Are Labor Market Indicators Telling the Truth? Role of Measurement Error in the U.S. Current Population Survey
IMF Working Papers, February 26, 2019
Source details
- Canonical URL
- Are Labor Market Indicators Telling the Truth? Role of Measurement Error in the U.S. Current Population Survey
Other formats
Bibliographic details
- Authors: Ippei Shibata
- Published: February 26, 2019
- Series: IMF Working Papers
- DOI: https://doi.org/10.5089/9781498300452.001
Key findings
- Measurement error from classification (respondents misreporting true labor force status) is substantial in the U.S. Current Population Survey (CPS).
- Once corrected using a latent variable approach, the unemployment rate is on average 0.8 percentage points (ppts) higher than the official unemployment rate.
- The unemployment rate correction reaches a maximum of 2.0 ppts higher between 1996 and 2018.
- Correcting for misclassification alters previous assessments of business-cycle drivers:
- Job separation accounts for more of the unemployment fluctuations after correction.
- Participation accounts for fewer of the unemployment fluctuations after correction.
- The methodology requires labor force status observed for three periods and can be applied to other labor force surveys with that feature.
Methodology and analysis
- Uses a latent variable approach to correct for measurement (classification) error in CPS labor force status.
- Quantifies contributions to business-cycle fluctuations in the unemployment rate from:
- Job separation
- Job finding
- Participation
- Re-assesses transition probability contributions to unemployment fluctuations after correcting for misclassification.
Policy and research implications
- Policymakers relying on labor market indicators should account for classification error in labor force surveys to avoid understating unemployment.
- Re-evaluation of policy conclusions based on uncorrected CPS transition dynamics may be warranted, given that separation plays a larger role and participation a smaller role than previously estimated.
- The proposed methodology offers a tool for researchers and statistical agencies to improve labor market statistics when three-period observations are available.
Publication metadata (selected)
- Author: Ippei Shibata
- Date: February 26, 2019
- Series: IMF Working Paper No. 2019/040
- Pages: 40
- DOI: https://doi.org/10.5089/9781498300452.001
Source: Are Labor Market Indicators Telling the Truth? Role of Measurement Error in the U.S. Current Population Survey, Ippei Shibata, IMF Working Papers (2019), accessed from the IMF publication summary.
Content in this bundle
- Working Paper