It’s the automotive industry’s $13.2-billion question: Why isn’t the Canadian federal government matching the money it gave to Volkswagen to keep the Stellantis-LG battery plant in Windsor, Ontario? They have, after all, promised the German automaker some $13.2 billion in ongoing subsidies. Why shouldn’t Stellantis get the same treatment? At least, that’s Stellantis’ argument, and since it’s been feeding the media — mostly the Toronto Star — with little tranches of outrage, Stellantis’ is the only story being told. But there are questions to be answered, not the least of which is how much all this subsidization is going to cost over the next decade. First, though, let’s start with the seemingly outlandish possibility that—
Maybe Volkswagen is not getting the full IRA
The American Inflation Reduction Act (IRA), as Motor Mouth has detailed ad nauseum, is devoting untold billions to the production of electric-vehicle batteries in the United States. Incentives offered to automakers to build their assembly plants are nothing new — in either Canada or the U.S. — but the IRA’s promise to subsidize the end product of said plants, continuously, for almost a decade, well, that really is beyond the pale.
More specifically, the IRA’s clause 45X promises — again, as has been detailed here previously — US$45 per kilowatt-hour for every battery a U.S.A.-based plant builds, from the time production begins, through until 2032. In actual fact, that US$45/kWh subvention is broken into two parts: US$35 per kilowatt-hour is promised if the plant builds the battery cells in the good old U.S. of A; and another US$10/kWh is forthcoming if those same cells are then built into the modules that comprise all EV batteries, again in the U.S.A. Those are the monies it is claimed the Canadian federal government matched in attracting Volkswagen to St. Thomas, Ontario, and the same stipend that Stellantis now says it is seeking if it is to keep its proposed tie-up with LG in Windsor.
The only problem with that rather simplistic explanation is that—
Motor Mouth: Stellantis pulling out of Ontario battery plant
Motor Mouth Podcast: Will EV battery subsidies doom the North American auto industry?
The numbers don’t add up
InsideEVs, Reuters, and Motor1 are all claiming that Volkswagen’s St. Thomas plant will have a 90-gigawatt-hour annual manufacturing capacity when it is running at full steam. Volkswagen also says the plant will start producing battery cells by 2027. Meanwhile, Reuters also claims that Canada matched the IRA’s US$35/kWh for battery-cell manufacturing — VW’s St. Thomas plant is seemingly not going to assemble the cells into modules, forgoing the last US$10/kWh of IRA funding — to land the deal.
You shouldn’t have to reach for your calculator to realize that those numbers don’t add up. Ninety gigawatt-hours at US$35/kWh, for instance, adds up to US$3.15 billion per year. That’s CDN$4.3 billion. Again, per year. The IRA runs to 2032 so, at full capacity starting in 2027, St. Thomas should be worth CDN$25.8 billion, give or take, say, Ontario’s annual deficit or so. Admittedly, it’s going to take some time for Volkswagen to get St. Thomas running at full capacity.
As well, the IRA calls for reducing the monies paid out in the program’s last three years — 25-per-cent annual reductions starting in 2030 — ostensibly to wean automakers off the crack cocaine, oops, I mean “incentives” they will have been mainlining for a decade. That would reduce the potential graft to CDN$19.4 billion, but that’s still a long way from the comparatively measly CDN$13.2 billion the Canadian government is said to have proffered. The exactitude of those calculations notwithstanding, either Volkswagen is estimating the market won’t need St. Thomas’ full capacity for quite some time; or else it isn’t going to get the full IRA.
But all this posturing over what Volkswagen did or did not get, and what Stellantis should — or should not — be compensated for, obfuscates another problem the federal Liberals may be just beginning to cotton to. Namely that the monies being proffered are unprecedented in the industrial history of either country. They are such unbelievably large amounts, in fact, that we really need to, once again—
Follow the money
Earlier this year, the American Department of Energy’s Vehicle Technologies Office estimated that by 2030 — and remember the IRA will be in force until 2032 — there will be 1,000 gigawatt-hours of battery production in North America, up from just 55 gWh in 2021. When you, again, do the math, that works out to about US$34 billion in battery-manufacturing subsidization in 2030 alone. Do some more rough math, factor in again the gradual withdrawal to 2032, and you still have the capacity for all this battery-building largess to cost more than US$175 billion by the time the automakers have to go cold turkey (that is, assuming the American government really does cut them off then).
And that doesn’t factor in the IRA’s other subvention of EVs, clause 30D, which offers up to US$7,500 per EV assembled — and with materials sourced — within the USMCA agreement (namely the United States, Mexico, and Canada). That same study says that 2030’s 1,000 gigawatt-hours of USMCA battery production will be enough to produce somewhere between 10 and 13 million EVs, depending on whether those cells are divided up in approximately 77-kWh tranches (roughly what’s in a Hyundai Ioniq5) or 100-kWh bits (like a Tesla Model S).
Between the American government’s IRA payouts — and Canada’s attempts to match them — this EV subsidization boondoggle could cost North American governments a total of US$350 billion
Currently, few cars are eligible for the full US$7,500 because of those restrictive assembly requirements. But, considering the levels of subsidization involved — and the resultant flurry of construction occurring south of the border — it’s virtually assured that, in the near future, all save the most expensive electric vehicles sold in North America will be eligible for both the battery manufacturer’s subvention and the consumer tax credits.
More importantly, the IRA’s rebates are not, as with previous American EV tax credits, limited to a company’s first 200,000 sales, which means that many, if not most, of those 10 to 13 million EVs expected to be sold in 2030 will be eligible for some portion of the US$7,500 tax credit; or, in Canada, our $5,000 federal subsidy.
Trying to figure out how much that will all cost makes for an imprecise calculation at best, but if we factor in the various state and provincial subsidies, but then depreciate from the total because some cars will be too expensive to qualify and others won’t earn the full amount, a fair number would seem to be US$5,000 a car.
At the lower end of DoE’s estimation, if governments are still subsidizing as proscribed, that’s another US$50 billion in EV subvention. Again, in 2030 alone. Work in the consumer tax-credit numbers for the entire 2023-to-2032 timeframe, and we’re looking at somewhere in the order of upwards of another $175 billion. In other words, between the American government’s IRA payouts — and Canada’s attempts to match them — this EV subsidization boondoggle could cost North American governments a total of US$350 billion.
It couldn’t possibly come at a worse time
The only headline bigger than the stalemate between the Canadian government and Stellantis is the debt-ceiling crisis being fought south of the border. According to the latest reports, President Biden will have to freeze spending in all departments except defense. The IRA has so far escaped cuts, but it’s hard to believe that more than a third-of-a-trillion dollars of EV subsidization has eluded the attentions of the GOP’s right-wing Freedom Caucus, who, if you’ve been following American politics at all, you know need but a few members to vote to derail this agreement.
And just for a little context regarding how big these numbers are, let me point out that, according to numbers posted by the Council on Foreign Relations, all of the military aid to Ukraine — which the Freedom Caucus likewise opposes — is costing less than what the American government will spend just on battery subvention in 2030. I’ll remind you that Russia-Ukraine war is the greatest conflict Europe has seen since the Second World War, and the U.S.A. is, by far, Ukraine’s largest sponsor.
Closer to home, besides the heightened interest rates that make servicing our rapidly expanding national debt more difficult, there are reports that other battery manufacturers — including South Korean-based SK Innovations, which is reportedly looking at three sites in Ontario — are seeking to establish assembly facilities in Canada. Even the least politically astute amongst us can probably figure out that the deal the feds offer Stellantis will set the precedent for what subsequent battery manufacturers will be looking for.
And, finally, I must apologize. Despite my many crunchings of numbers, I am leaving you with more questions than answers. But considering the sparseness of information being proffered by either side — and the not-hardly-subtle omissions in the information that is released — perhaps I have narrowed the focus on some specific terms of the negotiations.
For instance, did Volkswagen really get the full IRA, or did it accept some lesser, though still substantial, per (kilowatt-hour) diem? More to the point, would Stellantis really walk away from its (already partially-constructed) battery plant? Just as importantly — at least to Stellantis CEO Carlos Tavares — would this proposed 45-gigawatt factory be welcome south of the border with the IRA’s full US$45/kWh subvention (unlike VW, Stellantis’ plant will assemble battery modules)? And — as if this discussion isn’t fraught with enough economic calamity — if Stellantis really does can the Windsor battery plant, what does it mean for the long-term future of the rest of the company’s Canadian workers?
Unfortunately, with all the disinformation being doled out piecemeal by both sides, we won’t know the answer until either construction starts anew; or Stellantis cuts a deal with the American government. Nonetheless, for me, one conclusion stands out. And it’s the same thing I said when I first broke this story a week ago. A major factor in this contretemps is that Stellantis — which, at least locally, still sees itself as a domestic automaker — doesn’t seem to have been given the same heads-up regarding the contents of President Biden’s Inflation Reduction Act as some of its competitors. It’s pissed, and the jobs promised in Windsor are the bargaining chip. Regardless of what VW did or didn’t get, methinks that Stellantis is driving a hard bargain.
Author’s note: Some reading this piece may object to me conflating American and Canadian subsidies into one final sum: e.g. that US$350 billion. This recent squabbling, for instance, shows that whether the IRA’s battery subsidy is paid directly to a U.S. manufacturer; or Canada has to pony up a similar amount to attract a battery plant to our shores, those US$45 per kilowatt-hour will force almost all batteries to be built in North America. Our federal government’s CDN$5,000 consumer incentive, meanwhile, may not match the US$7,500 the American IRA is offering, but considering that U.S. auto sales outnumber the Canadian by a factor of 10 to one, and that there are state and provincial subsidies available as well, my use of US$5,000 per EV seems reasonable.
By the same token, anyone thinking some of the electric vehicles sold in North America will be, like other cars, imported from other nations is truly underestimating the Inflation Reduction Act’s competition-killing level of subsidization. Whoever is paying for it, those US$350 billion in subvention virtually preclude imported EVs from being able to compete in North America. Oh, hyper-expensive luxury electrics will come from abroad, mainly because their customers are unlikely to be swayed by such picayune monies. But, for the meat of the market, EVs produced in North America will have a $5,000 to $15,000 advantage depending on what you are buying and where. No amount of cost-cutting production techniques or low-wage assembly workers can overcome such an advantage.