Image Image Image Image Image Image Image Image

Center for Labor Research and Education


Scroll to top


Potential Impact of Governor Brown’s Pension Reform Plan on Low Wage Workers

On October 27, 2011, Governor Brown released his “Twelve Point Pension Reform Plan” designed to reduce pension and retiree health costs for state and local government. The plan includes several sensible proposals, such as the elimination of pension spiking and capping defined-benefit (DB) pension benefits at a reasonable level. However, two key elements of the plan may impose a disproportionately large burden on low-wage workers. One is a “Hybrid” pension for new employees that will combine a reduced DB pension benefit with a defined-contribution (DC), 401(k) type plan in which employees bear all the risk. The other is the proposal to increase the age at which employees can receive full retirement benefits to 67 across the board for new, non-safety employees. Low-wage workers are ill-equipped to bear the risks and increased costs of a 401(k) style plan, need to replace a greater portion of earnings than do middle-or high-wage workers in order to meet basic expenses in retirement, and begin their working careers much earlier, and experience shorter life expectancy, than professional workers. Because of these factors, the two proposals could have significantly greater impacts on workers at the bottom of the wage spectrum.

“Hybrid” Risk-Sharing Pension Plan for New Employees

The Governor’s hybrid pension proposal has three key dimensions. First, retirement benefits for new employees will be composed of a reduced DB benefit and a 401(k)-type DC plan. Second, benefits would be targeted to replace, in combination with Social Security, 75 percent of pre-retirement earnings after 35 years of service for non-safety employees and 30 years for safety employees. Third, DB, DC, and Social Security would each provide roughly one-third of the total benefit, i.e., 25 percent of pre-retirement earnings. This proposal is intended to distribute pension risks and costs between employers and employees while attempting to provide retirement income that allows workers to maintain their standard of living after a lifetime of service. In reality, the proposals would have a disproportionate impact on the retirement security of lower wage workers and those in occupations requiring higher levels of physical activity.

401(k) type plans entail a host of risks and costs that lower-wage workers are ill-equipped to absorb.

401(k) plans were originally instituted as a tax shelter for corporate executives, and were never meant to be a vehicle through which to save and invest for basic living expenses. These plans not only shift investment and market risks from employers to employees; they entail fundamentally greater risks and inefficiencies overall compared to DB pensions. These risks and inefficiencies can be mitigated, though only partially, by higher up-front contributions during the accumulation phase–about 72 percent more compared to a DB pension. This is a high price tag for public employers in the state. Even higher contributions would be required on the part of employees if the possibility of an ill-timed market downturn is taken into account. Unfortunately, low-wage workers have extremely limited capacity to absorb such costs. As retirees, they will experience disproportionate hardship if the market does not perform as expected because their target retirement income provides only for basic expenses, not luxuries that can be easily cut back.

  • Individual 401(k) accounts typically garner lower investment returns than do DB pensions, which can diversify their investments over a wider array of vehicles and invest over a very long time period. Retirement benefits that rely heavily on 401(k)s also require funding to last several years past average life expectancy, while DB plans pool longevity risk and thus only need to be funded for the group’s average life expectancy. For men born in 1970, average life expectancy in the US after age 65 is about 19 years, or age 84.1 Private financial planners typically advise such individuals to plan savings to last an extra 6 years, to age 90, in order to eliminate most of the risk of running out of money when they are most vulnerable.
  • Because of these and other factors, it costs 46 percent less to provide comparable benefits through a DB pension than through a 401(k).2 Conversely, it costs 72 percent more to provide the same retirement income through a 401(k) as through to a DB pension. Indeed, the University of California recently chose to maintain its DB pension system rather than convert to a 401(k) system for new hires because it found that the latter would be significantly more expensive.
  • In addition, 401(k)s entail significant risk that workers’ retirement income will be substantially lower than anticipated because of unfavorable market conditions when they retire, i.e., steep market downturns that depress average returns despite a lifetime of prudent saving and investing. For instance, if a 401(k) account earns a healthy 6 percent average real return, it would take 5.5 percent of a typical low-wage worker’s earnings to meet the 25 percent replacement goal. This is equal to $138 a month for a worker earning $30,000 a year.3 But if real returns averaged 3 percent for a particular cohort of workers because the market crashes and fails to quickly recover just as these workers retire, the 401(k) would replace only 13.5 percent of their income, leaving them significantly short. DB plans manage this problem by averaging high and low returns across age groups, but the only strategy available to workers in individual 401(k) accounts is to contribute a much higher percentage of earnings: 10 percent in this scenario, or $250 a month.
  • It is unclear how the 401(k) contribution burden will be estimated under the governor’s proposal, or exactly how it will be distributed between employers and employees. To the degree that the plan depends on savings by employees to meet the targets, lower wage workers have less disposable income and are at greater risk of falling short of the necessary savings amounts.
  • If the 401(k) portion of the Governor’s proposed hybrid plan functioned to provide supplemental income, with Social Security and the DB pension covering basic expenses, the above risks, costs, and outcomes might be acceptable. However, the lowest paid public employees in the state–physicians’ assistants, classroom aides, janitors, groundskeepers, etc.–typically can only expect to meet basic living expenses with their target retirement income, just as they have very little discretionary income with which to hedge against poor market performance. Consequently, a retirement benefit that depends significantly on a 401(k) plan means much greater hardship than for them than it would for a higher paid worker with the same benefit structure. In some cases, it may mean reliance on public assistance.

The target earnings replacement ratio of 75 percent (combining Social Security, a DB pension, and a 401(k) plan) is insufficient for lower-wage workers.

  • To begin, 75 percent is a relatively conservative estimate of income replacement sufficient to maintain pre-retirement standard of living for middle and high income households. It is important to note that studies that seriously examine the cost of health care arrive at much higher income replacement needs. For instance, the target increases to 92 percent for a middle-income, dual earner household when health care is included, and to 98 percent when both health care and long term care insurance are included.4 Since the Governor’s 12 point pension reform proposal includes significant reduction of retiree health benefits, the corresponding increase in health care expenses for future pensioners needs to be taken into account when calculating income replacement targets.
  • Furthermore, the percentage of earnings required to maintain a household’s pre-retirement standard of living varies significantly depending on income level, with low-wage workers needing higher replacement rates than other workers. This is because the claims on income that typically decrease or disappear in retirement–income taxes, savings, and work related expenses–take up a smaller share of a typical low-wage worker’s pay. Thus low-wage workers need a higher percentage of their income replaced during retirement in order to maintain their pre-retirement standard of living.
  • Studies typically estimate retirement income replacement needs for low wage workers– those in the bottom third of the income distribution–to be between 80 percent and 90 percent depending on marital status.5 (Among full time workers in California, someone at the 25th percentile of the wage distribution earns about $25,000 a year.) Accounting for health care and long term care expenses would push up replacement needs for these workers past 100 percent.

Depending on how benefits are distributed between the DB and 401(k) plans in coordination with Social Security, the proposal could result in a regressive pension formula for public employees in the state.

The governor essentially proposes that Social Security, DB, and DC benefits will each replace about 25 percent of pre-retirement earnings. This represents a DB pension benefit formula of roughly .7 percent of final pay per year of service, based on a 35 year career. First, this represents a dramatic reduction from the current range of DB benefits, which starts at 1.5 percent. Second, there remain crucial questions about how state sponsored benefits will be coordinated with Social Security; depending on plan design, low-wage workers could end up receiving proportionally less benefit from their employer sponsored pension, including the secure DB pension, than high wage workers. Third, it is unclear how the proposed new pension formula would be applied to part-time workers.

  • How will pension benefits be adjusted in relation to the Social Security benefit structure, which replaces a higher share of earnings for low-wage workers than for high-wage workers?
    • Will the 75 percent replacement rate apply to all workers, regardless of income, or will it be adjusted by income level? Will the DB payout be adjusted in relationship to social security, lowering the DB replacement rate for lower wage workers, or will it be stable across income levels? If the net income replacement burden (after deducting estimated Social Security benefits) is split evenly between DB and DC plans, low-wage workers would see a dramatically lower DB pension formula relative to their earnings than high-wage workers.
    • How will the proposed new tier of retirement benefits account for potential reductions in Social Security?
  • How will the 35 year career standard for reaching 25 percent income replacement through the DB pension apply to part-time workers? Currently, pension benefits for part-time workers are calculated based on pro-rated pension service credits (.5 service credits for 1 year for a half time worker) and full time equivalent pay. A policy requiring 35 years’ worth of service credits rather than 35 calendar years to achieve the target income replacement goal from the DB pension would have the effect of severely penalizing part-time workers, making it impossible to earn full pension benefits proportionate to their employment.

Given these considerations, a fair pension system would take into account the higher income replacement needs of low wage workers, and their limited capacity to absorb the high costs and risks associated with a 401(k) type plan. The current proposal has the potential to penalize low-and middle-wage workers if it Social Security’s progressive benefits as an offset in tandem with a single income replacement target regardless of wage level.

Increasing the Retirement Age

Governor Brown proposes to raise the retirement age for new hires across the board to 67, in line with the scheduled increase in the age of eligibility for full Social Security benefits. Public safety workers are exempted, although their retirement ages also will be raised after consideration of the impact on public safety given the physical demands of specific occupations. The rationale is that life expectancy, and thus the number of years of pension benefit per retiree, has increased; and raising the age at which workers can receive their maximum DB pension benefit will help decrease pension expenses. However, when combined with important differences in career timing and life expectancy by income and occupation, the costs of this policy will be borne disproportionately by low-wage workers and blue collar workers who start their working lives earlier and who die younger than professional workers.

An across-the-board increase in full retirement age to 67 for non-safety workers would force lower-wage workers and workers in manual labor occupations to work longer careers, and live through shorter retirements, relative to their college educated counterparts.

  • Highly paid white collar professionals start their careers later after spending four or more years in higher education and are able to sustain long careers. In contrast, people working in physically demanding jobs, which tend not to require higher education, typically enter the paid workforce years earlier. Janitors, groundskeepers, and hospital orderlies who started working at age 18 or 20 may be many years shy of full retirement benefits under the Governor’s plan when their bodies give out after 40 or even 45 years of work. Furthermore, they may face difficulty transitioning to another, less physically demanding occupation due to skills mismatch.
  • In addition, this proposal would translate to lower retirement benefits for lower-earning, less educated workers who do not live as long and who have benefited least from the historical growth in life expectancy. The growing gap in life expectancy by education and income in the US is well documented. For instance, between the generation that retired in the 1960s and the one that retired in the 2000s, life expectancy at age 65 increased by 5.5 years to age 86 for men in the top half of the income spectrum.6 Their counterparts in the bottom half of the income spectrum gained one year, to age 81. Significant variations in life expectancy by socioeconomic status (e.g., income and education combined), controlling for race, have also been documented in California.7


Lower wage workers could be significantly disadvantaged if the Governor’s proposals for a hybrid pension and increased retirement age for new hires is applied without regard for wage level or occupational characteristics–through a severely downgraded pension benefit that fails to provide meaningful retirement security and a longer working career/shorter period of retirement compared to higher paid workers.


  1. Felicitie C. Bell and Michael L. Miller, Life Tables for the United States Social Security Area 1900-2100, Actuarial Study No. 120, SSA Pub. No. 11-11536, Washington, DC: Social Security Administration/Office of the Chief Actuary, August 2005.
  2. Beth Almeida and William B. Fornia, A Better Bang for the Buck: The Economic Efficiencies of Defined Benefit Pension Plans, Washington, DC: National Institute on Retirement Security, August 2008.
  3. Estimate assumes that the 401(k) account balance will be used to purchase a single premium, inflation protected life annuity at current rates.
  4. Alicia H. Munnell, Anthony Webb, Francesca Golub-Sass, and Dan Muldoon, “Long-Term Care Costs and the National Retirement Risk Index,” Issue in Brief No. 9-7, Chestnut Hill, MA: Center for Retirement Research at Boston College, March 2009, p. 10, Figure A-2.
  5. See for example Alicia H. Munnell, Anthony Webb, and Luke Delorme, “A New National Retirement Risk Index,” Issue in Brief No. 48, Chestnut Hill, MA: Center for Retirement Research at Boston College, June 2006; and Aon Consulting, 2008 Replacement Ratio Study, http://;
  6. Dean Baker and David Rosnick, “The Impact of Income Distribution on the Length of Retirement,” Issue Brief, Washington, DC: Center for Economic and Policy Research, October 2010.
  7. Christina A. Clarke and Amal Harrati, “Life Expectancy in California’s Diverse Population: Recent Estimates by Race/Ethnicity and Neighborhood Social Class,” pp. 42-56 in Nari Rhee (Ed.), Meeting California’s Retirement Security Challenge, Berkeley, CA: UC Berkeley Center for Labor Research and Education, October 2011.