Timothy T. Taylor Home Page
Journal of Economic Perspectives
Articles and Writing
Economics Textbook
Classroom Teaching
The Teaching Company
High School Pedagogy

Articles and Writing

January 28, 1990
"The State Lottery is Almost a Sure Bet: You'll Lose"
San Jose Mercury News
By Timothy Taylor
<< Back to 1990 menu

THE California lottery is a thorn under my skin, an itch I can't scratch, a blister that won't heal, a constant irritation to me. The parade of news stories about jackpots and winning numbers makes me tense and peevish, because I see the lottery as inherently deceptive and predatory.

My friends who play the lottery find this attitude a bit quirky, but they're inured to a certain amount of quirkiness on my part. For them, the lottery is a minor indulgence. They buy a few tickets when they think of it, or when the Lotto jackpot gets particularly high.
But the lottery doesn't make its money from the average player. The 10.8 million households in

California bought $2.8 billion worth of lottery tickets last year, for an average purchase of $260. However, that average includes many who never buy a ticket. About half of lottery sales are made to just 10 percent of households, and two-thirds of sales are made to 20 percent of households, according to "Selling Hope: State Lotteries in America," a just-published book by Charles T. Clotfelter and Philip J. Cook of Duke University.

These estimates imply that about a million California households are spending an average of $1,300 a year on lottery tickets, which qualifies as more than a harmless diversion.

The heavy players aren't mainly Gucci and Rolex types who can easily afford the losses, either. Clotfelter and Cook cite evidence that a heavy player is at least as likely to be poor as to be rich.

In general, the lottery hits the poor more heavily than other ways the state has of collecting revenue. Economists classify a tax as regressive if the percentage of income going to the tax declines as income rises. A 6 percent sales tax is regressive, for example, because a poor household that spends all its money pays the 6 percent on all of its income, while a higher income household that spends only half its income (and invests the rest) pays 6 percent on half its income, or 3 percent on its total income. According to a variety of economic studies, a lottery is about twice as regressive as a sales tax.

Christopher Schmitt, a Mercury News reporter, documented the regressivity of the California lottery in an article published last May. He compared the sales of lottery tickets by zip code district to the average incomes of people in those districts. He found that sales per resident of poor districts were twice as high as sales in more affluent districts.

But at least the suckers volunteer to contribute their money, right? Probably not; not if "voluntary" is defined as a matter of informed consent, understanding the odds and payoffs, rather than as a matter of being enticed by unrealistic fantasies.

For example, consider the odds of winning Lotto, the game which accounts for nearly three-quarters of lottery sales. Lotto involves choosing six numbers between 1 and 49, and the odds of choosing correctly are approximately 1 in 14 million.

Do you have a firm grip on how large 14 million is? Let's say that the newspaper were to print all 14 million combinations of six numbers. It prints them five columns per page, 160 lines per column. How many pages will it take to print all 14 million combinations?

On a typical weekday, this newspaper might have 72 pages, including the classified ads. Let's assume that the Sunday paper is twice as big. At that rate, it would take seven months worth of newspapers, 17,500 pages filled top to bottom with nothing but combinations of lottery numbers, to list all 14 million choices. Pile up your newspapers until next August, and then contemplate the stack and your odds of winning.

Or remember the pathetic story of the fellow who took out a $20,000 loan to buy Lotto tickets. Betting on 20,000 numbers raised his odds from 1 in 14 million to 1 in 700. In other words, if he had played $20,000 every week for 700 weeks (13 years, 6 months) he would on average win once. But of course, anyone who spends $14 million on lottery tickets will on average win once.

To add a bit more challenge, Lotto will add four numbers later this spring, so that it will involve choosing six numbers between 1 and 53. The number of combinations will then expand to 23 million, or enough to fill the newspaper for nearly a full year.

If you are blessed enough to hit the right combination out of that truckload of newsprint, you'll find that the announced Lotto prize overstates its actual value. The announced prize is before tax, and federal income tax will take 28 percent off the top. In addition, the announced prize involves some disingenuous fiddling with time.

Here's how: Imagine that someone owed you $100. They offer to pay you by putting $37 in a savings account. They explain that over 20 years at 5 percent interest, the $37 will have grown to $100, and you'll have your money. This would be a self-evident fraud; you don't have to be an accountant to recognize you aren't getting the $100 you were promised.

But the lottery announces its prizes not as an amount that you'll receive today, but as the sum of delayed payments. A $5 million Lotto jackpot, for example, is actually $250,000 a year for 20 years, before taxes. But as the previous example illustrates, $100 dollars received a decade or two from now isn't worth as much as cash in hand today. A rule of thumb is that the announced value of the Lotto jackpot should be discounted by two-thirds to adjust for the tax bite and the fact that the payments are delayed.

Now, overstating the value of the prizes may not seem like a big deal. One-third of an announced $10 million jackpot is still real money. But it's clear that big jackpots attract more players, which gives the lottery an incentive to overstate the size of jackpots. By announcing the sum of payments made over 20 years, before tax, they are doing their best to exaggerate without actually lying.

But at least the lottery raises money for California education, right? Well, sort of. The lottery will contribute about $1 billion to education this year, but it's not clear whether this is additional money for education funding, or whether the lottery is just freeing up funds to be used elsewhere.

For example, total state spending on education (including the general fund, bond funds, special fund and lottery money) was about 47 percent of total state spending both this year and in 1984-85, the year before the lottery started, according to a spokesman at the state Department of Finance. It's very difficult to demonstrate convincingly that lottery funds have actually increased spending on education, largely because it's hard to say what would have happened to education spending if the lottery hadn't been adopted.

In addition, the lottery does not provide a huge amount of money in relation to the size of the California budget.

Governor Deukmejian's recent budget lists $29 billion in education spending from all sources (four-fifths of it for K- 12). The $1 billion in lottery money is 3.4 percent of that amount. The lottery raises about as much money as the state tobacco tax, or about as much as would be raised by an increase of one-third of 1 percent in the state sales tax.

It's terribly ironic that the lottery was promoted as a way to benefit education, when classes in math and statistics and economics would strongly discourage people from believing in lucky numbers. It's as if the state sold horoscopes and magic crystals to finance science classes, or sponsored showings of "Gone With The Wind" to finance instruction in black studies.

I'm not opposed to legalized gambling. Although I personally would rather visit the dentist than Las Vegas, I can accept that others have different preferences. But it nags at my gut that the state of California doesn't just allow gambling, but promotes it.

In Deukmejian's most recent budget, lottery expenses for "advertising, promotion, and public relations" total $83 million, but that promotional budget is probably less important than the television and radio and newspaper coverage of lottery winners. Every drawing is a mini-drama, and every winner is a custom-made human interest story.

Of course, it's a bit tougher to interview the thousands of people who lost hundreds or thousands of dollars. But since the lottery pays out only half of what it takes in, every headline about a dollar won represents the missing story of two dollars lost by others.

The California lottery pushes its product, aiming at increased participation and sales every year. Deukmejian's budget notes that Lotto and Scratchers "have reached maturity and will be freshened with new features to sustain sales levels."

Through the lottery, state government relies on encouraging the poor and those with a gambling problem to lose their money at a game with terrible odds and overstated jackpots. I don't ask that the lottery be abolished. But instead of measuring success by increased lottery revenues, I think the lottery should be counted a success when Californians become realistic enough to reduce the number of tickets they buy, and responsible enough to find other ways of financing education.

<< Back to 1990 menu