Solving the yield/risk conundrum facing retirees
As millions of individuals approach retirement age, they face the challenges of saving enough money for their retirement while maximizing yield and minimizing risk.
Historically, fixed indexed annuities (FIA) have provided an important solution to solve the yield/risk conundrum facing retirees and near retirees. Recent innovations in product design, including the addition of living benefit features and customized indexes, has led to very strong sales growth within the FIA market.
This has broadened the appeal of the traditional FIA among conservative annuity purchasers who prefer to limit their downside risk of loss, while receiving some upside equity market potential with lower overall volatility. The addition of
living benefit (LB) guarantees provides a competitive alternative to variable annuities sold with LB guarantees. Low interest rates have also made FIAs more attractive relative to traditional fixed annuities, government bonds, and certificates of deposit. In today’s marketplace roughly three quarters of FIA contracts are issued with guaranteed withdrawal benefit (GLWB) riders, a trend that produces another source of fee revenue, but also increases the risk profile of the
product. Interestingly, this trend comes at a time when some variable annuity (VA) writers are shifting away from living benefit guarantees and moving the customer’s focus back on the pure tax deferral aspect of the product. These refreshed VAs, known as investment only VAs, are being issued with no living benefit features at all, but offer many more investment options as a sales inducement.
Given that living benefit FIAs have only been sold for a few years, it is still too early to tell how the industry’s risk appetite and the product designs will evolve over the years. However, with that said, FIAs with living benefits are a leading product in today’s marketplace and are likely to remain so for the foreseeable future.
Those insurers who currently offer FIAs like the risk/reward profile of the product which allows them to more efficiently manage equity market risk through annual changes to caps and participation rates, in addition to repricing crediting rates based on changing portfolio yields. This risk profile is generally consistent with A.M. Best’s product creditworthiness scale,
which classifies basic FIAs at the lowest end of high risk products within in its product risk continuum. We’ll see later that some of the more advanced features offered by some insurers has increased the risk profile of the product.
A.M. Best recently completed a survey of the top ten FIA writers, which represents roughly 70% of the overall FIA market, with the remaining market fragmented among forty other U.S. life companies selling FIA products in the U.S. today. Our survey focused on product design, risk profile, investment spreads and industry concerns. Product design features included a review of crediting strategies, living benefit designs, and surrender charges. Risk profile features including questions on guaranteed minimum interest rates, indexes options, hedging strategies and emerging areas of risk. Finally, additional questions were asked regarding sales, distribution systems, and investment strategies.
The FIA market remains highly concentrated. However, as shown in Exhibit 1, Allianz Life Insurance Company of North America remains the leader among top ten insurers with roughly one-third of total sales, despite a significant increase in competition from newer entrants in recent years. Interestingly, despite two decades of sales of the FIA product, many large companies still do not offer an FIA product as part of their product suite.
A.M. Best believes that historically these companies have avoided selling the product over concerns regarding suitability
of the product given its historically high surrender charge and long surrender charge period.Distribution systems have largely remained unchanged since the product was first introduced as the product continues to be sold by independent marketing agents (IMOs), banks and broker dealers. This adds some element of uncertainty around which carrier’s products are sold and/ or lapsed. The market remains concentrated despite newer entrants such as Security Benefit Life Insurance and Athene Holding Limited in recent years.
Product Features – Crediting Strategy:
Respondents were asked to indicate the percentage of new premiums that had living benefit riders in 2014. Results are divided into five bands as follows: 0 to 20% of sales, 21 to 40 %, 41 to 60%, 61 to 80% and 81 to 100% of sales.
While the average sales with living benefit guarantees has remained at 70% in recent years, we note that, increasingly, some carriers are selling a much higher percentages of “with guarantee” products, and that 30% of companies are selling the with guarantee version at 81-100% of current year sales.Nine of the top ten companies offered a living benefit (LB) rider for new sales, with a majority of the companies reporting that at least 40% of their new business included a living benefits rider. Not surprisingly, there was a correlation sales market rankings for new sales with leading companies generally reporting a higher percentage of new sales in products with an LB guarantee. Another question asked about the company’s crediting strategy’s within FIAs for 2014 sales, using the following six categories:
- annual point to point with no cap,
- annual point to point with a cap/participation rate,
- monthly point to point with no cap,
- monthly point to point with a cap/participation rate or daily average with a cap or participation rate, and other.
The most common crediting rate strategy for 2014 (excluding customized or “other” strategies) was the annual point-to point with cap/participation rates, representing nearly a third of all FIAs issued. The next most common crediting rate strategy was themonthly point-to-point with cap/participation rate, which was slightly above 20%. A significant percentage was in the ‘Other’ category. In some cases, “Other” simply represented a multi-year strategy tied to one of the other
five strategies; however, some companies reported either a volatility- controlled or a fixed interest strategy. In addition to the
current sales crediting strategy we also asked respondents to indicate the aggregate crediting strategy for their in-force book, which was which was largely consistent with 2014 sales with a modest increase noted in the “Other” category.
The respondents were also asked about their utilization of a multi-year strategy and reported a range of 18 months at the low end, to five years at the upper end, with most companies centered in the two year period. Only two companies did not offer a multi-year period. We then examined the indexes used as the basis for a crediting rate, with the vast majority of respondents using the well-known S+P 500 index, followed by a fixed or declared rate, which offers less volatility. Recently, some carriers have introduced returns based on proprietary indexes which offer a customized “blended” or “volatility-controlled” index. Interestingly, some customized indexes were not necessarily equity-based, as two companies offered a commodity index, including a gold-based ETF. Another company offered a blended index based on international indexes, including the Asian and European equity markets.
A.M. Best notes that to the extent customized indexes are being used, product comparability and ultimate performance is difficult and there is potential risk in that the policyholder does not understand the intricacies of the strategy, which could raise suitability issues down the road. Additionally, customization of indexes can lead to higher hedging costs, with multiple hedge targets and potentially higher basis risk. Surrender charge adequacy for an insurer’s existing book was viewed as strong, 20% of the companies reporting surrender charges within a range of 5.1% to 7.5 %, 50% reporting surrender charges between 7.6% and 10%, and another 30% reporting surrender charges in excess of 10%. As expected, the dominant market leader had lower surrender charges with newer market entrants in the middle range.
Interestingly, some longer-established companies with large surrender charges also have longer surrender charge periods and we have observed that newercompany did not provide the data) reported spreads in the 251-300 basis point range. An additional two companies reported relatively low spreads of less than 150 basis points, with one company reporting spreads above 350 basis points, and two companies reporting current spreads of 151- 200 basis points or 201-250 basis points, respectively. Within the 2011-2014 period, no company had spreads within the 201-250 range.
The survey also inquired about bail-out provisions to address potential persistency issues. The question focused on whether or not bail-out provisions were made available to policyholders when the index crediting strategy upon renewal was lower than the current crediting strategy. Based on survey results, 80 % of companies do not offer a bail-out provision. The lack of a bailout feature increases the persistency of the book under stable interest and equity market conditions as there is no policyholder put option. However, most companies do offer a limited bail-out provision for medical purposes, such as long-term care and critical illness. In terms of a possible persistency drag on the book, if a company is selling at high issue ages (70 plus), a bail-out feature for medical issues could dampen persistency somewhat.
Next, a three-part question was asked regarding the impact of withdrawals (during the benefit accumulation phase) on benefit bases, death benefit guarantees and benefit changes in during 2014. With respect to withdrawals, all companies indicated that the impact of the withdrawal on the benefit base was on a pro-rata basis as opposed to a dollar-for-dollar basis. Respondents indicated that death benefit guarantees were not enhanced beyond account value in six out of
nine responses, with three companies indicating some level of death benefit enhancement.
The response rate for benefit changes as a result of product redesign indicated a high level of within the respondent population, with only two of the nine companies not making any product changes during 2014. The nature of the changes was as follows: one company increased benefit payout percentages, two companies reduced benefit payout percentages, two companies increased rider fees, one company increased the rollup period and one company reduced benefit rollup percentages, while one moved to gender-based pricing. We found the lack of gender-based pricing and lack of plans to implement this somewhat surprising given its pricing importance in other product lines such as long-term care.
We asked respondents about GLWB rollup rates, which ranged from a very low 3% to a very rich 7% with compound interest, with one carrier offering 9% with simple interest. With respect to additional bonus or index credits, most companies offer this on top of the rollup rate. The spectrum of additional bonus or index credits was broad, which also makes comparability among carriers challenging. These additional sales inducements included bonus, annual ratchets, and percentage of index credits with a broad range of bonus credits in the 8 to 15% range, although the metric they were based on varied (i.e. a bonus on premium, a bonus on an index or a bonus on the lifetime withdrawals etc).
This variation likely reflects underlying policyholder dynamics, distribution considerations and an insurer’s overall risk appetite. In general, GLWB income payout percentages generally increase as the policyholder ages over time, with a ceiling at a certain attained age, which is typically age 90. Generally, the payouts are stratified by age, with some companies having very welldefined (i.e. narrower) age bands. In some cases, the income benefit increases in a linear
fashion (i.e. .10% per year) then levels off at a maximum issue range. There was considerable diversity among companies, although generally speaking most income benefit increases are capped at age 90, with the largest payout percentage noted at 7.8% at this age level. Only 20% of the companies offered an increasing payout option. Most of the survey responses
indicated that withdrawals during the deferral period affect the benefit base or benefit payouts on a pro-rata basis.
The popularity of living benefits is high as, according to LIMRA, roughly 70% of policyholders elect the living benefit feature. Maximum issue ages are generally capped at age 80 although issuance at this age would require careful scrutiny for suitability, and probably only be appropriate for immediate income benefits. Our survey asked respondents to show the income benefit for a sample newly issued policy given the following assumptions: male, with a $1.0 premium assuming no additional premiums, a 6% crediting rate, and no withdrawals until benefit election at various election ages.
The maximum benefit, minimum benefit, and standard deviation at various issue ages and election periods, and demonstrates a fair amount of variability in income benefits among companies. We believe this is representative of differences not only in product design related to the relative richness of the guarantee, but also due to variations as to when the living benefit can be exercised. Additionally, it likely reflects differences in a company’s risk appetite and its relative experience with this product line.
With respect to benefit election periods available at various issue ages, there is variation noted. For example, most
companies do not permit election of an immediate annuity at younger policy issuance ages. We believe that product
designs that prohibit the election of an immediate annuity are generally less risky and note that if an individual wished to
purchase an immediate annuity these are readily available in the traditional fixed annuity market. For those companies that
do not allow election of an immediate income benefit, the survey results revealed that the slope of the benefit payout in later
years was much steeper to compensate for this.
While there are material differences in the risk profile of these benefits from company to company, we do
note that the relative risk seems to remain largely the same . For example, Company A has the riskiest product at older ages, and this trend held true at an older issuance age versus age 50. Therefore, the risk profile and relative competitive market positions generally hold constant.
The average age of the in-force block was high, even among newer market entrants, which reflects the fact that these products are generally being sold to the near, or at, retirement market segment. The range of in-force was between 50 to 70 plus years, although 70% of respondents reported having an in-force block between 66 and 70 years of age. An additional
question was asked regarding reserves; statutory versus GAAP and DAC balances. Consistent with our expectations, the living benefit component of the reserves was the major driver under both accounting regimes, and DAC balances generally fell within a range of 10 to 25% of total GAAP reserves, with the variability dependent upon the relative seasoning of the book and differences in commission costs by company.
Another question addressed asset/liability matching (ALM). Not surprisingly, most respondents indicated that the duration mismatch for ALM was short, which is reflective of the low interest rate environment and the expectation of rising interest rates. For most companies, the duration mismatch generally fell within 1 to 2.5 years, with one outlier noted at 4.4 years. In the aggregate, the net ALM mismatch on an average basis for the industry was a modest 0.5 years.
Current Industry Issues and Concerns
Our survey concluded by asking respondents to rank the following industry issues on a scale of 1 to 10 (with ten being the highest impact): adverse regulatory changes, entry of new competitors, aggressive or mispriced products, low interest rates, rapid rise in interest rates, increased election of living benefits relative to pricing, increased hedging costs, volatile equity markets, increase in longevity, lower than priced for lapses. While there was a lot of variability in the rankings, all of these issues were important concerns, although some issues clearly were identified as more important (i.e., adverse regulatory changes, equity market volatility and lower than priced for lapses as evidenced by the mode rank order).
The FIA market is a large and growing segment within the U.S. annuity market, with increasing levels of product complexity as companies continue to innovate in response to rising competition. This creates the potential for an “arms” race within the FIA market similar to what we observed a decade ago with the rapid increase in variable annuity sales. As product complexity increases, the risk profile of an FIA is likely to increase.
At its core, the FIA market with guarantees is a hybrid product, offering blended returns based on the underlying index composition and its crediting methodology. Its competitive advantage historically has been the company’s flexibility to re-price the product at a predetermined time frame, typically on an annual basis. However, the more recent introduction of living benefit guarantees has increased the riskiness of the product and essentially reduced some of the pricing flexibility of the product given the embedded guarantee.]
Increasingly, the lines of distinction between VAs and FIAs are becoming blurred. The relative complexity of crediting rate methodologies, different crediting indexes, and recent product innovation may increase hedging costs and raise suitability issues, particularly if there is a custom index that may not be relatively well-understood.
Additionally, as is the case with variable annuities, there is increasing complexity of underlying investment or index choices as product innovation shifts to more customized indexes/asset allocation strategies for FIAs and VAs, respectively. The long-term performance and ability to hedge these new custom indices is untested and will need to be evaluated over time. To the extent that customized, multi–blended, asset allocation strategies are utilized, there is the potential for higher hedging costs given multi-hedge targets.
Finally, we expect further innovation in product features over time, risk/return for companies active in the market. As always, A. M. Best monitors the impact of these features with respect to creating rapid sales growth and potentially larger product concentration for insurers.