Perhaps the major selling point for online gaming proponents is that states can realize a new stream of income via taxation of online gaming profits. Whether such tax revenues will materialize in the amounts projected, and whether online revenues will significantly cannibalize revenue streams from established brick and mortar casinos are open questions.
One of the most important questions facing states as they enter the online gaming fold is the proper tax rate for online gaming revenues. Striking a balance between maximizing tax revenues while encouraging reinvestment of corporate profits into expanded facilities and additional jobs can be difficult, and states will undoubtedly set different gaming tax rates, much as they already have for brick and mortar gaming. For example, Nevada taxes gaming revenues at 6.75% of gross gaming revenue ("GGR")—essentially gaming revenues less winnings paid to players, without adjustment for other expenses—while Pennsylvania imposes a staggering 55% GGR tax rate. Other states have sliding scale GGR tax rates, while many states (including Nevada) enhance GGR tax receipts with various licensing fees.
At the recent World Regulatory Briefing (summaries of Day 1 and Day 2 by Marco Valerio for Online Poker Report), one panel of attorneys spoke on the topic of taxation of online gaming. One gaming media member in attendance Tweeted an interesting assertion from one of the panel members:
Ben Blair is suggesting that the taxes for land based and online gaming have to match as a matter of law. #WrB13
— Casino City Vin (@casinocityvin) October 23, 2013
Ben Blair is a tax attorney with the large international law firm Faegre Baker Daniels. Although Blair's resume is impressive, it appears he erred in this assertion. Of course, in fairness to Blair, this solitary Tweet isolates his comment from the context of the panel discussion (NOTE: See new Update at end of this post for additional context for Blair's comment). However, this Tweet is a good jumping off point for consideration of a significant online gaming taxation issue—Must online gaming be taxed in the same manner and at the same rates as established brick and mortar gaming?
At first glance, the answer seems obvious. An online slot machine or poker game is essentially just a virtual version of a brick and mortar slot machine or poker game. So of course online gaming must be taxed in the same manner as brick and mortar gaming. Right?
Tax law, however, is much more complicated. Congress and state legislatures generally have wide latitude in setting tax policy. [FN1]. Obviously a tax law cannot discriminate based on a protected class—a law taxing women, Hispanics, or Rastafarians at a higher rate would be unconstitutional. Nor can a tax law favor in-state residents over similarly situated out-of-state residents. [FN2].
When a tax law draws distinctions between groups of people to further purely economic policies, however, courts usually show legislatures substantial deference. For example, Congress encourages home ownership through a law permitting deduction of mortgage interest and property taxes while not permitting deduction of rental expenses. Similarly, Congress encourages capital investment through lower tax rates for income from capital gains than income from employment. At a state legislative level, important local industries are often given preferential tax treatment, and specific companies are often granted favorable tax breaks to encourage those companies to move to or remain in a state, or to expand operations within a state. The U.S. Supreme Court has long permitted legislatures broad discretion in using tax laws to advance the legislatures' preferred economic policies, even if the result is a taxation scheme which plays favorites among individuals and companies, so long as the tax policy is not wholly arbitrary:
"The [Equal Protection Clause] imposes no iron rule of equality, prohibiting the flexibility and variety that are appropriate to reasonable schemes of state taxation. The State may impose different specific taxes upon different trades and professions and may vary the rate of excise upon various products. It is not required to resort to close distinctions or to maintain a precise, scientific uniformity with reference to composition, use or value."
Allied Stores of Ohio, Inc. v. Bowers, 358 U.S. 522, 526-27 (1959). [FN3].
B. Riverboats v. Racetracks—Establishing the Contours for Permissible Taxation of Gaming
Turning back to online gaming, could a legislature draw a permissible distinction between online gaming revenue and brick and mortar gaming revenue such that the two types of revenue could be taxed at different rates? Interestingly, the U.S. Supreme Court has already addressed a strikingly similar issue in the context of differential tax treatment for various types of traditional gaming facilities.
Back in 1994, Iowa's gaming industry was in significant economic distress. The state's racetracks were hemorrhaging money, while its riverboat casinos which offered traditional slot machines and table games were facing unexpected competition from newly authorized Illinois riverboats which did not have the same betting and loss limits and excursion requirements as were imposed on Iowa's riverboats. [FN4]. In response, the Iowa legislature enacted sweeping revisions to Iowa's gaming laws, permitting racetracks to offer slot machines while eliminating bet and loss limits for riverboat casinos.
An interesting provision in Iowa's gaming expansion laws related to taxation of gaming revenues. The legislature imposed a different tax rate on slot machine revenues depending on whether the revenue came from a racetrack (36% maximum tax rate) or a riverboat (20% maximum tax rate). The state's racetracks predictably filed suit, claiming the different tax rates for slot machine revenues violated constitutional equal protection guarantees, and placed them at an unfair economic disadvantage to the riverboats.
The legal challenge reached the Iowa supreme court in 2002. Racing Ass'n of Central Iowa v. Fitzgerald, 648 N.W.2d 555 (Iowa 2002) ("RACI-I"). The court noted that the state legislature was entitled to broad discretion in setting tax policy, but could not discriminate on an arbitrary basis. The court's analysis began with a consideration of whether racetracks and riverboats were "similarly situated" within the state's gaming economy; if not, then the legislature could set different tax rates for the different types of gaming businesses:
"[T]he heart of the tax statute is in its disparate treatment of the main activity taking place at both riverboats and racetracks. That is, the essence of the tax is that it treats racetrack slot machines differently than riverboat slot machines. Where the same activity is being taxed at significantly different rates, a mere difference in location is not sufficient to uphold the discriminatory tax.
Both facilities exist for the purpose of operating gambling games. The bulk of both entities' revenue is from slot machines.... If the bulk of their revenue was not based on slot machines, but rather we were called upon to determine whether horse revenue is different from gambling table revenue, our conclusion may be different. However, because most of the entities' revenue is derived from the same source, it is of little consequence that there are other different types of gambling games available at racetracks and riverboats. More importantly, both facilities are authorized to operate within the same gaming industry. Both operate slot machines that are the primary revenue-making vehicle for both facilities. Racetracks and riverboats serve gambling games to the same group of consumers. Both gaming facilities are equally important to the State in terms of revenue production. We conclude racetracks and riverboats are of the same class for purposes of this equal protection challenge. If the taxing statute treats them differently and there is no rational basis for such differential treatment, the statute violates equal protection."
RACI-I, 648 N.W.2d at 559.
The court then turned to the analysis of whether the legislature had a rational basis related to a legitimate policy purpose for taxing gaming revenues at different rates. This is a highly deferential standard of review, and courts almost never invalidate a law on such a "rational basis" review. Nonetheless, the RACI-I court (in a 4-3 split decision) found that "the effect of the tax is contrary to the legislative purpose of promoting agriculture and economic development" and the differential tax rate "frustrates the racetracks' ability to contribute to the overall economy of this state." The majority determined:
"[T]he inescapable conclusion is the differential tax is not rationally related to the main purpose of the legislation or to the intent behind authorizing racetracks to operate in this state. The stated purpose was allegedly to save the racetracks from economic distress. There can be no rational reason for this differential tax, unless the reason for it was to drive the racetracks out of business, thereby helping the riverboat industry. Unless we recognize the desire to discriminately tax one business for the purpose of supporting another similarly situated business as a legitimate government interest, we can find no other basis for upholding this law."
RACI-I, 648 N.W.2d at 561.
The State of Iowa appealed to the U.S. Supreme Court, which granted review. In a remarkably brief unanimous opinion, the Court reversed the Iowa supreme court, finding that the differential tax rate was constitutional under a rational basis review. Fitzgerald v. Racing Ass'n of Central Iowa, 539 U.S. 103 (2003) ("Fitzgerald"). The Court first observed that Iowa's legislature could have rationally enacted its gaming tax scheme to provide economic aid to riverboat casinos:
"[T]he 1994 legislation, seen as a whole, can rationally be understood to do what that court says it seeks to do, namely, advance the racetracks' economic interests. Its grant to the racetracks of authority to operate slot machines should help the racetracks economically to some degree—even if its simultaneous imposition of a tax on slot machine adjusted revenue means that the law provides less help than respondents might like. At least a rational legislator might so believe. And the Constitution grants legislators, not courts, broad authority (within the bounds of rationality) to decide whom they wish to help with their tax laws and how much help those laws ought to provide."
Fitzgerald, 538 U.S. at 108. Once the Court determined that the Iowa legislature may have intended to provide economic support to the riverboat casino industry, justifying the differential tax scheme was rather straightforward:
"Once one realizes that not every provision in a law must share a single objective, one has no difficulty finding the necessary rational support for the 20 percent/36 percent differential here at issue. That difference, harmful to the racetracks, is helpful to the riverboats, which, as respondents concede, were also facing financial peril. These two characterizations are but opposite sides of the same coin. Each reflects a rational way for a legislator to view the matter. And aside from simply aiding the financial position of the riverboats, the legislators may have wanted to encourage the economic development of river communities or to promote riverboat history, say, by providing incentives for riverboats to remain in the State, rather than relocate to other States. Alternatively, they may have wanted to protect the reliance interests of riverboat operators, whose adjusted slot machine revenue had previously been taxed at the 20 percent rate. All these objectives are rational ones, which lower riverboat tax rates could further and which suffice to uphold the different tax rates."
Fitzgerald, 538 U.S. at 109.
In an incredibly rare turn of events, however, the U.S. Supreme Court did not get the last word on Iowa's gaming tax laws. On remand to the Iowa supreme court, attorneys for the racetrack asked the court to reevaluate the laws under the state constitution's equal protection clause. [FN5]. The court did so, holding that the Iowa state constitutional right to equal protection was broader than the federal right to equal protection (an increasingly common practice known as "judicial federalism"). Under this broader state equal protection right, the court applied an elevated level of scrutiny of the legislature's tax scheme, utilizing what is often referred to as "rational scrutiny with a bite" analysis.
As it did in its first review of the tax scheme, a divided Iowa supreme court found that the scheme of differential tax rates for gaming revenue from racetracks and riverboats was arbitrary and not in furtherance of any legitimate legislative interest. Racing Ass'n of Central Iowa v. Fitzgerald, 675 N.W.2d 1 (2004) ("RACI-II"). In analyzing the statute, the RACI-II court considered and rejected the three major justifications for the tax scheme suggested by the U.S. Supreme Court: a) economic development of river communities; b) reliance interests of established riverboat casinos; and c) assisting the financial position of riverboat casinos.
The RACI-II court spent most of its analysis on the third justification—the different economic positions of the racetracks and riverboats. The court noted that a legislative study had concluded that racetracks might be marginally more profitable than riverboats, and suggested a racetrack tax rate of 24%. Nothing in the record, however, established an economic basis to support a tax rate of 36% for racetracks, a rate 80% higher than for riverboats. "[T]here was no credible factual basis for the legislature to believe that the racetracks would be able to pay nearly twice the amount of taxes as the excursion boats on the same amount of revenue." RACI-II, 675 N.W.2d at 14. In the end, the majority concluded:
"[T]he item taxed—gambling revenue—is identical. The higher tax rate is triggered by the location where such revenues are earned. Yet there is no legitimate purpose supported by fact that justifies treating one gambling enterprise differently than another based on where the gambling takes place, other than an arbitrary decision to favor excursion boats."
RACI-II, 675 N.W.2d at 16.
Despite the RACI-II court's split decision (where the minority—including current Chief Justice Mark Cady—indicated they would defer to the judgment of the legislature in setting tax policy), it is notable that the court did not declare that preferential tax treatment for riverboats would always be unconstitutional. Rather, the legislature had merely failed to establish a record for why a differential tax scheme was appropriate as a matter of policy in that particular situation.
C. Will Online Gaming Face Taxation Hurdles?
Based on the RACI v. Fitzgerald line of cases, it appears that legislatures wanting to tax online gaming revenues at a rate higher than brick and mortar casino revenues could likely do so with only minimal justification. Most state courts will not apply the more demanding analysis of the RACI-II court, and instead will follow the highly deferential approach of the Fitzgerald decision. But even for states where courts may take a deeper look, a few plausible justifications exist for imposing a higher tax rate on online gaming revenues:
- Online gaming has lower overhead and higher profit margins than brick and mortar gaming;
- Brick and mortar casinos create more jobs and have a greater economic and cultural impact than do online gaming sites, and thus should be given preferential tax treatment;
- Because online gaming is more readily accessible, it exacerbates the social costs of gambling substantially more than does brick and mortar gaming, and thus the online gaming industry should pay for ameliorating those social costs (the "click your mouse, lose your house" argument); and,
- Online gaming poses unique law enforcement and compliance risks related to fraud, theft, money laundering, and underage gambling that results in regulatory costs that should be borne by the online gaming industry.
Now, this is not to say that a state legislature will or must impose a higher tax rate on online gaming revenues. In fact, Nevada taxes online gaming at the same rate as brick and mortar gaming. Rather, the point is merely that state legislatures, tempted with a windfall of online gaming tax revenues, can likely find a legally sufficient basis for justifying a higher tax rate on online gaming revenues.
New Jersey is a prime example, taxing online gaming revenues at 15% of GGR, nearly double the 8% of GGR tax rate for brick and mortar casino gaming revenues. However, the primary justification for the New Jersey legislation legalizing online gaming is that it will be a new source of revenue that will not only save the economically troubled Atlantic City casinos, but allow them to compete against new casinos opening in neighboring states. This justification certainly passes the rational basis test set forth by the U.S. Supreme Court in Fitzgerald, and most likely passes the enhanced review described by the Iowa supreme court in RACI-II.
A complicating factor for states like New Jersey—where established brick and mortar casinos are also the leading online gaming vendors—is that even if a higher tax rate is imposed for online gaming revenues, there likely will not be any party with standing willing to challenge the differential tax rate in court. Unlike in the RACI v. Fitzgerald case where the racetracks and riverboats were competitors, all New Jersey online gaming must occur under the auspices of established and licensed brick and mortar casinos in Atlantic City. Considering all of these casinos will be subject to the same tax rates for online gaming revenues, it is hard to imagine any of these casinos proving some kind of discrimination sufficient to establish an equal protection violation in court. And more to the point, when every online gaming operator is also operating a brick and mortar casino (or partnered with one), what casino will want to file a court challenge in the first place?
Even an online gaming affiliate with no brick and mortar operation—say a Party Gaming or PokerStars—would not likely wish to incur the political and legal fallout of a court challenge to an online gaming tax scheme when their corporate goals are primarily focused on gaining access to the American online gaming market, increasing online gaming market share, and expanding into additional states. Rocking the boat with a lawsuit would likely be viewed as counterproductive, particularly when tax costs can be passed along to customers in a variety of ways.
D. Conclusion—Online Poker Players Are Threatened by High Tax Rates for Legalized Online Gaming
There is no question that high tax rates for online gaming could easily destroy online poker as a viable profit-making activity for many winning poker players, most notably through increased rake and reduction or elimination of rakeback or similar promotions. With most courts giving significant deference—in some cases, a rubber stamp—to legislative tax policy decisions, it is clear that the time to fight for reasonable online gaming tax rates is when online gaming is first being considered for legalization in any state. Once tax rates are set and gaming revenues are being allocated by legislatures to other budget priorities, the odds of reducing online gaming tax rates diminishes significantly.
Further, the problem of tax rates for online gaming revenues will only become more acute as states begin to form interstate compacts to expand the liquidity pool for online gaming (a key consideration for poker players). [FN6]. Imagine what will happen when a state like Pennsylvania with a high brick and mortar gaming tax rate (55%) authorizes online gaming and looks to enter into interstate gaming compacts. If Pennsylvania imposes the same 55% tax rate—or an even higher tax rate—on online gaming revenues, it is not too big a leap to assume other states in the same online gaming compact will increase their online gaming tax rates, even if they do not go nearly as high as Pennsylvania.
Of course, it is theoretically possible that a legislature could implement a different, lower tax structure for online poker than for online gaming (this is significantly more likely in states which adopt poker-only online gaming as a first step to full online gaming legalization). Such a poker-specific tax structure likely would pass muster in any court challenge, given the different economic dynamics of poker. But legislatures are highly unlikely to be willing to draw such distinctions, so poker players are likely better off simply advocating for lower tax rates for online gaming in general than trying to carve out special tax treatment for poker (particularly considering poker's lobbying influence is dwarfed by that of gaming industry heavyweights like major corporate and tribal casino interests). [FN7].
Legalization and regulation of online poker is a key step in the development and maturation of the American poker industry. Yet legalization may come at too high a cost for many poker players if appropriate and reasonable taxation is not a priority in the legislative process. As more states look at legalizing online gaming, poker players would be wise to keep an eye on taxation issues.
UPDATE (4 Dec. 2013; 10:15 a.m. CT): Noted tax and gaming law attorney Brad Polizanno who was another panel member with Blair messaged me that the context for Blair's comment was a reference to the Internet Tax Freedom Act (ITFA). Passed by Congress in 1998, the ITFA bans states from imposing taxes on internet access, and also imposes a moratorium on "multiple or discriminatory taxes on electronic commerce". The imposition of differential tax rates on online gaming as compared to brick and mortar gaming may arguably run afoul of the ITFA, which defines a "discriminatory tax" as:
any tax imposed by a State or political subdivision thereof on electronic commerce that--
(a) is not generally imposed and legally collectible by such State or such political subdivision on transactions involving similar property, goods, services, or information accomplished through other means; [or]
(b) is not generally imposed and legally collectible at the same rate by such State or such political subdivision on transactions involving similar property, goods, services, or information accomplished through other means ...
To date, there have not been any significant cases deciding the issue of when goods or services are sufficiently "similar" to invoke the protection of the ITFA. Obviously a book or article of clothing purchased online is similar (if not identical) to the same book or article of clothing purchased in a shopping mall. But when online "goods and services" are merely a virtual representation of a tangible item, the issue of similarity gets legally murky. The analysis likely would turn on how the question is framed. If the issue is whether online slot machines operate the same as physical slot machines, they likely will be considered "similar" for purposes of the ITFA. (Of course, online poker could arguably be considered as a different sort of game than live poker in ways that would not apply to slot machines.) If, however, the issue is framed as whether the online gaming experience is a service "similar" to the brick and mortar gaming experience, then there are distinct differences which might preclude a finding of similarity. The Washington supreme court in its Rousso decision has already rejected an argument that online gaming is the virtual equivalent of brick and mortar casino gaming; the Rousso analysis, however, was in the context of a Dormant Commerce Clause argument, and the court did not consider the ITFA.
Four important points about the ITFA should be kept in mind by poker players as they watch further developments of online gaming legalization. First, New Jersey legislators clearly felt comfortable enacting a major piece of legislation with a differential tax rate for online gaming. Though the New Jersey legislature may not ultimately be correct on the point, the legislature's action suggests that its legal counsel saw no conflict with the ITFA. Second, as noted earlier, it might well be difficult to find a company willing to bring a challenge to the differential tax rate, whether based on an equal protection or an ITFA violation. Third, even if an ITFA challenge is brought and is successful, the legislature could cure the ITFA defect by merely imposing the same, higher tax rate on all gaming revenues. Fourth and finally, the ITFA tax moratorium is a creature of federal policy. Currently, the ITFA tax moratorium is set to expire in November 2014, though Congress is likely to extend the ITFA tax moratorium in some fashion. However, any bill to extend the ITFA tax moratorium would create openings for changes in the law. Given recent litigation by Amazon.com and other online retailers seeking clarity on collection of state sales taxes, and with state governments looking for new streams of revenue, the ITFA tax moratorium will likely generate significant debate by various interest groups. The advent of online gaming and the prospect of multi-state online gaming compacts also makes online gaming taxation a likely point of discussion during the ITFA extension debate. In short, even if the ITFA applies to online gaming in its current form, there is no guarantee it will apply after November 2014.
[FN1] The U.S. Supreme Court has described the broad deference owed by courts to legislatures in the area of tax policy:
"The broad discretion as to classification possessed by a legislature in the field of taxation has long been recognized․ [T]he passage of time has only served to underscore the wisdom of that recognition of the large area of discretion which is needed by a legislature in formulating sound tax policies․ It has been pointed out that in taxation, even more than in other fields, legislatures possess the greatest freedom in classification. Since the members of a legislature necessarily enjoy a familiarity with local conditions which this Court cannot have, the presumption of constitutionality can be overcome only by the most explicit demonstration that a classification is a hostile and oppressive discrimination against particular persons and classes․
No scheme of taxation, whether the tax is imposed on property, income, or purchases of goods and services, has yet been devised which is free of all discriminatory impact. In such a complex arena in which no perfect alternatives exist, the court does well not to impose too rigorous a standard of scrutiny lest all local fiscal schemes become subjects of criticism under [an equal protection analysis]."
San Antonio Indep. Sch. Dist. v. Rodriguez, 411 U.S. 1, 40-41 (1973) (footnotes omitted) (quoting Madden v. Kentucky, 309 U.S. 83, 87-88 (1940)).
[FN2] See, e.g., Metropolitan Life Ins. Co. v Ward, 470 U.S. 869 (1985) (holding that state law taxing domestic insurance companies at a lower rate than out-of-state insurers violated equal protection).
[FN3] “Where the public interest is served one business may be left untaxed and another taxed, in order to promote the one, or to restrict or suppress the other.” Carmichael v. S. Coal & Coke Co., 301 U.S. 495, 512 (1937). "[I]t has repeatedly been held and appears to be entirely settled that a statute which encourages the location within the State of needed and useful industries by exempting them, though not also others, from its taxes is not arbitrary and does not violate [equal protection guarantees]." Allied Stores of Ohio, Inc. v. Bowers, 358 U.S. 522, 528 (1959).
[FN4] As a college student in Iowa in the early and mid-1990s, I remember the days of going to a casino to play blackjack (poker rooms were not permitted at that time) with betting ranges from $1 per hand up to the big bet of $5 per hand. Also, each gambler was given a punchcard which was used by the casino to track daily buy-ins, which were capped at $200 per person. Just an interesting reminder of how the gambling legalization fight is actually rather recent, and the anti-gambling moralists still remain a strong force to be reckoned with in the imminent online gaming political battles.
[FN5] The lead appellate attorneys for the racetracks were Mark McCormick and Edward Mansfield. At the time of the RACI litigation, McCormick was a retired former justice of the Iowa supreme court. Mansfield was a prominent appellate attorney who was subsequently appointed to the Iowa supreme court. Both were members of a large Des Moines law firm started by David Belin, former legal counsel to the Warren Commission that investigated the assassination of President John F. Kennedy (Belin was renowned for his vigorous defense of the Commission's conclusion that Lee Harvey Oswald acted alone). When asking a state supreme court to sidestep an explicit ruling of the U.S. Supreme Court, it doesn't hurt to have a couple of heavyweight attorneys in your corner.
[FN6] Noted tax and gaming law attorney Brad Polizanno recently presented an excellent paper discussing many of the key tax issues facing states looking to enter into online gaming compacts.
[FN7] A good argument could be made that online gaming revenues should have a lower tax rate in that online gaming places less stress on the community (i.e., no traffic or "urban blight" issues, less demand on fire/police resources, and creation of higher skilled, higher compensated jobs) when compared to traditional casinos.