Obama's Universal Savings Plan May Not Be Right for Everyone
Click here for a printer friendly version of this article.
In early September, the Obama Administration launched an initiative to “help” people save. President Obama used his weekly radio address to announce administrative actions to expand automatic enrollment in employer-sponsored retirement plans, convert tax refunds into savings bonds, and allow employees to accept certain forms of compensation using deposits into retirement account. These administrative actions are but the first part of a comprehensive plan to create a universal system of savings accounts – through an expansion of “automatic enrollment.”
Auto-enrollment is the practice of “making it easier” for employees to participate in an employer-sponsored retirement plan by making it harder for them not to participate. Most auto-enrollment plans also feature automatic increases that “make it easier” for enrolled participants to increase their retirement contributions by doing it for them. Studies in peer reviewed journals have found that workers suffer from cognitive biases that lead them to accept default options even when they are not in the individuals’ economic interest. The data show that participation rates in employer-sponsored retirement plans are much higher when the employees are automatically enrolled in the program than when they are required to elect to participate.
The finding that “inertia” impacts individual decisions is both empirically robust and obvious. In the context of workplace savings, the default choice for retirement plan participation is likely impacted by the overall inscrutability of the average paycheck. There are a number of line items on every worker’s paystub that reflect automatic deductions from gross income (FICA, health insurance, life insurance, state taxes, etc.) that may not be readily understood by the worker. The plan here is simply to add one more.
But the notion that government is “making it easier” for a worker to save for retirement by encouraging employers to take money out of paychecks unless the worker objects misrepresents what’s actually occurring. The issue is not about simplifying saving, but rather the presumption that people are making bad decisions.
The auto-enrollment initiative, therefore, depends in large part on the notion that Americans do not save enough. The assumption of inadequate savings is not supported by the behavioral literature that “discovered” the benefits of auto-enrollment. While the country as a whole may save less than is prudent, the question of whether a specific household saves “enough” for retirement is quite separate from the question of whether that household would be more inclined to participate in a 401(k) if automatically enrolled in one.
For example, the Obama Administration estimates that participation in retirement savings plans among the “poorest households” would increase from 15% to 80% through a universal account system. But diverting the compensation of these liquidity-constrained households from current consumption towards retirement creates its own set of problems. The United States has a health insurance “crisis” largely because low-income workers are asked to devote an unusually large and fast-growing percentage of their total compensation to homogenized health coverage options. Is the U.S. government not concerned that these policies could perpetuate the crisis by requiring these same households to invest in homogenized savings plans that may or may not fit their needs? If the problem is low national savings, why is the proposal (or the rhetoric surrounding it) so focused on low income households? It’s intuitive, but worth repeating: savings is overwhelmingly generated by higher income households. It’s the reason stimulus measures tend to focus on getting money out to households with the highest “marginal propensity to consume.”
According to Federal Reserve research, the decline in savings among households in the top income quintile accounts for nearly all of the decline in the aggregate U.S. saving rate since 1989. But this decline is not the product of how workplace savings plans’ default options are configured. The fall in savings by higher income earners coincided with a rapid rise inthe prices of assets they owned. First, stocks in the 1990s and then housing in the past decade. The fact that capital gains are not reflected in the national savings rate brings into question the utility of this economic indicator. And using low national savings rates as a rationale for increasing the deferred portion of low-income workers’ compensation underscores how complicated this issue is and how easy it is to misdiagnose.
Policymakers intent on taking action to increase America’s savings rate have been reluctant to address the defects in the Administration’s plans. While sensible objections may not be enough to slow this train, at the very least it will be amusing to watch Congress pass reforms to “boost savings immediately” following a period when household savings grew at the fastest rate in a generation. If (or when) reform does come, the plan should be tailored to ensure that automatic savings does not do more harm than good to the targeted, low income households.