Working Paper: NBER ID: w17505
Authors: Vasily Kartashov; Raimond Maurer; Olivia S. Mitchell; Ralph Rogalla
Abstract: This paper assesses the impact of variable investment-linked deferred annuities (VILDAs) on lifecycle consumption, saving, and portfolio allocation patterns given stochastic and systematic mortality. Insurers have taken two approaches to manage systematic mortality risks, namely self-insurance and risk transfer to purchasers of the annuity products. We demonstrate that self-insurance leads to high loadings, so that households offered a choice would favor the risk transfer scheme. Reservation loadings on the actuarially fair VILDA price for non-participation are 0.5-8%; if insurers cannot hedge within this range, they will transfer systematic longevity risks to the annuitants. Our findings have implications for new payout products that may be attractive to older households seeking to protect against retirement shortfalls.
Keywords: longevity risk; deferred annuities; portfolio choice; lifecycle consumption
JEL Codes: G11; G22; G23; H55; I3
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| self-insurance (G52) | less favorable outcomes for consumers (D18) |
| non-participation in VILDAs (J17) | significant cost to households (D19) |
| systematic longevity risk (G52) | alters optimal lifecycle behavior (D15) |
| VILDAs (Y90) | household financial behavior (D14) |
| access to VILDAs (Y50) | higher consumption levels throughout lifetimes (D15) |
| VILDAs (Y90) | enhance consumption opportunities (D12) |
| VILDAs (Y90) | consumption increases at age 80 (D15) |
| VILDAs (Y90) | stable consumption benefits across outcomes (E21) |