Both taxpayers and governments struggle to stay on top of the various complex sources of tax law and to apply them in a myriad of different contexts. Given the potential for confusion and disagreement (not to mention the sometimes very large financial stakes involved) it would make sense to have a process for taxpayers to appeal government decisions to an expert body that can provide authoritative, reasoned and rational solutions to tax disputes. For this reason Canada, like the United States, has a specialized tax court dedicated to hearing appeals from decisions of the tax administration. Yet there is some evidence in both Canada and the US that judges in tax cases may be influenced by their own personal policy preferences or other factors extraneous to the “true” legal merits in deciding appeals from decisions of the tax administration. This paper examines in more detail appeals from tax assessments in Canada to understand the relative influence of judicial tax expertise and the policy preferences of judges on appeals to the Tax Court of Canada and the Federal Court of Appeal.
Our analysis reveals three main results: (1) policy preferences of judges matter, but not that much; (2) resources matter — a lot; and (3) there are dynamics relating to affirmation of appeals that are difficult to explain, although a desire to avoid the apprehension of bias is possible. First, there is robust evidence that policy preferences matter in the sense that there are measurable and statistically significant differences, but that they do not matter very much. Liberal appointees are only slightly more likely to favour the government than are Conservative appointees. This result is consistent with the findings in the literature where it is frequently found that there is no effect or only weak effect of party affiliations with voting patterns.
Second, the analysis consistently shows that resources matter in (1) representation; (2) the resources of taxpayers; and (3) on appeal. With respect to representation, self-represented litigants fare worse than taxpayers who are represented by tax agents, and taxpayers who are represented by agents fare worse than taxpayers who are represented by counsel. With respect to the resources of taxpayers, individual taxpayers are less successful than private corporations, and private corporations are less successful than public corporations. On appeal, the government is more selective about the cases that it appeals and is generally more successful than taxpayers, with the exception of public corporations.
Finally, we find that where Tax Court judges were appointed by Prime Ministers of different political parties the Federal Court of Appeal judge was 8% more likely to vote to affirm the Tax Court decision than where the Federal Court of Appeal judge and the Tax Court judge were appointed by a Prime Minister from the same party. It is not clear what is driving this result. One possibility is that judges may be sensitive to not wanting to appear to be politicized or partisan, and so may be more cautious about voting to reverse a judgment of a Tax Court judge with different political ties. If this is correct, it would be a striking illustration of the weak politicization of Canadian courts.