Canada’s Proposed Digital Services Tax: A Poor Solution in Search of a Problem
Making electoral promises is easy. Delivering on those promises while trying to rebalance budgets thrown out of whack by pandemic responses is somewhat trickier for governments. “Making the rich pay” is not exactly sound economic policy, but it is tempting for governments to go after big companies with supposedly deep pockets. And few targets are as tempting as the tech giants, sometimes referred to as the GAFA (Google, Amazon, Facebook, and Apple) companies. In the last few years, several countries, such as France, have begun imposing new digital services taxes (DSTs) that target these GAFA companies.
The federal government of Canada has succumbed to the temptation, too, releasing draft legislation to impose a 3 percent DST, to be retroactively applied to revenues earned on or after January 1, 2022. The government has confirmed, in the 2023 budget (at 189), its commitment to go ahead, if necessary, with the new tax measure.
The federal government would apply the DST at the beginning of 2024 (retroactively levied on 2022 revenues), and it would do so only if a multilateral agreement for the “acceptable” taxation of multinationals, including the GAFA companies, has not come into force by December 31, 2023. This international agreement, led by certain OECD countries, is intended as a response to criticisms that certain multinational companies do not pay enough tax in the countries where they generate profits.
The Threat of Less Profitable, or Unprofitable, Digital Services
The 3 percent DST, partly modelled on a similar measure in France, would apply to companies’ gross revenues, and it differs in this respect from the corporate income tax, which applies to profits. This difference is crucial: a 3 percent tax on revenues is anything but trivial. As discussed in a January 2020 Montreal Economic Institute (MEI) research paper, the global profit margin of Amazon from 2009 to 2018 averaged just 2.5 percent; this is less than the planned 3 percent tax rate. Indeed, if such a tax on revenues had been applied to all of the activities of the TSX 60 companies over the same 10-year period, it would have completely swamped the profits of nearly one-quarter (22 percent) of them—in particular, those with profit margins below 3 percent. Such a tax on gross revenues would be genuinely punitive, because even a company suffering losses would have to pay it. Such a tax can turn a company, or an entire portion of its activities, from profitable to unprofitable, and it can even threaten a company’s solvency.
Why, then, is the government proposing this highly punitive tax on digital companies while claiming that it wants to facilitate the digital transformation of the Canadian economy? The main justification is that the tech giants are not paying their “fair share.” Yet, as detailed below, this is simply not the case, and the new surtax, proposed with a view to solving a non-existent problem, would end up hurting the Canadian economy.
Do the Tech Giants Pay Their Fair Share?
The proposed DST would apply to companies with annual global revenue of at least €750 million and annual Canadian digital services revenue of at least $20 million. The tax would apply to revenue from online marketplaces, online advertising services, social media services, and user data.
The government intends to impose the DST at the beginning of 2024 in the event that a multilateral agreement has not taken effect, but the tax will be levied retroactively, on revenues from the beginning of 2022. The OECD’s work on the multilateral agreement has progressed slowly; a complex progress report was released in July 2022, and a draft agreement in January 2023. The agreement is intended to reform international rules in response to widespread criticisms that certain multinational companies do not pay enough tax in the countries where they generate profits.
Canada’s DST, if applied, would remain in place until the new rules contemplated by the multilateral agreement take effect, if they ever do. The government has declared its support for the OECD process and its desire to put an end to the fiscal “race to the bottom” and to force large corporations to pay their “fair share.” But while countries such as France and Canada are concerned with extracting more tax revenues, the question remains: Do the large tech companies that are the DST’s primary targets really get away with paying less tax than their peers? Is the government’s premise correct?
The answer to these questions is no. In the 2020 MEI research paper mentioned above, it was calculated that the average tax rate of the GAFA companies, over 5 years and 10 years, was actually higher than that of the large corporations that make up Canada’s TSX 30 and TSX 60. By this measure, then, these large digital companies actually pay more than their peers, and they should not be singled out for higher taxation.
Moreover, the GAFA companies not only pay more overall tax than other large companies, but also have an effective tax rate that is already higher than the 15 percent rate contemplated by the proposed pending international agreement. The MEI research calculated an average effective tax rate of 24 percent for the GAFA companies, whether over 5 years or 10 years. Canada’s DST, then, would effectively be piling a punitive surtax onto these companies in order to hit a target that has already been reached.
A Surtax That Would Harm the Digital Economy and Its Users
As we attempt to foresee the effects of the proposed DST on the Canadian economy, it is useful to consider what happened in France after the introduction of a similar measure. Once the so-called GAFA tax was implemented in that country, prices went up by 2 percent for clients of Google and by 3 percent for clients of Apple and Amazon. To a large extent, in other words, the tech giants simply passed the tax on to their local customers in the form of higher prices. In Canada, as in France, the direct clients of the large digital companies are individuals and businesses whose bills may increase if the government adopts the proposed DST. Even with free platforms, the tax can be passed on to businesses that pay for advertising on these platforms and may then be passed on, in turn, to the businesses’ own Canadian customers.
In May 2021, the parliamentary budget officer estimated that the federal government would collect more than $4.2 billion over five years with its proposed tax. Because it is reasonable to expect that a large part of the tax will be passed along to consumers through higher prices, the funds will ultimately come out of consumers’ pockets. That amount would simply represent a higher cost for the whole digital ecosystem.
This estimate does not account for compliance costs, which are likely to be non-negligible, nor does it take account of dynamic effects such as the impact on economic activity and the loss of investment and productivity related to a drop in profits, or the drop in demand for services related to price increases. The overall negative economic consequences would thus certainly be greater in the long term. The losses for the rest of the Canadian economy, beyond the digital sector, may be more difficult to quantify, but they would nonetheless be very real, because the prices of some goods and services would be pushed higher, at a time of historically high price inflation rates.
It is also very likely that Canadian companies will experience the negative effects of the DST more directly. According to a 2019 study, the French DST affected not only the GAFA companies but also some 27 other companies, including French digital businesses such as Criteo. A similar situation would no doubt arise in Canada, because the proposed DST would likely apply not only to foreign tech giants but also to some domestic companies.
According to Statistics Canada, Canadian companies brought in $398 billion from online sales in 2021. On average, large Canadian companies declared $79 million in gross revenues from e-commerce, a figure that is much higher than the DST threshold of $20 million. And this figure does not include the other three revenue sources targeted by the DST—namely, online advertising services, social media services, and user data.
Last but not least, Canada’s proposed DST seems to go against the spirit of the 2021 multilateral standstill agreement, which stated that no DST-like tax measure was going to be imposed “until the earlier of 31 December 2023 or the coming into force of the [Multilateral Convention].” Retroactively taxing 2022 and 2023 revenues under legislation enacted in late 2023, as currently proposed, is arguably not what the framers of that clause had in mind.
Of course, because the primary targets of the DST would be US companies, it is widely expected that the United States would retaliate by using trade-based measures against Canadian companies, a risk that was pointed out recently in a submission of the Canadian Chamber of Commerce to the finance minister.
Canada Should Scrap the DST
The adoption of the proposed DST would not be good policy for Canadians. This tax, in addition to imposing billions of dollars of direct costs on the Canadian digital economy and its users in the form of higher prices, will serve as a disincentive to future investment in that sector in Canada. Digital companies, as demonstrated above, already pay at least their “fair share” of taxes.
There is simply no good reason for the DST, and it should therefore be scrapped, even if the multilateral tax agreement is not implemented by the end of 2023. The high costs of the DST for Canadian companies and consumers are not justified, and the threat of the tax, now weighing on those Canadian companies, is already undermining Canadian innovation. That threat, and the uncertainty it creates, should be removed as soon as possible.
Wanting Canadian companies and multinationals to operate under a global fiscal framework is understandable. This desire, however, must not serve as a fig leaf for new tax measures that would mostly end up hurting Canadians and Canadian companies.
Valentin Petkantchin is Vice President, Research at the MEI. The views reflected in this opinion piece are his own.