There has been a clamour of voices recently expressing concern that bubbles could be forming in several corners of the financial market. The most widely cited concerns pertain to U.S. stock market valuations, the sustainability of artificial intelligence capital expenditures (cap ex), and the stability of the private credit market. What follows are a few musings on each through an economic lens.

Stock market concerns

While the U.S. stock market remains near record highs, there has been notable jitteriness in recent weeks (see next chart). Much of the market’s strength over the past several years has been wrung from a handful of mega-cap technology stocks. In turn, the question of whether they can continue to ascend or even hold their current valuations will be determined by whether AI truly blossoms into the next general-purpose technology that propels economic prosperity and corporate earnings sharply higher over many years. On the whole we are fairly optimistic about this prospect, though valuations are admittedly high and the portfolio concentration risk is considerable.

As such, our instinct has been to incrementally reduce exposure to the U.S. market after such remarkable gains, though certainly not to abandon it altogether. The tangle of cross-investments now being made between the major players makes it increasingly difficult to properly evaluate the individual merits of each country.

Critically, there are important differences between this run-up and the stock market bubble of the late 1990s. The companies of today are, for the most part, highly profitable and their business models are not as flimsy as many of the darlings of the late 1990s. Valuations are similarly not as elevated.

Furthermore, even in a scenario in which a bubble does form, timing is important. Tongues began wagging about a possible bubble in 1996 when then-Fed Chair Greenspan spoke of “irrational exuberance.” While he was not wrong, the stock market continued to rise from that point for another three-plus years, with the S&P 500 doubling over that time frame. This is to say that even when stock market bubbles are successfully identified, they are capable of extending for years.

Equity volatility reveals anxiety in markets

As of 10/31/2025. Sources: S&P Global, Macrobond, RBC GAM

Artificial intelligence cap ex concerns

Second, and relatedly, there are concerns about the sustainability of the enormous cap ex currently going into artificial intelligence. The Magnificent 7 companies alone are estimated to be spending US$373 billion on this in 2025, and even larger sums in the years to come (see next chart).

Magnificent 7 cap ex growth is expected to slow over the coming years

As of 10/31/2025. Sources: Bloomberg, RBC GAM

We would note that analysts forecast decelerating AI cap ex growth over the coming few years, meaning that the economic growth tailwind should become less powerful in future years than it was in 2025. Still, there is a difference between that and claiming that widespread malinvestment has occurred – the foundation for any kind of “bubble” argument.

The effectiveness of the AI investment remains to be determined and is massively sensitive to nearly unknowable considerations. For example:

• How quickly will demand for AI services rise in the future? 10% per year? 100% per year?

• How rapidly will the fantastically expensive computer chips powering data centres depreciate?

• Will AI make the leap from being good at summarizing the internet to being able to reach novel and revolutionary conclusions/discoveries all by itself?

For the moment, major tech companies are in a sprint to ensure they aren’t left behind. And even if the spending eventually proves to have been inefficiently high, it is unlikely that the companies will stop or that this will even become clear within the next year or two. Fortunately, most such companies have robust profits from other business units and the fraction of their free cash flow being dedicated toward AI investments is not distressingly high.

Private credit

It is difficult to speak intelligently about the private credit market given its opacity. It is certainly not ideal that such lenders are anecdotally becoming less discerning in their lending. Fascinatingly, the total sum of non-bank lending in the U.S. economy has actually been falling as a share of gross domestic product (GDP) over the past five years (see next chart). This is admittedly a broad metric, but one that encapsulates the sort of underregulated and higher risk lending that occurs in the private credit space.

Furthermore, while it is a bigger industry than a decade ago, the difference is smaller than you might imagine – merely up from 21% to 25% of GDP. And it wasn’t much smaller in the early 1990s than it is today.

U.S. non-bank loans have declined since the pandemic

As of Q2 2025. Shaded area represents recession. Sources: U.S. Bureau of Economic Analysis (BEA), the U.S. Federal Reserve, Macrobond, RBC GAM

While the lack of liquidity in private credit is often cited as an argument for the potential fragility of the sector, in actual fact it acts more like a shield, rendering the industry less vulnerable to sudden changes in sentiment of the sort that might sink publicly traded securities.

There have been two recent bankruptcies of some prominence in the U.S., both with a connection to shadow banking. While some fraction of their troubles may reflect excesses in that space, their woes can also be traced back to more idiosyncratic drivers such as alleged fraud, changing immigration policy and the imposition of tariffs on the North American auto sector. The fact that credit spreads remain narrow in public markets argues that the broader credit environment is relatively benign.

None of this is to say that these three areas do not have their questions or even problems, but it is not certain that they represent sizeable bubbles.

-EL

The U.S. government shutdown continues and is on the cusp of breaking the record for longest such episode (prior record: 35 days in 2018—2019). Betting markets now believe there is just under a 50% chance that the shutdown is resolved by mid-November, and a greater than 80% chance it is resolved by the end of November.

Politicians remain unwilling to compromise. Republicans are not averse to the “less government” that results from the shutdown, while Democrats are using what limited leverage they have to try to reverse health spending cuts.

Given this, what might bring a resolution? Rising frictions for the average American. And what might prompt this? Several things.

Unhappy essential government workers are toiling without pay (at least until the shutdown is resolved and they are issued back pay). Already, air traffic controllers and Transportation Security Administration (TSA) workers appear to be slowing down the pace of their work in protest of their missing paycheques, causing airport delays.

The federal government stopped funding food stamps around November 1. Some states are stepping in to fill the financial hole left by the federal government, but not all. A federal judge has ordered that the federal money resume flowing, but it is unclear whether the Department of Agriculture can or will comply. A remarkable 41.7 million Americans rely on this program – about 12% of the population. Hunger can be a powerful force.

The Affordable Care Act registration window opened on November 1. With diminished health care subsidies – a key concern for Democrats withholding their support for the budget – discontent with politicians may increase, amplifying pressure.

State and local elections on November 4 will reveal whether voters place the blame for the shutdown on one party or the other, potentially increasing pressure for a compromise solution.

We continue to believe the U.S. economy is about 1.0% to 1.25% smaller than it would otherwise be during the shutdown period. On the assumption that the shutdown is resolved around the middle of November, we have subtracted 1 percentage point from Q4 annualized GDP growth, with the first quarter of 2026 then reclaiming that loss.

-EL

With the U.S. government shutdown now in its second month, the list of delayed economic data releases has swelled to 28, including key payroll and GDP figures. A slew of alternative indicators – including some that previously received little attention – is helping to fill that gap.

Alternative labour market data is among the most prevalent:

• ADP released its usual monthly private sector payroll report for September, which showed an unexpected 32,000 net decline as well as downward revisions to prior months. The payroll processor also launched a new weekly employment estimate, suggesting an average of 14,000 jobs were added in the four weeks to October 11 – a monthly pace of about 60,000.

• While national initial jobless claims releases are delayed, our aggregation of state-level data shows a fairly steady trend in new unemployment insurance filings in October (see chart). Combined with the ADP data, this suggests a low-churn labour market persists (few firings but limited hiring). A further decline in Indeed job postings also points to slowing demand for labour.

• Revelio Labs, which uses publicly available social media and job posting data to proxy non-farm payrolls, estimates 60,000 jobs were added in September. Revelio doesn’t have a particularly strong track record predicting payroll releases – over the past three years its monthly estimates have been off by about 100,000 on average.

• The Chicago Fed’s real-time unemployment rate forecast for October was 4.35%. This is little changed from its September estimate but up 0.2 ppts from the latest U.S. Bureau of Labor Statistics (BLS) release for August. The quality of that estimate is impaired by lack of some official data inputs and doesn’t fully capture the increase in unemployment associated with furloughed federal workers. Bloomberg produces its own unemployment nowcast which is in line with the Chicago Fed estimate.

U.S. jobless claims remain low

RBC GAM estimates, as of the week ending 10/25/2025, based on latest state level data available. Actual claims data as of the week ending 09/20/2025. Sources: U.S. Department of Labor, Haver Analytics, RBC GAM

Many closely watched economic surveys are released privately or by the Federal Reserve, which is independently funded. These sources are thus unaffected by the government shutdown:

• The Fed’s October Beige Book suggested economic activity was little changed, on balance, with consumer spending inching down in recent weeks. The economic outlook is mixed, but many respondents expect elevated uncertainty to weigh on activity. Overall, our Beige Book sentiment indicator deteriorated slightly with the latest release (see chart).

• Regional Fed surveys point to mixed manufacturing conditions across districts, while the more comprehensive Institute for Supply Management (ISM) manufacturing index surprised to the downside in October and remained in contractionary territory for an eighth straight month. The ISM services index slipped to 50 in September – suggesting no change in economic activity – with the October edition to be released in the coming days. S&P’s U.S. Purchasing Managers’ Index (PMIs), which aren’t watched as closely as ISM, point to some improvement in October and stronger activity overall.

• University of Michigan and Conference Board consumer confidence surveys both show softening sentiment in October. The Federal Reserve of San Francisco’s Daily News Sentiment Index also weakened on net in the past two months but improved slightly toward the end of October.

Beige Book Sentiment Indicator is weakening

As of October 2025. The indicator quantifies the sentiment of local contacts by assigning different weights to a spectrum of positive and negative words used to describe overall economic conditions in the Fed Beige Book. Sources: U.S. Federal Reserve, RBC GAM

BLS brought back some employees to compile and release September Consumer Price Index (CPI) data, which was needed to calculate cost of living adjustments for social security benefits. But the White House warned that inflation data might not be collected for October. That could be a negotiating tactic, but if the next round of CPI is delayed, we’ll be watching alternative indicators like the PriceStats daily inflation index, which shows tentative signs of CPI flattening out in October (see chart).

U.S. Daily PriceStats Inflation Index shows signs of levelling

PriceStats Inflation Index as of 10/31/2025. CPI as of September 2025. Sources: State Street Global Markets Research, RBC GAM

Anonymized credit and debit card transactions help give a sense of the direction in consumer spending in the absence of retail sales and Personal Consumption Expenditures (PCE) data. Bloomberg’s tracker, which covers more than 20 million customers, points to some loss of spending momentum in October (see chart). Bank of America’s card spending data tells a similar story. Softening consumer spending makes sense, in our view, as unpaid federal employees are likely to cut back on non-essential purchases – on top of the weakening consumer sentiment noted above.

Card spending data suggests some loss of consumer momentum

As of 10/27/2025. Sources: U.S. Bureau of Economic Analysis (BEA), Macrobond, RBC GAM

Getting a sense of broader economic growth is more challenging as many of the inputs into GDP estimates haven’t been released. The Federal Reserve of Atlanta has updated its GDP nowcast based on private sector releases like existing home sales and ISM surveys, but we think its 3.9% annualized growth estimate for Q3 is somewhat stale and could be subject to revision once official data is released. For its part, the Federal Reserve of Dallas’s weekly economic index points to a slight loss of momentum in September and October.

Our overall interpretation of the alternative data is that there hasn’t been a major break in pre-shutdown trends, with the labour market in particular continuing to gradually soften. Given the Fed’s focus on the employment half of its mandate, one might have assumed that would be enough for the Fed to continue cutting rates in December. But Chair Powell suggested otherwise, noting the lack of official data – if it persists – could make the Fed more cautious about moving, using the metaphor of slowing down when driving in the fog.

That might be a moot point, with betting markets suggesting the government shutdown will be resolved ahead of the Fed’s next meeting on December 9-10. It will take time for statistics agencies to get back in gear and release delayed data, but we think it’s more likely than not that some key indicators will be available for that gathering. We continue to lean toward the Fed cutting once more by year end.

-JN

U.S.-China deal

Renewed antagonism culminated in the U.S. threatening an additional 100% tariff on China in October. However, the two countries made a breakthrough in recent days. The U.S. has instead cut its tariffs on China by about 10 percentage points, from a weighted average rate of 42.6% to 32.0%. China agreed to export more rare earths and to buy more soybeans, while the U.S. promised to export more computer chips.

It remains to be seen how long this accord lasts, as détentes reached earlier in 2025 did not survive for long. We continue to assume that the world’s two great powers will remain broadly frosty toward one another, that tariff rates will not significantly decline from here, and that specific export and import promises may not be fully delivered.

Supreme Court to hear tariff challenge

The U.S. Supreme Court will hear the challenge to IEEPA tariffs (loosely, the tariffs that have been applied at the country rather than sector level) starting on November 5. A verdict is expected by the end of the year. Betting markets continue to think it is more likely than not that the IEEPA tariffs are overturned. We are doubtful that the U.S. government would have to reimburse the parties that previously paid such tariffs, however.

The White House has several other tariff laws at its disposal. Section 122 tariffs could rapidly (if temporarily) replace IEEPA tariffs. Section 301 tariffs are capable of more enduringly replacing them after a period of investigation. All of this is to say that we would assume the White House can still achieve its aims after a brief period of chaos if IEEPA tariffs are indeed repealed.

U.S.-Canada squabble

After Ontario ran a television advertisement in U.S. markets that was critical of tariffs, the White House cut off trade negotiations with Canada and threatened a 10% tariff increase.

It is unclear what the 10% tariff hike refers to – whether on all Canadian products or just non-USMCA-compliant products. We think the latter, which would bring that rate up from 35% to 45%. But in practice the U.S. has not actually implemented the change, and it would have only minimal impact on Canada given that few products entering the U.S. are not USMCA compliant.

The two countries had been close to a deal on steel and aluminum trade. This is now on hold, but we expect it will resume after tensions cool. It sounded as though Canada was negotiating lower tariffs on steel and aluminum products up to a certain quota level. Presumably, U.S. steel and aluminum exports into Canada would enjoy a similar reprieve.

An energy component to the prospective deal was also reported. It is quite promising that the U.S. was willing to ease restrictions, as Canada has otherwise made little headway. The big negotiations will be the broader USMCA deal, which will presumably happen at some point in the next nine months. We discussed USMCA scenarios in the last #MacroMemo.

New sector tariffs

U.S. forestry tariffs took effect in mid-October. Truck tariffs began on November 1. On the whole, their introduction has had a surprisingly minimal effect on the average U.S. tariff rate (see next chart), though the consequences are nevertheless acute for the affected sectors.

Average U.S. tariff rate is little changed at 17.5%

Effective tariff rates estimated based on tariffs in effect as at the specified date and up to 11/01/2025; threatened rates not included. Excludes the de minimis effect – suspension of de minimis exemption for China and Hong Kong in May 2025 and effective 08/29/2025 for all other countries. Expected tariff assumes instantaneous and complete implementation, i.e., does not account for shipping delays, implementation lags, etc. Sources: Evercore ISI Tariff Tracker, International Monetary Fund, Macrobond, RBC GAM

South Korea and Japan

During President Trump’s recent trip to Asia, deals with South Korea and Japan were also reached.

The U.S. struck an enhanced deal with South Korea, reducing the country’s blanket tariff rate from 25% to 15%. For its part, South Korea agreed to several hundred billion dollars of investments in the U.S., with greater clarity over where the money will flow.

Japan also negotiated an expanded deal, building upon the earlier accord that saw Japan agree to a 15% tariff and make a plan to invest US$550 billion into the U.S. There is now greater clarity over where that money will go – largely toward supporting Japanese corporate projects within the U.S.  – and there is a stipulation that the investments must be made by the end of President Trump’s term.

There is also a new agreement to cooperate on new-generation nuclear power reactors (Japan will build these in the U.S. using U.S. technology), and similarly to cooperate in expanding rare earth production.

-EL & SK

Economists have been searching for a letterform description of the post-pandemic recovery. Is it a quick V-shaped rebound? A long U-shaped trough? A period of L-shaped stagnation? A W-shaped double dip?

Many economists have settled on a K-shaped recovery, evoking an uneven rebound with diverging trajectories among industries, businesses and household income cohorts. While the recovery has run its course, we still talk about a K-shaped economy in which certain groups are thriving while others struggle. This is particularly evident among households. How do we reconcile resilient consumer spending and rising household wealth with higher delinquency rates on consumer loans and growing food insecurity?

Several factors are supporting higher income households while headwinds for lower income Americans persist or even intensify:

• The post-pandemic surge in inflation disproportionately impacted lower-income consumers who spend more of their paycheques on necessities and have limited flexibility to reduce spending. While the rate of inflation has largely normalized – although not all the way to the Fed’s 2% target – the price level hasn’t returned to its pre-pandemic path. The cost of living remains elevated, representing an ongoing challenge for lower-income households.

• Congress and the White House recently extended tax cuts that disproportionately benefit higher income households, paid for in part by spending cuts that hurt lower income Americans. The Congressional Budget Office (CBO) estimates the One Big Beautiful Bill Act reduces resources for the lowest income households by 3% of income while adding to top earners’ resources by a similar amount (see chart).

• On top of a regressive tax bill, tariffs also disproportionately hurt lower income consumers who spend relatively more on goods. The Yale Budget Lab estimates tariff costs represent 3.5% of earnings for the lowest income households, while the highest income cohort sees less than one-third of that hit.

• Strong equity market gains – the S&P 500 has returned 21% over the past year and 17% on average over the past 5 years – have largely accrued to the wealthiest Americans (see chart). The top 1% of households by net worth account for half of corporate equity and mutual fund share holdings, while the next 9% of wealthiest households own 37%.

• Today’s higher interest rate environment benefits savers at the expense of borrowers. Lower income households are in the latter camp with the bottom half of the income distribution spending more than it earns. The least wealthy 50% of households account for slightly more than half of consumer credit outstanding but have limited interest-bearing assets.

• A softening labour market, with particularly challenging conditions for new entrants and job losers, hits lower income households harder. Wealthier households tend to be more reliant on non-wage sources of income.

• Wages at the lower end of the spectrum (25^th percentile) grew faster than the higher end (75^th percentile) for much of the past decade, but that relationship has flipped over the past year with higher earners now seeing faster wage gains (see chart). At 3.6% year-over-year and slowing, the current pace of wage growth at the 25^th percentile is only slightly ahead of 3% inflation, implying limited real wage gains.
  

Recent policy changes disproportionately hurt lower income Americans

As of 09/03/2025. Sources: Yale Budget Lab, Congressional Budget Office, RBC GAM

The richest Americans have disproportionately benefited from stock market gains

As of Q2 2025. Sources: U.S. Federal Reserve, Macrobond, RBC GAM

Lower-income wages are growing more slowly than higher-income

As of 09/11/2025. Sources: Federal Reserve Bank of Atlanta, RBC GAM

The K-shaped economy is also evident in the corporate world, with AI at the centre of diverging trajectories:

• The biggest companies are getting bigger. Since the S&P 500 hit its cycle low in October 2022, the cap-weighted index has outperformed the equal-weight index by nearly 30%. This is evidence of increasing concentration among the largest names. Similarly, the S&P 500 has outperformed the Russell 2000 small-cap index by around 30%.
  
  

• Within the S&P 500, AI-related companies are seeing outsized share price gains. Goldman Sachs’ basket of AI tech, media and telecom stocks is up 250% over the past three years, while AI data centres aren’t far behind with a 225% gain. Meanwhile, large-cap tech stocks outside the AI space have only gained half as much as the S&P 500, and companies at risk from AI are close to flat (see chart). Outside of the Magnificent 7, profit margins for the rest of the S&P are down in 2025 due in part to tariff costs.
  

• AI is a significant driver of business cap ex. Computer and software investment fully accounts for net growth in private, non-residential fixed investment over the past year. Data centre construction has also contributed, growing by one-third on an inflation-adjusted basis. We think AI-related cap ex might be adding something in the range of one-half percent to U.S. GDP growth this year.
  

• The job market impacts of rising AI adoption are only gradually emerging – as discussed in a recent #MacroMemo. A growing number of companies have suggested AI-related efficiencies will limit their need to hire in the coming years, and in some cases result in layoffs.

AI equity themes are outperforming while non-AI underperforms

As at 10/20/2025. Sources: Goldman Sachs, Bloomberg, RBC GAM

The upshot of a K-shaped economy is growing dependence on the spending of wealthier households and rapid growth in AI-related cap ex. The two are connected, with AI-led equity market gains driving positive wealth effects for richer consumers. The top 10% of earners now account for nearly half of consumer spending. This is a record share dating back to the late-1980s.

Relatively lofty equity market valuations driven by AI exuberance don’t mean a correction is imminent. But any significant drawdown could reverse the wealth effects that have underpinned resilient consumer spending. Even if equity markets remain resilient but upward momentum fades, wealth effects will diminish over time.

Beyond the vulnerability associated with narrowly based economic growth, a K-shaped economy has political implications. There’s the potential for inequality to breed discontent and social unrest – and further exacerbate political divisions. Inequality itself is a partisan issue in the U.S., with a recent PEW study finding 76% of liberal voters think inequality is a very big problem, compared with just 30% of conservatives. That’s the largest left-right gap among 15 countries surveyed.

While there is risk that the upper half of the K falters, there is also potential for headwinds buffeting the lower half to ease. We don’t foresee a near-term reversal of regressive tax and tariff policies. But if inflation cools more than expected and gives the Fed scope for additional rate cuts, that could ease both the cost-of-living challenge and debt servicing burden for lower income households.

Our forecast anticipates neither an upper-K downturn nor a lower-K resurgence. But we budget for some softening in consumer spending going forward as the economy’s reliance on wealthier households’ spending becomes more challenging. Nonetheless, we think the U.S. economy will continue to grow around a 2% pace next year with an ongoing (albeit less prominent) tailwind from AI cap ex, growing fiscal support and waning tariff effects as 2026 progresses.

-JN

China’s economy continues to grapple with deflation. This is most obvious in factory-gate or producer prices. They have been falling at an average annualized rate of 3% since mid-2022 and were down 2.3% year-over-year in September. Declines are widespread with 12 of 15 core industrial groups seeing lower prices relative to a year ago.

Consumer prices haven’t fallen as consistently and are close to flat over the past three years, albeit with intermittent declines including a slight 0.3% dip over the past year. Consumer deflation is more narrowly based, led by lower food and fuel prices. Core inflation has accelerated to 1% year-over-year in September (see chart).

Chinese producer prices are falling while consumer prices are close to flat

As of September 2025. Sources: China National Bureau of Statistics (NBS), Macrobond, RBC GAM

As a reminder, while falling prices might sound good for consumers, they can have a pernicious effect on the economy. Declining producer prices are eroding profit margins (see chart), with the ranks of loss-making manufacturers growing by more than one-third over the past three years. Anticipating lower future prices, households and businesses might delay purchases, resulting in a negative feedback loop of job losses and additional spending restraint.

Falling producer prices are dragging down margins

As of September 2025. Sources: China National Bureau of Statistics (NBS), Macrobond, RBC GAM

That is not to suggest China is experiencing a destructive deflationary spiral. Real GDP grew at a solid 5.2% pace year-to-date through Q3, albeit with some loss of momentum recently. Rising core inflation suggests limited risk of households putting off purchases to wait for lower prices. It’s hard to delay buying food and fuel, and lower prices for such essentials can stimulate spending elsewhere. Still, persistent producer price deflation is uncharacteristic of a healthy economy, and policymakers have reason to be concerned.

Several factors are contributing to falling factory-gate prices and lack of consumer price inflation:

• Spare capacity: China’s use of industrial capacity has fallen back to levels last seen in 2015-16 (pandemic aside). At that time the government’s supply-side reforms curbed over-production and put an end to persistent producer price deflation (see chart). Sectors with the greatest increase in spare capacity – including autos, pharmaceuticals, non-metallic mineral products and coal – are also seeing some of the most intense producer price deflation.

• Involution: This term typically refers to intense social competition but has been adapted to describe price wars among Chinese firms. This race-to-the-bottom dynamic is particularly evident in cleantech sectors like electric vehicles (EVs), batteries and solar panels – emerging industries in which China has sought to become a dominant global supplier. The same is true of some heavy industries like steelmaking and cement. EVs have become the posterchild for involution, with a growing number of domestic firms pushing down prices and profits – and prompting government rebuke.

• Weak property sector: Housing costs make up one-quarter of China’s CPI basket. These costs have been close to flat in recent years amid an ongoing downturn in China’s property market. Real estate is a key savings vehicle for Chinese households, so beyond restraining housing inflation, falling property prices are generating negative wealth effects and weighing on demand and consumer prices more broadly.

• Lower food and energy prices: Like other developing economies, food carries a relatively large weight (nearly one-third) in China’s CPI basket. Prices of staples like pork, eggs and fresh vegetables are all down by more than 10% year-over-year, due in part to fading base effects from previous price increases. As is the case globally, falling oil prices are a disinflationary force in China.

Excess capacity is weighing on China’s factory-gate prices

As of July 2025. Sources: China National Bureau of Statistics (NBS), Macrobond, RBC GAM

China successfully fought producer price deflation in the mid-2010s by imposing production cuts on industries with excess capacity – most notably steel and coal. and the country also encouraged mergers to dampen competition. Doing so this time around could be more challenging. The private sector (rather than state-owned enterprises) will play a greater role in many of the industries saddled with excess capacity or experiencing intense price competition.

Still, we think an anti-involutionary push will act as a modest headwind to China’s economy in the near-term. Chinese GDP growth slowed by about one-half percent in 2015-16 as supply-side reforms weighed on industrial production and fixed investment, with demand-side stimulus softening the blow somewhat.

To the extent that Chinese deflation reflects soft demand in the world’s second-largest economy, it could be a disinflationary force globally. However, as noted above, consumer price deflation is narrowly based and largely reflects idiosyncratic price declines (see chart). China is a net importer of food and agricultural products, so falling prices of some domestic staples shouldn’t spill outside its borders. Chinese demand is an important factor in global energy prices, but recent declines in oil prices are arguably more reflective of supply. China’s property sector downturn and flat housing prices are largely a domestic story.

As at 11/03/2025. Sources: China National Bureau of Statistics, Macrobond, RBC GAM

There has been some concern that China might find an outlet for overcapacity by flooding other countries with cheap goods, particularly as it looks to find new markets for what were formerly exports to the U.S. China has had success with the latter – its overall exports continue to grow despite a sharp decline in shipments to the U.S. – but there is limited evidence of widespread dumping. Export prices have been falling at about the same rate as overall producer prices (see chart) suggesting Chinese firms aren’t necessarily going out of their way to undercut foreign competitors.

Nonetheless, some countries are putting up barriers to Chinese imports to protect domestic industries, particularly in the auto sector where new Chinese EV brands are challenging traditional automakers. Canada has imposed a 100% tariff on Chinese EV imports – although that is reportedly being reconsidered in exchange for reduced tariffs on Canadian canola exports – while the European Union (EU) imposes additional countervailing duties of 17-35% based on the producer.

Chinese Producer Price Index drives export prices

As of September 2025. Sources: China General Administration of Customs (GAC), China National Bureau of Statistics (NBS), Macrobond, RBC GAM

-JN & VL

We opine regularly about the medium-term outlook for major emerging market economies. In our view, China can modestly exceed growth expectations over the next year, India can continue to expand quickly, and much of Southeast Asia has good prospects for the years ahead.

But we have tended to spend much less time critiquing individual data points in many of these emerging market economies or weighing in on the short-term outlook. As a remedy, we are pleased to introduce our Emerging Market Economic Momentum Index. It gauges which countries have enjoyed the best recent economic data and indeed which have the best short-term growth prospects. The index is created via a scorecard approach using three input variables:

1. Each country’s Manufacturing Purchasing Manager Index (PMI) gives a sense for how this bellwether sector is faring. The PMI also provides some insight into the near-term outlook. We focus on whether the index is stronger or weaker than the historically normal range for that country, rather than the absolute score.

Manufacturing PMI flags India for near-term growth prospects

As of October 2025. Box plot and whiskers show from top to bottom: maximum, 75th percentile, median, 25th percentile and minimum. Sorted by current level in descending order. Based on monthly data since 2006 or earlier if available. Sources: Haver Analytics, RBC GAM

1. The Citigroup Economic Data Change Index reveals whether the latest economic data for a country is coming in stronger or weaker than the average of the past year – in other words, whether the economy is accelerating or decelerating. Again, we focus on the score versus the country’s historical range, rather than the absolute value.

Citigroup Economic Data Change Index highlights Taiwan, Turkey and South Africa

As of October 2025. Box plot shows from top to bottom: 75th percentile, median and 25th percentile. Sorted by current level in descending order. Based on monthly data since 2004 or earlier if available. Sources: Citi, Bloomberg, RBC GAM

1. The Citigroup Economic Surprise Index conveys whether recent economic data has been better or worse than expected. This is distinct from “good” or “bad”, but it is still very useful. Efficient financial markets should have already factored in whether they expect the economy to be strong or weak, so a large fraction of any subsequent financial market movement should be on the basis of whether the actual result deviated from the expected result.
   

Citigroup Economic Surprise Index favours Taiwan, India and Thailand

As of October 2025. Box plot shows from top to bottom: 75th percentile, media and 25th percentile. Sorted by current level in descending order. Based on monthly data since 2004 or earlier if available. Sources: Citi, Bloomberg, RBC GAM

These three indicators are weighted equally in our Emerging Market Economic Momentum Index.

The inaugural edition of our index (see next table) finds India leading the way, followed by healthy indicators in South Africa, Taiwan, Turkey and Thailand. Conversely, Russia trails substantially, followed by Mexico, Poland, Brazil and China.

Our Emerging Market Economic Momentum Index places India out front

As at 11/03/2025. Sources: Macrobond, Haver, Citi, RBC GAM

There is, of course, no guarantee that the country hierarchy will remain in this particular order for long. This is by its nature a short-term guide to economic activity, and so fairly volatile. The hope is that it provides some guidance for investors incorporating macroeconomic factors into their tactical geographic decision-making.

-EL & AM

U.S. Federal Reserve

The Fed delivered the expected 25 basis point cut on October 17, reducing the fed funds rate from 4.00-4.25% to 3.75-4.00%. The Fed will also end its program of quantitative tightening on December 1. This means the central bank will no longer be tapping on the brakes (via a shrinking monetary base and the sale of government bonds) at the same time that it taps on the gas (via rate cuts). The latter decision was motivated not just by the balance sheet returning to a theoretically desirable size, but also by some recent stress in funding markets that suggests insufficient liquidity.

There was significant dissent at the meeting. Recent White House-appointee Miran favoured a larger 50-basis-point cut, while Kansas City Fed President Schmid preferred to leave the fed funds rate unchanged.

This was a “lesser” Fed meeting that did not update the Fed’s economic or dot plot outlook. The language of the statement was largely as expected, acknowledging moderate economic growth, slower job gains and still-elevated inflation.

The press conference yielded a hawkish turn, as Chair Powell revealed that there are strongly differing views about how to proceed from here (perhaps not entirely surprisingly given the afore-mentioned dissents). As a result, a further reduction of the fed funds rate in December is not a foregone conclusion. The market had previously priced nearly 100% of a rate cut in December, and this has since scaled back to about 79%. That’s probably about right, but we continue to flag the potential that the Fed doesn’t cut quite as much as the market imagines over the span of the coming year.

EL & JN

Internet outage

Amazon Web Services (AWS), the world’s largest cloud-computing service provider, was temporarily laid low by a failed software update on October 20.

Reflecting the extent to which even the world’s most sophisticated companies have outsourced much of their computing activity, this had amazingly far-reaching implications. Hundreds of major internet services were temporarily interrupted:

• social media (Facebook, Snapchat)

• online retail (Amazon, among others)

• voice assistants (Alexa)

• banks and financial platforms (Fidelity, Coinbase, Robinhood, Venmo)

• videogames (Fortnite, Roblox)

• home security (Ring cameras)

• airlines (United)

• sports ticketing

This was not the first such computer failure. Just over a year ago, a large fraction of the world’s Windows PCs suffered failures after a faulty CrowdStrike update.

Given that the AWS problem was resolved within 24 hours, this was mostly just a headache for the affected companies and their users. Still, some estimates put the economic cost at several billion dollars. This is large, if not large enough to make much of a dent in a global economy generating more than US$100 trillion of value per year.

The more urgent takeaway is that the concentration of corporate computer services onto a small handful of cloud providers represents a profound economic risk. Should these services go offline for a lengthy period of time, massive economic damage spanning dozens of economic sectors could accumulate. This might happen either due to internal errors or external hacking. The hacking could be by a profit-motivated syndicate, but it might instead be by an enemy in the form of cyberwarfare, crippling an economy as part of a broader strike.

It is unlikely that companies will respond by abandoning cloud computing, re-internalizing their computer services. Nor does it seem likely that there will be a significant increase in the number of cloud providers, so the concentration risk remains. As such, the onus is on cloud providers to continue to improve their resiliency against such episodes.

-EL

Canadian economy

The Canadian economy unexpectedly shrank in August by 0.3% month-over-month (MoM). The flash estimate had pointed to a flat reading. Although the new flash estimate for September is for a modest +0.1% rebound, the net conclusion is that of further economic weakness in Canada.

Third-quarter GDP is now tracking just +0.4% annualized growth, meaning that the country will probably only barely avoid recording two consecutive quarters of declining output. Further, the handoff to fourth-quarter GDP is now poor, meaning that Q4 GDP may rise by less than the Bank of Canada’s +1.0% projection. Our own forecast is below that number.

Given the headwind from tariffs, the August weakness originated from a logical place: goods production. Goods output fell by 0.6% due to broadly based declines including a 0.5% drop in manufacturing. It should be noted that services output also fell, by 0.1%, but this was due in part to a flight attendant’s strike.

-EL & JN

Bank of Canada

The Bank of Canada also cut its policy rate on October 17, from 2.50% to 2.25%. Monetary policy can now be characterized as slightly accommodative, which is consistent with appreciable slack in the economy, inflation that runs only moderately above target, and the expectation that some degree of excess capacity will persist through 2027.

The Bank of Canada suggests that its work is now done – there is no active plan for further easing if the Bank’s economic forecast is achieved. But our own forecasts are modestly below the consensus and we continue to believe there could be incrementally further monetary easing in Canada, though not necessarily as soon as in December.

-EL & JN

Canadian budget preview

We will keep Canada’s budget preview relatively short given that the budget itself will be released imminently, obviating the need for further speculation.

A starting observation is that Canada’s budget timing has now permanently shifted: after no budget whatsoever over the past a year and a half, future budgets will be tabled in the fall rather than the spring.

This budget should be substantially different than its predecessors. This is in part because there is a new Prime Minister and government in place, but more importantly because the circumstances have changed: Canada is grappling with an economic emergency as its access to the U.S. market is challenged. Bold action and a new economic vision are appropriate.

A much larger deficit is anticipated, possibly reaching C$100 billion (versus a mere C$42.2 billion projection a year ago). Interestingly, the largest projected deficit may actually be for 2025 rather than 2026. This is for a few reasons.

1. This year, 2025, has been a challenging year economically, with the result that government revenues have likely undershot initial expectations and automatic stabilizers have increased spending.

2. Some infrastructure money is already flowing, as are other newly created programs such as the Trade Diversification Corridors Fund and the Strategic Response Fund.

3. The budget likely assumes faster productivity gains in future years, which flatters the fiscal balance.

A key change in the budget will be the separation of operational and capital outlays. There will be a focus on balancing the operating budget – including significant program and staff spending cuts – whereas the capital budget will be allowed to run free. Of course, the overall money in / money out dynamic will be in significant deficit due to all of the capital investments expected. The idea behind separating operational and capital outlays is that these capital investments should in theory generate an economic return over time, rendering them “free” or at least less expensive. Key priorities include:

• Increasing infrastructure and resource-sector investment, both a means of strengthening the economy and diversifying away from the U.S. The already passed Bill C-5, the One Canadian Economy Act, helps in this pursuit by streamlining the approval process for such projects. The Parliamentary Budget Office estimates Canada will deliver another C$159 billion of infrastructure investments over the next five years.

• Trade diversification, with the goal of doubling non-U.S. trade over the next decade. It is an ambitious goal, and Canada is already nibbling away at it. The country recently secured invitations to formally engage in negotiations with India and China. Canada has also recently struck or is pursuing trade deals with Indonesia, Malaysia and the Philippines, and potentially also penning a broader trade deal with the Association of Southeast Asian Nations (ASEAN) in 2026. The budget can also be expected to continue supporting industries directly targeted by U.S. tariffs.

• Military spending, with some additional expenditures overlapping with infrastructure spending.

As per the Liberal campaign platform, efforts to stimulate residential construction are also expected.

The government already delivered tax cuts earlier in the year, abandoning the carbon tax and lowering the lowest-bracket tax rate. There is the hope that there could be a policy of accelerated depreciation to encourage Canadian businesses to do more capital expenditures.

Canada is expected to announce its own stablecoin strategy. This form of cryptocurrency is increasingly taking centre stage around the world given the way it removes volatility while maintaining the advantages of the blockchain.

On the net, the additional fiscal impulse and focus on high-fiscal-multiplier infrastructure could provide something like a 0.5% boost to Canadian GDP growth. But the timing of this is tricky. The spending cuts could hurt immediately whereas infrastructure spending takes time to trickle out.

The final burning question is whether the budget will pass. We assume yes, but this is far from certain given that the Liberals are two seats short of a majority, and they lack the official partner that they had in prior governments. That said, the Conservatives have reliably supported initiatives with which they have been aligned over the past six months.

The Liberal government has indicated that they would be willing to fight an election over the budget if need be. Current polling has the Liberals only a few points in front of the Conservatives, but still in a position to capture a similar number of seats as they did last spring, albeit with a considerable error margin. There is even a 46% chance that the Liberals would win a majority government in a hypothetical election held today – an attractive proposition, but hardly likely enough to actively pursue.