Dependence Duration and Labor Market Conditions: Evidence from the United States

How to cite

Notowidigdo, Matthew; Lange, Fabian; Kroft, Kory. 2016. "Dependence Duration and Labor Market Conditions: Evidence from the United States." AEA RCT Registry. AEARCTR-0001453. https://doi.org/10.1257/rct.1453-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Matthew Notowidigdo* , Fabian Lange , Kory Kroft

* Primary investigator

Registry ID
AEARCTR-0001453
DOI
https://doi.org/10.1257/rct.1453-1.0
First registered
Last updated
Status
Completed
Trial dates
2011-08-01 – 2012-07-31
Countries
United States of America (North America)
JEL codes
J64
Registry record
http://www.socialscienceregistry.org/trials/1453

Abstract

This project studies the role of employer behavior in generating "negative duration dependence" -- the adverse effect of a longer unemployment spell -- by sending fictitious resumes to real job postings in 100 U.S. cities. Our results indicate that the likelihood of receiving a callback for an interview significantly decreases with the length of a worker's unemployment spell, with the majority of this decline occurring during the first eight months. We explore how this effect varies with local labor market conditions and find that duration dependence is stronger when the local labor market is tighter. This result is consistent with the prediction of a broad class of screening models in which employers use the unemployment spell length as a signal of unobserved productivity and recognize that this signal is less informative in weak labor markets.

Experimental design

This study sought to determine how the length of time out of work affects a worker's job market opportunities. This was done using a large-scale resume audit study, in which fictitious resumes were submitted to real, online job postings in each of the 100 largest metropolitan areas (MSAs) in the U.S., and "callbacks" from employers were tracked for each submission. For each resume, both the employment status and the length of the current unemployment spell were randomized from 1 to 36 months (if the worker is unemployed). The study was conducted in a single major online job board in the U.S. focusing on three job categories: administrative/clerical, customer service, and sales. Within these job categories, roughly 12,000 fictitious resumes were sent to roughly 3,000 job openings located in the largest 100 Metropolitan Statistical Areas in the U.S. according to population. In choosing a job to apply to, MSA and job category combinations were randomly sampled without replacement from the distribution. Upon being assigned an MSA and job category, jobs were sought out within the pre-determined MSA for the pre-determined job type. Once a job was identified, four fictitious resumes were constructed and sent to the job opening. The design of these resumes was based on roughly 1,200 real resumes that were manually collected from various online job message boards.

Design details

Randomization method
Fictitious resumes were created following a six-step process: 1. Candidate gender distribution was predetermined for each job type 2. Candidate names were randomly generated based on common frequency census data 3. Fictitious home addresses, local phone numbers, and email addresses of candidates were created based on listings from real resumes from that Metropolitan Statistical Area 4. Fictitious resumes were given job histories, educational histories, and objective summaries to match the job for which the candidate was to apply 5. A measure of quality was defined for each resume 6. Employment status and length of the current unemployment spell was randomized for each candidate. Individuals were randomly assigned to the employment status "Employed" with probability 0.25. Individuals that were not assigned to the Employed treatment were unemployed and were randomly assigned an unemployment duration according to a discrete uniform distribution of between 1 and 36 months.
Randomization unit
individuals
Clusters
(no clusters)
Observations
12,054 resumes submitted to 3,040 jobs
Arms
9,236 resumes had ongoing unemployment spells of at least one month, with the remaining 2,810 resumes conveying that the worker was currently employed
Primary outcomes
(Dependent variable for all main endpoints was "received callback for interview") The Effect of Unemployment Duration on Probability of Callback Sensitivity to Alternative Controls and Alternative Specifications How Does Duration Dependence Vary With Labor Market Conditions? Duration Dependence by Local Labor Market Heterogeneity in Duration Dependence by Resume and Job Characteristics Callbacks for the Currently Employed and Labor Market Conditions

Papers from this trial

  1. Kroft, Kory, Fabian Lange, Matthew J. Notowidigdo. 2013. "Duration Dependence and Labor Market Conditions: Evidence from a Field Experiment." The Quarterly Journal of Economics. https://www.povertyactionlab.org/sites/default/files/publications/3814%20Duration%20Dependence%20and%20Labor%20QJE%202013.pdf

Cited by

No citation data available for this registration yet.

Keywords