The personal income tax (PIT) is the single largest source of revenue for Canadian governments, but it also has a major impact on the level and growth of personal income. The PIT influences a wide range of important economic decisions that individuals make, such as hours of work, location of employment, retirement dates, acquisition of education and training, occupation choice, risk-taking, entrepreneurship and savings. All of these decisions affect the amount and types of income generated in the economy.
When governments raise the PIT rates, there are both direct and indirect costs imposed on the private sector. The direct cost is the additional PIT that taxpayers have to pay. The indirect cost is the loss of income-generating activities arising from the taxpayers’ responses to the tax rate increases. The marginal cost of public funds (MCF) measures both the direct and indirect costs to society in raising an additional dollar of tax revenue through a tax rate increase. In the absence of other distortions in the economy, such as monopolies or pollution, the cost of raising an additional dollar of tax revenue is normally more than a dollar because of the additional welfare loss arising from further distorting decisions that affect the size of the economic pie.
The MCF can help inform tax policy decisions.
By indicating which are the highest and which the lowest cost sources of tax revenue, the MCF can point the direction for revenue neutral changes in the tax mix that would lower the total cost to society of raising a given amount of tax revenue. It can also be used in cost-benefit analyses of public expenditure programs because the additional benefit from an additional dollar spent on a program should be greater than the cost of financing it through the tax system. As we will see, the MCF of the personal income tax can be very high. In Ontario, our estimates show that it is more than $6 per $1 of revenue raised.
With costs this high, marginal public expenditures would have to bring very high value to justify raising extra income tax revenues to pay for them. No other province has so high a marginal cost of funds, though in every province the MCF exceeds $2 per $1 raised: to the direct cost of taxation has to be added more than another full dollar of indirect cost.
Several factors affect the magnitude of the MCF for any tax. Tax bases that are very tax sensitive have a higher MCF. If the taxpayer can readily shift activity away from a tax base, there is a greater distortion in the allocation of resources when its tax rate is increased, and the MCF for that tax base will be higher. A measure of the tax sensitivity of a tax base is the percentage reduction in the tax base from a one percentage point increase in its tax rate. This parameter is called the (own) semi-elasticity of the tax base, and it is a negative number because of the inverse relation between a tax base and the tax rate that is applied to it. In addition, the interaction between tax bases can affect the size of the welfare loss from a tax increase.
For example, an increase in the PIT rate, by lowering a taxpayer’s disposable income, will normally lead to a reduction in the consumption of goods and services that are taxed under a general sales tax, leading to a larger welfare loss per dollar of total tax revenue generated from the PIT rate increase. This interaction can be measured by the percentage change in the sales tax base when the PIT rate is increased by one percentage point, known as the cross semi-elasticity of the tax bases. Another factor that affects the size of the MCF is the tax rate itself.
Generally speaking, the higher the tax rate the higher the MCF because the welfare loss from taxation depends on the size of the tax wedge, the gap between the pre- and post-tax returns on an investment, or the gap between the before and the after-tax wage rates that workers receive. Finally, taxes that generate more tax revenue will have higher a MCF because distortions in the larger tax bases have a larger impact on economic activity and represent larger efficiency losses.
Given estimates of the semi-elasticities of the tax bases, the tax rates, and the shares of tax revenues generated by different taxes, it is possible to calculate the taxes’ MCFs. Here I will present some estimates of the MCFs for provincial personal income taxes based on the econometric estimates of the long-run tax sensitivities of the aggregate PIT bases found in a 2016 study by Ferede and Dahlby. That study found that the average own semi-elasticity of the provincial PIT bases is -3.5, implying that a one percentage point increase in the top provincial marginal income tax rate reduces a province’s PIT base by an average of 3.5 per cent. Quebec’s PIT base has the lowest tax sensitivity, with semi-elasticity of -2.62, perhaps because of lower population mobility than in other provinces. Interestingly, British Columbia has the most tax sensitive PIT base with a semi elasticity of -4.42.
[Insert Figure 1 here]
Figure 1 (above) shows the estimates of the marginal cost of raising an additional dollar of tax revenue through an increase in the top provincial marginal tax rate in each of the 10 provinces based on their projected 2017 tax rates. Ontario has the highest MCF for a PIT rate increase, where raising an additional dollar of PIT revenue costs the private sector $6.77, The MCF in Ontario is high because it has one of the highest top marginal PIT rates, 20.53 per cent, on top of the federal rate of 33 per cent, and a relatively sensitive PIT base with a semi-elasticity of -4.15.
Note that while British Columbia has the lowest top provincial marginal income tax rate, 14.7 per cent, it has a higher MCF than Alberta, 2.85 versus 1.77, even though Alberta’s top marginal rate is slightly higher at 15 per cent. This arises because, as previously noted, the econometric evidence indicates B.C. has the most tax sensitive PIT base, though it’s not clear why B.C.’s PIT base is so sensitive.
There’s an important lesson drawn from this—we should not simply judge a province’s “tax competitiveness” based on how its tax rates compare with other provinces because tax base sensitivity varies between provinces, and a province, such as B.C. with a relatively low tax rate, can nonetheless have a very costly tax system.
The 2016 Ferede and Dahlby study does not estimate the MCF of the federal PIT. Although not strictly comparable with provincial MCFs for 2017 shown in Figure 1, the MCF for the federal PIT was 1.30 in a study by Baylor and Beauséjour (2004, Table 4) and 1.17 in a 2012 study, “The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Canadian Provincial Governments” in the journal International Tax and Public Finance by Dahlby and Ferede (Table 6a). These estimates of the MCF for the federal PIT are lower than those for the provincial PITs because provincial PIT increases create incentives to shift income-generating activities across provincial borders, an option not available at the federal level.
The estimates of the MCFs in all of the provinces except Alberta exceed $2 for every additional dollar of PIT revenue, indicating that the indirect cost of raising an additional dollar of PIT revenue is more that the additional amount the taxpayer pays. However, the studies by Dahlby and Ferede (2012) and Ferede and Dahlby (2016) indicate that the provincial corporate income taxes have even higher MCFs, and are more costly sources of tax revenue than the PIT. On the other hand, the provincial sales taxes generally have lower MCFs that the PIT. Tax reforms that shift more of the tax burden from provincial corporate income taxes to provincial sales taxes, or even to the provincial PITs, would improve provincial economic performance.
And much of the benefit of lower corporate income taxes would accrue to workers, offsetting the concerns about the distribution effects of these change in the tax mix, because a 2016 study by Ebrahimi and Vaillancourt have shown that workers’ wages and salaries increase when provincial CIT rates are reduced.
For more on Canada’s personal income tax, see here.
Commentary
The high cost of raising revenue through Canada’s personal income tax
EST. READ TIME 7 MIN.Share this:
Facebook
Twitter / X
Linkedin
The personal income tax (PIT) is the single largest source of revenue for Canadian governments, but it also has a major impact on the level and growth of personal income. The PIT influences a wide range of important economic decisions that individuals make, such as hours of work, location of employment, retirement dates, acquisition of education and training, occupation choice, risk-taking, entrepreneurship and savings. All of these decisions affect the amount and types of income generated in the economy.
When governments raise the PIT rates, there are both direct and indirect costs imposed on the private sector. The direct cost is the additional PIT that taxpayers have to pay. The indirect cost is the loss of income-generating activities arising from the taxpayers’ responses to the tax rate increases. The marginal cost of public funds (MCF) measures both the direct and indirect costs to society in raising an additional dollar of tax revenue through a tax rate increase. In the absence of other distortions in the economy, such as monopolies or pollution, the cost of raising an additional dollar of tax revenue is normally more than a dollar because of the additional welfare loss arising from further distorting decisions that affect the size of the economic pie.
The MCF can help inform tax policy decisions.
By indicating which are the highest and which the lowest cost sources of tax revenue, the MCF can point the direction for revenue neutral changes in the tax mix that would lower the total cost to society of raising a given amount of tax revenue. It can also be used in cost-benefit analyses of public expenditure programs because the additional benefit from an additional dollar spent on a program should be greater than the cost of financing it through the tax system. As we will see, the MCF of the personal income tax can be very high. In Ontario, our estimates show that it is more than $6 per $1 of revenue raised.
With costs this high, marginal public expenditures would have to bring very high value to justify raising extra income tax revenues to pay for them. No other province has so high a marginal cost of funds, though in every province the MCF exceeds $2 per $1 raised: to the direct cost of taxation has to be added more than another full dollar of indirect cost.
Several factors affect the magnitude of the MCF for any tax. Tax bases that are very tax sensitive have a higher MCF. If the taxpayer can readily shift activity away from a tax base, there is a greater distortion in the allocation of resources when its tax rate is increased, and the MCF for that tax base will be higher. A measure of the tax sensitivity of a tax base is the percentage reduction in the tax base from a one percentage point increase in its tax rate. This parameter is called the (own) semi-elasticity of the tax base, and it is a negative number because of the inverse relation between a tax base and the tax rate that is applied to it. In addition, the interaction between tax bases can affect the size of the welfare loss from a tax increase.
For example, an increase in the PIT rate, by lowering a taxpayer’s disposable income, will normally lead to a reduction in the consumption of goods and services that are taxed under a general sales tax, leading to a larger welfare loss per dollar of total tax revenue generated from the PIT rate increase. This interaction can be measured by the percentage change in the sales tax base when the PIT rate is increased by one percentage point, known as the cross semi-elasticity of the tax bases. Another factor that affects the size of the MCF is the tax rate itself.
Generally speaking, the higher the tax rate the higher the MCF because the welfare loss from taxation depends on the size of the tax wedge, the gap between the pre- and post-tax returns on an investment, or the gap between the before and the after-tax wage rates that workers receive. Finally, taxes that generate more tax revenue will have higher a MCF because distortions in the larger tax bases have a larger impact on economic activity and represent larger efficiency losses.
Given estimates of the semi-elasticities of the tax bases, the tax rates, and the shares of tax revenues generated by different taxes, it is possible to calculate the taxes’ MCFs. Here I will present some estimates of the MCFs for provincial personal income taxes based on the econometric estimates of the long-run tax sensitivities of the aggregate PIT bases found in a 2016 study by Ferede and Dahlby. That study found that the average own semi-elasticity of the provincial PIT bases is -3.5, implying that a one percentage point increase in the top provincial marginal income tax rate reduces a province’s PIT base by an average of 3.5 per cent. Quebec’s PIT base has the lowest tax sensitivity, with semi-elasticity of -2.62, perhaps because of lower population mobility than in other provinces. Interestingly, British Columbia has the most tax sensitive PIT base with a semi elasticity of -4.42.
[Insert Figure 1 here]
Figure 1 (above) shows the estimates of the marginal cost of raising an additional dollar of tax revenue through an increase in the top provincial marginal tax rate in each of the 10 provinces based on their projected 2017 tax rates. Ontario has the highest MCF for a PIT rate increase, where raising an additional dollar of PIT revenue costs the private sector $6.77, The MCF in Ontario is high because it has one of the highest top marginal PIT rates, 20.53 per cent, on top of the federal rate of 33 per cent, and a relatively sensitive PIT base with a semi-elasticity of -4.15.
Note that while British Columbia has the lowest top provincial marginal income tax rate, 14.7 per cent, it has a higher MCF than Alberta, 2.85 versus 1.77, even though Alberta’s top marginal rate is slightly higher at 15 per cent. This arises because, as previously noted, the econometric evidence indicates B.C. has the most tax sensitive PIT base, though it’s not clear why B.C.’s PIT base is so sensitive.
There’s an important lesson drawn from this—we should not simply judge a province’s “tax competitiveness” based on how its tax rates compare with other provinces because tax base sensitivity varies between provinces, and a province, such as B.C. with a relatively low tax rate, can nonetheless have a very costly tax system.
The 2016 Ferede and Dahlby study does not estimate the MCF of the federal PIT. Although not strictly comparable with provincial MCFs for 2017 shown in Figure 1, the MCF for the federal PIT was 1.30 in a study by Baylor and Beauséjour (2004, Table 4) and 1.17 in a 2012 study, “The Effects of Tax Rate Changes on Tax Bases and the Marginal Cost of Public Funds for Canadian Provincial Governments” in the journal International Tax and Public Finance by Dahlby and Ferede (Table 6a). These estimates of the MCF for the federal PIT are lower than those for the provincial PITs because provincial PIT increases create incentives to shift income-generating activities across provincial borders, an option not available at the federal level.
The estimates of the MCFs in all of the provinces except Alberta exceed $2 for every additional dollar of PIT revenue, indicating that the indirect cost of raising an additional dollar of PIT revenue is more that the additional amount the taxpayer pays. However, the studies by Dahlby and Ferede (2012) and Ferede and Dahlby (2016) indicate that the provincial corporate income taxes have even higher MCFs, and are more costly sources of tax revenue than the PIT. On the other hand, the provincial sales taxes generally have lower MCFs that the PIT. Tax reforms that shift more of the tax burden from provincial corporate income taxes to provincial sales taxes, or even to the provincial PITs, would improve provincial economic performance.
And much of the benefit of lower corporate income taxes would accrue to workers, offsetting the concerns about the distribution effects of these change in the tax mix, because a 2016 study by Ebrahimi and Vaillancourt have shown that workers’ wages and salaries increase when provincial CIT rates are reduced.
For more on Canada’s personal income tax, see here.
Share this:
Facebook
Twitter / X
Linkedin
Bev Dahlby
Senior Fellow, Fraser Institute
STAY UP TO DATE
More on this topic
Related Articles
By: Ben Eisen and Jake Fuss
By: Jake Fuss and Alex Whalen
By: Jake Fuss and Grady Munro
By: Ben Eisen and Jake Fuss
STAY UP TO DATE