Monologue
The inflation outlook and the implications for interest rates have emerged as the key theme for financial markets through the remainder of 2026 and into 2027. We tackle this head-on in our comprehensive Memo this week, cross-checking the near-term inflation views we expressed in our Q2 Outlook Report against a monetarist analysis of credit conditions. Without giving away the nuanced takeaways (scroll down!), the piece argues that we could be entering a tightening cycle, but a very different one from 2022.
At last week's monetary policy meetings, global central banks shifted their forward guidance in a more hawkish direction, but the jury is still out on whether we'll actually see interest rates rise in the US. Much depends on the duration of the Strait of Hormuz blockade (and how open it is thereafter), so there is genuine uncertainty on this question. In our updated scenarios, the odds of a hike within the next year are around 50%.
Beyond the geopolitical imponderables, there is another factor making the forward interest rate path difficult to read: how central banks are thinking about their response to the inflationary consequences of the Iran War. Despite decades of study since the 1970s oil embargo, economists have yet to agree on a standard playbook for supply shocks.
A fascinating discourse is just emerging, which I'm going to take the liberty of branding "the Atlantic debate."
In the new world corner, incoming Fed Chairman Kevin Warsh expresses what's essentially the established wisdom, augmented with modern tools. At his Senate nomination hearing, Warsh argued that policymakers should "look through" large, one-off supply-driven price increases, with rates rising only if underlying inflation moves persistently above target. His preferred indicators of underlying inflation are so-called "trimmed" measures, in which a certain percentage of the highest and lowest price movements in each month's inflation basket are removed from the index.
There are some problems with this approach, notably that, by the time "underlying inflation" (whatever the measure) has risen meaningfully above target, it's probably too late (see chart). Warsh is essentially supporting a backward-looking approach that will, by definition, fail at inflection points. Apparently missing the irony, Warsh said in the same hearing that "the fatal policy error going back four or five years [the Fed being too slow to raise rates] is still a legacy that we’re dealing with."
Across the pond, Bank of England Chief Economist and Executive Director for Monetary Analysis and Research Huw Pill has quietly developed a fresh policy framework (outlined in a recent speech) which uses scenarios to guide decision-making under "radical uncertainty," by which he means uncertainty that can't be reduced to known probabilities or risks. Under "radical uncertainty," we don't know whether we have the right model of the world, and even if we do, we're unsure about the right parameters.
The reason I say "quietly" is that the approach is philosophically abstract, so it may not be receiving the attention it deserves. In essence, Pill argues that taking a "robust" approach to uncertainty by choosing a response that performs acceptably well over a range of plausible scenarios is better than the "optimality" approach of picking (guessing) one model and following it unerringly.
In the case of the current supply shock, where there is uncertainty about the scale of the shock itself and whether wage- and price-setting behaviors in response have changed since 2022, Pill's framework favors earlier interest rate hikes over Warsh's "wait and see" approach. Across scenarios, the cost of being wrong about potential persistently high inflation is higher than the cost of over-tightening. Pill put his reputation where his mouth is in last week's interest rate meeting, casting a lone dissenting vote in favor of raising Bank Rate.
Between these two Atlantic poles, a middle ground may be emerging from (where else) Canada. Bloomberg reports that as part of the regular five-year review of its monetary policy framework, the Bank is considering adding explicit consideration of the current output gap when responding to supply shocks. Governor Macklem hinted that the Bank may even spell out its reaction function under different circumstances. This doesn't go as far as Pill's "robust" scenario-based approach, but it would force the Bank to think beyond textbook orthodoxy and outline different policy paths for the same class of shock depending on domestic demand.
A very faint silver lining on the horizon of the Iran conflict is the potential for a breakthrough in how rate-setters respond to supply shocks. We'll be watching the debate – and the data – closely.
Dylan Smith
Founder and Chief Economist
Marginal Movers
Rising 👆
- Vigilance: We must be mindful of the risks of private credit — Bank of England Governor Andrew Bailey writes in a rare opinion piece that "There is no doubt that private credit offers significant benefits to the financial system... In recognising the benefits, however, we must be mindful of the potential risks. Among these is that private credit remains untested in a severe or prolonged economic downturn."
- AI spend (demand-side): How Much Are Firms Spending on AI (and What Will Happen to Headcounts)? — "Firms appear to be ramping up AI spending per employee significantly—by 50 percent, to roughly $2,000 per employee (or $280 billion in aggregate) this year."
Falling 👇
- Confidence: The Value of Reliable Statistics — "Our baseline estimate implies that preserving trust in the integrity and quality of official statistics generates economic benefits of about $25 for every $1 spent on the agency’s budget."
- Education: Influential study touting ChatGPT in education retracted over red flags — "In some cases it appears it was synthesizing very poor quality studies, or mixing together findings from studies that simply cannot be accurately compared due to very different methods, populations, and samples."
Macro Monitor
Marginal improvements in the US labor market
Coming against the backdrop of concerns that the Iran conflict might weigh on economic activity, a slew of data signaling a net improvement in the labor market in March and April comes as welcome news.
Current employment conditions have been described as "frozen," with a low-hiring, low-firing dynamic preventing a deterioration in the labor market while keeping the unemployed out of work for longer. This may be shifting, with JOLTS data recording the largest increase in hiring since the pandemic era in March (up by 0.7 million to 5.6 million hires), although this was offset by a pickup in layoffs after a recent trough. Alternative data from Challenger, Gray & Christmas, and initial claims reports both indicate that layoffs declined in April.
For the first time since June 2025, payrolls rose in sequential months in March and April (by a strong 185k and 115k, respectively). For the second month in a row, reported employment growth more than doubled the consensus Wall Street estimate. Private data providers Revelio and ADP both posted their largest monthly increase in employment since early 2025, confirming the official signal.
The nascent improvement in labor demand is combining with a restricted supply environment (due to tighter immigration policy) to arrest the upward drift in the unemployment rate, which stayed constant at 4.3% in April. A weekly unemployment index compiled by Morning Consult reached its lowest (i.e., most positive) level since mid-2025 in early May.
Wage growth remained roughly steady in April, but sectors most exposed to the AI infrastructure build-out and immigration-related shortfalls are heating up.
Canadian labor market loosens
Canada's labor market is going in the opposite direction from the US. The unemployment rate rose by 0.2 percentage points in April, to 6.9% (still below the recent August/September 2025 peak of 7.1%), on a combination of a small drop in overall employment (-18k) and an increase in the labor force as more people enter the labor market (+33.5k).
The employment-to-population ratio, our preferred summary indicator of the state of the labor market, fell by 0.1 percentage points to 60.5%, lower than at any point between 2000 and 2020, including the aftermath of the global financial crisis. Despite the Canadian population declining slightly in 2025 due to tighter immigration policy, the economy is not generating enough jobs to absorb young workers entering the labor market or older people returning to the workforce for economic reasons.
Aside from lackluster demand, a major reason for the subdued labor market is firms' unwillingness to hire amid the uncertainty created by upcoming CUSMA/USMCA trade negotiations and the impact of the Iran conflict. Additionally, the softening real estate industry and disruptions in the technology sector are harming employment. Consequently, we're seeing job losses in construction (-15.7k), information services (-24.8k, though far firmer in YoY terms), resources (-5.5k), and manufacturing (-1.5k), and net hiring in services industries such as business services (21.5k), healthcare (17.5k), accommodation and food (13k), and education (6k).
The Bank of Canada's core mandate is inflation, but it takes the labor market into account. We see the oil price shock in Canada as being smaller and more transitory than in the US, as underlying economic conditions are less inflationary. Today's signal from the labor market reinforces our view that the Bank of Canada can remain on hold longer than the Fed, and may not need to raise rates at all. If we were to get a resolution in Hormuz while trade negotiations drag on, we'd pencil in a rate cut in Canada this year.
Higher rates are here
In case you've missed it, two G10 central banks have now increased interest rates since the start of the Iran conflict. Norway (a net oil exporter) this week joined Australia (a net importer), raising its policy rate by 25 basis points.
See the appendix for arcMacro proprietary Factors and the Key Macroeconomic Indicators tracking chart.
Market Monitor
Public markets
US equity markets hit new highs again this week, with good jobs data complementing a strong earnings season to help investors shrug off geopolitical uncertainty.
This said, concentration has reached new extremes, with the market returning to the "AI hyperscalers vs. everyone else" dynamics that defined 2025. Five tech stocks – Alphabet, Nvidia, Amazon, Broadcom, and Apple – have accounted for more than half of the S&P’s recent gains. These lopsided dynamics suggest the market is fragile and prone to a correction. Next week's inflation data will provide a test; higher-than-expected inflation or evidence of rapid pass-through to core measures would force investors to price in higher interest rate expectations (hyperscalers are now major borrowers).
Beyond the stock market, small weekly moves belied extreme churn — especially in commodity markets, as hopes for a deal to open the Strait of Hormuz once again came to nothing.
Private Markets
Those following the private credit sector would do well to read the Financial Stability Board's Report on Vulnerabilities in Private Credit (it makes a good companion to our two-part report on the risks the industry may pose to the financial system and markets).
The report is both comprehensive and subtle. The upshot is that we can look forward to more active monitoring and better data as regulators start to peer through the industry's opacity — an innovation to be welcomed.
The chart below captures the most important fact about the industry: it's just too small to be a systemic risk to the banking system.
See the appendix for the market monitor table
Memo
Rising Business Borrowing: Inflationary Bottleneck, or Disinflationary Productivity Driver?
Bottom line: We conduct a monetarist cross-check of our near-term inflation views. Decomposing the ~10% annualized surge in bank lending in early 2026 sheds light on the nature of potential near-term inflation pressures. Unlike in 2022, new credit is being directed almost exclusively to productive uses rather than inflationary consumer borrowing. The question is therefore whether temporary supply-side bottlenecks – in chips, labor, and energy (exacerbated by tariffs) – will outweigh productivity gains. We think the answer in the short term is yes, but the pattern of borrowing has made us more optimistic that a supply-side expansion can help tame inflation in the medium term.
What it means for investors: This is not 2022, but near-term inflation risks remain elevated. Interest rates may need to rise to cool pressures, in which case we anticipate a short and shallow hiking cycle compared with the post-pandemic period. Much still depends on uncertain developments in the Middle East.
In our recent Quarterly Outlook, we argued that inflation risk in the US is higher than markets appreciate, making interest rate hikes more likely than forward pricing implies.
In the spirit of cross-checking our approach in the face of uncertainty, we go old-school and look at what the monetary side of the economy is telling us.
To do so, we apply some recent advances made by scholars who are braving the monetarist winter by studying credit expansion (notably Mian, Sufi & Verner (2020) and Bezemer & Zhang (2019)). Building beyond the standard monetary aggregates approach, these authors provide cross-country evidence on the inflationary consequences of different categories of credit extension. We apply their basic approach to the US, with less econometrics.
Is the money talking?
In some corners of the financial world, monetarist analysis of inflation dynamics is back in vogue. After all, growth in aggregate M2 money supply provided a strong leading signal ahead of the 2021-2023 inflation surge.
After turning negative when the Federal Reserve belatedly raised interest rates in early 2022, M2 growth is now back to something like normal. In year-over-year terms, M2 growth of 4.6% puts it between the 1990s tech boom average of 4% and the 2000s average of 6%.
At face value, this signals a benign inflation outlook. However, academics and practitioners have long known that looking at M2 alone can be misleading. It's simply too broad, and its drivers too disparate, to provide a clean signal. One problem is that it can capture both economic fundamentals and policy responses. Looking at the chart above, one could easily confuse the 2012 rise in money supply growth as an inflationary signal. In fact, it captured the Fed's balance-sheet expansion in response to the risks of outright deflation.
Can we improve on M2?
Yes, by getting more granular. The underlying logic of the monetarist approach to inflation is sound. When the money supply expands in response to credit expansion or public spending, aggregate demand rises. If that demand is not matched by a corresponding rise in the supply of goods and services, prices will rise. This is the classic case of more money chasing the same quantity of goods.
Following this logic, we narrow our focus to bank lending (which drives money creation via the multiplier process) and acknowledge that different types of credit are more inflationary than others, because they have different relative effects on aggregate supply and demand.
We define three types of credit:
- Consumption lending has no effect on supply; it only affects demand, making it the most inflationary category. This category includes credit cards, consumer loans, and auto loans.
- Productive lending provides businesses with the financing they need to invest in their operations. This creates demand (e.g., the purchase of new machines), but also expands aggregate supply. The impact on inflation is indeterminate, depending on timing and scale. This category includes all commercial and industrial bank lending, as well as lending for corporate construction and development.
- Financial lending is any lending to businesses or households to fund the purchase of financial assets, including non-construction real estate loans. This has minimal effects on consumer prices in the near term, but can fuel asset prices and eventually trickle through into higher inflation over time.
Accounting for shadow banking
Before we decompose lending growth into these three components, there is a complication to deal with. Between 2015 and 2025, lending to "nondepository financial institutions" (NDFIs) ballooned by $2.5 trillion, rising from 4% of all bank credit to 14.5%. NDFIs are a catchall category encompassing any financial institution not licensed as a bank to take deposits, ranging from insurers to asset managers to consumer mortgage lenders to private equity and private credit funds. Lending to NDFIs has constituted the largest share of the increase in bank lending in 2026.
The problem we face is that publicly available bank lending data do not break down lending to NDFIs by their ultimate use. To decompose NDFI lending into the three categories listed above, we rely on a recent FDIC report that provides the splits as of Q3 2025. The report attributed roughly a quarter of total NDFI lending, respectively, to household mortgage lending, business credit, and private equity; another 10% to direct consumer lending; and the remaining 15% to unclassified. We assume that the proportions are stable and decompose aggregate NDFI accordingly. This is an imperfect assumption, but more accurate than excluding this lending channel now that it is economically significant.
Framing the inflation outlook: bottlenecks vs. productivity growth
The chart below shows the results of our decomposition of contributions to the 3-month growth in bank lending into our three categories. Total bank lending is growing at a rate of nearly 10% on an annualized basis, twice the average rate in 2025 and significantly above the average since 2015.
The surge in credit growth is being driven by productive lending, which is replacing financial lending, while the contribution from consumption lending stays stable. Unlike in 2022, supply-side pressures are not coinciding with strong consumer demand fueled by money supply and credit growth. Instead, lending to consumers looks roughly in line with levels needed to sustain consumption at current levels.
The chart gives us a clear framing of the current question we should be asking ourselves: Will surging corporate borrowing prove inflationary or disinflationary?
The answer may well be "a bit of both." In the near term, we think inflationary bottlenecks are already emerging and will be significant enough to keep inflation well above target over the next 6-18 months. The contributing factors are:
- AI infrastructure shortages: The pace of the data center and related infrastructure buildout is already creating shortages in required inputs, including microchips and related equipment.
- Labor market tightness: Immigration restrictions risk creating pockets of labor market tightness, especially where high concentrations of migrant workers overlap with rising demand. In such industries, we're already seeing wage growth rising (see chart).
- Electricity prices: Secular trends and cyclical pressures are converging to raise electricity prices, which will have significant knock-on effects in general price levels. Key factors include data centers, the shift to renewables, vehicle electrification, higher oil and gas prices, and aging power infrastructure (see chart).
- Tariffs: Businesses are passing on tariff costs that were initially absorbed in margins, adding another near-term source of inflation.
Over a longer horizon, we have some sympathy with incoming Federal Reserve Chair Kevin Warsh's view that a wave of inflation-attenuating productivity growth may be on the horizon.
In fact, a technology-based infrastructure boom of the type we're currently seeing can ultimately only be disinflationary. Either it will generate higher productivity, enabling a given level of economic growth to be achieved with less inflation (all else constant), or the boom will turn into a bust, which would be even more disinflationary.
The catch is that the timing of potential productivity improvements is uncertain, and the process of achieving them (companies still need to invest to integrate AI into their processes) can sustain inflation well into the medium term.
Overall, then, we see the signal from monetary analysis as being consistent with our view that the current economic environment points to sustained above-target inflation. Over the longer term, productivity gains will compete with fiscal sustainability concerns regarding expected inflation (both imply higher long-term interest rates).
How will the Fed respond?
The combination of short-term inflationary pressure and longer-term productivity growth implies a modest hiking cycle for the Federal Reserve. Inflation did not return to the 2% target and is now rising again. We currently assign a 50% chance that the Federal Reserve will need to hike to maintain its credibility as inflation continues to drift away from target.
At the same time, the nature of the inflationary risk we describe in this piece does not require a severe hiking cycle of the type we saw in 2022/23, because rampant consumer demand and exuberant financial markets do not need to be brought to heel. Rather, a signal that the Fed remains vigilant – perhaps via 25-75 basis points of hikes – will be enough to constrain excesses while supply bottlenecks are resolved and productivity improves.
Of course, there is an important caveat: developments in the Strait of Hormuz are a critical short-term factor, and could mean a steeper hiking cycle if shortages from a sustained blockage cause underlying inflation to rise or expectations to become de-anchored.
Appendix
Proprietary Factor and Regime Model and Key Indicators
Disclosures
AI Declaration
All written content, analysis, and opinions are original and ascribed to the author. AI tools were used for proofreading and summarization purposes only. AI tools may also have been used in the development (codebase) of the analytical models reported in this document.
Disclaimer
This publication is for informational and educational purposes only and does not constitute financial or investment advice. Nothing in this report should be construed as a recommendation to buy, sell, or hold any security or financial instrument. Always consult a qualified financial advisor before making investment decisions.