再探特朗普解雇劳工统计局长背后：美国就业数据统计有什么问题？

Core Viewpoint - The recent drastic downward revision of U.S. employment data by the Bureau of Labor Statistics (BLS) has raised questions about the credibility of official economic statistics, especially following President Trump's controversial dismissal of BLS Director Erika McEntarfer due to disappointing job numbers [1][4]. Data Revision and Employment Statistics - The BLS reported only 73,000 new jobs added in July, significantly below market expectations, and revised down the previous months' data by 258,000 jobs, with June's figures adjusted from 147,000 to a mere 14,000 [1]. - A senior statistician explained that the large downward revision is primarily due to data collection lag, where initial surveys did not capture all responses, leading to significant corrections later [1][6]. - The BLS's employment data collection methods have faced scrutiny, with claims of systematic overestimation of job growth followed by substantial downward revisions [4][5]. Model Limitations and External Factors - The BLS's predictive models have been criticized for failing to adapt to significant external economic changes, resulting in inaccuracies during economic transitions [3][6]. - The response rate for employment surveys has dropped from around 60% pre-pandemic to 42.6% as of March this year, indicating challenges in data collection [6][7]. Budget Cuts and Staffing Issues - The BLS has faced budget cuts and staffing reductions, which have been attributed to the Trump administration's hiring freeze and budget constraints, impacting the agency's statistical capabilities [11][12]. - The BLS is expected to face an 8% budget cut in the next fiscal year, along with a reduction of over 150 staff members, which could further compromise data quality and timeliness [12][13]. Data Collection Challenges - The BLS's reliance on estimation methods, particularly in the context of reduced data collection efforts, has led to increased error margins in reported statistics [8][12]. - The agency's data collection for the Consumer Price Index (CPI) has also been scaled back, with a significant reduction in local data collection efforts, leading to a reliance on broader regional data [7][8].