The dubious case for casinos
I got way off my usual research agenda this morning for a business panel on CBC radio. The topic was the economics of casinos, the result of the City of Surrey voting down a new casino proposal. I have often disparagingly compared stock markets to casinos, but in fact I knew relatively little about the actual business of casinos. I don’t even buy lottery tickets. Perhaps it is the economist in me that knows the odds are stacked against me.
Like drugs, when it comes to gambling my preference is that we shine daylight on it, regulate it and tax it, and make sure that we put the funds into mitigating harms. Gambling has long been with us, from poker parties and hockey pools through to lottery tickets and horse racing. But casinos are a form of gambling that has seen a massive expansion across North America in recent years. Back in 1989, there was Atlantic City and Vegas, period. Now they are almost everywhere, with 17 in BC and thousands across the continent.
A big part of the casino bandwagon has been governments seeking revenues to pay for services without having to raise taxes (reminiscent of natural gas or oil sands royalties). This avoids having the conversation we need to have about how to pay through taxes for the services we want. Gambling revenue in general is a bad way of funding our public policy priorities. My grad school prof, Richard Lipsey, in a co-authored paper comments that the biggest addicts are governments, who are in a conflict of interest as both promoters and regulators of gambling.
It is not at all clear that there is a net revenue gain to governments from casinos. True, the BC government now gets more than $1 billion per year in net income through the BC Lottery Corp, but economists note that there will also be lower revenue from sales taxes on other expenditures that were diverted to gambling. Although higher rates of taxation on casinos relative to sales taxes may lead to a positive gain, one economist reckons that:
[C]asino expenditures come at the expense of noncasino expenditures to such a large extent that, despite the high tax rates applied to casino revenues, the reductions in non-casino spending lead to declines in sales tax revenues that are even larger.
Casinos mostly redistribute income from gamblers to wages, profits and taxes – that is the business model. And because a large portion of those revenues come from a smaller percentage of problem gamblers, they are a tax predicated on harm to their (and their family’s) well-being.
There is an interesting similarity between proposals for casinos and proposals for pipelines or LNG plants. As a matter of public policy we need to assess the benefits against the costs. On the benefit side, proponents from the industry or government tend to overstate the gains, while neglecting losses in other parts of the economy and the cost side of the ledger.
There is a Wal-Mart effect here, where income for the casino and new jobs are offset by losses in other sectors, some related (horse racing) and some not (bars, restaurants and music venues). Only if there is a net increase in tourism would there be a net gain, and the social problems are mostly exported when they leave town. In any event, such opportunities have been greatly reduced by beggar-thy-neighbour competition among new casinos and other on-line options.
In addition to cannibalizing other parts of the economy, casinos don’t make for a solid economic development strategy. They don’t improve the live-ability of our cities, as they are big boxes without proper frontages to the street, even lacking windows. They are not consistent with the development of walkable and bikeable communities.
In terms of costs, we know that about 1% of the population are severe problem gamblers, and a wider swath are moderate problem gamblers. In BC, the BC Medical Association estimates that there are 31,000 of the former and 128,000 of the latter. And it is these folks who are responsible for two-thirds to four-fifths of bets. Compounding addictions include alcoholism and drug problems (with cheap drinks the norm at casinos). Beyond the individual there are private costs to families and friends who get roped into financial problems, and employers in the form of lost productivity and embezzlement. Then there are “externalities”, social costs in the form of mental health care, social services, policing and the judicial system. Casinos are also an ideal place for organized crime, in the form of loan sharking and money laundering.
These are harder to measure due to under-reporting, and given the proliferation of casinos in recent years it seems like there is still a lot more research to be done to nail these numbers down. Still, attempts to add these costs together suggest that social costs likely outweigh any economic benefits from casinos.
Seems to me it hardly takes an economist to conclude that a business which operates purely by taking customers’ money in return for no product at all isn’t adding a lot to the economy.
It’s also a vivid demonstration of the poor fit between efficient market theories and the real world. I mean, theoretically, given the economic model and its model of self-interested human behaviour, gambling as an industry cannot exist. Gambling as an activity maybe can, if you posit informal betting between actors with mismatches in information (although the idea of mismatched information itself violates strict neoclassical tenets). But gambling as an industry, with things like slot machines which by definition take in more on average than they give out, is impossible to square with “homo economicus” or anything very close to it.
At best, even if one could say that efficient markets were something like a useful approximation in general, that generalization clearly could not apply to casinos. Oddly, the backers of casinos tend to subscribe to and invoke precisely the kind of economic tenets which the mere existence of casinos falsifies.