NATIONAL GAMBLING IMPACT STUDY COMMISSION

CHARLES CLOTFELTER

CHAIRPERSON JAMES: I'd like now to recognize Dr. Clotfelter and again, I apologize, I did not realize that you had arrived when I introduced the panel. But thank you for being here. And to remind everyone, he has coauthored Selling Hope: State Lotteries in America and is a Professor of Economics and Law at Duke University.

Professor Clotfelter.

PROFESSOR CLOTFELTER: Thank you very much. I'm delighted to be here, and I was just barely in the door when you said my name, so I was just about effectively gone.

I am delighted to be here and to be on this panel. Rebecca Paul, who just preceded me, is one of the real giants in this industry, having been the director of three big lotteries, successively and successfully. And I am an economist, as my written statement says, so you have to take what I say with that particular admonition, and also a professor and we have a public policy program named after Terry Sanford, who was governor of North Carolina and also a Senator and president of our university. And that's the main place I hang out, but also economics department and the law school.

I started being interested, as an academic, in lotteries back in about 1972 when I was a graduate student in this state. At that time, Massachusetts was one of only a handful of states that had lotteries. As Rebecca Paul said, New Hampshire started the modern lotteries in 1964. But as she also, I noticed, pointed out lotteries were used in colonial times and every day at Harvard I passed by a dormitory, Howorthy Hall, that has a brass plaque that said it was paid for in part by a lottery way back in colonial times. So the lottery as a form of raising money has a long history here.

I was studying taxes and the financing of government and what struck me about the lottery was that it was different than taxes in that as the proponents said at that time you didn't have to pay that particular tax, and some people called it a painless tax, because it was completely voluntary. Which would be one point that I would make as a teacher to say that the lottery had two parts to it, one is the provision of a service or a good, and the other part is the effective taxation of that. And so we can't look upon the lottery as a tax because it is a service that's being provided by the state with some benefits.

Most of the work that I have done on lotteries, I've done with my colleague Phillip Cook, at Duke. And together we did spend a few years trying to do, what we aimed to do was to have a comprehensive study of lotteries from soup to nuts, and we wanted to cover everything that at least economists could cover. The questions of the psychological affects on players, we don't pretend to be experts on. But we did review a good bit of work and then did some original work ourselves.

And what we concluded and found was that lotteries have been with almost no exception some of the most sufficiently and honestly run agencies of state government. They have been set up by states with different laws, but mainly with the objective of raising revenue for the state. In some cases the revenues have been earmarked as in the case of Georgia. In some cases they have been not earmarked and the go directly into the general fund.

As an academic I would look upon lotteries as being distinctive in three ways. One is that they are surprisingly important. When we used to give seminars to fellow academics, most of them don't play the lottery, they gamble in other ways, they would think the lottery is kind of a little footnote or maybe something rather cute.

But one of our points was that they're really surprisingly important. If you rank the things that states do, and in a GNP sense, the number of dollars spent, it's one of the biggest operations of state government. The biggest would be education, but we go down the line, lotteries are, overall, probably in the top five or six for most states.

If you then restrict yourself to things that the states produce and sell to citizens, it's the second biggest thing. The first biggest thing would be higher education. We produce colleges and universities and education and sell them to citizens, but the second biggest thing would be lotteries. And if you think about things that are visible to citizens, it would have to be one of the biggest items. So that's number one. They're much more important than one might think.

Second, they are distinctive in that they do advertise their product. So what the state does and does prominently is going to be something big. It's like the three hundred pound gorilla, whatever that individual does it's going to make a difference. So the advertising for good or bad is going to be quite prominent.

And the third distinctive thing is the financing of lotteries and here I come back to the two part product. There is a provision and then there's the making money off of the thing. Now we coin the term implicit tax because the profit that comes from running a lottery is not a tax officially, but it has the same effect, the dollars are still green and they go into the state treasury. So we look upon this as implicit tax and that will come up again in some of the things that I wanted to say.

Before making the two points that I wanted to, based on the letter that I got and eliminating the things that I really didn't know much about, I wanted mainly to talk about who plays and why people play. Those are really the two things that I think I can contribute something.

But I did notice on the outline today that there was a topic today that said per capita sales. So let me just say a word in my statement. I do have a spreadsheet at the end, which I made up based on the industry magazine in this industry is International Gaming and Wagering Business, it's an excellent source of information. I've relied on it a great deal in work that I've done.

The spreadsheet has information on the per capita sales in the most recent data period which is 1996. If one takes the total sales and divides by the total number of people that live in the states, the 37 states and the District of Columbia, as Ms. Paul said, one gets a per capita sales of, well a lot. The number I'm looking at excludes these VLTs. And maybe I should just make a professor's point right now that would be obvious to anybody in the industry that these video lottery terminals, you've probably seen them, but if you haven't, they're just a little box that might be located in a bar or a tavern. And one sits down and punches buttons and puts a card or money in and it might be, if it's a poker game then you're picking I want to draw two and you win, you'll find out right away if you win. And one can play a lot of games quickly in there.

These are so different from most of the lottery products that it is apples and oranges to add these numbers together. One probably could also make that statement about Keno as well. So, for thestatistics that I presented, I left out the states with video lottery terminals, because you'll see the ratio of prizes to sales is much higher than they would be in table games in casinos or slot machines. So, if you only look at the non-VLT states, which I think is the most meaningful comparison, the per capita sales in 1996 is $146.

Back in 1975 when I started out looking at this, the comparable number in 1996 dollars was $28 per capita. Now anything that's grown in real terms from $28 per capita to $146 per capita, is growing pretty rapidly. If you look between the period 1975 and 1989 per capita sales grew at an annual rate of eleven percent. And of course the stock market has been doing better than that. But eleven percent used to be a very impressive number. And in a few years it will become impressive again. But eleven percent real is a great growth rate. Between 1989 and now, it has been almost flat, about one percent a year.

So as far as the per capita numbers, I could say, and Ms. Paul also made a good point, that all these per capita numbers need to be looked at a little skeptically because I'm dividing by the number of people that live in the lottery states. In the case of Georgia, she's selling to South Carolinians, and some of my fellow state members in North Carolina are going down to Atlanta and playing. So that the number in Georgia, which is $221 per capita, probably overstates the sales to Georgians. And we just don't know how much. And the same thing would apply to the $146 number because I'm only dividing by the population of 37 states and the District of Columbia. So with that as a caveat the numbers grew a lot and then haven't grown all that much in the last few years, if you take out the VLTs.

I was in Oregon about a month ago and I, these people were telling me how much the VLT sales were and I said you've got to take me to one of these places. So we went across the street to the tavern and there was a woman there and I said do you mind if I look over your shoulder, and she said not at all, but you're going to see me lose a lot of money. And so boom, boom, boom, and I said what happened, oh well, I lost $20 and then she walked from here to that green light to an ATM machine and got another $20 and went back and put it in, and she proceeded to play again. I think she was having a bad day. That's not scientific research, that was just observation.

Let me then address the two questions that I had on my list. First, who plays and this is something that Ms. Paul also addressed. And I might make a comment or two to say how what I'm saying compares. What I'm going to say depends on research that's at least ten years old. So this is quite relevant to this Commission. We did look at the 1976 National Gaming Commission but most of the information that we took for this book we collected from a variety of sources, including surveys in California, data from Massachusetts, there have been a number of studies done by economists.

One kind of study uses information based on geographical reporting unit. And caveats that Rebecca Paul raise are very real, they need to be, you need to be concerned about. If you take data on sales for example, these sales could be attributed to people that came from somewhere else. So if we looked at sales at the downtown locations say at, where did I pass one in the taxicab, near the Parker House, they could be people in Wellesley buying these tickets. So one cannot blithely divide sales by the number of people who live in an area. I assume that the people that have done these studies have taken that into account. Certainly in the case of Cook and myself we did, we were sensitive to that.

In any case, the data are old and a new national gambling study would give us much better information. So I would think that at the very least a study that you all would recommend would allow people to answer questions that you pose to this problem. Who plays, do men still play more than women, what's the relationship to education and that sort of thing. So what I'm going to tell you is the best that I know based on information that is old. And certainly the marketing information that Georgia Lottery and others have would be much better to have. If I had it, then I would be happy to report upon it.

Play is very common, as Rebecca Paul said, I don't know if she said this, a lot of people play. In our coverage a majority of adults had played the lottery, some 60 percent in a California survey. It's also very concentrated, as would be not surprising. In our studies some ten percent of the most active players accounted for half the play. And if, you could probably get that down to say one percent probably accounted for much more than one percent of the play, because it gets very concentrated.

Men do play more than women. Those in the middle years play more then the very young and the very old. Catholics play more then Protestants. Education, there is a fairly monotonic affect, the more education you have the less you play on average. That's again on average, there would be exceptions in all of these cases. Laborers tend to play more then advanced professionals. Income is a very important number and it comes back into discussion of regressivity.

Our data showed no relationship between the dollars played and income. So that is on average people in the middle played, bet as many dollars as those at the bottom or those at the top. It varied a little by game, especially when the jackpot got bigger, the number at the top got up a little bit higher. Because there is no systematic relationship between -- okay this is where I've got to be an economist, you have to drink that extra sip of coffee, if the amount by income does not change then the amount spent as a percentage of income is different.

And again, this is the [Rebecca] Paul and [Bill] Gates question. As one's income gets bigger, then a constant amount of spending which is essentially what it is in the lottery case is going to be smaller for those at the top than those at the bottom. And that's about as sophisticated a mathematical question as I'm going to pose to you.

Related to the who plays question is why do they play. And I could show you some very sophisticated economics articles that look at questions about aversion to risk and why people would buy insurance and why would they also bet at the same time. But economists really haven't answered the question, I mean to most people's satisfaction. You would have to note, as has already been pointed out, that gambling has been with us a long time. The casting of lots is referred to in the Bible, we have certainly seen from Jamestown to Harvard, to Yale we found that many of our institutions have been--

COMMISSIONER WILHELM: Some would argue on not a unmixed blessing.

DR. CLOTFELTER: And I'm always happy to point out in my own state that our dear University of North Carolina was financed in part by lottery revenues.

But the urge and interest in gambling has always been around. And I would almost leave it at that. People enjoy it, we don't know exactly why, certainly the chance of winning is part of it, but there's also a camaraderie involved in that. If you asked yourself how many, at this very moment, how many pools are now operating based on the NCAA Basketball Tournament, it would give you a suggestion that it's alive and well all over the place. So whatever it is, it's been a very popular thing.

While we don't know precisely why people play, we do and can pick up some information. I've got two minutes and I'm going to finish in two minutes. If we want to come back to that, there's a lot of evidence that people think that they can control outcome with things. And so many people would refer to that as superstitious behavior.

Let me then go to the other question, is the implicit tax regressive? Part of the lottery is this implicit tax, the portion that is the implicit tax is regressive, it is decidedly regressive. The definition of regressive, as used by economists, is a tax that hits poor people at a higher rate then rich people. And since the ratio, as I explained, goes down with income, any constant percentage tax or implicit tax also goes down. You can measure this eight ways from Sunday and it will always be regressive. Now, that may not be important to you, but as defined it is certainly regressive.

But I would finish with a point that Rebecca Paul made, that there are also benefits from this. People play this and we shouldn't discount the reason they play as being nothing. It could be that a lottery has as much social importance as buying a cookie or playing a video game. And if people are in many ways the judge of their own best behavior then why shouldn't they also have the right to make this choice. Our impression is that for the vast majority of players it is not a big problem, but for some, yes, it is.

I thank you for the chance to testify.

CHAIRPERSON JAMES: Let me offer this as a suggestion for the Commission. I think there's a lot of information that has been put before us this morning, and a great deal of opportunity for us to ask questions of our panelists. And also for them to engage in a dialogue among themselves. But I thinkthat would be greatly enhanced, our discussion would be, if we took a fifteen minute break right now. And so as a result of that, I'm going to ask us to go into recess for fifteen minutes and then come back for discussion. Thank you.

(Whereupon, at 10:35 a.m.,
there was a short recess.)

(10:54 A.M.)