Statistics Canada reported no growth in real gross domestic product for July, a reading taken before the heaviest United States tariffs entered the data. Economists say the damage arrives in the fourth quarter, and the Bank of Canada meets on October 28 with a dilemma on its hands.
OTTAWA, September 30, 2026 | Canada’s economy stopped growing in July. Statistics Canada reported on Monday, September 29 that real gross domestic product by industry was essentially unchanged from June, ending a three month run of expansion and delivering the first clearly negative signal in a year that had, until the summer, held up better than most forecasters expected.
The composition of the flat reading matters more than the headline. Construction rose 1.3 per cent and utilities gained 1.7 per cent, the latter driven substantially by cooling demand during a hot July. Those gains were offset by declines in the goods producing sectors most exposed to trade. Manufacturing fell 0.9 per cent, with petroleum refinery output down 6.2 per cent. Mining, quarrying and oil and gas extraction fell 0.5 per cent, a second consecutive monthly decline. Retail trade also weakened.
The critical point for anyone reading the number as a tariff indicator is chronology. July ended before the United States imposed 50 per cent duties on roughly US$20 billion of Canadian goods on August 22, before Canada’s counter-tariffs on approximately C$27.6 billion of American goods took effect on September 8, and before the American import prohibition on Canadian alcohol, dairy and motorcycles came into force on September 29. The July data measure an economy that had already lost momentum without the heaviest measures having landed.
Statistics Canada’s advance estimate points to growth of 0.2 per cent in August, a figure the agency will revise next month and one that captures only the first nine days of the 50 per cent duty regime.
Reading the industry accounts properly
Monthly gross domestic product by industry is a supply side measure. It adds up value added across roughly twenty industry groupings and is published with a two month lag, which makes it the most timely broad measure of Canadian output and also the one most frequently over-interpreted.
Two cautions apply to the July release. The first is that a flat headline can conceal substantial rotation beneath it, and in this case it does. A 1.3 per cent construction gain and a 1.7 per cent utilities gain are large monthly moves, and both are sectors with limited direct trade exposure. The goods producing sectors that carry Canada’s export relationship moved the other way. An economy in which the trade exposed portion contracts while the domestically oriented portion expands is not a stable economy, even when the two effects happen to cancel in a given month.
The second caution concerns the utilities gain specifically. Cooling demand during a hot July is weather, not activity. It will reverse. Stripping it out leaves a headline that would have been modestly negative, which is the number most private forecasters are actually working from.
There is a third, more technical point. Statistics Canada’s advance estimate for the following month is produced from a narrower information set than the full release and is revised more often than not. Treating the 0.2 per cent August advance as a settled fact would be a mistake in an ordinary month and is a larger mistake in a month containing a structural break in trade policy.
What the economists are saying
Marc Ercolao, an economist at TD Bank, characterised the stall as consistent with the broader volatility that has marked Canadian growth through 2026, noting that production disruptions unrelated to trade accounted for some of the July weakness. That framing cautions against reading the single month as a structural break.
The more pointed commentary concerns what comes next. Peter Shannon, a senior economist at KPMG, said the tariffs will meaningfully hit in September with the full effect arriving in the fourth quarter, and estimated the drag at roughly half a percentage point of annual gross domestic product. KPMG Canada has published a range of 0.3 to 0.5 percentage points over the coming year.
Private sector forecasters have been trimming. BDC Economics reduced its 2026 growth projection from 1.0 per cent to 0.9 per cent. The revisions are modest in isolation and collectively describe an economy expected to grow, but barely, through a period in which its largest trading relationship is being restructured against its will.
Several economists have made the point that confidence may matter almost as much as the tariff bill itself. A firm that cannot forecast its landed cost into its principal export market defers capital spending and hiring regardless of what the current duty rate happens to be. That channel does not show up cleanly in any single line of the industry accounts, and it operates with a lag that outlasts the measures that triggered it.
The second quarter comparison
The July reading is jarring in part because of what preceded it. Canada posted quarterly growth of 0.8 per cent in the second quarter, an annualised rate of 3.3 per cent, a figure strong enough to have prompted questions about whether the economy needed further monetary support at all.
Part of that second quarter strength was mechanical. Importers and exporters on both sides of the border pulled shipments forward ahead of announced tariff deadlines, inflating trade volumes in a way that borrowed activity from subsequent quarters. Inventory accumulation ahead of the August measures had the same effect. Economists warned at the time that the strength was partly artificial and that payback would follow.
July looks like the beginning of the payback. Coverage of the release noted merchandise exports down 2.3 per cent and a sharp narrowing of the merchandise trade surplus, from roughly C$4.2 billion to under C$800 million, a swing consistent with the front running thesis and with the initial effect of retaliatory measures on both sides.
The labour market has not yet broken
The clearest counterweight in the data is employment. Canada added roughly 75,000 jobs in July and the unemployment rate stood at 6.4 per cent, a labour market that remains functional by any historical standard and considerably better than a flat GDP print would normally imply.
The divergence between output and employment is a familiar pattern at the start of a trade shock. Firms facing an uncertain but potentially temporary disruption hold onto skilled workers rather than incur the cost of rehiring later. Labour hoarding compresses measured productivity and delays the employment effect of an output decline, and it typically breaks when management concludes the disruption is structural rather than cyclical.
This week supplied evidence that the conclusion is being reached in some quarters. Stelco announced it will indefinitely idle finishing operations at its Hamilton Works and lay off approximately 350 members of United Steelworkers Local 1005, citing American tariffs and weak coated steel pricing. Announcements of that kind are the mechanism by which a trade measure migrates from the trade accounts into the labour force survey.
The Bank of Canada’s problem
The Governing Council meets on October 28, and the July data have not made the decision easier.
Deputy Governor Toni Gravelle has described the situation facing policymakers as a true dilemma. The conflict is straightforward to state and difficult to resolve. Weakening output and a trade shock concentrated in the goods producing sectors argue for monetary support. Persistent inflationary pressure, driven in part by elevated oil prices, argues against it. Tariffs themselves push in both directions at once, raising the price level while reducing demand, which is the classic supply shock problem that monetary policy is least well equipped to address.
The Bank has noted that the economy retains excess supply, meaning capacity and labour are not being fully used. That assessment supports the case for easing. Against it sits the risk that a central bank cutting into a tariff driven price increase validates the increase and complicates the return to target.
Markets will read the August advance estimate of 0.2 per cent, the September data when they arrive, and the October inflation print as the relevant inputs. The Bank has been explicit that it is not prepared to look through a trade shock of uncertain duration, which in practice means a preference for waiting until the fourth quarter data clarify the magnitude.
Where the damage is concentrated
The aggregate figures obscure how unevenly the tariff effect is distributed.
Manufacturing is the primary transmission channel, and within manufacturing the exposure is concentrated in Ontario and Quebec. Steel, aluminum, automotive, machinery and fabricated metal products account for a disproportionate share of Canadian merchandise exports to the United States, and all sit under American measures. The 0.9 per cent July manufacturing decline predates the heaviest of those measures.
Energy is a different story with a similar outcome. The 6.2 per cent drop in petroleum refinery output reflects maintenance and operational factors more than trade policy, but the second consecutive decline in mining and extraction points to softening volumes in a sector where Canadian producers have limited pricing power and where the American refining complex remains the dominant customer.
Agriculture and food processing occupy an unusual position. The American measures have hit dairy and beverages hard, while Canada’s January arrangement with China delivered meaningful relief in the other direction, cutting the Chinese duty on Canadian canola seed from 75.8 per cent to 15 per cent effective March 1 with further reductions on canola meal, peas, lobster and crab scheduled through the end of 2026. For prairie agriculture the net trade position in 2026 is better than it was in 2025, which is not true of almost any other Canadian export sector.
Services, which account for the larger share of output, have been comparatively insulated. That insulation is one reason the aggregate number is flat rather than negative, and one reason the headline understates the severity of what is happening in the goods economy.
The regional picture
National aggregates flatten a set of regional experiences that diverge sharply, and the divergence is widening as the tariff programme matures.
Ontario carries the heaviest exposure. The province’s manufacturing base is concentrated in automotive assembly and parts, steel, machinery and fabricated metals, every one of which sits under an American measure. The Stelco idle announced this week in Hamilton is a visible instance of a broader compression running through the southern Ontario industrial corridor. Ontario also absorbs the largest share of the counter-tariff burden on the import side, since the province accounts for the bulk of Canadian imports of American machinery, appliances and steel derivative products.
Quebec’s exposure runs through aluminum, aerospace, forest products and, as of this week, recreational vehicles. The American prohibition on Can-Am Spyder and Canyon motorcycles built by Bombardier Recreational Products closes the dominant market for a Quebec manufactured product line, and the province’s aluminum smelters face American duties on a commodity for which the United States has no realistic domestic substitute at volume.
The prairie provinces are the exception. Agricultural exports to China have improved materially under the January arrangement, energy volumes remain the dominant driver of provincial output, and the manufacturing exposure that is hurting central Canada is comparatively limited. Saskatchewan in particular has seen the canola trade move from a 75.8 per cent Chinese duty to 15 per cent within a single year.
British Columbia occupies a middle position, with softwood lumber under longstanding American duties, a significant Asia Pacific trade orientation that provides partial insulation, and a services heavy urban economy that has continued to grow. Atlantic Canada’s exposure is concentrated in seafood, where Chinese tariff relief on lobster and crab offsets part of the American pressure.
The policy consequence of that divergence is that a single national response will fit no region well. The C$1.5 billion Regional Tariff Response Initiative is delivered through the regional development agencies precisely to allow for that variation, and how the funds are allocated across regions will be one of the more closely watched fiscal decisions of the fourth quarter.
How this compares to earlier trade shocks
Canada has absorbed trade shocks before, and the historical comparisons offer limited comfort in either direction.
The softwood lumber disputes, which have recurred across four decades, demonstrated that a sector can operate for years under punitive American duties without disappearing, sustained by currency movements, cost discipline and eventual partial refunds. They also demonstrated that the adjustment is paid for in mill closures and community dislocation long before any legal resolution arrives.
The 2018 and 2019 Section 232 steel and aluminum tariffs are the closer precedent in mechanism, but the difference in magnitude is decisive. Those measures were set at 25 per cent on steel and 10 per cent on aluminum, were lifted within roughly a year after negotiation, and did not extend to alcohol, dairy, vehicles or an outright prohibition on entry. The current programme is broader, higher, and rests on a statutory authority that contemplates escalation to exclusion rather than capping at a duty.
The most relevant historical lesson is about duration rather than depth. Trade disputes between Canada and the United States tend to outlast the political cycles that create them, and firms that plan on the assumption of rapid resolution have consistently been wrong. The planning horizon that matters for Canadian exporters is years, not quarters, and the July GDP figure is the first month of what is likely to be a long series.
What it means for importers and exporters
For Canadian exporters, the July data confirm what the order books already showed and offer a planning discipline rather than news. The measured effect of the August and September measures has not yet appeared in the national accounts, which means that any internal forecast anchored to published macro data is, by construction, looking at a picture that is at least two months stale. Firms should be building from their own shipment and order data rather than from the published series.
The front running effect deserves particular attention in demand planning. Customers on both sides of the border pulled forward purchases in the second quarter. Inventory built in that period is still working through the channel. An exporter reading a soft fourth quarter order book should be distinguishing between demand destruction and inventory digestion, because the two call for very different responses.
For Canadian importers of American goods, the counter-tariff schedule that took effect on September 8 covers 629 tariff codes at 15, 25 and 50 per cent, touching steel, dairy, appliances, agricultural equipment, pulp and paper and electronics. Classification and valuation exposure has increased materially, and the Canada Border Services Agency has signalled an enhanced compliance posture, particularly on steel and derivative products. Importers who have not reviewed their tariff classification since the summer are carrying an unquantified liability.
For firms of any size, the support architecture remains the C$7.5 billion package announced on August 25, comprising a C$1.5 billion Regional Tariff Response Initiative for small and medium enterprises, a C$500 million liquidity stream through the Business Development Bank of Canada, a C$2 billion Canada Strong Diversification Fund for capital projects opening non-American markets, and C$3.5 billion in Rapid Response Supports covering expanded employment insurance, retraining and a new Worker Retention and Retraining Program. Uptake has been slower than the government projected, in part because the application processes were designed before the scale of the disruption was clear.
The policy backdrop
Prime Minister Mark Carney said on Monday that he has no plans to apply further trade pressure on the United States as the American midterm elections approach, while adding that he would never rule anything out and that Canadian timing is not dictated by any specific aspect of the American political calendar.
That posture has implications for the economic outlook. A government that is not planning further escalation is signalling that the current tariff structure is the planning assumption through at least the fourth quarter, which is useful information for firms trying to size their exposure. It also means that the next meaningful change in the trade environment is likely to come from Washington, from the courts, or from the joint review of the Canada United States Mexico Agreement rather than from Ottawa.
The government’s stated strategic objective is to double non-American trade within a decade, pursued through deeper association with the European Union, a comprehensive economic partnership agreement with India targeted for conclusion around the December G20 meetings, and the arrangement with China concluded in January. None of those files changes the fourth quarter arithmetic. All of them change the shape of the Canadian economy that emerges from it.
What to watch
Three data points will define the next two months.
The August GDP figure, published in late October, will be the first reading that contains a meaningful stretch of the 50 per cent duty regime. The advance estimate of 0.2 per cent is a starting point, not a forecast, and the revision will be informative in itself.
The Bank of Canada’s October 28 decision will reveal how the Governing Council is weighting the supply shock against the demand shock. A cut would signal that the output risk has become the dominant concern. A hold would signal that the Bank intends to wait for fourth quarter data before committing.
And the September trade figures, when they arrive, will show the first full month of Canadian counter-tariffs alongside the American measures. Those numbers will tell exporters and importers more about their actual operating environment than any forecast currently available to them.
Beyond those three markers, the longer question is whether the flat July reading marks the top of a cycle or the bottom of a dip. The honest answer is that the data cannot yet distinguish between the two, because the measures that will determine the outcome had not taken effect when the observations were collected.
What the release does establish is that Canada entered the heaviest phase of the trade conflict without momentum. An economy growing at a 3.3 per cent annualised rate has room to absorb a half point drag. An economy growing at zero does not. That is the substance of the deputy governor’s dilemma, and it is the reason a flat month has attracted attention out of proportion to the size of the number itself.