When he introduced a bill last year to increase the tax rate on Montana's
video gambling machines, Rep. Butch Waddill argued that other states (like
South Dakota) tax gambling machines at far higher rates. And he argued
that Montana, facing a substantial budget deficit, needed the revenue.
"This is a significant way to get some money," he told the Helena
Independent Record. "And it's not taking food out of anyone's
While Waddill's bill didn't passthe gambling industry opposed it
and others were skeptical it could raise as much money as promisedhis
comment raises an interesting issue. Where does the money spent on lotteries,
pulltabs or video gambling come from? More specifically, do these forms
of gambling draw greater expenditure from the rich or the poor? And do
they compete with essentials in the family budget, or simply come out
of discretionary income, money that would otherwise be spent on DVDs or
a seat at a ballgame?
Most economic studies have found that gambling expenditure and, therefore,
gambling taxes are regressive in the sense that the poor spend a higher
proportion of their income on gambling than do the rich. Indeed, even
though participation rates are similar among demographic groups, research
shows that there are significant differences in the levels of spending
among groups, and generally speaking, members of less advantaged groupsthe
poor, racial minorities and less educated spend considerably more per
The 1999 National Survey on Gambling Behavior, for example, conducted
by the National Opinion Research Center (NORC) at the University of Chicago,
found that while whites, blacks, Hispanics and others have roughly equal
lottery participation rates, blacks spend far more annually on a per capita
basis. And while equal percentages of high school dropouts and college
students had participated in a lottery within the past year (48 percent),
the dropouts had spent $334 per capita in that year, while the average
college grad had bought just $86 of tickets.
When it came to income, households with incomes in the $50,000 to $99,999
range actually participated in lotteries at a higher rate (61.2 percent)
than households with less than $10,000 in income (48.5 percent), but those
poorest households spent an average of $520 on lotteries in a year, while
the richer households spent $301. Industry groups have critiqued these
NORC numbers, but most independent, academic studies support their general
One of the more interesting studies of lottery economics
was published by the National Bureau of Economic Research in November
2002, just as North Dakotans were voting themselves a new lottery. The
analysis, by economist Melissa Schettini Kearney at Wellesley College,
looked at household spending data from 1982 to 1998 (a period in which
21 states adopted lotteries) and compared changes in household gambling
expenditure in states that adopted lotteries to those in states that didn't.
Kearney found that total household gambling increases after a state introduces
a lottery, meaning that spending on gambling doesn't just shift from slot
machines to scratch-off tickets, for example, but that lottery money comes
from some other pocket of the family budget.
Kearney found, in fact, that after a state lottery is introduced, the
average household reduces its monthly nongambling consumption by about
$38 a month. And "among households in the lowest income third,"
she wrote, "the data demonstrate a statistically significant reduction
in expenditures on food eaten in the home (3.1 percent) and on home mortgage,
rent, and other bills (6.9 percent.)"
Sounds bad, perhaps, as it implies that people do in fact sacrifice food
and housing for illusive gambling gains. But Kearney pointed out that
a new lottery could be "welfare-enhancing" in the sense that
adoption of a state lottery lowers the price of gambling and so people
can buy more of it, if they so choose.
Prices are lower both because consumers don't have to drive to a neighboring
state to buy a ticket and because whatever negative stigma was previously
attached to lottery participation is decreased by the state's support
of it. "They're just responding to the price change as predicted
by standard economic theory," she said. And welfare is enhanced,
in an economist's view, whenever a consumer is given more choices at lower
But economic theory also assumes perfect information, and Kearney looked
at this side of the question as well. Do consumers understand the gamble
they're buying? Are they making informed choices? By analyzing weekly
data from 91 lotto games between 1992 and 1998, Kearney found statistical
support for the idea that lottery sales are driven by the expected value
of a gamble (prize amount times the probability of winning). She also
found that consumers respond to "entertainment" features of
lottery games that are unrelated to potential gains (for example, the
number of drawings per week or the number of digits chosen). Taken together,
these results imply that lottery gamblers are "at least partlyand
potentially fullyinformed, rational consumers."
But Kearney offered two caveats. First, some households undoubtedly reduce
monthly consumption by far more than the average $38 drop; and second,
within each household, individuals may have very different preferences
regarding gambling vs. food. "So a parent might have the money and
decide to spend less on food and more on lottery tickets," explained
Kearney. "That makes the parent happier, but doesn't benefit the
children, so on net, this might decrease social welfare."