On September 11, the U.S. Bureau of Labor Statistics released data on real income: in August, the average hourly wage for all private non-farm employees increased nominally by 0.3%, but during the same period, the Urban Consumer Price Index rose by 0.4%. As a result, the average hourly wage after adjusting for inflation decreased by 0.1% on a month-to-month basis. The average weekly working hours increased by 0.3%, which offset the decline in the purchasing power of hourly wages, leading to a 0.2% increase in real weekly wages on a month-to-month basis. These figures bring a more relatable question to mind: although the amount on our paychecks has increased, it doesn't necessarily mean we can buy more goods and services.

The year-on-year performance also showed divergence. From August 2025 to August 2026, the actual average hourly wage for all employees decreased by 0.3%, while the average hours worked increased by 0.6%, resulting in a 0.3% increase in the actual weekly wage. For employees in production and non-managerial positions, the actual hourly wage also decreased by 0.1% month-over-month, and so did the actual weekly wage by 0.1%, as their average hours worked did not increase; however, the year-on-year actual hourly wage decreased by 0.1%, but the actual weekly wage increased by 0.1% due to a 0.3% increase in hours worked. Changes that are similar in terms of average values have different impacts on different groups of workers.

Overtime supports weekly income, but it cannot conceal the fact that purchasing power per hour is declining.

Actual hourly wage measures how much goods and services can be purchased with one hour of labor income, while actual weekly wage also takes into account how many hours a person works in a week. In August, the purchasing power per hour for all employees decreased, yet they earned higher weekly incomes due to increased working hours. For individuals, this may mean that they need to work more hours to improve their income, rather than relying on an increase in the value of their labor. If the increase in working hours comes from peak seasons or overtime, it may not be sustainable; however, if companies consistently increase full-time schedules over the long term, it may reflect that demand remains strong.

Production and non-managerial employees better illustrate this distinction. Their average working hours did not change that month, so the decrease in actual hourly wage was directly reflected in their actual weekly wage. Management and high-paying positions have different weights in the overall average, and changes in the industry structure can also affect the average wage. If there is an increase in employment in high-paying industries in a certain month, the average hourly wage may rise, even if employees in the same position do not receive a raise; conversely, an increase in low-paying positions may also lower the average. Therefore, this data is suitable for observing macro purchasing power and does not equate to the individual wage trajectory of each employee.

A nominal hourly wage increase of 0.3% is not necessarily bad news in itself, but the problem lies in the fact that CPI increased by 0.4% that month. Gas prices contributed more than one-third of the CPI monthly increase, and the impact of energy costs quickly eroded the wage gains. If household consumption relies more on driving, the actual pressure may be higher than average; however, those who use public transportation or families with lower fixed housing costs may be less affected. The official measure of real wages uses the overall price index and does not recalculate the purchasing power for each individual family.

A year-on-year decrease of 0.3% in actual hourly wages means that the nominal wage growth over the past year has not kept up with inflation; however, actual weekly wages have increased by 0.3%, mainly due to a 0.6% increase in working hours. If this trend continues, it will lead to two constraints: workers will have less disposable time, and businesses will face higher labor costs. Consumption can be temporarily sustained by more working hours, but it is not as stable as growth in both productivity and actual hourly wages. If the economy then cools down and working hours are cut first, weekly income may decline faster than hourly wages.

To assess consumption resilience, one cannot rely solely on wage growth; factors such as working hours, prices, and distribution also need to be considered.

There is often a superficial contradiction in macro data where "salaries are rising, yet residents still feel financially strained," but actual income is the bridge that connects the two. Nominal wages determine the amount received, inflation determines purchasing power, and hours worked determine the total weekly income. If only a 0.3% increase in wages is reported, it will overlook the faster rise in prices; if only a 0.1% decrease in actual hourly wages is reported, it will ignore the overall increase in weekly wages due to more hours worked. A complete analysis must present all three factors simultaneously.

Whether consumption can continue depends on household savings, debt, and necessary expenses. A slight increase in actual weekly wages may support retail and service consumption, but rising energy and housing costs will direct more income towards necessities, reducing discretionary spending. High-income households have more assets and savings buffers, while low-income households rely more on their current wages; the average does not reflect this disparity. Credit card balances, delinquency rates, and consumption data across different income levels can help determine whether the growth is widespread.

For businesses, the combination of hourly wage and hours worked affects cost planning. If orders increase, increasing working hours may be more flexible than hiring new staff immediately; however, if the overtime rate is too high, both marginal costs and employee fatigue will rise. Industries such as retail, catering, and logistics also need to include fuel and wages in their budgets. Simply considering the nominal increase in wages as the entire change in labor costs overlooks benefits, overtime pay, and the structure of job positions.

For policy observers, a decline in real income does not automatically imply interest rate cuts; nominal wages and service inflation may still remain sticky. Similarly, a single-month energy shock does not necessarily mean a long-term deterioration in wage purchasing power. The data on real income itself is also subject to revisions in wage and price statistics, so changes of 0.1 percentage points should be viewed more as a direction indicator than as a precise measure of individual welfare. It is changes in purchasing power and working hours over several months that can better indicate whether the labor market is truly weakening. The September data will be released on October 14th, and at that time, attention should be paid to whether gasoline prices fall, whether nominal hourly wages can continue to grow, and whether average working hours remain stable. The most accurate conclusion for now is that wage increases in August did not keep up with inflation, and overall weekly income was temporarily sustained by longer working hours, while production workers and non-managerial staff did not benefit from the same buffer in working hours.