Thursday, February 28, 2019

New paper: “Why Has Poverty Declined for Widows?”

The Center for Retirement Research at Boston College has released a new Issue in Brief:

“Why Has Poverty Declined for Widows?”

by Alicia H. Munnell, Geoffrey T. Sanzenbacher, and Alice Zulkarnain

The brief’s key findings are:

  • Since the mid-1990s, the poverty rate for widows has dropped sharply.
  • Potential reasons include rising levels of education and work experience for women generally and a higher marriage rate among women with more education.
  • The findings show that, so far, the drop in widows’ poverty has primarily been driven by the general increase in women’s education and work experience.
  • Going forward, the poverty rate should continue to fall not only because of education and work patterns but also because of marriage selection.
  • Despite the progress, widows will remain at greater risk of poverty than married women.

This brief is available here. Read more!

Monday, February 25, 2019

New NBER Working Papers on Retirement

Retirement Implications of a Low Wage Growth, Low Real Interest Rate Economy

Jason Scott, John B. Shoven, Sita Slavov, John G. Watson

NBER Working Paper No. 25556
Issued in February 2019
NBER Program(s):Aging

We examine the implications of persistent low real interest rates and wage growth rates on individuals nearing retirement. We begin by reviewing the concept of r star – the long-term real, safe interest rate that is neither expansionary nor contractionary – and presenting recent estimates suggesting that this value has declined. We then examine the implications of low returns and low wage growth for individuals currently aged 45 and 55. We find that low returns and low wage growth have substantial welfare effects, with compensating variations that are often in the hundreds of thousands of dollars. Low returns increase optimal Social Security claiming ages and the marginal benefit of working longer, while low wage growth decreases the marginal benefit of working longer. Low economy-wide wage growth has a much larger welfare effect than low individual wage growth due to wage indexation of the initial benefit and the progressivity of the Social Security benefit formula. When individual wage growth alone is low, wage indexation is unchanged, and the progressivity of the benefit formula provides insurance. When economy-wide wage growth is low, wage indexation is less generous and there is no insurance benefit from progressivity as average wages fall along with individual wages.

Household Responses to Transfers and Liquidity: Evidence from Social Security's Survivors Benefits

Itzik Fadlon, Shanthi P. Ramnath, Patricia K. Tong

NBER Working Paper No. 25586
Issued in February 2019
NBER Program(s):Aging, Labor Studies, Public Economics

We use administrative tax data that cover the U.S. population to identify the causal effects of Social Security’s survivors benefit receipt on American families’ behavior and financial well-being. We analyze over a quarter of a million widowed households in which the husband died between 2002-2007, and we exploit a sharp age discontinuity in benefit eligibility to study the responses of financially vulnerable households to government transfers. We first study how households respond to unanticipated benefit receipt in the immediate periods following a large financial shock to investigate the protective role of transfers. We find significant impacts of the program on newly-widowed families’ net income and labor supply behavior, which points to considerable allocative inefficiencies in the life insurance market and to a high valuation of survivors benefits in protecting Americans against mortality shocks. Second, to investigate the particular role of liquidity and benefit timing, we then study how already-widowed women’s labor supply responds to anticipated survivors benefit receipt. We find considerable responses to cash-on-hand via benefit availability that underscore allocative inefficiencies in the credit market and the value of liquidity itself provided by government transfers. These responses and their heterogeneity highlight mechanisms that underlie the labor supply behavior of older vulnerable households, and they point to liquidity constraints, rather than myopia or benefit-schedule misperceptions, as the likely operative channel. Our results have implications for survivors benefits in the U.S., and, more generally, for retirement behavior and response mechanisms to transfers among older vulnerable populations.

Employer Concerns and Responses to an Aging Workforce

Robert L. Clark, Steven Nyce, Beth Ritter, John B. Shoven

NBER Working Paper No. 25572
Issued in February 2019
NBER Program(s):Aging

Economist and public policy analysts have devoted considerable research to examining the work and retirement decisions of employees. Much less effort has been spent on understanding the concerns and challenges of employers if their workers delay retirement and remain on the job until older ages. In this study, we report findings from three employer surveys with the objective of learning how organizations are responding to the aging of their workforces. The surveys provide several important observations. First, employer concerns about workforce aging vary considerably across the economy. To some firms, these demographic changes are of immediate concern and are viewed as a significant risk to the organization while other firms remain more concerned about potential productivity and cost effect of an older labor force. Second, most employers expect the importance of workforce aging to increase in the next five years. In response, a significant proportion of organizations are making changes to working conditions and compensation policies. Third, firms remain reluctant to adopt formal phased retirement policies but are more willing to offer part-time employment, return to work, and other policies on a case by case basis.

Read more!

Washington Post endorses “progressive price indexing” for Social Security

In an editorial, the Washington Post opposed Social Security expansion and endorsed a Bush-era reform to reduce benefits for middle and income retirees in the future.

The Social Security 2100 Act, co-sponsored by over 200 House Democrats, would

divert scarce resources toward a vast majority of Social Security recipients who are not only not poor but, in many cases, perfectly comfortable. We are all for making the overall tax system more progressive than it already is, including by taxing high earners, as the Social Security 2100 Act would do. You can tap “the rich” only so many times, however; and the priority should be to use that money for children, who are almost twice as likely to be poor as senior citizens.

Instead, the Post argues, we should consider a proposal that was part of President George W. Bush’s Social Security reform proposals, so-called “progressive price indexing.”

There is an alternative proposal to make Social Security both protective of the elderly poor and more solvent over the long term. The plan would retain Social Security’s current benefit formula for the 30 percent of workers with the lowest lifetime earnings, while reducing the growth rate of initial benefits for the top 70 percent. Phased in over the next 30 years, it would save a relatively modest $77 billion in the first decade. Savings would accumulate more quickly thereafter, though, to reduce Social Security’s total claim on national output in 2048 from 6.3 percent under current law to 5.7 percent, according to the Congressional Budget Office. The CBO refers to the proposal as “progressive price indexing.” Given that it allocates the nation’s limited retirement-income resources to those who need them most, instead of promising more benefits to practically everyone, the “progressive” label does indeed apply.

I certainly wouldn’t oppose a reform package based on progressive price indexing. At the same time, it’s a proposal that’s very difficult for the public to understand and doesn’t have any end goal of what we want our Social Security program to look like. Progressive price indexing says simply that we’re going to reduce benefit growth for middle and high earners, not what we (as a society or as policymakers) think the appropriate benefit levels are for Americans of different means.

That’s why over time I came to favor a different, and more radical approach: a flat dollar benefit that provides a true guarantee against poverty in old age, while requiring middle and high earners to save more for retirement. This flat dollar benefit approach is similar to New Zealand’s retirement system, along with reforms that have been enacted in the U.K.

Read more!

Monday, February 18, 2019

Social Security expansion could be a disappointment for the poor.

Over at National Review Online, I write on how Social Security expansion as proposed by Rep. John Larson and his 200 House co-sponsors and Sen. Bernie Sander’s separate bill may end up disappointing poor retirees, the ones you think would most need the extra benefits.

On paper, plans like the Social Security 2100 Act offer very large benefit increases to very low earners, who make up about the poorest fifth of retirees. The SSA actuarial memos for these plans offer examples showing benefit increases of up to 44%.

But these examples look at individuals who work a full career and receive benefits based on their own earnings. But that’s not always the reality for low-income retirees. Many failed to work a full career; a short working career is the easiest route to a low income in retirement. Others low-wage workers end up being “dually-entitled” in retirement, meaning that they receive a benefit based upon their own earnings plus an additional “auxiliary benefit” based on the earnings of a higher-earning spouse. Their total benefit is capped at half of the higher-earning spouse’s check. Social Security expansion might boost a retirees’ own earned benefit, but in most cases that would simply result in a lower auxiliary benefit. While higher-earning retirees would be eligible for much smaller benefit increases, because these high earners don’t receive auxiliary benefits they might be more likely to receive what they’ve been promised.

Social Security expansion also favors high earners in a another way: they live longer. This counts not just for increases in initial retirement benefits, but also in the higher COLAs that Social Security expansion plans offer.

I’ve argued for boosting benefits for low-income retirees. We’re a rich nation and can afford to keep seniors out of poverty. But since Social Security is a complex program, we should think carefully about how we go about doing it.

Read more!

Tuesday, February 12, 2019

New study: “Retiring Earlier than Planned: What Matters Most?”

The Center for Retirement Research at Boston College has released a new Issue in Brief:

“Retiring Earlier than Planned: What Matters Most?”

by Alicia H. Munnell, Matthew S. Rutledge, and Geoffrey T. Sanzenbacher

The brief's key findings are:

  • More than a third of older workers retire earlier than planned: the question is why?
  • This study looks at: 1) the impact of unexpected changes in health, employment, family, and finances on early retirement; and 2) the prevalence of these shocks.
  • The findings suggest that:
    • Health shocks play the largest role, mainly because they are widespread.
    • Job loss without finding a new job, while not as prevalent, is also important.
    • Family transitions have a modest impact, while financial shocks appear to have little effect.
  • A key caveat is that all the shocks combined explain only about a quarter of earlier-than-planned retirements, so clearly other factors are also at play.

This brief is available here. Read more!