Despite providing real-time labor market data, firms like Revealio Labs depend on foundational government statistics to reweight their datasets for accuracy. This calibration process is only needed about once a year, allowing their models to function for a considerable time during government data blackouts without significant degradation.
Related Insights
The recent government shutdown will create a permanent void in crucial economic data for October. While statistics like payrolls might be collected retroactively, survey-based data such as the Consumer Price Index (CPI) and household unemployment figures are likely lost forever due to recall bias, creating a black hole in the historical record.
A shutdown doesn't just delay data reports; if it extends into mid-month, it prevents the government from conducting the surveys needed for future reports. This disrupts the entire data collection pipeline, causing a ripple effect that can obscure economic trends for months after the government reopens.
The absence of official government data during shutdowns creates a 'data void' that heightens economic anxiety. Economists and the public are forced to over-rely on anecdotal evidence, like conversations with Uber drivers, which makes the economy feel more volatile and difficult to assess accurately.
Private firms like ADP have business incentives that may conflict with the public's need for consistent economic data. ADP's recent decision to stop providing weekly data to the Fed during a government shutdown highlights this tension and the irreplaceability of official government statistics.
While historical ADP charts seem to track official Bureau of Labor Statistics (BLS) data, this is misleading. In the moment, ADP's estimates are often inaccurate. The firm revises its historical data months later to align with the official BLS numbers, creating an illusion of real-time accuracy.
Former BLS Commissioner Erica Groshen explains that data revisions are a designed feature, offering users a choice between fast but less precise initial data and slower but more accurate final data. It's an intentional balance between timeliness and accuracy.
Revealio Labs scrapes 105M US professional profiles, primarily from LinkedIn. To correct for biases (e.g., overrepresentation of tech workers), they reweight the data using BLS industry and occupation statistics. A Bayesian model then adjusts for the typical 3-month lag in users updating their job status, enabling a real-time 'nowcast'.
The government's failure to release key economic reports (jobs, GDP, inflation) creates a dangerous information vacuum, forcing the Fed and businesses to operate without instruments. This void presents a significant business opportunity for private companies to develop and sell alternative economic data streams and forecasting models to fill the gap.
The Federal Reserve is not 'flying blind' during government shutdowns that halt official statistics. It uses a composite of alternative indicators for the labor market and inflation, providing enough of a signal to stick to its pre-planned policy path, such as proceeding with scheduled interest rate cuts.
A government shutdown lasting several weeks poses a greater threat than just delayed reports. Data collection for time-sensitive indicators like the Consumer Price Index becomes impossible or unreliable, as prices can't be collected retroactively and people's recall fades, potentially forcing agencies to skip a month of data entirely.