
"The digitization of everyday life has inundated us with more economic data than ever. And yet, government statistics on real economic conditions seem to be growing ever less reliable. Doubts began spreading even before President Donald Trump fired the commissioner of the Bureau of Labor Statistics. For years now, response rates to surveys that are at the heart of the official statistics have been dropping. Data-gathering agencies release key metrics long after the fact, and they are chronically underfunded."
"So do the choices made by policy makers in the White House and at the Federal Reserve. An incredible range of sophisticated private sources-including real-time payroll data, online transaction records, and consumer-spending databases-offer alternative sources of fine-grained, up-to-the-minute data that could, under the right conditions, be used to supplement a survey-based approach and provide a more dynamic and accurate picture of the state of the economy."
"These issues were underscored last week, when the Bureau of Labor Statistics released preliminary revisions to its jobs data from March 2024 to March 2025 suggesting that the economy added about half as many jobs over that period as originally reported, for a difference of 911,000. This was an unusually large change, but major revisions to jobs numbers happen every year. In that sense, they are normal."
Digital sources now produce extensive, real-time economic information while official government statistics show growing signs of unreliability. Survey response rates have fallen, reporting is delayed, and statistical agencies face chronic underfunding. Business and policy decisions depend on accurate, timely measures of hiring, spending, and investment. Private data sets—such as payroll feeds, transaction records, and consumer-spending databases—can provide fine-grained, up-to-the-minute indicators to supplement survey-based methods. Recent large revisions to jobs data highlighted the mismatch between preliminary reports and later estimates, raising questions about continued reliance on slow, outdated data-collection techniques.
Read at The Atlantic
Unable to calculate read time
Collection
[
|
...
]