|
 |
Third, using 1995 estimates,
the authors show that Canadian government gambling revenues now constitute
about two per cent of total government revenues. What I find interesting
about this statistic is that it invites study on the relationship between
the revenue and the expense side of government ledgers. A high proportion
of Canadian provincial budgets is spent on health care. An aging population,
technological advances and a competitive international market for health
care practitioners will heighten cost pressures further. If governments
expand gambling in ways that are later found to cause even tiny increases
in health care expenditures, the revenue "growth" becomes illusory, particularly
with the advent of intensified competition from offshore locations and
the Internet.
The expense summary of provincial
lottery corporations is nicely done. An interesting minor addition would
be a detailing of marketing and promotion costs. Agencies such as the
British Columbia Lottery Corporation rank among the largest advertising
accounts in their provinces. It is important to maintain awareness of
these expenditures into the future given the well-understood example of
tobacco and alcohol marketing.
The authors then review family
expenditure surveys using an impressive number of domestic and international
sources. They present a multivariate analysis in order to identify the
key determinants of purchasing decisions. Their evaluation of the incidence
of gambling taxes reveals that they are second only to tobacco in terms
of regressivity.
The final part of the study
has attracted some interest in the popular press. It finds that the benefits
of gambling in Canada greatly outweigh the costs. The authors begin by
reviewing the methodological issues. Appropriately, they point out the
need for an "incremental" approach. In the case of problem and pathological
gambling, for example, it is important to try to separate out the costs
created by other illnesses such as alcoholism in order to get useful results.
Unfortunately, the underlying causal linkages remain uncertain, and so,
remain "problematic" for cost estimation projections. The difficulty is
exacerbated by our limited knowledge of incidence, due to the inherent
limitations of self-report data and poor (or unreported) response rates,
evident in the two British Columbia incidence studies with which I am
most familiar.
The authors' cost estimations
include only "real" resource costs. What economists call "transfers" are
not included. "Real resource costs ...do not include any form of transfers,
including the proceeds of crime (theft), government transfers (welfare),
inter-family transfers and bad debts (transfers from creditors to debtors),
since transfers do not use additional resources" (p. 41). This is good
economic practice but it is also a good reason why so many students of
the overall effects of gambling dismiss economic studies that take this
line as irrelevant. Such studies do tell us something, but they manifestly
do not tell us everything about the social impacts of gambling.
To put the economist's definition
of economic gain and loss into perspective, consider a tax of $100 on
the 100,000 poorest people in the country, the proceeds of which are used
to pay a lump sum to the richest person. This would be neither good nor
bad (economically) as it merely transfers wealth from one set of people
to another. Closer to the problem at hand, if gamblers are driven to embezzle
money from others, and they seek out poorer and less well-informed people
as the easiest victims, but the government does not respond with additional
police and other resources that are costs to the justice system, there
is no "loss" to be set against the gains of gambling.
Similarly, if an unemployed
or retired person commits suicide because of insufferable gambling losses,
there is no economic cost (in fact, there may be an economic gain in the
sense that the person will not consume more medical and hospital resources).
A cost is recorded only if the person who committed suicide is an employed
person, and then, only if he is not replaced by someone who is employed
(p. 41). Of course, it is interesting to uncover the narrowly defined
economic costs of gambling (and I am not ridiculing the attempt), but
as these examples show, the costs are much less than the social losses,
many of which show up in economists' calculations as "mere" transfers.
Thus, the authors' conclusion that gains from gambling exceed losses must
be interpreted with extreme caution.
The "population health paradigm"
— defined as a conceptual framework for thinking about why some people
(and hence, some societies) are healthier than others — will help
shed more light on gambling as a desirable fiscal tool for governments.
Vaillancourt and Roy's treatment of income distribution and socio-economic
status is useful in this regard because both have been found to be important
determinants of population health.
The authors' assertion that
more research is needed to help understand provincial and regional issues
associated with gambling expansion is well taken. Their conclusion suggests
that the risks of expanded state-sponsored gambling in Canada can be justified
by societal and government benefits. On that point, I must respectfully
disagree (Henriksson, in press; Henriksson & Lipsey, 1999). That said,
I found this study to be a useful contribution to the literature.
References
- Henriksson, L.E. (In
press).
- Gambling in Canada: some
insights for cost-benefit analysis. Management and Decision Economics.
- Henriksson, L.E. &
Lipsey, R.G. (1999).
- Should provinces expand
gambling? Canadian Public Policy, XXV(2), 259—275.
The author thanks Professor
Richard Lipsey for his insightful comments and suggestions.
This book review was not
peer-reviewed.
Submitted: December 20,
2000
Correspondence:
Lennart E. (Len) Henriksson, PhD
Faculty of Commerce & Business Administration
University of British Columbia
2053 Main Mall, Vancouver, BC, Canada V6T 1Z2
Tel: (604) 822-8289
E-mail: len.henriksson@commerce.ubc.ca
|
|
