America's AI boom is keeping growth alive - and keeping the Fed in a bind
Capital spending and consumer demand are still carrying the U.S. economy, but inflation remains above target, hiring has slowed sharply and borrowing costs are biting harder. The next CPI report is the hinge point for a two-speed expansion.
That is the most useful way to read the U.S. economy in mid-October. Real output is still expanding. Consumers are still spending. Corporate investment, especially around artificial intelligence, data centers, computing capacity and related infrastructure, is providing a powerful tailwind. The International Monetary Fund's newly released World Economic Outlook puts U.S. real GDP growth at 2.3% for 2026, a pace that would be difficult to square with a conventional recession story.
At the same time, the labor market has lost much of the cushion that made earlier inflation fights easier to absorb. Employers added only 29,000 jobs in September, while the unemployment rate stood at 4.2%. Inflation is still running above the Federal Reserve's 2% goal, and August data showed both the consumer price index and the Fed's preferred personal consumption expenditures price index rising 3.4% from a year earlier. Mortgage rates have returned to levels that keep homeownership out of reach for many buyers.
The result is an economy with an accelerator and a brake pressed at once. The accelerator is investment, asset wealth and continued demand. The brake is expensive credit, sticky prices and slower hiring. For the Federal Reserve, the uncomfortable question is whether it can cool the first set of forces without breaking the second.
AI investment is becoming macroeconomic, not just a technology story
For most of the past two years, artificial intelligence was discussed as an earnings story for chipmakers, cloud providers and software companies. In 2026 it has become something larger: a demand shock that reaches construction, electricity, networking equipment, machinery, financing and land. That helps explain why broad economic activity has remained firmer than the slowdown in hiring would normally suggest.
The IMF's October outlook explicitly points to technology investment as one reason the global economy has proved more resilient than expected. The fund forecasts world growth of 3.1% this year and says economies and sectors benefiting from the AI investment wave are doing relatively well. For the United States, both the IMF and the Federal Reserve's September projections put 2026 growth at 2.3%.
Official U.S. data show the same basic resilience. The Bureau of Economic Analysis' third estimate put second-quarter real GDP growth at a 2.2% annual rate. Consumer spending, investment and exports all contributed. That is not breakneck growth, but it is enough to keep aggregate demand from cooling as quickly as policymakers might prefer.
The important distinction is that investment can be productive and inflationary at the same time. More computing capacity can raise future productivity, lower costs and expand potential output. But building that capacity now also requires power, skilled labor, equipment, financing and materials. If those inputs are constrained, the near-term effect can be stronger demand and higher prices before the productivity payoff arrives.
Data centers and computing infrastructure support orders, construction and related business activity even as other sectors slow.
When new demand arrives faster than electricity, components and skilled labor can expand, prices can rise before new capacity is ready.
Higher productivity could eventually make growth less inflationary, but the timing and scale of that payoff remain uncertain.
A supply-side boom can still create a demand-side problem on the way there.
The Fed's September minutes captured that ambiguity. Some participants saw strong productivity and investment as reasons the economy could grow faster without overheating. A couple of participants also flagged AI-related demand and energy-market disruptions as potential sources of broader price pressure. Both can be true at once.
Inflation is no longer only about rent and services
August inflation data showed a broad enough problem to keep the Federal Reserve on alert. The consumer price index rose 0.4% for the month and 3.4% from a year earlier. Core CPI, which excludes food and energy, rose 0.3% for the month and 2.4% over 12 months. Gasoline rose 3.9% in August, the broader energy index rose 2.1%, and shelter increased 0.3%.
The Fed's preferred PCE measure told a similar story. In August, the PCE price index rose 0.3% from July and 3.4% from a year earlier. Core PCE rose 0.2% for the month and 3.0% over 12 months. Inflation has come down from earlier peaks, but it has not returned to the 2% neighborhood the Fed wants to see on a sustainable basis.
Import prices add another layer. Bureau of Labor Statistics data show nonfuel import prices up 5.5% over the year through August, the fastest 12-month increase since 2022. Import prices for capital goods were up 7.3%, while nonfuel industrial supplies and materials were up 12.6%. Those categories matter because they feed into the cost of equipment, production and construction before consumers ever see a retail price tag.
August CPI, year over year. Energy and shelter both contributed to the monthly increase.
August PCE inflation, year over year; core PCE was 3.0%.
Nonfuel import prices, year over year through August, signaling pressure beyond oil alone.
Consumers are still spending, but the cushion is thinner
Household demand has been the most persistent source of U.S. economic resilience. In August, personal consumption expenditures rose 0.9% in current dollars and 0.6% after adjusting for prices. Personal income rose only 0.2%, while disposable personal income rose 0.3%. The personal saving rate was 4.1%.
One month does not establish a trend, and households can legitimately draw on savings or credit when income temporarily lags spending. But the August gap is a useful warning. Consumption cannot permanently outrun income. If borrowing stays expensive and job creation stays weak, households with less wealth and less cash buffer will eventually have to make harder tradeoffs.
Pay growth is no longer an obvious offset to inflation. Average hourly earnings rose 3.0% from a year earlier in September. Because the September CPI had not yet been released at publication time, a clean September real-wage comparison was not yet possible. But August CPI inflation was 3.4%, underscoring how narrow the margin has become for workers who depend primarily on wages rather than asset gains.
The consumer has not cracked. But resilience is getting more expensive to maintain.
Hiring has slowed enough to change the risk calculation
The September employment report is the clearest evidence that the economy's strength is not evenly distributed. Nonfarm payrolls rose by just 29,000, and the unemployment rate was 4.2%. Over the previous 12 months, payroll growth averaged 45,000 per month, far below the pace that characterized the post-pandemic expansion.
That does not automatically mean the labor market is collapsing. Demographic changes, slower labor-force growth and productivity gains can reduce the number of jobs needed to sustain output. But slower hiring matters because it reduces workers' bargaining power and makes households more sensitive to any further shock. A worker who is confident of finding another job can tolerate higher prices and borrowing costs more easily than one facing a thin vacancy market.
The Fed therefore has to separate two questions that often travel together but are no longer moving in sync: Is demand too strong for inflation to fall? And is the labor market strong enough to withstand tighter policy? The answer may now be yes to the first and only barely to the second.
High rates stop being abstract when they hit the monthly payment
Housing is where monetary policy becomes tangible. Freddie Mac's weekly survey put the average 30-year fixed mortgage rate at 7.40% on Oct. 8, with the 15-year rate at 6.73%. At those levels, even buyers with solid incomes face monthly payments far above what the same home would have required during the low-rate years.
Existing-home sales rose 1.5% in September to a seasonally adjusted annual rate of 3.99 million, according to the National Association of Realtors, but sales were still 2.2% below a year earlier. The median existing-home price was $417,500, up 1.3% from a year earlier, and inventory stood at 1.60 million homes. The combination of expensive financing and prices that have not fallen much keeps affordability under pressure even when more homes come onto the market.
This is also why another Fed rate increase is not a narrow Wall Street event. It can raise financing costs for homebuilders, landlords, small businesses and households, while higher Treasury yields can push mortgage rates higher even without a one-for-one move in the policy rate.
Why the Fed tightened even as job growth weakened
On Sept. 16, the Federal Open Market Committee voted unanimously to raise the federal funds target range by a quarter percentage point, to 3.75% to 4.00%. The statement described economic activity as expanding at a solid pace, called capital investment robust and said inflation remained elevated. The central bank's message was straightforward: the inflation problem was serious enough to justify more restraint.
The accompanying projections made the tradeoff even clearer. The median 2026 forecast put real GDP growth at 2.3%, the unemployment rate at 4.1%, headline PCE inflation at 3.7% and core PCE inflation at 3.4%. In other words, policymakers expected a reasonably strong economy and inflation materially above target at the same time.
The September minutes, released Oct. 7, said all participants supported the rate increase and many judged that a somewhat higher policy path would be prudent. Several participants viewed the policy rate as not restrictive or only mildly restrictive. That language matters because it tells investors and borrowers that the Fed does not necessarily think the current rate level is enough to finish the job.
Growth is solid, AI-related investment is strong, consumer spending remains positive and inflation is still well above 2%.
Weak payroll growth and expensive credit increase the risk that another hike causes more labor-market damage than intended.
If AI investment raises productive capacity quickly, the economy can sustain faster growth without generating as much inflation.
If external price shocks spread into wages and services, the Fed may have to tighten even as hiring remains soft.
The wealth effect is real - and very uneven
Strong equity markets can support spending by making asset-owning households feel wealthier, but the effect is concentrated. Federal Reserve distributional data for the second quarter show households in the top 10% of the wealth distribution holding about $56.9 trillion of corporate equities and mutual-fund shares, out of roughly $64.6 trillion across the shown groups. The bottom half held about $0.37 trillion. Those figures do not include every form of retirement exposure, but they illustrate why a market boom can lift aggregate consumption without making most households feel equally flush.
One reason spending stays firm: the people with the most assets can keep spending
This helps reconcile two facts that otherwise seem contradictory. Surveys and household anecdotes can show intense frustration with prices, rent and credit costs, while aggregate consumption remains strong. The households most exposed to housing payments, auto loans, credit cards and day-to-day grocery inflation are not necessarily the households driving the largest dollar share of discretionary spending.
At the top of the distribution, rising portfolios and high incomes can keep travel, services and luxury spending resilient. At the lower and middle parts of the distribution, the same economy can feel like a squeeze: borrowing costs are higher, savings buffers are thinner and wage gains no longer clearly exceed inflation.
For monetary policy, this creates a transmission problem. Higher interest rates are supposed to slow demand. But when spending is supported by wealth gains and investment booms that are less sensitive to household borrowing costs, the Fed may need to hold rates higher for longer to get the same cooling effect. That, in turn, can intensify the pain in the most rate-sensitive corners of the economy.
September CPI could reset the entire rate debate
The Bureau of Labor Statistics is scheduled to release the September consumer price index at 8:30 a.m. Eastern on Wednesday, Oct. 14. At publication time, those figures had not yet been released. That distinction matters: the next inflation print will be new evidence, not confirmation of any forecast now circulating in markets.
Investors will focus first on the monthly change in headline CPI because energy costs have been volatile. They will then look through to core inflation, especially shelter and service categories, to judge whether the price shock is broadening. A hot headline number driven mostly by gasoline would be uncomfortable but different from a hot core number driven by persistent services. A soft core reading would give the Fed more room to wait, even if energy keeps headline inflation elevated.
The sequence after CPI is also important. Producer prices are scheduled for Oct. 15, and import and export prices for Oct. 16. Together, those releases can reveal whether cost pressure is moving from commodities and imported inputs into the domestic production pipeline.
September CPI scheduled
Read the composition, not just the headline
Energy-heavy inflation and broad service inflation have different implications. The Fed cares about whether price pressure is spreading and persisting, not just whether one monthly number looks high.
The next phase depends on which side of the economy gives way first
No single indicator can settle the outlook because the forces pulling on the economy are unusually different. The most useful framework is to watch three paths rather than make one heroic forecast.
Productivity catches up
AI investment expands productive capacity, inflation cools without a collapse in demand, and the Fed can stop tightening after a limited number of moves. Growth stays near trend while hiring remains modest rather than contracting.
Inflation broadens
Energy, imports and investment-related demand spill into more services and wages. The Fed raises rates again or keeps them high for longer, housing weakens further and the labor market absorbs more of the adjustment.
The investment boom cools
Corporate spending slows before productivity gains arrive. With hiring already weak, aggregate demand loses its strongest support and the economy shifts from an inflation problem toward a growth problem.
For households, companies and investors, the economy demands more selectivity
The phrase "soft landing" is no longer enough to describe the challenge. A landing can look soft in GDP and harsh in household cash flow. It can look strong in corporate investment and weak in hiring. It can reward owners of appreciating assets while squeezing renters, first-time buyers and borrowers. That is why the next several weeks of data will matter less as a verdict on recession and more as a map of where the stress is moving.
Households should expect borrowing costs to remain a central constraint until the Fed sees convincing evidence that inflation is returning toward 2%. Businesses need to distinguish between demand that is durable and demand that is being pulled forward by an investment rush. Investors have to weigh the genuine productivity potential of AI against the possibility that the capital boom itself keeps rates higher and financing more expensive.
For policymakers, the hardest outcome is not simply high inflation or weak growth. It is the combination now visible in parts of the data: strong aggregate spending, elevated prices, concentrated investment and a labor market that is losing momentum. That mix leaves less room for error.
Focus on cash flow, not just the headline economy. High mortgage, auto and revolving-credit costs can outweigh the benefit of solid GDP growth.
Watch imported input costs, electricity constraints and labor availability. The AI buildout can raise both demand and operating costs.
The key question is whether earnings and productivity rise fast enough to justify high valuations while interest rates remain restrictive.
CPI composition, not one headline print, will determine whether policymakers see inflation as a temporary shock or a broader persistence problem.
Quick questions
Is the United States in a recession?
Current official data do not describe a recession. Real GDP grew at a 2.2% annual rate in the second quarter, and both the IMF and the Federal Reserve project 2.3% real growth for 2026. The concern is uneven growth and a weakening labor market, not a confirmed economy-wide contraction.
Why would the Fed raise rates if hiring is weak?
Because inflation remains above target and demand is still resilient. The Fed is balancing its price-stability mandate against growing labor-market risk. September's unanimous rate increase showed that policymakers judged inflation risk to be the more urgent problem at that meeting.
Why does AI investment matter for inflation?
It can raise demand for electricity, construction, equipment and skilled labor before new capacity is fully available. Over time, AI could also raise productivity and reduce inflation pressure by increasing output. The timing of those two effects is one of the biggest uncertainties in the outlook.
What should readers watch in the September CPI report?
Separate energy from core inflation and then look inside core services, especially shelter. A broad acceleration would matter more for the Fed than a one-month jump concentrated in gasoline. The report was scheduled for 8:30 a.m. Eastern on Oct. 14 and had not been released at publication time.
Why does the economy feel worse than GDP looks?
GDP is an aggregate measure. It does not show how evenly gains are distributed or how much households pay to finance homes, cars and credit-card balances. Asset owners can benefit from market gains while borrowers and renters experience much tighter conditions.
Sources and data
- U.S. Bureau of Labor Statistics, Consumer Price Index - August 2026 inflation data.
- U.S. Bureau of Labor Statistics, Employment Situation - September 2026 jobs and unemployment data.
- U.S. Bureau of Labor Statistics, Import and Export Price Indexes - August 2026 import-price data.
- U.S. Bureau of Labor Statistics, 2026 release calendar - scheduled October CPI, PPI and import-price releases.
- U.S. Bureau of Economic Analysis, GDP third estimate - second-quarter 2026 real GDP.
- U.S. Bureau of Economic Analysis, Personal Income and Outlays - August income, spending, saving and PCE inflation.
- Federal Reserve, Sept. 16, 2026 FOMC statement - target rate increase and policy assessment.
- Federal Reserve, Sept. 16 meeting minutes - policy discussion, inflation risks and views on restrictiveness.
- Federal Reserve, September 2026 Summary of Economic Projections - median GDP, unemployment and inflation forecasts.
- Federal Reserve, Distributional Financial Accounts - second-quarter 2026 household asset distribution.
- Freddie Mac, Primary Mortgage Market Survey archive - Oct. 8 mortgage rates.
- National Association of Realtors, Existing-Home Sales - September 2026 sales, prices and inventory.
- International Monetary Fund, October 2026 World Economic Outlook data - U.S. and global real GDP growth projections.
- International Monetary Fund, October 2026 World Economic Outlook briefing - global resilience, AI investment and risk discussion.
Comments
Post a Comment