A Critical Review Of The Study “The Relative Effectiveness Of Tax Policy When Considering Mitigating Factors Influencing Tobacco Consumption Of American Youth”

A Critical Review of the Study “The Relative Effectiveness of Tax Policy When Considering Mitigating Factors Influencing Tobacco Consumption of American Youth”

I. Summary

Kristen Collett wrote an article called The Relative Effectiveness of Tax Policy When Considering Mitigating Factors Influencing Tobacco Consumption of American Youth that was published in the August 2003 edition of Issues in Political Economy. While her article has many relevant findings that will have significance for future studies, Collett neglected to include several variables which would have had an important impact on the results. In this paper I will compare and contrast Collett’s article with a similar study written by a more experienced author.

According to Collett, the youth of America have yet to realize the harm tobacco consumption causes, as youth smoking increased in the 90’s and is still rising.

She says that previous studies show that the most effective way to reduce tobacco consumption is through increased excise taxes.

For her study, Collett used data from the State Tobacco Activities Tracking and Evaluation (STATE) System to get information on the 51 states used in her observation. The STATE database contains current data on tobacco prevention and control as well as data on each state’s consumption.

Collett ran a regression to determine the correlations that exist between the dependent variable, and the seven independent variables. The dependent variable was consumption of 18-24 year olds. The independent variables included, tobacco growing states, excise tax, state control policies, advertising restrictions, consumption of 25-44 year olds, 45-64 year olds, and those over 65 years old.

The correlation between consumption of 18-24 year olds and tobacco growing states was negative, this means that states that grew tobacco had higher consumption. This was not a surprise because the states which have the lowest tobacco excise tax are also the states that grow tobacco. The correlation between tobacco consumption of youth and excise tax was positive. Collett did not expect that result because it implies higher consumption near a higher excise tax, which Collett feels, “…contradicts the law of demand.” The correlation between consumption of 18-24 year olds and state control policies was positive meaning in states where it is illegal for minors to purchase tobacco, consumption was at its highest. Collett expected the positive correlation between consumption of 18-24 year olds and both advertising restrictions and consumption of 25-44 year olds. This means that consumption is highest when there is increased consumption by an older age group and states where there aren’t any advertising restrictions. Collett did not expect the negative correlations between consumption of 18-24 year olds and consumption of 45-64 year olds, and 65 + year olds.

Collett then conducted a least squares analysis and focused on the R-square and adjusted R-square values. “…the R-square was 0.37, meaning that tobacco вЂ" growing states, excise tax, state control policies, advertising restrictions, and consumption of 25-44 year olds, 45-64 year olds, and 65+ year olds explain approximately 37 percent of the tobacco consumption of 18-24 year olds” (CITE). This means that 63 percent of the consumption is explained by other variables. Each of the variables in the regression had a P value that was too high to be significant, with the exception of consumption by 18-24 year olds. This variable had a P value of 0.001 indicating it had a strong significance.

After reviewing the results from her from regression, Collett realizes that

the majority of the variables she used did not have any real significance on the consumption of 18-24 year olds. So she dropped all the variables except consumption of 25-44 year olds and excise tax. After running a regression on the new model, Collet found the R square to be 0.33. This means the two variables explain 33 percent of consumption of 18-24 year olds. The P stat and T stat values indicated that consumption of 25-44 year olds was significant at the 5 percent level while excise tax was significant at the 17 percent level. This means Collect was 95 percent confident that consumption of 25-44 year olds is significant and 83 percent sure that excise tax is significant. P values of respectively, 0.020 and 0.009 indicate the regression was successful in finding relevant data.

Collet’s results did not support her hypothesis which was “…that tobacco consumption of 18-24 year olds will be higher in states that grow tobacco, have a lower excise tax, where it is illegal for minors to purchase cigarettes, do not restrict advertising, and have higher consumption of older age groups” (CITE).

II. Analysis

While Collet’s study is meaningful, it is lacking in robustness. The main criticism I have is the lack of relevant variables. The variables Collet included aren’t enough to let us see the whole picture.

For a more unimpaired look into this problem I refer you to, The Impact of Prices and Control Policies on Cigarette Smoking Among College Students, by Christina Czart. For this study, the authors used data from the 1997 Harvard School of Public Health College Alcohol Study. In regards to the size of their sample, Czart says, “After eliminating individuals with missing information for gender, age, smoking, ethnicity, parental education, and hours worked, a final sample of 15,148 individuals remains. Approximately 60 percent of the sample is female, and the sample is predominantly white” (CITE).

EVALUATION OF VARIABLES

Table 1

Variables used by Kristen Collet

Tobacco вЂ" growing states Consumption of 25-44 year olds

Excise Tax Consumption of 45-64 year olds

State control policies Consumption of 65+ year olds

Advertising Restrictions

Table 2

Variables used by CZART

Measure of smoking State and local policy measures

Smoke presently Price of cigarettes

Avg. daily number of cigarettes Local private workplace smoking restrictions