New papers from the Social Science Research Network

"Phantoms Never Die: Living with Unreliable Population Data"

ANDREW J. G. CAIRNS, Heriot-Watt University - Department of Actuarial Science & Statistics
Email: a.cairns@ma.hw.ac.uk
DAVID P. BLAKE, City University London - Cass Business School - The Pensions Institute
Email: d.blake@city.ac.uk
KEVIN DOWD, City University London - Sir John Cass Business School
Email: kevin.dowd@hotmail.co.uk
AMY R. KESSLER, Prudential Retirement
Email: akessler@bear.com

The analysis of national mortality trends is critically dependent on the quality of the population, exposures and deaths data that underpin death rates. We develop a framework that allows us to assess data reliability and identify anomalies, illustrated, by way of example, using England & Wales (EW) population data. First, we propose a set of graphical diagnostics that help to pinpoint anomalies. Second, we develop a simple Bayesian model that allows us to quantify objectively the size of any anomalies. Two-dimensional graphical diagnostics and modelling techniques are shown to improve significantly our ability to identify and quantify anomalies. An important conclusion is that significant anomalies in population data can often be linked to uneven patterns of births in cohorts born in the distant past. In the case of EW, errors of more than 9% in the estimated size of some birth cohorts can be attributed to an uneven pattern of births. We propose methods that can use births data to improve estimates of the underlying population exposures.
Finally, we consider the impact of anomalies on mortality forecasts and annuity values, and find significant impacts for some cohorts. Our methodology has general applicability to other population data sources, such as the Human Mortality Database.

"How Does the Probability of a 'Successful' Retirement Differ Between Participants in Final-Average Defined Benefit Plans and Voluntary Enrollment 401(k) Plans?"
EBRI Notes, Vol. 36, No. 10 (October 2015)

JACK VANDERHEI, Employee Benefit Research Institute (EBRI)
Email: vanderhei@ebri.org

This paper begins with a review of the previous academic literature and summarizes previous Employee Benefit Research Institute (EBRI) research analyzing the conditions under which voluntary-enrollment (VE) 401(k) plans are likely to provide an accumulation of retirement assets at least equivalent to those provided under a counterfactual final-average defined benefit (DB) plan. New research is then presented to show the percentage of “successful” retirements by income quartile for workers currently ages 25-29 who will have more than 30 years of simulated eligibility for participation in a 401(k) plan. Results are first presented for both voluntary-enrollment 401(k) plans and final-average DB plans with a 1.5 percent accrual rate. Sensitivity analysis is provided by also analyzing the comparative success rates of final-average DB plans with accrual rates of 1.0 and 2.0 percent. Using baseline assumptions (defined in the study), it appears that the DB plan has a higher probability of achieving a real replacement rate (when combined with Social Security payments) of 60 percent than the VE 401(k) plans for the first three income quartiles. If a 70 percent replacement rate is used as a threshold, participants in the third- and fourth-income quartiles have a much higher probability of success with the 401(k) plans than the DB plans. When the threshold is set at a higher (and according to many financial planners, more realistic) replacement rate of 80 percent, the 401(k) plans have a much higher probability of success than the counterfactual DB plans for all groups except for the lowest-income quartile (where the results are virtually even).

"The Impact of Temporary Assistance Programs on the Social Security Claiming Age"

GEOFFREY SANZENBACHER, Boston College Economics Department
Email: geoffrey.sanzenbacher.1@bc.edu
APRIL YANYUAN WU, Mathematica Policy Research, Inc.
Email: wuuv@bc.edu
MATTHEW S. RUTLEDGE, Boston College, Center for Retirement Research
Email: rutledma@umich.edu

Delaying claiming past the early eligibility age of 62 has taken on increased importance. Individuals turning 62 with no job and limited income may be able to use temporary assistance programs such as Unemployment Insurance (UI), Medicaid, and the Supplemental Nutrition Assistance Program (SNAP) as sources of support prior to collecting Social Security benefits. To what extent do these programs allow recipients to delay Social Security claiming? The challenge in answering this question stems from the fact that program users’ dire economic straits may make them more likely to claim benefits from both Social Security and these programs, generating a misleading correlation between Social Security claiming and temporary assistance benefits. This paper constructs instruments for program generosity that vary with an individual’s state of residence but should not reflect the characteristics or circumstances of the individual.