Innovators in Action 2014

Pioneering State-Level Pension Reform in Michigan

Interview with Douglas B. Roberts, Ph.D., former Michigan State Treasurer

In 1996, Michigan became the first state to enact major reforms to its state employee pension system. It closed the defined-benefit system to new hires and created a parallel defined-contribution system. Michigan’s pioneering reforms have served as a model for similar actions taken later in Alaska, Utah and other states.

Michigan’s pension reform efforts were a major policy initiative developed and promoted by the administration of former Governor John Engler. Former state Treasurer Douglas B. Roberts, Ph.D. led the administration’s policy development, from crafting the details of pension reform legislation, to meeting with relevant stakeholders, educating the legislature, and engaging in activities critical to its successful enactment. Roberts also played a similar, major role in the passage of Proposal A, a 1994 constitutional amendment that overhauled school finance in Michigan.

Reason Foundation Director of Government Reform Leonard Gilroy recently interviewed Roberts on the factors that prompted Michigan’s historic pension reforms, how proponents made the case for reform, lessons learned from the reform effort, the rationale for Proposal A, and more.



Leonard Gilroy, Reason Foundation: You served as Michigan’s State Treasurer from 1991–1998 and again from 2001–2002, and that’s a fairly unique position in Michigan relative to other states. Can you explain why?

Douglas B. Roberts, Ph.D., former Michigan State Treasurer: In most states, the state treasurer is an elected position. Some state treasurers who are elected often later run for higher office, such as governor or senator. In Michigan, the position is appointed by the governor. As a result, it’s a fundamentally different job and a very special position. The Michigan treasurer is constitutionally the sole fiduciary for the pension system. In addition, the position is responsible for all of the tax collections, for all cash management, all bonding activities and even things like Pell grants. Michigan’s treasurer also represents the governor to promote the governor’s policies before the legislature. The position functions more like the U.S. secretary of the treasury. When I spoke or testified before the legislature, I represented the governor and his positions. Michigan’s treasurer holds a completely different job than in most states, and that continues to this day.

Gilroy: Michigan was the first state to take on the issue of reforming public sector pensions—enacting reforms back in 1996—and you were Governor Engler’s point person on the issue. Can you explain what factors contributed to the push for reform?

Roberts: The single most important factor is that we had a very strong, extremely knowledgeable, and extremely policy-interested governor in Governor John Engler. That makes all the difference in the world. At that time, both the school employee pension system and the state employee pension system in Michigan were fully funded. The change was therefore not driven by an immediate financial crisis, but instead by the vision that the state could, and in time, face one.

It was an issue driven entirely by the knowledge that over time, the system did not benefit taxpayers, in our judgment, nor did it benefit most of the public employees themselves. The idea was to move away from an outdated system—and by that I’m talking about a system where you work for an employer for 30 years, then you retire and collect a pension. Even in the late 1990s, it was clear that society was changing. Our technology and our workforce were changing. People were no longer working for a single employer over their lifetime, but taking advantage of our strong sense of upward mobility. An individual might change jobs five, six, or seven times over a career. It’s very difficult to vest in a pension system when you’re changing jobs so frequently, assuming that your workplace requires 10 years of work to qualify for pension benefits.

Michigan’s pension system, which was very similar to other pension systems in the country, required a vesting period of 10 years. This is a very important point. According to information that I developed while state treasurer, 50 percent of state employees who worked for Michigan government didn’t work the necessary 10 years to vest in the pension system. This meant that 50 percent of state employees who ever collected a state paycheck at any point in their lives did not receive pension benefits at all. That was one of the drivers of reform.

Another driver was the uncertainty of market performance. If the state was responsible for investing all of the money—which it was— and if we invested wisely and returns exceeded the amount of money necessary to cover the pension benefits, which had happened in the past, then additional benefits were paid. Conversely, benefits did not decrease when markets went down. The only ones to lose were taxpayers, who paid for the increased benefits.

With those factors in mind, the issue became how to shift from a defined-benefit system to a defined-contribution system. To gain support for our position, we proposed that state employees should have a choice: those currently in the defined-benefit system could stay within that system OR could choose to move to the new defined-contribution system.

This choice did not apply to future employees. That limited the impact. If people were close to their 10-year vesting period, then I’d tell them “for goodness sake stay and get vested.” But for those recently hired, I’d advise them to consider their next few years of employment status and make a decision based on that. Those who were going to leave state government before vesting would find the defined-contribution system to their benefit. They would at least leave with retirement benefits.

There was yet another position. Our research determined that, at that time, there were about 14 major public pension systems run by about 200 individuals within the U.S. Those 200 people controlled pension funds with a total value in excess of $1 trillion. That gave them the opportunity to exercise influence over the private sector via proxy voting. Should the public sector have such influence over the private sector? Large mutual funds can exercise similar influence over pension systems via proxy voting. However, the heads of those mutual funds are seldom concerned with political outcomes such as who is elected governor.

In leading the campaign, Governor Engler insisted on high standards. He asked the executive staff to draft all the bills necessary to implement such a proposal—down to every detail of every statute that had to be amended. He also wanted those drafts on his desk before he would go public with the proposal. That way, when questions and opposition were raised, questions could be answered and objections could be countered.

I would tell anyone looking to do this that they need to assign someone to sit down with a roomful of pension experts and lawyers to go line by line over everything that’s changing. There are often hundreds of pages of pension laws, and it took a long time to go through every one of them. But once we did, those changes became the basis of the reforms we ultimately enacted.

All of these things combined to create the momentum we needed to propose—and win—the reforms.

Gilroy: The reforms enacted consisted primarily of closing the defined-benefit (DB) pension system to new hires and launching new defined-contribution (DC) system for state employees. Can you describe the rationale for shifting from the DB to DC system?

Roberts: The model for shifting from defined-benefit to defined-contribution was already in place—in the private sector and in several university systems. Governor Engler’s goals were to create a fiscally sound retirement system to the benefit of the taxpayer and the state employee, and one that protected the state’s assets regardless of market performance. He also sought a system in which state employees weren’t bound by “golden handcuffs.” He wanted them to be able to leave state employment when they wanted and to keep their retirement savings whenever they left government employment.

At the same time, he sought a retirement system the state could afford for the long term, regardless of market ups and downs and one which would relieve future leaders of constant worry about pension liabilities.

Pensions represent future costs, and the future can be unpredictable. Forecasting future costs can be imprecise. You can change the assumptions upon which actuarial decisions are made. Accounting assumptions can change. As we know, economic conditions can change. But short-term decisions made in the past don’t always fit the longer-term view of the future. The administration’s goal was to leave office without leaving problems for future leaders to solve.

Gilroy: What steps did you take to build the case for reform?

Roberts: We first built a base of arguments. Would the transformation be a benefit to the taxpayer? I think the answer is clearly “yes.” Is it a benefit to the employees? We built the case that reform would be in the best interest of a majority of employees.

We also built a coalition. For example, at the time we were proposing the pension reforms, we were also proposing an “early out”—an incentive to retire early from the existing defined-benefit system. State employees close to retirement in age and in years of service were provided the opportunity to retire early. That way it was easier for some state employees to testify before the legislature that they were in favor of the early retirement option and that the transition would not affect any current employees.

Even with all our hard work, passage in the legislature wasn’t easy. We weren’t able to win legislative approval for a shift in benefits paid to school employees, and the measures barely passed to make the change for state employees. At the time, Republicans controlled the governor’s office, the House, and the Senate. In my opinion, such reform was passed, although narrowly, because of two facts—the system was fully funded and the new system didn’t affect any current employees.

But I think we ultimately made the case that in the long run we took great care to design a new system that would benefit the state and current state employees. We met individually with state legislators, mostly Republican, to win their support. We met with legislative leaders and didn’t take votes for granted. We had support from the business community, and business leaders contacted legislators as well. Our story was always one of positive benefit to current state employees. It’s hard to generate opposition from those who aren’t affected by a change.

Gilroy: How much of an understanding of the pension situation did the legislature have when you started talking to them to make the case for reform? What convinced them to ultimately support your efforts?

Roberts: This reform came at a time in Michigan before the state enacted term limits. Michigan now has one of the strictest term limit laws in the country. At the time, we had a number of legislators who’d been on the job for a while, and we had chairs of committees covering pension issues who were really knowledgeable and understood the proposal. That did help in terms of trying to convince those who may have been undecided.

That said, it did ultimately become partisan, meaning that Republicans were for it and Democrats were against it. One of the most helpful things was that Governor Engler had proposed similar measures once before. It took two legislative sessions for the changes to be enacted. The administration tried not to offend anyone in the earlier effort. How you win is important. How you lose is also important. When you’re losing on an issue, don’t offend the other side. Sooner or later you’re going to need to convince your opponents all over again. How you lose can be as important as how you win.

We had lost the first time around. So we backed away and made a few minor changes. One of the things that we learned was that we had to make sure that we could answer all the questions and all the objections. We didn’t want to get beat up for not having answers. Somebody has to be saying that you can agree or disagree with my answer, but here it is.

We didn’t give up. We were one vote away from approval in the legislature. We worked to get the vote and the measure passed. It was close, but let’s face it, an awful lot of things in history happen by one vote.

Gilroy: What types of opposition to reform did you run into along the way, and how did you address those challenges?

Roberts: We all have different views of investing. We know that personal investments can be challenging and even professional money managers can be wrong. One of the arguments was that we are not all equally good at managing finances. The argument was that the employees would make unwise investment decisions with their retirement money. Our response was to have confidence in people’s ability to take personal responsibility and to welcome the freedom to manage their own finances. We also pointed out that private industry and university staff and faculty members have long been successfully managing their own personal defined-contribution plans long before the state had established its system.

Gilroy: Some policymakers have shied away from pension reform, fearing high transition costs involved with the switch from a defined-benefit to a defined-contribution system. Was this an issue in Michigan?

Roberts: Transition costs often refer to the fact that if a defined-benefit system is underfunded, then replacing it with a defined-contribution plan increases total costs. During the transition period, both systems must be paid for, which is an added cost. In Michigan, the system at the time was fully funded. However, if the system to be reformed is not fully funded, then costs can be an issue. Some will argue that they can’t afford reform. I would argue that that represents only a short-term horizon. If you’ve already made a commitment to a defined-benefit system, then you have to cover those costs. And if you shift from a defined-benefit to a defined-contribution system and the defined-benefit system is underfunded, then you are going to have to continue to pay into the defined-benefit system and you’re going to have to pay into the defined-contribution system at the same time. That’s absolutely true.

Here’s one potential way to handle that: I would propose closing the defined-benefit system, issuing pension bonds, and putting that money into your pension system to make it 100 percent funded. You do have to pay off those bonds, but given the current interest rates—unless the whole system collapses—you will find this is a much lower-cost way to manage the transition. Your payments to secure total funding will then be at a fixed, known rate. You can now set the defined-benefit system aside as fixed, close it off and shift to the defined-contribution system.

Most pension systems today are still assuming a seven to eight percent rate of return. States can borrow today at a much lower rate. Now of course this has to be through taxable bonds, not tax-exempt, because of arbitrage laws. New Jersey and other states have chosen similar solutions. You issue the bonds and put the money into the pension system, where it will earn interest at a greater rate than the interest you’re paying on the debt.

Gilroy: What have been the key outcomes of the reform? How would you assess the financial risks that Michigan taxpayers face today, post-reform, with regard to the state’s pension system?

Roberts: Consistent with the system’s design, when examining the current total state employment, the number of people on the defined benefit system is shrinking and the number of people in the defined contribution system is growing. I don’t think that there’s any question that as markets go up and down, this is a major plus. The new plan has continued in effect for almost 20 years. Ultimately, all of those who elected to stay in the defined-benefit system will retire and following that, there will be no new retirees in the system.

In addition to the “early out” adopted in 1997, another was adopted in 2002. The two offers together did increase the state’s costs. As of the end of September 2005, there were 45,800 retired state employees in the state employee pension system. Out of that number, 8,000 retired early. That’s 17 percent of all retirees at that time had opted for an “early out.” They retired earlier than the actuaries expected, and they earned an increased benefit to do so. Therefore the “early out” provisions were an added cost. Without these added costs, the benefit to the state of moving to a defined-contribution system would have been more apparent.

Another issue I want to raise concerns the investments of the $40 billion fund. The benefit of this fund is for current retirees and future retirees. It is not state money. Let me repeat: it is not state money. People do not understand that concept. The money belongs to the retirees, and a state official—the state treasurer—is responsible for investing the retirement fund. But people will come to the pension system and ask the fund to invest money in economic development. I’m not against that if it’s a good investment, but if it’s a bad investment I’m not going to invest in it because it’s the retirees’ money. A fiduciary is a legal term that means that you must act in the retirees’ best interest, not in the best interest of the state or taxpayers. But you have problems with pension systems being abused and people making investment decisions because of political considerations.

Over time, I was asked to review external investment proposals that would tap the state’s retirement resources as though they were fully a state-only asset. Those proposals viewed the retirement resource as taxpayers’ money. While it’s true that the taxpayers did contribute to the revenue for state government employee retirement, once that money was moved into the pension system, then the fiduciary is in a very difficult position. The fiduciary also has to say, “Yes, I am a state employee,” and “Yes, I want to promote economic development in Michigan.” But the fiduciary must also say “I have a legal, moral and ethical responsibility to protect the people who this money now belongs to.”

That’s the same philosophy that questions whether the influence of large pension systems is positive or not. As I mentioned before, in 2004, 14 public pension systems controlled $1.256 trillion in assets. Those assets were controlled by 210 people. Once again I ask: Is such influence positive?

Gilroy: Are there any other lessons learned from Michigan’s pension reforms that you would offer to those in other jurisdictions who may be contemplating similar efforts?

Roberts: Do your homework. Do your research. Know the laws. Design your system for the broadest possible benefit, to deliver responsible government and to see your community well into the future. Don’t give up if at first you don’t succeed. Keep your allies—and your opponents—on your best side. Be a leader who leads, who isn’t afraid to be the first to innovate.

States aren’t the only ones with unfunded accrued liabilities—also called legacy costs—to consider. For example, in Michigan—and it’s different in other states—there are numerous local pension systems. In Michigan, we have a state employee pension system, a school employee system, and a couple of smaller systems for state judges and state police. But we also have numerous pension systems covering people employed by local governments.

Your state may have numerous pension systems but not great oversight over them. As a result, if they run into problems the state may be placed under tremendous pressure to either bail them out or to solve problems for them. So you need strong oversight, or at least an early warning that problems are coming.

Keep looking ahead. Technology, our modern workplaces, new devices, communications, pinpoint decision-making and world-wide competition make it imperative to assure freedom in retirement investing. As new generations of workers move from one job to another, the idea of defined benefits with 10-year vesting requirements is out-of-date.

The last thing I would say is that if your first proposal loses, be aware that the battle’s not over. Pensions are a long-term issue that need a long-term solution. And in politics, those are the toughest to pass. Politics is often about “what have you done for me lately,” and pension issues that stretch 10 to 20 years out, especially in states with term limits, may not command the serious attention they deserve.

Gilroy: You’ve said that in addition to pension reform, you’re also very proud of achieving major school finance reforms. Can you explain the nature of those reforms?

Roberts: In 1994, Michigan—without a court order—adopted a major constitutional change in the way we financed schools in the state. We called this Proposal A, and it was adopted by a vote of the people. Michigan had been trying to address school finance since 1972. I was part of six attempts to amend the state’s constitution to reform school finance that failed up until 1994. In Michigan we had 550 school districts, and each district was required to gain voter approval for property tax votes. Therefore, some districts that valued education voted higher property taxes. Some districts were home to more valuable property and were able to collect higher property tax revenues. Some districts didn’t have as much high-value property and relied on state subsidies. However, the disparity between the haves and have nots was large and was growing worse.

We substantially removed property taxes from school financing and imposed higher sales taxes. Sales taxes at the time were constitutionally limited to four percent. We voted to raise sales taxes to six percent and, in exchange, to reduce property taxes significantly. The result is substantial state control over what school districts collect in taxes and have available to spend. Today, we still have a disparity in spending among various districts, but not nearly the disparity that existed in 1993. And just as important, that disparity isn’t growing.

We also substantially reduced property taxes, which were always a major impediment in the state. I think that people voted for Proposal A to some extent because of the school finance aspect, but in larger part because we were reducing property taxes.

That particular proposal also included major changes to the way we deliver education in this state. That was the beginning of charter schools in Michigan, and it was the beginning of schools of choice. If you live in one district and another district is willing to accept new pupils, then parents can send their children to another district’s school.

We changed so many things. They were major, major accomplishments. I’m thrilled to have been a part of it. I think the system is fairer for the state. There are still a lot of people in Michigan who don’t like the reforms, but my response is that it was 20 years ago. The beauty of our government is that we can change for the better of all.

Gilroy: Is there any other advice can you offer to policymakers looking to tackle reforming large, entrenched systems?

Roberts: What you need is a very strong leader who is basically willing to go where other people haven’t gone. And let’s face it: for a lot of politicians, if someone else hasn’t done it, then they want to wait until someone else goes first. Not only was Governor Engler willing to go first, but he was extremely knowledgeable about government, extremely interested in policy, had spent years in the Michigan House of Representatives and the Michigan Senate, and had a deep understanding of the budget—better than anyone. When you have someone like that who’s interested in policy, they become an integral part of the policy decisions—not just at the end but during the development. That is huge. I recognize that you can’t just go out and find that, but when you do have it, take advantage of it for goodness sake.

The other thing I would say is that you should push your initial proposals a little bit further than you think is possible.


Douglas B. Roberts, Ph.D., is the former Michigan state treasurer and currently serves as the director of the Institute for Public Policy & Social Research (IPPSR) at Michigan State University. He holds more than 28 years of experience in Michigan government, including 10 years as state treasurer, time as director of the Senate Fiscal Agency, deputy superintendent of the Department of Education, deputy director of the Department of Management and Budget (DMB) and acting director of DMB. He also served for two years as a vice president with Lockheed Martin IMS. He holds doctorate and master's degrees in economics from Michigan State University and a bachelor's degree in economics from the University of Maryland. He has been affiliated with IPPSR since April 2003.

Other articles in Reason Foundation's
Innovators in Action 2014 series are available online here.

Leonard Gilroy is Director of Government Reform





;