Implementing-Defined-Contribution-Pension-Reform

1
Implementing Defined Contribution Pension Reform: Five Case Studies
Rich Danker
February 2013
Abstract
States and cities are considering shifting from the defined benefit to defined contribution model
of pension funding for their employees, a transition made by the private sector beginning in the late
1970s and the federal government in the mid-1980s. This reform would come with the benefits of the
removal of a contingent liability from the government sponsors, long-term savings for taxpayers and the
spread of portable retirement plans for workers. It would also equalize the quality and level of pension
compensation between public-sector employees and the rest of the labor force. Experience shows that
to be successful such overhaul requires a thorough explanation to taxpayers that the defined
contribution model is a better deal, a credible replacement plan for beneficiaries, and strong political
leadership to withstand opposition. Five case studies – the private sector, the federal government, and
the states of Michigan, Utah, and Rhode Island – are presented here and are derived from research and
interviews with some of the principal players to tell the stories of successful pension reform.
Introduction
A New York Times article has called spiraling public employee compensation costs “the public-
policy issue of our time”1
. Fair-market valuation shows that public pension plans are underfunded by
approximately $4.6 trillion2
. Expanding workforces, sweetened benefits, and battered investment
positions through the last decade have come due in the form of higher pension costs for states and
cities. In the absence of a clear legal route to reduce benefits, governments will likely have to raise taxes
and redirect public resources to make their pension beneficiaries whole. Some have called for a new law
1
Bai
2
Biggs

2
to allow the states to enter bankruptcy to shed these liabilities or for a bailout program from Congress.
It is clear that the defined benefit model of pension funding in current form is unsustainable.
Fortunately, there is already a narrative of successful pension reform in the private sector, the
federal government, and a handful of states. It shows how the defined contribution model provides
retirement compensation in a way that is fair, fiscally sustainable, and in sync with the labor market. The
nation’s states and cities, which sponsor a total of 3,400 pension plans, have a clear replacement for the
current pension structure that they can no longer afford. Achieving this in the face of the political
resistance by public-sector unions will require winning arguments associated with every major aspect of
pension reform, including 1) emphasis on truthful data 2) design of an appealing defined contribution
plan 3) the moral case for equalizing pension compensation among all employees 4) the savings offered
to taxpayers.
Defined Benefit vs. Defined Contribution
The essential difference between the defined benefit (DB) and defined contribution (DC) models
of pension compensation is the obligation of the returns. The sponsor of a DB plan promises the
participant a certain level of perpetual monthly payments in retirement, typically calculated from four
main factors – service time, salary, salary multiple, and annual cost of living allowance (COLA). In a DC
plan (also known as a thrift plan), the participant is entitled to the money that has been contributed to
his retirement account plus investment returns. This can typically either be accepted upon retirement as
a lump sum or converted into an annuity. In the DB plan, the pension return is guaranteed; in the DC
plan, it is a matter of investment earnings.
In the public sector, this means that for the 84 percent of state and local workers enrolled in a
DB plan3
, the government sponsor is liable for making payments it has promised them in retirement. Or,
in other words, taxpayers are responsible for meeting their pension benefits. These benefits are
considered by almost all states to be contracts and offer the protections for them afforded in the
contracts clause of the U.S. Constitution and equivalents written into state constitutions. Some states
also have constitutional or statutory provisions that specifically protect pensions from modification.
Defined benefit pensions are a guarantee from the government to the employee of a certain amount of
pension compensation in perpetuity.
3
Wiatrowski

3
Defined contribution plans, by offering no guarantee of the size of the pension payout in
retirement, remove the contingent liability from the government to its employees. Pension
compensation is derived from contributions, not promises, and is tied to the market performance of the
invested contributions. Once an employee retires, the employer is finished securing his pension and the
taxpayers bear no further responsibility. The DC pension is delivered upon retirement, while the DB
pension must be maintained through retirement (although the DC account can be converted into an
annuity).
This fundamental difference in obligation makes for different funding mechanisms and
structures. Defined benefit plans pool contributions and invest them with an anticipated rate of return
(usually around 8%). Defined contribution plans place the contributions in the employee’s retirement
account where he can invest them according to a menu of options. While the split in contributions
between the employer and employee varies, it is important to note that for public workers, the
government sponsor is their single source. Government pays the entirety of both components of
contributions, either directly or in association with employee salaries. Unlike the largest public defined
benefit pension plan, Social Security, there is no contributory tax from participants to fund state and
local DB plans. All contributions come from the government sponsor, some of which may be shown as a
contribution of employee wages but are nevertheless still made by the employer. Any increase in the
employee contribution rate is still borne by the sponsor and can even be offset from the employee’s
point of view with a salary increase.
In the DC model, contribution levels from the employee perspective may fluctuate depending on
how much the worker chooses to have withdrawn into his retirement account. The government is still
the source of these monies, but they are taken as automatic savings rather than wage earnings. Since
the government bears no liability for making pension payouts in this system, the employee’s chosen
contribution rate is of no concern to its funding responsibility.
The government’s funding obligation for the DC pension begins and ends with its contributions
to the employee’s retirement account. There is no embedded contingent liability. In the DB pension, the
government is obligated to fund both contributions and any difference left over when the benefits come
due. This responsibility to meet both a statutory contribution rate and specified pension payout for each
participant places a tremendous financial liability on taxpayers. The difference between the two models
is stark. The DC model is a form of deferred compensation – the DB model is a government entitlement.

4
The Private Sector: Beginning of the Revolution
The DB approach had been the pension model for most firms offering retirement plans,
particularly in the postwar period of corporate expansion and conglomeration. However, in many cases
these pensions were inadequately funded or poorly administered. Beginning in the 1960s there was a
growing political movement for strong federal standards to regulate private-sector pensions. On Labor
Day 1974, President Gerald Ford signed the Employee Retirement Income Security Act. ERISA, as it
became known, is a comprehensive regulation of employer pension plans. It set minimum requirements
for funding, vesting, and benefits while imposing fiduciary responsibilities. One of its most prominent
features was the Pension Benefit Guaranty Corporation, an independent agency whose main purpose is
to operate an insurance program for employee pension benefits funded by premiums on the employers.
ERISA also extended to the issue of financial accounting. It mandated that private pension plans
follow actuarial methods and assumptions in calculating funding levels. Firms no longer had the
discretion to determine their own accounting policies but were subject to the newly-created Financial
Accounting Standards Board (FASB), a non-profit group formed in 1973 to set generally-accepted
accounting practices (GAAP) for U.S. corporations.
The most important accounting guideline imposed by ERISA pertained to the discount rate in
measuring pension liabilities. The discount rate is a crucial factor in calculating (or “discounting”) future
cash flows to the present value. In pension financing, the obligations to be paid out to retirees in the
future are calculated as present-day liabilities using this variable. The lower the discount rate is, the
higher the value of a liability. FASB directed plan sponsors to discount their pension obligations using
interest rates from Treasury bonds, considered to be risk-free securities because of the federal
government’s credibility not to default on them. This forced companies to book their pension debt at
market value, preventing them from undervaluing what they owed their employees.
Just as ERISA imposed stringent standards on private defined benefit pension plans, changes in
the federal tax code beginning later in the 1970s made the alternative DC model more appealing to both
company and worker4
. The 1978 Revenue Act made employee contributions to qualified retirement
accounts tax-exempt. This new section of the tax code took effect in 1980 and spawned an eponymous
4
Butrica, Iams, Smith & Toder

5
investment product, the 401(k), which became the de facto vehicle for private-sector DC retirement
plans. Pre-tax contributions could be invested there and grow tax-free. It was equally favorable from a
tax standpoint to the employer and employee. ERISA had also created the Individual Retirement
Account (IRA) for those not covered by employer plans, and the Kemp-Roth tax overhaul in 1981 made
them available to all workers. However, the IRA came with significantly smaller caps than the 401(k) and
the contributions were subject to Social Security taxes.
During the early 1980s many blue chip firms designed and opened 401(k) plans, including
Johnson & Johnson, PepsiCo, JC Penney, and Honeywell. By the middle of the decade 294 large pension
plans had been shut down5
. Surveys showed that nearly half of large companies were either offering the
DC option or considering it. The results were dramatic. By 1984 there were already 7.5 million plan
participants with $92 billion in assets. In 1990 there were nearly 20 million participants with $385 billion
in assets.
There were several factors besides the preferential tax treatment that made the 401(k)
appealing. The advent of a long-term bull market expanded the number of retail investors and piqued
the interest of baby boomers in managing their retirement savings. Meanwhile, the U.S. economy
moved away from manufacturing and toward the services and technology sectors, where firms
considered the 401(k) a better fit for their financial profiles and workforces. The labor market was
becoming more fluid as the hallmarks of the postwar economy (unionization, career employment) faded
away. The portability and personalization of 401(k) accounts had innate appeal for workers in the new
economy. By the height of the tech boom in 1998, they had 37 million participants with $1.5 trillion in
assets. The Watson Wyatt Retirement Attitude Survey taken in 2005 found that the 401(k) was viewed
as the most important source of future retirement income compared to other potential sources6
.
The market meltdown of 2008 rocked investment accounts. A survey by the Employee Benefit
Research Institute (the pension industry’s think tank) found that accounts with between $50,000 and
$100,000 experienced about a 15 percent decline in average balance for the year while those with more
than $200,000 saw a 25 percent drop in value7
. This was in the context of a 37 percent decline in the
S&P 500 Index of stocks, a nearly 20 percent average loss in home values, and a freezing of the money
5
Keller
6
Tower Watson 2005
7
VanDerhei

6
markets. Most asset classes have since recovered to pre-crisis levels (though not housing)8
. This
experience did not deter the spread of individual retirement accounts. A Towers Watson survey of
Fortune 100 companies found that in 2008, DC plans became more common than DB plans for new
hires. Furthermore, just one quarter of these mega-employers had traditional DB plans, with the rest
offering some kind of individual account. This trend increased in the next two years (post-financial crisis)
surveyed9
.
After a three decade shift in the style of private-sector retirement compensation, it is clear that
the DC model is here to stay and the DB model is fading. Two icons of the old economy, General Motors
and Ford, are offering buyouts to one-third of their workforces in a massive effort to shed pension
liabilities. Large firms, once the traditional sponsors of lifetime pension benefits, have found the
individual account to be a better fit for their workers and their balance sheets.
Federal Government: the Great Leap Forward
Just as the defined contribution model was catching on in the private sector thanks to changes
in the tax code, the federal government started looking at its own retirement costs. Ronald Reagan
campaigned in 1980 to downsize government but struggled to achieve this against opposing interests in
Congress and within his own administration. Reagan in his second year in office tapped the industrialist
J. Peter Grace to lead an exhaustive investigation of how the government spent money and what could
be done to cut costs. He tasked the members of the Grace Commission (formally called the President’s
Private Sector Survey on Cost Control) to “work like bloodhounds” and not “leave any stone unturned in
your quest to root out inefficiency”10
.
The section of the commission’s report on retirement costs today reads like the contemporary
assessments of state and local pension plans. The federal pension programs, it found, provide “benefit
formulas more liberal than can typically be found in the private sector; allow retirement, with reduced
benefits, at an earlier age than is typically found in the private sector; and provide full protection against
inflation”11
. It noted that the generous federal pension policy was introduced in the 1920s to
8
Standard and Poors
9
Tower Watson 2010
10
Reagan
11
President’s Private Sector Survey on Cost Control 280

7
compensate for lower wages paid by the government relative to the private sector, but when wages
were increased through the Federal Salary Reform Act of 1962 the pension benefits were not adjusted
downward to offset this shift. The Civil Service Retirement System (CSRS), the major sponsor for civilian
federal workers, was providing benefits approximately three times as large as the top private plans. Its
outlays had increased 18 percent annually in the last decade. The government continued to enhance
benefits without funding them.
The report recommended bringing the CSRS in line with the private sector by raising the
retirement age from 55 to 62, basing benefits off five rather than three years of final salary, lowering the
benefit multiplier, and reducing cost of living allowances COLAS’s below the rate of inflation. This would
have kept the current system in place at reduced costs for a present value savings of $30 billion through
2003. The Grace Commission helped highlight a national fiscal dilemma, even though pension reform
was a small segment of its work. In 1982, the unfunded liability for the CSRS was $575 billion (compared
to $170 billion held by the states), which represented an implicit liability of $5,000 per U.S. adult12
.
The report confirmed for the Reagan White House that something had to be done. But it
committed to holding off on acting until after the 1984 election. During the first term it was also
preoccupied with Social Security reform. Another national commission, led by Alan Greenspan,
produced a report with recommendations to resolve the Social Security trust fund’s short-term funding
crisis which was coming to a head in 1983. After the Greenspan Commission’s report, the administration
and Democrats in Congress quickly agreed to a set of Social Security compromises: they increased the
retirement age over decades, raised the OASDI tax, and delayed the year’s COLA by six months13
.
By curbing benefits and increasing contributions, the Social Security deal was a precursor to
pension reform. In light of the trust fund surpluses and debate over a “lockbox” in 2000, Reagan and
House Speaker Tip O’Neill’s pact probably went too far in expanding the short-term funding stream for
Social Security. But the entitlement program greased the skids for federal retirement overhaul in more
ways than one. New federal hires were put into Social Security, a move supported by the right and left
(though not the labor unions). Conservatives believed it was only fair to everyone else to have
government workers participating in the program, while liberals supported it as a way to help solidify
the trust fund. Strong bipartisan agreement on Social Security provided traction (and cost savings) for
federal civilian pension reform.
12
Yoder Jr.
13
Light

8
The reform model immediately became clear in Congress with the bills that were introduced in
the wake of the Social Security reform of 1983 and the Grace Commission report of 1984: a substantial
new defined contribution plan paired with a modest defined benefit plan for new hires (along with their
Social Security). This would more closely resemble the retirement compensation that workers in the
private sector could expect at the time.
The Reagan team was content to let Congress do the work. Many of its key personnel from the
first term (who had stared down the striking air traffic controllers) had left or were leaving, depriving the
administration of the resources to design and push its own plan. The White House’s sticking points were
that the overall normal cost (the percent of the employer’s payroll used to fund pension compensation)
had to be contained and the new DC plan’s money management couldn’t be politicized.
After voting on different versions of their own bills, the House and Senate held conference
meetings intermittently for six months between 1985 and 1986 to hammer out legislation acceptable to
both chambers and the White House. The final plan lowered employer normal costs from 25 to 22.5
percent of payroll by reducing the annuity formula, raising the retirement age slightly, capping the
COLA, and eliminating the crediting of unused sick leave. For the thrift (i.e. defined contribution)
component, it increased the contribution limit from five to ten percent of pay and introduced employer
matching for up to five percent of contributions. Investment funds in fixed-income and equities were
made available as options in addition to the existing fund in government securities.
The conference report passed both chambers by voice votes and was signed into law by
President Reagan a few weeks later in June of 1986. The end result for federal employees was a
trimmed down annuity benefit paired with a larger defined contribution benefit plus Social Security,
applying to all civilian employees hired after 1983. The new model was configured as the Federal
Employees Retirement System (FERS), with the Civil Service Retirement System restricted to the workers
hired before 1984.
How was Congress able to agree on a new retirement system for one million federal employees
within a relatively short period of time and pass it with the unanimity of a voice vote? There were four
major policy reasons: 1) recognition of the unsustainable course of federal pension outlays as
highlighted by the Grace Commission and other analyses 2) the preceding enrollment of new federal
employees in Social Security 3) the restricting of the major changes to recently hired and future workers
4) the design of an attractive DC plan at the centerpiece of the new pension compensation model.

9
On the political side, the 99th
Congress was the one in which the frustrating quest to tighten the
belt of government through the 1980s began to bear fruit. The Gramm-Rudman-Hollings spending caps
passed at the end of 1985 helped reduce the budget deficit by nearly one-third. The Tax Reform Act of
1986 eliminated tax expenditures once thought untouchable while lowering rates, consolidating
brackets, and eventually bringing millions off the tax rolls by doubling and indexing the personal
exemption and expanding the standard deduction and earned income tax credit. In short, the 99th
Congress was the ideal climate in which to confront the $570 billion in unfunded federal civilian pension
liabilities and replace the system that had produced this chilling figure. The consensus decision to start
putting federal employees in Social Security formed a basis for the normalization of their retirement
compensation toward the private-sector style (as well as a modest cost savings since federal workers in
the aggregate would contribute slightly more than they received from Social Security due to their
income levels). Additionally, both the Social Security change and then FERS were applied to workers
hired after 1983. Newly hired workers make for ineffective political opposition; future workers even
more so. No method but the “soft freeze” of the CSRS was considered by Congress and the political
firestorm of chopping benefits for career workers never came up.
Congress was also convinced that a thrift savings plan could do the job as the backbone of the
new system. According to the Senate staffer who was the principle bill writer, it saw the benefits of the
DC model: the encouragement of employee savings, the early vesting, the portable nature, and of
course relief from the threat of unfunded liabilities. The fact that it was already functioning and
increasingly popular in the private sector underpinned its confidence in this reform method.
The design of the thrift’s investment program consumed the most time of any of the bill’s
provisions. The Transactions of Society of Actuaries (the Social Security Administration’s research
journal) report of the process explains that “The question was how to invest in private securities without
giving the executive branch investment powers that might be used for political purposes.” Indeed, one
of the Reagan White House’s chief concerns was that the money management be completely
depoliticized14
. Federal retirement assets belonging to its workforce could not become a source for
favors traded by Washington power brokers. The pay-to-play scandals that have embarrassed state
pension funds in recent years show this to have been a prescient concern.
14
Schreitmueller

10
The decision to allow thrift investment options in private securities was a break from both Social
Security and the CSRS – both of those programs’ investment pools were restricted to Treasuries. But
stocks and corporate bonds held obvious upside, especially in a world (like today) where Treasuries
produced a nominal return after inflation. Private-sector investments offered higher potential returns
and injected massive large pools of capital into the economy. Because the thrift savings program was
based on individual accounts rather than the pay-as-you-go method and not liable for a defined benefit,
it had a risk tolerance that a program such as Social Security could not bear.
After considering an IRA-style plan that would have let participants pick from a myriad of
investment products on the market, Congress adopted the simple passive investment approach
introduced by Senator William Roth of Delaware. The Reagan administration initially opposed this
policy, but backed down after Congress won the public argument that the passive approach would be
insulated from politics and was superior to the IRA model because it offered much lower investment
costs. An independent thrift board was set up to administer the plan. Aside from determining the
maturities of the debt security investments and selecting which major index funds to replicate, it was
given no active investment responsibilities15
.
As with any massive bipartisan reform deal, there was discontent from some corners. Robert
Myers, the longtime Social Security Administration chief actuary and public retirement expert, for one,
was very hard in his assessment. He criticized the delay in the House and Senate working together, the
general absence of the White House in the process, and the failure of FERS to increase the retirement
age to the level in the private sector (it was Myers who set the retirement age of 65 for Social Security in
the 1940s; FERS only nudged the minimum retirement age up by one year to 56). He found that the
members of Congress who took the most interest in the bill were those who represented large numbers
of federal civilian employees, which kept the pressure on for generous benefits. Myers thought the bill’s
design would have worked best if it had been delegated to a blue ribbon committee, similar to the
Greenspan Social Security commission he staffed.
Others were disappointed that fewer eligible federal employees made the jump from the old
retirement system to the new one at the onset. Although 20 percent of workers in CSRS enrolled in the
new thrift program, most resisted joining FERS even when it made sense. The CBO calculated that 40
percent of workers would be better off making the switch, yet just two percent did during the first year
15
Schreitmueller

11
of eligibility. Richard Schreitmueller, in his follow-up to comments on his report for the Transactions of
Society of Actuaries journal, blamed this on a failure of communication. The government simply did not
market the new system to the degree necessary to publicize its features, which included a temporary
two-for-one employer match to the thrift account as a sweetener in the beginning of the program16
. This
has been a problem in the past with other government benefit programs such as Medicaid.
The 1986 law also left behind a substantial pension deficit attached to the CSRS. As of fiscal 2010
it measured $673.1 billion. This is an unfunded liability of the federal government that must be financed
through the remaining life of the system by the Treasury using cash on hand or new borrowing. Because
the CSRS was only closed to new entrants – not shut down – it will continue to pay benefits until the last
surviving beneficiary of the program dies, which the government estimates to be sometime around
207017
.
The FERS routinely incurs small deficits in funding when pension outlays differ from actuarial
projections. But unlike the CSRS, its defined benefit by law is prefunded, wherein the employer must
contribute the entire normal cost of a pension, including projected raises and COLAs in future years. The
federal DB pension is expensive – the employee pays just 0.8 percent of the 12.7 percent of pay that it
consumes – but prefunding requirement leaves no cause of concern about swelling unfunded liabilities
that states and localities are facing18
. The federal government pays the full tab for its pensions year in
and year out.
The most consequential legacy of federal pension reform has been the DC component. The
Thrift Savings Plan became a highly-respected and popular program that now offers more investing
choices for its members. In addition to the original funds in common stocks, corporate bonds, and
government debt, there are now index funds in small-cap and international stocks as well as a “Lifecycle
Funds” option that shifts investment in those five categories according to the member’s time horizon in
the program. Upon retirement, members can receive payment in lump-sum, payment based on a fixed
dollar or number of months until the account is exhausted, or purchase a lifetime annuity. The average
16
Schreitmueller
17
Losey
18
Roth

12
annual expense ratio is just 0.03 percent19
(mutual funds in 401(k) plans were recently found to have an
average cost of 1.27 percent20
).
The fact that the federal government could make its massive DC plan not just function but thrive
gave confidence to other large employers to embrace the model. It further legitimated and popularized
a form of retirement compensation that was still coming into its own when Congress began debating
civil service pension reform in the mid-1980s. The creation of the thrift plan also solidified the tax
climate in the eyes of the private sector. As Schreitmueller writes in his definitive account of the reform:
Adoption of a thrift plan for federal workers gives a sense of permanence to favorable tax
treatment of 401(k) arrangements, with government officials having a direct personal interest in
the rules governing such plans. Thus, the thrift plan represents another step along the road
toward reliance on a defined contribution plans and away from reliance on defined benefit
plans.
He similarly predicted that the federal overhaul would inspire other large public employers to
reform their retirement systems in the same direction21
. Unfortunately, this has been a much slower and
tougher road for the states than that bright vista of 1986 suggested.
Michigan: “Somebody Had to Go First”
Despite its notoriety in free-market circles as an organized labor political stronghold, Michigan in
1997 became the first state to implement comprehensive reform of a statewide retirement system from
the DB to DC model. Much of this accomplishment had to do with thorough policy work, savvy legislative
handling, and the persistence of popular three-term governor John Engler. On the last point, the bill’s
legislative sponsor, former Rep. Kim Rhead, said that “our governor wouldn’t take no for an answer” and
“just wouldn’t let it [pension reform] die”22
. Engler and state treasurer Doug Roberts, the top fiscal
player in Michigan’s executive branch, pushed through a pension overhaul of the Michigan State
Employees Retirement System that few in the beginning thought was possible.
19
Isaacs
20
Wasik
21
Much of this account of the Federal Employees’ Retirement System Act of 1986 is sourced in Schreitmueller’s
article, the most thorough narrative of the legislative and policy history
22
Rhead

13
Roberts’ argument for reform was basic: it was a better deal for the taxpayers. “What are the
benefits [of the current system] to taxpayers?” he kept asking. Taxpayers assumed all the financial risks
that came with promised returns, future pay raises, longevity of employees, and benefit increases. That
risk simply outweighed the benefits of providing such generous retirement compensation. A better idea,
he thought, would be to pay employees more for the work they were doing now rather than in
retirement compensation. This would reduce contingent risk associated with pension financing and
more immediately allocate compensation to highly-valued workers.
In scoping out how the retirement system affected state workers, Roberts discovered that only
about 50 percent were meeting vesting criteria to receive a pension. These were primarily long-serving
union members. When he walked around his own treasurer’s department, he would find cubicles in
which people had their retirement dates circled. It was clear that the status quo was good for these
workers23
. But in the long run the system was unsustainable in terms of offering benefits to a transient
workforce and paying for new retirement promises. “We looked at [MSERS] in 20-25 years and it was
going to be a train wreck,” Rhead said. “We’d be unable to repay these benefits that were
constitutionally guaranteed”24
.
The financial situation facing Michigan and the rest of the country was markedly different in
1996 from today. The stock market was soaring thanks to inflows from retail investors looking to
participate in the Internet boom. While this bolstered Michigan’s pension funded ratio (it was close to
100 percent) it also made the concept of individual retirement accounts more appealing. Workers in
401(k) and similar plans could take more risks in equities than public pension plans typically were
allowed by their government sponsors. Roberts also believed that this model was more appropriate for
the capital markets. Governments, he thought, shouldn’t be running businesses from inside as they
were inclined to do as sponsors of pension funds with massive equity stakes in public companies. As the
state’s money manager, he had particular credibility to make this argument.
As noted above, MSERS was close to fully funded at the time but projected to run out of money
for benefits decades down the road. That the system wasn’t facing a near-term pension financing crisis
enabled the reform bill to be a “soft freeze” – closing the DB plan for new workers only and thereby
23
Roberts
24
Rhead

14
grandfathering in current workers – which offered significant political insulation. Engler and his allies
were pushing “a long-term solution to a long-term problem,” as Roberts put it25
.
The governor in fact made a concerted effort not to have a large public debate about pension
reform that could enliven organized labor. Engler said he did not want to make it into a “cause célèbre”
that would draw the full-throated opposition of the left in Michigan. One component of the bill that
Engler cited as particularly important to tempering opposition was an “early out” provision that gave
early retirement to certain state workers26
. Approximately 5,100 participants accepted this deal – which
the governor’s office presented as a one-time offer paired with the defined contribution overhaul27
. This
proved helpful in deterring older, politically active union members from opposing the overall MSERS
reform package.
Engler had his staff spent most of 1996 doing the technical work associated with the bill
(including retaining an outside counsel to offer legal guidance) and mastering the arguments needed for
the legislative debate. The governor was willing to allow for a lengthy period of debate (which gave the
unions a “chance to complain” in the words of Rhead28
) while he aggressively lobbied Republican
legislators for votes29
. In this context Roberts described his boss as “the best politician I’ve ever been
around.” He was also the rare type of politician, Roberts added, who was not intimidated in the pursuit
of new ideas30
. This was around the same time Engler was being considered by Bob Dole as vice-
presidential candidate on the 1996 Republican ticket.
At the end of the 1996 lame duck Michigan legislative session, the pension reform bill passed by
a bare simple majority in the House and one more vote than that in the Senate, both chambers
controlled by the Republicans. All new employees hired in 1997 were automatically enrolled in MSERS
defined contribution accounts. Current workers were given the option of transferring their accumulated
pension benefits to a DC account. They would get a 401(k) in the place of a pension annuity of equal
actuarial value. It was akin to a buyout. About five percent of state employees took this deal31
.
25
Roberts
26
Engler
27
Papke
28
Rhead
29
Engler
30
Roberts
31
Papke

15
Did it work for Michigan? A recent analysis by Richard C. Dreyfuss for the Mackinac Center
estimated that the state has saved $167 million in pension normal costs since the bill took effect in
1997. Furthermore, he projects that the unfunded liability of the old system -- $4.1 billion was the latest
reported figure when he published this paper in 2011 – would be between $2.3 and $4.3 billion higher32
.
By Roberts’ criterion of serving the taxpayers, it has undoubtedly been a better deal. Pension costs for
the new system have become affordable, controllable, and transparent. The contingent liability has
been cut off. The pension deficit associated with Michigan state employees (though not teachers, who
are in a separate system) – unlike most other states – is on a path to extinction.
Employees in the new system get a minimum employer contribution of 4 percent of salary plus
up to another 3 percent in matching. Their own contribution limit extends up to $17,000. Vesting is
relatively quick – 50 percent at two years of service, 75 percent at three years, and full ownership of the
account at four years. This is in contrast to the stringent vesting period in MSERS which began at age 55
with 30 years of service or 60 with ten years. Unlike the old system, the workers don’t have to worry
about staying on the job long enough to collect retirement compensation.
The Michigan bill is, as Dreyfuss concludes, “a model for reforming other government pension
systems”33
. It achieved Engler’s goals of replacing the old way with a “flexible, funded system” in which
“retirement savings could start on day one”34
. It was unveiled in a methodical manner so as not to
provoke gratuitous confrontation with labor groups who saw a sacred benefit – guaranteed pensions –
undone in a state at the epicenter of the their movement. Moving exclusively to the defined
contribution model was an untested idea at the time. As Roberts put it, “Somebody had to go first”35
.
Michigan’s blueprint would become increasingly popular in free-market policymaking circles, but it took
a complete reversal of market performance to compel another state to follow it.
Utah: A Market Crash and “Transition Costs”
In 2009, Dan Liljenquist was a newly-elected 35-year old Utah senator who had been assigned to
chair the retirement committee, traditionally a backwater in the state’s legislative affairs. But Liljenquist,
32
Dreyfuss
33
Ibid
34
Engler
35
Roberts

16
a former Bain consultant, knew he was taking the assignment at a very important time. The financial
crisis that had peaked the previous year had decimated investor portfolios – from individuals all the way
up to large pools of managed money like public pension funds. Utah’s pension assets lost 22 percent of
their value ($4.8 billion) in 2008. A system that was over 100 percent actuarially funded was now on
track to be just 70 percent funded in 2013 when those losses were smoothed over the coming five
years. Utah wasn’t going to run out of money for pensions anytime soon, but its defined benefit was on
the way to becoming a major fiscal drag on the state.
Liljenquist dug into the problem with all the analytical skills of a first-rate consultant. The first
thing he did was collect data. He was told by system administrators that plan costs were up by 2.5
percent that year, but they had no projection for how much they would have to go up in the future to
absorb the losses from the 2008 crash. Then he went to the system’s actuary to get those projections
and have the actuary stress test different investment returns – 6%, 7%, 7.5%, and 8.5% -- over the next
40 years. Those scenarios all had the pension system hollowed out of funding between 2040 and 2050 if
current contribution rates were kept in place36
.
The point Liljenquist vividly proved was that the status quo was unacceptable. To plug the hole
from 2008, the state would have to commit 10 percent of its general fund ($400 million) for the next 25
years toward increased pension contributions. To equate that in human rather than budgetary terms, he
showed that this was equivalent to the salaries of 8,000 teachers. Public services would have to be
sacrificed in order to maintain Utah’s defined benefit pension system. Liljenquist, like his predecessors in
Michigan, was trying to show that this was a bad deal for his state’s taxpayers.
Liljenquist had two principles that guided his search for the appropriate pension reform plan: 1)
meet every penny of current commitments and 2) cap the liability. He wanted to make each retiree
whole while closing down an unsustainable system. These two ideas came with four action items. To
deal with what it owed, the state would have to continue to make the Actuarial Required Contribution
(ARC) recommended by the actuary. In addition, the practice of double-dipping would have to be ended
because post-retirement reemployment was driving up pension costs. Capping the liability meant that
Utah would need a plan to pay off the pension deficit incurred from the 2008 market crash as rapidly as
possible. Finally, the legislature would have to design a new system with lower and predictable costs.
36
Liljenquist (report)

17
Liljenquist was not married to any particular replacement plan. His proposition was, “Here’s
what we can afford” – about 10 percent of payroll – “let’s see what it buys.” He preferred a straight
401(k)-style replacement plan where the government/taxpayers were off the hook from accumulating
any further debt. If that wasn’t possible, his second choice was a cash balance plan funded by a private
annuity. This would mean that workers were promised a certain level of funds in their retirement
account but it would be low enough that the state could hire an insurance firm to cover it. A cash
balance plan was like a 401(k) with a floor beneath it37
.
The system Utah got was more like what the federal government has – a “hybrid plan” wherein
the individual account is anchored by a scaled-down defined benefit. Employees hired into the new
“Tier II” system after June 2011 got the option of a straight 401(k) account or one with the DB
component – a “hybrid of a hybrid.” Normal cost was capped by statute at 10 percent of salary. This
meant that workers who chose the DB would get no more than 10 percent of their salary from the state
to fund it – any deficit would have to be made up from their own payroll contributions (to put that in
perspective, the normal cost for employees in the old system was 23 percent). While the DB formula
was a relatively modest 1.5 percent multiplier for each year of service (2.5 percent for police and fire),
this cap on employer funding for it provided an immediate incentive for entrants to select the 401(k)
plan only rather than adding the expensive DB. This was the intent of the new system’s design: direct
workers into the DC accounts as their single source of retirement compensation from the state.
Why did Liljenquist and his committee design a bill that retained a DB element when they were
trying to replace it with the opposite model? The answer has to do with accounting rules. By maintaining
the old system in a stripped down form, the state could make lower payments toward its unfunded
pension debt because it could base those payments as a percentage of its payroll rather than a level
dollar amount. As a percentage of payroll, the payment amounts (the amortized, as opposed to normal,
component of the ARC), could be graduated rather than fixed since payroll would grow gradually. It is
akin to making a larger payment on your mortgage as your income grows.
Keeping the old system open did the trick of maintaining a large payroll base for the purposes of
associating pension amortization payments with new state workers. A chunk of the employer
contribution for each new hire would continue to go toward paying down Utah’s unfunded liability. If
this hadn’t been the case and the system was officially closed, the state would have no new employees
37
Liljenquist (interview)

18
to associate these contributions with and would need to switch to the level dollar method, i.e. a higher
contribution rate. Keeping the old system thus spared the state from a hefty fiscal note, explained
Liljenquist38
.
Liljenquist, for all his mastery of the details, did not know when he was writing pension reform
legislation that this analysis was incorrect – rules for percentage of payroll vs. level dollar amortization
only mattered for reporting purposes and had no bearing on actual cash outflows. As the threatening
fiscal note implied, the Utah legislature’s professional staff believed the same line of thinking, as did the
actuary on contract with the state. In fact, the entire public pension profession – an elite of
administrators, actuaries, accountants, lawyers, and others defined by their expertise – accepted the
notion that closing down a pension system really did mean having to pay off the deficit associated with
it more quickly. This was despite the fact that from both a compliance and economic perspective, it
made no sense.
The “transition costs” myth – the falsehood that closing down a DB pension system came with
the fiscal burden of increased annual amortization payments – was not comprehensively debunked in
public until Robert Costrell’s paper “’GASB Won’t Let Me’ – A False Objection to Public Pension Reform”
was released in May 2012. Costrell, an economist at the University of Arkansas and fellow at the George
W. Bush Institute at Southern Methodist University, explained in detail that guidelines for amortizing
pension deficits – the crux of the debate – apply to financial reporting rather than funding. A public
pension sponsor is required by the Government Accounting Standards Board, the independent oversight
body, to amortize unfunded liabilities on the level dollar method if a plan is closed. But GASB has no
authority over actual funding. This explains how many states routinely make pension contributions well
short of the Annual Required Contribution: the ARC is a reporting requirement, not a funding
requirement. Pension sponsors determine for themselves how they finance their obligations, including
when it comes to past debts, and do so by statute or through the pension boards39
.
The transition costs misconception also made no sense in terms of cash flow. The amortization
requirement from GASB implies from a funding perspective that each new employee enrolled in a
retirement system contributes toward the unfunded pension debt. That would be true if this were Social
Security, where payroll contributions from participants fund current benefits in the pay-as-you-go
model. But unlike Social Security, with state and local pensions the employer and employee
38
3939
Costrell

19
contributions come from the same source – the government sponsor – and therefore payroll makes no
difference when it comes to making pension payments. Whether or not the contributions are associated
with employees and tacked on to their pay makes no difference when it comes to cash available to make
them. Associating pension payments with payroll (and therefore accelerating them for a closed system)
is nothing more than a function of the accounting process of amortization.
Despite the clarity that can be found in the reporting vs. funding contrast, many states
considering reform have been operating under the same impression conveyed to Utah that closing a
defined benefit system means higher payments. Costrell in his paper identifies California, Minnesota,
Kentucky and Virginia as states where the “transition costs” claim was levied as a successful argument
against pension reform40
(Nevada is another). The perverse outcome is the idea that closing a sputtering
pension system is actually more expensive than keeping it open. Keith Brainard, research director of the
National Conference of State Legislatures, said, “Generally speaking the more underfunded a plan is, the
more expensive it is to try and switch.” A headline in the Orange County Register surmised that it was
“cheaper to do nothing” when pension reform was floated in California in 201241
. Another in Virginia
cited by Costrell blared, “Retirement System Too Far in Debt for Reform.” Imagine such as assessment of
Social Security. Only in the insular world of public pensions – a world finally getting scrutiny in recent
years because of its dysfunction – would such logic be taken at face value.
Another irony of “transition costs” is that GASB as part of its new rules beginning in 2014 plans
to discontinue the ARC and with it the required reporting of amortized costs. Costrell notes that the
organization in its revision cited the fact that those involved have misread GASB as setting “de facto
contribution policy standards” when it has no such intent42
. GASB, it seems, will have the last word in
emphasizing that reporting is entirely different from funding. Looking back on it, the misuse of the ARC
was of a piece with the will of the public pension elite – the lawmakers, trustees, administrators,
consultants, and labor leaders – to maintain the status quo, which meant disregarding the ARC when it
came to keeping up with funding but reimagining its constraints when the topic is closing a defined
benefit plan.
In the end, despite the roadblock thrown up by the problem described above, Utah got what it
wanted with a new retirement system. Liljenquist, who wryly notes in his presentations on the
40
Costrell
41
Sforza
42
Costrell

20
experience that “future employees are not an effective lobbying force,”43
credits the “soft-freeze”
approach (also taken by the federal government and Michigan) with putting the odds in his favor. So did
the provision to delay opening the new system by 12 months until June 2011, which gave the legislature
the chance to come back in session the next year and clean up the bill (they made benefits more flexible
and added a liquidation option if the plan became “critically unstable”). The ban on double-dipping, by
his account, actually produced the biggest uproar because it caused substantial income losses for some
state retirees, like the man who told Liljenquist the new law would cost him $400,000 a year.
In the end, the retirement bill passed the House 46-26 on February 26, 2010 and the Senate 18-
8 on March 1, clear victories that nonetheless indicated a political battle. “I was nervous because I
thought that I might cost some people their careers,” Liljenquist reflected. “But it really helped them.
Some legislators said it was the most important vote they’ve ever taken.” He reports that exit polls of
Utah statehouse elections later in the year showed that pension reform was the top issue that
persuaded undecided voters to break for the GOP44
.
Rhode Island: the Progressive Case for Pension Reform
On that same Election Day in November 2010, Gina Raimondo was elected treasurer of Rhode
Island by an overwhelming margin. Like Liljenquist, she came fresh into politics from the elite business
world, having worked at a private equity firm on Wall Street and then founding the first Rhode Island-
based venture capital fund. Raised in a blue collar family in Smithfield, Rhode Island, her education path
was Harvard, Oxford, and Yale Law School. In short, her background had the perfect combination of
Rhode Island real deal and outsider big deal that would make people listen. Overhauling pensions in the
state would require just that.
For a Democrat, Raimondo was taking up the pension issue at exactly the right time. The funding
crisis that was accelerated by the market crash of 2008 raised the specter of being more than simply a
bad deal for taxpayers: it was one that could deprive them of essential public services. On her tour
through local chambers of commerce, rotary clubs, union meetings, and in media interviews, Raimondo
frequently mentioned the plight of Vallejo, California, a hardscrabble city in bankruptcy that left its
pension system intact while cutting to the bone police, fire, and other basic services. Her essential
43
Liljenquist (presentation)
44

21
warning, punctuated after methodical but clear explanations of the state’s finances, was that this could
happen in Rhode Island if the state didn’t act. “That’s not the future we want for our children,” she told
one of the assemblies45
. She even told Bloomberg that reading in 2009 about potential cuts in libraries
and bus service spurred her to run for office. “I literally put the paper down and said, ‘I have to do this, I
have to run,’” she said46
.
In public office Raimondo was doing something new: making the progressive case for pension
reform. In a state with more government than private sector workers, she could hardly do what
reformers in Michigan and Utah had done in depicting it as a better deal for the taxpayer. As a Democrat
in a public-sector oriented economy, Raimondo had to make the case to beneficiaries that downsizing
the system was a better deal for them. The heightened sense of crisis in 2011 enabled that argument.
When Central Falls, Rhode Island tipped into bankruptcy in August, she had the Vallejo example right in
front of her audiences.
Raimondo combined the listening skills of an effective politician that Dan Liljenquist had in Utah
(where he replied to 3,500 emails on the issue47
) with the authority over the financial picture that Doug
Roberts demonstrated in the same role as Michigan’s treasurer. Her stops through Rhode Island
communities, usually on weekday evenings so people could attend after work, were a mix of empathy
and stone-cold analysis. “I feel your anger,” Raimondo was quoted in the New York Times as telling a
crowd. “In many ways, I’m angry myself. Many of the shenanigans that went on in past years were just
wrong”48
.
In a general sense she was referring to a system that made promises without doing what was
necessary to keep them, the failure of the state’s fiduciary role. But in a specific sense she was
implicating the practice that led to this breakdown: the understating of Rhode Island’s pension liabilities
that resulted in chronic underfunding. By using a high interest rate to calculate its future obligations in
current dollars, the retirement system reported far less of what in was on the hook for than what an
insurer would say. But this wasn’t particular to Rhode Island: virtually all government pension sponsors
engage in this practice because it is sanctioned by GASB. The regulator’s guideline says that the plans
may use projected investment return as the discount rate to value their liabilities. Most project a return
45
Walsh
46
McDonald
47
48
Walsh

22
of approximately 8 percent, which becomes their discount rather than the yield from Treasury bonds
that market valuation would use. The idea behind market valuation is that a liability should be valued
using an interest rate that reflects its risk characteristics. Since public pension obligations are conceived
of as risk-free due to their protection in the contracts clauses of the federal and state constitutions, the
interest rate from a risk-free security like a Treasury bond is the best proxy to discount future cash
outlays into present liabilities. Such valuation assures that the state will put away enough money to pay
for the pensions when they come due.
Raimondo recognized that this was more than a debate about accounting principles: it was
about the principle of telling taxpayers the truth on government debt. In most areas of public spending
it was fairly clear to see the financial obligations a state or city had – for schools, roads, or hospitals –
but with pensions the faulty valuation made it opaque. The state was reporting $7.3 billion in unfunded
liabilities, but analysis from the Mercatus Center using market valuation said the real figure was $12
billion49
. Raimondo knew that to get political traction the state needed to at least acknowledge what
was told as truth in financial economics.
This was the motivation behind the treasurer’s “Truth in Numbers” report. It was a 17-page fess-
up that explained the ways costs exploded over time, the ignorance of actuarial practices by the
retirement system and legislature, alternative valuations, and how taxpayers would be soaked if nothing
was done. The most pointed section was the projection of normal costs using different discount rate
scenarios. Raimondo’s office had its actuary show the percentage of pay that would need to go to
pensions according to the discount rates of 8.25 (the state’s rate of the time), 7.5, 6.2, and 4.4 percent.
The low end of 4.4 percent didn’t match the risk-free Treasury interest rate (3.5 percent at the time),
but it was relatively close. The report showed potential normal costs for state employees of 9.3, 11.4,
14.4, and 22.2 percent, respectively. This range indicated that according to a discount rate closer to
market valuation the state needed to pay nearly two-and-a-half times more as a percentage of regular
state workers’ salary to fund its pension system. The situation for teachers was even worse.
“Truth in Numbers” also made the funding burden clear in terms of how it hit taxpayers (just as
Dan Liljenquist had done so well in Utah). Expanding circles showed how the chunk of each taxpayer
dollar that went to pensions had grown from three cents in 2003 to 16 cents for fiscal 2013 and would
hit 20 cents in 2018. Back in percentage of salary terms, the taxpayer (or employer) total component
49
Norcross & VanMetre

23
was set to go from 23 to 35 percent over the next two years while workers’ contribution was fixed
around 9 percent. Three-quarters of this amount would go to amortized (past) costs rather than normal
(current) costs. The unfunded liability, the report made clear, was the “lion’s share of the problem”50
.
After spending the first half of 2011 on raising awareness about the pension crisis, Raimondo
focused on the solution, which she outlined at the end of the “Truth in Numbers” report when it came
out in June. In recommending savings, it insisted that changes had to apply not only to new hires but
current employees and retirees as well. In that regard Rhode Island would be acting more aggressively
than Utah or Michigan had, but unlike those states it was already in a funding crisis (the reported
unfunded liability growing nearly 60 percent in two years) and needed to reap big savings right away.
The biggest was in suspending the cost of living allowance (COLA), which would cut $1 billion off the
state’s unfunded liability right away. Raising the retirement age by two years, to 67, would also produce
a large immediate savings. Raimondo’s report called for a defined contribution plan, but only as an
addition to the current model in the form of a hybrid system. The reason, as in Utah, was “transition
costs,” which the treasurer’s office indicated would be $882 million in new contributions over the
coming five years for closing the DB system51
. Once again, it was apparently more costly to end a
taxpayer guarantee of pension benefits than to sustain it. It is worth noting that Rhode Island used the
same actuary as Utah (Gabriel, Roeder, Smith) to crunch the numbers.
By fall of 2011, Raimondo and the forces for pension reform got an unambiguous win when their
plan cleared a special session of the legislature in November with overwhelming support (57-15 in the
House, 35-2 in the Senate). The new law suspended COLAs until the system was 80 percent funded (a
level not in sight), shifted more than half of employee contributions to individual retirement accounts,
raised the retirement age, and reduced the pension multiplier for calculating benefits.
In total, the changes dropped the reported unfunded liability from over $7 billion to around $3
billion and improved the funded ratio from 48 to 60 percent. Thanks to changes to current obligations
that delivered immediate savings, the Rhode Island Retirement Security Act brought the state off the
precipice of committing itself to pouring huge new amounts of money into the pension system. Under
the new law, it would pay $177 million for pensions in fiscal 2013 rather than the $305 million that had
been previously projected under the old system. For a state with just over $3 billion in general revenue,
the first year’s annual savings of $128 million was a significant amount that would help it avoid the
50
Raimondo
51
Ibid

24
desperate cuts to public services that Raimondo and her progressive allies feared. This fact cut to the
heart of the bipartisan appeal of comprehensive pension reform toward the defined contribution model.
New Jersey and Other States: False Starts and Incremental Changes
Pension reform has become something of a political buzzword over the last few years, and the
news coverage might suggest that it has happened in a flurry across the country. That impression is far
from true, as the action in other states has been modest rather than thorough.
One of the most common prescriptions has been increases in employee contributions. Other
popular changes include raising the retirement age or service requirement, tightening the benefit
formula, cutting or suspending the COLA, and closing loopholes that enabled spiking or other abuses. All
of these tactics can be helpful in enhancing the funding position and saving money (particularly the
COLA suspension, enacted by Colorado, Minnesota, South Dakota, Florida, and New Jersey, in addition
to Rhode Island). However, they are all improvements to paper over a flawed system – the defined
benefit model – that postpone the day of reckoning when these states will need to figure out how to
fund their plans in the long-term.
The National Conference of State Legislatures counted 43 states making “major changes” to
their pension plans from 2009-201152
. One state that garnered a large share of the headlines was New
Jersey, mainly thanks to Gov. Chris Christie. While the Republican star was effective in helping make
public pensions a national issue (in part owing to his town hall harangues that went viral on YouTube),
the content of his reform program was limited. It included most of the tweaks listed above and reduced
the state’s unfunded liability by 30 percent. But the reforms only improved the funded ratio from 56.4
to 62.2 percent, and it is not projected to hit 80 percent (the criterion of plan health) until 2041. The
State Budget Crisis Task Force, a project led by former Federal Reserve Chairman Paul Volcker and
former New York Lt. Gov. Richard Ravitch, reported last October that in the coming years New Jersey
will be “forced to make extraordinarily difficult choices about spending priorities and taxes.” In the next
five years the state’s scheduled pension contributions are scheduled to go from $1.03 billion to $5.5
billion, an increase equivalent to 40 percent of New Jersey’s annual education spending53
.
52
Snell
53
Ravitch & Volcker

25
Christie talked to voters about how delaying pension reform would inflict painful choices, but he
did not propose a plan to stave those off. Closing the state’s defined benefit system may have been a
nonstarter with a Democrat-controlled legislature, but the governor did not even make a case for it. As
the Wall Street Journal editorial page put it, “By dropping the issue without a fight, Mr. Christie has
given away too much even before the unions get to the table”54
. In summary, the New Jersey pension
saga of 2010-11, like many that reverberated across the country, had a lot of sizzle without substance.
There were short-term savings from changes on the margins but not the kind of overhaul adopted by
Michigan, Utah, and Rhode Island that actually contain the pension liabilities. By going for the
incremental, most states pushed the heavy lifting of pension reform down the road.
Lessons Learned
The five examples of implementing defined contribution pension reform were all driven by their
own circumstances. In the private sector, stricter accounting rules and new tax code provisions for
individual retirement accounts made the switch a relatively easy call for most firms, particularly in the
context of an increasingly more fluid labor market. For the federal government, the attention brought
by Ronald Reagan’s “bloodhounds” on the Grace Commission to the outsize benefits of the Civil Service
Retirement System prompted Congress to try to instill retirement compensation parity for public
workers.
At the state level, the context of each of the three states reviewed was different. Michigan, the
early adopter, took on pension reform smack in the middle of the bull market when the 401(k) account
had particular appeal to public workers who wished to have more of their savings in equities. Utah, on
the other hand, acted in the aftermath of a market crash and financial crisis that decimated retirement
portfolios by approximately 25 percent, including the state’s own. Rhode Island was also reeling from
the blowout, but unlike Utah it took the hit from a weak funding position and would require large
injections of cash just to remain solvent. What we see here are various degrees of financial strain that
each pension sponsor faced and that correlated positively with how long they waited to act. This is the
first and most obvious lesson to be learned from the pension reform experience: waiting just costs more
money. It can also mean more leftover entanglements stemming from austerity decisions, as the legal
challenge to Rhode Island’s 2011 law from the state’s public unions shows.
54
Wall Street Journal

26
In each case of legislative reform, in Congress and in the states, the process went relatively
quickly. Lawmakers understood the need to act within a time frame that maximized appreciation and
understanding of the pension problem both within the legislative chambers and among the public. The
preparation and groundwork done in advance enabled this swiftness. Doug Roberts and Dan Liljenquist,
for instance, spent months before the legislative session preparing their case. Gina Raimondo assigned
her treasurer’s department to this upon taking office. She then spent the better part of a year going
around the state with her presentation, as Liljenquist had done in Utah in meeting after meeting with
every relevant group. Because it must happen in a compressed and cluttered legislative calendar, state-
based pension reform favors the advocates who are fast on their feet and can understand how to latch
onto the political tailwinds.
In most fiscal policy debates the side that owns the numbers tends to prevail, and it was no
different in the battles that were had over pension reform. In each case there was some authority that
backed up the specifics, if not the theme, of the impetus for moving against defined benefit systems. For
the private sector, that was the ERISA law and the IRS. Their combination of actuarial methods and tax
changes revealed the scope of the DB pension burden and brought the individual retirement account
into focus. For the federal government it was the Grace Commission, whose finding that the federal
government was providing retirement compensation roughly three times as large as private plans
shocked and galvanized Congress into acting. In the states, it was the treasurers in the cases of Michigan
and Rhode Island and lead legislator Dan Liljenquist in the case of Utah. They spent months analyzing
the pension dilemma and putting this analysis into public presentations that were of superior quality.
Liljenquist was so convinced that this set the foundation for success that in his presentations on the
topic his first lesson is to “demand the data” – an instruction to would-be reformers to acquire the
detailed, comprehensive scenarios and modeling from pension actuaries that is needed to tell the whole
story of any retirement system. It’s clear that the side that has the authority over the data begins the
debate with a clear edge.
In addition to mastering the upfront debates over the DB public pension problem, the legislative
success stories also had credible replacement plans. Today, the Federal Thrift Savings Plan is considered
the gold standard for public defined contribution benefits. Michigan followed this example by offering
appealing but prudent investment choices for its 401(k) enrollees. Utah and Rhode Island introduced
hybrid plans with generous employer contribution levels. In each case reform designers were mindful of

27
expectations of fringe benefits and competent delivery that workers were used to in public
employment.
In each of the political battles there emerged some kind of powerful consensus for pension
reform. In the business community, it was embracing the modern form of retirement compensation that
marked a certain break from the old economy. In Congress, it was zeal to cut federal spending that
culminated with automatic federal spending cuts imposed by the Gramm-Rudman-Hollings Act in 1985.
In Michigan and Utah, it was a belief among the Republicans in control that the pension status quo was
no longer sustainable. Finally, in Rhode Island it was a revenue crisis that threatened public services, the
distinct possibility that the small state could meet the same fate that had hit cash-strapped
municipalities around the country. The votes weren’t always of the same type: in Michigan and Utah
they were party-line while in Congress and in Rhode Island they were bipartisan. The consensus in each
case depended upon a sense of urgency that propelled pension reform through institutions often
gripped by paralysis and resistance to new ideas.
It took strong personalities to foster that sense of urgency in the states, another common
theme. The principals in each – Gov. John Engler in Michigan, Sen. Dan Liljenquist in Utah, Treasurer
Gina Raimondo in Rhode Island – had strong willpower, compelling points of view that amounted to a
vision, and indefatigable energy for the cause. They had sufficient conviction of the importance of what
they were doing to accept bruising, high-profile political fights that would have to be hard-won if they
were successful. We know from their experience that comprehensive pension reform is not for cautious
political leaders.
The states reviewed above are not just important for the lessons learned – including the hazards
connected to GASB from the “transition costs” issue and high discount rates – but also for the simple
fact that they solved the threshold dilemma for the other states. Most governors and legislators are
reluctant to be the first mover on a large-scale reform, preferring to have precedent to take into
account. Michigan, Utah, and Rhode Island, by acting rapidly in three different scenarios, established
that precedent for others to follow. As public pension funding positions worsen, the appeal of the
defined contribution model will become more clear and urgent. Despite the tremendous financial
distress many states will face by the time they realize the need to act, they will have the advantage of
the reform roadmap carved by the three that came before them.

28
Works Cited
Bai, Matt. "How Chris Christie Did His Homework." New York Times 24 February 2011.
Biggs, Andrew. "Public Pensions: How Well Funded Are They, Really?" July 2012. State Budget Solutions.
28 August 2012.
Butrica, Barbara, et al. "The Disappearing Defined Benefit Pension and Its Potential Impact on the
Retirement Incomes of Baby Boomers." Social Security Bulletin (2009).
Costrell, Robert. ""GASB Won't Let Me" -- False Objection to Pension Reform." Laura and John Arnold
Foundation Policy Perspective May 2012.
Dreyfuss, Richard C. Estimated Savings from Michigan's 1997 State Employees Pension Plan Reform.
Midland, MI: Mackinac Center, 2011.
Engler, John. Interview. Rich Danker. 25 October 2012. In person.
Isaacs, Katelin. Federal Employees' Retirement System: Benefits and Financing. Report for Congress.
Washington, D.C.: Congressional Research Service, 2012.
Jr., Edwin Yoder. "Government Living Beyond its Means." Washington Post 27 July 1986.
Keller, Bill. "Another Stab at Pension Reform." New York Times 15 July 1984.
Light, Paul. "The Crisis Last Time: Social Security Reform ." New York Times 5 March 2005.
Liljenquist, Dan. Interview. Rich Danker. 27 November 2012.
—. The Pension Battlefield. Presentation to American Legislative Exchange Council. Washington, 2012.
Document.
Losey, Stephen. "Fed Pensions Underfunded by $673 Billion." Federal Times 16 October 2011.
McDonald, Michael. "Gina Raimondo Math Convinces Rhode Island of America’s Prospects With Debt."
Bloomberg Business Week 10 January 2012.
Norcross, Eileen and Benjamin VanMetre. Rhode Island's Local Pension Debts. Arlington, VA: Mercatus
Center, 2011.
Papke, Leslie. Which Public Employees Switch to a Defined Contribution Plan? East Lansing, MI: Michigan
State University, 2011.
Raimondo, Gina. Truth in Numbers. Providence, RI: Ofice of the Rhode Island General Treasurer, 2011.
Ravitch, Richard and Paul Volcker. Report on New Jersey. New York: State Budget Crisis Task Force, 2012.

29
Reagan, Ronald. "Remarks at a White House Luncheon With the Chairman and Executive Committee of
the Private Sector Survey on Cost Control." 10 March 1982. Ronald Reagan Presidential Library.
http://www.reagan.utexas.edu/archives/speeches/1982/31082d.htm. 20 September 2012.
Rhead, Kim. Interview. Rich Danker. 16 October 2012. Phone.
Roberts, Doug. Interview. Rich Danker. 1 November 2012. Phone.
Roth, Allan. CBS MoneyWatch . 16 May 2010. 1 October 2012.
Schreuitmueller, Richard. "The Federal Employees' Retirement System Act of 1986." Transactions of
Society of Actuaries (1988).
Sforza, Teri. "Pension reform: Cheaper to do nothing?" Orange County Register 3 January 2012.
Snell, Ron. "STATE PENSION REFORM, 2009-2011." March 2012. National Conference of State
Legislatures . 12 February 2013.
Standard and Poors. "S&P/Case-Shiller Home Price Indices." June 2012. S&P/Dow Jones Indices.
http://www.standardandpoors.com/indices/sp-case-shiller-home-price-
indices/en/us/?indexId=spusa-cashpidff--p-us----. 17 September 2012.
Tower Watson. "Prevalence of Retirement Plans by Type." June 2010. 17 September 2012.
Towers Watson. "Workers' Retirement Plan Preferences and Expectations." March 2005. Watson Wyatt
Insider. http://www.watsonwyatt.com/us/pubs/insider/showarticle.asp?ArticleID=14461. 17
September 2012.
Van DerHei, Jack. "The Impact of the Recent Financial Crisis on." February 2009. Employee Benefit
Research Institute. http://www.ebri.org/pdf/briefspdf/EBRI_IB_2-2009_Crisis-Impct.pdf. 17
September 2012.
Waitrowski, William. Changing Landscape of Employment-based Retirement Benefits. Washington, D.C.:
Bureau of Labor Statistics, 2011. Web.
Wall Street Journal. "The Christie Example." Wall Street Journal 20 September 2010.
Walsh, Mary. "The Little State with a Big Mess." New York Times 22 October 2011.
Wasik, John. "Paring 401(k) Expense." New York Times 11 September 2012.

Implementing-Defined-Contribution-Pension-Reform

More Related Content

What's hot (18)

Viewers also liked (15)

Similar to Implementing-Defined-Contribution-Pension-Reform (20)

Implementing-Defined-Contribution-Pension-Reform