Wade Pfau on SPIAs

[Nords]

I really enjoy reading Wade Pfau's blog, and sometimes I wonder if we're watching a new generation's economist spreading his wings to take the financial research world by storm. I think the next big advances in retirement forecasting success will rely on being able to model variable withdrawal rates and various types of [-]annuities[/-] "insurance" against portfolio failure.

Or maybe he's perfectly happy to use his blog to crowdsource his research a little before he throws it to the FPA Journal's peer-review wolves.

His latest post explains why you might actually want to buy a SPIA now, before interest rates go up:
Pensions, Retirement Planning, and Economics Blog: Are Annuities (SPIAs) Okay When Interest Rates are Low?

I think every retiree should annuitize a portion of their portfolio, even if it's just Social Security or a military pension, but I wouldn't have considered this to be a buying opportunity. The logic is counterintuitive but worth considering.

Note the extensive quoting of Bogleheads posters.

 

[Onward]

Good article. It's an interesting point about declining mental faculties as a reason to shop for an annuity sooner.

 

[73ss454]

I can't figure out annuities and I think I still have all my marbles. Can't imagine trying to figure them out at 75.

I guess if I don't have any of my marbles left and have no idea what's going on I won't need any money. I'll let the state take car of me I guess.

 

[MichaelB]

Thanks for the link, Nords.

 

[rescueme]

For me, one of the key points noted is "You don’t buy annuities because they are good investments, you buy them because they provide a guaranteed income for life." He's correct. An SPIA is insurance, not an investment. Unfortunately a lot of folks don't understand that fact. An SPIA is akin to any other income vehicle, such as a defined benefit (e.g. pension) plan or Social Security, and get tripped up on trying to equate it in the same manner as an investment

I'm one of the strange folks that purchased an SPIA at an early age, at a low (but higher than today) interest rate. However in our case, rather than covering an "income gap" later in life, we're using it to cover an income gap in early retirement, due to no pension and caused by our own decision related to delaying SS. This will result in "trading up" from the SPIA to a superior inflation adjusted annuity (AKA SS) along with not wanting to just depend only on our portfolio for current retirement expenses. Additionally, the additional SPIA income (regardless of value due to inflation impact) will just be icing on the cake for us.

It's not a tool for just old folks (well, much older than DW/me) but it may work, depending on your financial plan/desires. You have to have a defined reason to use such a financial vehicle rather than have it on your "checklist" for something to execute at a certain time of life...

 

[chemist]

Nords I agree Wade Pfau is a brilliant young man. I loved his paper where he finally answered the paradox of why a guy retiring in early 2008 with $2MM can withdraw $80K a year whereas the same guy if he waited to retire until early 2009 with $1MM can only withdraw $40K. He also eliminated the paradox by figuring the savings rate required as a percent of income to retire. His papers are always thought provoking and he explores novel ground.

 

[RetireBy90]

I think every retiree should annuitize a portion of their portfolio, even if it's just Social Security or a military pension, but I wouldn't have considered this to be a buying opportunity. The logic is counterintuitive but worth considering.

I agree - but torn between discussing this back and forth (I do like to argue) and dropping it cause it seems to be a religous issue and I've never been able to convince anyone anyway.

I was moved by the points about the cost of delaying a purchase while waiting for rates to return to more normal ground, in that I had never given much thought to 2 points in particular (in my own words here): 1) you loose out on the protection while you wait, and 2) you are going to suck up the low rate on your bond investments anyway, so your just transferring the sux rate from your bonds to the annuity.

Thanks for the link, interesting and provoking read.

 

[Midpack]

Interesting with some useful thoughts, and I have read (and linked here to) Wade Pfau articles before. Unless the market provides historic or greater returns, I can't rule out SPIA's as part of my portfolio one day. And the longevity insurance aspect of SPIA's can't be cost effectively duplicated by an individual investor that I know of. If you want to maximize spending and die broke or with a fixed bequest, a SPIA is probably the only way. Without a SPIA, my plan could fail early or leave a large estate - not desirable in my case. But the articles seems to leave out three key determinants.

1. What age group is this blog aimed at? Waiting for interest rates to increase to lower annuity cost (he seems to phrase it as increasing payouts which is backwards IMO) may not be prudent for someone who is 70 years old, but at 55 I think it would be wise to wait in most cases based on...
2. ...he doesn't address relative nest egg size. For someone at 65 for example, if their plan has them withdrawing 5% to meet their income needs, they probably can't risk waiting. For someone age 65 withdrawing at 2%, I think it would be a mistake to annuitize much if at all, especially if Soc Sec is/will be available to them. How to know when to annuitize has been very well spelled out by Otar in Unveiling the Retirement Myth and Fullmer http://www.schulmerichandassoc.com/Modern_Portfolio_Decumulation.pdf. Knowing when is the key overriding issue IMO, and it requires being proactive.
3. A lesser point, but he mentions that by waiting for interest rates to rise the portfolio may suffer a loss (on bonds), and that's almost certainly true.
   1. It's implied in his chart showing SPIA returns vs age, but he doesn't come right out and say - all else being equal, the cost of an annuity for a given annual income decreases with age. Again, showing % return increasing with age is a confusing way to make his point IMO. The objective with an annuity is to secure a certain annual income stream (and see what it costs at various ages, interest rates), not a fixed upfront annuity cost at any age (for a higher annual income).

Again, I agree Wade Pfau is a great resource. This work seems incomplete?

 

[Nords]

Again, I agree Wade Pfau is a great resource. This work seems incomplete?

It's a blog, and I'm sure we'll read the full research in the FPA Journal someday soon...

 

[Bikerdude]

I guess if I don't have any of my marbles left and have no idea what's going on I won't need any money. I'll let the state take car of me I guess.

Did you mean to say " I'll let the state take the car from me I guess."

 

[73ss454]

Nah, miss spoke. They won't get my car, that's for sure. I meant "care of me" (heh)

 

[Nords]

Again, I agree Wade Pfau is a great resource. This work seems incomplete?

Ah, here you go:
Pensions, Retirement Planning, and Economics Blog: The Safety-first, Goals-based Approach to Financial Planning

Midpack provided a good critique. For some of her points, it is just a matter that the analysis is still incomplete and not comprehensive. But one really good point in particular is that my graph showing how the annuity payout rate increases with age is not a particularly useful way of looking at the information. Retirees may have in mind a set amount of spending they wish to obtain from their annuity, and so what they care about is how much it will cost to obtain this annuitized income stream as they age. The following figure shows the cost of purchasing $1 of real income for each year that one lives as a function of age:
Any reading suggestions or other comments are appreciated. Midpack has earned a spot in the acknowledgments section of any research article I might eventually write about this.

 

[Independent]

This makes a lot of sense to me:

Boglehead alec also makes a good point that if you are waiting for interest rates to rise, and then they do rise, you will probably experience capital losses on your bond portfolio and so, again, you don’t get any benefit from the higher payout rate since you have less assets by that time.

I don't know if I've posted that idea on another thread, but I think about it every time I read that today isn't a good time to buy.

It assumes, of course, that the money you will spend on a fixed SPIA will come from the fixed interest side of your portfolio, but it seems to me that's what a typical AA would dictate. Financially, an SPIA is a fixed income security with a longevity kicker.

(FTR, in our case, it appears that we can get enough annuity by deferring SS, and that has a better payout rate.)

 

[haha]

This makes a lot of sense to me:



I don't know if I've posted that idea on another thread, but I think about it every time I read that today isn't a good time to buy.

It assumes, of course, that the money you will spend on a fixed SPIA will come from the fixed interest side of your portfolio, but it seems to me that's what a typical AA would dictate. Financially, an SPIA is a fixed income security with a longevity kicker.

(FTR, in our case, it appears that we can get enough annuity by deferring SS, and that has a better payout rate.)

The only thing is, an FIA has a rather long duration, whereas an individual may keep his other fixed income assets' average duration very short if he so desires. The other difference is that some think if trends change they can get out of bond funds or treasuries fast. While this may or may not be true, it is definitely not true for a fixed annuity.

Ha

 

[jdw_fire]

The only thing is, an FIA has a rather long duration, whereas an individual may keep his other fixed income assets' average duration very short if he so desires.

Ha

your principle will still be reduced by taking monthly WDs while waiting to buy the annuity

... The other difference is that some think if trends change they can get out of bond funds or treasuries fast. While this may or may not be true, it is definitely not true for a fixed annuity.

Ha

getting out of the annuity isnt the point. you are getting into the annuity because you want to put the longevity risk off on someone else (the insurance company). this was only a discussion of timing (i.e. when to buy the annuity, not if); will delaying the purchase be more financially advantagous than buying immediately.

 

[haha]

your principle will still be reduced by taking monthly WDs while waiting to buy the annuity




getting out of the annuity isnt the point. you are getting into the annuity because you want to put the longevity risk off on someone else (the insurance company). this was only a discussion of timing (i.e. when to buy the annuity, not if); will delaying the purchase be more financially advantagous than buying immediately.

Thanks so much.

 

[Independent]

The only thing is, an FIA has a rather long duration, whereas an individual may keep his other fixed income assets' average duration very short if he so desires. The other difference is that some think if trends change they can get out of bond funds or treasuries fast. While this may or may not be true, it is definitely not true for a fixed annuity.

Ha

Both true. I should have specified "higher duration" bonds because that's where the ins company is going to invest the money.

And, the SPIA eliminates changes down the road. My guess is that this is the primary reason that people don't use them - don't want to "lose control" is a bigger factor than standard calculations. I wasn't trying to cover all the pros and cons, just the one comment about getting a better deal by waiting for interest rates to go up.

 

[rescueme]

your principle will still be reduced by taking monthly WDs while waiting to buy the annuity...

That was one of the "big things" for us. What will you use as your source of required income until things "turn around". Wait on better interest rates, wait on better equity/bond returns, wait on whatever - and life goes on but you still have the requirement for a base income stream of which flux is not desired.

Sure, we could have taken the risk of using alternatives (CD's, Bonds, whatever) for early retirement income, but one of the big advantages is we don't need to "time" any withdrawl from an "alternative investment" (regardless of market risk).

Today's the 15th. We get a deposit to our MM account (to help pay current retirement expenses). Next month (on the 15th), the same thing will happen. It's the same as any pension, SS or other "annuitized" product does, for the rest of your life.

There are no decisions to be made, or action required on our part. Just another reason we went with an SPIA rather than going alone...

 

[Midpack]

Finally read the response today. While what he says is most certainly correct, the odds are interest rates are going to increase in the years ahead (who knows when though), so it's unlikely the "wash" he describes is going to occur. Interest rates almost can't go any lower for the foreseeable future unless I'm missing something. I would think the benefit of the more likely interest rate increase would at least deserve equal mention.

To understand this, consider a 65-year old male. The cost of obtaining a real annuity of $1 is $18.48.* For a 70-year old male, the cost has fallen by $3.35 to $15.13. That’s an 18% drop. But that is assuming interest rates remain constant at 1%. If interest rates change during those 5 years, so would the 70-year old numbers.* That might be a wash though, as interest rate decreases would make the annuity more expensive but also boost the returns on your bond holdings. You would also need to consider whether your unannuitized wealth would hold its value or not over those 5-years as you spend it down and experience the unpredictable market returns. After all, you did miss $5 worth of real income from the annuity by waiting, and that presumably needs to come from your portfolio instead. These combined factors would tell you, mathematically and probabilistically speaking, whether it is worthwhile to wait.

 

[haha]

Finally read the response today. While what he says is most certainly correct, the odds are interest rates are going to increase in the years ahead (who knows when though), so it's unlikely the "wash" he describes is going to occur.

Assuming equally high credit quality of the annuity and the existing bond portfolio, the asnwer to "a wash" or not a wash depends entirely on the average duration of the existing bond portfolio. If portfolio duration is 5 years, that means that it will take 5 years before better returns on new bonds will compensate for capital losses on old bonds.

If you are selling a product that is less useful with a very low interest rate environment, you have to think of a reason to offer why that doesn't matter. But it almost always does.

The last auction of new 30 year TIPS showed the high yield at 0.77% WooHoo!

Ha

 

[Gone4Good]

your principle will still be reduced by taking monthly WDs while waiting to buy the annuity

And your SPIA payout increases as you "consume" your average life expectancy while you wait. My guess is that this is pretty close to a wash.

 

[Gone4Good]

Assuming equally high credit quality of the annuity and the existing bond portfolio, the asnwer to "a wash" or not a wash depends entirely on the average duration of the existing bond portfolio. If portfolio duration is 5 years, that means that it will take 5 years before better returns on new bonds will compensate for capital losses on old bonds.

Agreed.

And considering I can get a near-zero durration 5-yr CD from Ally paying about as much as a 10-year treasury bond, my incentive to go extremely long duration and lock in today's rate is close to zero. The only way it makes sense to do that is if you think rates are going to go even lower over the next five years.

 

[jdw_fire]

And your SPIA payout increases as you "consume" your average life expectancy while you wait. My guess is that this is pretty close to a wash.

i see you agree with the point that it is a "wash" and therefore it doesnt really pay to wait for higher interest rates to buy the SPIA.

 

[Midpack]

i see you agree with the point that it is a "wash" and therefore it doesnt really pay to wait for higher interest rates to buy the SPIA.

No where near that simple for reasons that have been explained already. If you're 65 years old @ 6% WR, unfortunately you probably can't risk waiting to annuitiize at least somewhat.

If you'd like to explain the case for not waiting - for a 55 year old @ 2.5% WR with Soc Sec in the future and knowing interest rates will increase eventually (they are not going to decrease meaningfully next), it would be instructive.

 

[MooreBonds]

No where near that simple for reasons that have been explained already. If you're 65 years old @ 6% WR, unfortunately you probably can't risk waiting to annuitiize at least somewhat.

If you'd like to explain the case for not waiting - for a 55 year old @ 2.5% WR with Soc Sec in the future and knowing interest rates will increase eventually (they are not going to decrease meaningfully next), it would be instructive.

For a 55 year old living off of a 2.5% WR, it may not make sense.

The only reason I would purchase a SPIA is for the same reason I put $7,000 into a deferred annuity in my late 20s - to diversify a tiny bit. Some people like me enjoy having a very broad diversification, and might see "peace of mind" value in having a small part of their expenses hedged with a fixed income allocation. The 'yield' for a 55 year old from a SPIA would be roughly 4% or so, which isn't that far from an approximate average of short-term interest rates over the past 30 years. What short-term rates will average over the next 30 years is anyone's guess, but as rates rise at some point in the future, you have to average out the current yield against the maximum likely yield down the road. As technology and the ease of moving money across the globe only advances, you also have to factor in that impact on fixed income financial markets (i.e. it's easier for lenders to have access to a greater pool of capital, which will tend to ease rates lower rather than higher over time).

If you have complete confidence in your AA/portfolio, and ability to maintain a 2.5%-3% WR from here on out, and don't desire to have that diversification "peace of mind" aspect (as well as potentially wanting to maximize the benefit for your estate), then I'd say a SPIA is probably not for you.