Retirement bonds, aka SeLFIES

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Sun Jun 24, 2018 9:45 am

Arun Muralidhar on goals-based investing and the use of financial instruments, such as retirement bonds, that match cash flows.
The growing trend of goals-based investing recognizes that investors hold assets to satisfy key goals — planning for retirement (individual or institutional), saving for a child's education, or purchasing a house or a car. These goals share one interesting, often overlooked, facet: Every investor would like a guaranteed, target level of real cash flows (linked to the inflation of the goal), that starts on a given date and runs for a fixed, foreseeable period of time. ...

There are two investment approaches to goals-based investing, to achieve this goal.

The traditional/mainstream approach takes existing assets and attempts to create "optimized" portfolios, based on modern portfolio theory. Robo-investing or target-date funds utilize this approach. However, this approach is much more elegant in theory than practice because there are no currently available assets with the desired cash flow profile, and the industry's track record for forecasting returns and other inputs for these models is poor, at best. Moreover, most investors are financially illiterate and do not understand compounding, diversification or the impact of inflation. These shortcomings make this approach complex, error-prone, expensive and risky. The problem is well beyond the scope of the average investor and could lead to another pension crisis, as individuals are being asked to take control of their investment decisions in DC plans. ...

The second approach, called the KISS (keep it simple and straightforward) or cash flow-driven approach, takes a very different tack: It recognizes each goal is unique and predominantly about cash flows. Rather than trying to create ever more complex MPT approaches, it focuses on creating goal-appropriate, cash flow-generating financial instruments that investors need, which then trivializes the investment problem and eliminates previously mentioned challenges.
Link to article - http://www.pionline.com/article/2018040 ... ghtforward

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Wed Jun 27, 2018 2:13 pm

Robert Powell writing about retirement bonds for financial security in TheStreet.
Arun Muralidhar is on a mission to solve the retirement crisis. In his just-published book, 50 States of Gray, Muralidhar discusses what Nobel prize winner and MIT professor Robert Merton describes as "innovative proposals for pension design and a new pension bond." …

Part of the problem has to do with the funds used in the defined-contribution plans, specifically target-date funds (TDFs), a qualified default investment alternative, or QDIA. "All target-date funds are doing is switching people from stocks to bonds as they age," he said. "But the challenge with that is that there is absolutely no guarantee of the outcome you'll get if you're invested in a target-date fund." What's more, the TDF fund you buy from one company could be completely different from the one you buy from another company, both of which will give you highly random outcomes defined either as wealth or income at retirement, he said. Just because people are the same age doesn't mean they should have the same portfolio. Ultimately, he says, it's harder to determine how much wealth you'll need for retirement than how much income you might need. Thus, the capital asset pricing model, or CAPM, doesn't work for retirement.

The solution, according to Muralidhar, are bonds called Bonds for Financial Security (BFFS) or Standard-of-Living-indexed Forward-starting Income-only Securities or (SeLFIES). Government agencies, such as the U.S. Treasury, can issue these bonds to improve retirement security and fund infrastructure. The idea behind this new bond is people want certainty about what one gets -- in terms of income -- and when they get it. "It's the exact cash flows that a well-designed bond could have," he said. …

To be fair, pre-retirees and retirees could certainly create a Treasury-Inflation Protected Securities (TIPS) bond ladder to achieve some of what Muralidhar has described. But, Muralidhar says, it's hard to do, and it's complex, and it's illiquid, and it's expensive.
Link to article - https://www.thestreet.com/retirement/ir ... s-14633315

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

grok87
Posts: 8454
Joined: Tue Feb 27, 2007 9:00 pm

Re: Retirement bonds

Post by grok87 » Wed Jun 27, 2018 5:19 pm

bobcat2 wrote:
Wed Jun 27, 2018 2:13 pm
Robert Powell writing about retirement bonds for financial security in TheStreet.
The solution, according to Muralidhar, are bonds called Bonds for Financial Security (BFFS) or Standard-of-Living-indexed Forward-starting Income-only Securities or (SeLFIES). Government agencies, such as the U.S. Treasury, can issue these bonds to improve retirement security and fund infrastructure. The idea behind this new bond is people want certainty about what one gets -- in terms of income -- and when they get it. "It's the exact cash flows that a well-designed bond could have," he said. …

To be fair, pre-retirees and retirees could certainly create a Treasury-Inflation Protected Securities (TIPS) bond ladder to achieve some of what Muralidhar has described. But, Muralidhar says, it's hard to do, and it's complex, and it's illiquid, and it's expensive.
Link to article - https://www.thestreet.com/retirement/ir ... s-14633315

BobK
Thanks Bob.

I think parts of his idea are interesting. I would take issue with the idea that TIPS are illiquid or expensive. They are not as liquid as treasuries but then again nothing is. They are certainly much more liquid than corporates, munis, etc.

this quote from the article is interesting
wrote: What's more, TIPS only give you protection against inflation. "It doesn't give you protection against standard of living changes," he said. "And so that's where we were arguing that if people are trying to create a particular cash flow profile, why have this complex engineering when there is an incentive for the government to issue the original instrument and therefore, why not issue SeLFIES."
the article doesn't elaborate but it sounds like he is proposing the SELFIES be indexed not to CPI but to wage inflation in the same way that Social Security is indexed to wage inflation during ones working years. I think that's an interesting idea but i imagine it would make the bonds very illiquid. i.e. most institutional investors would stick to cpi linked tips and there would be limited demand for bonds linked to wage inflation.
Keep calm and Boglehead on. KCBO.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Wed Jun 27, 2018 9:11 pm

grok87 wrote:
Wed Jun 27, 2018 5:19 pm
this quote from the article is interesting
wrote:What's more, TIPS only give you protection against inflation. "It doesn't give you protection against standard of living changes," he said. "And so that's where we were arguing that if people are trying to create a particular cash flow profile, why have this complex engineering when there is an incentive for the government to issue the original instrument and therefore, why not issue SeLFIES."
the article doesn't elaborate but it sounds like he is proposing the SELFIES be indexed not to CPI but to wage inflation in the same way that Social Security is indexed to wage inflation during ones working years. I think that's an interesting idea but i imagine it would make the bonds very illiquid. i.e. most institutional investors would stick to cpi linked tips and there would be limited demand for bonds linked to wage inflation.
I agree that indexing to living standards rather than inflation is a bridge too far, at least at the beginning. :( It will be hard enough to get governments to issue these bonds initially without adding that twist.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Mon Jul 02, 2018 1:52 pm

Robert Merton writing about SeLFIES retirement bonds in IPE (Investments & Pensions Europe).
These innovations ensure even the most financially illiterate individual can be self-reliant with respect to retirement planning (without requiring a forecast of expected returns, optimisers/retirement calculators, or even intermediaries). For example, if investors want to guarantee $50,000 annually, risk-free for 20 years in retirement, to maintain their standard of living, they would need to buy 10,000 SeLFIES (50,000 divided by 5) over their working life.

The complex decisions of how much to save, how to invest, and how to draw down are simply folded into a calculation of how many to buy. In addition to being simple, liquid, easily traded, and with low credit risk, SeLFIES can be bequeathed to heirs, unlike high-cost, inflexible and illiquid annuities. The inheritability of SeLFIES overcomes investor fears that premature death means leaving money on the table. Buying SeLFIES would be similar to creating an individual DB scheme, with the guaranteed pay-out determined simply by the number purchased. SeLFIES greatly simplify retirement investing by allowing participants to be self-reliant in managing their portfolios. ...

Moreover, SeLFIES could become the safe asset in these target-date strategies. They could also be used as safe, liability-hedging assets in dynamically managed target-income strategies – allowing investors to target a higher retirement standard of living/income by investing in risky assets early in their life cycle, but dynamically locking in gains by investing in SeLFIES. Further, simple statements would illustrate the level of real, locked-in retirement standard of living, based on the number of bonds purchased. In today’s DC plans, statements focused on wealth accumulated give investors no sense of retirement standard of living or what to do to achieve their retirement objectives.
Link to article - https://www.ipe.com/reports/special-rep ... 63.article

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

grok87
Posts: 8454
Joined: Tue Feb 27, 2007 9:00 pm

Re: Retirement bonds

Post by grok87 » Mon Jul 02, 2018 3:08 pm

bobcat2 wrote:
Mon Jul 02, 2018 1:52 pm
Robert Merton writing about SeLFIES retirement bonds in IPE (Investments & Pensions Europe).
These innovations ensure even the most financially illiterate individual can be self-reliant with respect to retirement planning (without requiring a forecast of expected returns, optimisers/retirement calculators, or even intermediaries). For example, if investors want to guarantee $50,000 annually, risk-free for 20 years in retirement, to maintain their standard of living, they would need to buy 10,000 SeLFIES (50,000 divided by 5) over their working life.

The complex decisions of how much to save, how to invest, and how to draw down are simply folded into a calculation of how many to buy. In addition to being simple, liquid, easily traded, and with low credit risk, SeLFIES can be bequeathed to heirs, unlike high-cost, inflexible and illiquid annuities. The inheritability of SeLFIES overcomes investor fears that premature death means leaving money on the table. Buying SeLFIES would be similar to creating an individual DB scheme, with the guaranteed pay-out determined simply by the number purchased. SeLFIES greatly simplify retirement investing by allowing participants to be self-reliant in managing their portfolios. ...

Moreover, SeLFIES could become the safe asset in these target-date strategies. They could also be used as safe, liability-hedging assets in dynamically managed target-income strategies – allowing investors to target a higher retirement standard of living/income by investing in risky assets early in their life cycle, but dynamically locking in gains by investing in SeLFIES. Further, simple statements would illustrate the level of real, locked-in retirement standard of living, based on the number of bonds purchased. In today’s DC plans, statements focused on wealth accumulated give investors no sense of retirement standard of living or what to do to achieve their retirement objectives.
Link to article - https://www.ipe.com/reports/special-rep ... 63.article

BobK
Thanks Bob for keeping us updated on this. Much appreciated.

So in the article he says the idea is to index the bonds not to inflation but to (nominal) aggregate per capita consumption. And then goes on to say:
“Merton” wrote: SeLFIES would require an appropriate measure of consumption to be articulated for the index; specifically, how consumer-durable purchases are treated and whether or not to include leisure time, not normally included in consumption. This is the same challenge embedded in TIPS.
So this idea is even more out there than I thought!
To be honest it smacks to me of a bunch of academics trying to create something unrealistic.

Having something akin to the current social secuirty system of indexing for average wage growth during a deferral period and then for CPI during the payout period would make some sense. Even that Though would lead to a very fragmented secondary market. Tips strips is i think all that is really needed here...
Keep calm and Boglehead on. KCBO.

Telloverture
Posts: 9
Joined: Mon Apr 23, 2018 8:05 am

Re: Retirement bonds

Post by Telloverture » Mon Jul 02, 2018 10:09 pm

This proposal continues Merton’s advocacy of shifting focus away from an accumulation “number” for retirement and instead focusing on a guaranteed income stream (see https://hbr.org/2014/07/the-crisis-in-r ... t-planning.) Similar to a defined benefit pension or social security, you are not meant to focus on the accumulated value but on the income you have secured through your contributions.

Apart from the behavioral/nomenclature aspects, much of what the professors propose can already be done, namely with I-Bonds.

When you break it down, this proposal still has an accumulation phase (time until income starts) and a distribution phase (time during which payments are received.) For an individual investor key issues that need to be addressed are:

1) During accumulation
  • a. Guaranteed return
    b. Inflation protection
    c. Tax incentives
    d. Liquidity
2) During distribution
  • a. Guaranteed payments
    b. Inflation protection
    c. Tax incentives
    d. Liquidity
    e. Longevity
    f. Bequests
For 1) in the U.S. we have I-Bonds. They provide a guaranteed return, are fully indexed to inflation, are tax-deferred, and provide almost complete liquidity at par even in small denominations. All this for a 30-year term. No need for TIPS ladders (unless you want to buy more than the annual I-Bond maximum.)

The biggest problem with I-Bonds is their low real return. The Merton proposal would need to spell out if and how a higher return could be achieved.
In addition, the SelFIES proposal envisions indexing to per-capita consumption instead of inflation. Personally, I can’t readily wrap my head around that one so I’ll leave it alone for now.

For 2) we have immediate annuities, a decidedly imperfect solution. They provide guaranteed payments (the guarantee only as good as the issuing insurer), can provide (expensive) inflation protection, provide only limited tax benefits, provide no liquidity or value at death (unless expensive riders are purchased). Their biggest benefit is that they provide longevity credits (extra return for pooling many annuitants).

So one might think substantively this is where the Merton proposal has the greatest merit – providing a unitized, liquid, government-guaranteed immediate annuity adjusted for inflation that avoids the myriad charges for expenses, hedging, mortality, and profit that insurance companies embed in their annuities.

But of course with a fixed 20-year term, there is no need for an annuity at all. You can just roll over your I-Bond portfolio (into TIPS, if necessary) and draw the balance down over 20 years. You don’t need an insurance company for a period-certain annuity.

Bottom line, I don’t see this proposal adding much to the existing solutions.

BTW I don’t think deferred income annuities fit the bill here either – I haven’t seen any that provide a real return during the waiting period (accumulation phase), though inflation protection of the payments is available (at a significant cost.)

gmaynardkrebs
Posts: 922
Joined: Sun Feb 10, 2008 11:48 am

Re: Retirement bonds

Post by gmaynardkrebs » Mon Jul 02, 2018 11:41 pm

Telloverture wrote:
Mon Jul 02, 2018 10:09 pm
This proposal continues Merton’s advocacy of shifting focus away from an accumulation “number” for retirement and instead focusing on a guaranteed income stream (see https://hbr.org/2014/07/the-crisis-in-r ... t-planning.) Similar to a defined benefit pension or social security, you are not meant to focus on the accumulated value but on the income you have secured through your contributions.

Apart from the behavioral/nomenclature aspects, much of what the professors propose can already be done, namely with I-Bonds.

When you break it down, this proposal still has an accumulation phase (time until income starts) and a distribution phase (time during which payments are received.) For an individual investor key issues that need to be addressed are:

1) During accumulation
  • a. Guaranteed return
    b. Inflation protection
    c. Tax incentives
    d. Liquidity
2) During distribution
  • a. Guaranteed payments
    b. Inflation protection
    c. Tax incentives
    d. Liquidity
    e. Longevity
    f. Bequests
For 1) in the U.S. we have I-Bonds. They provide a guaranteed return, are fully indexed to inflation, are tax-deferred, and provide almost complete liquidity at par even in small denominations. All this for a 30-year term. No need for TIPS ladders (unless you want to buy more than the annual I-Bond maximum.)

The biggest problem with I-Bonds is their low real return. The Merton proposal would need to spell out if and how a higher return could be achieved.
In addition, the SelFIES proposal envisions indexing to per-capita consumption instead of inflation. Personally, I can’t readily wrap my head around that one so I’ll leave it alone for now.

For 2) we have immediate annuities, a decidedly imperfect solution. They provide guaranteed payments (the guarantee only as good as the issuing insurer), can provide (expensive) inflation protection, provide only limited tax benefits, provide no liquidity or value at death (unless expensive riders are purchased). Their biggest benefit is that they provide longevity credits (extra return for pooling many annuitants).

So one might think substantively this is where the Merton proposal has the greatest merit – providing a unitized, liquid, government-guaranteed immediate annuity adjusted for inflation that avoids the myriad charges for expenses, hedging, mortality, and profit that insurance companies embed in their annuities.

But of course with a fixed 20-year term, there is no need for an annuity at all. You can just roll over your I-Bond portfolio (into TIPS, if necessary) and draw the balance down over 20 years. You don’t need an insurance company for a period-certain annuity.

Bottom line, I don’t see this proposal adding much to the existing solutions.

BTW I don’t think deferred income annuities fit the bill here either – I haven’t seen any that provide a real return during the waiting period (accumulation phase), though inflation protection of the payments is available (at a significant cost.)
As you note, Merton spelled much of his thinking out in the HBR article a few years ago. I agree with you that his proposal does not solve the problem of low real returns. It's more about getting retirement savers to focus on the correct metric, which apparently most do not. Why that is the case is a good question. Merton says current law is a prime impediment, and he may be right. My own theory, however, is that the mutual fund industry will not willingly embrace so dramatic a change to the current system, which has rewarded them beyond their wildest dreams, quite possibly at the expense of their clients.

As far as I bonds, I agree with you in theory that they should work, but as you note, the practical limits are far too low. TIPS are too heavily taxed (outside of IRA/401Ks) to be of much use. I think Merton's best point is when he says that no one thinking about a private retirement system on a clean slate would come up with the system we have now, which was designed in a different era, for a different purpose. Merton's proposal has its flaws, but the system we have now is also flawed, and in fact, has failed large numbers of people.

Telloverture
Posts: 9
Joined: Mon Apr 23, 2018 8:05 am

Re: Retirement bonds

Post by Telloverture » Tue Jul 03, 2018 8:05 am

As far as I bonds, I agree with you in theory that they should work, but as you note, the practical limits are far too low. TIPS are too heavily taxed (outside of IRA/401Ks) to be of much use. I think Merton's best point is when he says that no one thinking about a private retirement system on a clean slate would come up with the system we have now, which was designed in a different era, for a different purpose. Merton's proposal has its flaws, but the system we have now is also flawed, and in fact, has failed large numbers of people.
The IBond investment limits are not that low for a median household. How many (joint filing) households are able to save $25k a year? I am not sure how the fixed rate is determined, but this is where perhaps some of the professors' ideas could come into play. Indexing to per capita consumption could provide some incremental real return.

gmaynardkrebs
Posts: 922
Joined: Sun Feb 10, 2008 11:48 am

Re: Retirement bonds

Post by gmaynardkrebs » Tue Jul 03, 2018 8:49 am

Telloverture wrote:
Tue Jul 03, 2018 8:05 am
As far as I bonds, I agree with you in theory that they should work, but as you note, the practical limits are far too low. TIPS are too heavily taxed (outside of IRA/401Ks) to be of much use. I think Merton's best point is when he says that no one thinking about a private retirement system on a clean slate would come up with the system we have now, which was designed in a different era, for a different purpose. Merton's proposal has its flaws, but the system we have now is also flawed, and in fact, has failed large numbers of people.
The IBond investment limits are not that low for a median household. How many (joint filing) households are able to save $25k a year? I am not sure how the fixed rate is determined, but this is where perhaps some of the professors' ideas could come into play. Indexing to per capita consumption could provide some incremental real return.
I grant your point, but there's a real problem right there -- $25k/year in I-bonds at today's real rates isn't enough, even for a median income family. How you get people to save more is a daunting problem in a consumption society like ours.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Fri Jul 06, 2018 8:57 am

Arun Muralidhar will be making a presentation at DC Bogleheads on Sunday July 29 on these BFFS retirement bonds and the MMM flex model of retirement planning. See this link for more information about the meeting.
Link - viewtopic.php?f=9&t=139752&p=4004533&hi ... n#p4004533

If you have questions about these bonds, and cannot attend the meeting, post your questions about the bonds here. I'll cull from the list of questions the best ones and query Arun about them at the meeting.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

Dottie57
Posts: 4656
Joined: Thu May 19, 2016 5:43 pm

Re: Retirement bonds

Post by Dottie57 » Sun Jul 08, 2018 11:10 am

I just looked at my SS statement and see that I put 142k into SS and employers put in 138k. My portion was a bit above 50%.

Total of 280k. Taking this money at age 61 and putting it into an SPDA for 9 yrs would generate about 30k per year. It would be nice to get the Money back with interest and get an SPDA. Without interest I would have a annuity money of about 30 k at age 70.

I used taking the money at 70 because that is when I plan on taking SS.

I am liking the idea of retirement bonds more and more.
Last edited by Dottie57 on Sun Jul 08, 2018 11:18 am, edited 1 time in total.

gmaynardkrebs
Posts: 922
Joined: Sun Feb 10, 2008 11:48 am

Re: Retirement bonds

Post by gmaynardkrebs » Sun Jul 08, 2018 11:18 am

Dottie57 wrote:
Sun Jul 08, 2018 11:10 am
I just looked at my SS statement and see that I put 142k into SS and employers put in 138k. My portion was a bit above 50%.

Total of 280k. Taking this money at age 61 and putting it into an SPDA for 9 yrs would generate about 30k per year. It would be nice to get the Money back with interest and get an SPIA. Without interest I would get an SPIA of about 30 k at age 80.

I used taking the money at 70 because that is when I plan on taking SS.

I am liking the idea of retirement bonds more and more.
What would you get at age 70?

Dottie57
Posts: 4656
Joined: Thu May 19, 2016 5:43 pm

Re: Retirement bonds

Post by Dottie57 » Sun Jul 08, 2018 11:24 am

gmaynardkrebs wrote:
Sun Jul 08, 2018 11:18 am
Dottie57 wrote:
Sun Jul 08, 2018 11:10 am
I just looked at my SS statement and see that I put 142k into SS and employers put in 138k. My portion was a bit above 50%.

Total of 280k. Taking this money at age 61 and putting it into an SPDA for 9 yrs would generate about 30k per year. It would be nice to get the Money back with interest and get an SPIA. Without interest I would get an SPIA of about 30 k at age 80.

I used taking the money at 70 because that is when I plan on taking SS.

I am liking the idea of retirement bonds more and more.
What would you get at age 70?
Right now SS says about 42k plus a bit more. However if SS is cut by the percent the fund is short it would be 30k +.

PaulF
Posts: 148
Joined: Wed Aug 06, 2008 10:07 pm
Location: Wisconsin

Re: Retirement bonds

Post by PaulF » Sun Jul 08, 2018 3:06 pm

Dottie57 wrote:
Sun Jul 08, 2018 11:10 am
I just looked at my SS statement and see that I put 142k into SS and employers put in 138k. My portion was a bit above 50%.

Total of 280k. Taking this money at age 61 and putting it into an SPDA for 9 yrs would generate about 30k per year. It would be nice to get the Money back with interest and get an SPDA. Without interest I would have a annuity money of about 30 k at age 70.

I used taking the money at 70 because that is when I plan on taking SS.

I am liking the idea of retirement bonds more and more.
You do realize that the "I" in FICA stands for "insurance," right?

Dottie57
Posts: 4656
Joined: Thu May 19, 2016 5:43 pm

Re: Retirement bonds

Post by Dottie57 » Sun Jul 08, 2018 6:46 pm

PaulF wrote:
Sun Jul 08, 2018 3:06 pm
Dottie57 wrote:
Sun Jul 08, 2018 11:10 am
I just looked at my SS statement and see that I put 142k into SS and employers put in 138k. My portion was a bit above 50%.

Total of 280k. Taking this money at age 61 and putting it into an SPDA for 9 yrs would generate about 30k per year. It would be nice to get the Money back with interest and get an SPDA. Without interest I would have a annuity money of about 30 k at age 70.

I used taking the money at 70 because that is when I plan on taking SS.

I am liking the idea of retirement bonds more and more.
You do realize that the "I" in FICA stands for "insurance," right?
Yes, I do.

Given the lack of $ in the trust funds with possible cuts in payouts , I am not too happy.

PaulF
Posts: 148
Joined: Wed Aug 06, 2008 10:07 pm
Location: Wisconsin

Re: Retirement bonds

Post by PaulF » Sun Jul 08, 2018 9:45 pm

I'd wager that had you become disabled some years ago, you would prefer the present system to one of retirement bonds.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Sun Jul 08, 2018 9:59 pm

PaulF wrote:
Sun Jul 08, 2018 9:45 pm
I'd wager that had you become disabled some years ago, you would prefer the present system to one of retirement bonds.
These retirement bonds are not meant to be in lieu of Social Security, but in addition to Social Security. They would simply be a new type of federal government bond - much like when the government began issuing new bonds in the form of TIPS in 1997 and I-bonds in 1998. These bonds would be marketable like TIPS or nominal Treasuries.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Thu Jul 26, 2018 1:04 pm

Arun Muralidhar will be making a presentation at DC Bogleheads this coming Sunday July 29 on these BFFS retirement bonds and the MMM flex model of retirement planning. See this link for more information about the meeting.
Link - viewtopic.php?f=9&t=254682

So if you are interested and in the DC area, go to the meeting. All are welcome. If you can't attend but have questions about these bonds, post your questions about the bonds here. I'll attempt to ask them at the meeting.

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Fri Sep 28, 2018 11:13 am

Christopher Robbins writing in Financial Advisor about the proposed new retirement bonds.
“Existing products often offered as default options in the retirement context have no ability to secure retirement goals,” says Martellini. “They can’t generate a fixed level of income in retirement. The investor is buying a lottery ticket, and if they’re lucky, markets will do well and they’ll enjoy good performance, but the negotiating power of their wealth in retirement is completely out of their control.”

Martellini and his colleagues coined the phrase “flexicurity” to define the ideal investment solution for retirees. At heart, most retirement investors want security and a guaranteed stream of income, but they also want the flexibility to adjust their investments and their potential income stream over time. For the retirement-focused portfolio, goals like outperforming other investments or reaching a target asset level are more aspirational than essential.

Products like annuities offer secure sources of income with little flexibility. Other products like bonds and target-date funds offer flexibility with little security.

“There’s no absolute meaning for the concept of a safe asset; it’s very investor and goal specific,” says Martellini. “If I need $100 in 10 years, I should buy a 10-year discounted bond. If I need $10 a year for 20 years, 10 years from now, the safe asset is not the 10-year pure discount bond. It’s a retirement bond-matching portfolio.”
Link to article - https://www.fa-mag.com/news/a-new-retir ... section=47

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds

Post by bobcat2 » Tue Oct 23, 2018 11:15 am

Robert Merton delivers the Dr. RH Patil Memorial Lecture at the National Stock Exchange in Mumbai, India.

Merton's entire discussion is well worth watching. He begins by talking about the development of modern finance beginning with Harry Markowitz and MPT in 1952. Then he moves on to discuss how the field of modern finance can be used to address important global economic challenges.

Merton chooses examples of how financial innovations available today or likely in the near future can be used to to address three global economic policy concerns. One of the three policy concerns is the global retirement crisis. He then discusses at length the proposed new retirement bond - SeLFIES. The portion of his presentation about SeLFIES begins at roughly the 23 minute mark of the lecture. Merton's explanation of the SeLFIE bond is both insightful and clearly described.

Link to presentation - https://www.bloombergquint.com/videos/n ... gs.czO6W74
Merton’s entire talk was about solving some of these issues from the supply side. He was dismissive of financial literacy saying that it is better to design a car that you can just drive rather than teach details of internal combustion to the driver. He says that it is better to design products that an average person can plug and play rather than hold them accountable to learn finance and evaluate products—especially for retirement. This is important because academia informs policy. Policy directs regulation. Regulation constructs markets. Markets design products that you and I buy.

Merton’s latest work is around product design in a government bond that will allow an average person to target her retirement. The solution is called the Standard of Living indexed, Forward-starting, Income-only Securities—or SeLFIES. This is a government bond that begins to pay interest after a certain number of years, for a certain number of years. The buyer targeting her retirement will have to make two assumptions—the year of retirement and how long she will live.

Assume that a 40-year-old in 2018 decides that she will retire in 2038. She currently spends ₹12 lakh a year and would like to maintain this standard of living in real terms when she turns 60. She would also like to have a product that looks after inflation during the 30 years she plans to live post retirement. Merton’s solution is to have a deferred interest paying bond that you can buy in 2018, with the payouts beginning at retirement. The payouts will be indexed to a standard of living index so that buyers don’t see a drop in their lifestyle. The payout of these bonds, says Merton, is exactly the cash flows that a typical infrastructure project throws up—a long investment period with no cash returns and then an annuity of returns over the life of the project. Long term infrastructure projects can [thus] be funded by these bonds.
Links to accompanying articles.

https://blogs.timesofindia.indiatimes.c ... -security/

[urlhttps://www.livemint.com/Money/tC0GQYK2BUyTgbwL ... stick.html][/url]

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

User avatar
triceratop
Moderator
Posts: 5725
Joined: Tue Aug 04, 2015 8:20 pm
Location: la la land

Re: Retirement bonds, aka SeLFIES

Post by triceratop » Wed Oct 24, 2018 6:25 pm

Merton's talk had three components: (1) an overview of how financial engineering has benefited society (2) a proposal for a new type of retirement bond, called a SeLFIE, and (3) a discussion of models of trust in a fintech / AI future. I enjoyed all three sections of the talk immensely.

Merton stressed at the end of his Fintech portion that "everything is a model" and that "whenever someone says 'I can build something that cannot fail', to get off that boat" because everything is a model so "everything can fail, and therefore everybody to give you a complete answer should tell you 'how do you deal with it when it fails'". As an idea, this seems related to a very well-known exchange from The Big Short where (paraphrased for the family forum that this is) the protagonist asks his counterparty how they will take advantage of them.

Merton never addressed this question with respect to the SeLFIES (therefore, his SeLFIE proposal is incomplete), an irony of which he is probably aware. It is immediately apparent to me how the SeLFIES idea can fail but I will not mention it because it takes us outside the allowed topics of this forum. However, I suggest we think about Merton's question posed in the third portion of his talk to the financial engineering product proposed in the second part of the talk.

All that said, it was an incredibly engaging lecture that I am very pleased to have watched. Thank you bobcat2 for posting!

P.S. thanks to jalbert for indirectly leading me to rediscovering this thread!
"To play the stock market is to play musical chairs under the chord progression of a bid-ask spread."

jalbert
Posts: 3824
Joined: Fri Apr 10, 2015 12:29 am

Re: Retirement bonds, aka SeLFIES

Post by jalbert » Thu Oct 25, 2018 5:03 am

Not sure I follow why these bonds should be so much cheaper than other fixed income investments with similar credit risk. Real rates are low in France right now. Unless the govt is subsidizing the yield of these bonds, when the 20-years of payout cash flows are discounted back to a present value using prevailing real rates, the bonds would be anything but inexpensive. If these are replacing govt subsidized pensions then a subsidized rate may be possible.
Risk is not a guarantor of return.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

Re: Retirement bonds, aka SeLFIES

Post by bobcat2 » Thu Oct 25, 2018 12:00 pm

jalbert wrote:
Thu Oct 25, 2018 5:03 am
Not sure I follow why these bonds should be so much cheaper than other fixed income investments with similar credit risk.
Merton didn't say that the bonds are cheaper by producing high returns compared to other bonds. What Merton did say is that these bonds are extremely simple both to understand and to implement. Calculation of how many bonds to purchase before retirement is very simple because these bonds are indexed for both inflation and changes in living standards and there are no 3rd party costs, which would include costs for making those calculations. Nor does the investor need to deal with coupon reinvestment issues between purchase date and retirement date. These attributes make the bonds low expense, because this simple strategy for meeting retirement income goals is easy to implement in that it doesn't require spending a lot of time acquiring financial knowledge and implementing that knowledge and also the absence of 3rd party costs. He is not claiming that these bonds will have higher returns than other bonds with similar credit risks.

For example, if at age 35 and planning to retire at age 65 the investor wants $30,000 per year in safe retirement income from these bonds she needs to buy 6,000 of these bonds between now and 30 years from now. She could purchase nine bonds per pay period (216 bond purchases a year) and be slightly ahead of schedule for meeting her retirement income goal. That's an extremely simple strategy compared to other investment strategies for targeting $30,000 in safe income for retirement.

BobK
Last edited by bobcat2 on Thu Oct 25, 2018 12:21 pm, edited 1 time in total.
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

ThrustVectoring
Posts: 535
Joined: Wed Jul 12, 2017 2:51 pm

Re: Retirement bonds, aka SeLFIES

Post by ThrustVectoring » Thu Oct 25, 2018 12:18 pm

jalbert wrote:
Thu Oct 25, 2018 5:03 am
Not sure I follow why these bonds should be so much cheaper than other fixed income investments with similar credit risk. Real rates are low in France right now. Unless the govt is subsidizing the yield of these bonds, when the 20-years of payout cash flows are discounted back to a present value using prevailing real rates, the bonds would be anything but inexpensive. If these are replacing govt subsidized pensions then a subsidized rate may be possible.
Term premium is a thing, too. These bonds have *extremely* long durations.

And on the note of making the bonds cheaper, you could make these more efficient at funding retirement by adding in an actuarially fair tontine system or other mortality credit system. Basically, these bonds would pay the listed premium if the owner is still alive. Like, one of the problems with standard retirement savings is that people die before they spend down their portfolio, and then the remaining money goes not towards funding retirement, but towards an inheritance. This would fix that problem.

And at that point, you might as well lift the 20-year payout period. In the US, you could easily call these something like Social Security Supplemental Bonds. The feds know your age and gender so they can give you a fair price for $100/mo of income starting at a particular age. It'd be real easy for folks to understand, too.

I think that'd be the ideal system for a lot of folks. If social security isn't enough for you, buy supplemental bonds until the numbers work.
Current portfolio: 60% VTI / 40% VXUS

RetiredArtist
Posts: 34
Joined: Wed Aug 26, 2015 4:38 pm

Re: Retirement bonds

Post by RetiredArtist » Thu Oct 25, 2018 12:38 pm

aristotelian wrote:
Tue May 08, 2018 10:56 am
Isn't paying into Social Security essentially buying "retirement bonds"? Sounds like this would be like adding a voluntary component.

Might Australia have a system like this, in which your employers pay into your government administered "super" (annuity)? Is there a voluntary component? Our system is SO complex, and can lead to financial predation.

User avatar
bobcat2
Posts: 5235
Joined: Tue Feb 20, 2007 3:27 pm
Location: just barely Outside the Beltway

LIVE bonds that can complement SeLFIES

Post by bobcat2 » Thu Oct 25, 2018 12:41 pm

ThrustVectoring wrote:
Thu Oct 25, 2018 12:18 pm
And on the note of making the bonds cheaper, you could make these more efficient at funding retirement by adding in an actuarially fair tontine system or other mortality credit system. Basically, these bonds would pay the listed premium if the owner is still alive. Like, one of the problems with standard retirement savings is that people die before they spend down their portfolio, and then the remaining money goes not towards funding retirement, but towards an inheritance. This would fix that problem.

And at that point, you might as well lift the 20-year payout period. In the US, you could easily call these something like Social Security Supplemental Bonds. The feds know your age and gender so they can give you a fair price for $100/mo of income starting at a particular age. It'd be real easy for folks to understand, too.

I think that'd be the ideal system for a lot of folks. If social security isn't enough for you, buy supplemental bonds until the numbers work.
:thumbsup :thumbsup


Actually Arun Muralidhar has proposed adding a retirement bond beyond normal life expectancy that looks something like a cross between a life annuity and a tontine. He calls that bond a Longevity-Indexed Variable Expiration (LIVE) bond. Here is a link to his preliminary paper on the LIVE bond.
Managing Longevity Risk – The Case for Longevity-Indexed Variable Expiration (LIVE) Bonds

from the Abstract -
... we suggest governments should create Longevity-Indexed Variable Expiration (LIVE) bonds. These bonds, targeted to individuals (and institutions) would pay income-only, and start paying only after the average life-expectancy of the economy (having addressed retirement income through life expectancy with a complementary BFFS/SeLFIES bond). Each bond will be cohort specific and based on tax collections of that cohort. In this fashion, the government is fully hedged (because the bond will be a form of a collateralized debt obligation), and hence a natural issuer, with low credit risk. Since BFFS/SeLFIES cover the life-expectancy of those living less than the average, only those individuals who live beyond the average (usually richer portions of the population) and with limited resources need purchase LIVE bonds.
Link to paper - https://papers.ssrn.com/sol3/papers.cfm ... id=3224236

BobK
In finance risk is defined as uncertainty that is consequential (nontrivial). | The two main methods of dealing with financial risk are the matching of assets to goals & diversifying.

jalbert
Posts: 3824
Joined: Fri Apr 10, 2015 12:29 am

Re: Retirement bonds, aka SeLFIES

Post by jalbert » Thu Oct 25, 2018 1:21 pm

Term premium is a thing, too. These bonds have *extremely* long durations.
Was there a proposal to auction them so that the market would establish the term premium?

If someone were buying one at age 25, the bond would have a 60-year life and both the govt and individual would be taking the risk of a mispriced term premium, each in the opposite direction, as there is no bond market to price that term premium, and it is unclear how accurate it would be if there were.

While I think these are interesting as a way to simplify the planning around funding future real liabilities, they also look like a prelude to replacing defined benefit govt-sponsored pensions with defined contribution retirement bonds, much as defined contribution plans like 401Ks have replaced private sector pensions.
Risk is not a guarantor of return.

Post Reply