Tens of thousands of Hungarians rallied to protest a proposal to create a tax on internet usage in Hungary. In response to the first protest on Sunday, the Hungarian government limited its proposal.
When I first started researching state-run lotteries, like many people, I had never thought of lotteries in terms of tax policy. For me, the lottery conjured up images of smiling Powerball winners displaying $10 million checks for the TV camera. Occasionally I heard of lottery players suffering financial ruin or gambling addiction. But as I researched, I learned that in between these two extremes is the less glamorous but equally important issue of the lottery’s effect on state finance.
Lotteries are fundamentally different from other gambling in one important way: they are provided by the state, and only by the state. Therefore, they raise policy questions entirely separate from issues of morality and addiction. The lottery is more than a controversial way to add a little money to state coffers; it is a tax and should be evaluated as such. When we subject it to the tests of good tax policy, it fails.
We’ve all heard people say glibly that lotteries are a tax on the poor, a tax on people who are bad at math, even a tax on the stupid, but most people don’t actually believe the lottery is a tax.
In 1732 Henry Fielding wrote the following:
A Lottery is a Taxation,
Upon all the Fools in Creation;
And Heav’n be prais’d,
It is easily rais’d,
Credulity’s always in Fashion;
For, Folly’s a Fund,
Will never lose Ground;
While Fools are so rife in the Nation.1
Fielding could well have been talking about 21st century American lotteries. His contemporaries did not purchase computer-generated lotto tickets from the 7-Eleven, nor did they watch Powerball drawings on TV or check the winning numbers on the internet. However, the basic public policy issues inherent in lottery finance have remained much the same since Fielding’s day.
To understand how the lottery evolved into an instrument of state tax policy, we must travel back almost 400 years. The history of state-run lotteries in America is a long roller coaster ride that has veered several times between outright prohibition and enthusiastic promotion.
It all started in 1612 when King James I of England created a lottery in London to aid Jamestown, the first British colony in America. The colonists enthusiastically embraced England’s lottery tradition and started their own, organizing both public and private lotteries. Proceeds were used for various public works projects, such as bridges, libraries, roads and lighthouses. The line between public and private was often blurred and proceeds also benefited private universities and churches. There were over 160 colonial lotteries before the start of the Revolutionary War,2 and they even provided some of the funding for the war.
Lotteries soon came to be seen as more of a civic responsibility than a form of entertainment or gambling.3 The colonists viewed them as a more palatable revenue raiser than explicit taxation. And, in fact, some people did see the lottery as a type of tax. In 1892, A. R. Spofford, Librarian of Congress, wrote, “[The lottery was] not regarded at all as a kind of gambling; the most reputable citizens were engaged in these lotteries. . . . It was looked upon as a kind of voluntary tax. . . .” 4 Adding to the lottery’s appeal was the shortage of other sources of public funding. Taxes were unpopular and, prior to 1790, there were only three incorporated banks. Lotteries therefore helped fill a void in both public and private financing.
They were especially popular during the period following the adoption of the Constitution and prior to the establishment of effective local taxation. From 1790 to 1860, 24 of the 33 states used them to finance jails, courthouses, hospitals, orphanages, libraries, schools, colleges and churches. While townships and institutions were sometimes granted permission to hold lotteries, they were mainly the province of the states, although there were a few dismal exceptions: between 1792 and 1842, Congress passed a series of federal lotteries to improve roads and infrastructure in Washington, D.C. Tickets sold well, but the agents conducting the lottery absconded with the proceeds.
The 19th century brought a proliferation of shops that sold only lottery tickets; in 1831, in Philadelphia alone, there were over 100 such shops.5 Despite their rapid geographical and financial growth, state-authorized lotteries were not without problems. Private firms gradually took over more of the day-to-day operations, and fraud became rampant. Many lotteries awarded fewer prizes than advertised or none at all. States found themselves unable to regulate the industry and began to consider prohibition. Anti-gambling sentiment became part of the social reform movement already underway, which included abolition of slavery and promotion of women’s rights and temperance. People began to see gambling in general and lotteries in particular as a way of taking advantage of the poor. States started banning them, first in the Northeast, then in the South and West.
There was a short-lived lottery resurgence after the Civil War. The South needed funds to rebuild, and lottery revenue raised for Reconstruction was seen as a voluntary tax.6 In 1868, Louisiana legislators accepted bribes in exchange for granting the Louisiana Lottery Company a 25-year charter as the sole proprietor of the state lottery, and by 1878 Louisiana had the only legal—albeit scandal-ridden—lottery in the country. It finally came to an end in 1890 when Congress enacted a prohibition against interstate commerce involving lottery tickets, followed by a prohibition against all mail related to lotteries. By 1894 there were no legal state lotteries and 35 states had constitutional prohibitions against them.
There followed a 70-year period of prohibition against lotteries, from 1894 to 1964. However, at no point were they forgotten by the public or by elected officials. A group of New York philanthropists founded the National Conference for Legalizing Lotteries to support enactment of state and federal lotteries to fund hospitals and other charitable causes. During the Great Depression, there was a flurry of proposals for lotteries at both the state and federal levels to fund unemployment relief, and during World War I, members of Congress introduced lottery bills to help defray the costs of the war. Public support for these proposals was usually high, with every poll taken after 1938 showing more support than opposition.7 Meanwhile, illegal lotteries, known as “policy” and “numbers,” flourished, and in 1930, approximately thirteen percent of Americans participated in the Irish Sweepstakes by purchasing tickets that had been smuggled into the country. 8
In 1931 Nevada re-legalized casinos, and 33 years later New Hampshire ushered in the modern era of state lotteries. In 1964, after 27 years of annual lottery bills in the state legislature, New Hampshire introduced a lottery, approved by 76 percent of the voters in a public referendum.
In the years leading up to 1964, there was not only a growing acceptance once again of gambling in general, but also a growing opposition to tax increases. New Hampshire had no sales tax or income tax and relied heavily on property and excise taxes, with over half of the state revenue coming from excise taxes on alcohol, tobacco and horse racing.9 The lottery was thought to be part of a continued resistance to a sales tax, and also a way to increase state aid to education, and deal with the state’s budget deficit.
New Hampshire paved the way for New York to establish a lottery in 1967, followed by New Jersey in 1970. It was approved in New York, despite considerable opposition, in part because legislators had recently increased the sales tax and were reluctant to increase sales or income taxes further. Lotteries spread rapidly to other states, first in the Northeast and then the West and Midwest, with the South the last holdout. In many states a referendum or initiative was necessary to remove a constitutional ban. In the 1980s, convenience stores and private suppliers of lottery equipment began lobbying for the passage of lotteries and conducting petition drives among voters in states where legislators were hesitant or opposed. Almost all of the referenda and initiatives passed, although some states approved lotteries through legislation alone.
Currently, 40 states and the District of Columbia operate lotteries, with two more soon to start. Oklahomans approved a referendum in November for a lottery to support public education. They had rejected the referendum in 1994 but may have been swayed this time by the expensive pro-lottery campaign conducted by a large coalition of education groups.
North Carolina will also soon have a lottery, as the legislature approved one in August. It has been a controversial issue in the state for twenty years. North Carolina’s road to lottery enactment was longer, more contentious, and more partisan than many states’ and highlights some of the concerns people have long had about lotteries. Southern states have been the most reluctant to enact them, but worries about "losing" money to neighboring states have prompted many to succumb. South Carolina and Tennessee recently gave in, leaving North Carolina completely surrounded by lotteries, and the only state on the East Coast without one.
The governor was an outspoken supporter of a lottery to fund education, but the House and Senate had long been divided, mainly along partisan lines. The two sides fought bitterly throughout the spring and summer, even including a projected $425 million in lottery revenue in the state’s budget, before the lottery legislation had even been approved. Approval of the Fiscal Year 2006 budget was delayed over a month into the fiscal year, in part because of debate over the proposed lottery. After the budget was approved, the House, worried that the Senate might reject the lottery bill, sought another source of education funding and passed a bill authorizing certain counties to raise their sales taxes. Finally, when it looked as if the lottery bill were dead for the year, the Senate passed it with a tie vote that was broken by the lieutenant governor. Ticket sales could begin in as little as six months.
North Carolina Democrats largely supported the lottery, but the North Carolina Republican Party platform states: "A state lottery turns government into a bookie, succeeds only on the basis of false advertising, capitalizes on broken dreams and personal irresponsibility, and places the burden of taxation most heavily on those who are least able to afford it."
In 1999 the National Gambling Impact Study Commission, which was created by Congress in 1996 to “conduct a comprehensive legal and factual study of the social and economic impacts of gambling,”10 recommended a moratorium on the expansion of all gambling in the U.S. Clearly its advice has not been heeded, as many states are waging battles over lottery enactment and expansion. Utah is the only non-lottery state not seriously considering enacting a lottery, and many states are expanding existing lotteries, ostensibly to deal with budget woes. They’re joining multi-state games such as Powerball, attempting to sell tickets on the internet,11 and installing video lottery terminals at racetracks. Legislators in states that have recently enacted and expanded lotteries have repeatedly stressed the need for new revenue sources to close budget gaps, but they may be confusing a need for revenue with the desire to spend more, and voters seem to be convinced of the need.
Even states that are not experiencing fiscal crises are likely to enact lotteries simply to avoid losing money to neighboring states. A state whose neighbor has a lottery is more likely to enact one itself, due to concerns that its citizens are spending money in other states, and the pragmatic notion that people are going to gamble anyway, so they may as well spend their money at home. Groups pushing for lottery enactment have gone so far as to run advertisements capitalizing on the desire to keep revenue in the state. When North Carolina was embroiled in its lottery debate, a group called the North Carolina Lottery for Education Coalition created TV commercials featuring a fictional South Carolina convenience store clerk named Bubba who enthusiastically sold lottery tickets to North Carolinians who crossed the border just to buy them. Bubba thanked North Carolinians for spending money on his state’s lottery and helping pay for South Carolina students’ education. The “loss” of revenue to South Carolina became one of the centerpieces of the fight for lottery enactment in North Carolina.
In many non-lottery states, it is no longer a question of whether a lottery will be enacted, but a question of when. Twenty years ago, researchers searched for the reasons certain states enacted lotteries. Now the question is, Why do certain states not enact lotteries? What unusual factors have prevented lottery enactment in Alabama, Alaska, Arkansas, Hawaii, Mississippi, Nevada, Utah and Wyoming? In some of these states, it is easy to pinpoint factors. In Utah, there is widespread religious opposition to gambling. In Nevada, the casino industry has a vested interest in preventing the competition lotteries would pose. Alaska’s and Hawaii’s geographic isolation means they don’t have to worry about “losing” money to neighboring states.
In order to appreciate the debates over lottery revenue, we need to understand the magnitude of the industry. In Fiscal Year 2004, lotteries generated over $48 billion in consumer spending, up from almost $45 billion in Fiscal Year 2003.12 This figure is not so surprising when we consider that, in Fiscal Year 2002, the average American spent more money on lotteries than on reading materials or attending movies.13
Slightly over half of the money spent is returned to players in the form of prizes. Part of the remainder covers operating costs—including vendor commissions, equipment, administration and advertising—and the rest is transferred to state coffers. The states call their portion “profit,” but, as will be discussed later, it is actually tax revenue. Nationwide, in Fiscal Year 2003, 31% of spending—or almost $14 billion—was transferred to state coffers.14 The breakdown of prizes, operating costs and government transfers varies from state to state. Nationwide, from the New Hampshire lottery’s inception in 1964 through Fiscal Year 2002, lotteries have paid 53 percent in prizes and transferred 35 percent to the state. The remaining twelve percent covered operating costs.15
In Fiscal Year 2003, the amount of money raised by lotteries varied considerably by state, and comprised 2.3% of own-source general revenue in the average lottery state. The amount ranged from a high of 7.7 % in West Virginia to a low of 0.3% percent in Montana.16 Many people believe this revenue helps state governments keep taxes lower. They are mistaken on two counts. First, the lottery itself is a tax; second, state and local taxes as a percentage of income are slightly higher in lottery states.17
Is this $14 billion really tax revenue? No government—state or federal—labels it as such. However, despite the lack of a formal definition as a tax by a government agency, lottery “profits” constitute an implicit tax. When state governments removed lottery prohibitions from their constitutions, they did so only for themselves. Seeing lotteries as a potential goldmine for state coffers, they maintained the ban on private lotteries and created for themselves a monopoly and, in effect, a source of tax revenue.
Lottery proponents argue that a tax is a mandatory or compulsory payment, and playing the lottery is voluntary, so lottery revenue cannot be a tax. But they're confusing the purchase of a product with the payment of the tax on the product. True, the purchase of a lottery ticket is voluntary, but the tax portion of the ticket price is not, just as a sales or excise tax is compulsory on a voluntary purchase of alcohol, clothing or books. The voluntary nature of the purchase does not make the tax any less of a tax. Using the lottery supporters’ rationale, we’d have to say that because the purchase of a book is voluntary, the sales tax on the book is not really a tax. Just try to buy a $20 book and hand the cashier a $20 bill, but refuse to pay the $1 sales tax and leave the store with book in hand. The only difference between the lottery tax and sales or excise taxes is that the lottery tax is built into the price of the ticket, rather than reported separately.
At the heart of the spurious if-it’s-voluntary-it-can’t-be-a-tax argument is the assumption that, since the lottery is a recreational activity rather than a necessity, only people who can afford it and enjoy it—those who are willing and able to pay —will participate. Presumably, government revenue that is contributed enthusiastically and voluntarily is preferable to revenue that is contributed under duress. This argument seems to suggest that the lottery is akin to a sort of user fee, or a charge paid to the government for a specific service, by the people who use that service.
Understanding the different categories of government revenue can shed some light on the classification of lottery revenue. The Census Bureau Government Finance and Employment Classification Manual, the definitive publication on classification of government revenue, considers lottery proceeds “miscellaneous general revenue,” and lotteries a “general government activity.”18
The Census Bureau calls user fees “current charges” and defines them as: “Amounts received from the public for performance of specific services which benefit the person charged and from sale of commodities or services other than utilities and liquor stores.”19 Examples include the sale of postage by the U.S. Postal Service; fees from turnpikes, ferries, bridges, public campgrounds and public parking facilities; and tuition at public institutions of higher education.
The line between taxes and fees is sometimes blurry and courts in many states have tried to create a distinction between the two. To do so, they consider the intent of the governing body that created the tax or fee. If the governing body’s intent was simply to meet the needs of a person who paid for a service or product, the payment is probably a fee rather than a tax. However, if the intent was to raise revenues to benefit the community at large, then the payment is a tax. The lottery clearly falls into the latter category since legislators create lotteries to raise money for projects that benefit the community at large—not to provide a good to the public that cannot easily be supplied in the private market.
Courts also ask, How is the revenue from the tax or fee used? Is it put into a special fund that covers regulatory or other costs of providing the good or service in question? In that case, the payment is probably a fee. Or is it put into the state’s general fund, as lottery revenue often is? If so, and if the amount of revenue generated is more than the amount needed to provide the good or service, and if the revenue is used to fund unrelated government activities, courts are likely to consider it a tax rather than a fee. The National Conference of State Legislatures, in its Guidelines for User Charges, states that user fees “should cover the cost of the services provided. They should not be used to generate excess revenues that are diverted to unrelated programs or services.”20 Using these criteria, we can see that lottery profits clearly do not constitute a user fee.
In Fiscal Year 2003, after lottery agencies paid winners their prizes, the majority of the remaining money—called the “takeout”—was transferred to state coffers. Operating costs accounted for only 27 percent of the takeout; the rest was kept by state governments as “profit” and used to fund unrelated projects. If all of the takeout were put back into the operation of the lottery, with no “profit,” then we could consider the revenue to be a fee rather than a tax. However, the whole point of the lottery has always been to raise money for unrelated public projects21 such as education, roads and parks, or simply for states’ general funds.
Clearly, then, lottery revenue is not a type of user fee. Does that necessarily mean it is tax revenue rather than miscellaneous revenue, as the Census Bureau claims? There is no reason to put any type of revenue in a miscellaneous catch-all category if it would fit better in a more clearly defined category. Lottery profits certainly fit the Census Bureau’s definition of a tax, which is: “Compulsory contributions exacted by a government for public purposes, other than for employee and employer assessments and contributions to finance retirement and social insurance trust systems and for special assessments to pay capital improvements.“22 There is nothing in this definition that excludes lottery revenue. One example the Census Bureau provides in the tax category is pari-mutuel sales tax. Pari-mutuels include horse racing, dog racing and jai-alai, and there is no reason lottery proceeds could not be treated the same way the pari-mutuel gambling proceeds are treated; it would simply require legislators who are wiling to call the lottery profit what it is—a tax.
Another way to examine the question of whether the lottery is a tax is to compare lottery sales and alcohol sales in Alcoholic Beverage Control states, in which the sale of alcohol is limited to and regulated by the government. In control states, the government raises revenue (which it considers non-tax “profits”) from the operation of liquor stores, as well as from license taxes and alcohol excise taxes. The Census Bureau and state revenue departments do not consider profits from the sale of alcohol to be tax revenue,23 but they do classify alcohol excise taxes as tax revenue. Tax-wise, what, really, is the difference between lottery profits and alcohol excise taxes in control states, other than bureaucratic semantics?
With both lottery tickets and alcohol, the consumer purchases a product from the government and pays a price above and beyond what it costs the government to provide the product, with the “profit” going to the state. In both cases the government legalized a previously illegal product, granted itself a monopoly on the sale of that product and collects revenue from the sale of the product. A private vendor could easily sell lottery tickets and, in many states, private vendors do sell alcohol. With both lotteries and alcohol, the government has granted itself a monopoly on the sale of a product that does not need to be sold by the government. There is nothing inherent in lottery tickets—or alcohol—that mandates it be sold exclusively by the government and taxed heavily, with profits going to unrelated government programs.24 The state of Maine clearly recognizes the similarity between these products: there is a Maine Bureau of Alcoholic Beverages and Lottery Operations, as well as a Liquor and Lottery Commission.
Because the government has a monopoly on both products, the total price can be set at any amount the state chooses, and the rate can easily be raised or lowered to raise revenue or even to discourage consumption. In the case of lotteries, the takeout rates are increased to raise revenue since the state does not want to discourage lottery consumption, although excise taxes on any form of gambling could in theory be raised to discourage participation. If the rationale of the heavy implicit lottery tax were simply to discourage consumption (as is sometimes said to be the case with tobacco taxes), then the government would not advertise the lottery. The National Gambling Impact Study Commission recognized the similarity between lotteries and products like alcohol and tobacco when it correctly referred to the lottery as a “sin tax.” 25
We have established that the lottery is a tax, but why does it matter whether it’s a tax? Lottery supporters counter that, regardless of the tax implications, lotteries are a voluntary activity in which the majority of consumers wish to participate, and, as such, should be allowed to continue. This argument may be plausible when applied to private gambling, but it simply doesn’t work for state-run gambling.
Suppose the state outlawed the sale of bread at private grocery stores and started selling bread itself at $20 a loaf? What if the state started selling and advertising cigarettes to raise money for public schools? The purchase of bread and cigarettes would still be voluntary, but people would be outraged. The voluntary nature of state-run lotteries does not absolve them from the same scrutiny to which we subject other government activities. Lotteries are a government enterprise and a source of tax revenue, and must be evaluated as such.
So, let’s turn to tax policy. What exactly is a good tax?
First, a good tax is simple, easy to understand and easy to comply with. It’s certainly easy enough for a lottery player to comply with the tax by purchasing a ticket, but the administrative burden of operating lotteries makes the tax system overall more complex and less efficient. The problem is compounded in states that use lottery revenue specifically to lower other taxes. The Wisconsin Lottery Web site states that, since 1988, $2 billion—or 32% of revenue—has been returned to eligible Wisconsin taxpayers in the form of property tax credits. In addition, when retailer commissions, prizes, and property tax relief are added together, at least 95% of total lottery revenue has “gone back to the people of Wisconsin.” The Web site boasts further, “That money stays in Wisconsin's economy for the good of everyone.” Is that really something worth bragging about—that the money stays in the state’s economy? A private business could keep that revenue in the state’s economy just as easily. A state-run gambling enterprise is not necessary to keep money in the state. The only thing the Wisconsin lottery really accomplishes is a redistribution of some of the tax burden, shifting it from property owners to lottery players. Perhaps some people think homeowners as a group are more deserving of a tax break than gamblers are, but the tax code should not be used to impose such moral judgments.
The second test we can use to judge taxes is regressivity. A regressive tax is one that is paid disproportionately by low-income people. Lottery supporters often misunderstand—perhaps on purpose—the meaning of regressivity. If a low-income person pays more tax than an upper-income person, as a percentage of his income, then the tax is regressive. Some lottery agencies and associations claim to study regressivity when in fact they focus simply on rates of participation—how many people in each income bracket play the lottery, and how often, which has nothing to do with regressivity. They cite studies showing that the poor spend less money on lotteries than the middle class and wealthy do, but this also is not a true measure of regressivity since it does not take into account money spent as a percentage of income. They claim that regressivity is an issue only with regard to taxes, not with voluntary activities like lotteries,26 and they are content to rely on meaningless statistics that show, the poor play the lottery less often than the middle or upper class in certain states, or that the majority of a state’s lottery consumers are middle-income. What good are these figures if we don’t know how much people spend—and how much they earn? $500 worth of lottery tickets in one year may be a drop in the bucket to an upper-income person, but it is a significant portion of a poor person’s income.
Extensive evidence shows that lotteries are regressive.27 The National Gambling Impact Study Commission determined that, during the time period studied, not only were lotteries regressive, but the poor spent more as a dollar amount.28 Other studies show that the poor do not spend more as a dollar amount, but they do spend more as a percentage of income. The level of regressivity seems to vary depending on the type of game and location, but there is a consensus among most researchers that state lotteries are indeed regressive.
There are also concerns that states actively market the lottery to the poor. In 1986, the Illinois Lottery placed a billboard in a poor Chicago neighborhood that read, "From Washington Street to Easy Street." Even though similar billboards were placed around the city with different street names, angry Chicagoans boycotted the lottery, alleging it was taking advantage of the poor. Some even claim the billboard read, “Your Ticket out of Here,” although the Illinois Lottery denies that. Regardless of the exact wording, there was widespread sentiment that the lottery was taking advantage of the state’s most vulnerable citizens.
Lottery proponents counter the regressivity claim by arguing that the poor spend a disproportionate amount of their income on other consumer goods as well, but this argument fails to take into account that, unlike other consumer goods, lottery tickets are sold and advertised by the government. Should the government be in the business of selling, marketing and profiting from an item on which the poor spend—albeit voluntarily—a higher percentage of their income?
The third tax policy concern with lotteries is that they are not economically neutral. One principle of sound tax policy is that the tax system should not favor the consumption of one good over another, or distort consumer spending. Neutrality, in a broad sense, means fairness: the tax code should treat all goods and services the same, and, since tax revenue pays for general public services, it is important that taxes be levied as broadly as possible rather than on a subset of the population who happen to enjoy a particular product or service. Singling out one product for a high tax rate is economically inefficient. Consumers will likely shift away from that product and find less highly taxed substitutes. This distortion of consumer behavior ultimately damages the economy.
Other types of commercial gambling are taxed, but the high government “profit” rate on lotteries makes the payout rate (the percentage of spending returned to players as prizes) lower than in other forms of gambling. The higher the government’s cut, the lower the payout and the higher the implicit tax.
Politicians, reluctant to make the politically unpopular move of raising income or sales taxes, reason that voters will be more accepting of a high tax on a recreational activity like gambling--especially one that many people consider immoral or unhealthy. The same rationale applies to other types of “sin taxes,” like high excise taxes on alcohol and cigarettes, but how much of a sin do legislators really believe the lottery to be, given how heavily they advertise it?
Because the lottery tax is built into the price of the ticket, the rate is not easy to calculate and varies from year to year and state to state, but it's always staggering. In Fiscal Year 2003, 31 cents of every dollar spent on lottery tickets was kept by state governments. This translates to an implicit tax rate of 45%—far higher than any state’s sales tax and higher than the effective tax rate on most states’ casinos.29 In 2003, ten states had implicit lottery tax rates over 50%, and the average resident of a lottery state paid $55 in lottery tax.
Another principle of sound tax policy is transparency. The tax system should be as clear and simple as possible; taxpayers should understand what is being taxed and what the tax rates are. Transparency, simply put, is honesty. The policymakers who write tax laws should not attempt to hide the cost of taxes from the public.
The lottery tax, however, is a hidden tax. The state creates a monopoly for itself and builds the tax into the price of the tickets, then advertises the lottery as a recreational activity rather than a revenue-raising activity. The government never has to admit that the money it keeps is tax revenue. Lottery agencies are willing to provide consumers with information on the breakdown of profits, prizes and administrative costs, but they do not call the profit “tax revenue.” This simple choice of words is what creates the lack of transparency. Minnesota does consider part of its ticket sales to be an “in-lieu-of-sales tax” of 6.5 percent.30 But this simply creates more confusion, implying that the rest of the government’s share is not tax revenue.
Lottery agencies can raise or lower the implicit tax rate in numerous ways: by introducing new games, by changing the percentage of the ticket price that ends up in state coffers, by increasing the ticket price, or by introducing an entirely new type of product, like video lottery terminals. Ticket prices are set based on the amount of revenue desired, not on a market price.
Politicians often try to create a false dichotomy between lotteries and taxes. They talk about having a lottery to keep taxes low or to avoid a tax increase, not realizing—or refusing to acknowledge—that the lottery is a tax. In the recent North Carolina lottery battle, the governor proclaimed, “It’s either going to have to be lottery, a lottery for education, or it’s going to have to be a tax.” 31 And the following headline from a lottery trade publication says it all: “Governor Lays it on the Line, Choose Tax or Lottery.”32 States could make the implicit lottery tax explicit simply by acknowledging that it is a tax, and by requiting lottery vendors to give customers receipts clearly itemizing the tax portion of the ticket price. Or lotteries could be run privately and taxed by state governments the same way casinos and pari-mutuels are.
There are several reasons legislators and lottery officials would rather not label the lottery a tax. It would be politically unpopular: a legislator who wants to create a lottery would be “raising taxes” rather than “raising money for education.” They want the money to spend, without having to admit to raising taxes. The lottery lets politicians have their cake and eat it too. Legislators often give the impression that a lottery will cause money for education or other worthy causes to simply materialize out of thin air, as if the state will be richer without any citizen being made poorer. Regardless of what we call lottery revenue, it still has to be paid by someone—someone who will have a few dollars less in his pocket afterward.
If lottery proceeds were officially labeled tax revenue by the Census Bureau and by state revenue departments, anti-tax sentiment might stem the tide on the growth of lotteries. State revenue departments would have to take the first step, as the Census Bureau will not classify as a tax any payment that does not pass the “visibility test”: a payment must be visible to the taxpayer in order to be considered a tax, and if a state government does not itemize a payment as a tax, the Census Bureau will not consider it a tax.33 If more people viewed the lottery as a tax, perhaps it would lose some of its luster. Of course, pro-lottery legislators do not want to see this happen, as it would mean they would have to explicitly—and honestly—raise taxes or cut spending.
If the lottery were properly labeled a tax, lottery agencies would also be subject to greater oversight and scrutiny. In many states, they are not bound by the same rules that other government agencies must follow. Some lottery agencies are part of a department of state government, usually the department of revenue. However, the majority of states have established separate agencies, which do not have to abide by all of the rules that govern other state agencies. In a few states, the lottery agency is independent and quasi-public. It is argued that this independence is necessary for the lottery to operate as a business and generate as much revenue as possible. This means, in a few states, paying managers salaries comparable to those in the private sector but not generally allowed for government employees. The Tennessee Lottery recently came under fire for paying its CEO $700,000 in salary and bonuses in her first year on the job—more than the governor makes. She has since received a pay cut, and her compensation is now a more modest $577,000.34
Lottery officials and pro-lottery legislators have a vested interest in maintaining the current system, and, because they have a monopoly on lottery provision, they’re able to. They do not make explicit the true nature of the lottery tax because, quite simply, they do not have to.
The lack of transparency is a problem with traditional lotteries, but video lottery terminals, or VLTs, are even worse. They often resemble casino games so closely that players may not realize the games are run—and taxed—by the state lottery agency. When we think of "playing the lottery," we think of buying a ticket and waiting days to find out whether we won, not playing slots at the track.
Last year New York added racetrack video gaming machines (VGMs) to its lottery line-up. Lottery officials were careful to distinguish their video slot machines from traditional slot machines, which are unconstitutional in New York, although the difference was mainly a technicality. Even after the machines were installed, there was concern that they would be found unconstitutional. In July 2004, a New York appellate court ruled that the machines are lottery games rather than traditional slot machines, and therefore constitutional. However, the fact that the issue had to be resolved in court speaks volumes.
The uses of lottery proceeds pose yet another fiscal policy concern. Most states earmark, or allocate, lottery revenue for specific programs. (The rest simply transfer the proceeds to the state’s general fund.) Proceeds have been earmarked for programs as diverse as parks and recreation, senior citizens programs, salmon restoration, and pension relief funds for police officers and fire fighters, among other things.
The most common program for which revenue is earmarked is education. Twenty-three states earmark all or part of their proceeds for public education, including elementary, secondary, college and vocational education. While lotteries have ostensibly raised a large amount of money for education, it is not clear that the funds are always used for education. Skeptics say that earmarking is at best ineffective and at worst a misleading political tactic to persuade voters to approve lottery referenda and play the lottery. Legislators can simply shuffle funds; lottery revenue allows them to use the money that would have been allocated for education for other purposes. Voters and consumers, however, may be under the impression that lottery funds will significantly increase the total amount of money spent on education.
A number of studies have attempted to prove or disprove the suspicion that earmarked funds are fungible and tend to simply replace rather than supplement education expenditures,35 but of course it is impossible to know how much money would have been spent on education in the absence of lottery funds. However, there is not much evidence that these funds add significantly to the total amount spent on education. In 1999 the National Gambling Impact Study Commission concluded, “When expenditures on the earmarked purpose far exceed the revenues available from the lottery, as is the case with the general education budget, there is no practical way of preventing a legislature from allocating general revenues away from earmarked uses, thus blunting the purpose of the earmarking.” 36
Lottery agencies are aware of the potential for misallocation. South Carolina, for example, states in its 2002 lottery legislation: “[P]roceeds of lottery games must be used to support improvements and enhancements for educational purposes and programs as provided by the General Assembly and . . . the net proceeds must be used to supplement, not supplant, existing resources for educational purposes and programs.” 37 There is, however, no reliable way to enforce this rule.
The Montana legislature is keenly aware of the perils of earmarking. In 1995, after nearly twenty years of earmarking proceeds for education, the state legislature began transferring revenue to the general fund instead. The president of the Montana Education Association stated that it was an “illusion” that lottery funds significantly benefited public schools.38 Georgia, which earmarks funds for its HOPE scholarship program, revealed in 2003 that $1.8 billion worth of lottery proceeds that should have been used for education had instead been spent on other projects, including museums, security fences and metal detectors, and renovations of historic buildings.39
Clearly, the lottery fails the tests of good tax policy. However, no state has abandoned the lottery in the past century and, unfortunately, none is likely to do so soon. If they did, however, they would improve their tax systems by increasing accountability, transparency and economic neutrality, as well as decreasing regressivity. Legislators would find that they do not truly need the tax revenue raised by lotteries; they would either get by without it or raise it through explicit taxation enacted legislatively—and honestly. They could allow lotteries to continue in the private market or even ban them entirely. In either case, the cessation of state-run lotteries would result in more principled state tax systems.
1 Henry Fielding, The Lottery (London: J. Watts, 1732), Scene 1, quoted in Charles T. Clotfelter and Philip J. Cook, Selling Hope: State Lotteries in America (Cambridge, MA: Harvard University Press, 1989), 215.
2 164 lotteries are mentioned in John Samuel Ezell, Fortune's Merry Wheel: The Lottery in America (Cambridge, MA: Harvard University Press, 1960), quoted in Louis Jordan, “Colonial Currency,” Robert H. Gore, Jr. Numismatic Endowment, University of Notre Dame (http://www.coins.nd.edu/ColCurrency/CurrencyText/MA-44descrip.html), and Jordan notes that more have been discovered since 1960.
3 Clotfelter and Cook, Selling Hope, 20.
4 A. R. Spofford, “Lotteries in American History,” Annual Report of the American Historical Association, 1892 (Washington, DC: Government Printing Office, 1893), 174-175, quoted in Clotfelter and Cook, Selling Hope, 35.
5 Spofford, 190, quoted in Clotfelter and Cook, Selling Hope, 37.
7 Clotfelter and Cook, Selling Hope, 43.
8 Ibid, 38.
9 David Weinstein and Lillian Deitch, The Impact of Legalized Gambling: The Socioeconomic Consequences of Lotteries and Off-Track Betting (New York: Praeger, 1974), 14-15, quoted in Clotfelter and Cook, Selling Hope, 142.
10 Public Law 104-169, 104th Congress, available at http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=104_cong_public_laws&docid=f:publ169.104.pdf.
11 See Alicia Hansen, “Internet Lottery Sales: Click Here to Pay Higher Taxes,” Tax Foundation Commentary, March 18, 2005, http://www.taxfoundation.org/legacy/show/409.html; Dennis Cauchon, “Lotteries may gamble on Internet,” USA Today, April 21, 2005, http://www.usatoday.com/news/nation/2005-04-21-lottery-online_x.htm?POE=NEWISVA; and Georgia General Assembly, House Bill 346, http://www.legis.state.ga.us/legis/2005_06/search/hb346.htm.
12 North American Association of State and Provincial Lotteries, “Fiscal Years 2002, 2003 and 2004 lottery sales and profits,” http://www.naspl.org/sales&profits.html . (Puerto Rico’s lottery revenue has been subtracted.)
15 Teresa LaFleur and Bruce LaFleur. LaFleur’s 2003 World Lottery Almanac (Boyds, Maryland: TLF Publications, 2003), 19.
19 U.S. Census Bureau, “Government Finance and Employment Classification Manual,” published December, 2000, http://www.census.gov/govs/www/class_ch7_current.html.
21 The only use of profits that is relevant to the operation of lotteries is the gambling addiction treatment and education programs that some states run, using a small portion of the profits. There is an element of hypocrisy to a state profiting from a product whose use necessitates state-run addiction treatment programs.
22 U.S. Census Bureau, “Government Finance and Employment Classification Manual,” published December, 2000, http://www.census.gov/govs/www/class_ch7.html#S7.23.
23 State-run liquor store profits and lottery profits are misclassified in the same way: Profits from the operation of state-run liquor stores (money that is left over after operating costs and taxes—including excise, sales and license taxes—have been subtracted) go into the general fund or are earmarked for specific programs. (For examples, see http://www.beveragenet.net/sw/2005/0510_stw/0510cs_1.asp and http://www.acca-online.org/acca_mag/countycommissioner/2001/2001-4%20coline.htm.) Liquor store profits end up in the same place and are, for the most part, used in the same way as alcohol excise taxes. Why, then, should they not both be labeled tax revenue? (As with lotteries, some states use a small portion of liquor store profits for addiction education programs; Mississippi even levies an “alcohol abuse tax” on liquor.) See http://www.census.gov/govs/www/class_ch7_utility.html#a90 for classification of state-run liquor store revenue,
24 Many lottery proponents argue that lotteries must be government run to be fair. According to the North American Association of State and Provincial Lotteries, “[T]he public has a right to demand both the security and integrity of lotteries, to ensure that everyone stands an equal chance of winning, that all advertised prizes are in fact paid out, and that the lottery does not resort to unscrupulous business practices.” (http://www.naspl.org/faq.html#lotreg) But why should consumers have a greater right to security and integrity with lotteries than with any other business transaction? With private lotteries, consumers would have legal recourse in the case of fraud, as they do with other businesses, and—as with any business transaction—the buyer must beware. Why should the government guarantee the integrity of gambling, of all things? Lotteries are, after all, a gamble, and gambling in any form involves risk.
25 National Gambling Impact Study Commission Final Report (Washington, DC: Government Printing Office, 1999), ch.2, p. 3, http://govinfo.library.unt.edu/ngisc/reports/2.pdf.
26 See, for example, North American Association of State and Provincial Lotteries, “Demographic Study Highlights,” http://www.naspl.org/demograf.html, and “Who plays lotteries?” http://www.naspl.org/faq.html#whoplays.
27 See, for example, National Gambling Impact Study Commission Final Report, ch.7; Clotfelter and Cook, Selling Hope, 222-230; Mary O. Borg and Paul M. Mason, “The Budgetary Incidence of a Lottery to Support Education,” National Tax Journal 41 (March, 1988): 75-85; Clotfelter and Cook, “State Lotteries,” 12; Christopher Cornwell and David B. Mustard, “The Distributional Impacts of Lottery-Funded Aid: Evidence from Georgia’s Hope Scholarship,” unpublished paper, University of Georgia (2001); and Virginia Lottery, “Who Plays the Lottery,” (1997), quoted in Charles T. Clotfelter, “Do Lotteries Hurt the Poor? Well, Yes and No,” Duke Policy News (Summer 2000) (summary of testimony given to North Carolina House Select Committee on a State Lottery, April 19, 2000).
28 National Gambling Impact Study Commission Final Report, Executive Summary, 14, http://govinfo.library.unt.edu/ngisc/reports/exsum_16-20.pdf.
29 To calculate the implicit tax rate, divide tax revenue (profits) by the value of the ticket (prize money plus administrative costs). The implicit tax rate is the state government’s tax on ticket sales and excludes income taxes subsequently collected on winnings. For a chart of each state’s implicit tax rate, see Alicia Hansen, Tax Foundation Background Paper, No. 46, “Lotteries and State Fiscal Policy” (October 2004), 15.
33 The visibility test: “One important feature of tax revenue is the need to pass a ‘visibility test.’ That is, the tax levy must be visible to the taxpayer as being a tax and not buried under the guise of another revenue. Take, for instance, a tax on utility services provided by the government levying the tax. If the utility bill does not itemize the tax but incorporates it into its user charge rate (therefore being invisible to the customer as a tax), then that so-called 'tax' is reported as a utility revenue for Census Bureau purposes.” U.S. Census Bureau, “Government Finance and Employment Classification Manual,” published December, 2000, http://www.census.gov/govs/www/class_ch7.html#S7.21.
34 Skip Cauthorn, “Lottery adjusts Paul pay,” The (Nashville, Tenn.) City Paper, November 30, 2004, http://www.nashvillecitypaper.com/index.cfm?section=9&screen=news&news_id=37535.
35 See, for example, Charles J. Spindler, “The Lottery and Education: Robbing Peter to Pay Paul?” Public Budgeting and Finance 15 (Fall 1995): 54-62, quoted in Charles T. Clotfelter, Philip J. Cook, Julie A. Edell and Marian Moore, “State Lotteries at the Turn of the Century: Report to the National Gambling Impact Study Commission” (Washington, DC: Government Printing Office, 1999), 6; Noel D. Campbell, “Do Lottery Funds Increase Educational Expenditure?: Evidence from Georgia’s Lottery for Education,” Journal of Education Finance 28 (Winter 2003): 383-402; Donald E. Miller and Patrick A. Pierce, “Lotteries for Education: Windfall or Hoax?” State and Local Government Review 29 (1997): 34-42; William N. Evans and Ping Zhang, “The Impact of Earmarked Lottery Revenue on State Educational Expenditures,” working paper, University of Maryland (November 2002); and Clotfelter and Cook, Selling Hope, 166.
36 Clotfelter and Cook, “State Lotteries,” 6.
37 South Carolina Education Lottery, “How Education Wins,” May 13, 2004, http://www.sceducationlottery.com/HowEducationWins.asp.
38 Evans and Zhang, 30.
39 James Salzer, “Special projects shrink lottery proceeds,” The Atlanta Journal-Constitution, November 11, 2003, http://www.ajc.com/metro/content/metro/hope/11hopewaste.html.
Join the Tax Foundation's fight for sound tax policy Go
The official weblog of the Tax Foundation.
For information on your state, select it from the drop-down menu.
Contact information for Tax Foundation policy staff Ask
Tens of thousands of Hungarians rallied to protest a proposal to create a tax on internet usage in Hungary. In response to the first protest on Sunday, the Hungarian government limited its proposal.
In the past three decades, the importance of “pass-through” businesses has grown substantially. The combined net income of sole proprietors, LLCs, Partnerships, and S corporations has increased fivefold and now accounts...