In Part I we saw that the goals of taxation were two-fold: to collect revenue for the government and to discourage certain activities. However, the government hoarding Jacques Plante as an Edmonton Oiler hockey cards in an attempt to reduce the after-tax income of rich-people is not aligned with our goals.

If there were no other considerations when creating tax policy, then designing a tax system would to meet our two goals is relatively easy- we identify a number of negative externalities and set tax rates on those activities as to eliminate the externality. If any additional revenue is needed, we identify a number of other activities (‘sins’) we would like to discourage and tax those. I use the qualifier relatively because a number of complications exist:

1. How much tax revenue to collect?

   In other words, when do we stop introducing new taxes? When is enough… enough?

2. What should the rate of taxation be to correct for the externality?

   The optimal tax rate is determined by calculating the marginal cost the activity that generated the externality places on others. This is easier to do in theory than it is in practice.

3. Which activities do we consider a ‘sin’?

   Joe wants to see a sin tax placed on alcohol, Mary wants to see one placed on abortion. Alex wants to see a sin tax on eating meat, Dave wants to see one placed on marijuana consumption. Betty wants to see a sin tax on gay marriage, Howard wants to see one placed on cutting down trees. As a society, how do we decide which activities are harmful to society and which are beneficial?

4. Can we cost-effectively collect the tax?

   As a society we might decide we want to place a tax on using racial slurs or picking your nose (though Dr. Fredrich Bischinger would disapprove). As a practical matter it is going to be difficult, if not impossible, to enforce and collect these taxes. Even taxes that are possible to collect may come with a high collection and enforcement costs. A tax where 90% of the money generated goes to collection and enforcement will provide little revenue to the government (though may still ‘work’ in the sense of discouraging the activity taxed).

There are at least four other real-world considerations that we need to consider when designing a tax system:

1. Ability to Pay

   Instead of a setting a tax rate based on an activity (earning income, emitting pollution), we could simply send each person in Canada a tax bill for $10,000 a year. Such a tax (known as a head tax) would be difficult, if not impossible, for low-income/low-wealth individuals to pay.

2. How Useful is an Extra Dollar to a Person?

   Related to the above point (in fact, some economists would combine them into a single ability to pay principle). Giving twenty dollars to Bill Gates will make no difference to his life. Giving me twenty dollars would provide me with some additional happiness – I might use it on a bottle of wine or a Nintendo Wii game I wouldn’t have otherwise bought. Giving $20 to a person well below the poverty-line is likely to provide a great deal of happiness, as it may make the difference between being able to pay the electricity bill or being able to eat three meals a day.

   Economists refer to this phenomenon as declining marginal utility of wealth – the more wealth you have, the less happiness you receive from an extra dollar. The flip-side of this is that taking a dollar away in taxes from a less wealthy person incurs more hardship than taking a dollar away in taxes from Bill Gates.

   Due to the declining marginal utility of wealth (and a person’s ability to pay) it is desirable, all else being equal, to have a disproportionate share of taxes paid for by the rich. However,

   1. We can make any tax disproportionately paid for by the rich, if we rebate some of the tax revenue back to low-income households. The Goods and services tax/harmonized sales tax (GST/HST) works this way, with low-income households mailed a GST rebate cheque.
   2. There are some taxes that we are simply willing to overlook the fact that they disproportionately affect the poor (known as regressive taxes). Sin taxes on cigarettes are the classic example, since a disproportionate amount of low-income persons smoke. However, what matters when analyzing a tax system is not how much one tax affects the poor vs. the rich, rather we are concerned with the relative tax burden of the entire tax system. We can make up for the regressive nature of the cigarette tax through non-regressive (progressive) taxes elsewhere.

3. Does the tax violate the Canadian constitution?

   Section 15(1) of Canadian Charter of Rights and Freedoms states:

   15. (1) Every individual is equal before and under the law and has the right to the equal protection and equal benefit of the law without discrimination and, in particular, without discrimination based on race, national or ethnic origin, colour, religion, sex, age or mental or physical disability.

   We cannot create a tax for practicing a particular religion, either directly (a $500 tax for being an Anglican) or indirectly (a $500 tax for wearing a yarmulke). Greg Mankiw has advocated taxing individuals in proportion to their height – I suspect this would violate the Canadian constitution. (Note: I use the term ‘suspect’ since I am not a constitutional scholar, so I leave the issue to those more qualified. For what it’s worth, I did study the field as an undergraduate under the direction of a very young Ian Brodie).

4. Is the tax politically feasible?

   This should be self-explanatory. This is an issue I will leave to the political pundits. Too often, however, the argument of political feasibility is used to argue against any change in the tax system. But clearly new taxes are introduced from time to time, such as the income tax in 1917, gas taxes in the 1920s, employment insurance in the 1940s, Canada Pension Plan ‘contributions’ in the 1960s, the GST in the 1980s, etc. You can make a convincing argument how any of these taxes was politically infeasible around the time they were introduced, yet clearly they were enacted. As such, we will not spend a great deal of time discussing political feasibility.

Now we understand the constraints we are operating under, we can start to examine the activities we would like to discourage and the activities we would like to avoid discouraging. We will begin that discussion in Part III.