Workflow
美国非农就业数据统计服务
icon
Search documents
李稻葵:美国非农就业数据,到底有没有造假?这对我们会有啥影响?
Sou Hu Cai Jing· 2025-08-21 04:02
Core Viewpoint - The recent revisions of U.S. non-farm employment data have raised questions about the reliability of these statistics, suggesting potential political influences and methodological inadequacies in the data collection process [1][3][19]. Group 1: Employment Data Revisions - The U.S. non-farm employment data for May was revised down from 144,000 to 19,000, a reduction of 88% [1]. - The June data was also significantly adjusted, dropping from 147,000 to 14,000 [1]. - The Labor Statistics Bureau conducts monthly surveys of over 100,000 businesses and government agencies, but response rates have declined, leading to larger discrepancies in reported data [3][4]. Group 2: Statistical Methodology Issues - The statistical models used for employment data may no longer reflect the complexities of the current U.S. economy, which is characterized by rapid changes in business dynamics [9][12][14]. - The "business birth-death model" assumes a stable economic environment, which is not applicable in the current context [10][11]. Group 3: Political Influences - Historical patterns show that extreme overestimations of employment data often occur during election years, while significant underestimations are common in the first year of a new presidency [15][16]. - The recent firing of the Labor Statistics Bureau director by Trump is seen as a politically motivated action, reflecting the intertwining of economic data and political agendas [19][20]. Group 4: Implications for Monetary Policy - Weak employment data has increased expectations for interest rate cuts by the Federal Reserve, which could provide more room for policy adjustments in other economies, such as China [21].