October 2005
Volume 6, Issue 3

Director’s Corner

With no shortage of legislative proposals in the offing, the Retirement Research Consortium’s annual conference in Washington, D.C. included a panel of Washington-area experts outlining a range of possibilities. We were fortunate to have as panelists Eva Cargill from the Office of Congressman Wexler, Kim Hildred from the Committee on Ways and Means, Tom Miller from the Joint Economic Committee, Jennifer Olson from the Office of Senator Graham, and Chris Socha from the Office of Senator DeMint. The discussion following the panel presentations, moderated by James B. Lockhart III, included questions from the audience and was lively and informative. The RRC routinely draws approximately 300 attendees from academe, government agencies, the private pension sector, and, of course, from the Social Security Administration itself. This year the Boston College center had primary responsibility among the members of the Retirement Research Consortium for organizing the effort, and I congratulate it and the Office of Policy at the SSA for the conference’s success. Discourse between the academic, policy, and legislative community is, it seems to me, potentially invaluable and the annual RRC conference provides an excellent, established venue for timely exchanges among these groups. A full write-up of the conference is included in this issue of the MRRC newsletter.

John Laitner, Director MRRC

Regular Features


Inside This Issue
2005 RRC Conference Held

New Research Awards

Researcher Q & A

New Working Papers

Commissioner Barnhart Message


 7th Annual Retirement Research Consortium Conference Held in Washington, D.C.

The seventh annual conference of the Social Security Retirement Research Consortium was held at the National Press Club in Washington, D.C. on August 11th and 12th, 2005. The two-day conference, entitled “Towards a Secure Retirement System,” was planned and arranged this year by the Center for Retirement Research at Boston College (CRR) in conjunction with the Michigan Retirement Research Center (MRRC), the NBER Retirement Research Center, and the Social Security Administration (SSA). To open the conference, greetings were offered by Alicia Munnell, Director of the CRR, John Laitner, Director of the MRRC, and David Wise, Director of the NBER center. Introductory remarks were made James B. Lockhart III, Deputy Commissioner of the Social Security.

The luncheon address was delivered by Charles Blahous, who is Special Assistant to the President for Economic Policy and a member of the National Economic Council.

Charles Blahous and Jo Anne B. Barnhart, Commissioner of SSA

A panel discussion entitled “Social Security Reform-Congressional Proposals and Views” was lead by Deputy Commission Lockhart. The panel included Eva Cargill from the Office of Congressman Wexler, Kim Hildred from the Committee on Ways and Means, Tom Miller from the Joint Economic Committee, Jennifer Olson from the Office of Senator Graham, and Chris Socha from the Office of Senator DeMint.

James B. Lockhart, III Deputy Commissioner Of SSA

Over the two days, researchers from each of the three consortium centers presented papers on current projects, all of which are funded by SSA. The topics covered in the seven panel sessions included papers addressing the financial needs of retirees, planning and saving behavior, effects of unexpected events in retirement, factors affecting labor force participation of older workers, risks and returns on Social Security private accounts, and the implications of individual investment behavior for private accounts.

MRRC Display

Papers presented at the conference are available on the MRRC website. http://www.mrrc.isr.umich.edu/

back to the top.

 New MRRC Research Awards

The MRRC is pleased to announce the following research awards for 2005-2006.

Enhancing the Quality of Data on Income and Wealth F.
F. Thomas Juster, Daniel H. Hill, Michael Perry, and Honggao Cao

How Changes in Social Security Affect Retirement Trends
Alan L. Gustman and Thomas L. Steinmeier

Alternative Measures of Replacement Rates
Michael Hurd and Susann Rohwedder

Household Savings and Retirement: Expectations and Realizations
Steven J. Haider and Melvin Stephens

Planning and Financial Literacy: How Do Women Fare?
Annamaria Lusardi

Trading Behavior in Personal Accounts: Lessons from 401(k) Pension Participants
Olivia S. Mitchell

Personal Social Security Accounts: Quantifying the Macro and Efficiency Effects II
Shinichi Nishiyama and Kent Smetters

Labor Supply of Older Americans: Effects of Tax Rates and Tax Treatment of Pension and Social Security Income
Lucie Schmidt and Purvi Sevak

Probabilistic Thinking and Early Social Security Claiming
Robert J. Willis and Adeline Delavande

Home Production by Dual Earner Couples and Consumption During Retirement
Chris House, John Laitner and Dmitry Stolyrov

Designing Optimal Pension Payouts Using Life Annuities and Phased Withdrawals: A Portfolio Approach
Raimond Maurer

Financial Risk, Retirement, Saving and Investment
Alan L. Gustman & Thomas L. Steinmeier

The Effect of Abolishing the Earnings Test on Labor Market Behavior Before Age 65
Pierre-Carl Michaud and Arthur van Soest

Pathways to Bad Retirement Outcomes
Susann Rohwedder

The Role of Spousal Health Insurance Provision in Retirement Decisions Jody Schimmel

Assessing the Risks and the Costs of the Risks Posed by Social Security's Potential Insolvency Lawrence Kotlikoff

Evaluating Social Security Reforms using Probabilistic Population Forecasts in a Life-cycle Model of Social Insurance Warren C. Sanderson, Hugo A. Benitez-Silva, & Debra Sabatini Dwyer

Health Risk in Retirement and Wealth Accumulation
Purvi Sevak and Lina Walker

Who Annuitizes and How Do System Rules Shape These Choices? The Case of Chile
Estelle James and Alejandra Cox-Edwards

Retirement Savings Portfolio Management
Jeff Dominitz and Angela Hung

Rising Economic Risk and the Labor Supply of Older Workers
Nicole Maestas

The Dynamics of Work-Related Health and Labor Market Status: An International Comparison
Arie Kapetyn, Pierre-Carl Michaud, James P. Smith, Arthur van Soest, and James Banks

Efficiency Gains and Social Security Reform
John Laitner and Daniel Silverman

back to the top.


Request a Social Security Speaker

The Social Security Administration seeks opportunities to communicate with the public about important issues surrounding Social Security. Speakers are available nationwide for conferences, seminars, classrooms, workshops, etc.

If you are interested in having a Social Security representative speak at your organization's event, visit the Social Security website for an on-line request form. http://www.socialsecurity.gov/organizations/

Also, look for the Social Security display at professional meetings and conferences.

back to the top.

Researcher Q & A

The MRRC recently asked researchers a series of questions about their work. In this issue Kent Smetters discusses research conducted with Shinichi Nishiyama.

How would you describe your MRRC-supported work?

Smetters photographKent Smetters is an Associate Professor of Insurance and Risk Management and Undergraduate Program coordinator at the Wharton School, University of Pennsylvania. In 2001-2002, he was Deputy Assistant Secretary of Economic Policy at the U.S. Treasury and currently serves on the Blue Ribbon Advisory Panel on Dynamic Scoring for the U.S. Congress.
Nishiyama photographShinichi Nishiyama is an Assistant Professor of Risk Management and Insurance in the Robinson College of Business at Georgia State University. In 2000-2005, he was a principal analyst of fiscal policy studies at the Congressional Budget Office, and specialized in the macroeconomic analyses of president's budget and Social Security reform.

All of my MRRC-sponsored research has focused in one way or another on reform of Social Security.

Previously, Jan Walliser and I proposed and examined the impact of the transition to a fully funded system. More recently, Shinichi Nishiyama and I have focused our attention on potential efficiency gains and losses associated with personal Social Security accounts.

Most reform options being considered involve having winners and losers. When policy makers debate different reform scenarios, it is important for them to have information on who wins and who loses. The best possible reform would be something economists call Pareto improving, where the reform improves things for some without making anyone else worse off. More realistically there will be losers, but we want to minimize that if possible. Another way we think about this is to ask: how much money is left over after we’ve compensated all those people who have been harmed by the policy change? In our earlier work we developed a model that would allow us to examine this question. Our recent project sponsored by the MRRC uses this model.

How does your research differ from previous literature in this area?

Two major sources of uncertainty we all face are about our lifetime wages and how long we’re going to live. We’d all like to be able to buy some type of insurance against these two types of shocks. Of course, we can buy life insurance, but it’s very hard to buy insurance against wage shocks because of what we call adverse selection or the moral hazard problem—in this case the risk that people would buy this type of insurance and then intentionally lose their jobs in order to collect the insurance. So there really isn’t a market for wage insurance.

Most previous models – although there are a couple of exceptions -- have implicitly assumed that these markets do exist by assuming that wages could be perfectly predicted. Using these simplistic models, researchers were able to show very large gains from Social Security reform. Over time, improved computing technology has allowed us and others to create models that are computationally more complex. And now we can ask, with a more realistic model that includes uninsurable wage shocks, how do the results change? And the answer is that they change quite a bit. In fact, we now show that there could be efficiency losses from partial Social Security privatization unless the privatization is designed correctly. Our model is, I believe, the first to rigorously compute efficiency changes inside of a more realistic environment by incorporating a technique first used by Alan Auerbach and Laurence Kotlikoff into this more advanced model.

Interestingly, efficiency losses are possible despite the fact that privatization does produce gains in capital stock and GDP both over the long run as well as over the short run. The old models also found more saving because as you replace a pay-as-you-go system with private accounts, naturally there is more saving. That’s still going on in our new model, but there’s another reason we see more saving. Under privatization there is a reduction in the insurance that is being provided by the Social Security Administration after privatization. This encourages households to engage in additional precautionary savings. Also, as money accumulates in these accounts, the interest rate drops because capital is less scarce.

Were your results unexpected?

We tested these results in a number of different ways. In our baseline model, we assumed that the U.S. was a closed economy—that there was no relationship of our economy to other nation’s economies. When we allow the economy to be open, some of the extra capital that accumulates under privatization can go abroad, and interest rates do not fall. In this context we show that the efficiency losses under private accounts get even bigger. This is a bit of a surprise, but here’s the explanation.

As I said before, under privatization in the short run you have some losers and in the long run you have some winners. In the short run you have to have some additional debt in order to compensate the losers. You pay for the debt with some of the gains from the winners in the long run. If you have a low interest rate, that allows you to borrow the money at a cheaper rate and allows you to make this transfer between generations more easily. In the open economy, interest rates do not fall, and the cost of this intergenerational transfer is greater. Therefore, the efficiency loss to privatization is magnified.

In the baseline model, we also assumed that there were no private annuity markets—that people could not buy the type of insurance that traditional Social Security provides. An even bigger surprise in our work is that we find that privatization actually performs worse in a model that does have private annuity markets. The reason is that if the annuity market does not exist, then people are kind of stuck and they have to increase their precautionary savings, which depresses interest rates. The reduction in rates is good for efficiency reasons. When there are private annuity markets, there is less precautionary savings, which keeps interest rates high, and, hence, raises the cost of the intergenerational transfer I just mentioned.

The current system provides risk sharing because Social Security benefits are progressive. In our model, we are also able to examine various options for including some risk-sharing with private accounts. One way is for the government to subsidize the contributions made by poorer households through matching. It turns out that’s not very effective. The problem is this has to be phased out at some income level. The phase out range creates implicit marginal tax rates, which means that it discourages you from working more because as your income increases you lose some benefits. We try a model in which there is no phase out and we match the contributions even of Bill Gates. The cost of this is prohibitive, though, and it creates distortions because the government has to raise that money.

The alternative approach is to make the part of the system that remains unchanged (under a system of partial privatization) more progressive. That turns out to be very effective and the reason is there’s no real implicit marginal tax rate since redistribution is based on lifetime wages and not current wages. In fact, if the remaining Social Security system is made sufficiently progressive then we can even produce efficiency gains from partial privatization. Interestingly, this approach is similar in spirit to the President’s proposal, although we have not computed his exact plan.

Where are you headed next with this research?

In our previous work, we made two simplifications. The first is that we assumed that there’s no correlation between mortality and income. But, in reality, we know that richer people live longer. Our new model will incorporate this relationship. There’s some controversy in the literature as to how progressive Social Security really is because richer people are compensated less but they live longer. So we’ll now get this in there.

The second issue is that the model has up to now assumed that there is no human capital investment. People arrive in our model already endowed with all their human capital, all the investments their parents made in them. We will explicitly model that because Social Security could have a positive or negative effect on the accumulation of human capital. On the one hand, if Social Security gives you a worse deal as you get richer, that could discourage your investment in the types of human capital that would make you richer, like education. On the other hand, you never really know when you make an investment in something like schooling if it’s going to pay off in terms of wealth. So the risk sharing aspect of Social Security (that actually provides a kind of wage insurance) could encourage or discourage human capital investment.

back to the top.

New MRRC Working Papers

Below is a selection of abstracts from recently released MRRC working papers. Visit our website for full papers and other current papers. http://www.mrrc.isr.umich.edu/publications/papers/.

Enhancing the Quality of Data on Income and Wealth
Honggao Cao, Daniel H. Hill, F. Thomas Juster and Michael Perry

Over the last decade or so, a substantial effort has gone into the design of a series of methodological investigations aimed at enhancing the quality of survey data on income and wealth. These investigations have largely been conducted at the Survey Research Center at the University of Michigan, and have mainly involved two longitudinal surveys: the Health and Retirement Study (HRS), with a first wave beginning in 1992 and continued thereafter every other year through 2004; and the Assets and Health Dynamics Among the Oldest Old (AHEAD) Study, begun in 1993 and continued in 1995 and 1998, then in every other year through 2004. This provides an overview of the main studies and summarizes what has been learned about correcting longitudinal inconsistencies that arise.

Knowledge and Preference in Reporting Financial Information
Honggao Cao and Daniel H. Hill

This article models respondent behavior in a financial survey with a framework explicitly integrating a respondent’s knowledge of and willingness to reveal his or her financial status. Whether a respondent provides a valid answer, a “don’t know”, or a “refusal” to a financial question depends on the interaction of his or her financial knowledge and preferences regarding revealing the knowledge. Using asset response and nonresponse data from the Health and Retirement Study (2000), we found that knowledge and preferences play interrelated roles in reporting financial information, that a respondent’s age, gender, education, and race and ethnicity are important predictors of respondent behavior, and that race and ethnicity affect respondent behavior only via their influence on preferences, while gender only via its influence on knowledge.

back to the top.

Commissioner Barnhart Sends Message of Strength on SSA 70th Anniversary

On Sunday, Social Security marks its 70th anniversary. When the program was created in 1935, America was a vastly different nation. We were in the midst of the Great Depression, and at least one-third of all older Americans were dependent upon others for their financial support. Retirement was something that happened when you could no longer work -- not something you planned for in advance.

When Franklin Roosevelt signed the legislation into law, I'm not sure even he realized the significant role the program would play in the next seven decades. I daresay no one foresaw Social Security becoming part of the fabric of society. Yet that is precisely what happened. And, with good reason. Since its inception, Social Security has paid approximately $8.4 trillion in benefits to nearly 200 million people.

As Social Security established a prominent position helping ensure economic security for Americans, the passage of seven decades has brought substantial and unanticipated change, especially to the population the program was created to serve.

The number of older Americans living now is greater than anyone could have imagined in 1935. Then, only 7.5 million people were age 65 or older. Today, approximately 36 million, or roughly one in eight people, are older Americans.

These numbers are going to continue to grow even more rapidly in the coming decades. In less than three years, America's 78 million baby boomers will begin to reach retirement age. By the middle of this century, about one of every five Americans will be 65 or older. This increase in life expectancy is a wonderful success story for our nation. More and more Americans are working longer and enjoying a lengthy retirement. But increases in life expectancy mean challenges for Social Security.

The Social Security program is largely a pay-as-you-go system -- with today's workers paying for today's beneficiaries. This system has worked well over the years -- especially when there was a relatively large number of workers to support each individual receiving benefits. But today's demographics are working against us. Our parents, grandparents and great-grandparents can feel confident about the promise of a secure future. Their benefits are secure and will be paid.

Unfortunately, the same cannot be said for my teenage son and his friends. I believe Social Security's 70th anniversary is the perfect opportunity for us to signal to younger generations of Americans that we, as a society, are committed to strengthening this important program -- for them.

In today's rapidly developing world, it's no surprise that government programs also will need to adjust to our changing circumstances.

Under President Bush's leadership, this issue is being discussed on Capitol Hill and in living rooms across the country.Looking ahead, the financing problems facing Social Security, coupled with the program's complexity and scope, will be challenging to address. Reflecting back, our nation has a proud history of grappling with difficult issues. And we do it best when we work together. I believe Social Security -- a program that touches the lives of almost every American -- deserves nothing less.

Jo Anne B. Barnhart is the Commissioner of Social Security.

This message was published in the Orlando Sentinel August 11, 2005.

back to the top.

University of Michigan
Institute for Social Research
426 Thompson Street
3026 ISR
Ann Arbor, MI 48106-1248

Phone: 734-615-0422
Fax: 734-615-2180
Email: mailto:mrrcumich@umich.edu