In Simple Words
Think of shopping at a large supermarket with quality items scattered across many aisles. A shopper prefers organic produce and selects it when spotted, but walking down every aisle takes physical effort, so they leave many top-tier items on the shelves. Lowering the effort needed to fetch items allows shoppers to acquire more top-quality goods rather than selecting random products.
Jobseekers on a Lahore job platform
- Text-only outreach
- Applications favoured higher-pay jobs
- Assigned text plus calls
- Applications still favoured higher-pay jobs
Applications across salary groups. No clear difference in their distribution across the two arms; this does not establish equal job quality.
See source →Research question
Standard labor economics assumes search is either fully directed toward top-value vacancies or purely random. In practice, jobseekers actively target higher-value opportunities but miss numerous high-return vacancies because initiating applications requires non-trivial cognitive effort.
Participants
N = 1,116,952 jobseeker-vacancy matches across 1,343 vacancies in Lahore, Pakistan.
Study Design
Field experiment comparing treatment vs control application distributions across salary and match quality quintiles, contrasted with a randomized vacancy listing order experiment (N = 938,284 matches).
Results
Jobseekers apply to higher-salary jobs at higher rates, but control jobseekers apply to under 0.1% of top-salary quintile matches. Marginal applications induced by phone call initiation match the distribution of inframarginal applications across salary quintiles (p = 0.739), whereas forcing random search via vacancy ordering yields sharply lower interview returns (2.4% vs 6.3%, p = 0.077).
Takeaway
Deploy search and matching platform algorithms that surface high-value matches while simultaneously eliminating step-initiation barriers for those targeted matches.