Viewpoints

From January to August the average tariff rate on all imported goods increased 9 percentage points, from 2.3% to 11.3%, while the price of those imports, recorded by the Bureau of Labor Statistics (BLS) excluding tariffs, was essentially unchanged. Hence, the price of imports, including tariffs, rose 1.113/1.023 - 1 = 8.5%, reflecting the full amount of the increase in tariffs (Chart 1, solid lines). This suggests the cost of the tariffs has been borne entirely by American businesses and consumers, with none of that cost passed backwards to foreign suppliers.

State real GDP growth rates in 2025:2 ranged from -1.1% in Arkansas (Mississippi fell at a 0.9% rate) to 7.3% in North Dakota. States in the Plains benefited from a turnaround in farm output after declines in the first quarter, and there were also increases in mining output in some of those. Arkansas and Mississippi were dragged down by losses in farm output. Strong increases in finance and information helped boost growth in states such as California, New York, and Massachusetts.

Personal income growth rates ranged from 10.4% in Kansas to 0.9% in Arkansas. In the majority of states, perhaps most notably Massachusetts, increases in transfer payments significantly swelled the income gains related to state GDP growth. Much of the transfer growth reflected the recent retroactive payments to some Social Security beneficiaries.

This release also included estimates of consumer spending by state for 2024. Given the marked lag in the release of these numbers, and their annual frequency, they are likely not of much interest in assessing current and perspective conditions, either for the nation as a whole or for individual states, though they may be of help in modeling state revenues.

The Federal Reserve Bank of Philadelphia’s state coincident indexes in August were on the soft side, but a touch better than the initial July results. In the one-month changes, while four states (Kentucky, Rhode Island, Colorado and Alabama) had increases above .50 percent, seven were down. Over the three months ending in August, Alabama was on top, with two other states (Kentucky and Colorado) also seeing increases above one percent. Five states saw declines, with Maryland down .60 percent. Over the last twelve months, Iowa was down, and four others saw increases of less than one percent. Alabama was the only state with an increase higher than four percent (Idaho was up 3.62 percent), and eight others were at or higher than three percent.

The independently estimated national estimates of growth over the last three and twelve months were, respectively, .38 and 2.25 percent. Both measures appear to be a bit lower than what the state numbers would have suggested.

The AI boom has revived the old promise that a wave of general-purpose technology will lift labour productivity everywhere. But sector mix matters. The chart below shows where recent job growth has actually happened: healthcare and social care. Since end-2023, employment in that sector has grown 3–4x faster than overall payrolls in the US, UK and Japan, pushing healthcare’s share of total jobs toward 14–15%. The euro area has been softer, but the direction is similar. Ageing populations and long-COVID backlogs are doing the heavy lifting. But the composition effect is awkward for macro optimists: healthcare is labour-intensive, highly regulated and only partly tradable—characteristics that usually come with lower measured productivity growth.

State labor markets were little-changed in August. Utah saw a statistically significant increase in payrolls and DC saw a drop. No other state had a significant change.

The unemployment rose a significant .2 percentage points in Delaware and Maryland, while increasing .1 percentage point in Minnesota. Colorado’s rate fell .3 percentage point, and Alabama was down .1.California. The highest unemployment rates were in DC (6.0%), California (5.5%), Nevada (5.4%), Michigan (5.2%), New Jersey, Ohio, and Oregon (all 5.0%). Alabama, Hawaii, Montana, North Dakota, South Dakota, and Vermont had unemployment rates under 3.0%, while South Dakota’s 1.9% was the lowest in the nation.

Puerto Rico’s unemployment rate edged up to 5.6% and the island’s job count rose 3,100.

At the start of the year, the consensus view was grim: the US-led world economy was heading into recession, inflation would spiral out of control, and global trade faced collapse. Yet, none of this has come to pass. Business-cycle indicators still show resilience, and neither a global trade downturn nor sustained inflationary pressure has materialised.

Global Exports: Resilient Despite Tariffs

The trade war has been disruptive but not disastrous. In the US, GDP contracted in the first quarter, largely because of a sharp rise in imports as manufacturers front-loaded inventories ahead of tariff deadlines. By the second quarter, domestic demand shrank 0.4% quarter-on-quarter as inventories slumped. Exports and residential investment declined, but consumer spending actually strengthened—hardly evidence of households buckling under tariff pressure.

The labour market has softened but isn’t collapsing. Unemployment at 4.3% is still below the pre-pandemic average of 5.1%, while August retail sales rose a brisk 5% year-on-year.

Surprisingly, the trade war’s impact on global trade has been modest. World exports dipped 6.5% from late 2024 to February 2025, only to rebound 13%, leaving shipments up nearly 6% year-to-date through May (Figure 1). The sharpest weakness came in April and May, immediately after tariff hikes, but the recovery was swift.

By comparison, past crises inflicted far greater damage. During the Global Financial Crisis, global exports collapsed by 47% in just eight months. Between 2014 and 2016, shipments fell 28% amid crashing commodity prices, a slowing China, and a strong US dollar. The pandemic caused a 33% peak-to-trough fall when economies shut down worldwide. Against this backdrop, the current downturn appears relatively mild.

Chart 1 shows that no matter how you slice it and dice it, consumer price inflation is trending higher. In the three months ended August 2025, the annualized rate of consumer inflation has ranged from 3.3% (FRB Cleveland 16% Trimmed Mean) to 3.65% (CPI excluding food and energy components). Again, consumer inflation is trending higher.

The Case Shiller Home Price Index fell 0.3% in June, its 4th consecutive monthly decline, lowering its yr/yr increase to 1.9% from 6.6% in February 2024 (Charts 1 and 2).

The flattening of home prices reflects slower demand for home sales and a moderation of transactions. This has resulted from the surge in home prices--the Case Shiller Home Price Index surged 38.9% from mid-2020-mid-2022 and until recently it has continued to rise--combined with higher mortgage rates. The home affordability index hovers near an all-time high. The current fundamentals in housing suggest that the Case Shiller Home Price Index will continue to be flat-to-down in the next year.

The recent consecutive monthly declines in home prices have potentially important implications for overall inflation. Rental costs and OER (Owners’ equivalent rent of residences) are the largest combined components of the CPI and PCE Price Index. OER and rental costs comprise 33.6% of the CPI (OER, 26.2%; rental costs, 7.4%) and somewhat less in the PCE price index. PCE inflation (the PCE inflation reflects the prices of all goods and services consumed, while the CPI excludes consumption that is financed by third party payers, including Medicare, Medicaid and employer-financed health insurance), which results in a higher weight for health care costs and a lower weight for the shelter components.

Through July, rental costs were up 3.5% yr/yr, down from 5.1% a year earlier and OER is up 4.0% yr/yr, down from 5.3% yr/yr. (Chart 3). Both OER and rental costs are derived by the Bureau of Labor Statistics from the trend in home values. A historical note: prior to May 1983, the CPI included direct measures of mortgage rates. This measurement presumed that homeowners bought their homes at current mortgage rates each month. Beginning win May 1983, the BLS shifted its measure of inflation and replaced the mortgage rate with measures of OER and rental costs. Since then, there have been modest modifications.

Model-based predictions of rental costs and OER. By summer 2021, the Fed had lowered rates to zero and engaged in massive purchases of Treasuries and MBS, and the government had increased deficit spending by a whopping 25% of GDP. Inflation had just begun to accelerate, and Case Shiller Home Price index was soaring. Along with my former colleague Mahmoud AbuGhzalah, I estimated a model that predicted changes of OER and rental costs based on the Case Shiller Hole Price Index. The model proved prescient, as described below. Based on that model, I now predict the OER + rental costs will continue to moderate sharply in the next year to yr/yr increases close to one-half their current pace. And if the Case Shiller Home Price Index continues to be flat-to-down a touch—which seems likely based on their current lofty prices and soft housing demand, the monthly changes in rental costs and OER will subtract from inflation, and their yr/yr increases will head toward zero in 2027.
Our model was a VAR (vector auto regression) with three variables, the Case Shiller Home Price Index, rental costs and OER in which all variables are regressed on their lagged values. This allows the model to estimate the impacts of a shock to one of the variables. Most importantly, it estimates the lagged impacts of a change in the Case Shiller Home Price Index on OER and rental costs. We estimated the model using monthly data going back to 1983 when the BLS changed its method of calculating the CPI to include OER and rental costs.

The model provided a robust statistic fit, with percentage changes in OER and rental costs optimally following the Case Shiller Home Price Index by 12 months and more. In Summer 2021, the model predicted that OER and rental costs, which at the time were rising 3.25%-3.5% yr/yr, would soar and contribute to very high inflation. That’s exactly what happened. This increase in housing inflation somehow came as a surprise to the Fed. When inflation and the shelter component soared in the second half of 2021 and 2022, the Fed began measuring inflation excluding this important sector. When OER and rental cost increases reached 7.5% while the increases in Case Shiller had begun to ebb, we re-estimated the model and it predicted OER and rental costs would recede. They did.

The four recent consecutive monthly declines in Case Shiller (April-July 2025, when its monthly change averaged -0.3%) and its rapidly declining yr/yr increases now predict a continued deceleration in OER and rental costs. Based on the moderating increases of the Case Shiller Home Price Index to date, the monthly and yr/yr increases in OER and rental costs should continue to decelerate significantly further, and subtract from inflation. I expect the Case Shiller index to remain negative to flat for many months to come, and for its yr/yr to eventually decline toward zero. This will contribute to lower inflation, more so in the CPI than the PCE, and will likely be reflected in lower inflationary expectations. As that unfolds, the Fed can be expected to drop its measure of inflation that excludes the housing component.