The Crying of Prop. 1, 2013

The Crying of Prop. 1

Stephen Q. Shafer, M.D., M.A. , M.P.H.

Chairperson, Coalition Against Gambling in New York

 

8561188366_eda5d758cf_mbuckets

 

 

 

 

 

Summary Social costs due to  increased  problem gambling after  “up to seven” new casinos open in New York State will almost certainly exceed revenues from the State’s taxing new casinos.  These costs are real but externalized,  thus easily hidden or denied.  Just a 10% increase in the statewide prevalence of problem gambling would almost wipe out the gains in revenue to the State treasury and create thousands more gambling addicts than “permanent good jobs.”   A  25%  increase would nearly negate the entire sum ($1.2 billion)  targeted for recovery via in-state casinos.  In their quest for revenue without increases in conventional taxes,  state officials implied by silence  that  the number of  new problem gamblers anticipated either cannot be estimated or need not be. It is a nullity, off  the board.

No public policy can be evaluated properly without considering costs. Yet that’s what happened in the legislature and on the campaign trail.  This paper gives a public health physician’s   viewpoint of  the dishonesty in marketing “Proposal 1” right into the polling booth.

Introduction  The victorious campaign to legalize casinos in New York State  played up  hoped-for benefits and  played down  likely costs.  While conceding when pressed that problem gambling is a problem, promoters never  acknowledged  the flip side to making casinos more convenient to New Yorkers.  This step, to become law on January 1,  will (not might, will) create new addicted gamblers and new problem gamblers as well as service current ones.  Costs quite possibly in the hundreds of millions of dollars per year  will extend  to society from this sector.

Neither government nor business interests have made public any consideration whatsoever of   these costs.  The deliberate silence moved me, as a physician trained in public health, to compare these costs to the much-publicized  benefits.  I focused on a narrow question:  will the inflow of casino money to the State Treasury equal or outweigh the costs of the new casinos, externalized to New York’s people?  This is only one type of benefit, only one category of costs.

Analysis   What inflow is expected?    Proposal 1 promoters  have repeatedly said that gamblers from  NY  “spend”  $ 1.2 billion / yr  at  casinos in adjoining states and Canada.  [1]   The basis for this figure is not known to me.  In 2012 patrons from NY left behind at the two Indian casinos in Connecticut $259 million. [2]   Presumably the other billion was left behind in Pennsylvania, New Jersey, and Canada.  If this amount were lost in NYS  instead of elsewhere,  taxing it at the rates in Table 1  would yield to the State treasury the following amounts:.

 

                                                       Abbreviations used in the textNYS     New York StateQSEC   Quantifiable Socio-Economic Costs

 

Table 1.  Division of  $1.2 billion  “spent”  in out-of –NYS  casinos between State Treasury and  casino  ownership if all  $1.2 billion were kept in NYS.

Tax rate     State Treasury                Owners

$ millions               $ millions

20%                   240                            960

25%                   300                            900

30%                   360                            840

35%                   420                            780

40%                   480                            720

45%                   540                            660

At  30%  (reasonable guesstimate for NYS non-tribal casinos with 70% slots 30% other), the State Treasury would gain $360 million.  This is not the only possible benefit to the state, but is the most easily measured and the most talked-about, as in  “educating our children,” “property tax relief.”

Now to costs:   the principal (but far from the only) cost of “up to seven” casinos is the creation of new addicted and problem gamblers through a “distance  effect. ”     For a brief review of the literature, see pages 2-4 of a Dec 2012 paper which looked at the hypothetical scenario of five new casinos in New York City   [3]

The entry of  new gamblers during a specified time (incidence) into a category like “addicted”  is not well  measured by the prevalence (proportion active at a given time) [4].  This is because individuals leave the active prevalence pool over time through recovery, death, imprisonment, totally disabling illness or out-migration.  As hard as it is to measure the  prevalence of problem gambling,  it is far harder to measure incidence. Thus prevalence, fraught with methodological problems,  is the usual benchmark.

To assess costs of new addicted gamblers and problem gamblers we need  head counts and  per-head figures for cost.  The cost figures cited most often are from Earl L. Grinols, Distinguished Professor of Economics at Baylor. His book shows clearly how he came to them. [5]   I call these Quantifiable Socioeconomic Costs  (abbrev. QSEC), though Grinols does not use that term.  In  2012 dollars  QSEC  are $13787 / yr  per pathological (addicted ) gambler; those per problem Gambler are $3600 /yr .  Note well:  QSEC do not include suicide, divorce,  mental anguish, family disruption.  The costs of   these calamities are un-quantifiable; no monetary values can be assigned.  Thus they are even easier to disregard than QSEC.

Regarding the head count: the number of active addicted gamblers in NYS can be estimated by applying to a rounded-off figure of 15 million adults statistics for prevalence among adults of addicted (1.14%) and of problem gamblers (2.8%).  These are not recent but are well-established from a meta-analysis. [6]   In a national sample reported in 2004 [7] the prevalence of pathological and problem gamblers combined was 3.5% .

If at baseline there are 171,000 active addicted gamblers (15 million * 1.14%) and prevalence goes  up  by 5%, NYS  has  at least 9,000 new addicts.  Table 2 shows the number of new cases for a given increment in prevalence,  and the QSEC attached.  Table 3 works the same way for problem gamblers.  Table 4 combines the QSEC  for both types, arrayed by % increase.

Table 2.  Number of  new addicted  gamblers in NYS and Quantifiable Socioeconomic Costs of new addicted gamblers, by increase in prevalence  over baseline.  e.g. 9,000 * $13,787 /yr = $ 124 million/yr

Increase                       New addicted  gamblers   QSEC of increase in $ millions/yr

5%                                             9,000                         124

10%                                         17,000                         234

15%                                         26,000                         358

20%                                         34,000                         469

25%                                         43,000                         593

30%                                         51,000                         703

Table 3.  Number of new problem gamblers in NYS and Quantifiable Socioeconomic Costs of new problem gamblers, by increase in prevalence  over baseline. e.g. 21,000 * $3,600 /yr = $ 76 million/yr

Increase                       New  problem  gamblers   QSEC of increase in $ millions/yr

5%                                           21,000                         76

10%                                         42,000                         151

15%                                         63,000                         227

20%                                         84,000                         302

25%                                       105,000                         378

30%                                       126,000                         454

Table 4.  Quantifiable Socioeconomic Costs of new addicted  gamblers + new problem gamblers by increase in prevalence  over baseline.  e.g. for 5% increase in both,  total is $124 + $76 = $200 million/year.

Increase                       QSEC of increase, in $ millions/yr

5%                                           200

10%                                         385

15%                                         585

20%                                         771

25%                                         971

30%                                       1157

From Table 4 we see that if prevalence of addictive gambling and of problem gambling both  rise by only 10%, the QSEC attached to that rise are more than the $360 million the State would recover by taxing 1.2 billion at 30%.

We must consider, though, that persons who became gamblers because of the convenient casinos (some addicted,  some  problem gamblers, most in neither type) will also lose money there  that can be taxed.  How much might that add to State treasury  revenue?  Would that be enough to “cover” the QSEC springing from new casinos?  To answer that we need an estimate for  losses by type of gambler.

Grinols and Omorov in a 1996 paper [7]  estimated annual losses to casinos by persons living within 35 miles of  Las Vegas or Atlantic City  at $14,200/year (1992 dollars)  per one hundred  persons.    This includes people who do not go to a casino from one year to the next.  Converting to 2012 dollars gives   $ 23, 400 per 100 persons.  This figure was used in my Dec 2012 paper to reckon that losses to casinos by the 8 million residents of Greater New York came to $1.87 billion/year. I assumed that all losses were to casinos outside the reach of NYS taxation.   If  $23, 400 were applied to the entire NYS population it would mean that losses to casinos come to  $3.5 billion /year,  three times higher that of the commonly-cited  estimate of 1.2 billion /year.  Assuming that  $ 1.2 billion for the whole state is correct, a likely explanation is that since no resident of Greater NY  lives within 35 miles of a full-service casino,   the $23,400/100 persons/yr  figure was high; the longer distance lowers willingness to travel then spend.

$1.2 billion lost at out of state casinos by residents of NYS means that the average loss per year is $1.2 billion/15 million, or $8000/100 persons/year.  I took liberties with the famous table in Grinols and Omorow [8] , keeping the ratios of annual loss per gambler between types very much like those in the original but lowering the values so that the annual loss / 100 persons comes out to $8000, not $23,400.  The results in dollars of 1992 are in Table 5.

Table 5.  Hypothetical structure of casino revenues in 1992 dollars, by type of gambler   This table is formatted like the one in Grinols and Omorov [ref 8]  but the input values  in columns 1 and 3 have been altered, producing figures in column 4 very  different from those in the original.  Percentages in column 5 are similar to those in the original.

Prevalence in population

Type of gambler

Annual loss per gambler

Annual loss per 100 adults

Cumulative % of casino gross

1.14%

Addicted

1480

1687

35

2.8%

Problem

250

700

49

6.06%

Heavy bettor

118

715

64

50%

Light bettor

35

1750

100

40%

Non-bettor

0

0

100

100%

All types

 

4852

 

 

Converting  $4852 dollars of 1992 to $8006 dollars of 2012 gives the amount lost last year out of state /100 persons in population. For a population of 15 million the annual out of state loss is reckoned at $1.2 billion.  If new casinos caused no rise in prevalence of problem gambling AND stopped 100% of leakage,  this table could also represent the gamblers’ losses ( ~ = “gaming revenue”)  at casinos in New York after new casinos are at full steam.  Of this $1.2 billion the State Treasury gets  $360 million at 30% tax rate.

Leaving the baseline of  Table 5, consider the changes in  losses at casinos by gamblers after new casinos open in state,  assuming a 10% rise in prevalences of gambling addiction and problem gambling.  In Table 6,  note  new figures in column 1, same figures in column 3.  The annual loss to casinos rises, as does the revenue to the state treasury.

Table 6.  Hypothetical structure of casino revenues in 1992 dollars, by type of gambler, reflecting changes in relative frequencies of type in population due to new casinos    Assumption: compared to baseline there is a 10 % increase in prevalence of all types of gambler and a decrease in proportion of non-bettors from 40% at baseline to 34%.

Prevalence in population

Type of gambler

Annual loss per gambler

Annual loss per 100 adults

Cumulative % of casino gross

1.25%

Addicted

1480

1850

35

3.08%

Problem

250

770

49

6.7%

Heavy bettor

118

791

64

55%

Light bettor

35

1925

100

34%

Non-bettor

0

0

100

100%

All types

 

5336

 

 

 

Converting $5336 to dollars of 2012 gives $8804 / 100 adults/yr .  This, multiplied by 15 million adults,    yields $1.32 billion as the amount that would be lost by gamblers from New York at casinos in  New York after new casinos are built.  In this scenario the non-tribal New York casinos realize from the losses of  new gamblers an extra $120 million  ( = $1.32B – $1.2B) of which the State gets by taxation 30 % , or $ 36 million.

If  the prevalences of addicted gambling and problem gambling rise by 25% with new casinos, not just by 10%, the annual loss to casinos rises in proportion.  See Table 7.

Table 7.  Hypothetical structure of casino revenues in 1992 dollars, by type of gambler, reflecting changes in relative frequencies of type in population due to new casinos   Assumption: compared to baseline there is a 25% increase in prevalence of  all types of gambler  and a decrease in proportion of non-bettors from 40% at baseline to 29%.

Prevalence in population

Type of gambler

Annual loss per gambler

Annual loss per 100 adults

Cumulative % of casino gross

1.43%

Addicted

1480

2116

36

3.5%

Problem

250

875

51

7.8%

Heavy bettor

118

920

64

62.5%

Light bettor

35

2188

100

25%

Non-bettor

0

0

100

100%

All types

 

6099

 

Converting $6099 to dollars of 2012 gives $10063 / 100 adults/yr. This figure, multiplied by  15 million adults,  yields $1.51 billion as the amount that would be lost at New York casinos by New Yorkers.  Casinos take  in $310 million above baseline (=$1.51B – $1.2B) from the new gamblers, of  which  the State Treasury collects 30%  or $93 million above the baseline intake of $360 million.

Discussion The QSEC to New York society associated with generating 17,000 new addicted gamblers and 42,000 new problem gamblers (10% increase in prevalence of both) are  $385 million, very nearly as much as the revenue to the state ($396 M) from taxing the losses by established and new gamblers at its new casinos.    If the State attended to costs, not just to revenues,  it would see this barely breaks even. The QSEC to New York society associated with generating 43,000 new addicted gamblers and 105,000 new problem gamblers (25% increase in prevalence of both)  are  $971 million, more than twice as much  as much as the revenue to the state ($450 million) from taxing the losses by established and new gamblers at its new casinos. If the State paid attention to costs,  it would see a fiasco.  This $971 million quantifiable cost almost equals the total of  $1.2 billion supposedly at stake.

The advertisements run by NY JOBS NOW implied that every dollar of the 1.2 billion that is lost to a casino in New York rather than across a border will benefit New York.    One ostensible benefit,  outweighed by QSEC,  is tax revenue to be disbursed  back to the populace as “aid to education” or  “property tax relief.”  Another,  not touted so loudly,  might be  to keep the rest of the money within the state where  it will go to overhead and profits of businesses with structures in NYS..  Much of that overhead will, we presume, pass to persons now living in the state as wages and as property tax paid to municipalities (if no abatements).  This would be a benefit to the state.  In fairness and transparency, however,  it  must still be weighed against QSEC, which in the 25% increase scenario may equal or exceed it.   That leaves profit.  Is this a benefit to NYS?    The casino buildings  will be in our state.  Where will the profit-takers be?

It  is too early to know if any of the owners will be NY companies.  Front-running candidates as of  Nov 18  include  Foxwoods, a Connecticut company;  Claremont Partners, “ a partnership of  mostly offshore investors based in the  Isle of Man;”  EPR Properties, based in Kansas City;  Empire Resorts, based in Kuala Lumpur; Concord Associates, New York; Muss Development, New York City; and Jeff  Gural, who lives in New York City.  RH Land Development has its New York State location in Rochester.  Traditions at the Glen Resort has one location, in Johnson City.    Vista Hospitality has its American offices in Binghamton. Caesar’s Entertainment, Las Vegas, is reported to be interested after having been dismissed in Massachusetts.  Rumor says the Stockbridge-Munsee Tribe of Wisconsin is interested. A list like this will change week-to-week.

Conclusion It is sad indeed that voters were sold the amendment by being shown none of the debits, only the income.  In place of a cost-benefit analysis,  the electorate got the travesty of a benefit-only analysis.  New York State’s leaders and legislative followers sought  revenue at any cost, as long as the latter was out of sight.

Most of the debit side of the ledger springs from the formation of new gambling addicts and problem gamblers, creating a public health problem never recognized as such.  In a progressive state such as New York has been and should be, this would have been addressed by Health in All Policy.  HiAP is a fairly new concept, not the law of the land but gaining ground in North America (e.g. California and Ontario).  It came out of a 1998 resolution by the World Health Organization .  Basically, it requires that large – scale governmental policy have a health impact assessment before adoption.  New York State deliberately bypassed the responsibility to offer voters a traditional cost-benefit analysis of policy.  Just as deliberately the state passed up the challenge of  applying innovative HiAP.

It is impossible to predict exactly how many new problem gamblers will develop in the “up to seven casinos” future.  It is also impossible  that there will not be more of  them as our state gets more convenient casinos. Legislators hinted at the threat by putting some funding for treatment and prevention into the Upstate Gaming and Economic Development Act passed last June.  Then they withdrew attention.  No one in Albany made any estimate of the scope of the problem, thus zeroing it away.  If it has no size, it’s nothing.

200px-A-dressing_the_White_Queen

“Alice laughed: “There’s no use trying,” she said; “one can’t believe impossible things.”
“I daresay you haven’t had much practice,” said the Queen. “When I was younger, I always did it for half an hour a day. Why, sometimes I’ve believed as many as six impossible things before breakfast.”     Lewis Carroll,   Through the Looking Glass

 

 

Afterword on prevalence Even if the prevalences of  addiction  and  problem gambling could be shown by a crystal ball to be absolutely stable over (say) four years following the introduction  of casinos, that first phase has created  new disordered gamblers.   Every year, some gambling addicts and problem gamblers leave the “active prevalence” pool  through  recovery, death,  imprisonment, disabling illness or out-migration. Suppose this rate is truly 10% a year;  after  four years,  prevalence of active addiction  that began at (say)  1.14% should be down to 0.75%.  If  it is not, “replacement addicts” have entered the pool,   Some are in-migrants, some relapses, many newly-minted.

 

Studies that purport to find no statistically significant increase in the prevalence of (say) problem gambling over time after exposure rises may also be underpowered.  A study to detect with reasonable statistical power a near-doubling of  prevalence from 1.14% to a prevalence of 2.17% would require 2000 interviews at each point.  This is impossibly expensive for a public agency to do.  A study to detect a 25% increase from 0.0114 to 0.0142 would require 20,000 interviews at each point.  The increase seems less than minuscule, yet in a population of 15 million it represents 43,000 new addicts.  No wonder the gambling promoters have no fear of someone’s proving an increase in the prevalence of addiction and thus no fear of being held responsible for any ominous trends.

References

  1. From the Home Page of  NY Jobs Now http://www.nyjobsnow.com/index.php#benefits

NEW JOBS. MONEY FOR SCHOOLS. LOWER PROPERTY TAXES.

New Yorkers currently spend more than $1.2 billion a year at destination casinos in neighboring states. Allowing casinos in New York will keep a lot of that money right here in New York where it belongs — helping to generate economic activity, fund our schools, and provide tax relief.

On the ballot this November, Proposal 1 will ask voters to approve the casino plan passed by Governor Andrew Cuomo and the State Legislature this spring.

2. Center for  Policy Analysis, University of Massachusetts at Dartmouth: New England Casino    Gaming Update 2013.  Economic Development series no. 74

3. Shafer, Stephen Q. New Commercial Casinos Will Mean Thousands of New Gambling Addicts  Dec. 2012  http://cagnyinf.org/wp/new-casinos-equal-1000s-of-gambling-addicts/

4. Shafer, Stephen Q.  Measure Something: Prevalence of Pathological and of Problem Gamblers. http://cagnyinf.org/wp/9_nov_2013_measure_prevalence/

5. Earl L. Grinols. Gambling in America Cambridge University Press 2004 pp. 171-174.

6. Shaffer HJ, Hall MN, Vander Bilt J. Estimated Disordered Gambling Behavior in the United States and Canada Report to National Gambling Impact Study Final Report 1999  https://divisiononaddictions.org/html/publications/meta.pdf

7.  Welte JW et al. The Relationship of Ecological and Geographic Factors to Gambling Behavior.  J. Gambling Studies ( 2004) 20: 405-442

8. Grinols EL and  Omorov  JD.  Development or Dreamfield Delusions? Assessing Casino Gambling’s Costs and Benefits.  J. Law and Commerce 1996-97, vol 16 p 59

The drawing is captured from Wikipedia.  Original by John Tenniel for  Through the Looking Glass and What Alice Found There, by Lewis Carroll

photograph of buckets is by Kevin Krebs from clickr.com photos 8561188366_eda5d758cf_

The author retired in 2010 as Clinical Professor of Neurology at Harlem Hospital Center, Columbia University. He has an M.P.H. in Epidemiology and an M.A. in Political Science gained while (1976-78) a Robert Wood Johnson Clinical Scholar in the Department of Medicine,  Columbia College of Physicians and Surgeons.  He is Chairperson of Coalition Against Gambling in New York,  a non-profit  registered in Buffalo.

Permission is hereby given to reproduce this post in whole or part as long as the permalink above is cited.

Leave a Reply