Failure of unravelling theory? A natural field experiment on voluntary quality disclosure

Classic "unravelling" theory holds that buyers should treat with maximal scepticism sellers who withhold verifiable information relating to their quality, as buyers infer from such non-disclosure that the seller possesses the lowest possible quality. This study is the first to use a natural field experiment to test this proposition, and the first to test it in a labour market context. We sent out 12,301 job applications, varying the information on degree classification - a signal of academic quality - that the applicant presented to the employer. Our results do not support unravelling theory. Applications which left degree classification undisclosed were significantly more likely to receive positive responses from employers than those disclosing the lowest possible degree classification. Employers treated non-disclosing applicants similarly to those disclosing mid-scale classifications, suggesting the extent to which adverse inference is drawn from missing information is limited. Evidence is presented against the alternative interpretation that non-disclosure success is driven by recruiters' usage of software tools.