The government shutdown provides an unexpected test case for an increasingly urgent question: What would happen if official economic statistics became less useful, whether through declining resources, loss of expertise, or political interference? While a temporary shutdown differs from long-term erosion of statistical quality, it offers valuable lessons about what private sector data can and cannot do.
When the September jobs report was delayed, multiple private sector alternatives stepped in. ADP continued publishing payroll figures, newer entrants like Revelio Labs joined the field, and firms including the Carlyle Group weighed in with payroll growth estimates. Historical analysis suggests these sources are rough and imprecise predictors of the official nonfarm payroll number, with methodological differences in coverage, revision timing, and data collection making direct comparisons tricky.
More importantly, while private firms can estimate overall employment from payroll records or job postings, they struggle to measure unemployment rates and other household survey indicators. Companies that can see who is on payroll or searching for a job might have a much harder time seeing who isn't working, isn't searching, or has left the labor force entirely. But including these people is essential for calculating key labor market measures like the labor force participation rate and unemployment rate.
This limitation is especially problematic right now. Given uncertainty surrounding immigration rates and labor force growth, interpreting the nonfarm payroll number has become unusually difficult. Without knowing how fast the labor force is growing, it's hard to determine the break-even level of job growth that keeps the labor market steady. Policymakers and analysts have been relying more heavily on household survey measures like the unemployment rate—precisely the indicators that private sector sources struggle to replicate.
There do exist alternative survey efforts about jobs, like the Real-Time Population Survey and the US Survey of Working Arrangements and Attitudes. The polling firm Morning Consult regularly publishes an unemployment index; however, a pop-up window notes that this tracker is only “temporarily available for free” and its methodology explanation is paywalled, illustrating how private sector data are typically less transparent and open to researcher scrutiny than official statistics are.[1]
The shutdown also exposed significant gaps in private sector inflation measurement. Several well-regarded sources track prices, including the Pricing Lab, which has been helpful in estimating the impact of tariffs on prices. However, these generally focus on goods sold at retail establishments. The problem? Goods represent only 36 percent of consumer expenditures, while services account for the remaining 64 percent of the Consumer Price Index (CPI).
Some service prices can be tracked privately: The Zillow Observed Rent Index helps predict CPI shelter prices, which constitute 35 percent of the CPI expenditure basket. But the remaining harder-to-measure segment of services excluding housing is still 29 percent of CPI and an even larger share of the Federal Reserve's preferred inflation measure, the Personal Consumption Expenditures price index, where housing carries much lower weight. The private sector simply cannot replicate the comprehensive view of inflation across goods and services that statistical agencies provide.
Perhaps the most revealing lesson involves data that fewer people pay attention to but that forms essential scaffolding for the statistics many people do watch closely. In late September, the Bureau of Labor Statistics (BLS) delayed the Consumer Expenditure Survey by about five weeks for technical reasons (a separate issue from the shutdown, but illustrative nonetheless).
The Consumer Expenditure Survey doesn't produce data that investors trade on. It's annual, not high-frequency, so it isn't current enough for many business decisions. But it's critically important for establishing how to weight different goods and services in inflation measures, based on actual consumer spending patterns. Without this survey of how spending patterns change over time, inflation measures would become less accurate. This is exactly the kind of essential economic infrastructure the private sector is unlikely to produce. Firms lack visibility into the full range of consumer spending, and they lack incentives to invest in datasets without immediate business or financial applications.
The shutdown reinforces what statisticians and economists have long argued: Private sector data is valuable but cannot replace official statistics. Private sources offer speed, specificity, and innovation that complement government data. They can track emerging trends, provide granular detail for particular sectors or geographies, and offer near real-time updates. But they cannot match the breadth, representativeness, consistency, transparency, or public commitment of official statistics.
The shutdown also reminds us that even when private firms do produce useful data, their plans and strategies can change. Recently ADP stopped providing its payroll data to the Federal Reserve, ending a long-standing arrangement, but also introduced a public-facing weekly payroll report. Private companies can change their business models, start or stop data-sharing agreements, or simply go out of business. Official statistics, by contrast, provide the continuity and commitment that economic policymaking and decision-making require.
As I've written previously, private companies can be excellent partners in producing economic statistics, particularly as survey response rates decline. But partnership requires a strong public sector foundation. The shutdown shows us what's at stake if we get this wrong.
Finally, this shutdown inspires a call to action: Administrations should designate the production of all principal federal economic indicators as essential activities that continue during shutdowns, and Congress should consider legislation to ensure this. During this shutdown, the White House directed BLS to call back employees to publish the September CPI because that report determines the annual Social Security cost-of-living adjustment. But that’s hardly the only essential application of federal statistics. Users of other statistical products should speak loudly about how important official data are for public spending, policy decisions, and business operations.
Note
1. Pop-up and paywall blocker on Morning Consult site viewed on October 29, 2025, at 9 am ET.
Data Disclosure
This publication does not include a replication package.
