Alabama Lottery Proposal Analysis

¶ … benefit analysis of the proposed state lottery for Alabama. Assessing the costs and benefits of this lottery is challenging. The costs are ill-defined, and often lumped in with other gambling costs in general. Yet, lotteries are not the same thing as casino gambling or sports gambling. So there is a lack of hard data available. The benefits are clearer. These take into account the return on money already being spent by Alabamans out-of-state, and the multiplier effect of this new spending. For the state government, the lottery is a clear winner, but for the Alabama economy as a whole, the cost-benefit analysis looks less positive.

The state of Alabama is examining ways in which it can increase its revenues. One of the ideas that has been floated is that of a state-run lottery. In February 2016 a bill was passed that will allow the state to set-up a statewide vote on the issue in November (Cason, 2016). This is not the first time that the state's voters have been asked to consider the lottery idea -- the lottery concept was rejected in a 1999 referendum (Ibid.). Lotteries are utilized throughout the Western world as a mechanism for increasing government revenue. They are popular because the entity that runs the lottery has control over how much prize money is paid out, leaving the rest of the revenue as profit for the state to add to its revenues. If the state can access information from other public lotteries, it will have more than enough data with which to set payouts and prices to a level that is immediately profitable.

As the 1999 vote indicated, however, there is a substantial amount of opposition to the concept of a lottery. Gambling is generally considered to be a sin, or otherwise a moral failing. Thus, there is significant moral opposition to a lottery in the state of Alabama. Moreover, lotteries tend to be drain on the incomes of people, which can affect low income people who buy lottery tickets hoping to escape poverty, but end up deeper in poverty when they fail to win anything. Thus, there are social costs associated with a lottery, and these are sometimes significant. Indeed, lotteries often run contrary to state provisions against gambling, though there are certain differences between casino gambling and lotteries that are sometimes conflated by opponents. Nevertheless, that there are social costs associated with lotteries is a known fact (Rychlak, 1992).

The concept of a cost-benefit analysis is to gather and present information regarding a project. In particular government projects seek to balance a variety of different costs and benefits. Ideally, only projects with a positive net benefit are undertaken. There are numerous challenges associated with conducting a cost-benefit analysis. First, the data is always forward-looking, which means that it is speculative. With a lottery, there is probably a lot of data in existence from other governments, but that data might not be publicly-available. The reality is that a lot of what goes into a cost-benefit analysis is speculative. This means that great care must be undertaken to ensure that the data involved is of the best possible quality; otherwise the CBA will be a garbage-in, garbage-out scenario. In particular to public projects, there are many costs and benefits that are difficult to quantify. Where a corporation might look strictly at cash flows, a government has to take qualitative costs and benefits into account, and it can be difficult to convert these to hard numbers (Boardman et al., 2011). The objective of this particular analysis is not to make the decision, but rather to provide the necessary understanding cost-benefit decisions involved with the Alabama lottery concept for politicians and the citizens of Alabama to make an informed decision on the matter at such time as they are asked to vote on it.

Benefits

The lottery idea has been proposed because it has been demonstrated to deliver an intriguing set of benefits to many other governments, both in the U.S. and around the world. Typically lottery revenue is allocated to a variety of state budget items. However, governments will often frame this money as being for education, or other popular expenditure, as a pragmatic matter of framing the trade-off in such a way as to increase lottery support (Pierce & Miller, 1999). Studies have shown that around 29% of Powerball lottery revenue goes to education, for example (Gandel, 2016). This figure is in line with the experiences of other jurisdictions -- publicly available figures from the Province of Alberta show around one-fifth of lottery funding goes to education, so the relatively low number is not limited to the U.S. (ALF, 2016).

If education is just one category of expenditure, accounting for a bit less than one-third of lottery profits, that raises the question of what the other expenditures are. Lottery funds are used for a variety of other public expenditures, and can cover a wide range of social programs and other government spending. The use of education as a means of justifying lotteries is common in many jurisdictions, and some have speculated that perhaps this is because education is more acceptable to people than general government expenditures, given that many view government as bloated and wasteful (Allen, 1991). Ultimately, lottery funds are a new revenue source, one that allows the state to withhold tax increases, or even lower taxes, while still meeting balancing the state budget (Allen,1991).

Thus, the reality of the trade-off is that the state of Alabama can avoid tax increases, and can also avoid program cuts, while maintaining its budget. The state has chronic issues with respect to funding critical services, and has been looking for new revenue ideas for quite some time (Cason, 2016). Raising taxes is at odds with the ideology of the Republican-led government, so the lottery represents an opportunity for a new revenue source to be introduced to shore up the state's budget without an increase in taxes, and as such it is politically expedient.

There is no particular value to be ascribed to the political benefits of the lottery, however. The only benefits that should be taken into account for this analysis are the revenues that the state will receive. Indeed, the decision on how to spend those revenues is actually incremental to the decision to approve the lottery. The state government is free to use that money however it sees fit. How it sells the idea of the lottery to the public is a separate matter altogether. Given that it is not known how those funds will be used, it is difficult to speculate on the benefits. For example, there are benefits to education if that money is spent on education. But at this point, regardless of rhetoric, it is unclear if education would avoid cuts if the lottery is approved.

This raises the question of whether there are any benefits at all to the lottery. The thinking is this. At present, the people who would spend money on this lottery are Alabamans. This is money from the state, that goes from the people to the state government, to be cycled back into the economy through a variety of different programs and expenditures. There are going to be transaction costs associated with running the lottery, so the funds available to the state will be lower than the funds that are presently in the hands of Alabamans that would be spent on the lottery.

One issue of consideration is that Alabamans probably already gamble, when they go out of state. Even if one assumes that the market for lottery gambling is distinct from the market for other types of gambling (casino, sports betting, racetrack, etc.), most Alabamans live within two hours of a state that allows lotteries (Florida, Georgia or Tennessee). Thus, there is likely to be a net loss to the state with cross-border lottery shopping. Studies have shown that even when a state does offer lottery, there is some loss when other states offer lotteries perceived as superior (Garrett & Marsh, 2002). Thus, Alabama is already losing out on lottery revenue. This is especially noticeable during major lottery events like the $1.5 billion Powerball, when Alabamans traveled across state lines in numbers in order to buy tickets.

The group Tax Foundation estimates that in 2010 states earned $17 billion from lotteries (Hansen, 2010). This estimate is reasonable considering that Garrett & Marsh (2002) came up with a figure of around $11 billion in 2001. Today's number will doubtless be higher, perhaps conservatively inflated to around $20 billion. In the Garret & Marsh study, they found that Kansas lost around $10.4 million dollars of revenue in cross-border lottery shopping, and this was after Kansas institute its own lottery. In Alabama, the figure is likely to be the entire lottery expenditure. For Kansas this averaged around $56.79. Alabama's major cities are close to, but do not abut, lottery states. Therefore, the figure will probably be lower. A reasonable estimate, however, is that the $56.79 figure is still an absolute gain for Alabama, because it is not currently earning any money from revenue. If it is losing $20 per capita, it is losing $96 million dollars already that its citizens are spending in other states on lotteries.

If Alabamans are already spending in the ballpark of $100 million, and they will spend closer to $266 million (based on the Kansas spending figures) once a state lottery is instituted the following are impacts to the state's budget. First, the $266 million figure is gross. The net gain to states from lottery revenues is roughly 30.5% (Garrett & Marsh, 2002), which gives the state a total of $81 million. That is new money for the state. These are conservative figures; the supporters of the bill are using a figure of $300 million when the talk to the media (Cason, 2016).

The latest state budget has appropriations of $15 billion for the general fund (down from $15.5 in FY 2015) and $5.95 billion for the educational trust, most of which flows to K-12 schools. Lottery funds would make a contribution to allowing the state education budget to at least remain stable. More important, the state education trust budget is expected to increase to $6.8 billion in 2017. The state is selling the initiative as being for college scholarships, but last year money was moved from the Education Fund to the General Fund on account of a shortfall in the latter. The optics would be poor of using this money for the General Fund. Instead, new spending on scholarships is proposed, and is certainly being considered to be a viable selling point.

Money from the proposed lottery represents a relatively small portion of the state budget initially, but nevertheless represents found money for the state. It would be, however, just one of many potential revenue streams.

If some of that money was already being spent on lotteries in other states -- 37.5% of it -- then Alabama is simply bringing that money home. This money will have a variety of economic impacts on the state. First, some will go to the state government. Second, some will go back into the economy as winnings and administration. There will be a multiplier effect for this. The multiplier effect will be much lower for the new lottery spending that will result from this initiative. Most lottery players are low income earners, so they are probably already spending on other things in the state anyway. The lottery would represent, at best, a shift in spending, with limited net gain to the state's economy as a whole. But there is still $100 million that would come back into the state economy.

The multiplier effect is related to the marginal propensity to consume. What this means is that new money that enters the economy is spent at a certain rate, this marginal propensity. If $100 million is spent, and the marginal propensity to consume is 0.8, then after the initial $100 million is spent on the lottery, $80 will be re-spent in the economy, and then $64 in the next round (Investopedia, 2016). There are no precise figures for the multiplier effect of consumer spending, but Parker et al. (2013) identified a figure of between 50-90% for new money that enters the economy as consumer spending. Knowing that lotteries are more popular among the poor, and that they already are expected to spend all of their money, the incremental multiplier would be at the low end of this, no more than 50%. All the same, this represents an economic benefit of bringing that $100 million in out-of-state lottery spending back to Alabama of around $150 million on the state's GDP.

If the economic benefit of this lottery is around $150 million each year on the out-of-state lottery spending and $166 million on in-state spending (conservatively assuming no particular multiplier), the economic benefit of the lottery is around $316 million per year.

Costs

The benefits discussed above are net of the lottery's operating costs, most of which remain in state in the form of wages paid to Alabamans to run the lottery and to Alabama residents who win prizes. The costs can generally be divided in a few different categories.. Of particular concern to critics of the lottery plan are the social costs associated with gambling. Alabama already faces some of these costs, to the extent that its residents gamble out-of-state or illegally bet on sporting events. The ethical argument against the lottery is that the state would be officially sanctioning gambling.

More pragmatically, the state will likely increase its social costs, because lotteries will be easier to play for Alabamans. Out-of-state lottery purchases were estimated at the low end because while most Alabamans live within two hours of lottery states, few live immediately adjacent to one. One of the unfortunate things for this analysis is that most of the analysis done on social costs is on the rather generic "gambling," rather than specifically on lotteries, i.e. Collins & Lapsley (2003). Given that lotteries have one of the lower social costs among the different forms of gambling, it would be entirely disingenuous to work with studies that do not isolate lotteries specifically. As Walker & Barnett (1999) note, identifying the social costs of gambling is inherently challenging. Some gambling is addictive behavior, with higher social costs, while most gambling has near zero social costs, and that includes most lottery playing. Moreover, where studies do exist, they are frequently subject to bias -- for example a study on the social costs of gambling in Wisconsin that openly talks about "the evils of gambling" (Thompson, Gazel & Rickman, 1996). One cannot take papers rooted in moral indignation as worthy of evidence-based policy decisions. Morality, as anybody with even a minimal understanding of philosophy knows, is a tricky subject, usually in the eye of the beholder, and having no place in rational econometric analysis.

The first approach that is hypothesized is the cost-of-illness approach, which evaluates the social costs of gambling in the manner of alcohol and drug addiction. Costs include the cost of treatment, education, law enforcement, prevention and research. This approach seems a bit false for lotteries, which are not particularly addictive in the destructive way that, say, slot machines can be. Lotteries are regularly scheduled and most buyers are habitual, but that does not mean that people who play lotteries are apt to self-destructive behavior, or behavior destructive to others.

The economic approach might make sense for this analysis, being that this is a cost-benefit analysis with benefits framed strictly in economic terms. Aggregate wealth is the important aspect of this. This approach has its critics, but if applied here would reflect in the money pulled out of society for lotteries. For example, if someone had a lottery habit so destructive that they missed work, or that it broke their home. Again, such outcomes are more associated with stronger forms of gambling, and lotteries contribute only a small amount of such destructive behavior.

The third approach costs that Walker identified was the public heath approach. This perspective is considered to be the best of the three. This approach is focused on how gambling affects health, families and communities. What is interesting about all three of these approaches is that for the gambling to have significant costs, it must be addictive, and that addiction must lead to destructive outcomes. From a purely economic perspective, the lottery adds money to the Alabama economy that Alabamans are currently spending out of state, and the rest of the money is simply diverted from other activities. Ultimately, there is an offset with that spending -- diverting money to lottery tickets is just part of the economic multiplier because people work in lotteries, lottery ticket sales help retail outlets, people win lotteries and states spend the remaining revenue.

Where there is little research is on the actual negative social costs associated with lotteries. This is back to square one, where other forms of gambling must be excluded because not all forms of gambling are equal with respect to their social costs. Studies on the social costs of gambling usually refer to casino gambling or sports betting, where costs include suicide, divorce, unemployment, organized crime, child abuse and neglect, and whatever governmental costs might be accrued to deal with these issues. The reality, however, is that lotteries almost never bring about these costs. While opponents of the lottery might frame the lottery as gambling equivalent to other forms, that is simply not reality, and there is no point in engaging any sort of slippery slope argument, because future decisions about gambling are not incremental to this decision.

There is some evidence that lottery players are distinct from non-players. We know that they are younger and have less income and education. The heavy players of lotteries have the least amount of income. The heaviest lottery players tend to have addictive tendencies and engage in other forms of gambling (Burns, Gillett, Rubenstein & Gentry, 1990). In that sense, the presence of a state-run lottery does not likely hold much destructive influence as those individuals probably already play lotteries and possibly already have gambling problems. The bigger concern would be the low income heavy players, who frequently fantasize and buy for that purpose, to the detriment of their ability to spend on other needs in life. These are the greatest risk for social costs incremental to this decision.

Rychlak argues that lotteries do spread compulsive gambling. He cites studies showing correlations between the introduction of lotteries and increases in problem gamblers. He speculates that this is the case because many state-run lotteries have adopted the traits of games that breed gambling addiction -- a sense of mastery, excitement and low odds. Lotteries have low odds, of that there is little doubt. The expected value of a given ticket is between 1-3% of its cost, depending on the game and jackpot. The sense of excitement is not really there, but other studies noted that fantasy plays a strong element in problem lottery players, rather than excitement. The sense of mastery is not an element of most lotteries, as the player has no perception of control over his or her own fate. However, as games mimic these traits, they will create more problem gamblers.

Rychlak waffles between isolating lotteries, and conflating them with all gambling. This makes it easier to draw out bigger numbers for the social costs. He finds that a high percentage of compulsive gamblers have played the lottery, but that is no surprise, and is not a causal relationship. Most alcoholics have consumed beer in their life, too. He also notes, however, that there are impacts on the poor, who tend to be the biggest lottery buyers. He points out that lotteries focus advertising on days when, for example, Social Security checks are released. The lottery is clearly regressive, and even when someone is not compulsive as in missing work, they are still diverting money that they need for other things to the lottery. For adults, this is a matter of personal choice. But when state agencies need to fund people's food (food stamps), or when this affects children who then need state funding, or when the lottery player develops a gambling problem and the state funds recovery programs, these are clear social costs associated with lotteries.