Tontines may make sense despite their history of disrepute | Hacker News Reader

Tontines may make sense despite their history of disrepute

washingtonpost.com

187 pointsgjvc184 comments

Previously on HN:

https://news.ycombinator.com/item?id=10299482

At least we can reuse that debaitified title.

I don't see how this is really different from an annuity beyond getting a bonus payment when someone dies and not being funded in a way that you're sure the money won't run out prematurely.

With an annuity, a company sells you a plan that will pay you $X/mo for a lump sum of $Y. Those that die below actuarial estimates end up paying in more than they get out while those that continue living get the profits from it.

I guess the difference is that with an annuity, you're expecting the monthly payments to be pretty close to what you're paying in over a normal life expectancy. So, a million dollar annuity with a 10-year life expectancy might net you $8,000-$8,500/mo. By contrast, a tontine would see you put in a million dollars and then maybe get $3,000 per month with a 10-year life expectancy with those that outlive others getting more money as time goes on. So, if you live long and others die, your payment might go up to $10,000 or $15,000. Is that the point of the tontine?

But why on earth would that be the future of retirement? An annuity is predictable. The risk is borne by the company selling it. With a tontine, you get unlucky and everyone lives long and you're in a bad situation. In the best case, your payments go up, but why would you want the chance at $10-15k after others have died rather than $8k predictably always?

An annuity already spreads the risk and uncertainty of life expectancy. That's the problem we're trying to solve. An annuity already lets you profit off those who die early who have paid in more than they take out - and if you live long you get more than you paid in. If you want to add the morbidity of the tontine to your retirement plan, take out life insurance policies on your friends and cash in when they die young and lose money if they live a long life. Combined with an annuity and you basically have the tontine. You get regular payouts from the annuity and you profit when your friends die via life insurance policies on them.

But looking at that, why bother with the life insurance policy? Why wouldn't you just get the annuity and have a predictable, steady income?

* You might not like companies that sell annuities and think they aren't a good investment. However, they're sound compared to some hacked-together, crowd-sourced thing that isn't backed by someone reliable with deep pockets. I won't personally buy an annuity, but they're based off spreading the risk in a conservative enough way to guarantee predictable funds (along with a reasonable profit for the company). An under-funded tontine that has people live longer than expected would simply run out of money and leave people destitute.

The thing that I feel is relevant is: "The risk is borne by the company selling it." That risk means that they have to give you a very bad rate to cover their risk and still make a profit. With a Tontine the risk in the return is borne by the individual who can therefore share in the upside as well as the downside.

"The risk is borne by the company selling it" is true of all insurance contracts. That isn't enough to explain why people dislike annuities but buy other types of insurance.

In the US, I think it's fair to say that the typical insurance policy is required somehow or another.

I guess there are lots of people that rationally purchase homeowners insurance, but there's also lots of people that purchase it as a requirement of their mortgage. Auto and medical both have lots of government rules that push people to purchase them.

Yes, but the compulsory liability insurance in my state is something ridiculously low (like $20K, IIRC). I carry a significant multiple of that, which is an example of buying optional insurance.

Life insurance is another purchase that's typically voluntary.

Different kinds of risk. Normal insurance doesn't care if the stock market drops 30%. Something guaranteeing payout for decades has to buffer against that possibility.

I thought much the same thing. The price and returns of the annuity assumes that some people will die before life expectancy, and others will die later. I don't see what's unethical about that.

There are 3 big problems with pensions. Firstly, people contribute too late, so they don't get the benefits of 50 years of compound interest and stockmarket booms. Second, many companies mismanage them - see the current Philip Green at BHS debacle in the UK for an example. Thirdly, too many people don't contribute anything at all (for whatever reason), but still need some kind of income in their old age.

I don't see how tontines get round any of these problems.

This is completely correct. The only reason these sorts of silly financial instruments are available, and the only reason you're not at the top of the thread, is because the level of basic economic literacy is so low.

The question of why annuities are relatively unpopular is an interesting social/institutional one. From talking to some financial planners, it sounds like some crooked origins gave them a bad reputation (like used cars) that they have struggled to shake. Even today there are an unusual number of scummy annuities, although it's much lower than in the past. I suspect it's because they have several moving parts, and are a bit harder for the financial unsavvy to value and compare between companies, so the market gets more swindlers.

EDIT: I previously also wrote "and the only reason this article is in the Washington Post without directly addressing these points," which is incorrect, as roymurdock helpfully pointed out.

The article touches on this somewhat, but I think said differently:

In an annuity, you're gambling against your own death.

In a tontine, you're gambling against the deaths of others.

From a purely cultural perspective, I think most Americans would rather gamble against the deaths of others.

The article addresses both your point and the GP's point directly:

Economists have long said that the rational thing to do is to buy an annuity. At retirement age, you could pay an insurance company $100,000 in return for some $5,000-6,000 a year in guaranteed payments until you die. But most people don’t do that. For decades, economists have been trying to figure out why.

James Poterba, an economics professor at MIT who has extensively studied American retirement, says it’s still a mystery why annuities are so unpopular. A number of theories have been advanced — people might like to leave some money for their children, or they might worry about medical expenses late in life. Poterba and his colleagues have also pointed out that Social Security is already like an annuity, promising constant payments for life.

But there’s also some evidence that people just irrationally dislike annuities. As behavioral economist Richard Thaler wrote in the New York Times: “Rather than viewing an annuity as providing insurance in the event that one lives past 85 or 90, most people seem to consider buying an annuity as a gamble, in which one has to live a certain number of years just to break even.”

Here is where tontines come in. If people irrationally fear annuities because they seem like a gamble on one's own life, history suggests that they irrationally loved tontines because they see tontines as a gamble on other people's lives.

I agree that financial literacy in the US is low, and the basic accounting/finance classes should be required at the high school level, but this article is good and draws upon a decent amount of actual economic literature. I recommend reading it to the end.

I actually did read this article to the end back when it was written last year, but I wrote the above comment based on a faulty memory. Thank you for the correction. The problem with the article is less than I claimed, but it is still written in a confusing and sensational way.

I'd bet it's mostly to do with fear of sudden, large expenses. Part of the reason to save up a large stock of funds is to be able to pay off big, unexpected bills. To shell out a large sum for an annuity also requires them to shell out (some?) of that ability. The elderly probably couple that with a disdain for taking on financial debt.

I'd like to put forward another possible reason why someone might opt to not get an annuity...

If you are financially savvy, or can pay someone else to be, does it not make more sense to invest your money yourself? There are a number of "too big to fail" companies, like GM, and Wells Fargo, that offer dividends. If you composed a portfolio of high dividend, stable, stocks... Could you theoretically have a "guaranteed" monthly income AND the ability to liquidate, in the event of an emergency?

Perhaps the guaranteed payments of an annuity exceed the guaranteed returns of an investment portfolio. The companies managing the annuities have the capital to invest in businesses and ventures that a typical person can't. Such as financing, large real estate developments, or investments in startups, small businesses, or private equity.

With a stock portfolio, the only guarantee cash-flow is dividends since capital appreciation varies a lot year-to-year. Dividends are usually in the 2-3% range. A large, diversified real-estate portfolio might see 8-15% yearly returns in cash, which can be distributed, saved, or reinvested as needed.

> If you composed a portfolio of high dividend, stable, stocks... Could you theoretically have a "guaranteed" monthly income AND the ability to liquidate, in the event of an emergency?

No. You can buy an annuity, or US Treasury bonds, but guaranteed high returns don't exist, and adding in a liquidity requirement doesn't help.

Note that GM declared bankruptcy in 2009. Even if it is still "too big to fail," that doesn't mean investors can't lose all their money.

I never claimed guaranteed high returns. An annuity doesn't guarantee you "high" monthly income. It guarantees you AN monthly income.

My comment was to address the question regarding why more people don't use annuities, and I'm not sure your response counters that.

Regarding GM, yes, I'm aware they went bankrupt. But correct me if I'm wrong - Investors didn't lose their shares? The government bailed them out, and they're operational today. That's what I mean by "too big to fail".

> In the case of GM, investors who held shares of the company before its reorganization saw their equity holdings cancelled in March, 2011. Per the company's reorganization plan, they did not receive any shares of the "new" GM.

http://www.fool.com/investing/general/2016/02/23/what-a-corp...

No, that doesn't insure against old age. Consider this method of getting a guaranteed income: take your pot of money M and divide it by N, where N is the maximum number of years you could expect to live (if, say, you lived to 121). Then you can get a guaranteed income of M/N per year. But this is silly, because most of the time you die much earlier, and the money goes to waste.

The whole point is to transfer money from worlds where you die early to worlds where you die late. An annuity does this by pooling your money with many people living to different ages. For a large pool and accurate actuarial tables, the number of people who die every year can be predicted, so one can determine an amount of money to pay out each year such that each surviving person gets the same amount each year, with long-surviving people getting more total money (as desired).

I really wanted to like annuities. It's a great concept, but when I looked into annuities, they were all pretty bad deals as compared to just investing in the market yourself. It is true that you offload some risk, but the price you pay for lowering your risk is just too high.

Can you quantify that at all? What's "too high"?

The unpopularity of annuities seems like a direct consequence of loss aversion, which says losses are more psychologically powerful than equivalent gains. [1]

If I trade $100k in savings for an annuity, I'm starting off at a deep "loss", which becomes a "gain" only at some point in the uncertain future. Not surprising that most people won't take that bet.

[1] https://en.wikipedia.org/wiki/Loss_aversion

A lot of people also don't understand annuities. They don't understand how their bank (or whoever) can guarantee them a paycheck every year at a certain amount. It seems it's often not explained as a form of insurance.

Also I understand them but the rates seem a bit rubbish. I just got a quote for a 55 year old and they offered 3.7% (this is in the UK). For comparison you could buy rental property and get a yield like 6% plus capital appreciation. With property, you suddenly want a lump sum, you can sell or mortgage. With the annuity you are out of luck. Also the rent goes up with inflation, this annuity I think not.

This Vanguard annuity paying $1k/month starting at age 65 costs $168k, i.e., 7.1%.

https://investor.vanguard.com/annuity/fixed

If you just directly attached this to a dummy pot of $120k, you'd have a $1k/month annuity at age 55 for $288k, which is already 4.2%. It also comes with the significant bonus that your estate keeps the $168k, plus what's left of the $120k, if you die before 65.

So basically, I just think there are much better annuities out there than what you found.

I meant to reply earlier, but <something, something, work>. Anyways, now that I'm off:

jessriedel found an annuity with a good return. But I also have another point to make:

Annuities mitigate risk. This is the same reason people invest in various government bonds (directly or wrapped up in other investment packages). You're getting a guaranteed return on an annuity. Your rent (and the underlying property) are not guaranteed. The risk in them offers greater opportunity for profit, yes. So why not do both? If you have the money, you ought to consider both (if you can find a good annuity or similar investment).

If you don't have enough money, your choice will come down to your risk tolerance. I'm 33. I'm physically fit, rarely ill, with no debt, a college degree, a stable job, and even if it weren't stable I'm a more than competent programmer and mathematician so I can find work somewhere. I can invest in riskier things than, say, my 60 year-old parents with the almost paid off house, who would have to move (emotionally challenging at their age) to find gainful employment (also difficult at their age) should they lose their money in a bad investment. For them (also note: annuities are targeted at this age group due to the way the sellers of them make profit off of them), an annuity with a decent return is better than purchasing and renting out a place if they have to choose one.

If they have the resources, they don't have to choose. If you're young, take the high-risk investments, as you age move your earnings to safer, stabler, lower earning investments, and retire with confidence.

Isn't there some counter-party risk that the bank (or whoever) might go under? With regular insurance, that's not really an issue--I'll just find another provider--but for an annuity, your money would be gone.

This risk also pretty hard to evaluate, as everyone seemed to discover around 2009.

Having worked both in Finance on Wall Street - for an ethical shop with quality employees in a modest market (Munis) - and in Insurance/Risk Management, I can off-hand say that I think the actuaries and quants in Insurance are easily comparable in talents and abilities with Finance peers. Thus, I think some 'structured products' from the Insurance industry - rather than Finance as most people might think of it - are worth consideration. Sure, both avenues are regulated, but scams will show up in any industry. What I'm getting at is Insruance isn't supposed to be "flashy" - the numbers are supposed to work.

As an aside, I do see the two trying out new, potentially risky synthetics called Insurance Linked Securities. Not sure it has been much of a market in the past - existed, sure, but there's a lot of sloshing Central Bank injected money looking for returns...and pension funds...and, well, everybody. Eek!

A few potential reasons they're interesting compared to annuities: they're more honest about the risk, the risk is easier to understand, the risk is incremental rather than binary, and the lottery aspect might encourage better saving behavior/make them more marketable than annuities.

Annuities are typically sold as "risk free", but you have the risk of the insurance company becoming insolvent. A tontine's risk of everyone living longer is obvious. Since annuities have general credit exposure to the insurance company, to understand the credit risk you need to understand all of the insurers lines of business. A tontine with segregated assets doesn't depend on the other business of the manager for solvency. Some of these issues with annuities can/are mitigated by regulation, but that's another thing you have to understand in order to understand the risk. Customers understanding the risk should lead to better decisions about how much to rely on it.

Because annuities have guaranteed payments, if life expectancy increased the insurance company would continue paying out an an unsustainable rate, and when it ran out of money benefits would go from 100% to 0%. A tontine would instead reduce payments incrementally so that it's always sustainable.

Finally, people don't always make financial decisions in the way economic theory suggests they ought to. In particular, people buy way more lottery tickets than economists think they should, and way fewer annuities. If you sprinkle a little bit of lottery into an annuity it might become more marketable, leading to better average financial decision making.

My favourite example of death related investment is the French housing system, viager. You can buy the rights to a house, but you only get the deeds once the current owner dies. In return for paying the current (often elderly) owner a modest monthly rent plus a lump sum, you get the house for a low price (e.g. 50% off). The lump sum is a fraction of the discount price and the rent is, keeping it morbid, based on French life expectancy.

The fun part is you aren't allowed to ask explicitly how healthy the owner is, so it's a total gamble. You might be stuck paying rent to someone who lives until they're 100 or they could die in an "accident" tomorrow. So in addition to checking the walls for damp when you go round, buyers surreptitiously look for medication and disability aids. Presumably some people also go to the trouble of stalking the current owners to gauge how close they are to the coffin.

To make things more exciting, if your parents bought a house en viager and they leave it to you in their will (but the owner hasn't died yet), you're on the hook for the payments. If you don't keep them up, the owner can re-sell the house.

http://www.connexionfrance.com/explaining-the-viager-system-...

If you're unlucky, you end up with http://www.nytimes.com/1995/12/29/world/a-120-year-lease-on-... (1995):

"Andre-Francois Raffray thought he had a great deal 30 years ago: He would pay a 90-year-old woman 2,500 francs (about $500) a month until she died, then move into her grand apartment in a town Vincent van Gogh once roamed.

But this Christmas, Mr. Raffray died at age 77, having laid out the equivalent of more than $184,000 for an apartment he never got to live in.

On the same day, Jeanne Calment, now listed in the Guinness Book of Records as the world's oldest person at 120, dined on foie gras, duck thighs, cheese and chocolate cake at her nursing home near the sought-after apartment in Arles, northwest of Marseilles in the south of France."

So she lived in a nursing home for 10 years but still didn't let the guy live in the apartment? I know she had the right to it but come on...

If she wanted to give someone money, wouldn't it be better to give it to the poor?!

He has no rights to the house until she dies. This is perfectly within her contract with him.

So she can get his payments, and money from a subletter? I guess it's part of the game.

Yes, it is. It's still also her being a douche.

Contract rights have nothing to do with the morality of this...

The morality is already clearly established: in exchange for a very small monthly sum, you get the property when the owner dies. The property may be a dedicated rental unit already, so whether or not the owner lives there has no bearing. So it's not just _within_ the contract, it's the _purpose_ of the contract.

>The morality is already clearly established: in exchange for a very small monthly sum, you get the property when the owner dies.

That's, again, the contract. Seems some people can't tell between legal and moral, unless we talk about the 70s and segregation or something like that...

No, I (and presumably the other posters) can tell the difference - I just disagree on the morality, partly because I think the legal aspects are related to the moral aspects.

The guy purchased the right to the house once the current owner dies, but you believe it's immoral for her to continue living in it, or renting it out, or loaning it to friends, while she is alive. I think it's immoral to demand the house while she's alive, since that wasn't the deal.

Sometimes legality and morality are widely separated, but here they seem related.

It is a lottery with two participants that willingly took part in it. She won big, but if she had died the day after the contract was signed, he would have won big: he would have gotten the house for about $500. Would you suggest he should have paid her inheritance a lump sum if that would have happened?

I don't see much harm in "winning big" over somebody who just died. Being dead, it's not like she would have missed the money.

On the other hand, having someone pay tons of money and then die without anything to show for it, I do see, even if he "willingly signed into it".

There's also a threshold over which it should just be handed over. That such a threshold wasn't in the contract doesn't make it any less odious.

Her heirs would have disagreed. Consider the case where your mother owns a house and 'sells' it for $500 just before dying.

Also, I expect people buying such rights are generally well-off, and the people selling it to be relatively poor. After all, the buyer must be able to pay money on a house they cannot live in for an indeterminate period, and the seller typically chooses this construct to be ensured of both a roof over their head and money to live from for one's entire life, no matter how long that is.

The odds in this lottery get better every month for the buyer, so, assuming that the buyer can afford to buy into this, they should have no reason to want to get out of this.

He would have had to make a down-payment on the house too. You pay e.g. half the total payment up front and then the remainder in monthly installments. Still, it would have been a quarter of the retail price, not half, in that case.

The morality is that he got exactly what she offered and exactly what he asked and paid for, and the arrangement was perfectly consensual. It would be immoral for him to demand more, without additional payment.

I fail to see anything immoral here.

You'd think the contract would say that in addition to being alive you have to actually live in the house as well. Otherwise you could just sublet it out and have someone else pay you for it while you live somewhere else. Seems kind of counter to the intention of the idea.

I guess that is the risk for this approach. This could almost be the premise for an Edgar Allan Poe poem. It's almost too perfect

HN discussion on that story from a wee while back:

https://news.ycombinator.com/item?id=11367422

This exists in Italy as well, but it usually doesn't include a rent. You just buy the house at a discount. It's very very common in areas which are rapidly gentrifying.

What is the Italian name for this type of contract?

nuda proprietà

That's what I supposed (I am Italian myself) but it looks different from the French thing. From what I understand with "nuda proprietà" you buy the apartment but cannot use it until the person in residence dies or voluntarily leaves. But you the owner will not pay anything to the "tenant". So there is no extra spending even if the person using the place lives much longer than expected.

Well nuda proprietà is actually a bit more general a concept: it just means that one person has ownership and the other has all the other benefits. Remember when Berlusconi touted that as the solution to his conflict of interest?

Don't we just call this a reverse mortgage in the US?

from my understanding, reverse mortgage is a mortgage where the owners slowly encumber their property with debt. If the owners die (or decide to sell), the property is still the owners, but it has the mortgage debt on it so they have to pay that off as part of the process.

these days, it is better to just move into a smaller flat and rent out the larger property at market rates.

This also happens in New York— one can buy an apartment with a rent-stabilized tenant in a building which has been converted to a coop. The tenant has a right to lease renewals until they die (and any family living with them at time of death has a right to lease renewals as well).

Here is one: http://streeteasy.com/building/243-west-98-street-new_york/7...

I'd be suspicious of a building that uses Comic Sans on its awning.

Being French, I never heard of someone doing this with strangers. On the other hand, it's a well known way to bypass succession fees : parents sell (or find a way to pass on without monetary exchange) the house in viager to their children, and continue living there until they die.

That way, even in a case where there is disagreement in the family later on, parents can't be forced out of the house.

Same thing in Italy concerning succession. It exists as well with strangers, but I never heard a regular payment was involved: you just buy the so-called "nude property" and then wait.

"bare property" :)

It's a low-profile business but transactions between strangers aren't as rare as one might think. Paris has a few notaires specialized in rente voyagère working in tandem with financial advisors. I know Toulouse and Aix also have those.

It can be a great solution for an old homeowner with meager retirement earnings and no natural beneficiaries.

I thought you couldn't do it with members of your family, has it changed? Didn't have a look at it for a while...

It is also a very strong motive for murder. I wouldn't want to go into such a deal with a stranger who "wins" a very large sum of money when I die...

Statistically, people overwhelmingly are not murderers.

Even people with motive.

But such a thing would attract them. Like rich people attract kidnappers.

Generally people who kidnap believe they can get away with it by never being in the frame as it were. Given that you'd be one of the prime suspects on motive alone in this case, it seems like a bet on being able to commit a "perfect" murder.

I don't think it's a likely occurrence

Is there a faster way of becoming prime suspect? The kidnapping equivalent would be allowing random people to take out 'kidnap insurance' on you (with your agreement). If it so happens that you are kidnapped, it's pretty obvious which door the investigators will knock first.

But this attracts the few who are.

Actually it wouldn't be. You are buying an asset for 50% off which will continue to go up in price. And you get to pay a rentor a cheap rent to take care of your asset their entire lives.

It's a great deal if you can get it. The only time this will get problematic is if housing prices drop and continue to drop.

My American university has bought property in this manner. There were several old houses on the fringes of the campus with elderly owners. The university bought the properties years ago, but the owners continued to live in them until they died. Once the property transferred, the university bulldozed the houses to build new research facilities.

You forgot to add "under mysterious circumstances".

Usually in this situation the university becomes the landlord and the inhabitants are protected just under the usual tenancy and eviction laws.

I believe it is called a reverse mortgage in the United States.

https://en.wikipedia.org/wiki/Reverse_mortgage

In the states, the bank is lending you money and essentially building equity in your home. If your heirs pay the loan off when you die, they often may keep the house.

It's a no-lose proposition for the bank, presuming they charge a profitable interest rate. But, they are not guaranteed to get the house at the end of the deal.

>Reverse mortgages allow elders to access the home equity they have built up in their homes now, and defer payment of the loan until they die, sell, or move out of the home. Because there are no required mortgage payments on a reverse mortgage, the interest is added to the loan balance each month. The rising loan balance can eventually grow to exceed the value of the home, particularly in times of declining home values or if the borrower continues to live in the home for many years. However, the borrower (or the borrower’s estate) is generally not required to repay any additional loan balance in excess of the value of the home

https://en.wikipedia.org/wiki/Reverse_mortgage

Israel Horovitz made a movie about it:

https://en.wikipedia.org/wiki/My_Old_Lady_(film)

There is one problem with that. People can predict their own life expectancy. People who expect to live shorter will avoid such scheme.

To give an example: state pension in my (European) country is unfair to men. We live shorter and retire latter. In average women gets 45% more money for the same contribution. As result men are avoiding state pension and use private funds.

In Finland this is solved by it not being possible to avoid state pension. You can collect your additional stash, but if you're getting a salary, you will be paying to the pension fund.

Yes, women live longer, but in general have a bit smaller pension due to their time off the workforce because of motherhood, and (still) slightly lower salary levels. I don't consider this unfairness towards men any significant issue.

That's an excellent point. It actually helps balance existing inequality. Maybe not 'ideal', but fairly practical.

No! You don't balance unrelated, uncorrelated things.

Do we now abandon efforts to reduce gender roles, or pay-gap differences because that would upset the balance. Are women who don't become mothers owed a greater pension?

Refusing to fix one problem, due to some illusion of karma is ridiculous. Should we also abandon cancer-cures out of concern for those who are short-lived by other causes?

Can you elaborate? Why is it wrong to base individual pensions payments on individual contributions? I'm not following.

Are you just saying it wrong to lump people into two different groups arbitrarily? Like we don't base social payments on two arbitrary cohorts like "people over/under 68 inches of height", so why would we base it on gender?

The original poster was implying that other inequalities wrt gender "balanced out" the equality wrt pensions and lifespan. I was rejecting arbitrarily linking two unrelated things (problems), and consider them as cancelling each other out.

The poster also stated "In Finland this is solved by it not being possible to avoid state pension" so this isn't basing individual pensions on individual contributions, because much like life insurance, and individual pension would adjust for life-expectancy, whereas I believe the Finnish system is a flat rate wrt gender (this is what I inferred?).

The point about arbitrary groupings is a good one, but it will fall on deaf ears in places like the US, where established groupings have high political importance.

The existing inequality caused by not working as much and choosing lower paying fields? You can bet if men benefitted from this instead it would be considered institutional sexism that would be targeted and dismantled.

> time off the workforce because of motherhood, and (still) slightly lower salary levels

This has nothing to do with life-expectancy though, so isn't relevant.

Well, I don't see the whole point of life expectancy and pensions relevant. Pension or payments definitely should not be balanced according to statistical life expectancy per gender.

Not even privately purchased annuities? Otherwise you may have just discovered why many people don't buy them; they are a suboptimal purchase for men, if the ungendered life-expectancy is used.

No, not talking about the private ones, but rather the compulsory state-driven pension system that is in place in most of the Europe.

Pensions, much like life-insurance, should probably factor it in, which would include other things too (weight, smoking).

I wrote "for the same contribution". If women pays equal money, she will get more from system, because she lives extra seven years.

Plus children are counted as a bonus for state pension. So you do not lose pension for spending three years home with kid. (men would)

We also have compulsory state pension system. It does not stop people from opting out.

> We also have compulsory state pension system. It does not stop people from opting out.

This doesn't sound like compulsory, right?

You can always leave to other country, or minimize your contributions with creative accounting.

Also very common is to build your own house by yourself DIY style. It is significantly cheaper because you do not have to pay taxes from work you do for yourself.

The United States taxes its citizens' income worldwide, at least above a certain income threshold, and from my understanding requires yearly filing regardless. IMO it's unlikely that simply moving to a different country would allow you to dodge a federal mandatory pension (provided it happened in the US).

Social security has an awful lot in common with a federal mandatory pension. If anything, it would probably be easier to opt out of a pension system (where pension implies that the individual benefit is more fully funded by the individual payments).

People that can show they are making payments into a similar program elsewhere often don't have to make social security payments.

https://www.ssa.gov/international/agreements_overview.html

Social security is the United States' federal mandatory pension, and it does not generally apply to foreign income (unless in a country that also has a mandatory pension with which the US has a tax treaty).

https://www.irs.gov/individuals/international-taxpayers/soci...

Oddly enough, you can opt out of the Social Security - some employees of government agencies don't have to contribute and some religions are also exempt.

Personally, I feel that anybody should be able to permanently opt-out if desired.

I worked for one of those organizations in the past, which seems to have skewed my view quite a bit. Thanks for the corrections.

> As result men are avoiding state pension and use private funds.

Does that help? I imagine there could be laws against private insurers "discriminating" based on sex (or other differences, e.g. smoker status etc.).

Some countries allow driving insurance to "discriminate" though (women pay less), so maybe they would allow pension insurance as well.

"Discrimination" isn't when it is based on actuarial models. Most countries' court systems have already ruled on this.

Basically, I can charge a woman x% more for disability insurance because I can mathematically prove that women are statistically more likely to claim.

Hm... If you made this argument for e.g. black Americans (something something more likely to be criminals), most people would scream stuff much worse than just "discrimination". Yet both of these are based on math.

Actually, using actuarial science in that way is still very much open to debate.

https://en.wikipedia.org/wiki/Actuarial_science#Actuaries_in...

If you name a group, I can give you a mathematical model that will unfairly discriminate against that group, purely through how I divide up the population. "Math" is a mechanism, not an indicator of right or wrong.

Actuarial Science is a hard science that is a few hundred years old, not a social construct that you can manipulate.

There is no desire to "show" anything beyond having an accurate representation of probability.

Actuarial Science is also a mechanism, not something that indicates right or wrong. It gives you answers to questions. But questions themselves can be biased.

Same issue exists with insurance, of course - and people who write annuities know this, and account for gender.

You can always add classes based on age or membership in the tontine to make it easier.

For example, a tontine of 1000 euros with 10 members, but since the last member is in the age bracket of 20-30 instead of 30-40, they get a smaller annuity than the rest.

The Wikipedia article about tontines mentions this:

> Because younger nominees clearly had a longer life expectancy, the 17th and 18th-century tontines were normally divided into several "classes" by age (typically in bands of 5, 7 or 10 years): each class effectively formed a separate tontine, with the shares of deceased members devolving to fellow-nominees within the same class.

https://en.wikipedia.org/wiki/Tontine

A hybrid tontine system is the most common way to save for your pension in Denmark. It is not regarded as sleazy.

The companies that run the schemes, has historically been run by unions, and has historically given good yields on their wealth. The largest company of the type is state-owned and it gets money from an obligatory tax applied to all people working in the country.

The system will stop working, if politicians begins to tax these pension funds for infrastructure investment purposes, or the pension funds shifts people to financial products that prioritize individual savings/insurance.

This "tontine" system is probably the best investment, in terms of securing a good life after 65 years old, if you are among the survivors.

>if you are among the survivors

Can you expand on this?

The members of the scheme who die relatively young subsidise those who live to an old age: both have paid the same amount in, but the latter draws a pension for a longer time.

And by then it doesn't really matter!

Wait, do the payments actually increase as more people die? If not, it's not a tonine, just regular pension insurance.

Well, the money invested by the deceased has to go somewhere..

I'm not sure how the system works in Denmark but in Sweden you choose whether to give the remaining pension to your family or to have it split among the other people in the pension fund that also opted for the split.

"There was rampant shadiness back in the day" ...this is an actual quote from a major paper? Maybe I'm just getting old, but this seems like unbelievably sloppy language for someone being paid to write.

This is a blog/column. I'm not sure this article appeared in the actual paper.

> There was rampant shadiness back in the day, including countless swindles perpetrated on the public, both related and unrelated to the tontine. Bribery of newspapers and politicians was also routine, as was embezzlement and plain theft. Magazine exposés riled the public with stories of secret payments and secret blackmail of public officials.

And WaPo thinks our current financial industry can do better? Sounds a lot like NINJA loans and synthetic CDOs ...

Further reading:

"Tontine Pensions: A Solution to the Chronic Underfunding of Traditional Pension Plans", Forman & Sabin 2014 http://repozytorium.put.poznan.pl/Content/343995/Szczepanski... "Tontine Insurance and the Armstrong Investigation: A Case of Stifled Innovation, 1868-1905", Ransom & Sutch 1987 https://www.dropbox.com/s/8ubxv59z524fq5s/1987-ransom.pdf http://cyber.sci-hub.bz/MTAuMjMwNy8yMTIyMjM2/10.2307%4021222...

It seems to me that the middle/upper middle classes moving from defined benefit (old school pension plans) to defined contribution (401k plans) was a chance for them to build up (and learn to sustainably manage) real capitalistic wealth that could be passed down to their kids. It's not a stretch to predict that to the extent the vast majority of these folks store their wealth in annuities, or similar plans that effectively evaporate upon death, the result will be much smaller estates/inheritances which would reinforce or even worsen the wealth inequality that exists today.

Couldn't this whole scheme be developed in totally decentralized fashion using bitcoin, ethereum or similar? Would be quite a good idea.

I was about to say something about needing an enforcement body of some sort. But then I realized you could literally use a dead man switch.

Couldn't the members could write a script (or even just have a friend) that keeps them "alive" forever?

Maybe using some form of biometrics. Or also hiring some sort of private investigator who would keep track of everyone in the tontine primarily via social media and sometimes by physically checking on them.

I'd imagine it would be impossible to get around this problem completely though.

Probably not, because keeping a trusted account of the alive/dead status of each member doesn't sound feasible.

A potential method would be some sort of "still alive" signal. If it isn't received after X years (how many years missing before you're declared legally dead?), the system would stop paying to your account.

This doesn't prevent others from obtaining your credentials and maintaining your account (see people not reporting deceased family members to keep their social security checks).

I don't think I would like to invest on something called "tontine"; "tonto" means dummy in Spanish. White Snow's Dopey is called "tontin" in Spanish. Lone Ranger's friend Tonto had to be renamed "Toro" as in bull.

I thought that Tonto's name was the joke/point?

My biggest concern is the direct benefit off the death of others causing a perverse incentive to kill. There are people who will mug and risk killing you over the money in your wallet, so I could definitely see people being killed over the money in one of these.

Makes sense, it's like crowd-sourced life insurance. - http://blog.streeteye.com/blog/2015/09/tontines-strange-name...

But insurance companies love selling insanely high-priced annuity products, and US laws around insurance, securities registration, taxes make it hard/impossible.

Regulation can be premature optimization / standardization. (or insurance company rent-seeking)

Damn, didn't take long to get to the real reason. Governments are scared shitless about the public pension default issue. The hundreds of billions of unfunded liability with regards to public employee pension payments has been well known and no good solution has been found. From straight up state employees to those working for transit authorities. The numbers of truly staggering.

So I guess need a new means to sell people on a lower payout? Perhaps this will also end up in a way taking over SS?

Pension funds in many cases operate similarly; their finances improve if beneficiaries die early, and in some cases they are required to pay out surplus returns above a certain level. The difference seems to be that pensions operate continuously (ie, they'd only see a demographic effect on payouts if there were a global decline in life expectancy) rather than on a defined batch of customers.

Can you construct an insurance instrument that behaves like a tontine without breaking the law?

I'm imagining some giant network of life insurance policies with thousands of beneficiaries on each policy and automatic rebalancing after every death.

I don't know, it just feels like a tontine (illegal in the US) can be constructed through what I assume are legal mechanisms.

The title sounds like an ad for local news. Please change the title to what it is.

So, it's a shame you can't do this over a blockchain, to verify honesty of the payouts -- no way to make sure that someone's still alive.

The government keeps legally meaningful records of deaths. It can be done.

Blockchain assassination market was a thing a few years ago, IIRC. Obama and Bernake had prices on their heads.

I wonder if these two products can be combined somehow, Highlander style.

I mentioned this in another comment, but I think it would be feasible to hire somebody for a small price to investigate the statuses of the participants once per year as well as implementing some sort of biometric dead man's switch.

The old British comedy "The Wrong Box" tells a story about a tontine and its consequences.

The Simpsons did it in Raging Abe Simpson and His Grumbling Grandson in "The Curse of the Flying Hellfish".

Also, Archer: The Double Deuce

And for that matter, in Archer, too. Woodhouse and his army buddies set up a tontine.

of course, www.tontines.com is a landing page to godaddy

I like reading about old financial products because the conditions can be recycled as novel

... and fuck men who live shorter.

Please don't post this sort of angry unsubstantive comment to Hacker News. It's destructive of the spirit of the site, which is intellectual curiosity.

We detached this comment from https://news.ycombinator.com/item?id=11861314 and marked it off-topic.

She probably rented it out. Socialist Europe tends to frown on property standing empty, but there's no reason the guy should've got the rental income.

Drive-by name-calling ruins HN threads, as seen below. Please don't toss matches into political petrol on this site.

We detached this subthread from https://news.ycombinator.com/item?id=11861319 and marked it off-topic.

I wasn't name-calling, I was describing. I'm a socialist and a European. The government of France is literally the socialist party.

Fair enough. The phrase doesn't hop contexts so successfully.

Everything old is new again.

"a modern tontine would be particularly suited to soothing the frustrations of 21st-century retirement"

You know what would be even better? Higher interest rates.

Interest rates in most of the developed world have been on the decline for almost 30 years. There is no way to just magically increase interest rates in the market. Contrary to the popular belief, the Fed doesn't set interest rates in the economy, and it has even lower power over long-term interest rates [1]. Also, you're interested in interest rates above and beyond inflation, i.e. 'real' rates of return, not nominal.

[1] http://aswathdamodaran.blogspot.co.uk/2015/09/the-fed-intere...

I disagree with almost everything you said here. And a blogspot link does not prove your point. It's just someone's opinion.

Interest rates aren't directly set by central banks, but of course they influence the cost of borrowing for commercial banks. That's why when central bank rates fall, commercial rates fall too. And visa versa. And why when commercial banks don't lower rates in turn there is political pressure to "pass on the savings".

As for longer term rates, they do have power over that by selling one maturity of instrument for the purchasing of another maturity. This has colloquially been called "operation twist".

I am fully aware of real vs nominal. But I'm curious. How do you know what real rates are if you do not have a consistent, repeatable, uncorrupted measure of inflation?

"It's just someone's opinion"

In this case it's Damodaran's opinion, so yeah, it matters, like, a lot.

"Interest rates aren't directly set by central banks"

The "risk-free" interest rate for a country is. All other rates are derived from it (see CAPM). Interest rates for any loan, or cost of capital for any enterprise is the risk-free rate plus the risk premium.

"what real rates are if you do not have a consistent, repeatable, uncorrupted measure of inflation?"

You can use a lot of proxies for that, and you'll see cases where there are major disagreements between what the market sees as inflation and what the government says it is (see Venezuela and Argentina, for example, http://www.economist.com/node/21548242).

That aside, the market usually has a fair trust of statistical institutions that calculate inflation, because that data is quite transparent.

> Interest rates aren't directly set by central banks, but of course they influence the cost of borrowing for commercial banks. That's why when central bank rates fall, commercial rates fall too. And visa versa. And why when commercial banks don't lower rates in turn there is political pressure to "pass on the savings".

Sure, but that's all nominal. Real interest rates are necessarily the rate of concrete economic growth; central bank interest rates cannot affect that except to the extent that they actually impact the real economy (i.e. if they make investment easier/harder that can affect real productivity).

> I am fully aware of real vs nominal. But I'm curious. How do you know what real rates are if you do not have a consistent, repeatable, uncorrupted measure of inflation?

You need a measure of what you value - which is up to you. Whatever that measure is, you can figure out how much value you get by spending money now vs by investing and spending money later. The interest rate is just a consensus/average of those.

The central bank can set their rate at 1000%, but if no one has any potential investments that can yield 1000%+ returns, companies just won't borrow money. In other words, no one will be getting 1000% returns on their savings. On the other hand, if you'd be willing to lend your savings at say 5%, you might find action.

People invest in the CB and earn their guaranteed return. This would drive inflation as bonds mature and soon investment returns would eventually look something like (1000 + $realReturn)%.

Maybe not at such an extraordinary rate, but Volcker raised rates to 10-21% in the early 80s and US-based stock returns during that era were substantially higher than at any other time.

http://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/...

Referring to that, in the 16 years after the 10 T-Bond return hit a peak of 33% in 1982, stocks yielded over 30% in six separate times. In the next 16 years, from 1999 to 2015, it's done that once, 8 years after a -30% return from the crash of 2008. The S&P also yielded 30+% only twice in the 16 years before 1983.

It's not concrete evidence, but it does suggest some relationship between central bank yield and stock returns.

I get the impression that you don't know who Aswath Damadoran is. You should look him up - he's one heck of a good writer.

It is absolutely the case in the USA that the Fed pushed down long term interest rates by their manipulation of shorter term rates, thereby modifying the yield curve.

The Fed themselves state this, eg http://www.frbsf.org/education/teacher-resources/us-monetary...

Only the most obtuse person would argue otherwise, for example by claiming that market forces set the rates...but the Fed has truly unlimited credit so only an idiot would bet against them.

I liked your comment until you said "only the most obtuse person."

It wasn't meant for you, but for those who argue that the market is the sole determinant of what happens.

The Fed is like the engine of a plane, the market the wings. To argue that all you need are wings and the engine is irrelevant, is what too many (on other forums) argue.

> You know what would be even better? Higher interest rates.

Why would you want to give money to people who already have money?! Interest payments should come from productive uses of the saver's money, which currently there seem to be very few (banks often just buy "risk-free" sovereign debt, which, being risk-free, doesn't deserve any interest).

Interest rates are set by the economy as a whole.

Because of arbitrage, the interest rate on low-risk assets (including bonds and various forms of debt) will always be more or less the same. You can't have higher returns without a higher cost of borrowing. And of course, raising interest rates suddenly will have negative effects - assets will be devalued (higher borrowing costs drives house prices down, for example), which hurts retirees even more than low interest rates on savings.

> You know what would be even better? Higher interest rates.

Higher interest rates have to be supported by inflation. Inflation comes primarily from two sources: productivity gains and raw market growth. Modern economies are hitting diminishing returns with productivity gains, and most (if not all) industrialized economies are close to zero population growth.

Where are the higher interest rates supposed to come from? Currency manipulation?

Reply on news.ycombinator.com