Taking off the wrapper: The Hidden Dynamic of Variable Annuity Funds

by Vijay Yadav , 26.10.16

Vijay Yadav, Professor of Finance at ESSEC Business School, shares new research providing convincing evidence that variable annuity affiliated funds can outperform, by sizeable margins, pure open-end funds.

From the paper "Better than Expected: Hidden Dynamic of Variable Annuity Funds" (V. Yadav, M. Massa), Review of Finance, Jan 2016

The financial crisis and the demise of AIG brought to the fore the role of insurance companies in financial markets and their interactions with other financial players. Still today, insurance companies play a major role in the financial system, not only as investors and asset managers, but also as providers of intermediation between asset managers and investors.

Variable annuity funds (VAF), for example, are tax-deferred retirement vehicles whereby individual investors sign a contract with an insurance company. These contracts typically have two stages:

An accumulation stage during which the annuitant makes regular purchase payments to the insurance company, who in turn offers to the investor a menu of investment options. These subaccounts are managed by various fund management firms.
A distribution stage during which the investor receives regular payments from the insurance company. The amount of money distributed depends on the amount of purchase payments and the performance of subaccounts in which the money was invested.

Taking off the wrapper

A typical variable annuity wrapper – in other words, the all-in-one package sold by a given insurance company – may offer around 50 subaccounts to choose from. And while investors are strongly discouraged from withdrawing money from the wrapper by high withdrawal fees and tax penalties, they can reallocate money within the wrapper from one investment option to another at almost no cost. Subsequently, poorly performing subaccounts are punished by investors who can easily withdraw money and reallocate it to subaccounts with superior recent performance.

What does this mean for investors? First, it reduces search costs for investors and facilitates comparison across the subaccounts offered within the same wrapper. But perhaps more importantly, it increases competition between the fund management firms offering these subaccounts, who must now work to retain investor’s money.

Variable annuity funds: subject to bargaining

So, how are variable annuity funds affected by the balance of bargaining power between insurance companies and fund management companies? Through an analysis of the complete sample of variable annuities within the US over the period 2001-2011, our research finds that the flow-performance sensitivity for variable annuity funds is higher than that for pure open end funds (OEFs), suggesting that investors react more to performance for VAFs than for OEFs. In other words, VAFs get rewarded more generously by the investors for good performance in terms of new inflows. This phenomenon creates strong incentives for fund management companies to offer their best funds in variable annuities.

Indeed, we find that the variable annuity funds deliver higher performance than otherwise similar open-end mutual funds. We measure risk-adjusted fund performance using many alternative definitions of performance. Pure VAFs outperform pure OEFs by 66 basis points (bps) per year in the case of market-adjusted return, 65 bps per year in the case of CAPM alpha, 67 bps per year in the case of 3-factor alpha and 70 bps per year in the case of 4-factor alpha. This effect is economically very sizable in the mutual fund industry where the average market- adjusted fund performance is close to 0. We also compare the performance difference between VAFs and OEFs belonging to the same family and find that VAFs outperform OEFs belonging to same family by around 60 bps per year in the case of 4-factor alpha.

Variable Annuity Funds: Outstanding

While our analysis is mainly based on the overall set of US equity funds, we also confirm the results in other major fund categories: Balanced Funds, International Stock Funds, and Sector Stock Funds. We show that, regardless of the investment style, there is a strong positive relationship between fund performance and VA-affiliation.

The results provide convincing evidence that the performance of VA-affiliated funds is way better than that of pure OEFs. We explain this performance difference in terms of self-selection: only the better OEFs are chosen to be part of insurance wrappers. We find that the OEFs that have been added to insurance wrappers had much better performance than the other OEFs in the three years period prior to being added in the wrapper. Moreover, these funds continue to perform significantly better than the pure open end funds following the transition. For example, in terms of 4-factor alpha, these funds had outperformed other pure OEFs by 18.3 bps per month (2.20% per year) in the three year period before transition and they outperformed pure OEFs by 6.8 bps per month (0.82% per year) after the transition.