OLIVIA S. MITCHELL, University of Pennsylvania - Insurance & Risk Management Department, National Bureau of Economic Research (NBER)
Email: mitchelo@wharton.upenn.edu
GARY R. MOTTOLA, Vanguard Center for Retirement Research
Email: gmottola@vanguard.com
STEPHEN P. UTKUS, Vanguard Center for Retirement Research
Email: steve_utkus@vanguard.com
TAKESHI YAMAGUCHI, University of Pennsylvania - The Wharton School
Email: tyamaguc@wharton.upenn.edu

Important behavioral factors such as default and framing effects are increasingly being employed to optimize decision-making in a variety of settings, including individually-directed retirement plans. Yet such approaches may have unintended "spillover" effects, as we show with regard to the introduction of lifecycle funds in U.S. 401(k) plans. As anticipated, lifecycle funds do reshape individual portfolio choices through large default and framing effects. But unexpectedly, they also create a new class of investors which holds these funds as part of more complex portfolios. Our results are directly relevant to those interested in retirement plan design and retirement security; they also highlight the importance of assessing such spillover effects in other consequential settings where behavioral economics techniques may be employed.

Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

GORDON L. CLARK, Oxford University Center for the Environment
Email: gordon.clark@ouce.ox.ac.uk
ROBERTO DURÁN-FERNÁNDEZ, University of Oxford
Email: roberto.duran-fernandez@geog.ox.ac.uk
KENDRA STRAUSS, affiliation not provided to SSRN
Email: kendra.strauss@geog.ox.ac.uk

It is widely observed that being in the market gives financial traders access to knowledge and information not available to remote traders. A truism of the geography of finance, it is also a perspective that can shed light on the interaction between market location, global financial movements, and personal welfare. In this paper, we develop an explanation of the premium attached to being in the market, drawing upon previous contributions on the relevance of tacit knowledge and the insights provided by behavioural finance with respect to time-space myopia. To illustrate our model of 4 types of behaviour, mixing together various combinations of time and space conceptions of market performance, we analyse the intended retirement investment portfolios of nearly 2400 participants in a defined contribution pension plan sponsored by a London-based investment bank. Having demonstrated the empirical significance of the UK house-price bubble, respondents' retirement investment portfolios are analysed focusing upon the relative significance of property in relation to a range of other investment instruments. It is shown that, amongst similarly located respondents, there was a range of investment strategies dependent, in part, upon respondents' age, household status, job classification, and income. These results allow us to distinguish between different types of behaviour even amongst well-placed respondents, providing evidence of the co-existence of sophisticated, naive, and opportunistic investors against the base-case of time-space myopic behaviour. Implications are drawn for conceptualising a rapprochement between the insights of the behavioural revolution for economic geography (and in particular, the geography of finance) relevant for public policy.

JACK VANDERHEI, Employee Benefit Research Institute (EBRI), Temple University - Risk Management & Insurance & Actuarial Science
Email: vanderhei@ebri.org

As part of EBRI’s 2008 analysis of the likely impact of the Pension Protection Act’s safe harbor automatic enrollment and automatic escalation provisions, we developed a stochastic simulation model to project future 401(k) balances as a function of various plan design variables as well as assumptions with respect to various employee behavioral responses. In this paper I report on the results I obtained using the EBRI simulation model to determine how target-date funds (TDFs) would likely impact 401(k) participants assumed to be automatically enrolled. I realize that TDF use in 401(k) plans is not limited to those automatically enrolled; however, based on our simulation results, it appears that this 401(k) auto-enrollment will represent the majority of TDF use in the future and hence I will concentrate my analysis on those results. Results are reported both at the time of retirement as well as at the time of job change for those who are assumed to cash out. Several scenarios are presented in terms of alternative rates of return as well as several different types of target date funds.

CRAIG COPELAND, Employee Benefit Research Institute (EBRI)
Email: COPELAND@EBRI.ORG

This paper assesses the current status of Americans’ savings for retirement by examining the incidence of individual account plans among families, as well as the average amount of assets accumulated in these accounts. The 2007 Survey of Consumer Finances (SCF), the Federal Reserve Board’s triennial survey of wealth, is the basis for this study, as it is a leading source of data on Americans’ wealth, provides detailed information on retirement plan incidence and account balances among families, and is the latest available. While 2007 SCF is the most comprehensive and current survey of Americans’ finances, its timing was unfortunate due to the significant downturn in the economy in 2008 just after the survey was released. To account for that change, this analysis provides estimates of the changes in asset values from the end of 2007 to mid-June 2009 for individual account plan balances. The account balances of the defined contribution plans and IRAs are adjusted based on the asset allocation reported within the plans by using equity market returns and bond market returns from January 1, 2008, to June 19, 2009. Among all families with a defined contribution plan in 2007, the median (mid-point) plan balance was $31,800, up 16 percent from 2004. According to EBRI estimates, this dropped 16.4 percent (to $26,578) from year-end 2007 to mid-June 2009. Losses were higher for families with more than $100,000 a year in income (down 22 percent) or having a net worth in the top 10 percent (down 28 percent). Among all families with an IRA/Keogh plan, the median value of their plan was $34,000 in 2007, up 3 percent from 2004. EBRI estimates this median value dropped 15 percent (to $28,955) from year-end 2007 to mid-June 2009.

In 2007, 40.6 percent of families included a participant in an employment-based retirement plan (either a defined benefit or defined contribution plan) from a current job. This was up from 38.8 percent in 1992, but virtually unchanged from 40.3 percent in 2004. A significant shift in the plan type occurred from 1992 to 2007, with the percentage of families with a plan having only a defined benefit plan decreasing from 40.0 percent to 17.4 percent. In 2007, 66.2 percent of families had a participant in a current or previous employer’s retirement plan or an IRA/Keogh, up slightly from 2004 (65.4 percent). The percentage of families that owned either an individual retirement account or a Keogh plan increased in 2007 to 30.6 percent from 29.1 percent in 2004. While regular IRAs account for the largest percentage of IRA ownership, rollover IRAs had a larger share of assets than regular IRAs in 2007. The increase in IRA wealth is expected to continue in the future, as more workers will be in defined contribution plans and will be in them for a longer period of their working lives.

NING TANG, University of Pennsylvania - The Wharton School
Email: Ntang@wharton.upenn.edu
OLIVIA S. MITCHELL, University of Pennsylvania - Insurance & Risk Management Department, National Bureau of Economic Research (NBER)
Email: mitchelo@wharton.upenn.edu
GARY R. MOTTOLA, Vanguard Center for Retirement Research
Email: gmottola@vanguard.com
STEPHEN P. UTKUS, Vanguard Center for Retirement Research
Email: steve_utkus@vanguard.com

Though millions of US workers have 401(k) plans, few studies evaluate participant investment performance. Using data on over 1,000 401(k) plans and their participants, we identify key portfolio investment inefficiencies and attribute them to offered investment menus versus individual portfolio choices. We show that the vast majority of 401(k) plans offers reasonable investment menus. Nevertheless, participants “undo” the efficient menu and make substantial mistakes: in a 20-year career it will reduce retirement wealth by one-fifth, in fact, more than what a naive allocation strategy would yield. We outline implications for plan sponsors and participants seeking to enhance portfolio efficiency: don’t just offer or choose more funds, but help people invest smarter.

GARY BURTLESS, Brookings Institution, Retirement Research Center, Boston College
Email: GBURTLESS@BROOK.EDU

The recent financial crisis and historical record suggest important lessons about the design of national pension systems. First, wide fluctuation in asset returns makes it hard for well-informed savers to select a saving rate or a sensible investment strategy for DC pensions. Workers who follow identical investment strategies but who retire a few years apart can receive DC pensions that are startlingly unequal. Second, it is hard for ordinary workers, as opposed to optimal planners, to make sensible choices about portfolio allocation. Their investment errors mean that actual returns fall short of the theoretical returns that could be earned by a well-informed, disciplined investor.