A Financial Product I’d Like to See
When I retire I’d like to be able to invest in a fund holding a low-cost index of the world’s stocks that addresses longevity risk. The idea is that it would be like a low-cost annuity based on stock returns rather than bond returns.
The easiest way to describe this idea is first as a simple tontine structure. Imagine a large number of 65-year old women each placing $100,000 into a fund and the money gets invested in a low-cost index of the world’s stocks. Each month the fund sells some fraction of the shares and divides the money among the surviving women.
The dollar value of shares the fund sells would be chosen based on expected stock market returns and mortality expectations. The payouts would be calculated so that if these expectations turn out exactly right, then the monthly payments would rise exactly with inflation. However, the actual monthly payments would be based on actual stock returns and the actual number of surviving women. No money would go to the estates of women who die; this is the trade-off to get higher payments while still alive.
In a real fund of this type, we’d actually open the fund up to different deposit amounts, different ages, men, and couples. We could even introduce the option of payments to an estate after an early death (in exchange for lower payments while alive). We’d need to have actuarial rules to choose fair payout levels. We’d also need rules for how to adjust payment levels if longevity statistics change over time.
Another possible approach to this type of product is to have an insurance company take on the longevity risk rather than having a tontine structure based on actual mortality. The insurance company would essentially promise to deliver a certain number of shares (or more likely the cash value of the shares) each month. The payment levels would be determined by stock market returns and expected mortality statistics (actual mortality numbers would not affect payments).
All these details can be worked out by people skilled at actuarial math. The main thing I want is to get investment returns based on stock investments along with the higher payments that come from eliminating longevity risk.
I wouldn’t put all my money in such an investment, but I would put in a substantial amount when I retire. I’d then maintain a cash cushion to deal with stock market volatility, and I’d keep some fraction of my portfolio to manage on my own. I might also buy a standard simple annuity to serve as a bond-like version of this type of new investment.
The easiest way to describe this idea is first as a simple tontine structure. Imagine a large number of 65-year old women each placing $100,000 into a fund and the money gets invested in a low-cost index of the world’s stocks. Each month the fund sells some fraction of the shares and divides the money among the surviving women.
The dollar value of shares the fund sells would be chosen based on expected stock market returns and mortality expectations. The payouts would be calculated so that if these expectations turn out exactly right, then the monthly payments would rise exactly with inflation. However, the actual monthly payments would be based on actual stock returns and the actual number of surviving women. No money would go to the estates of women who die; this is the trade-off to get higher payments while still alive.
In a real fund of this type, we’d actually open the fund up to different deposit amounts, different ages, men, and couples. We could even introduce the option of payments to an estate after an early death (in exchange for lower payments while alive). We’d need to have actuarial rules to choose fair payout levels. We’d also need rules for how to adjust payment levels if longevity statistics change over time.
Another possible approach to this type of product is to have an insurance company take on the longevity risk rather than having a tontine structure based on actual mortality. The insurance company would essentially promise to deliver a certain number of shares (or more likely the cash value of the shares) each month. The payment levels would be determined by stock market returns and expected mortality statistics (actual mortality numbers would not affect payments).
All these details can be worked out by people skilled at actuarial math. The main thing I want is to get investment returns based on stock investments along with the higher payments that come from eliminating longevity risk.
I wouldn’t put all my money in such an investment, but I would put in a substantial amount when I retire. I’d then maintain a cash cushion to deal with stock market volatility, and I’d keep some fraction of my portfolio to manage on my own. I might also buy a standard simple annuity to serve as a bond-like version of this type of new investment.
If you called in MJ's Money For Life Fund, you'd be able to sell it no problem, you just need some marketing folks to completely blur all information, and make it look like a "magic money tree"(TM) and you are home free.
ReplyDelete@Alan: That sounds like a lot of work. I was just hoping someone else would create this fund so that I could use it.
DeleteIntersting idea. I still think that having the option to purchase "additional" CPP would accomplish much of what you are proposing. All the actuarial and logistic structure is already in place. Making it revenue positive and using any profits to help the less fortunate, would perhaps make it politically palatable.
ReplyDelete@Garth: The difference between the fund I describe and a pension plan like CPP is that (at least for the tontine version), those running the fund take on no risk at all. In contrast, CPP's guaranteed payments create significant risk for taxpayers in the event that CPP is unable to generate expected returns for a long time. I don't mind taking on stock market risk myself if I can eliminate longevity risk.
DeleteTrue, but that same risk for taxpayers exists today. CPP can change the rules if they must, as they have done in the past.
DeleteIf this financial product ever exist, it wont be "low fees" unfortunatly. Remember that in Canada, plain and simple managed mutual funds carry MER of 2% and more and even index mutual funds are in the 0.5-0.7% range wich is still to much if you ask me!
ReplyDeleteThe idea is great because while inflation is a "silent killer" and makes us investing against it, people who die "earlyer than expected"(lets say before 110) will end up with large estate. The price to pay for a safe retirement.
@Le Barbu: Perhaps Vanguard would be interested in a fund or ETF with this structure. They seem to be the best hope for low fees.
DeleteI like this idea, an annuity with a twist. I've read the people with annuities live longer. I think there's a subtle incentive to take care of yourself since you're being paid a monthly amount to stay alive. Entering a tontine with a bunch of people you feel you're competing with might amplify this effect.
ReplyDeleteSeems like a good lawyer could draw up a contract for a group of friends that want to do something like this. I smell a business opportunity here if it could be simply adapted to various groups: softball teams, social clubs, family members, etc.
@Gene: I think there are laws against tontine structures right now, but I'm not sure of the details. I'd prefer large groups because it reduces the incentive to kill off other members of the tontine.
Delete@Michael
ReplyDeleteNot sure if you are still monitoring these older posts...
I just finished reading Milevsky's "King William's Tontine".
I think you would love it!
@Garth: The description I read online sounds interesting. My local library doesn't have it, so I'll add it to my list of books to buy if I get motivated enough.
DeleteThis article by Kitces is very good as well.
Deletehttps://www.kitces.com/blog/tontine-agreement-instead-of-annuity-lifetime-income-mortality-credits-and-book-review-of-milevsky-king-williams-tontine/
@Garth: Kitces are is very good. It seems to me that a tontine with a healthy mix of stocks and bonds and structured to make payments based on certain expectations of returns and mortality could benefit retirees greatly. However, it would have to be run by an entity that didn't skim off so much profit for itself that the retirees lose the tontine benefits compared to other investment approaches. Perhaps some mutual arrangement such as Vanguard's would work. In return for the much higher payments a tontine would offer, I'd be happy to take on some stock market risk along with the risk that people will suddenly start living longer than expected.
DeleteHere is a link to the academic paper by Milevsky and Salisbury. I have not read it but it is supposed to contain most of the info and implementation ideas found in the book...
ReplyDeletehttps://papers.ssrn.com/sol3/papers.cfm?abstract_id=2271259
At it's essence, it is all about accessing mortality credits while bypassing the insurance company's many layers of costs. I agree that a "Vanguard like" approach would seem to be ideal. Milevsky suggests that the actual costs should be less than 1%.
ReplyDeleteThe push back from the insurance industry would be formidable...