Report | Retirement

California retirement plan could serve as a national model

Policy Memo #193

Press release

A bill to create a publicly administered retirement savings program for private-sector workers in California who do not have access to an employer-sponsored plan has passed both houses of the state legislature and awaits the signature of Governor Jerry Brown. The California Secure Choice Retirement Savings Program (SB 1234), co-authored by California State Senators Kevin de León (D-Los Angeles) and Senate President Pro Tempore Darrell Steinberg (D-Sacramento), would require employers who do not offer a retirement plan to implement automatic paycheck deduction for employee contributions to a low-fee and relatively low-risk retirement savings plan. Though some of the details remain to be worked out by the plan’s board of trustees, workers would probably be assured of recouping their contributions while retaining some of the upside of investing in a balanced, low-cost fund.

Unless they opt out, workers would be automatically enrolled in the plan and 3 percent of earnings would be contributed on their behalf. These employee contributions would be pooled and professionally managed by an independent investment manager—either a private firm or the California Public Employees’ Retirement System (CalPERS) working on a contractual basis.

As the federal Thrift Savings Plan (TSP) demonstrates, pooling reduces administrative costs and can boost net returns by a percentage point (Davis, Kazzi, and Madland 2010), a seemingly small difference that can increase the size of a retirement nest egg by one-third.1 Pooled pension funds are also able to achieve higher risk-adjusted returns than 401(k) account holders because institutional investors are better able to take advantage of professional expertise and because combining the savings of workers who retire at different times reduces the need for liquidity.

Whereas many 401(k) participants take an all-or-nothing approach to risk, the Secure Choice fund would be invested in a balanced portfolio, with at most 50 percent invested in equities. Many 401(k) participants choose riskier asset allocations based on the false but widespread belief that cumulative stock returns average out over time. Even a 50-50 stock-bond portfolio carries significant risk (Burtless 2008), though the Secure Choice board of trustees would have the leeway to set more conservative investment guidelines or—better yet—use a reserve fund to smooth out returns for workers who retire in the wake of bull and bear markets.

The potential for intergenerational risk-sharing could be the most significant innovation of the California plan, which is loosely based on a model developed by the National Conference on Public Employee Retirement Systems (NCPERS) (Kim 2011). A key difference between the California plan and the NCPERS model is that the California plan is structured as a payroll deduction individual retirement account (“auto IRA”), whereas the NCPERS plan is a cash balance plan, a type of employer-sponsored defined benefit pension that falls under the Employee Retirement Income Security Act (ERISA) and allows for employer contributions.

The California plan also has many similarities to EPI’s Guaranteed Retirement Account (GRA) plan, which calls for the government to guarantee a return of 3 percent above inflation (Ghilarducci 2007).  The California plan would likely guarantee a more modest rate of return through private underwriters—possibly only insuring against losses (a guarantee of principal). This has political advantages over the GRA plan, since taxpayers would incur no risk. However, it would also be less efficient to the extent that government entities are less risk averse than private insurers and could provide the same guarantee at a lower cost (Munnell et al. 2009; Stubbs and Rhee 2012).

For workers participating over the course of a long career, it may not matter whether the guarantee is set at 3 percent real or 0 percent nominal, since, based on historical returns and stochastic simulations, it is unlikely that returns on a balanced portfolio would fall short of even the higher guarantee over 30 years (Munnell et al. 2009; Stubbs and Rhee 2012). Conversely, this implies that there may be little risk for taxpayers in the government-insured GRA plan, and what risk remains must be balanced against the fact that increased hardship in old age also costs taxpayers due to increased reliance on government safety net programs.

A modest rate-of-return guarantee does provide a real protection to short-term investors—workers who enroll in the plan a few years before retiring. It may also bring peace of mind to other workers. Nevertheless, the Secure Choice board could provide more retirement security through a lower guarantee and a higher target rate of return, since the plan calls for purchasing the guarantee through private insurers and the cost of a higher guarantee could erode returns. In other words, using the reserve fund to smooth out returns is more important than purchasing a minimum rate-of-return guarantee, though both are left to the board’s discretion.

In any case, the California plan is an enormous improvement over most plans to automatically enroll workers in 401(k)s or IRAs because it does not steer workers’ savings to risky, high-cost investments. Thus, it generally adheres to the principle of “first, do no harm,” unless workers interpret the 3 percent contribution as sufficient.  For most workers, it is not. An inflation-adjusted annuity based on the default contribution to a Secure Choice account would replace roughly 18 percent of earnings after a lifetime of work, assuming a 3.5 percent net rate of return above inflation (see Stubbs and Rhee 2012 for a discussion of plausible rates of return).2 Since Social Security will replace roughly 37 percent of earnings and most experts recommend a total replacement rate of at least 70 percent, most workers should be encouraged to contribute at least twice the default amount. Indebted workers in lower tax brackets are the exception, because federal subsidies for retirement savings will continue to be skewed toward taxpayers in higher tax brackets. With little or no tax benefit, it makes more sense for many low-income workers to opt out of the plan in order to pay down credit card balances and other loans first.

While the GRA plan and a plan proposed by Senator Tom Harkin, chairman of the U.S. Senate Committee on Health, Education, Labor, and Pensions (U.S. Senate HELP Committee 2012), would address the problem of upside-down tax breaks and provide other advantages like mandatory annuitization, both the GRA plan and the Harkin plan are, admittedly, bigger political lifts. In the meantime, the California plan, which enjoys both union and small-business support (SEIU Local 721 2012; Merman and Klinger 2012), could be a very important first step. It would also serve as a model for lawmakers in other states concerned about retirement security and seeking to take advantage of economies of scale and other advantages afforded by state-administered pension funds.


1. A worker making a one-time contribution and earning an 8 percent investment return net of fees will have a 32 percent larger accumulation after 30 years than one with a 7 percent net return. A worker who contributes the same amount each year over 30 years will end up with a 21 percent larger account balance with the higher return, and one who contributes a steadily increasing amount—growing by 2 percent per year—will end up with an 18 percent larger balance with the higher return. The differences are greater if assumed returns are lower or the investment period is longer.

2. Authors’ calculations based on Social Security’s prototypical medium earner beginning her career at age 21 in 2012 and retiring at 65 with a remaining life expectancy of 22 years (assumptions adapted from the 2012 Social Security Trustees Report and related tables as well as Clingman and Burkhalter 2011). Replacement rates are based on peak earnings. The Secure Choice annuity and Social Security benefit would replace a higher share of final earnings, since the prototypical worker’s earnings peak in her early 50s.


Burtless, Gary. 2008. “Stock Market Fluctuations and Retiree Incomes: An Update.” Washington, D.C.: The Brookings Institution, October 30.

Clingman, Michael and Kyle Burkhalter. 2011. “Scaled Factors for Hypothetical Earnings Examples under the 2011 Trustees Report Assumptions.” Actuarial Note No. 2011.3. Baltimore, Md.: Social Security Administration, July.

Davis, Rowland, Nayla Kazzi, and David Madland. 2010. The Promise and Peril of a Model 401(k) Plan: Measuring the Effectiveness of Retirement Savings Plans Offered by Private Companies and the Federal Government. Washington, D.C.: Center for American Progress.

Ghilarducci, Teresa. 2007. Guaranteed Retirement Accounts: Toward Retirement Income Security. Briefing Paper No. 204. Washington, D.C.: Economic Policy Institute.

Kim, Hank H. 2011. The Secure Choice Pension: A Way Forward for Retirement Security in the Private Sector. Washington, D.C.: National Conference on Public Employee Retirement Systems.

Mermin, David and Jeff Klinger. 2012. National Conference on Public Employee Retirement Systems: Findings from a Survey of 505 Small Business Owners in California. Washington, D.C.: Lake Research Partners.

Munnell, Alicia H., Alex Golub-Sass, Richard W. Kopcke, and Anthony Webb. 2009. “What Does It Cost to Guarantee Returns?” Issue in Brief 9-4. Chestnut Hill, Mass.: Center for Retirement Research at Boston College, November.

Service Employees International Union Local 721. 2012. “‘Retirement Security for All’ Legislation Moves Forward,” August 17.

Social Security Trustees. 2012. The 2012 Annual Report of the Board of Trustees of the Federal Old-Age and Survivors Insurance and Federal Disability Insurance Trust Funds (including single-year tables consistent with the report), April 25.

Stubbs, David M. and Nari Rhee. 2012. “Can a Publicly Sponsored Retirement Plan for Private Sector Workers Guarantee Benefits at No Risk to the State?” Policy Brief. Berkeley, Calif.: UC Berkeley Center for Labor Research and Education.

U.S. Senate Committee on Health, Education, Labor and Pensions. 2012. The Retirement Crisis and a Plan to Solve It.

See related work on Retirement

See more work by Monique Morrissey and Ross Eisenbrey