The Economics of Gambling:  Summary Points

Professor Earl L. Grinols, Dept. of Economics, University of Illinois

PA Senate Finance Committee (30 April 2003, 9:00 am, Room 8E-B, East Wing, Main Capitol, Harrisburg, PA)

PA House Finance Committee (30 April 2003, 1:30 p.m., Room 205, Ryan Office Buliding, Harrisburg, PA)


  • ·         Studying economics of gambling since 1990.
    • Ø  Independent research.
    • Ø  Not funded by gambling or anti-gambling organizations.


  • ·         Gambling attracts clientele disproportionately
    • Ø  30% don’t gamble at all; most gamble rarely, minority 10% account for 66-80% of wagers.
    • Ø  30-50 % of revenues derive from problem and pathological gamblers (e.g. 48.2% of gaming machine revenue, Aus. Inst. for Gambling Research, 2001; 37 % Montana keno machines; 1/3 Australia National Productivity Commission study, 1999.)
    • Ø  Convenience casinos, racinos, draw from nearby (over 70% from less than 35 miles)
    • Ø  Bulk of casino revenues are from slot machines.


  • ·         Gambling creates economic costs for society and taxpayers, including non-users.
    • Ø  Crime:  E.g. Aggravated assault, rape, robbery, larceny, burglary, auto theft, embezzlement, fraud.
    • Ø  Business and Employment Costs:  Lost productivity, lost work time, unemployment-related employer costs.
    • Ø  Bankruptcy
    • Ø  Suicide
    • Ø  Illness:  E.g. Stress-related, cardiovascular, anxiety, depression, cognitive disorders.
    • Ø  Social Service Costs:  Treatment, unemployment & other social services.
    • Ø  Direct Regulatory Costs
    • Ø  Family Costs:  E.g. Divorce, separation, child abuse, child neglect, domestic violence.
    • Ø  Abused dollars
    • Ø  (NB  Electronic Gambling Devices typically represent 60-80 % of typical Class III (casino style) revenues.)


  • ·         Gambling fails a cost-benefit test.
    • Ø  Even using conservative cost & benefit estimates, costs to benefits are greater than $3:$1.
    • Ø  Social costs (mid-range) from gambling are approximately $214 per adult annually (of which crime = $63).  Costs of introducing gambling depend on starting base, but typically exceed $130 per adult.
    • Ø  Social benefits (preferred number) are less than $42 per adult.
    • Ø  On a per pathological gambler basis, studies in different parts of the nation conducted since 1994 conservatively estimate costs to be $10,100 per year.


  • ·         Economic Development; Failure of Impact Studies
    • Ø  IMPACT STUDIES ARE NOT COST-BENEFIT STUDIES.  More people working next door to you may have nothing to do with the well being of citizens in your area.  Well being may actually decline.
    • Ø  The value of an additional job has been estimated to be worth as little as zero to the community, or between $0-$1,500.[i]  In a typical county of 100,000 adults the introduction of casinos would create additional social costs of $12.7 m annually and direct social benefits of $4.2 m.  Using $750 as the average value to the rest of the county of a job means that casinos would have to increase the total number of jobs in the county by more than 11,333 to improve well being of residents, an unlikely outcome.
    • Ø  Gambling promoters argue gambling creates regional jobs.  This is sometimes possible, as in the case of an Atlantic City or Las Vegas where the area has effectively converted itself into one large casino and entertainment center that serves primarily tourists.  In general, however, gambling:

—    Loses area jobs when local gambler dollars are removed from the area (when they otherwise would not have been) in the form of taxes or are spent by the casino owners or employees outside the area.

—    Creates area jobs when outside gambler dollars are spent locally by the casino and,

—    Loses net jobs when the first flow is larger than the second.

—    A full accounting of dollar flows, therefore, is needed to determine if gambling will create more jobs than it loses.


  • ·         Crime:  U. Illinois-U. Georgia, Grinols-Mustard Study
    • Ø  7.9%    of total crime (FBI Index I) due to gambling in counties with Class III gambling.
    • Ø  7.7%    of property crime
    • Ø  10.3%  of violent crime
    • Ø  E.g. For an average county with 100,000 population this implies 772 more larcenies, 357 more burglaries, 331 more auto thefts, 12 more rapes, 68 more robberies, and 112 more aggravated assaults.


  • ·         How can we conclude that casinos increase crime?
    • Ø  Casinos have largely been built since 1991, a period when crime has decreased substantially.  Because crime is caused by many factors other than casinos, we want to take those into consideration when we estimate the effect of casinos on crime.
    • Ø  A better comparison is to look at the drop in crime rates for counties that have casinos and for those that don’t.  Crime rates drop much more in noncasino-counties than casino-counties.  The divergence between the two sets of counties is greatest since 1992 after casino expansions.
    • Ø  In many areas such as Florida, casino-county crime rates as a fraction of noncasino-county rates have increased from 2-25% less in 1977 to much higher in 1996. FBI Index I Crimes (-6% to 11%), Violent Crimes (-17% to -2%), Aggravated assault (-8% to 9%), Rape (-9% to 13%), Robbery (-25% to -14%), Murder (-23% to 0%), Property Crimes (-4% to 13%), Larceny (-4% to 13%), Larceny (-2% to 12%), Burglary (-5% to 2%), Auto Theft (-20% to 14%).  All individual crimes switched from negative to positive, except robbery, which still moved in the direction consistent with the rest.
    • Ø  Indian reservations: compact-counties had higher crime rates than noncompact-counties, but the difference was stable until the early 1990s, when the crime rates in compact-counties increased by even more.  By 1996 compact-counties have significantly higher crime rates than non-compact counties.


  • ·         Gambling Taxes are Worse than a Conventional Tax Collecting Identical Revenue
    • Ø  A conventional tax implies social costs per extra dollar collected of  $1.25-$1.45.
    • Ø  Taxes on gambling revenues cost   $2.53  per dollar of tax collected if the tax rate is 50 percent, a number that rises above  $4.82  per dollar if the gambling tax rate is 20 percent.


  • ·         Gambling is a Slippery Slope
    • Ø  What is easy for one state to do, others can do.
    • Ø  Generates a race to the bottom.
    • Ø  End result:  States not gain at others’ expense; all reap social costs.


  • ·         National Gambling Impact Study Commission
    • Ø  ~3 m problem and pathological gamblers, 15 million more at risk (8.5 m potential according to other studies)
    • Ø  ~$5 b costs now, (vs. $20 b – $85 b potential based on other studies:  My research implies ~ $30-$53 b.)
    • Ø  Gambling is like losing the lost output of another 1990-91 recession very decade.


  • ·         For many, gambling merely transfers dollars from one pocket to another without creating a tangible product.
    • Ø  Some gamble for recreation (such gamblers are presumably provided recreation value for their lost dollars), but many gamble to acquire money.  Instead of creating a product or offering a service to earn money, this second group of gamblers doesn’t accomplish anything and yet expects to acquire money.
    • Ø  The more people there are who gamble to acquire money, the poorer society is.
    • Ø  If everyone gambled to acquire his money, we would all starve.

[i] Rappaport, Jordan and Chad Wilkerson, 2001, What are the Benefits of Hosting a Major League Sports Franchise?, Economic Review, Federal Reserve Bank of Kansas City, 86, 1, 55-86.


Check out this study re property values from Massachusetts.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.