community alike generally feel this venture has been a resounding success (see related story). As a student, you now don't have to wander all over the campus to do business with the Institute. You can go to one location, and putting that together with advances in the student information system, you can also do more things than you could before. So the two of them combined make for a much better life for the students. People forget that this isn't only about cost containment: we're trying to streamline structures and improve customer service. I think this is a good example of both.

Now, about some of our efforts in supplier consolidation: we closed the Office of Laboratory Supplies and went into partnerships with Office Depot in the office supply area, and with VWR and BOC Gases in the areas of scientific apparatus and supplies and industrial gases. We still haven't gotten to the point where people make full use of these partnerships, so we have work to go in getting the most out of the volume discount potential. People are still buying from other vendors, even when they can get the same product faster and cheaper from a consolidated vendor. So that's something that we have to review next.

The biggest success overall is the total revamping of the way the Plant operates. Custodial Services went from one big group of 250 people or so into about 26 self-directed teams. We have computers in all the team rooms so that they can deliver and receive information on e-mail. Many of them have been trained to use the computers, so they're picking up new skills and more opportunities for advancement.

The teams have a lot of autonomy. They decide how they are going to clean. Stock is delivered in off hours to the team room, so that custodians can begin cleaning as soon as they start a shift. Similarly, we have shifted to zones in maintenance. We have five zones on the campus. The team rooms are in the zones they serve.

As far as a group of people embracing change, the Plant has probably been the most impressive at the Institute. Why? Because the leader of the Plant consulted with the rest of the Plant people, many of whom are union employees, and together they went through a strategic planning effort before they embarked on reengineering. Also, everyone, including the union leadership, understood that they needed to find better ways to do business.

So you think that the experience with the Physical Plant was a more participatory process so you got more buy-in early on?


And that's been less true in some of the other areas?

The Plant is only one department -- a large department, to be sure -- but they spent a lot of time, with extensive participation, on reengineering and it paid off in the end. For other initiatives, which touch every department, it's harder. You can put together a task force or working group, and you can do pilots but, in the end, you've only got a relatively small group that's been involved in the discussion, so the rest of the organization can say "This doesn't have anything to do with me -- maybe if I ignore it, it will go away."

The biggest ongoing challenge is Management Reporting. Management Reporting has only accomplished a small piece of what they intended to do. Over the next six months, we will actually have the entire Institute as users of SAP, but then we must proceed to do two things. One, we need to take time to really clean up all the loose ends -- training, conversion and full implementation. Secondly, we must continue bringing up other modules, including a combined personnel and payroll module, and some additional modules for the physical plant. We've already converted the Plant stock system to SAP -- that's been done since last July.

But those two efforts will be fairly significant, and then we've got to move towards changing the way we do business, and one of those is the administrative system clusters -- and that's the thing that people don't like to hear about.

So let me ask about it…

Well, they don't like to hear it because it's threatening, but as you look ahead, it will be the only efficient way to do business in a large and diverse research university like MIT. We have one cluster in place with Materials and Chemical Engineering that's functional now. The staff is all trained in SAP. We have learned a lot from that venture. But clustering is going to have to continue because, as the Institute looks ahead, the balance between its income and expenses will -- all other trends remaining the same -- continue to widen dramatically.

Well, then, what do you say to the skeptics and veterans who see SAP and the entire Management Reporting initiative as a dubious venture? What kinds of things will SAP do for the Institute that aren't being done now?

The Institute functions reasonably well and so I understand when people say we don't need to change. But let me take the simplest of things -- one we're talking about right now. We're trying to interest the community in understanding and utilizing electronic journal vouchers, particularly as we get near the closing of this fiscal year. Electronic vouchers are faster, leave a clearer trail, and save substantially on paper. Their use will also help us close the books much more quickly.

As the number of transactions decline with consolidation and productivity goes up, we will have better, cleaner, faster electronic audit trails, which is something our research sponsors will expect from us. Also, none of our suppliers will shut us off when some department has a $32 overdue bill because a key manager is on vacation. And don't think that doesn't happen now, because it does.

And let's not forget that SAP will eliminate large parts of the so-called "Year 2000" problem for MIT. Other places are spending millions just to address this one problem, but, with SAP, the cure is just a side benefit of a system which is designed primarily to achieve other objectives.

I'm convinced on my way out that, from a financial prospective, the Institute will be forced to take full advantage of what these new systems offer, including clustering.


Number one, there's a significant change coming in how the government allows us to finance RAs and TAs. We're going to see it on July 1. The EB [Employee Benefit] rate will drop from approximately 40 to 27 or 28 percent. That means a dramatic change in the way RA/TA tuition and stipends must be financed and it's going to cost the Institute $16 or $18 million a year. Number two, those old "new" buildings that we have spent very little on for so long will require major expenditures. We figure that at a minimum we should start by spending at least $20 million a year in funds not now budgeted.

You said before that we think we can lean down about $40 million a year if we work on it, but you're now saying, "Well, we have to take $20 million of that and put it against maintenance."

Yes, and the only way we would ever get to the $40 million -- you know, we're part way there already -- but the only way we would get all the way is if we make the major changes in departmental clustering. And the longer we take to get there, the longer it will take to approach that figure.

That's a good transition to make to the next question but, before we do that, you cited cost shifting and long-term maintenance --

And capital projects. Even when we raise the money to build a new facility -- and we don't always raise all the money, especially for facilities like dormitories -- there are maintenance and debt service which generally find their way back into the operating budget.

So MIT must continue its push to realize greater productivity and resource management initiatives, and we can't do that unless we look at the advantages of administrative clustering, which will allow us to gradually reduce staffing. The opponents of clustering would say "Well, if you decrease the number of people, you will also decrease the quality of service."

No, that's wrong, because the reason that we stopped to invest in a tool -- which happens to be SAP; it could have been Oracle or some others -- was that we agree that, without changes in how we do the work, the critics could be right.

I think that with the full introduction of the broadest range of what SAP can help us do -- including things like payroll, personnel and a central data warehouse -- people will be able to get more information more easily, do what they do now far more efficiently, and the quality of what they provide to a service receiver should go up.

Another area where cost control and quality of service issues have come up is that of on-campus food service.

A I think the Institute is still in self-denial about whether in fact you can make a food system work if you don't have some degree of mandatory participation. I recently asked my colleagues at the 30 largest private universities how many of them had a non-mandatory food plan, and the answer was "zero." Here at MIT, we're still trying to get by without one, and I really think that is a big issue. Now, I understand what the pressures are, and I understand when people say "Well, we'll make it so good that people will want to participate more, and it will basically be economical." I wish them good luck, but I personally don't believe it.

Have there been other particular areas where the experience of other schools has been useful to MIT?

Well, I'd be hard-pressed to define them exactly, but certainly I would have to say yes. I've personally learned a lot from the people at Penn because they were the other school that was in this business of reengineering in a big way about the same time we started up. Now, most schools seem to be in it, but Penn has been very dynamic. They just outsourced their entire Plant operation -- including all Plant employees. I don't agree with everything they've done, but they've been very aggressive.

The final question is one of the most frequently asked questions of MIT faculty and staff: What about the possibility of another round of early retirement incentives? What factors would be weighed in deciding whether or not to do such a program, and is anybody looking at that at the moment?

No, we're not looking at it. I don't personally believe that we will have another incentive retirement program, at least for some time. However, there are discussions about changing some of the provisions of the pension plan that could have the indirect result of making it more attractive to retire earlier than the current so-called normal retirement age of 65. I think the big issue is the question of faculty renewal. There may be ways of including changes that naturally move us toward a younger faculty. Perhaps we can create a system that allows very senior professors to continue on a part-time basis but still makes room for younger people. But in general, I don't see a need to push for another incentive retirement program at this time. Even with all of the changes we've talked about -- in Management Reporting systems, clustering and other areas -- resulting personnel changes can, by and large, be accommodated through normal turnover, which is a lot higher than most people think.

President's Office | Reengineering at MIT | Tech Talk | MIT News Office | MIT