Lottery Winners Do Not Avoid Bankruptcy

The rags to riches to rags story of a poor, unemployed fellow who wins the lottery, blows the cash, and ends up just as poor and unemployed as he began is a common trope.  (Here is a classic in the genre). In a paper just published in the Review of Economics and Statistics (gated, free version here), Hankins, Hoekstra and Skiba argue that the rags to riches to rags story has a systematic component.

The authors link records of lottery winners to bankruptcy records. The use of the lottery is a great randomization device, although obviously it restricts the sample to people who play the lottery.

The central finding is this: people who win large amounts are just as likely to end up bankrupt as people who win small amounts. People who win a large amount, $50,000 to $150,000, have a lower bankruptcy rate immediately after winning but a higher bankruptcy rate a few years later so the 5-year bankruptcy rate for the big winners is no lower than for the small winners. Amazingly, by the time the big winners do go bankrupt their assets and debts are not significantly different from those of the small winners. The big winners who ended up bankrupt could have paid off all of their debts but chose not to.

N.B. the result is not that most lottery winners go bankrupt or that winning money doesn’t help people–the result, as Robin Hanson might say, is that bankruptcy isn’t about money.


Similar findings every year in the car industry find large down payments usually lead to higher, not lower delinquencies. Makes you wonder about pushes for higher mortgage down payments.

Cause and effect. Higher required down payments are almost surely linked to creditworthiness (or lack of). Of course, making someone come up with a big down payment may make them less financially stable and more likely to become delinquent. All in all, I would say the higher down payments required are beneficial to the lender, even if they might increase delinquency in some groups, because they provide a buffer against the losses.

Larger all things being equal, or larger because they are buying more expensive + newer cars?

Generally, people with decent-to-good credit can purchase a new car with 0 down. People who purchase cars at the "buy here, pay here" type of used car dealer have to put large (near 1/2 of the balance) down payments (and generally, the "required" down payment is the amount that the dealer paid for the vehicle). I used to work in the car industry.

People who were upside down (or underwater if you prefer the real estate term) in their cars have the balance added to the new car loan. So there are quite a lot of car purchasers who have "negative" down payments to mess up the statistics.

A decade ago, consumers would stop paying all their late bills in order to keep their house. Today, they'll pay everything else before they pay the house. This further muddles any statistical analysis.

"Makes you wonder about pushes for higher mortgage down payments."

This is a ceteris paribus problem. All other things are not equal. The folks with larger car downpayments may be higher risk borrowers compared to other auto buyers.

For houses, the data appears to be quite clear. Higher down-payments are associated with lower foreclosure rates. Why? Because buyers with a larger down-payment are less likely to end up with negative equity and can sell their houses if need be. More broadly, substantial down-payment requirements tend to dampen bubbles. With zero down-payments the sky's the limit. At 20% speculators both need more cash to speculate and may be more wary of losing it. Stated differently, zero or low down-payments are effectively "free" options on houses. Heads the speculator wins, tails someone else looses. Note that low / zero down-payments are strongly associated with fraud. See "ML-Implode Statement On H.R. 6694 And FHA "Seller-Funded Downpayment Assistance" Loans" (

Any wonder the market went crazy after 2001?

For some research see,, and The last link claims that higher down-payments are only weakly associated with lower foreclosure rates. See also for a study that shows that lower down-payments correlate with lower, not higher foreclosure rates. Why? Selection bias. The lower down-payment group had much higher FICO score. This would appear to be the housing (FHA) equivalent of the care loans you mentioned.

Mortgages are no-recourse loans meaning that if you stop paying, they can take your house, but not your car, your bank account of anything else. So if you have a small down-payment a relatively small downwards price change will put your mortgage "underwater" giving you an incentive to walk away from it and the bank takes the loss. The higher the down-payment, the more the market has to fall before you can walk away from the house with the bank taking a loss.

Recourse varies significantly by state.

Incorrect. Or rather, incomplete. What you will find is that a large down-payment (or equivalently, a lower Loan-to-Value) means lower risk once you adjust for non-collateral-related credit quality. As several folks noted below, this can be hidden by the fact that the lender will then tend to ask for larger down-payments from the worst risks, hiding or even reversing the effect in the observed portfolio.

Carefully check the charts over at

The data is from 2001 which is probably a good thing (a "normal" housing market).

At equal FICO scores, higher LTV ratios strongly correlate with higher foreclosure rates.

Down payments for housing aren't to reduce delinquency. They're designed to make sure delequency doesn't really matter (since outside of perhaps twice in history houses haven't fallen by 20%) so the bank can simply sell the house. Cars depreciate much more quickly than land and houses.

+nelsonal, that mortgage lenders emerge whole is not the same as "delinquency doesn't really matter." Increased mortgage delinquency implies an increase in the foreclosure-sale portion of the housing market. Foreclosure properties typically deteriorate at an accelerated rate, so more foreclosures has consequences for the quality of the housing stock.

Downpayments reduce delinquency because they increase the probability of the homeowner having positive equity after a fall in home prices. People generally stop paying only when their equity goes negative.

I'm not convinced this is generally true; it certainly isn't true in my personal experience. People generally stop paying when they can't afford to make the payments. Can happen due to reduction in income, increase in the required payment, or exhaustion of credit, and more importantly it can happen even when equity is still positive. If you don't have the money to pay the mortgage, the mortgage doesn't get paid no matter what the equity. Conversely, if you have a steady and adequate income, a house you like, and good credit, you'll probably keep making the payment even if the equity goes negative.

Negative equity will matter in marginal cases, where a person can make the mortage payment but only by way of extreme austerity elsewhere - in that case, preservation of equity will serve as an additional incentive for said austerity. I am unconvinced that this is the dominant factor in default and/or foreclosure.

That's through 2009 for private lable loans, which are generally poorer on either income or credit or both.
The last page has a similar chart through 1Q which only shows LTV beyond 100% (negative equity).


You observations match the data I have seen. However, if you have positive equity, you can sell your house to avoid foreclosure. This is why the housing the bubble masked the sub-prime debacle for years. Sub-prime borrowers were still getting into trouble back then. However, they had an easy exit (at a profit no less). When the market topped and started to fall, the escape valve failed. Of course, may statements also apply to prime borrowers, etc.

So, then, what IS bankruptcy about?

Some people suck at time management: they will always procrastinate and leave things until the last minute. It doesn't matter how much extra time you give them, they will always be running late.

Some people suck at money management. It doesn't matter how much extra money you give them, they will always be broke.

Throwing money at a problem doesn't solve it. It's not a new insight.

Some people suck at time management: they will always procrastinate and leave things until the last minute.

*looks around furtively*

$50-150K is a large amount (and I assume that is the headline number, not the actual lump sum/after-tax amount)? Hardly the stuff of rags-to-riches.

It might seem like it to the winners and their entourage.

Agreed. $150k would allow me to pay off my educational debt immediately with some money left over. That would change my life in many ways.

Using a 4% discount rate over 25 years (it would vary from state to state), minus 25% for federal taxes (plus whatever state income tax is applicable) leaves you with about $42,000 if you won $150,000. Even in my student-poor life in the armpit of North Carolina, that would have lasted me four years. It would barely enable a person to purchase one of the low-end luxury cars. Hey, I wouldn't turn it down either (I wouldn't turn down the $100 miniprize), but this doesn't even reach Obama's definition of rich.

I'm no expert on the CA lottery, but I think the 26 annual payments provision only applies to the Jackpot prize. I've never seen the Jackpot at $150k. CA taxes don't apply, so just the 25% federal tax. Also, OBVIOUSLY WE WERE SHOOTING THE BREEZE ABOUT NET WINNINGS NOT CALCULATING WHETHER WE COULD BUY THAT ISLAND IF WE WON $300,000,000 WHAT IS WRONG WITH YOU?

That windfalls don't improve resource management skills has a lot of implications.

Is that dry sarcasm I detect?

Not really. I had a long bloviating paragraph composed and decided to go with the Captain Obvious approach.

"... obviously it restricts the sample to people who play the lottery."
Surprised it's taken nine comments to point out the obvious ...

Clearly the most significant issue at hand, given the natural skew of lottery players toward lower incomes and thus implied relative worse money management skills in aggregate.

Yes, but it's still interesting to compare lottery winners who won small to those who won big.

I don't think you need to take the detour through "lower incomes".

People who play the lottery are likely to be bad money managers.

I'd be interested in the causes of these bankruptcies. As I recall most US bankruptcies are due to medical problems or divorce, those events may be dramatic enough to overwhelm this amount of money.

Elizabeth Warren dishonestly but successfully got a large part of the public to believe that most bankruptcies are due to the rapacious health care system in this country[tm]. "Most bankruptcies due to medical problems" is only true if you count all of the following to be medically induced bankruptcies: problem gambling, substance addiction, disability of a breadwinner and thus loss of income [as opposed to the medical bills], and death of the breadwinner [hey, that's a medical problem, isn't it?]. See page W5-65 of .

If you go bankrupt and you had $1000 [one thousand dollars] in medical bills over the past two years you are considered to have gone bankrupt due to medical bills. Cut me a break!

See for a bit of a critique.


Can I make a bad pun about your name?

Nothing supports the "poor character" argument for persistent poverty better than giving a dirtbag some money and watching him plumb new depths. It can also be seen in the trust fund babies who lose the fortunes amassed by their forebears.

Beying born into poverty certainly is an obstacle to success, but by no means is it an impenetrable barrier. Beying born into privilege is a good launch pad for one's own success, but it's not a guarantee.

"dirtbag"? Really? All that this shows is that your wealth is not strongly path dependent. Is it really your position that someone who never learned money-management skills is a "dirtbag"? Is it really your belief that people who grew up in "the hood" and therefore never learned the basic social graces one must display at an interview is a "dirtbag"?

Having met many people from "the hood" who have good social graces, good money management skills, and good job skills, I feel supported in my belief that the DIRTBAGS who remained in "the hood" are there because of their own poor character or poor choices.

BTW, you were the one who injected race into this, not me. "The hood" can just as easily be replaced by "trailer park" or "the sticks" or "skid row" or any other place where DIRTBAGS congregate and remain.

There is an abundance of role models and strong character to be found within reach of our poorest people. There is an abundance of opportunity with free public schools and social assistance. Quite frankly there's no excuse for failure. Success and character aren't measured by the size of one's wallet. Staying on the right side of the law, having any form of productive employment, and limiting ones vices is sufficient in my book to avoid being a DIRTBAG.

Michael Vick, for example, is rich and successful, but he displayed his true lack of character when he obtained wealth. Many politicians are DIRTBAGS.

It's not a social crime to be born poor. It's a social crime to remain poor for lack of trying. If anything, those people who rose from poverty through honest means display more character than most people I know.

Here's a video from a "brutha from da hood" PrometheeFeu who is down with what my main man Mike be saying.

Caution: Harsh language but really funny and succinct.

Oh, darn. The link didn't work.

Anyway, it's Felonius Munk Presents: Stop it B!

On YouTube.

although obviously it restricts the sample to people who play the lottery.

I think that's a large confounding factor, given that the lottery is commonly referred to as a tax on not understanding statistics.

I think that’s a large confounding factor, given that the lottery is commonly referred to as a tax on not understanding statistics

Perhaps, it's not entirely clear to me that playing the lottery is necessarily irrational or the result of ignorance -- sure the net expected value is negative, but (given that lotteries provide a way of, say, turning $1 into $1 million, for someone who has no realistic prospect of getting that kind of money otherwise) it could fulfill some perfectly rational risk-return function. That's excluding the hedonic value of gambling. Insurance (if it is actuarily fair) also has a negative net expected value, but we hardly call people who buy insurance policies ignorant of statistics.

I think this datum is an illustration of a general principle that unsophisticated, non-business people who suddenly get a lot of money often do not keep it (through poor resource management skills, short horizon, poor impulse control or whatever). Another example is professional sports -- as Tyler has noted, 78% of NFL players are bankrupt or under severe financial stress two years after their careers end (and the number is almost as high in the NBA), and there are plenty of similar examples in the entertainment industry (people like Elton John who generate huge incomes and then waste them).

I think this belief that the lottery is "a tax on people who can't do math", or the belief that insurance is a self-evident good is symptomatic of a larger technocratic view of society. In order to decide whether a risk is "worth it". You must refer to the risk preferences of the people who are considering whether to take that risk or not. Yet obviously, for a technocrat, this is an unacceptable situation since it means that no value judgement can be made about the actions of people who chose to take risks. No policy can be derived except laissez-faire. They must pick a particular level of risk. So they use an availability heuristic and select most often an expected value of 0 because it sounds good. (Also, the equation is easy to solve and it looks more scientific to have both sides of an equation be equal than having both sides be different by an arbitrary looking number) So since the expected value of playing the lottery is negative, it is "a tax on people who can't do math" as opposed to the expression of lottery players' hedonic desire to play or a risk preference.

I certainly don't ignore the 'hedonic value' of playing the lottery (is that how people with a Ph.D. pronounce 'fun'?) I have a good friend with an accounting who plays the lottery. He's clearly not innumerate, and he told me he knows he's taking a losing bet, but what he's really buying isn't a chance at winning but hope which, to him, is worth a more than a few dollars. I can get behind that.

But do you really think most lottery players are that self-aware? During college I worked summers at a construction company, which should be a legally mandated experience for any wannabe white collar workers. Those guys played the lottery every week, and not just $5 or $10 worth. Ten times that, easy, not to mention the scratchoffs they also bought to tide them over during the week. They seriously, genuinely believed that the lottery was the best path to financial independence. Where some people have a 401(k), these guys had fists full of lotto tickets.

For the kind of person who makes a career out of working manual labor in the construction industry, a lottery jackpot may very well be the best path to something resembling financial independence for them.

Under a certain set of (IMO unrealistic) assumptions, that might be true. But ...

First, the people I'm talking about were at least semiskilled, making over double the minimum wage at the time, plus fairly regular OT. (Some were actually skilled, but those guys generally weren't the heavy lotto players.) While we aren't talking about people who are well-off, neither were they destitute. They had the financial capability to save at least some money.

Second, this depends on your definition of financial independence. If you're talking about the proverbial "fuck you" money, then yeah, the lotto is their best shot at that. If you're talking about financial independence in terms of owning your own home and being debt free, there's no way the lottery ticket is the best path.

Example. One of the guys had a Pontiac Firebird (and this was in about 2001, 20 years after the Firebird's heydey) which he was paying a monthly note on. He was constantly bitching about the payments. Given that the dude was a financial wreck, I'm sure he was paying 10%-15% interest on that car note. Even given the most generous assumptions in favor of lotto playing, I cannot imagine how you could show that spending $150-200 a month on lotto tickets is a better bet than using that money to pay off the Firebird.

I've heard it said by other commenters on this blog that playing the lottery is in some ways the inverse of buying insurance. While buying insurance lowers your expected outcome in return for reducing your variance, the lottery lowers your expected outcome while increasing your variance.

Not really.

I totally get why expected negative return, but *possibility* of highly positive return, can be attractive to some people. A 1:3million chance of winning a million, may to some be better than a 1:1 probability of keeping the $1. That's not a mystery.

But a 1:3 million chance of winning a million *must* to any rational player who gets what he's doing, be less attractive than a 1:1 million chance of winning a million.

That is, if the high-return low-probability gamble is wanted, you should nevertheless prefer the gamble where the probability is as high as possible.

People play lotteries for substantial sums. Here in Norway 25% of those who gamble do so for atleast $2000/year.

There's simply better gambles to be had - if you're willing to bet $2000/year over 20 years, and your hoped-for income is winning a million.

Out of the money futures, for example, on the average give about a 1:1 expected return (thus they're BAD investments, since expected return unity on a highly risky gamble, is bad investing).

But as lottery-tickets they're spectacularly good ! 1:1 expected return is bad, but it's a hell of a lot better than the 1:0.5 you get in typical lotteries.

Very good post.

I was referring more to the social aspect of the expression itself and its reflection on which people choose to play the lottery -- rational or not.

There are also a class of ":unsophisticated, non-business people" who do not play the lottery. I would argue they are less likely to experience bankruptcy.

Paradoxically, the people who are best able to handle a windfall probably skew toward not playing the lottery.

Not true. It means that people have far different levels of risk aversion than you. Or it may mean the marginal utility of a disposable dollar is low. Or the slim hope of hitting the lottery bears more utility that what a dollar can buy.

Contrary to popular misconception, middle class people spend the most on lottery tickets. But contrary to state propaganda, the poor spend a greater proportion of their income on lotteries than higher income earners. If it's viewed as a tax, it's very regressive.

If it's evidence of ignorance of statistics you seek, find it in the people who choose lottery numbers based on birth dates. The relatively high concentration of choices from 1-12 and 1-31 increase the likelihood of splitting the prize money, increasing the expected loss. Any method of choosing numbers in a systematic way reduces your expected payout substantially. The morons think, "Winning half of $50 million is fine by me!" as if they've already won it.

Or the slim hope of hitting the lottery bears more utility that what a dollar can buy.

Yes, that's the essence of a tax on not understanding statistics. All utility is in the eye of the beholder but some is more rational than other.

But see above, my point is more about who chooses to play and how that might skew this study's sample.

You have no point.

A person with a higher preference for risk than you or with hyperbolic discounting of utility is NOT irrational. They might find themselves perpetually coming up short, but it's their life to live. It's no less irrational than enjoying a cigarette or driving really fast or going to a bad neighborhood to get some kicking barbecue.

Ya rolls ya dice, ya takes ya chances.

A person who takes a 1:3million gamble in the hopes of winning a million, in a market where 1:1million gambles of doing the same are abailable, is irrational.

If "making decisions in such a way as to maximize your likelihood of unhappiness" isn't the definition of irrational, then what is? As I read this post, it's impossible for anyone to act irrationally, because you can just explain it away as differing risk preferences or present value preferences.

I've heard that if 1 2 3 4 5 6 is ever drawn in a major national lottery, the jackpot will be split something like 10000 ways with a number of lesser prizes as well.


Every lottery has some smart ass who took statistics and realizes that 1, 2, 3, 4, 5, 6 has the same probability of occurring as any other combination.

But since smart asses are not unique, it's highly likely there will be numerous people who will make this foolish error in judgment. If you are going to play, your job is to minimize your expected loss.

You should never play birthdays, sequences, numbers from Lost, or any other set of numbers that are likely to be duplicated by anyone else. It's best to let the machine make a random draw. Of course, if the machine randomly chooses numbers concentrated between 1-31, you just randomly got screwed.

The same is true of athletes who win the 'pro sports lottery'. A large number of ex-pro athletes go bankrupt not long after their sports careers end:

As far as I can see, in the USA poverty is not simply a lack of money.

A comnfounding factor: Bankruptcy is of no use to the extremely poor, who have no assets and no means to take on debts. The very poor also may not be able to find a lawyer, or pay filing fees. Winning the lottery simply moves some people from the category of those who can't avail themselves of bankrputcy to those who can.

Hello...Good post...

Anyways, well...Economist Ludwig Von Mises stated that people with capital (money) can lose it just as easily if they cannot manage it well. Such examples can also include inheritance transfers of wealth. If the son or daughter cannot manage their parents capital well they will lose it just as easily as they gained it.

A person who earns money spontaneously will not know what to do with it or how to invest it. The person who earns it through hard work (and NOT exclusive government grants, lottery wins or inheritance transfers of wealth) WILL know (most of the time) what to do with the money in the future and how to effectively manage it.

Old Old saying captures it all. Income 18 pounds, Expenditures 18 pounds 1 shilling, result misery. Income 18 pounds Expenditures 17 pounds 15 shillings, Result, Happiness.

No matter how much money you give people that spend more than 100% of their income, bankruptcy will eventually come.

I think the lottery winner restriction is really interesting---these are some of the people we want to help most manage their finances. We found out one thing that doesn't make them rich--money, maybe there are things that do make them rich--education, financial advice, a structured life with a good job. Personally I find structure is immensely important to have in my life. If you win the lottery, you quit your job, and maybe your structure goes to hell. This doesn't need to turn into a not-worth-helping-out-poor-people-b/c-they'll-just-lose-it party but maybe a reason to think deeper about what it means to help.

Bankruptcy is about immaturity, lack of faith in our institutions, and delay of gratification.

With money, you can put it in a time capsule and forward it from present self to future self. There's a lot of people out there who when they do get money, believe they had better spend it as fast as they can, before they lose it.

If present self is not mature enough to realize that future self will need money and take appropriate steps to secure that reality, the only option is bankruptcy.

> Bankruptcy is about immaturity, lack of faith in our institutions, and delay of gratification.

Or maybe, sometimes it's about "mature", long time-horizon planning, hyper-rational, behavior from a young person (perhaps an econ graduate who takes their teachers seriously)?

To such people, bankruptcy law is society's way of granting you an option: take a huge bet and if you succeed spectacularly you are among our capitalist worlds' kings -
but if you fail you have a limited downside (back to zero! Is this so bad if you are young and single, especially so if a young man?)

Yes it's a mystery that other parts of society seem strangely willing to help you in placing this bet - but they do and that's arguably their problem; our hypothetical young man takes rational and mature advantages of his circumstances.

> as Robin Hanson might say, is that bankruptcy isn’t about money.

Perhaps you can you enlighten us by explaining?

Hanson's basic trick is that "X" isn't about "Y", but that (...argument-to-fool-a-toddler-for-10-seconds...) "X" is really about "status".
Are you perhaps suggesting bankruptcy is about status? This would be quite something! Please, please, give us a bit more.

Isn't this finding still consistent with the permanent income hypothesis? Large positive shock leads to ramp up in expenditures for liquidity/credit-constrained households, but over time they return back to their long-run low levels of expenditure (and savings).

Could the study be redone differentiating between winners who bought a lottery ticket and winners who were given their tickets by someone else? This wouldn't necessarily be a perfect control; people who are given lottery tickets as gifts may STILL be statistically different from the general population. But it would be a start. If there is still no effect on 5-year bankruptcy rates, I'd consider that extremely credible. (I'm sure data would start to look sparse at that point, but it might be worth a try.)

I've always thought that if I had something like a big lottery win, I'd be so paranoid about blowing it all that I'd fail to get much enjoyment from it. But there is something to be said about being secure in a low profile lifestyle. I'd still have to find some sort of work just to feel alive. But the earning level would no longer be a concern. I'd just need to feel I was doing something worthwhile and that it mattered whether I showed up.

Comments for this post are closed