Where Less Is More: Reducing Variable Annuity Fees to Benefit Policyholder and Insurer

• Published date: 01 September 2019
• Date: 01 September 2019
• DOI: http://doi.org/10.1111/jori.12237
• Author: Carole Bernard,Thorsten Moenig

©2018 The Journal of Risk and Insurance. Vol.XX, No. XX, 1–22 (2018).

DOI: 10.1111/jori.12237

Where Less Is More: Reducing Variable Annuity

Fees to Benefit Policyholder and Insurer

Carole Bernard

Thorsten Moenig

Abstract

In the United States, variable annuities (VAs) are popular long-term per-

sonal investment vehicles. Recently, however, sales have begun to dwindle.

In fact, financial advisers have long argued against investing in VAs due

to the products’ high fees. VA providers charge these fees—typically at a

constant rate throughout the policy period—to cover their expenses and the

costs of embedded guarantees, and lowering this constant fee rate could

make the VAunprofitable. Instead, we propose and analyze a simple change

to the fee “structure” that would lower fee rates overall (and thus make the

product more attractive to investors) without reducing the insurer’s profit.

The key insight is that this time-dependent fee rate (with moderate front-

loading) implicitly discourages policy lapses and exchanges, which reduces

the providers’ policy acquisition expenses. Taking into account financially

optimal lapse (and reentry) decisions, we determine the optimal timing and

rate of the fee reduction for a competitive as well as for an innovative VA

provider. An important characteristic of this feature is that it can be imple-

mented easily and effectively to both new and existing VApolicies.

Introduction

In the United States, variable annuities (VAs)combine the investment features of mu-

tual funds with favorable tax treatment and return guarantees (Hardy, 2003). Over

Carole Bernard is at the Department of Accounting, Law and Finance, Grenoble Ecole de Man-

agement, France, and the Faculty of Economics and Political Science, Vrije Universiteit Brus-

sel (VUB), Belgium. Bernard can be contacted via e-mail: carole.bernard@grenoble-em.com.

Thorsten Moenig is at the Department of Risk, Insurance, and Healthcare Management, Tem-

ple University,Philadelphia, PA. Moenig can be contacted via e-mail: moenig@temple.edu. The

authors acknowledge financial support from a 2015 research grant Ignacio H. Larramendi by

Fundaci´

on Mapfre in the area of insurance and social protection (Grant SP/15/BIL/009). We

would like to thank the editor, an associate editor, and two anonymous referees for helpful

comments on an earlier version of the article. We also thank Guillem Montoliu, Ralph Rogalla,

Adam Shao, Juan Zhang, and participants of the seminar series of the University of Montreal,

the 2017 IRFRC Annual Conference in Singapore,the 52nd Actuarial Research Conference in At-

lanta, the 2017 ARIA meetings in Toronto,the 2017 ASTIN/AFIR/ERM Colloquium in Panama

City,and the 44th EGRIE seminar in London, for their feedback and suggestions. All remaining

errors are ours.

1

761

Vol. 86, No. 3, 761–782 (2019).

2The Journal of Risk and Insurance

the past two decades, they have developed into popular long-term investment vehi-

cles, with providers holding on to around $1.9 trillion in net assets at the end of 2016.

However, VA sales have recently begun to dwindle, and since 2013 the industry is

experiencing (increasingly) negative net sales.1In fact, many financial advisers have

long tried to steer consumers away from VAs, largely due to the high fees associated

with the product (see, e.g., NASDAQ, 2009; Kiplinger, 2011; Wall Street Journal, 2012;

Forbes, 2015, among many others).2Therefore, reducingproduct fees would make VAs

more attractive to investors and would likely counteract the recent decline in demand.

In this article, we propose that a change in the fee “structure”—specifically, through a

time-dependent (i.e., moderately front-loaded) fee rate—could accomplish that while

at the same time increasing the products’ profitability to providers.

Currently, VA providers charge a “time-invariant” base fee (formally, the “mortality

and expense risk charge” and the “administration charge”) in order to recover their

expenses over the duration of the policy.The high level of this fee is a consequence of

the largepolicy acquisition expenses (e.g., commissions), in combination with frequent

policy lapses (Paris, 2017), which give providers less time to recover their up-front

expenses (Pinquet, Guillen, and Ayuso,2011; Moenig and Zhu, 2016). We demonstrate

that a partial front-loading of the fees can potentially reduce lapses by rewarding and

encouraging long-term consumer participation in the policy. In turn, this allows the

VA provider to reduce the fee rate across all policy years.

Specifically, we implement a VA with a basic death benefit guarantee and a binary

fee structure based on a one-time fee reduction. Under our model specifications,

this simple change eliminates virtually all lapse incentives of a financially rational

policyholder—without increasing the fee rate above the current level in any policy

year. In fact, in a competitive VA market, our proposed fee structure lowers the in-

surer’s average annual expenses—and thus the fee rates—by up to 35 percent (i.e., 8

percent of the single premium), under our model specifications. Moreover, we find

that an innovative VA provider willing to deviate from the current status quo may

gain a surplus of around 6 percent of the purchase premium. That is, our proposed

time-dependent fee structure can make both policyholder and insurer simultaneously

better off.

The policyholder’s “financially rational” incentive to lapse stems from the fact that

the death benefit guarantee commonly embedded in VAs resembles a (conditional,

long-term) put option on the VA account value and therefore loses in value if the

account value increases. As the policyholder continues to pay fees for this now over-

valued guarantee, he would benefit from lapsing the current VA policy and imme-

diately “reentering” the market by purchasing an otherwise identical product. This

would increase the guaranteed amount to the current VA account value, without

changing any of the other contract parameters. Moenig and Zhu (2016) show that this

1That is, investors are surrendering policies faster than new money is coming into the market.

Source: Insured Retirement Institute.

2Research has shown that the tax benefits of VAscan outweigh these fees, but only in the long

term (Milevsky and Panyagometh, 2001; Moenig and Bauer, 2016; Moenig and Zhu, 2016).

2The Journal of Risk and Insurance

762