As a professor and gambling researcher here in the Oz of odds, I've done many studies on the gaming industry and its economic impact around the world; as you read this, in fact, I'm on my way to Japan and Macau for some summer research into the casinos there. And so, last year, when Pennsylvania lawmakers began considering whether to allow slot machines at racetracks and other locations around the state, two interested parties sought my advice. Gov. Ed Rendell's office asked me to estimate the volume of state revenues that could be generated by the machines, and a citizens group asked me to analyze the potential wider economic impact.

Like many of their counterparts in Maryland, the District of Columbia and elsewhere around the country, leading Pennsylvania politicians see slot machine gambling as an easy way to pick up dollars for financing programs and balancing budgets. And they're right: The machines do make a lot of money, and they can be taxed. For the Rendell report, I based my calculations -- which I'll explain below -- on a total of 30,000 machines in Pennsylvania. They could reliably be expected to produce $3 billion a year in profits for the machine owners. The legislation provides for these profits to be taxed at a rate of something like 34 percent, easily allowing the state to realize $1 billion in new annual revenues. This impressive number must have been useful to the pro-slots governor, for when the legislation passed, it authorized not 30,000 machines, but more than 60,000. That should mean at least $2 billion in annual revenues. And that represents a lot of reduced property taxes.

For the citizens group's report on wider economic impact, however, I looked at much more important questions: Do the dollars and cents add up to economic sense? Would the introduction of slot machines actually be good or bad for the towns that get them?

Those questions had a very different answer -- and it's an answer that politicians looking to bulk up their budgets ought to take note of.

To understand why, let's follow the money.

Begin with this estimate: Each slot machine -- and I'm using the term to mean all kinds of slots, video lottery terminals and similar machines -- can be expected to win approximately $100,000 from players every year. That's how much the machine keeps after it pays out all those winnings and jackpots that successful players never stop talking about. (This is the average for machines in Atlantic City; slots in Illinois and Connecticut actually average net revenues of more than $150,000 each per year.)

That hundred grand is money that comes out of someone's pockets. Whose? Well, in Las Vegas, 90 percent of gambling revenue comes from tourists and visitors. They stay here for an average of four days, spending more money on non-gambling activities -- hotel rooms, restaurant meals, shows, shopping -- than at the tables and machines.

This reliable influx of cash has fueled an amazing economic boom in Vegas, growing every year since the 1950s (with the exception of 2001 and 2002, when fear of terrorism kept more Americans close to home). But it's only possible because the money comes from outside the region; without those out-of-towners, Las Vegas's own population of 1.7 million would support about four or five casinos, and maybe 7,000 or 8,000 slot machines. But we have almost 200 casinos and more than 150,000 machines. We need those visitors.

The situation is very different in Pennsylvania, Maryland and the District, which have more than enough residents -- some 13 million adults -- to occupy all the slot machines that have been proposed for all the venues. No slot owner is going to have much incentive to market outside the area. The mid-Atlantic region could certainly produce another $7 billion to $10 billion in gambling revenues without taking a penny away from what is now being spent in Atlantic City. Furthermore, as more regional venues authorize some form of gambling -- Delaware and West Virginia already have -- there will be less incentive for anyone to travel very far to play the slots.

But every new venue will definitely lure more local folks to gambling. As the research proves, the average American adult spends (that is, loses) $350 each year in games of chance. Those living in casino jurisdictions lose twice that much. Among Nevadans, the average is more than $1,000 a year.

In short, it's likely that the local populations of the three venues will produce virtually every penny of the revenues collected by these machines. And they will divert more of their disposable income, which might otherwise go to other local businesses, to the slots.

That's where the money comes from. Where does that $100,000 a year go? Let's take a hypothetical slot machine in downtown Philadelphia. If all the money taken from the gamblers in Philly stayed there, it would be a wash, and we could leave the debate up to the moralists and the politicians. But it just doesn't work that way.

First, the machine owner had to buy the machine. Typical cost: $15,000. Average market life for each machine: three years. Thus, Philadelphia spends (loses) $5,000 of that $100,000 a year. (Who gains it? Nevada, mostly. Two companies -- IGT of Reno and Bally's of Las Vegas make 80 percent of all the slot and video lottery machines in the United States.)

Machines don't require much maintenance (usually one worker for each 20 machines), but at least those wages will remain in the area. Then there are administrative costs (marketing, accounting, random expenses) and supplies, building costs and utilities. From what I've studied, these costs are likely to total $25,000 a year per machine, and 80 percent of that will be spent locally. That means 20 percent -- or $5,000 -- leaves the area.

The state is taxing the profit, you recall, at a rate of 34 percent. That means $34,000 will go to Harrisburg to fund statewide programs. (This being government, the revenue flow is more complicated than that, but this is close enough for a fair estimate.) The Philadelphia area has less than half the state's population, so maybe $15,000 returns, but $19,000 is lost to the Philadelphia economy. Perhaps local taxes of $4,000 will be levied on the facility, and this money will remain in the area.

Add everything up, and it leaves about $32,000 in net profit for the owner of the machine. Does this money stay in Philadelphia? Sorry, most gambling facilities tend to be owned by far-off concerns. A lot of the slots will be at racetracks, for example, and the owners of one Pennsylvania track aren't just out of town, they're out of the country. Let's generously suppose that one-half the ownership will be local. Swoosh, there goes another $16,000 out of the economy.

We haven't yet counted federal taxes, and they will be hefty. Taxes on that $32,000 will send another $5,000 out of the Philadelphia area.

In the end, of the $100,000 that the single slot machine will take from residents of the Philadelphia area each year, $50,000 will leave the regional economy. Given that a basic $18-an-hour job requires about $50,000 a year in wages and fringe benefits, that's like losing one job from the local economy per machine. Now, multiply that by the thousands of slot machines planned for the Philadelphia area alone.

Still, the Pennsylvania legislators chose slots and some in Maryland and the District are still pushing to make the same choice. Why? For the same reason I've seen politicians pass laws like this in venues around the country: Because all they care about is checking off a potential tax-reducing revenue item on their budgets. They'll get it, all right, and they're cynically willing to hurt the regional economy in the process.

Maryland's proposal has been defeated twice; the District is just beginning the debate. So should you want to put a slot-machine complex or two or three in our nation's capital? Add up the numbers. It might make sense if more than 50 percent of the gamblers would be from out of the area, and I don't mean Bethesda. But get real. Certainly the District attracts tourists, but those school buses lined up along the Mall aren't loaded with gamblers, and those young parents pushing baby strollers and leading little tykes into the Air and Space Museum are not your typical Vegas visitor. Out here, our gamblers average 50 years of age, and only 8 percent of them come with children under 21. Washington's tourists are not prospects for machine gambling. Don't even think about it. They want to see your national museums, your hallowed monuments to democracy, your icons of freedom. The slot machine doesn't make that lineup.

On the other hand . . . maybe the Smithsonian could come up with a special exhibition for the District's most infamous gambling establishment -- Edward Pendleton's Palace of Fortune, which operated a couple of blocks from the White House (at 14th and Pennsylvania) from the mid-1830s to 1858. Pendleton used his position to become the leading lobbyist in the antebellum capital. Hundreds of bills -- most of them providing favors for selected citizens -- were passed on his recommendations. He was especially persuasive with congressmen who had rather large debts as a result of their gambling. Also other congressmen could expect a run of good luck at the tables if Pendleton needed their vote on a special bill.

Considering that history, it occurs to me that one side benefit of renewed gambling in the District might accrue to my current home town, Las Vegas. If a casino opened right there in Washington where the politicians make their decisions, the Democratic and Republican parties could quit sending their bagmen -- sorry, fundraisers -- to Las Vegas, wasting airplane fuel and other nonrenewable resources as they entertain potential campaign contributors in the back rooms of our casinos. We can manage quite well without them, and our legitimate, revenue-producing visitors could use the hotel rooms, thank you.

Author's e-mail:

William N. Thompson is a professor of public administration at the University of Nevada Las Vegas and author of "Gambling in America: An Encyclopedia of History, Issues, and Society" (ABC-Clio).

Taking their chances: Gamblers in Edward Pendleton's Palace of Fortune, which operated near the White House in the mid-1800s.