Pressing Its Luck

Several MPPs expressed alarm when these changes were first introduced to the legislature, pointing out that it would be impossible to track how lottery money was used. As one MPP remarked when the environment was added to the list of lottery beneficiaries: 

There is no special accountability for the expenditure of the lottery funds so appropriated by the Legislature beyond the usual standing committee on estimates and standing committee on public accounts review after the event. In fact, I am saying that in Bill 114 the vague phrase “for the protection of the environment” is absolutely meaningless.23

These fears proved to be well grounded. The dangers of having “lottery funds go into the big, black hole of the bottomless pit of the consolidated revenue fund,”24 as another MPP put it, were obvious to lawmakers even then: 

What this bill is saying and what this government wants to do is take all the profits of the lotteries and pay them into the consolidated revenue fund, which is that whole pool of money that comes just like every other tax in the province comes and is paid into the consolidated revenue fund. Once it is in there, the cabinet “in such manner . . . may direct to be available [sic] for appropriation by the Legislature.” . . . You do not have to have a paranoid mentality to see what is possible with this bill.25

Indeed, one does not have to be paranoid to point out that slippery slopes are indeed slippery. 

By the time the Ontario Casino Corporation (OCC) was set up in 1993 to run Ontario’s first casino in Windsor, the government had abandoned the “games for games” pretense altogether: casino revenues would be piped directly into the Consolidated Revenue Fund.26

In what is surely one of the most tragic of ironies in Canadian history, Floyd Laughren, the NDP MPP for Nickel Belt who so perspicaciously described the implications of moving gaming revenues into the Consolidated Revenue Fund, was the finance minister who introduced casinos into Ontario. 

With the merging of the OCC and the OLC into the Ontario Lottery and Gaming Corporation (OLGC, later rebranded OLG) in 2000,27 the last connection between gambling money and specific public entertainment projects vanished. Gambling revenues are now pocketed by the government to use however it sees fit.28

This brings us to today’s Ontario, where gambling is treated not as a vice the state should restrain but rather as a source of income the state can—and should—encourage. The province treats its gambling manager as a revenue generator, an identity that OLG has fully embraced. According to the Ontario Ombudsman’s 2007 report on OLG and fraud, “The OLG had become fixated on profit rather than public service. It had come to define itself by its role as a cash cow.”29 Thus OLG’s ongoing modernization plan under the Liberal government, which began in 2010: as Globe and Mail columnist Adam Radwanski commented, “Now, for the first time, it will be purely about profit. With what is somewhat euphemistically described as a ‘more market-driven and consumer-responsive strategy,’ Ontario will try to milk every single dollar there is to be spent.”30

As this history shows, governments of all ideological stripes have participated in creating and have profited from the reality we have today. Addiction has a way of eating away at even the most principled among us. 

History, as it turns out, tends to repeat itself. Consider the following opposition member’s statement on the provincial government’s proposal to move gambling money into the general revenue fund: 

They’ve subverted [the act] by taking the dollars out of a transparent, accountable process and slipping it into the government’s back pocket, where they can begin to distribute money in any way they choose to do so without accountability to an act. Before, when the lottery dollars came in, they would have to demonstrate to the public that they actually did with the money what the lottery act said that they should do with the money. Now, instead, they could decide on any average day to reduce the amount of money put out in the lottery fund by, let’s say, 50 per cent and would not have to be accountable to any kind of an act.31 

Yet this quote comes not from 1989 Ontario, but from 2019 Alberta, which took the same approach to a cash crunch as its counterpart in Ontario had thirty years earlier: siphoning gambling profits into the province’s general coffers.32 Like a tumbler of whiskey placed in front of an alcoholic, a fund full of gambling money is irresistible for cash-strapped provinces. But could this enticing stream of non-tax revenue be too good to be true?

Where the Chips Lie 

Today, OLG is well on track to achieve its stated goal of “generating greater and more sustainable Net Profit to the Province.”33 One need look no further than OLG’s own financial statements to see the fruits of its aggressive new strategy. Ontario’s first year of government-sponsored gambling, 1975, generated a modest $43 million for the province; four decades later, gambling contributes over $2 billion to provincial coffers. 

OLG’s revenue comes from a range of gambling operations. According to its 2018-19 report, the largest share of gambling revenue comes from the lottery, which took in about half of total proceeds at $4.2 billion. The other major revenue source is land-based gaming (the business line that includes slots and casinos—such as Woodbine or Kawartha Downs—and resort casinos such as Fallsview), bringing in another $3.9 billion.34 While these two business lines take in similar shares of revenue, there is a significant difference in their profit margins: the lottery’s net profit to the province is less than $1.2 billion (barely 28% of proceeds), while land-based gaming contributes over $1.6 billion (42% of proceeds) to provincial coffers. 

These games are also growing at very different rates. The revenue after prizes (but before expenses) of land-based gaming has grown substantially since casinos were introduced, shooting from $693 million in 1994 to $3.86 billion in 2019. In contrast, lottery revenue after prize payments has grown far more slowly, boasting only a 21 percent bump over the same period.35 The last, very small, portion of gambling revenue is made up of charitable gambling (2% of proceeds) and online gambling (1% of the corporation’s total proceeds, but with the highest profit margin of OLG’s four business lines at 54%).36

37

This revenue is destined for many pots. OLG’s largest expense is prizes: $2.3 billion, or approximately 28 percent of total gaming proceeds, was paid out to winners across the province last year. Payments to land-based gaming service providers is a distant second at $1.2 billion. Another $694 million went to stakeholders, among which the OLG includes the government of Canada, Ontario First Nations, and municipal commissions. Personnel costs ($572 million) and marketing and promotion ($215 million) are among OLG’s highest indirect expenses. The corporation’s various other operating expenses, such as commissions and fees, ticket printing, facilities, and administration, totaled $940 million. Everything left over after all these bills have been paid—nearly $2.5 billion last year, about a third of total revenue—goes to the province.40

Setting the Hook? Gambling Revenue as Proportion of Government Revenue 

OLG sees its expansion as a success story for Ontario, proudly pointing out that its payments to the province “have increased by $908 million or a record 47.7 per cent” in the past decade.41 Though gambling money has declined somewhat as a share of provincial income over the last fifteen years, it has nonetheless held relatively steady since around 2012, when the modernization process began. OLG contributions have had a powerful impact on the province’s finances.42 In 2005, former premier Bob Rae, the architect of casino legalization in Ontario, confessed: “There is no doubt about it, we have come to rely on gambling revenues. . . . Perhaps in a better world we wouldn’t, but the fact of the matter is it’s here, it’s here to stay.”43

Ontario has become dependent on its yearly $2 billion-plus gambling cheque. OLG contributions made up 1.6 percent of the provincial government’s income in the 2018–19 fiscal year. To put that in context, the $2.4 billion of government payments from OLG just two years earlier matched the $2.4 billion government deficit.44But where is the money coming from? Who, and what communities, are the source of these funds?

P(l)ayer Profile: Demographics

Most of Ontario’s population is paying into the pockets of OLG in some form or another. Researchers have offered different estimates over the years of how many Ontarians gamble, from around two in three (63 to 66%)45 to just over four in five (83%).46 This is roughly consistent with estimates of the Canada-wide average (67-81%).47

Ontario’s population is not paying OLG equally, however. It’s common to hear the state’s use of gambling money described as a “voluntary” tax. We disagree for reasons we outline below, but regardless, this is a tax that disproportionately burdens the poor, the marginalized, and the addicted. 

Money In, Money Out: Gambling by Income

On the surface, the regressive nature of gambling revenue can be hard to see. The data collected by Statistics Canada seem to show that those who have more money are more likely to gamble and to spend more money when they do. Data published in 2011, for instance, found that only 46 percent of households that took home less than $20,000 after tax gambled, compared to three-quarters of those making $60,000 or more. Those in the highest income bracket spent $620 a year on gambling, more than half again as much as the nation’s lowest earners.48 

The Survey of Household Spending (SHS) provides a slightly different set of data—average expenditures on “games of chance” are broken down by income quintile, and data on the proportion of participating households is unavailable—but the numbers seem to tell the same story. Canada’s highest-earning households (averaging $100,000 in after-tax income) spent an average of $240 each year on gambling from 2010 to 2017; this is almost twice as much as those in the lowest-earning category (average after-tax income of $18,900), who spent only $134. The pattern holds for Ontario as well: households making $18,700 spent just $105 on gambling, while those making $103,900 spent more than double at $223.49

But first glances can be deceptive. Higher earners may be spending more of their paycheques at the casino, but gambling eats up a much higher proportion of the poor’s income. According to SHS data, households in Canada’s highest income quintile spend an average of 0.24 percent of their after-tax earnings on games of chance each year; those in the lowest quintile spent nearly three times as much at 0.71 percent. In Ontario, bottom-quintile households spend 0.57 percent of their after-tax income gambling, more than two and a half times as much as the 0.21 percent spent by top-quintile households.50

Less than one percent of a household’s annual earnings may not seem like a lot of money, even for a low-income family. We agree that $134 per year is a modest amount—spending ten bucks a month on entertainment is hardly unreasonable (though, as we’ll discuss below, these SHS figures understate actual spending by a dramatic margin). These seemingly low numbers, however, should not distract us from the high-stakes problem at play: when Ontario collects lottery and casino money, it is digging deeper into the pockets of the poor than of the rich. Gambling may be a “voluntary” tax (more on the accuracy of this description below as well), but it is a tax. And this means the province is paying its bills in a way that hits low-income families hardest. 

And yet this is exactly the opposite of how other tax revenue functions. Our progressive tax system is designed to tax the rich more heavily than the poor: those who have more pay more. The lowest income quintile in Ontario loses barely 2.5 percent of its total income to taxation, while the average household in the province’s highest income quintile turns over to the government over nearly ten times more of its total income at 23 percent.51 (The relative proportion of OLG revenue derived from low-income households may, in fact, be even higher: a recent study found that richer gamblers spent more on gambling activities that were outside OLG’s jurisdiction, such as social gambling and at casinos located outside Ontario.)52

The regressive nature of gambling taxes is no secret. Ontario has known for decades that its collection of lottery revenue “is a regressive tax and that it hurts poor people more than it hurts those people who are better off,” as one MPP put it back in 1989.53 Nor is Ontario an outlier: research both within Canada and internationally has consistently found that poor households spend a greater percentage of their income on gambling than their wealthier neighbours do.54

The Demographics Data Gap 

Statistics Canada’s data on spending and income across the country provide a window into the relationship between gambling and household earnings. But these figures are not without their limitations. As with other addictions, most of us don’t want to admit we have a problem. We gamble more than we think we do. According to the Survey of Household Spending, the average Ontario household spent $158 on games of chance in 2017. Multiply this figure by the number of households in Ontario that year (5.6 million55), though, and the province’s total gambling spending comes to only $885 million—a mere 12 percent of OLG’s reported revenue that year ($7.5 billion56). Put another way, the average Ontario household would have had to gamble more than $1,300—well over eight times more than they reported spending—in 2017 for OLG’s books to balance. 

If households across all income quintiles are underestimating their gambling spending by approximately the same margin in their SHS expenditure records, it would mean that Ontario’s lowest income households are spending more than 4.5 percent of their income on gambling.57 The Ontario government uses income taxes—which are set through a transparent process and for which it is held publicly accountable—to collect an average of $2,747 from households making $10,918 a month, and only $47 from households making $1,567 a month. At the same time, and away from the gaze of the public eye, it takes another $55 of what remains in its poorest citizens’ pockets—an average of $1,519 for the month—but just $82 more ($138) from those who have five and a half times more after-tax income ($8,262) left to pay their monthly bills.

This inconsistency is not uncommon: people consistently underestimate how much they actually spend gambling in self-reported household expenditure surveys, often dramatically.58 This is not to say that these survey respondents are necessarily making a conscious decision to fudge the numbers—it is likely that many people are simply unaware of how much their weekly lottery habit drains from their bank account, particularly since less than half of Canadians track their income and expenses in a budget.59 It is important to acknowledge this limitation, but what we believe to be the core issue at hand remains: gambling disproportionately burdens the poor, a finding that not only is consistent across provinces in SHS data, but has been repeatedly borne out by other research as well. 

Cause for Concern 

Demographics are tough to tease out.60 Finding up-to-date data on gambling in Ontario for this paper was challenging, especially since existing research has tended to focus on problem gambling specifically rather than on gambling participation in general.61 The latest comprehensive national investigation of gambling in Canada was published in 2002 by Statistics Canada. More recent research exists at the provincial level, but various problems including low response rates and small sample sizes, make these data less reliable than desired. While a group of researchers from the Alberta Gambling Research Institute is working on a major study to fill this gap, the project had not yet been completed at time of writing.62

Despite challenges in determining who’s wagering what in Ontario, there’s strong evidence that those at the margins of society are paying disproportionately into the coffers of OLG. Take, for example, Canadian and international studies that establish links between lower levels of education and higher levels of gambling participation.63 A comprehensive examination of demographic trends in Australia’s gambling industry, for example, found “a striking gradient in that losses are much higher for people with lower levels of education.”64A study of Lotto players in Germany likewise emphasized the significance of education: “Education proves to be one of the most important determinants both for participation in the lottery and for the amount of expenditures. . . . The demand for lottery tickets appears to be, first and foremost, a question of education, since the group of Lotto players, which is already overrepresented by lower levels of education, also displays distinctly higher expenditures within these levels.”65 Given that lower levels of education are also linked to lower job earnings,66 the overrepresentation of less-educated groups in the lottery market is likely to amplify the lottery’s regressive effect. Indigenous peoples also have disproportionately lower incomes compared to majority populations and as such are disproportionately affected by the expansion of gambling.67

Problem Gambling 

Research also suggests that Ontario’s gambling addiction is being fed disproportionately by gambling addicts. Those classified as problem gamblers make up an estimated 1–3 percent of Ontario’s total adult population, depending on how problem gambling is defined and measured in a given survey.68 This is more or less consistent with problem gambling rates in Canada and similar jurisdictions (i.e., large Western countries with legalized gambling, such as Australia and the United States), which also hover around 1–2 percent.69 As with other addictions, there is an abundance of evidence suggesting that problem gambling is more likely to afflict society’s vulnerable and marginalized, having been linked to lower income, minority ethnic status, less formal education, homelessness, alcohol abuse, and higher rates of psychiatric disorders.70 Even after controlling for the effect of these mental health risk factors, problem gamblers are more likely than the rest of the population to have attempted or thought about committing suicide.71

Even though problem gamblers represent a tiny fraction of Ontario’s population, studies estimate that the province collects 15–30 percent of its total gambling revenue from this group.78 These figures, while comparable to findings in other regions,79 are contested by the gambling industry.80 An in-depth analysis of the controversy surrounding problem gambler revenue is beyond the scope of this paper. But as gambling scholars Robert Williams and Robert Wood point out, it should come as no surprise that, as in other fields, from business to health care, relatively few high spenders account for an outsized portion of profits.81

Addiction by Design: Machine Gambling in Ontario 

While problem players are disproportionately lucrative for the industry, OLG dedicates only 2.6 percent of its net profit to prevention and treatment of problem gambling in Ontario.82It is interesting—and somewhat ironic—to note that the problem-gambling funding that OLG sends to the Ministry of Health comes from slot machines. The amount of problem gambling funding received by the Ministry of Health each year is “based on government policy that directs 2% of annual OLG slot machine revenue (before service provider fee deductions and excluding slot machine revenue from the Resort Casinos).”83 Yet slot machines and other electronic gambling machines (EGMs) have come under particular scrutiny for their addictive design and association with problem gambling.84 EGMs include a number of features acting to impair players’ rational control, such as losses disguised as wins—where audio and visual effects celebrate a player “winning” an amount less than he or she wagered, even though the player lost money—and near misses, where the display of symbols makes it appear that the player was close to winning even though the outcome of each play is completely random. These structural characteristics manipulate players’ emotional and cognitive perceptions of the game to keep them playing longer and spending more: 

Electronic gambling machines (EGMs) are computers utilising sophisticated techniques, designed to maximise spending and “time on device” per user. EGM designs very successfully employ psychological [principles] to maximise users’ bet sizes and machine usage. These characteristics have the effect of increasing the addictive potential of EGMs.85

There are over 27,500 slot machines in Ontario, and they are extremely lucrative to the province. In every year that the OLG recorded how much money was generated by slots (2000–2001 through 2010–11), these machines generated at least 88 percent of land-based gaming revenue. Given that land-based gaming has been responsible for nearly 60 percent of Ontario’s gambling income for the past two decades, EGMs are responsible for around half of all gambling money going into provincial coffers.86 If half of the province’s gambling tax is collected by machines designed to override players’ conscious control, can this tax really be called “voluntary”? 

OLG proudly touts the merits of its responsible gambling program, assuring players it will “educate you about gambling so you can make informed choices about the games you love to play.”87 But given that modern forms of gambling are deliberately designed to undermine and overcome rational decision-making, how effective is giving players information about the odds or reminding them to play within their limit?

A study conducted in 2011 found that only 20 percent of Ontarians reported having played electronic gambling machines in the past year.88 Yet OLG’s annual report for that year shows that fully 61 percent of its revenue came from slot machines.89 In other words, nearly half the Ontario government’s gambling profits are coming from only one fifth of the population.90 These profits are poured directly into the Consolidated Revenue Fund to be doled out to various ministries at the government’s discretion. OLG money is treated exactly the same way as general tax revenue—even though this “voluntary tax” is not collected equitably, as the data show. So what can be done to lift the burden of gambling from the shoulders of the poor and marginalized? 

Gaming Out a Government Gambling Policy 

Ontario’s government is addicted to gambling. OLG revenue is painted as a form of public funds collected for everyone’s benefit—collected, no less, on a voluntary basis while meeting a widespread demand for entertainment. If the people want to try their luck at the casino over the weekend, the story goes, why shouldn’t our schools and hospitals benefit from the profits? Surely these dollars are better in the hands of government than in the hands of organized crime. 

Yet beneath this rosy spin lies a system that preys on Ontario’s most vulnerable citizens. Through the gambling industry, the state is digging deeper into the pockets of those who have the least, so that it can keep public programs and services for the rest of the province artificially cheap. How, then, can the government be weaned off this unhealthy and unjust dependency? 

Cold Turkey 

One option would simply be to outlaw gambling altogether, returning to the sweeping restrictions of the pre-1969 Criminal Code. This could both cut off the revenue feeding the government’s addiction at the source and eliminate the social problems associated with rapid gambling expansion in one fell swoop. A blanket ban, however, is both unfeasible and irresponsible. Setting OLG’s expansion aside, there may be legitimate reasons to have the state regulate gambling, such as curbing crime and keeping games fair.91This paper does not address that particular question. 

We suggest, however, that the massive economic upheaval the COVID-19 pandemic has created, and its containment measures, represent a unique opportunity for the provincial government to cut its addiction to gambling money cold turkey. At time of writing, the total cost of the outbreak on Ontario’s finances is impossible to predict, but the $2.5 billion that Ontario stands to lose in gambling income is certain to pale in comparison. The province’s finances will need to be rebuilt after the COVID-19 crisis subsides, and this rebuilding should include structures preventing the government from depending on OLG profits to pay its bills. Social-distancing measures have already cut off the province’s flow of gambling money: OLG facilities have been shut down, and the amount of revenue going into public coffers from gambling is likely to be at historic lows.92 COVID-19 provides the province with an unprecedented opportunity to start clean. The costs to the treasury will never be lower. We urge the province not to relapse at this critical time. Relying on OLG profits is an unjust way to pay for (necessary) COVID-19 spending—Ontario’s financial recovery should not be financed by those most vulnerable to economic disaster. Instead, it should focus on building the institutional arrangements that encourage savings and investment. 

Getting Clean 

The key to breaking the province’s cycle of dependency is getting OLG money out of the Consolidated Revenue Fund. Because of the regressive means by which gambling dollars are extracted, these funds must be kept separate from general tax revenues: no government should be using a system that targets the poor more than the rich to finance its policy agenda. Moving gambling revenue out of the consolidated fund, and into a specific fund—preferably aimed at relief of poverty—would be the equivalent of the government’s admitting it has a problem, admitting that it has harmed the public it is intended to protect, and forming the first steps to making direct amends.94

The temptation, of course, will be to free up tax revenue by using OLG profits to pay for existing programs. But using lottery funds instead of HST to pay the bills for Ontario Works, for instance, only means that the government will avoid real recovery. It is imperative that Ontario not try to escape the discomfort of withdrawal by simply moving money around, and keeping it in its possession in the long-term.

(In)voluntary Taxation 

A common objection to this framework is that without the extra cash it gets from casinos and lotteries, the government will have to raise taxes. Why should the state take money from all its citizens when some of them are already supplying this money of their own free will? The problem with this “voluntary” or “free will” tax95 argument is not just that it ignores the evidence that shows it to be inequitable and regressive, but that it fails to account for the ways in which the system is structured to overcome individual players’ wills. 

As discussed above, Ontario collects half of this tax using machines designed to overcome players’ voluntary decisions on how much to spend. The state’s dependence on this revenue source calls into question the “voluntary” nature of how citizens part with their money in lotteries, casinos, and other means of gambling.96 In its role as regulator, operator, and beneficiary of gambling, the province will tend, as history has shown us, to focus on its own financial ends, even at the cost of its other roles. OLG’s marketing spending, which has totaled the equivalent of $7 billion over its forty-five-year history and averaged nearly $322 million a year since 2000, is a case in point.97 It’s not necessarily that the people want to gamble more—we don’t necessarily know. But we know with certainty that the state wants them to.98 

This aggressive promotion strategy is conspicuously inconsistent with the marketing restrictions for other government-controlled “vices.” Advertising regulations for alcohol and tobacco are strictly worded so as to prevent artificial stimulation of demand. A booze commercial is permitted only if it “promotes a general brand or type of liquor and not the consumption of liquor in general.”99 Tobacco products and accessories cannot even be visible to consumers until after they’ve been purchased.100 The government restrains the market for these products because it knows that unhealthy consumption—more of a risk when they’re more easily available—has social costs. As a moneydraining form of entertainment with a potential for devastating addiction, gambling falls into the same category.101 And when one examines the means of gambling this becomes even more true. EGMs, as researcher James Doughney has pointed out, are comparable to cigarettes in the way they facilitate harm to their users: 

The EGM and tobacco industries intend users to consume their products in precisely the ways that directly, and without further mediation, initiate the causal chain that results in known harms. . . . Both industries supply control-impairing products that, used as intended, will cause them to use the product in harmful quantities. Almost all smokers will smoke potentially lethal amounts. The EGM product, used precisely as intended, will cause users to lose control of time and money in sufficient numbers for the industry to flourish.102

Cigarette packages are plastered with labels that urge consumers to stop smoking, often with graphic images.103 Meanwhile, OLG provides 24/7 access to slots and urges Ontarians try them out with messages like “Fun wins every time!”104 But why should the state encourage the kind of fun and excitement that’s premised on making people poorer and risks impoverishing them altogether? 

 105

While we acknowledge the persistence of gambling as a fact, and a regulatory role for the state as a limited good, we do not consider its artificial expansion, nor the use of its proceeds as either a fact, or a good.106 Gambling ads encourage people to think they can win—“The next winner could be you!”— at games they are designed to lose.107 In doing so, the state is intentionally cultivating money-losing habits among its citizenry and neglecting more fundamental, positive, and ultimately growth-oriented economic habits.

Habit Forming 

Once OLG funds have been disentangled from legitimate tax revenue, then, how could they be used to advance the state’s responsibility to administer justice for the most vulnerable? We put forward a few specific policy options below. Ontario could, for example, redistribute gambling money back to the poor directly. The “voluntary tax” would flow in the same net direction as other tax revenue—from those who have plenty to those who don’t have enough. A common objection to transfer policies such as this is the “beer and popcorn” argument: won’t giving gambling money back to the poor just encourage them to keep gambling? The assumption that low-income families use government cheques to sustain irresponsible spending habits is widespread,108 but evidence suggests the opposite is true. A study on child benefit spending in Canada, for instance, showed that low-income families who received unconditional cash transfers spent more on household essentials—such as home meals, transportation, and child care—and less on alcohol, tobacco, and restaurant meals. It turns out that getting money back in chunks may actually be an incentive to spend it wisely.109

A second strategy is incentivizing savings. Families who struggle to make ends meet rarely have much to put away for a rainy day. Without any kind of small savings reserve to act as a buffer, they are particularly vulnerable if an unexpected crisis strikes.110 This in turn could force them to turn to unsavoury options in a cash flow emergency: research has found that a household with less than $500 in the bank is over two and a half times more likely than one with $2,000 to have taken out a payday loan.111

The need to build the wealth of financially insecure Canadians has never been more apparent given the widespread fragility exposed by the coronavirus-induced economic crisis. Sudden layoffs on an unprecedented scale have demonstrated the importance of assets like emergency savings to cover an unexpected loss of income.112 Yet claims for unemployment benefits are spiking into the millions at a time when close to half of Canadians do not have enough assets to cover their living expenses for three months.113

Stacking the Deck: Institutions and Savings Habits 

It’s important to emphasize here that we’re not suggesting that the poor’s lack of savings is due to a lack of discipline or self-control. As behavioural economists know well, we all have cognitive biases and lapses in willpower, which lead us to make irrational decisions. Which of us hasn’t eaten that extra donut when we knew we shouldn’t? Who among us has ever eaten one potato chip? Giving in to the same temptation hurts a low-income household more than a rich one, though, since the indulgent purchase takes up a greater share of the poor household’s income.114

Even more significant is the role of financial and social institutions in shaping economic behaviour. Left to our own devices, most of us don’t save nearly as much as we know we should. But there are a host of structures that aim to support saving: the government promises to stretch our kids’ school savings a little further by contributing to our RESPs (Registered Education Savings Plans115); banks offer personalized plans that encourage us to invest and build up wealth (ours and theirs); the Canadian Pension Plan’s automatic deductions nudge us to put away money for retirement; companies encourage savings by offering matching RRSP contributions as part of employee compensation. Evidence suggests that even the poorest households have a few dollars to spare, which could reasonably be used to start building an emergency savings fund.116 Yet even though almost all Canadians have access to banking services, the poor don’t benefit from tax-encouraged savings tools and products the way that the middle and upper classes—who have higher taxable incomes and thus more to gain from income-tax exemptions, deferrals, and deductions—do.117 This leaves low-income families vulnerable to institutions that discourage savings habits: the gambling industry illustrates the latter category perfectly, urging players to spend their extra dollars instead of putting them away.118

This wealth-draining system is the foundation of the government’s benefitting from gambling. Ontario makes money from casinos and lotteries only because gamblers come out poorer in the long run. A few lucky winners of astronomical sums may come out ahead, but the deck is stacked in favour of the house, and against the commons. Every dollar of OLG’s net profit to the province is a dollar a player has lost: “For government to win, its citizens must lose.”119

Making Sure the Right House Wins: A Broader View of Gambling and the State

It’s hard to see how the government’s encouraging people to gamble more in order to increase its cash flow is good for the citizens it is called to serve, or for its long-term prosperity. A discussion of the intrinsic moral nature of gambling is beyond the scope of this paper—we make no claim as to whether it’s wrong for individuals to gamble, and we do not believe that the authority to make such judgments belongs exclusively with the state. Our concern is instead with how state-run gambling fits in a justly ordered society. The ways in which gambling affects individuals should cause us to examine the broader injustice of a state that relies on a robust gambling industry to pay its bills: “To associate all the evils of gambling with personal choice is to overlook the complicity of a system that needs gamblers in order to flourish.”120

There are, effectively, two schools of thought that support Ontario’s supplementing of its income with gambling revenue, and we can broadly label them as “minimalist” and “activist.” The minimalist group includes those who believe the state should have as small a role as possible in society, getting out of the way so that citizens can be free to make their own choices. Proponents of an activist state, on the other hand, support a larger government, one that takes a more involved role in helping disadvantaged groups (chief among them the poor). Though each of these camps offers its own defense of OLG revenue as a public good, both are inconsistent with how gambling money actually flows in the province. 

Gambling in the Minimalist State: “Free” Choice? 

For minimalists, OLG revenue keeps taxes down through the voluntary decisions of individuals. Private citizens should be free to spend their money however they choose; the state has no business telling them not to buy lottery tickets or play slot machines. If some people want to give more money to the government, all the better for those who’d rather spend their paycheques elsewhere. And yet the province’s insatiable expansion of gambling only gives more money to the state, creating a bigger government—exactly the situation that minimalists want to avoid. 

The increasingly sophisticated “built-in addictive structure”122 of gambling equipment and facilities is another challenge to the minimalist approach, particularly its avid defense of free choice. As Matthew Crawford points out, “The defense of machine gambling by libertarians, as well as the industry’s own portrayal of it as free-spirited gaming, assumes an autonomous subject capable of acting in his own self-interest. On the other hand, the machines and every aspect of the casino environment are deliberately engineered to induce people to play ‘to extinction.’”123 The architects of the gambling experience know exactly which behavioural buttons to press to keep players glued to the game.124 How far should the free will argument go when the natural human response to these situations is impaired control?125

Gambling in the Activist State: Who’s Paying? 

On the face of it, state-run gambling appeals to those seeking an activist state. If Lotto 6/49 tickets bring in more money for the welfare system, the expansion of Ontario’s gambling sector should continue full speed ahead—the province’s schools, hospitals, and social programs need all the resources they can get. But as we’ve shown, using gambling money to fund these programs means that a disproportionate share of the cost is shouldered by the poor and marginalized. The activist’s preferred tax system is progressive, in which the government collects a higher percentage of income from wealthier citizens—who have more to spare—in order to ease the burden on low-income families who need a boost. Since provincial gambling revenue flows in the opposite direction, it’s hard to see how a government addicted to this kind of cash is helping its most vulnerable citizens. 

Habits of the Gambling State 

One of the responsibilities of government is to enable and encourage good habits, including economic ones, and to shape structures that promote the economic well-being of its citizens. Governments cannot simply ignore the underlying issues of economic justice at stake when it comes to its involvement with gambling. Ignoring these questions is certainly no problem for those who stand to benefit from the unbridled expansion of gambling, including the revenue-hungry OLG. For them, as Crawford points out, “our moral squeamishness about being ‘judgmental’ smells like opportunity.”126 

Ontario’s population is immersed in an environment where losing money on casino games and lottery tickets is encouraged as a normal, harmless form of fun. This environment fosters a distinct set of habits, attitudes, and dispositions that have social and economic consequences. When one person buys a lottery ticket or two, we have no problem laughing it off as a waste of money—we all know the odds. Yet we rarely think about what the embrace of gambling habits might be doing to our province as a whole.

Behind the seemingly innocent thrill of risk involved in putting money on the line lies a more insidious get-rich-quick impulse, the seductive appeal of getting something for almost nothing. Ontario, we suggest, should think more seriously about the daydream of easy escape at the heart of gambling marketing. “Dream to the Max,” urges Lotto Max.129 “It’s fun to think about round-the-globe travel, sunny skies, fulfilling wishes and dreams.”130 Or, the infamous Lotto 6/49 jingle: “Just imagine.” That OLG recognizes the power of such visions is an indication that it is aware of the power of imagination in shaping its audience’s habits and vision of the good life.131 But are these the kind of economic desires a government would want to encourage in its citizens? Promoting bad risks and false hopes rather than prudent saving habits and discipline hardly seems responsible for a province with an average household debt of over $150,000 and public debt over $338 billion, draining over $12 billion worth of interest payments from the budget every year.132 We need to keep in mind that the more money Ontario spends trying to sell the big win, the more it strengthens “the illusion of the quick fix, the easy way out of our human dilemmas, the plausibility of soft and painless solutions to profoundly complex problems.”133 Does the government really want to encourage a province of fools?134

Gambling is not the way to financial security, either for individuals or for the province.135 Encouraging Ontarians to gamble more just moves the province’s money around, circulating citizens’ dollars rather than directing them toward value-added production.136 Lotteries and casinos are also a highly inefficient way for the government to generate revenue: recall that Ontario gets only 41 cents of every dollar that leaves gamblers’ pockets.137

As important as it is to recognize that Ontario’s gambling addiction is financially irresponsible, it is far more important to point out the province’s social irresponsibility. These days, it’s common for discussions of government responsibility to focus almost exclusively on growing the economy. But this materialistic emphasis is a fundamental distortion of the proper role of the state in the lives of its citizens: “Contrary to notions current in North American society, the maximizing of national wealth and the raising of our living standards is not the principal responsibility of government. . . . It is the undeniable and irrevocable obligation of governments to rule with justice for all and with charity towards the weak and powerless.”139 A government should promote healthy economic growth, yes, but this should always be subordinate to its defining responsibility to administer justice for the common good.140

Turning Bad Habits into Good: Policy Options 

When the state does its part to advance a positive vision of economic life, it nurtures in its citizens virtues that benefit not just pocketbooks (both private and public) but society as a whole.141 Instead of investing in more casinos or new kinds of lotteries, the government might want to consider how it could get its citizens involved in activities that promote genuine economic growth for all. Contrary to what lottery ads would have Ontarians believe, people shouldn’t hope to get something for nothing. It’s time for government to fulfill its responsibility to turn bad habits into good. But how? Below we offer a series of policy recommendations. 

1. Return OLG Profits to the Poor Through Cash Transfers. 

The first option policy-makers could pursue is simple: give gambling money right back to low-income families. Instead of keeping OLG revenue for itself, the government could divvy up the annual profits and cut cheques to the province’s neediest households: let’s call it the Ontario Gaming Equality Benefit (OGEB). Such a system could be tied to income levels, based on the same sliding scale used for, say, the Canada Workers Benefit.142The province makes these kinds of unconditional transfers already—take Ontario Works or ODSP, for instance—which means that OGEB could be used to enhance existing programs. As with other transfers, OGEB could be designed to help families support their kids, giving a boost those who have to cover child-related expenses. This kind of policy would be straightforward, relatively inexpensive to administer, and certain to be more needed than ever in the aftermath of the COVID-19 economic crisis. But while OGEB is a simple way for Ontario to quit its bad gambling habit, it doesn’t set up any good habits to take its place. And it might also encourage governments to reallocate existing spending in its general expenditures.

2. Use OLG Profits to Incentivize Saving in Dedicated Low-Income Savings Accounts.

Alternatively, OLG profits could be used to incentivize savings by contributing to savings accounts for low-income families. This approach might follow the RESP model: the government tops up the savings account by a certain percentage of the account holder’s contributions. A more ambitious model could mirror the Individual Development Accounts (IDAs) offered throughout the United States, which are savings accounts designed to help low-income households develop assets. In addition to having their contributions matched by a mix of public and private funding, IDA holders typically take financial literacy classes; matched withdrawals can be used only for investments such as post-secondary education, home ownership, or small business start-up expenses.143Ontario could follow the IDA model closely, using OLG proceeds to help the poor build long-term assets (i.e., buying a home or starting a business), or adapt it as needed to help build short-term rainy day funds. 

Administering these kind of state-supported savings accounts—monitoring contributions and withdrawals, providing financial literacy programs, enforcing restrictions, and so on—would be expensive.144 Evidence-based, innovative policy development would be required to make sure that account holders could access their savings only for legitimate emergencies or investments: if restrictions on when account funds can be withdrawn are too lax, the savings account risks turning into another chequing account, particularly given the constrained incomes of the poor. But these challenges do not make policy development unfeasible. Since both models have been in use for several years, policy-makers could draw from an abundance of research and resources if they set up either kind of low-income saving system in Ontario. 

3. Prize-Linked Savings Incentives 

If Ontarians want the excitement of chance and the thrill of winning, why not channel those desires toward money-saving rather than money-losing activities? We offer two models of prize-incentivized saving that have met with notable success in other jurisdictions. 

3a. Allow Financial Institutions to Offer Prize-Linked Savings Products. 

Prize-linked savings (PLS) is a great example of how the province could help make saving fun. In savings accounts that follow the PLS model, account holders forgo some or all of the interest they would normally earn on their savings in exchange for the chance to win a prize. More deposits earn savers more entries in the draw, increasing their chances of winning and so rewarding greater saving. One such program, called Save to Win, has seen great success across the United States since its launch by Michigan credit unions in 2009. With only a small deposit of $25, members can open a twelve-month share certificate and earn entries into drawings for prizes, which usually range from $25 to $5,000. Account holders have saved nearly $200 million through this program over the past ten years.145 Meanwhile, Walmart customers moved more than $2 billion through the PLS program linked to the retailer’s prepaid cards in the program’s first two years of operation.146

Not only have PLS products been popular when introduced, research also suggests that they are particularly appealing to low-income savers at risk of financial shocks147 and to those less likely to use traditional banking institutions.148 Importantly, studies have found evidence that PLS does not act as a substitute for other forms of saving, but attracts new savings and even new savers—earning interest isn’t much of a draw for those who don’t have much to put away.149 A PLS account is cheaper and easier to administer than an IDA, making PLS products attractive to the institutions that offer them as well.150 Given the regulatory barriers to non-government institutions implementing games of chance, however, getting PLS products into the Canadian market could be a challenging process.151 Even if these barriers were removed, making PLS accounts accessible to as many low-income Ontarians as possible would still require widespread buy-in from financial institutions.152

3b. Establish a Lottery Bond Program. 

Another example that Ontario could follow is Britain’s NS&I (National Savings and Investment), which encourages savings by offering the chance to hit the jackpot. Customers can buy prize-linked Premium Bonds from NS&I for £1 each. Instead of earning interest on these bonds directly, buyers have a shot at winning between £25 and £1 million each month. The minimum investment is only £25—manageable even for those with very limited incomes—and every bond number gets a separate entry in the draw, an incentive to save more.153 And even though inflation may erode the real value of bond holders’ money over time, this slight shrinkage is still much better than giving up the money altogether through a standard lottery ticket. Premium Bonds are immensely popular in the UK: 21 million Britons had nearly £80 million ($137 million cad) invested in these bonds at the start of 2019.154

Using prizes to promote bond sales could be an innovative way for the government to raise much-needed cash as it works to recover from the financial damage of COVID-19. Ontario could use OLG profits to cover the costs associated with getting the lottery bond program off the ground. Requiring all lottery retailers to offer lottery bonds alongside standard lotto tickets would help raise awareness of the new product—promoting a good habit should hardly be given less effort than promoting a bad one. While this option would not uproot the reliance on luck that the lottery habituates in its players, it would at least co-opt this impulse for a more positive end. 

4. Increase Funding for Problem Gambling Services from OLG’s Marketing Budget. 

In the most recent fiscal year for which information is available about problem gambling funding, OLG spent $282 million on marketing and promotion expenses, but only $64 million on problem gambling programs.155 This means that OLG is spending four and a half times as much on marketing as it is on gambling addictions. The relationship between gambling advertising and behaviour is not yet well researched, but at least two recent studies have suggested that “advertising prompts more frequent and riskier gambling.”156 We propose that OLG use its advertising line to at least double its problem gambling budget, which would still leave over half its marketing funds untouched. This policy would strengthen problem gambling research, prevention, and treatment initiatives in Ontario and scale back Ontario’s artificial stimulation of the gambling market. 

Appendix: Data and Calculations 

Unless otherwise stated, all OLG financial information was taken from the corporation’s annual reports and financial statements. OLG annual reports from fiscal 2014–15 to present are available on the corporation’s website,157 and reports from 2002–3 are available through the Legislative Assembly of Ontario’s website.158 For all previous years, the financial statements of OLG and its predecessors (OLC, the Ontario Lottery Corporation, and OCC, the Ontario Casino Corporation) were included in the Public Accounts of Ontario, which were also the source of total provincial revenue data. Digital versions of Ontario’s Public Accounts are available at the Government of Ontario website and at the Internet Archive.159 The authors have made every effort to ensure consistency across OLG revenue and expenditure categories over time. However, data may not always be directly comparable between years due to changes in OLG’s financial reporting practices (including, but not limited to, classification of different business lines and revenue recognition policies). Figures have been adjusted for inflation where necessary. 

Demographic Calculations 

The primary source of demographic information was Statistics Canada; data taken from other surveys and secondary research are noted in the footnotes. All income data were taken from Statistics Canada’s Canada Income Survey (CIS).160 Since CIS figures are reported in 2018 constant dollars, all data from other sources needed to be adjusted for inflation where necessary to be accurately comparable. CIS provides average income data for each income decile; the two income concepts of interest for our calculations were adjusted total income and adjusted after-tax income.161 In order to make these data comparable to gambling expenditure data from the Survey of Household Spending (SHS),162 income deciles were converted to quintiles. To calculate what proportion of total income, on average, each quintile pays in income taxes, after-tax income was subtracted from total income and the difference divided by total income (see table 1). 

Average data of annual household gambling expenditure are published in Statistics Canada’s SHS under an expenditure category called “games of chance.”163 Expenditure data can be viewed by province and income quintile, the categories relevant for this paper. While Statistics Canada is a reputable source with rigorous data standards, gambling expenditure data nevertheless have several significant limitations and as such should be treated with caution. Due to changes in data-collection methodology, data were available only for 2010–17 inclusive, and many figures were suppressed as “too unreliable to be published.”164 As such, SHS figures used in the paper represent the average of all available data from 2010 to 2017. For example, average expenditure on games of chance for households in Ontario’s lowest quintile was unavailable for 2012, 2014, 2016, and 2017, so the average of 2010, 2011, 2013, and 2015 was used. 

As discussed above in the body of the paper, total spending on games of chance reported by Statistics Canada is dramatically less than total recorded OLG revenue (see table 2). The difference between reported and actual spending was calculated by comparing SHS data to OLG revenue (before prizes and expenses, since expenditure on games of chance in the SHS is not net of winnings). For each year, average annual household expenditure on games of chance for all quintiles (i.e., for all households) was multiplied by the number of households in Ontario to calculate the total amount of gambling spending reported in the SHS.165 This total was then divided by OLG’s total reported revenue for that year to find reported spending as a percentage of actual spending. This percentage was then used to estimate actual spending for each income quintile, based on the assumption that households in each income quintile underestimate the amount they spend on games of chance by, on average, approximately the same margin. Though SHS data does not provide information on how the gap between estimated and actual spending is distributed across income quintiles, we believe it is reasonable to assume that actual spending does rise with disposable income levels.166 While the authors are unaware of any research examining the effect on income on underreporting gambling expenditures, there is some evidence from studies on underreporting other patterns of risky or harmful behaviour (e.g. drug use) that those with lower incomes are more likely to deny or underreport the extent of their participation in such behaviours.167 More importantly, the magnitude of the pattern does not negate the problem with the pattern itself—that is, the provincial government’s collecting proportionately more revenue from low-income households compared to higher-income households. Estimated actual data should nevertheless be treated with caution. 

Since Statistics Canada collects expenditure data from Canadians only, it is more accurate to exclude from OLG’s total any revenue the corporation reported collecting from American patrons at its border casinos (see table 3).168 This non-Ontario revenue was calculated based on “rated U.S. play”—the estimated percentage of revenue that comes from American patrons at both OLG’s Niagara and Caesars Windsor casinos—as recorded in OLG annual reports. For each of the relevant resort casinos, rated US play was multiplied by the total revenue generated by the casino. OLG’s estimated total revenue from American patrons (i.e., the sum of products calculated for Niagara and Caesars Windsor) was divided by total OLG revenue, providing an estimate of the percentage of OLG revenue from American patrons. Since OLG stopped publishing revenue generated by each resort casino after 2015–16, the average percentage of OLG revenue from American patrons for all years with available data (1.79%, from 2009–10 through 2015–16 inclusive169) was used for 2016–17 and 2017–18. For the years in which rated US play for Niagara was not available (2011–12 to 2015–16 inclusive), the average of available years (4.08%) was used. Our calculations assume all other revenue came from Ontario residents.170

EGM (Slot Machine) Revenue Calculations 

The proportion of OLG revenue generated by slot machines was calculated for fiscal 1994-95 (i.e., the first year of the Ontario Casino Corporation) to present (see table 4). Segmented information in OLG annual reports included the total revenue generated by slot machines for fiscal 2000–2001 through 2010-11 inclusive. For each of these years, slot machine revenue was divided by total land-based gaming revenue; for years in which slot machine revenue data were not published in OLG’s annual reports, the average of all available years (88.35%) was used to estimate the proportion of land-based gaming revenue generated by slots. Land-based gaming’s net profit to Ontario was calculated by adding land-based gaming net income (after prizes and expenses) to win contribution (recorded as an expense, but paid directly to the province: OLG resort casinos and the Great Blue Heron Casino are required to remit 20 percent of gaming revenue to the provincial government). These totals were divided by OLG’s total net profit to Ontario. The proportion of net profit to Ontario from land-based gaming was multiplied by the proportion of land-based gaming generated by slot machines to estimate the proportion of OLG’s net profit to Ontario—that is, Ontario’s total income from gambling— derived from slot machines. 

Data Tables 

Data used for the calculations in this paper (i.e., averages of all available data for years 2010–17 inclusive) are presented below in tables 1 through 4. Unless otherwise stated, all financial figures are presented in 2018 constant dollars. Figures in italics are authors’ estimates based on other calculations.171 

172

173