یادگیری در حین انجام کار:طرح پس اندازهای بازنشستگی با حمایت مالی فعالیت کارفرما و انباشت ثروت خانوار

This paper investigates the impact of household exposure to employer pension plan features using the Health and Retirement Survey. We investigate whether exposure to active management (choice) or participation in plan-sponsored financial education seminars impacts household portfolio allocations and wealth. We consider interactions between pension design and investment patterns outside of workers’ pension plans, utilizing two parametric estimators: the random effects probit and the multivariate probit. We extend our results non-parametrically via propensity score matching. We find repeated evidence that both of the plan features improve asset allocations and financial outcomes for recent retirees, especially when used together.

If portfolio composition and wealth change as households gain experience with financial markets then it follows that the evolution of employer-based Defined Contribution (DC) retirement savings plans is related to the evolution of household savings and wealth. In this paper we consider two DC plan features: (i) investment choice and (ii) financial education. We find that that the ability to choose investments increases a household's propensity to hold wealth in relatively risky assets such as equities, after controlling for other factors such as preferences for risk and planning horizons. This is good news for most cohorts of recent retirees given observed equity premiums, however increasing equity holdings will not always be optimal. For this reason we also consider employer-sponsored financial education offerings and find attendance linked to greater household wealth at retirement. Both plan features: choice and education are thus observed to be valuable for enhancing private savings and capital formation outside of the employer plan setting. Because household savings are linked to both the nation's wealth and its population's social insurance needs, we encourage the reader to consider this study of employer and household habits in terms of both private and public pension and savings policy.

Saving for retirement has traditionally been couched in terms of a three legged stool: private savings, public and employer pensions. This paper has focused on two of these legs, employer plans and private savings – specifically the impact of pension design on the composition and level of private non-pension savings. Results suggest that both features are useful in improving wealth accumulation outcomes at retirement. The work here highlights interactions. While choice is read to be more important for wealth outcomes, as Fig. 1.C suggests, we interpret this to be a result of skew in the distribution of impacts. Education is interpreted as more important in balancing portfolios across asset classes, and reducing variation in outcomes when combined with choice. Further education as a treatment may vary considerably from case to case and education-related improvements may be remedial, affecting lower portions of wealth and earnings distributions—as in Lusardi (2004). Our multivariate probit results suggest a main driver in education-related improvements is the establishment of IRA accounts, important for preserving DC balances during employer changes (and at retirement). Much of the parametric and semi-parametric work adds nuance to the basic patterns reported in Table 1 and Fig. 1.Regarding public retirement programs, the impact of plan design on investment choices and wealth outcomes has implications for proposals aimed at creating individual accounts for workers. Individual account structures shift risk to participants. It appears that those with exposure to DC plans, choice and education are likely to better manage any reform of this sort. It follows that we recommend financial education be added to any future proposal, with some particular attention focused on those without previous exposure to financial markets.22 Research by financial economists on portfolio allocation has underlined the disparity between theoretical lifecycle models of portfolio choice, which call for a high level of equity participation, and actual empirical data that indicate limited stock market participation. The view advanced in this paper is that the theory and empirics can be reconciled if we relax informational assumptions imposed on agents in the classical model. We contend that low levels of savings and under-diversified portfolios are correlated rather strongly with inexperience. Accordingly, employer programs which contribute to participation and to understanding are found to have positive impacts for savings outside of it.