U.S. CPI Reliance on Estimated Data Hits Record 36% Amid Staff Shortages

The proportion of estimated data in the U.S. Consumer Price Index (CPI) has surged to 36%, marking a record high. This increase is primarily due to staff shortages and the suspension of data collection in certain regions. Economists warn that the overuse of estimation methods and the overall decline in the number of price quotes are "reducing the quality of the CPI and its ability to track actual inflation."

The credibility of a key U.S. inflation indicator is under increasing scrutiny as the country becomes more reliant on estimated rather than directly collected data to compile the CPI. This situation has sparked widespread criticism ahead of the Federal Reserve's decision.

On September 11, the U.S. Bureau of Labor Statistics revealed that in the August CPI report, the proportion of prices generated using a method known as "different-cell imputation" or "heterogeneous unit estimation" had risen from 32% the previous month to 36%. This figure is not only the highest since records began in 2019 but also significantly higher than the 9% recorded in February of this year.

This method is used when the price of a particular item cannot be obtained in a specific region, and data from other geographical areas for the same item is used as a substitute. It is considered a less accurate "second-best option" for handling missing data.

This revelation comes as the U.S. releases its August CPI data, which largely met market expectations and reinforced market bets on an imminent Fed rate cut.

Economists believe that the increased use of estimation methods will compromise data accuracy. Alan Detmeister, an inflation economist, stated that while it is not yet clear how significant the impact of this substitution is at the macro level, the more pressing issue is that the total number of price quotes in the CPI has been declining for the past decade, which will increase data volatility.

He emphasized in a teleconference last month: "Overall, this is reducing the quality of the CPI and its ability to track actual inflation."

The U.S. Bureau of Labor Statistics' growing reliance on estimation methods is closely linked to the staff shortages and data collection challenges it has faced in recent years.

Since the current administration took office, the number of employees at the Bureau has been steadily declining. Although the Bureau has not disclosed the exact number of departures, advocacy groups that maintain contact with current and former employees estimate that the Bureau's workforce has decreased by at least 20%. The 2026 budget proposed by the administration will further reduce the Bureau's resources.

In addition, the Bureau itself has encountered data collection issues. In June, the Bureau stated that due to resource constraints, it had suspended CPI sample collection in three metropolitan areas, but its analysis suggested that this would have a minimal impact on the overall CPI. However, by the end of July, the Bureau reported that sample collection in approximately 15% of other regions had also been suspended, but did not specify how this would affect the inflation rate.

Omair Sharif, president of Inflation Insights LLC, noted that these two notifications about the suspension of data collection indicate that approximately 19% of prices in the CPI are being estimated, higher than the 5.1% at the end of 2022. "If you stop collecting data from certain regions nationwide, the proportion of 'different-cell imputation' will increase."

Concerns about the quality of the Bureau's data come amid significant political pressure. Following an unusually large downward revision in the July employment report, the administration dismissed the Bureau's director and accused her of manipulating data for political gain without evidence.

Amid a series of controversies, the Office of the Inspector General of the U.S. Department of Labor, which oversees the Bureau, announced on Wednesday that it is launching an investigation into the "challenges" the Bureau faces in adjusting the CPI, Producer Price Index (PPI), and employment data.