Good governance depends on good information. Since its founding, America has gathered data on itself and used that information to improve our understanding of the country’s strengths, weaknesses, and needs. In recent years, however, some of these data systems—especially in the domain of labor market information—have been showing their age and falling short in their mission to equip businesses and workers with the reliable, timely information they need.
In her new MIT Reader article, New York University labor market economist and data expert Julia Lane argues that the labor market information systems we have built over many decades are simply unable to provide accurate and timely measurements of how changes in the economy affect the demand for workers and skills. This problem has been consequential for decades, but the arrival of generative artificial intelligence and other emerging technologies is adding considerably to the urgency of finding new solutions.
When gaps exist in economic data, real-world consequences arise on both the demand and supply side. In the job market, inaccurate or obsolete information about job and skill trends aggravate pre-existing gaps between the needs of businesses and the skills of workers. Moreover, without adequate data, educational institutions and students are less able to anticipate change and adapt training accordingly. In 2020, for instance, then-candidate Joe Biden told the public to “learn to code.” Four years later, generative AI has cut deeply into the demand for basic coding skills. In the wake of this development, advice has shifted to “become an AI prompt engineer.” Such advice has similarly been voided, as ChatGPT and other sites create fully functional “prompt generators.” Trying to predict even the near future by the present is a mug’s game. We need to get ahead of the curve.
To address these challenges, Lane proposes the creation of a new, independent National Center for Data and Evidence (NCDE). This center would operate outside the federal government, but would have access to its data sets. Modeled on institutions like the Urban Institute and MDRC, NCDE would be independently funded and non-partisan, tasked with securely hosting data, creating demand-driven labor market information tools tailored to local and regional needs, and supporting evidence-based policymaking. Working alongside federal statistical agencies, but independent of their bureaucracies, NCDE would be a dynamic, flexible institution designed to meet the demands of an economy undergoing transformation by AI and other emerging technologies.
No system, including the proposed NCDE, can repeal the “knowledge problem.” Market economies will always move much faster than any data system could match. But the data, analysis, and tools NCDE would provide could help business, government, educational institutions, and workers narrow the universe of possible economic futures, improve targeting of limited educational and workforce development resources, and support design of retraining programs tuned to emerging-industry skills.
In the 1990s and early-2000s, the American economy underwent an enormous transition as automation and trade reshaped the labor market. Millions of workers lost factory jobs, and we lacked the information and policies to help them transition to new work. We have been living with the social, economic, and political consequences of that failure for nearly two decades. Lane’s NCDE proposal can help us avoid a labor market “Groundhog Day” by maximizing the gains of the emerging technology boom, while avoiding some of the pain.
The post America’s Labor Market Data System: The Case for a Rebuild appeared first on American Enterprise Institute – AEI.