Funded Research Paper Abstracts

Printer Friendly Version

Health Paper Abstracts

Household Demand for Health Insurance Price and Spouse's Coverage
Marjorie Honig, 2004 • Full Text

Demand for employment-based health insurance has traditionally been treated as an individual rather than a household decision. Dual-earner households are now the typical U.S. married household, however, and most firms offer family coverage as one of the options available to employees. Findings from a model that jointly estimates married workers’ take-up of their own employer-based insurance with both their own and their spouses’ insurance offers indicate that both own insurance price and opportunities for coverage under spouses’ employer-based plans are statistically important determinants of insurance take-up in dual-earner households.

Relative take-up elasticities with respect to own price and spouse’s offer indicate that potential coverage by spouses plays the larger role for both husbands and wives in the decision of whether to take their own coverage. We find evidence of selection into jobs offering health insurance among wives in dual-earner households, but not among husbands. Our findings also suggest that dual-earners may not be aware of the potential trade-off between wages and health benefits. Data are taken from the 1996 panel of the Survey of Income and Program Participation.

Sinking with the Ship: How Does Involuntary Job Displacement Affect Health Insurance?
Kosali Ilayperuma Simon, 2005 • Full Text

Numerous job cuts during the 2001 recession and the subsequent sluggish recovery have no doubt added to the growing number of Americans without health insurance, yet little evidence exists on how involuntary job displacement is causally related to health insurance status. Prior research indicates that workers who are about to be displaced earn less than their non displaced counterparts, and that wage losses associated with displacement actually start prior to the date of separation. In this paper, I use panel and cross sectional data to test whether workers about to be displaced differ in their health insurance coverage from their employer relative to similar workers who are not displaced, and if so, how far back in time these health insurance losses occur. I also test the extent to which other sources of health insurance cushion the effects of displacement on health insurance. I find that workers who are subsequently displaced from jobs are less likely to have own-employer provided health insurance than otherwise similar workers who are not displaced, and that this difference starts up to two years prior to displacement. Other sources of health insurance play an important role in protecting the health insurance coverage of these workers, even in the pre-displacement period.

Post-displacement Health Insurance Coverage of Displaced Workers
Kosali Ilayperuma Simon, 2005 • Full Text

Workers who experience job displacement are especially vulnerable to uninsurance in a job-based health insurance system. Studies of the wage experience of these workers show persistent losses that continue well into the new job. The wage literature also highlights the role of spousal labor supply and the unemployment insurance system in protecting these workers. In this paper, I first test whether displaced workers are ‘permanently scarred’ by job loss in terms of their health insurance coverage. In so doing, I also investigate the effect of alternative sources of health insurance in protecting displaced workers against uninsurance once re-employed. I next investigate the extent to which spousal health insurance acts as a subsidy during the job search process.

Once re-employed, former displaced workers fare worse than other new workers who voluntarily left jobs in terms of finding own-employer health insurance on the new job, although all workers experience some gains in health insurance with time on the new job. The relative loss of health insurance for displaced workers after re-employment is correlated with demographic factors (such as gender and marital status) and job characteristics (such as hours worked). In terms of the job search process, I find evidence that workers with access to spousal health insurance have longer job search periods after displacement, but do not necessarily find better paying jobs.

The Dynamics of Health Insurance Coverage: Factors Correlated with Insurance Gain and Loss among Adults
Robert W. Fairlie and Rebecca A. London, 2005 • Full Text

In this study, we examine annual transitions into and out of health insurance coverage using matched data from the 1996 to 2004 Annual Demographic Files (ADF) of the Current Population Survey (CPS). Although the CPS ADFs have primarily been used as cross-sectional samples, we create a two-year panel by linking consecutive surveys. The large sample sizes and longitudinally matched CPS data allow us to explore the relationship between changes in detailed employment characteristics and health insurance transitions over a two-year period. To our knowledge, the matched CPS data have not been previously used to explore the dynamics of health insurance coverage. The CPS measures health insurance coverage over the entire year prior to the survey, capturing movement between year-long uninsurance and part- or full-year insurance over a two-year period. Our study therefore focuses on transitions between relatively long spells of uninsurance (at least one year) and any length spell of coverage. This allows us to focus specifically on intermittent insurance coverage that leads to longer spells of uninsurance, spells that are the most likely to result in adverse health or financial outcomes.

We address several research questions using the two-year CPS panel. First, we examine differences in the incidence of health insurance transitions across detailed demographic and employment characteristics. The focus is on identifying whether low rates of health insurance among certain groups, such as minorities, less-educated workers, part-time workers, and workers at small employers, are due to high rates of health insurance loss, low rates of obtaining health insurance, or both. Second, we examine the incidence of health insurance transitions between public and private coverage, and between each of these and uninsurance. Third, we examine whether dynamic factors, such as job loss, movement from full-time to part-time work, movement from a large employer to a small employer and other changes in job characteristics are associated with health insurance loss. We also explore whether changes in employment and job characteristics are associated with gaining health insurance. We then examine dynamic factors that are related specifically to the loss and gain of private health insurance. Although it is difficult to identify causal factors of health insurance transitions, the analysis of the relationship between changes in health insurance coverage and changes in potentially correlated factors using the large two-year panel data in the CPS improves on cross-sectional analyses and offers some of the first estimates of the relationship between changes in employment characteristics on dynamic health insurance outcomes.

Hospital Ownership Changes and Employers’ Decision to Offer Insurance
Eric E. Seiber, 2006 • Full Text

This study uses 1995 - 1999 Contingent Workers Supplement and hospital ownership data from the American Hospital Association’s Annual Surveys to how the employer sponsored insurance market responds to changes in hospital ownership. A Nested Multinomial Logit model estimates a two dimensional choice set of availability of an employer based plan and participation in that plan. The model indicates that access to a public hospital system produces a negligible effect on insurance availability and participation for all but the lowest income workers, and the data showed no appreciable difference for not-for-profit and for-profit hospitals.

 Pension Paper Abstracts

Cross-Trading by ERISA Plan Managers
Thomas H. McInish, Ph.D., C.F.A., 2002 • Full Text

ERISA prohibits cross trades, the exchange of assets between two accounts without going through a public market. There have been numerous exemption requests motivated by a desire to reduce transaction costs (typically one to four percent). Mutual fund cross trades under Rule 17a-7 achieve economically significant savings. Transaction costs comprise commissions, market impact, and opportunity costs of missed trades. Further, round trip trades incur a bid-ask spread, which covers order processing costs (the normal expenses of providing liquidity) and asymmetric information costs (dealer losses to informed traders). Dealer quotes reflect asymmetric information costs and trade size. Trade prices exhibit regularities, including U-shaped patterns in returns and volume. Without a market trade, it is impossible to know what price each counterparty would have paid/received. If both parties are equally motivated and seek to trade at the same time, it makes sense to cross at the spread midpoint. But if one party typically uses patient trading strategies or is accommodating the counterparty, determination of a fair crossing price is difficult. If the goal is to minimize risk, cross trading should be prohibited. In a cost-benefit context, steps such as having written implementation plans and strong monitoring can reduce risk of abuse.

How Much and In What Manner Should Americans Save?
Jagadeesh Gokhale and Laurence J. Kotlikoff, 2004 • Full Text

This study uses ESPlannerTM -- a life-cycle, financial planning model -- to investigate how much American households should save and to understand the best form in which to do so. ESPlanner’s saving and life insurance recommendations generate the smoothest possible survival-state contingent lifetime consumption path for the household without putting it into debt. Such consumption smoothing is predicted by economic theory and appears, in general, to accord with actual behavior. By running households through ESPlanner based on current policy as well as on alternative fiscal policies, one can easily compare the program's recommended consumption and saving response to hypothetical tax and transfer policy changes and assess the degree to which borrowing constraints (the inability to borrow significant amounts beyond one’s mortgage) may be playing a role in determining the size of those responses. The program can also indicate what method of saving, be it in 401(k) and other tax-deferred accounts, Roth IRAs, or in non-retirement accounts is most efficacious with respect to minimizing lifetime taxes and maximizing lifetime consumption.

The 964 households used in our analysis are drawn from the Federal Reserve's 1995 Survey of Consumer Finances. This data set provides detailed information on household earnings, assets, housing, demographics, and retirement plans -- all of which is used by ESPlanner in formulating its recommendations. The policies we consider are tax hikes, tax cuts, social security benefit cuts, and the elimination of tax-deferred saving. Our analysis distinguishes between immediate and future policy changes as well as between permanent and temporary ones.

Our results are strongly influenced by the fact that a majority—58 percent—of our sample of households, many of which are young, is borrowing-constrained and, thus, more responsive to current than future policy changes no matter how long their duration. Borrowing constraints refer to the inability to smooth one’s consumption without incurring additional non-mortgage debt. In running ESPlanner, we assume that households cannot borrow simply to smooth their living standards, as apart from buying a home. This ignores the ability of households to borrow relatively small sums on their credit cards. But including a relatively small credit card borrowing limit would make very little difference to our results. While we assume a zero-non mortgage debt limit in running our sample households through the program, the fact that 58 percent of the households can’t perfectly smooth their living standards without going into debt represents a finding, rather than an assumption, of our analysis.

Because so many of our sample households are borrowing constrained, their consumption and saving responses to policy changes are very sensitive to the particular policy being enacted. Income tax changes, for example, have little effect on the consumption/saving of low-income households for the simple reason their income tax liabilities are relatively small. And social security benefit cuts will have minor effects on the young because they lie so far in the future and the young are generally borrowing constrained. On the other hand, eliminating tax-deferred saving will have no effect on current retirees, but greatly influence the spending of the young, since such a policy would relax their borrowing constraints. We also show that a small segment of households would end up raising their lifetime taxes and reducing their lifetime consumption by contributing to tax-deferred retirement accounts. For these households switching to Roth IRAs appears to be advantageous.

Valuing Assets in Retirement Savings Accounts
James M. Poterba, 2004 • Full Text

Assets in retirement saving plans have become an important component of net worth for many households. While many studies compare household balances in tax-deferred retirement accounts such as 401(k) plans with the financial assets held outside these accounts, these different asset components are not directly comparable. Taxes and in some cases penalties are due when assets are withdrawn from some retirement saving plans. These factors can make a dollar held inside a retirement account less valuable than a dollar held in a similar asset outside these accounts, particularly for those who are considering withdrawing assets from the tax-deferred accounts in the near future. For younger households who do not plan to withdraw tax deferred assets for many years, the opportunity for tax-free compound returns in retirement accounts can make a dollar inside such an account more valuable than a dollar outside such accounts from the standpoint of providing retirement resources, even though the principal from the retirement account will be taxed at the time of distribution, while the principal outside such accounts is untaxed. This paper illustrates the potential differences in the value of a dollar of invested in a bond, or in corporate stock, inside and outside tax-deferred accounts. It draws on a range of data sources to calibrate the value of the tax burden, and the benefit of compound growth, for assets held in retirement accounts, and describes the differences in relative valuation for households of different ages.

Participation in Defined Contribution Plans
William E. Even and David Macpherson, 2005 • Full Text

During the 1990s, many pension plans shifted the responsibility for directing the investment of pension plan assets to the employee. This study examines the rapid growth of the participant directed pension plans using data from the Survey of Consumer Finances, the Survey of Income and Program Participation, and IRS Form 5500. Several relevant questions are addressed. First, what types of workers are most likely to be in a participant directed plan and what types of employers are most likely to offer such plans? Second, how does participant direction affect the allocation of assets and the risk/return performance of the pension? The study has two important findings. First, participant direction has a significant effect on asset allocation in pension plans, shifting pensions away from employer stock and towards other types of stock. Second, based on risk-adjusted rates of return, participant directed plans actually outperform employer directed plans.

The Causes and Consequences of Pension Fund Holding of Employer Stock
William Even and David Macpherson, 2004 • Full Text

This report examines the causes and consequences of investing pension funds in employer stock using a merger of data from Form 5500 pension filings and stock return data from the Center for Research on Security Prices (CRSP). Section II reviews the existing hypotheses and related empirical evidence on factors that lead pension funds to invest in employer stock. Results from the Capital Asset Pricing Model are employed in section III to derive a measure of the non-diversification costs of holding employer stock. Section IV provides a description of the data used in our study and an empirical analysis of factors influencing pension fund holdings is provided in section V. The effect of employer stock holdings on the risk and return of pension portfolios is examined in section VI along with projections of how legislated limits on employer stock holdings would affect the distribution of returns.

Defined Contribution Pension Plans and Retirement Wealth Adequacy
Gary V. Engelhardt, 2005 • Full Text

This project uses a newly developed defined contribution (DC) pension wealth calculator to generate new estimates of DC pension wealth for individuals in the Health and Retirement Study (HRS) with employer-provided pension Summary Plan Descriptions (SPD). There are four primary findings. First, pension wealth from voluntary saving (and accrued earnings thereon) comprises half of DC pension wealth calculated from the sample of matched SPDs in the HRS. Second, the DC/401(k) Calculator yields dramatically lower mean estimates of DC pension wealth for HRS participants than the Pension Estimation Program. In particular, DC pension wealth is calculated to be as much as 20-25 percent less when using an increase in the number of modeling assumptions offered to the user and arguably better input data, and wealth in 401(k)-type pension plans is implied to be as much as 40-50 percent less. Third, most of the reduction in estimated DC wealth occurs in the upper portion of the pension-wealth distribution. Fourth, the Pension Estimation Program actually understates DC wealth in the middle of the pension-wealth distribution. These results suggest that previous analyses that have used HRS pension wealth created from the matched SPD data have overstated retirement wealth adequacy among HRS participants.

New Trends in Pension Benefits and Retirement Provisions
Olivia S. Mitchell, 2000 • Full Text

Private sector pension plans have undergone substantial change in form and structure in the United States over the last two decades. This paper explores and evaluates these changes using information on pension plan characteristics gathered by the U.S. Department of Labor (DOL) since 1980 in their periodic Employee Benefits Survey (EBS) of medium and large establishments. We also discuss how future data collection efforts could be improved to better measure key changes in the form and design of employer-sponsored pensions.

Key findings are as follows: Many aspects of defined benefit plans changed over time. For example, vesting rules were loosened; plans eased access to normal retirement; and pension benefit formulas moved toward final rather than career earnings, with increased weight on straight-time pay. In addition, these plans became more integrated with social security; at the same time, the form of social security integration changed substantially. The evidence also indicates that defined benefit plan replacement rates fell over time and benefit caps limit years of service counted in the retirement formula. In addition, disability benefit provisions grew more stringent; and participants were increasingly permitted to take a lump sum from their defined benefit plan.

Defined contribution plans also have evolved over time. Here, plan participants were granted greater access to diversified stock and bond funds, and fewer were permitted to invest in own-employer stock, common stock funds, and guaranteed insurance contracts. Participation and vesting rules appear most lenient for workers in 401(k) plans; generally employees must contribute a fraction of their pay to their plans rather than relying only on employer contributions; and employee access to pension fund assets prior to retirement is growing.