STUMP » Articles » Mortality with Meep: French Annuities and the Revolution - Antiselection » 30 December 2018, 15:54

Where Stu & MP spout off about everything.

Mortality with Meep: French Annuities and the Revolution - Antiselection  


30 December 2018, 15:54

One of the major causes of the French Revolution was the parlous state of the government finances. Now, poor public finance is something that occurs often in history, especially European history. Both Britain and France had public finance issues in the 18th century, partly driven by their propensity to war (especially with each other), and having profligate royalty.

But Britain got its financial house in order, and France… did not.

So France kept coming up with “clever ideas” to finance their spending.

Like annuities. At a single rate.


Lessons from history II: The “Thirty Maidens of Geneva” and the French Revolution

One of the main causes of the French Revolution was a crisis of public debt. Unlike England, 18th century France never managed to balance its budget, which led to several sovereign defaults. In 1789, Louis XVI, who had committed himself never to default, chose to summon the Estates General in order to find new resources. This was the beginning of the French Revolution.

The old way to fund a deficit

It is also where the small girls of the Swiss bourgeoisie come into play. How? The standard form of public debt at the time was the life annuity. The buyer, or annuitant, pays a lump sum to the State, and in exchange he receives a given percentage of that sum every year until he dies. You lend 100 to the State in 1700, and you receive 10 every year until death. Around 1690 already, the rate paid to the annuitant depends on his age. The older you are when you buy the annuity, the more you receive every year, in order to balance the income on a lifetime. For the State, the cost of the annuity is more or less the same, whatever the age of the buyer.

Around 1760, for several reasons which are still discussed in literature, the French government decided to come back to the more basic technique of “fixed-income” annuities. This means that whatever the age of the annuitant, they would receive the same amount until their demise. At the beginning, this was not a problem because most buyers were adults. Life expectancies were shorter than now and the fixed rate was calculated to correspond to the rate that was applied to a 50-year old buyer in the previous system.

Tricks and loopholes

But in 1771, Geneva banks realized that there was nothing preventing from buying annuities in other people’s names. In fact to receive annual payments, you had to prove that you were still alive. Life certificates were expensive, so a lot of people chose to buy annuities whose lifespan would correspond to that of a famous person. When he was executed in 1793, £400,000 of annuities relied on Louis XVI’s life. Banks exploited another loophole by involving young girls between 4 and 7 (most child diseases hit between 0 and 3 years old, so the life expectancy of a small child is superior to that of a newborn). Annuities were placed in their names. Banks even paid those girls follow-up care.

Smart as they were, Swiss bankers pushed the mechanism even further. “To diminish the risks of losses by accident, the banker would choose a certain number of ‘heads’, 30, 40, 50, 60 (the most common was 30, hence the name of the ‘Thirty Geneva Heads Loan’)”. Annuities from several different people, or heads, were packaged into a single financial product, which was then divided into shares and offered to clients. Sound familiar? It is the same mechanism as mortgage securitisation**, the allegedly revolutionary technique that provided the raw material for the subprime crisis. The technique became ‘revolutionary’ in another way, as we shall see.
Back to 1782. Three quarters of the annuities issued then are estimated to have been placed in the name of small children. French banks (and possibly individuals) had massively imitated the Geneva scheme. The little trick thus contributed to the rise of the debt burden and indirectly to the fall of the Ancien Régime.

I want to talk about antiselection and mispricing.


Antiselection is also known as adverse selection, and it’s the reason insurance underwriting exists (among other things). This publication by Munich Re sets out a good definition:

One of the greatest threats facing life insurers is anti-selection (also called adverse
selection or negative selection). Anti-selection occurs when an underwriting information
deficit allows a higher-risk group (such as smokers) to purchase life or health insurance
at the same price as a lower-risk group (non-smokers). This imbalance produces four

1) The members of the high-risk group receive higher claims payouts, and are more
costly to the insurer;

2) Because they pay a relatively low price, members of the high-risk group purchase
additional insurance;

3) The insurer, to cover losses, raises rates for everyone;

4) The low risk customers abandon the company and seek coverage at a lower
price elsewhere. Anti-selection is most often attributable to two sources: improper
underwriting and risk selection by the insurer, and/or superior knowledge of the risk
either by the insured or some third party involved in the transaction.

So, the main way we deal with this in annuities is not to actually medically underwrite, but quote a price by age and gender. The mortality rates we use for individual annuities are much lower than the general population, because those who buy life annuities tend to be wealthier and healthier than the general population.

(There can be impaired lives annuities, but that’s beyond the scope of this post.)

So here’s something that has happened a couple times: insurers being disallowed to have pricing of annuities based on gender. This has happened in the U.S. under two different situations, and in Europe under a gender discrimination bill.

I’m not going to go into those details, but what happens when you have to give men & women the same annuity rates?

We use the female mortality tables (no, not exactly, but not too far off) — women live longer than men, have lower mortality, so we’re going to the “riskier” one to prevent antiselection.

Similarly for life insurance — men are more likely to die earlier, and thus we give higher premium rates for men’s insurance. So we’d use male mortality tables if we have the charge the same for men & women for life insurance.

Antiselection is not necessarily a huge deal in situations where the different risk classes being treated as one don’t have hugely different outcomes.

But the life expectancy (remaining life expected) for a 6-year-old girl and a 60-year-old man are hugely disparate. So antiselection was a huge risk in the French annuities. I have no clue why the French thought this single rate was a good idea.


Somebody emailed me with the following:

A despondent man once asked Voltaire whether or not he should go on living. Voltaire said, “I advise you to do so.”


“To enrage those who are paying your annuities.”

I wish I could find the quote in its original French. I cannot find that quote here, which does include the original French for his quotes.

Which then reminded me of the connection between annuities and the French Revolution… and also that I had visited St. John’s University a couple months back to give a presentation. And while there, I got a good gander at the Davis Library, which has a long history:

Today’s Kathryn and Shelby Cullom Davis Library at St. John’s University’s School of Risk Management comprises the world’s largest collection of risk and insurance literature, policies, and related documents, and serves students of risk management, insurance, and actuarial science around the world as a center for study and research. The Library’s beginnings date back to 1901, when insurance executives founded The Insurance Society of New York, and began providing study space and material to young people entering the insurance industry.

Since its earliest days, the Library has served as a site for insurance lectures. Over the years, The Insurance Society of New York developed a curriculum based upon these lectures. This ultimately led to the creation of The School of Insurance, followed by The College of Insurance, the predecessor to St. John’s School of Risk Management.

A number of individuals and organizations contributed to the Library’s growing collection throughout the 1900’s. Throughout its history, the Library has served students, faculty, staff, and sponsors, as well as members of The Insurance Society of New York. In 1974, Shelby Cullom Davis and his wife Kathryn secured the Library’s future with a major endowment, and the Library was renamed in their honor.

With the merger of The College of Insurance and St. John’s University in 2001, the Davis Library became part of the St. John’s University library system. Today, the Davis Library remains one of the most comprehensive risk, insurance, and actuarial science collections in the world, serving students, faculty, scholars and industry practitioners seeking specialized knowledge in these areas. It is a place for study and learning, research and insight.

And they have some amazing historical documents there, including books of names in Irish annuities. (It wasn’t only France that did this dumbass thing, but what was “lucky” for the Irish annuities is that a lot of people killed in the French Revolution have their names on the lists…) Alas, I don’t see these items here (and their digital collections site is a dead link), but I did manage to take several pictures myself.

Here’s a post on the annuities (and tontines):

In England all three State tontine schemes up from the first in 1693 to the last in 1776 failed and by the following year Tontines to all intents and purposes were banned. In 1777 the English Parliament because,

“it hath been found by experience that the making insurances on the lives or other events wherein the assured shall have no interest hath introduced a mischievous kind of gaming”,

enacted the Life Assurance Act of 1777 (also known as the Gambling Act :14 Geo. 3 c.48) an act

“for regulating Insurances upon Lives, and for prohibiting all such Insurances except in cases where the persons insuring shall have an interest in the Life or Death of the Persons insured.”

Not so in Ireland where the 1777 Life Assurance (Gambling) Act was not incorporated until 1866 (and still in force!). There were three Irish State Tontines, fully subscribed to, particularly by the burgers of Geneva, in 1773, 1775 and in 1777.

And it sounds like they still had issues, even with those dying at the hands of Monsieur Guillotine:

Adverse Selection in the Irish Tontines of 1773 , 1775 and 1777
Yikang LiPublished 2017
A “tontine” is a special kind of annuity in which all participants contribute equally to a subscription pool, and a fixed percent of the total capital raised is distributed equally among surviving nominees every year. In this paper, we examine the adverse selection in the Irish tontines of 1773, 1775 and 1777 because of the presence of a group of speculative investors, namely a group of Genevan bankers. These Genevan investors purportedly cherry-picked nominees with greater expected longevity. Their existence allows us to study a rather unconventional aspect of adverse selection, which arises from the informational asymmetry among different types of buyers. Using a newly compiled data set on the nominees and their subsequent mortality, we estimate that these Genevan investors earned on average 8.5% more per share than the other subscribers of the Irish tontines. The result suggests that speculative investors with access to superior information may earn higher returns at the expense of average investors, a phenomenon implicit but difficult to quantify in other insurance markets.

Financial optimization has been around for a long time.


Here are some pictures of the historical book providing the list of names for the 1775 Irish Annuities:

Marie Antoinette was one of the names:

Here are a few other names:

A summary of the number of covered lives:

And the legal documents:

I looked up some info on the act authorizing these annuities:

Money had to be sought from other sources. The deficit on the year was 138,840 [pounds], there were floating debts of 400,000 [pounds], and the national debt, which was discharged in 1754, and started again under the Duke of Bedford, amounted to994,890 [pounds]. Government proposed to raise 265,000 [pounds] by a tontine, granting annuities on lives at 6 per cent.



Wikipedia: Tontine uses and abuses:

Louis XIV first made use of tontines in 1689 to fund military operations when he could not otherwise raise the money. The initial subscribers each put in 300 livres and, unlike most later schemes, this one was run honestly; the last survivor, a widow named Charlotte Barbier, who died in 1726 at the age of 96, received 73,000 livres in her last payment.111213 The English government first issued tontines in 1693 to fund a war against France, part of the Nine Years’ War.613

Tontines soon caused financial problems for their issuing governments, as the organisers tended to underestimate the longevity of the population. At first, tontine holders included men and women of all ages. However, by the mid-18th century, investors were beginning to understand how to game the system, and it became increasingly common to buy tontine shares for young children, especially for girls around the age of 5 (since girls lived longer than boys, and by which age they were less at risk of infant mortality). This created the possibility of significant returns for the shareholders, with significant losses for the organizers. As a result, tontine schemes were eventually abandoned, and by the mid-1850s tontines had been replaced by other investment vehicles, such as “penny policies”, a predecessor of the 20th-century pension scheme.

CBS news in 2009: A Government Pension That’s Lasted 271 Years

An obscure pension created in 1738, which still pays out money today from French government coffers, may be the longest-lived and most amazing government debt in modern history.

The story starts with a French lawyer named Claude Linotte, born in 1686, who became an advisor to the dukes of Bouillon and a tutor to their children. The duke offered Linotte an annual payment of 2,000 livres for life; the canny barrister successfully countered with an annuity payable to “the date of death of the last survivor among the descendants of Mr. and Mrs. Linotte.” (A livre was the currency at the time.)

A working paper published last week by Chicago Fed economist François Velde recounts what happened next. The debt survived a ducal succession by the prince de Guéméné, an annexation of the dutchy, the French Revolution, and in a complicated transaction involving estates and the castle at Navarre, was assumed by the French government when Napoleon signed the documents in 1807 while on the Polish-Russian border.

Eventually an alert bureaucrat in the French ministry of finance noticed the unusual debt and, circa 1900, tried to convince Marie-Louise-Alexandrine Aubriot to relinquish her lineage’s perpetual claim in exchange for a lump sum. She refused; the payments continued.

This might have been a happy ending — at least for the descendants of Claude Linotte, not taxpayers — but for violence through successive devaluation that was done to France’s currencies over time. There was a de-linking of the franc and gold in 1914, followed by the 1960 transition to the new franc and then the euro.

Today the perpetual annuity is worth only 1.20 euros a year, or about $1.80. The working paper says Linotte’s legacy is “now two hundred and fifty years old, weather-beaten and scarred by time’s passage and inflation, bereft of two zeros by the 1960 conversion from old francs to new francs, but tenacious and enduring, alive with dynastic permanence, and yielding each year enough to buy a café-crème at the bistrot on the way out from the Treasury.”

I wrote to the author of the paper, François Velde, to say: “If, at the time of the creation of the annuity, 740 livres was worth a mark (8 oz) of gold, the value of the annuity paid in gold today would be worth approximately $12,700 a year. When a debtor is able to redefine the terms of his obligation through currency devaluation, we get this interesting but not entirely unpredictable result: a payment becomes worth only 1/10,000th what it should be.”

Velde wrote back: “You’re right: 790F in 1811 contained 3555g of silver ($2136 at today’s prices, €1427) or 229g of gold ($8600, €5747). Another way to put it is that I should have given some idea of what the annuity was worth in real terms 200 years ago.”

Unfortunately, the lessons of history tell us that stability of the sort conducive to bankers offering predictable returns on capital over the centuries are not the norm. There are reports of a still-honored Dutch water board perpetual annuity from 1624 as well, but that and the French obligation remain the exception — war, confiscation, and looting are closer to the norm.

It’s amazing the French government still paid that.

Related Posts
Mortality Monday: Broken Heart Syndrome
Taxing Tuesday: Time to Make the Donuts
Mortality Monday: Supreme Court Probabilities