Since the tech crash, it has been increasingly popular to purchase variable annuities with "guaranteed living benefit" riders to help protect retirement income against the risk of a market catastrophe. The basic premise of the contracts is rather straightforward: for a modest cost, the insured can transfer the risk of a severe market decline to the insurance company, and remain assured that guaranteed retirement income will be available for life, undiminished by the market decline. In the meantime, the annuity owner has the opportunity to stay invested in the market, with the hopes of generating an ever greater upside (albeit reduced by the costs of the annuity).

The problem, however, is that unlike most types of insurance - where the law of large numbers allows the insurance company to have relative certainty about the timing and magnitude of potential claims - in the case of annuities with income guarantees against a market decline, the insurance company faces virtually no risk exposure for any of its policyowners, until a major market decline actually occurs... and then, suddenly, the insurance company must set aside reserves for everyone, all at once. The end result is that market volatility can end out creating drastic volatility in the required reserves and profitability of the insurance company - to such an extent that in the past several years, many major insurers have decided to stop offering the guarantees altogether.

In the aggregate, the problem is that having a large group of prospective policyowners transfer their exposure to market risk to the insurance company may seem like the risk is being transferred from the policyowner's perspective, but the risk is actually being concentrated from the insurance company's perspective, in a remarkably undiversified manner. As a result, the policyowners actually face the danger of turning what was a systemic risk exposure to overall markets into a very specific, undiversified, unsystematic risk exposure to the credit quality of a particular (insurance) company instead. And for those who fear the danger of a "black swan" event, the reality suddenly becomes clear that the kinds of black swans that could be a blow to markets may actually be an even more severe blow to the insurance companies trying to guarantee against them, as was evidenced by the need of at least one annuity provider to access Federal TARP funds for liquidity in the immediate aftermath of the financial crisis.

Ironically, it turns out the best alternative may simply be the solution that advisors recommended before such annuity contracts were available in the first place: to invest more conservatively, and spend more conservatively, and simply stay the course and weather the storm. After all, conservative spenders following a safe withdrawal rate approach have largely recovered their declines since the financial crisis, while many insurance companies have left for good; similarly, while the safe withdrawal rate weathered the Great Depression, many insurance companies did not. Of course, in today's marketplace, the annuity companies have finally figured out how to proper structure, constrain, and price annuity guarantees to manage their risk exposures and make the products economically feasible to be offered; yet the outcome of such pricing may be leading to the point where their floor-with-upside guarantees are no longer an appealing trade-off for the (new) cost.

Fundamentals Of Insurance

The basic principle of insurance is very straightforward. Individuals pay premiums into a common pool of money with an insurance company, which the company invests for some growth (and subtracts a bit of cost for the overhead of the insurance company), and then uses a combination of the original premiums plus growth to pay out insurance claims. To the extent there's money left over (claims are less than the premiums plus growth), the insurance company generates a profit, while if the claims exceed the premiums plus growth, the insurance company has a loss. Since insurance companies are a going concern, with a steady stream of new premiums coming in, claims being paid out, new policies being established, and old policies that are allowed to lapse, the ongoing management of the insurance company is a bit more complex, but the fundamental equation remains the same: premiums + growth - costs = claims + profit margin. The insurance company sets the premiums, manages the costs, invests for growth, and aims for a certain profit margin that will be left after anticipated claims.

Of course, this also means that being able to effectively anticipate claims is absolutely crucial for the effective function of a life insurance company, which is no small feat in a world where any individual insurance claim - from the death under a life insurance policy to a fire burning down the house under a homeowner's policy - can seem exceeding random on a case by case basis. Fortunately, though, what we see from the "law of large numbers" is that, given a large enough sample size, the actual number of incidents average amount remarkably close to the expected value (assuming, for the time being, we can make a reasonable estimate of the expected value in the first place). What is "random" at the individual level is remarkably stable in the aggregate.

Thus, for instance, while in any individual scenario, the person does or doesn't die and the house does or doesn't burn down - a very random, binary outcome - with a large pool of insured individuals (or properties), the frequency of claims becomes remarkably consistent, allowing the insurance company to be able to set an effective premium that allows the whole mechanism to work in the first place. In turn, this allows for the availability of everything from life insurance to homeowner's insurance; for some newer types of insurance coverage, we may not get the expected value right initially - such was the case early on for disability insurance, and more recently for long-term care insurance - but the principle remains the same: in large numbers, claims can occur with relative consistently.

Of course, with enough consistency, the outcome is essentially an assured loss for any individual insurance policy owner; on average, the policyowner's expected value is reduced by the costs (and profit margin) of the insurer. But access to insurance allows the individual to turn what could be a relatively extreme financial impact - like the loss of a family's primary breadwinner or the house that they live in - into a much smaller, manageable cost. While technically the expected value of the insurance transaction is financially diminished by insurance company overhead and profit margins, it provides a way for an individual facing a binary outcome - the event does or doesn't happen - to participate in the much steadier law-of-large-numbers outcomes instead.

Insuring Against A Market Catastrophe

While insurance can be a highly effective way to manage risks that are highly uncertain individually but average out effectively in large numbers, the problem with trying to insure against a market catastrophe is that the risks don't "average out" over time; instead, they clump together. After all, the reality is that if a large number of people buy insurance against a market decline - e.g., through a variable annuity with a living benefit rider (GMWB, GMIB, etc.) - then nobody will have a potential insurance claim while the market is going up, but virtually everybody will be "in the money" at the same time when there's a severe bear market.

Of course, insurance companies had some acknowledgement of this risk, which is why policies that "insured" against market declines did not pay out an immediately liquid benefit, but instead merely allowed the policyowners to draw out lifetime income which meant, at some point, they may eventually deplete their own assets and then draw on the insurance company's guarantee. Nonetheless, where a severe market decline occurs, regulators require the insurance company to ensure it has reserves sufficient to pay out on its potential obligations, requiring a huge allocation to be set aside; thus, while the insurance company may ultimately be able to make good on its guarantees, avoiding a knockout punch default, the impact to profits for the reserve allocation is so severe the "technical" knockout punch leads the insurer to leave the business anyway (as occurred with several annuity guarantee providers after the financial crisis).

The challenge is compounded by the fact that, given market volatility, the sufficiency of reserves to back market guarantees themselves become highly volatile, as a base of hundreds of billions of dollars of assets are backed by guarantees funded by fairly tiny (relative to the assets) rider fees. If there is an extended bull market - and the insurer has many years to collect fees before facing a market decline that results in a significant insurance exposure - the consequences can be somewhat more contained. But the fundamental problem remains that - unlike virtually all other types of insurance - it's not feasible to slowly, steadily build reserves against a slowly, steadily rising base of guarantees; instead, because all the contracts are tied to the same underlying stock market risk, virtually all the policies become a potential claim at the same time. For instance, if $300B of guaranteed annuities experience a severe 25% market decline, the insurance company is suddenly exposed to as much as $75B of claims, for which gathering a 0.5%-of-$300B - which is "only" $1.5B of fees - just doesn't cut it.

Notably, in other insurance contexts, companies are very cautious not to back risks that could result in a mass number of claims all at once. This is the reason why most insurance policies have exclusions for terrorist attacks and war, and similarly why it's so difficult to get flood insurance in many parts of the country (and/or why the government must often make flood coverage available as the insurer of last resort). It's a crucial aspect of insurance that in the end, its exposure to risk is well diversified (allowing the law of large numbers to work) and not be overly concentrated.

Diversifying Risk Or Concentrating It?

Ironically, while consumers view insurance company guarantees as a way to diversify their exposure to risk, the truth is that they are actually transferring their risk to the insurance company, and in the process the insurance company is actually concentrating the risk, given that a market decline can put virtually all guaranteed contracts "in the money" at once. Thus, as seen in the financial crisis, while many investors could afford the market decline and stay invested for the recovery (as the market has now reached new highs), the concentration of those risks aggregated in some insurance companies was so severe that they've discontinued the products altogether (and in one case, actually need a "bailout" loan from the Federal TARP funds!). And unfortunately for the insurance companies, the problem seems to have been exacerbated by investors who bailed out of markets (but not the annuity guarantees), effectively "locking in" their losses and virtually ensuring that they will eventually become claims against the insurance company (and of course, with the policies "in the money" few people are surrendering or lapsing them anymore, either!).

In the aggregate, these challenges raise the question of whether investors can really protect against stock market "black swans" by buying insurance, or whether attempts to do so just converts the black swan from an investor risk to an insurance company risk (and a more concentrated one at that). Bear in mind, the fundamental purpose of insurance is not to make people whole from their losses in the aggregate; it's to redistribute and smooth the losses by turning large impactful risks into smaller, manageable ones by sharing some of excess premiums of the "winners" with those who have claims (the "losers"). Except when the risk strikes all policyholders at the same time - as is effectively the case when insuring against market declines and economic catastrophes - then the insurance company simply doesn't have the resources to make the protection work, as there are no "winners" to offset against the "losers" in such scenarios. Instead, none of the policyholders are actually drawing against the insurance company's risk pool and reserves against a market catastrophe... until they are, together, all at once.

The bottom line is that, ultimately, people seek out insurance guarantees to protect against "untenable" risks in the markets. The kinds of "black swans" that can result in an economic disaster and a market catastrophe. Yet the reality is that when such extreme events occur, the insurance companies that ultimately use the same capital markets find themselves exposed to the same systemic risks and events. Clients might seek out a variable annuity guarantee to protect against the next Great Depression, yet the reality is that there were a non-trivial number of insurance companies that failed in the last Great Depression (not to mention the long-forgotten "Insurance Holidays" instituted to protect more insurance companies from default!)! Yes, it's true that we regulate insurance companies better now than we did back then, but by definition "black swans" are events that aren't predictable, can't be foreseen, and therefore can't necessarily be regulated against in advance.

This doesn't necessarily mean that a random black swan is likely to come along and wipe out a large number of insurance companies, but the point remains that if the fear is that a black swan could be coming, an insurance company isn't necessarily a safe place to hide. In fact, the client risks turning a systemic risk for their overall portfolio into a more concentrated, less diversified individual risk into a single company (since, in the end, a guarantee is only as good as the credit quality of the individual company underwriting the guarantee, which isn't exactly "secure" if you're assuming a black swan economic catastrophe up front!). Yes, there are also state guaranty funds, but it's not clear they're remotely well-capitalized enough to bail out the insurance industry in the aggregate (not to mention that most states limit their annuity guarantees), and does anyone really want to bet whether deficit-strapped the states can/will come up with the additional state-guaranty-bailout-funds if called upon?

Where To Go From Here?

So what's the alternative, for investors and clients who really want to do something to manage their exposures to risk?

Simply put, there are two primary alternatives: 1) to just not take as much risk in the first place, managing risk not by trying to shift market risk to an insurance company, but just owning less risky stuff in the first place (i.e., having a more conservative portfolio); or 2) by simply spending conservatively enough that even if "bad stuff" happens in the portfolio, the retiree who in the end is only spending a few percent a year from the portfolio allow can allow the bulk of the assets to stay invested long enough for a recovery to occur.

After all, it's quite notable that in the end, this kind of safe withdrawal rate approach survived a Great Depression that many insurance companies did not, and those following a safe withdrawal rate approach since 2008 are doing fine (given the stupendous market rally that has occurred since the decline) while a large number of insurance companies had to permanently exit the marketplace. In other words, notwithstanding how scary the ride can be at the time, the reality seems to be that just prudently managing a portfolio and drawing a conservative amount from it each year has had more success withstanding bear markets, economic turmoil, and "black swans" than the insurance companies some investors are looking to for protection. To be fair, not all insurance companies have left the business since 2008, and none have actually outright defaulted on their guarantees at this point. But we have seen a number of very consumer-unfriendly practices from insurance companies trying to get off the hook for their guarantees, from "buy-backs" to changing available investment options and constraining asset allocations and more, and we still don't know how the next bear market will play out either.

Notably, this doesn't mean that advisors and their clients should eschew annuities and their guarantees altogether. The term "annuity" is still used to label an incredibly broad range of products, and some do not face these "concentration" risks at all, whether it's plain vanilla variable annuities used as a tax-deferral and asset location wrapper, fixed annuities that may have an appealing yield, or an immediate annuity that pools mortality/longevity risk (which really does obey the law of large numbers!). Nonetheless, when it comes to offerings like variable annuities with guarantees against market risk, it remains unclear whether in the end, clients are really shifting their systemic market risk away, or actually concentrating it into an even more dangerous unsystematic specific-company risk instead! That's not diversification, it's concentration!

Fortunately, many annuity companies have at least recognized the tenuous situation of their current market guarantees, leading to contracts that have less generous benefits, higher costs (or at least, costs that the insurance company is permitted to raise significantly in the future under the contractual provisions of the annuity), and "indirect" costs like asset allocation requirements that force investors to hold a significant portion of assets in fixed income investments that may cost more than their prospective return. Which means the danger of an annuity company "catastrophe" is perhaps diminished, but at the cost that it's no longer clear whether there's any value in using an annuity to secure an "income floor with upside" when there may no longer be any upside in the first place after the embedded costs.

  • JS

    Reduce equity exposure the closer one gets to retirement? Shift into what?…bonds? And then, what? … Wait for the disaster. I’d rather shift the risk to an insurer.

    • Michael Kitces

      What do you think the insurer is investing the money in, in order to provide that coverage? They’re buying… the same stocks and bonds you’re trying to avoid. Except they’re concentrating the risk of dozens/hundreds/thousands/millions of policyowners all at once. How does that improve the situation?
      – Michael

      • Joe Blow

        Wait in the case of variable annuities the insurer isn’t investing any money in stocks or bonds for the assets held in the contract. All of that is just passing through and going to mutual fund companies. Really the only thing the insurer is actually doing is collecting a fee for riders as the money passes through and then going into the derivatives market with the fee revenue and hedging the exposure that the riders are giving them. That’s it.

        • Steve Cooperstein

          Right on, Joe.

          • Brian

            I think an important point is that Hartford received a bail out. Most individual investors did not.

  • Joe Blow

    A) The hedge: The insurance carriers hedge a significant portion of this risk through index put option contracts. Where do you think the massive and outrageously higher fees for these riders go to? Insurance carriers have gotten really good at the mathematics of pairing derivatives and fixed income pools over the last decade with the growth of the Indexed products, but the math of pairing equities(that possess no fixed return) with derivatives to completely hedge a risk is little more of a wild goose chase.
    B) Persistency: Do not underestimate the importance and carriers experience with persistency rates in allowing insurance carriers to smooth out market volatility *on paper* for annuity/insurance policy holders. Insurance carriers have been doing that with bond assets for probably 100 years. At the end of the day if a carrier can accurately predict your persistency rates(which admittedly can never be an exact science particularly in the case of variable products) then an insurance carrier can get away with handing a person an insurance contract with a very consistent stated return each year based on projected growth over the average length a policy is held even though their bond portfolio is all over the board. At least with equities positions the individual holdings are segregated. In a lot of ways it’s the fixed side of the industry that actually carry’s more risk for insurance carriers because they’re actually guaranteeing the entire performance with disconnected bond pools and in the variable side the performance is based on the sub accounts and they’re only taking large fees and buying option contracts to hedge the risk (something that the average person could do if they understood it–granted the insurance carrier retains a small amount of fat tail risk though because they’re not going to hedge the entire thing all the way to 100% loss).

    ***The real issue is that annuity contracts guaranteed income withdrawal rate is massively lower than what you could safely take out of some couch potato 60-40 portfolio and you suffer in performance under the fees. People forget that the 3.5% or 4% safe withdrawal rate is not the same thing as an annuity contract offering 4%. The SWR refers to 3.5% or 4% **starting* withdrawal rate + inflation adjustments. So it’s actually 4%, 4.15%, 4.3%, 4.5%, etc. to mimic the effect of inflation adjustments in income while working. An annuity contract offering 4% for life of contract has no such adjustments.***

    That and the terrible idea of moving a capital gains asset with a step up in basis at death into an ordinary income asset with no step up in basis at death are the reasons why annuities should be generally avoided(with few exceptions).

  • Bill Smith

    The anti-income guarantee (or annuity) is clearly ingrained in some segments of the advisor world. Clearly income guarantees can be viewed as a threat to advisors and there is a very vocal group (albeit minority) of advisors who will come up with any reason to say that clients shouldn’t consider an income guarantee. It’s very interesting that almost all academics see the benefit of an guaranteed income stream for life. It really is only the fee-only RIA community who have refused any adoption of any guaranteed income stream no matter how good a product. Clearly, RIAs have a major financial interest in making the case that clients should keep assets with the advisor and not move to an insurer. The author fails to note that if the markets become very volatile, benefits are only paid out when the covered asset account is depleted which is a function of the client’s longevity. The combination of proper hedging and pooling of risk using mortality assumptions significantly if not completely limits the carriers underwriting risk. On the flip side, what protection does a client have w/o an income guarantee? Telling your client to cut spending or get a job is not optimal.

    • Joe Blow

      I’ve met numerous commission based and hybrid advisors that eschew annuities on purely mathematical grounds.

      A variable annuity guaranteeing a 4% withdrawal rate over 30 years of life expectancy for a large rider fee is vastly less superior to 4% year one, 4.12% year 2, 4.24% year 3, 4.37% year 4, 5.38% year 10, 7.22% year 20, etc. that the academic community has stated numerous times is considered a safe withdrawal rate in practically any environment.

      Also, an insurance carrier isn’t some magical entity (nor are the writers of derivatives contracts carriers buy from to hedge their guarantees). They are looking at the same data when they price these things out and they have to offer these guarantees at a cost high enough to provide themselves a cushion in case things don’t go the way they projected. And over a long enough time period markets tend to be much more predictable which means that paying some other entity to lay off that risk is largely only illusory piece of mind when you had time on your side the whole time.

      • Bill Smith

        Joe, keep in mind most if not all income guarantees provide a step-up based on market increases. If the fee drag on the investment account is low, there can be multiple set-ups. In a strong up market (e.g. past 4 years) a step-up after lock-in provide a huge upside benefit. Its all about sequence of return risk. You can say over 30 years markets always cooperate but the timing of your retirement and the risk of market drop during the immediate years pre/post retirement is very real.

        • Michael Kitces

          If the markets perform favorably enough that the guarantee was stepping up, the retiree could “step up” their safe withdrawal rate as well, and still come out with greater cash flows.

        • Joe Blow

          And those step up guarantee’s also have a pretty high fee associated with them. Furthermore, they also typically restrict investment options enough that prevent you from taking on much risk so that step up basically never goes in the money and the annuitant just ends with lower performance than they should have had because of this forced level of conservatism.

          The 3.5-4% withdrawal rate works in the worst of retirement timing situations in US history. Retire in 1929 with a ~3.7% withdrawal rate and 30 years later you still have money left. Retire in 1972 (the worst year in US history to retire–because inflation matters more in distribution phase than market declines) with a 3.5% withdrawal rate and 30 years later you still have money left and you’re withdrawal rate has grown to about 20% a year of the original balance you had in 1972.

          Take out your Variable annuity with 4% and it’s still a 4% withdrawal rate 20 years later even though inflation has caused that income to be worth considerably less. In 1972 a person taking out your variable annuity was screwed.

        • Joe Blow

          And Bill if you wanted to not completely screw a client in 1972 with an annuity this is a way you could do it.

          Instead of going into the variable annuity market for 4% withdrawal rate go into the fixed annuity market for lets say around 5.5% withdrawal rate for some recently retired individual.

          Now you only need 3.5% withdrawal rate in order to compete with academics on the issue. Well 3.5%/5.5% = 64% of the portfolio that could go into a guaranteed income stream to generate 3.5% of the entire balance the first year. You would then have 36% of the portfolio left over to invest any way you like to cover the generate the inflation adjustments (we’ll call this the side account). At first you would withdraw zero from the side account, but over time this account has to cover the inflation adjustments for all of the withdrawals. This has the effect of making the growth in the withdrawal rate of the side account more parabolic towards the end, but at least in theory should be able to closely mimic the academic safe withdrawal rate with similar or potentially higher consistency. Of course their is always a cost when you use a carrier and this one comes with much higher odds of substantially reduced inheritance to your heirs as the price to the insurance carrier(but maybe the client doesn’t care about that or doesn’t have any).

    • Michael Kitces

      I’m presuming this comment is directed at me to imply some kind of “anti-annuity bias” – you do realize that I am a co-author of “Advisor’s Guide To Annuities” (see and have written on the benefits of annuities for many, many years now.

      The reality is that the product selection available today in the variable annuity GMIB/GMWB world is drastically different than it was 10-15 years ago when it started. To acknowledge that the products, features, and risks have changed is not an “anti-annuity bias” – it’s acknowledging reality.

      You suggest that “proper hedging” must be easy, yet given how many insurers have left the marketplace, and the fact that Hartford had to get bailout funds – after adamantly stating in the 2005-2008 period that they were hedging these risks and were not exposed – makes it clear that the hedging is not nearly as easy in practice as implied. In reality many of the companies were retaining a portion of the risk (either that, or just they outright botched the hedging), as their departure from the post-2008 marketplace seems to have revealed.

      • Joe Blow

        Mike keep in mind the spread that I just illustrated above and then realize that the carrier has that spread for the rest of the annuitant’s life.

        Retire in the worst year(1972) and a person starting out at 3.5% withdrawal rate has about a 20% withdrawal rate 30 years later. A quick estimate has that at probably around a 6 or 7% average weighted withdrawal rate over the entirety of those 30 years. The insurance carrier fees the heck out of the holdings and offers you 4% fixed over those 30 years. That is a pretty large spread and cushion.

        Carriers that got their ratings crushed after 2008 did so because they held to much MBS and CDOs in their general funds which were held against their **fixed life** business. Granted, many took losses due to massive underpricing and reserves of their riders, but they have since jacked the costs of those riders in the other direction.

        Now the situation isn’t so much the risk these guarantees place on insurers, but more so “At what price is the guarantee no longer worth it.” I don’t know, but it has to be lower than outrageous prices you see today.

        • Michael Kitces


          Indeed, in terms of the particulars of the available riders and the cost/benefit trade-offs today, this has become my concern as well – as the riders have moved to arguably “better” pricing, they have shifted from “floor with upside” arrangements to what seem to be inevitable self-fulfilling prophecies of withdrawals-plus-fees that drive people straight to the guaranteed outcome. The “worst case scenario” of the rider guarantee is becoming the BEST case scenario (see

          Of course, indirectly this is the point I’m trying to make. When the riders are underpriced, there’s a risk that a severe market event will disrupt the annuity provider. When the guarantees are “fairly” priced, they’re so expensive they become unappealing. All of this traces back to the same underlying problem – when the insurance companies are investing in the same stuff with the same capital market assumptions, they can’t just manufacture returns or guarantees out of thin air. They just concentrate the same risks that were already on the table. And to the extent they try to hedge them with options, arguably many consumers could simply do the same?

          • Joe Blow

            I’m with you it’s just the “They concentrate the same risks” line that is throwing me off a bit. A bank that borrows short and lends long concentrates their risk.

            An insurance carrier that locks a pool of people into a long term contract can get away with making people believe they’re getting the excesses of the market smoothed out. All they’re really doing is window dressing the market **because they have people locked in for such a long time**.

            Let’s say I had xyz portfolio that was going to average 5% over time, but be everything from up 15% in good years to down 20% in bad years. Now if I offered to stand between you and that volatility and then gave you 3% fixed each and every year would it be a good deal? Now what if the catch was that you had to stay in that contract for the rest of your life to receive it?

            Sure carriers might take some losses on the really tough years when their regulator demands they hold more reserves against something that is paying out over 30 years at the exact same time their general fund is down, but it’s not going to take the carrier down. They’re just not going to make any money in those particular years. The real problem is that people think that less of volatility on paper of a product that requires lifetime participation means anything? It means nothing. That is why they wont underwrite the same risk, for the same price, on something that allows me to leave and keep the proceeds of that embedded option.

            **As to your question, in a very imperfect way customers could hedge their portfolio in a loosely similar way to the way variable annuity guarantees work. Depending on the terms of the exact rider that you were trying to mimic you would take out rolling at the money or out of the money 1 year put options on an index with a few percent of your money. You would hedge a large portion of it. Indexed annuities are much easier to mimic by customers(zero coupon bond + call option with the price difference between the starting zero price and the future bond face or coupon bond with rolling annual call options premiums equal to interest rate on the bond).

  • Co advisor, CPWA

    The concept of ANTIFragile is important to honor in this discussion. If the smoothing of the pattern of returns occurs instead of a quick shock rhe value of older annuities is real. The use of death Benefit annuities is not being explored as a real benefit to spousal benefits. The new joke is paying 4 to get 5%. Oh and the 4% is not gaurenteed, it can go up.

    Part of the problem with the RIA sales channel for annuities is that there is no AUM and this could reduce income significantly. Why are they not sold at Schwab?

    So mike thanks for advancing the conversation. ANTIFragile by TALEb should be required reading but that might be harder than the annuity sales liscence let alone the lack of required ce

    • Michael Kitces

      Actually, variable annuities would seem to violate Taleb’s Anti-Fragile principles, again because they concentrate the investment exposure in a single insurer, which is entirely reliant on pricing the cost structure right up front. The departure of so many insurers from this marketplace after the financial crisis is frankly an example to me of just now fragile they actually were.

      In terms of RIA options for annuities, there are a lot more now than there were. Schwab actually has had an annuity option for years (though no guaranteed retirement income rider for most of that time). A number of other carriers have been serving RIAs as well. The Aria RetireOne offering is specifically built for RIAs to offer a version of the retirement income guarantee without the separate annuity wrapper and without ‘disintermediating’ assets from the RIA. Though again, there is still debate about whether its guarantees really provide any material value at this point beyond just what’s available by investing and spending conservatively.

      While I certainly wouldn’t ignore the fact that at least some RIAs are likely negative on annuities because it does represent a risk to their asset base, the reality is that there ARE options for RIAs to find annuities to use. The problem is in the trade-offs (costs and available benefits) that are available today, and the sustainability of the companies to provide those guarantees; it’s not just about distribution channels anymore.

  • Steve Cooperstein

    I don’t really have time to comment on this in detail, but I just can’t hold back entirely.

    Michael, while I am fully impressed with all of you do and say, when it comes to insurance and income generation I think you need to do more research.

    Have you talked to any actuaries who have priced variable annuities with GLWB’s? While I am not a fan of the products because they fall so short of income annuities for generating income, I can assure you that insurance companies know how to hedge these guarantees, not perfectly, e.g., they use indexes for their hedges which will not exactly match the portfolio mix of the sub-accounts involved (i.e. Joe Blow is close on this).

    While i am not an insider and haven’t been involved since the downturn, I think companies are shying away because, as you almost say, capital requirements have been raised, making them less profitable than other venues for capital, and because top management is afraid of the seeming risks involved.

    BTW, most insurance companies didn’t go under during the Depression, and if the massive downturn you are warning against happened to a 4% (or less) unprotected equity/bond plan, the cut in the portfolio value would obliterate income.

    • Michael Kitces

      Yes, I’ve spent time with a number of actuaries pricing variable annuity GMIB and GLWB benefits. I’m aware of how these can theoretically be hedged. In practice, the effectiveness of the hedging varies, and if the asset base is large enough even “mostly good” hedging is not enough to avoid severe capital impacts (which happened to several companies in 2008-2009). As I’ve heard from one actuary, a number of companies were actually frightening in the extent to which they were self-insuring and basically not hedging at all prior to 2008, though most companies have at least learned that lesson now. But the caveat is that to more effectively hedge for real – especially in a higher risk environment – the cost of the contracts (and the flexibility of the benefits) has adjusted significantly, to the point where the guarantees are far less appealing than they were.

      If just “hedging your equity exposure with options” magically improved the risk/return parameters of the market, we could do it directly for the same results. The issue is still that at the end of the day, the insurance companies rely on the same capital market instruments and available returns that we do without them. Buying solutions like immediate annuities have risk-pooling features around mortality/longevity that genuinely improve the stability of the pool. But concentrating investment assets does not average out their risks the way that pooling mortality does in obeying the law of large numbers! While insurance companies may have a little bit more pricing power in negotiating lower costs to execute derivatives to hedge, ironically no one here seems to be making the case that the primary feature of an annuity is “insurance companies get better options pricing than consumers”.

      • Joe Blow

        Michael when a company’s pension plan let’s say drops 30% in one year why doesn’t it blow up the company immediately?

        More specifically when a pension plan takes a large loss in one year is the company forced to transfer enough assets from it’s balance sheet to the pension plan to restore it to the assets it had prior to the loss all in one year?


        A guaranteed income rider works in a semi similar way for an insurance carrier. The insurance carrier doesn’t have to immediately reserve all of the paper embedded options on their variable annuity contracts in one year when a crash happens. They just have to actuarially determine what to reserve over time. Crashes only impact their long term assumptions a bit(like a 30 year old targeting a particular number for age 65 after a crash might realize they only have to contribute an extra little bit per year to still hit that number with the same return assumption) which fluctuate based on how out of the money or in the money their pool is. The other thing they impact is that their regulator gets more cautious causing their reserve requirements to increase.

        Steve’s dead on. The main reason why some carriers exited the rider game and all others have jacked their fees for worse terms is that it’s no longer as profitable under their new reserve requirements. This is similar to the way banks are now all starting to sell their mortgage servicing rights to non bank financial institutions. It’s not because mortgage servicing rights threaten a banks balance sheet at all(it’s actually one of the safest things a bank can retain), but instead because Basel III raised the amount of capital they have to set aside for their value and it’s no longer profitable(it doesn’t contribute much to their bottom line).

        Insurance carriers use a combination of: index derivatives, reserves, and the distributed risk inherent in long term contracts to be able to pass these guarantees. The first is a partial hedge, the reserves is a capital cushion, and the last is just fake piece of mind because customers don’t understand that if you hold something for 30 years you’re average return is pretty predictable. So a company forcing you to participate for a long time making a few bad years look better on paper isn’t really doing much for you if the long term performance is going to be the same(or more accurately worse).

      • Steve Cooperstein

        Thanks Michael – more on point..

        As to the “we could do it without them” (besides the we/they :-) ), THEY are the only ones that can lower the cost of the income protection by only providing it till the person dies (just to be clear, this is not the same as the risk pooling in immediate annuities)

  • stephenwinks

    Excellent discussion.
    In recent years, many insurance companies have backed away from alluring performance guarantees which have proven not to be economically sustainable. Shouldn’t the economic viability of these future promises meet regulatory requirements before these products are approved by each state for sale? IF not, are there consumer protection remedies?
    Seems like this base point vetting would resolve a lot of misrepresentation problems.

  • Pingback: Fixing The Fixed Indexed Annuity()

  • Pingback: Can Variable Annuities Really Offer Insurance Guarantees Against … | Get Insurance Quotes Quick()

  • Michael Smith

    One of the things I’ve found interesting in these types of discussions is that I’ve yet to see a discussion of the costs of managing a portfolio with a withdrawal percentage. Many advisors will bash an annuity for its expenses (not Michael Kitces of course) but then will downplay their own fee structure and lack of transparency of internal costs. For example, many advisors are charging a fee of 1% – 1.5% (I’ve seen as high as 2%) plus using actively managed mutual funds with high expense ratios and turnover (see supplemental prospectus that no one reads for these added costs), trading and platform costs, and perhaps an additional layer if using a “money manager”, and you can easily get to 2.5% – 3% before you know it. Surely, these costs are a drag on a 4%-5% withdrawal rate, right?

    • Michael Kitces

      Costs are absolutely a drag on safe withdrawal rates as well. I’ve actually written about this separately a few times over the years on this blog and in my newsletter research, most recently at

      – Michael

      • Michael Smith

        Of course I should have known you’ve addressed already. I have some catching up to do since I just started following you this year! haha

  • Pingback: Rick Kahler: Annuity insurance For Market Crashes Has Risks | Kahler Financial Group()

  • DCdemos

    This is why you see many companies monitoring their flow of business much more closely these days. In the same way that a client would DCA into the markets, insurance companies are doing the same. They don’t want to take on too much business in the wrong year. Plus, these companies know that not everyone will/can take income from the products at the same time (If they’re not 59.5 etc). I’m sure this plays into their calculations too.

  • Pingback: The Apparent Paradox of Retiring Twins – Is it True? | Better Financial Education Blog()

  • Pingback: Can Variable Annuities Help In A Market Catastrophe? | Better Financial Education Blog()

  • Pingback: How do you picture retirement? | Better Financial Education Blog()

6 Tweets / Trackbacks

Michael E. Kitces

I write about financial planning strategies and practice management ideas, and have created several businesses to help people implement them.

For ConsumersFor Advisors

Blog Updates by Email

Nerd’s Eye View Praise

@MichaelKitces Twitter

Out and About

Thursday, March 5th, 2015

*Future of Financial Planning in the Digital Age @ Private Event

Tuesday, March 10th, 2015

*Understanding Longevity Annuities and their Potential Role in Retirement Income *The Impact of Valuation-Based Asset Allocation on Retirement Income *Future of Financial Planning in the Digital Age @ FPA San Francisco

Wednesday, March 11th, 2015

*CPWA Training on Retirement Planning @ IMCA