Munir Sheikh began his career as a public servant 40 years ago when he became an economist with the Economic Council of Canada. During the 1980s and 90s he was an economist at the federal Department of Finance, where he became senior assistant deputy minister in 2000. He went on to hold senior positions with Health Canada, the Privy Council Office and Human Resources Development Canada until Prime Minister Stephen Harper appointed him to the prestigious position of Chief Statistician of Canada in 2008.
Sheikh made national headlines in 2010, however, when he abruptly stepped down. His resignation took place in the context of the government’s decision to no longer require mandatory participation in Canada’s long-form census. The government misrepresented his views, stating publicly that he supported its decision. He told a House of Commons committee in July 2010: “The fact that in the media and in the public there is this perception that Statistics Canada is supporting a decision that no statistician would, it really casts doubt on the integrity of that agency.” He went on to add: “I, as head of that agency, cannot survive in that job.”
After his resignation, Sheikh publicly argued it was a mistake to cancel the long-form census because a voluntary survey would compromise the quality of census data. The truth did finally emerge that the cancellation of the long form census was wholly the government’s decision. Sheikh is now a distinguished fellow and adjunct professor at Queen’s University in Kingston, Ontario.
In recognition of his years of dedicated and principled public service, Sheikh was presented June 6 with the 2013 Corporate Knights Award of Distinction. In the exclusive commentary that follows, Sheikh weighs into discussions on the need for and impact of a carbon tax in Canada. His conclusion: If designed properly, taxing carbon makes sense for both the economy and the environment.
Arthur Cecil Pigou, a pioneer economist writing in 1920, argued that when the market system produces what he called “uncharged disservices” – which economists now call “externalities” – citizens are worse off and resources are misallocated. He also said an appropriate tax would improve a nation’s standard of living, or what he termed welfare. Pigou’s analysis of, and the policy solution for, externalities are now an integral and non-controversial part of the foundations of economic theory. These so-called uncharged disservices in the context of the current debate on the negative impacts of fossil-fuel use include, among others, climate change, higher ground-level ozone and acid rain.
But leave aside the environment for a moment. When the economics profession argues that, in the presence of these uncharged disservices, resources in the economy are misallocated, does it not logically follow that a proper tax policy that imposes a price on these disservices would allocate resources better and thus improve the economy?
If you now take into account any environmental benefits of a carbon tax, why should there be a trade-off between the environmental and economic objectives so that we can only get one at the expense of the other?
With economic theory so clear that a properly-designed carbon tax would improve resource allocation in the economy, why is there so much hesitation in implementing it to avoid the harmful negative effects of the “disservices” that come with burning fossil fuels?
Failure to act
There are indeed some plausible reasons for the failure to act. The most important is that there is a continued general belief that such a tax would have a negative impact on the economy. Another is that some sectors of the economy are relatively more dependent on the use of energy. They would suffer more than others as the cost increases for them would be above-average.
A third reason, related to the second and fairly important for a country like Canada, is the potential diversity of the regional effects of a carbon tax. It is argued that energy-rich regions, such as Alberta, would suffer more than others. Finally, there’s much concern that a carbon tax would be regressive, hurting lower-income citizens the most as they may spend larger proportions of their incomes on necessities, energy being one of them.
Combining these challenges would seem to suggest that there would hardly be anyone left in the country who would not get hurt by a carbon tax.
So we have a dilemma: the contrast between the non-controversial economic theory result and the list of above-mentioned negative consequences. How do we resolve this?
The economics discipline does have a solution: Quite simply, care must be taken to make it an appropriately designed carbon tax. Most of the problems listed above can generally be dealt with effectively by a range of policy instruments, so that the theory-predicted result can be achieved.
Keep in mind that there are potentially two sources of the problems listed above.
First, while a carbon tax would improve resource allocation by reducing the disservices of the use of fossil fuels, an overall increase in the tax burden on its own (without considering the effects of using the revenue generated by the tax) has the effect of reducing living standards. There is considerable literature available to show the cost of higher taxes for the economy in terms of lost output.
It is the overall increase in the tax burden that is critical from the perspective of negative national economic impacts. We would need to deal with it if we are to achieve the predicted theoretical result of a carbon tax.
Second, any carbon tax would naturally impose a disproportionate burden on those who produce or use fossil fuels more than others. Hence, there could be distributional effects across regions, sectors and people.
A consensus on the distributional consequences is harder to achieve. The range of views may include some liking most of the outcomes while others disagreeing with all of them. This is a difficult choice to make.
A useful scenario to consider is whether we have a set of policy instruments to minimize the distributional consequences of a carbon tax within the constraint that the economy not suffer.
This is where evidence-based policy development comes in. For this to happen we need models and data. We have both. With those, one can attempt to design a policy package that would accomplish the following set of objectives:
• Achieve the desired reduction in greenhouse gases using an appropriate level of a carbon tax;
• With the revenue generated from that carbon tax, set federal tax and spending levels such that there is no reduction in the overall level of economic output;
• Within this constraint, stabilize the existing income distribution, offsetting any negative consequences of a carbon tax, by using the mix of a range of taxes we now pay;
• Offset any distributional impact on economic sectors by reducing the taxes on the inputs they use; and
• Offset distributional impacts on the regions of the country.
I did some work along these lines using a set of economic, environmental and social policy models. This exercise yielded the following package that would mostly achieve these objectives: impose a federal carbon tax at the point of consumption; reduce personal income taxes tailored to offset the burden of a carbon tax by income class; reduce taxes on return on savings; and lower taxes on business inputs in a manner to try to offset sectoral distributional effects, which would also minimize any regional consequences.
I say mostly because there are those whose fossil fuel use is way beyond average and it is not possible or desirable from a policy perspective to micro manage policy beyond an appropriate aggregation level. As for the fossil-fuel sector, I assume that the difference between our production and lower demand would lead to a reduction in imports and an increase in our exports.
The economic variable used for such an exercise is the GDP, the most comprehensive aggregate economic output indicator currently available. It is well recognized that GDP excludes changes in natural capital and does not take into account the intergenerational economic effects of a policy change. With a more appropriate measure of production, it should be easier to get the result that economic theory predicts.
To be clear, such an exercise is only illustrative. There are at least four reasons for this observation. First, to develop a more precise package of policy action, one needs to capture in detail the latest structure of the Canadian economy, the tax system and income distribution. Second, in contrast to what is attempted above, policy makers may or may not want to offset all of the distributional effects across individuals, sectors and regions. Third, the constraint of revenue neutrality is imposed to demonstrate that a carbon tax need not reduce economic output. There may be uses of the carbon tax revenue that could strengthen the economy further or weaken it. Fourth, the package of policy change described above is not unique: there are likely to be other combinations of policy changes that may achieve similar or better outcomes.
In conclusion, the issue is not so much about a carbon tax per se, but the appropriate design of a carbon-tax-inclusive tax transfer system. This is an important topic that has been completely missing from the carbon tax debate. We need a more extensive discussion of the relevant issues using sorely needed analysis and the available evidence.