Rational Decumulation - Annuity to the Rescue

[chinaco]

Its been at least a week since anyone has had a chance to annuity bash. Just thought I would share this article. It does provide some interesting thoughts... although contrary to the thinking of many people on this board.

It is from the Wharton School of Business (and NY Life) about using annuities in retirement to achieve.

http://fic.wharton.upenn.edu/fic/papers/06/0614.pdf

Key findings revealed in the study also include:

• Immediate annuities offer the advantage of risk-pooling. The Wharton study found that because insurers can share the risk of outliving one’s savings across a large group of policyholders, income annuities can offer financial security throughout retirement using 25 to 40 percent less money than would be required to provide an equivalent level of financial security through a retirement portfolio that does not incorporate income annuities – a benefit no other financial product can provide.
• No other asset class can address the risk of outliving one’s nest egg without requiring much more money. Current mortality tables show that an average healthy American male at age 65 today can expect to reach approximately age 85 – but that same individual also has a 50 percent chance of living beyond age 85 and 25 percent chance of living beyond age 92. As a result, people who plan to cover their economic needs to their “life expectancy” – in this case, age 85 – still face a 50 percent chance of failure. The Wharton study explains that only lifetime income annuities can mitigate the financial risk of living too long by relieving consumers of the need to set aside the far greater sums they would otherwise need to allocate to other asset classes to ensure they would not outlive their retirement savings.
• Equities, fixed income and other investment products are not substitutes for income annuities. Professor Babbel and Professor Merrill demonstrate that, unlike income annuities, other asset classes fail to address the risk of living too long. In addition, the study shows that consumers who place their retirement wealth in investment products like mutual funds are subjected to greater risk, typically higher expenses, and returns that are unlikely to keep pace with annuity returns, when investment risk is taken into account.
• Covering basic living expenses with income annuities can enable significantly greater flexibility in other areas of a retirement plan. According to the study, if basic expenses are covered by income annuity payments, individuals can keep their discretionary funds invested in equities for a longer period of time, providing the benefits of historically higher returns. This approach can also enable retirees to delay taking Social Security benefits until they are fully vested, providing substantially higher payments.
• Several insurers have added consumer-friendly innovations to the income annuity product platform, virtually eliminating the traditional reasons for not purchasing. According to the professors, features such as annual inflation adjustments, access to capital in emergencies, legacy benefits, interest rate protection, and the ability to increase or decrease payments at a future date make historical reasons for not purchasing income annuities no longer valid.

 

[FinanceGeek]

I personally fall into the camp of believing that annuities are seldom a good deal for the purchaser. I think equity/fund investing is the way to go, especially with the advent of near-zero cost brokerage commissions, penny pricing (reduced spreads), and very low management fees on ETFs and other index fund offerings.

There are a few advantages of annuities, they fall into being able to invest in a tax deferred product without having earned income (buying a variable annuity for a child / grandchild is a great bequest mechanism), asset protection (annuities are not part of your bankruptcy estate in many locales), and as you mention the ability to receive guaranteed payments for life.

But I still think for most investors these advantages are outweighed by the higher costs associated with annuities, the inherent tax disadvantage of converting preferred tax rates into ordinary income rates at withdrawal, etc. You are right that some issuers of annuities are becoming more cost sensitive (I like Vanguard), but on an industry basis I don't think its as good a deal as self directed investing for the average FIRE type investor who is well schooled in the mechanics of investing.

 

[MasterBlaster]

FinanceGeek:

I believe that your objection is to deferred annuities. The article is in support of a different product - immediate annuities.

I have seen a number of studies suggesting that (perhaps) the best use of a nestegg is to buy both an income stream with some portion and then invest the balance in a more traditional portfolio. If ones goal is to never run out of money then the annuity payment along with a SS payment can provide a living floor if the other investments don't pan out or don't last as long as needed. With such a strategy one can take a more aggressive approach to depleting (decumulating) the portfolio side of the nestegg. With the portfolio-only strategy one must use a very conservative SWR to avoid never depleting the nestegg.

One problem with a portfolio-only strategy is that you take too little out of the nestegg when your are younger trying not to run out of money. And then when you are older with reasonable markets you end up with a huge portfolio that you cannot reasonably spend in your old age. That SWR rate keeps you from going broke but it also keeps you from living (monetarily) at your potential. The IA approach in some ways attempts therefore to safely spend the assets while you are younger.

Per the large fees, they can be somewhat mitigated by using a very low cost provider such as Vanguard.

 

[walkinwood]

Here's an article
http://www.early-retirement.org/forums/f28/new-thoughts-on-the-draw-down-phase-29273.html

that explores a "check" to see if & when you need an annuity. Its not a now or never proposition. Good article.

ww.

 

[chinaco]

Here's an article
http://www.early-retirement.org/forums/f28/new-thoughts-on-the-draw-down-phase-29273.html

that explores a "check" to see if & when you need an annuity. Its not a now or never proposition. Good article.

ww.

I read the article. I thought it did a good job of presenting an alternative way to look at the problem.

You are correct, the timing of the annuity (SPIA) is a factor to consider.

The pooling of money can have a desirable effect for removing longevity risk. Using a SPIA to create a base income can free one up to consume more of the portfolio earlier and not worry about being destitute. (DW and I have no desire to leave a large amount in the estate when we die)

The choices of how to fund the retirement income stream are a bit mind boggling. If an annuity is used, there are still some risks to consider. The insurance company could default. Plus, one still needs to deal with inflation in someway.

It seems to me that a Hybrid approach using a SPIA (some variant) + SS + Pension + Traditional Portfolio Management with securities is a viable approach.

The way I will fund our FIRE plan could indeed include a SPIA or funding for a SPIA late in life (as described in the article).

 

[FinanceGeek]

OK, we're talking immediate annuities. Mea culpa.

But, I still don't "get it". If I want to RE (lets say in my 30s...40s...50s) I can indeed invest a lump sum with one of the insurance companies and in return get guaranteed payments for life. A quick check of a representative annuity company (I used Berkshire Hathaway's EZQuote site at EZ quote) suggests that they are using around a 5% or so "safe withdrawal rate" assumption which indeed beats the 4% number we use around here.

But, two huge problems - if I do this my monthly income never grows with inflation (a huge problem if I'm going to FIRE and be in retirement for decades, even 3% to 4% inflation requires some planning for this retirement horizon), and I forfeit the ability to leave the funds I gave to the annuity company as a bequest to my heirs (although many annuities offer a guaranteed minimum payment option in case I die young that again only pays back the original investment without any growth).

Perhaps the optimal strategy here would be to consider taking the immediate fixed annuity without guaranteed minimum payment and using term life insurance to cover the possibility of you dying early. A quick check of the rate quotes suggests that you might or might not come out ahead on this one. Certainly if you're not eligible for the best term rates it'd be best to go with the guaranteed payment option since no medical underwriting is required.

Still doesn't sound good for someone who's going to FIRE compared to doing it yourself with index funds / ETFs.

I still think that the asset protection angle is the strongest reason I'd see someone who wanted to retire really young doing this.

OK update - I do see that some companies offer inflation protection to their single premium annuities through equity indexing, guess I need to read up on that more to see if that's an attractive option. I guess this site and most of its members is so focussed on the DIY option that this whole idea is very much "outside the box"...

 

[MasterBlaster]

FinanceGeek:

Looking at the Vanguard IA calculator I get a payment of around 5.7% for a 30 year old investing in an IA. If you take the IA with inflation adjustments you get a payment of around 3%.

Note for very long retirements that the SWR for stock-bond portfolio's is less than 4%, maybe 3.5% or so.

So if your point is that you are taking a huge hit by buying an IA, that is not really the case.

 

[FinanceGeek]

OK thanks I'll check out the VG site. It sounds like you have a good point.

Perhaps we're all really working too dang hard learning about MPT and SWR. I guess one could just turn it all over to a third party and be done with it! I long ago stopped doing my own car repairs, perhaps specialization of labor is not to be ignored in this field either.

 

[MileKing]

In my view, people seem to make the annuity discussion far more complex than it needs to be. Contrary to the findings of the study cited, immediate annuities are very much a fixed income product; in return for an upfront sum of cash, one receives a regular fixed stream of payments over some time period (one’s lifetime in the case of an immediate annuity). In this respect, they are very much like a long-term bond and should be treated as such when making investment / allocation decisions. As the immediate annuity provides guaranteed payments for life, they would seem to be a suitable substitute for one’s long-term bond positions provided the annuity payout (after all expenses) is greater than the bond’s payout. In situations where age + remaining bond term > life expectancy, the annuity is directly comparable to the bond and indeed has an advantage in that it eliminates longevity risk. Other than as a substitute for the long-term bond portion of one’s portfolio, I have to agree with FinanceGeek and conclude there is not much utility to immediate annuities.

 

[MasterBlaster]

OK so I retire at 55 with a 85 year life expectancy. So I set up my bond ladder spanning 30 years and spend the proceeds as they come due.

What happens if I am so unlucky as to live longer than my life expectancy ? Clearly half of the population will have this problem.

That in a nutshell explains the advantage of IA's and their utility.

 

[MileKing]

FinanceGeek:

Looking at the Vanguard IA calculator I get a payment of around 5.7% for a 30 year old investing in an IA. If you take the IA with inflation adjustments you get a payment of around 3%.

Note for very long retirements that the SWR for stock-bond portfolio's is less than 4%, maybe 3.5% or so.

So if your point is that you are taking a huge hit by buying an IA, that is not really the case.

Not sure why the 30 year old would want to invest in the example IA paying 5.7% when they could purchase a AA/AAA 30 year corporate bond paying about 6.2%.

 

[MasterBlaster]

Not sure why the 30 year old would want to invest in the example IA paying 5.7% when they could purchase a AA/AAA 30 year corporate bond paying about 6.2%.

Well as you know, Corporate bonds can change overnight. The new management comes in, loads up the company with debt. Then some number of years out the company goes under and the bondholders get stuck with nada.

That story happens all too often.

 

[Rich_by_the_Bay]

Well as you know, Corporate bonds can change overnight. The new management comes in, loads up the company with debt. Then some number of years out the company goes under and the bondholders get stuck with nada.

That story happens all too often.

A better analogy might be a long-term bond index fund, self-annuitized over a sure-thing life expectancy (like age 100). Still, there is some risk which is not present in an annuity.

 

[MasterBlaster]

Corporate bonds are not the same risk level as insurance company annuities.

We are comparing apples and oranges here. I can always find an investment that should pay higher returns. But I can't always find investments that pay higher returns without taking on more risk.

 

[MileKing]

Well as you know, Corporate bonds can change overnight. The new management comes in, loads up the company with debt. Then some number of years out the company goes under and the bondholders get stuck with nada.

That story happens all too often.

In order to compare purchasing a corporate bond to purchasing an IA, you need to hold the corporate bond until maturity (e.g., assume 30 years for long-term bonds). Thus, daily flucuations in price/yield are irrelevant. AA/AAA bonds are investment grade corporate bonds. I believe the default %'s on such bonds are extremely low, less than 1%. That should be very comparable to the chance that the insurance company will go out of business at some point taking one's "guaranteed" IA payments with it. If default risk on a single major corporation is particularly troublesome, one could invest in a long-term investment grade bond mutual fund. Vanguard's fund appears to be yielding 6.13% (net of expenses) at present. However, I should point out that bond funds churn their holdings and thus over a 30 year period there is no certainty that the yield would average 6.13%.

 

[FinanceGeek]

By going with the IA over do-it-yourself one gives up one set of risk/reward outcomes for a (probably) lower set of risks and rewards. After all the insurance company will establish its annuity payouts such that they expect to make money, the implication being that a reasonably savvy investor SHOULD be able to do better.

Personally I would not at all worry about Berkshire Hathaway or Vanguard defaulting on their annuities, short of some political/economic conditions that would wipe out any & all equity investors anyway.

Informative discussion! I have always mentally written off annuities but I can see from this discussion how they would have utility for some investors, especially less sophisticated or more risk averse ones.

 

[CyclingInvestor]

I have yet to see a study sponsered by an insurance company that did
not conclude that annuities were a good idea. It reminds me of the old
Tobacco Institute studies showing smoking was harmless.

 

[MasterBlaster]

Well there is much truth in that observation.

Perhaps quite a bit of cynicism too.

 

[W2R]

Personally I would not at all worry about Berkshire Hathaway or Vanguard defaulting on their annuities, short of some political/economic conditions that would wipe out any & all equity investors anyway.

Obviously, you are not a champion level worrier like me!! Give me a scenario, and I will worry about it.... I think that the one unresolved issue that stands between me and a small immediate inflation-adjusted lifetime annuity (maybe 1/4 - 1/3 of my TSP nestegg) is the possibility that MetLife might fold or default on their annuities before 2050.

I will probably get the annuity anyway, but I do worry! So many equally unsettling things have happened in the past 50 years.

 

[Rich_by_the_Bay]

but I do worry! So many equally unsettling things have happened in the past 50 years.

Want,

You should read Taleb's book "The Black Swan." It's terribly written and the author is a cynic, but it's about how we plan and plan and plan, yet most truly high-impact events happen from out of the blue and can't be planned for at all.

He speaks vaguely and confusingly about how to posture to minimize the impact of "black swans" but I didn't really get what he was saying. Nonetheless, your succinct observation is validated in the book and it was thought-provoking.

 

[W2R]

Want,

You should read Taleb's book "The Black Swan." It's terribly written and the author is a cynic, but it's about how we plan and plan and plan, yet most truly high-impact events happen from out of the blue and can't be planned for at all.

He speaks vaguely and confusingly about how to posture to minimize the impact of "black swans" but I didn't really get what he was saying. Nonetheless, your succinct observation is validated in the book and it was thought-provoking.

Sounds interesting. I'll look for it. It sounds like Taleb is another champion level worrier, like me! You know, he is SO right. The truly high-impact events can't be planned for at all. I have been trying, by building in safety net after safety net into my ER financial plans. Life is always an adventure, though.

 

[ats5g]

FinanceGeek,

You also might be interested in the better ages at which to annuitize money. I suppose it depends on your circumstances somewhat, but I don't see myself annuitizing any money before 60-65. The additional payout from annuitizing money [whether fixed, variable, or infl adjusted], just isn't great enough until then.

See Optimal Asset Allocation and The Real Option to Delay Annuitization: It’s Not Now-or-Never. For a less math intensive explanation see chapter 6 of Lifetime Financial Advice: Human Capital, Asset Allocation, and Insurance [starting on page 66]:

Thus—despite the preponderance of theoretical arguments in favor of annuitization—we are hesitant to advocate a single optimal age at which an investor should convert his or her savings account into an irreversible income annuity. Given the many trade-offs involved in this decision and numerous sources of uncertainty, we are comfortable with suggesting that prior to age 60 is too early whereas waiting until the age of 90 is too late. (At the advanced age of 90, the unisex mortality rate, q90, of 15 percent leads to mortality credits of 1,850, which are insurmountable on any investment frontier.)

For those worried about insurance company insolvency, see the portion of the paper in the OP that talks about state insurance guarantee limits. Additionally, note that it is usually the present value of your payments that is guaranteed, not necessarily the original amount of $$ you plunked down for the annuity. So, if the company you bought the annuity from at age 60 tanks when you 90, the present value of your payments at age 90 is going to be much less than the $$ you plunked down at age 60.

- Alec

 

[W2R]

For those worried about insurance company insolvency, see the portion of the paper in the OP that talks about state insurance guarantee limits.

Though it is probably very hard to believe (because it sounds like a justification, which it isn't), I actually did read the sections about that very slowly and carefully before my post. I even looked for that information.

I guess I have experienced too many of Rich_in_Tampa's "black swans". All of them seemed equally implausible and "taken care of" beforehand, or more so, but they happened anyway. Oh well! C'est la vie, I suppose.

 

[ats5g]

For Want2retire and anyone else interested,

National Organization of Life and Health Insurance Guaranty Associations

Search for your state's association

Search for your state's Guaranty Association Laws

- Alec

 

[aenlighten]

Who needs longevity insurance?

Just because some people will live past 100, doesn't mean you have the same chance. The long lived may benefit but they are likely the ones buying it anyway so it is likely between the long lived and the very long lived. Are your genes and behavior up to it?