Working Paper: NBER ID: w1859
Authors: Harry J. Holzer
Abstract: In this paper I investigate the use of different search methods by unemployed youth. I present a job search model which shows that search method choices should be related to their costs and expected productivities, as well as other factors such as nonwage income and wage offer distributions. I then present empirical evidence on the use of these methods and their effects on employment outcomes. These results show that the most frequently used search methods, which are friends and relatives and direct applications without referral, are also the most productive in generating job offers and acceptances. Econometric evidence then shows that the number of methods used is affected by factors which presumably reflect market opportunities as well as income sources and needs. While the use of specific search methods respond differently to these factors, they are chosen in a manner which generates positive average effects on employment outcomes for those who use them. The results are thus consistent with the search model presented here.
Keywords: job search; unemployment; youth; search methods; employment outcomes
JEL Codes: J64; J65
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| expected job offers (J68) | number of search methods used (C00) |
| marital status (J12) | number of search methods used (C00) |
| layoff status (J63) | number of search methods used (C00) |
| number of search methods used (C00) | probability of receiving job offers (J68) |
| number of search methods used (C00) | probability of accepting job offers (M51) |
| friends and relatives search method (C92) | employment outcomes (J68) |
| direct applications search method (C00) | employment outcomes (J68) |
| search method choices (C52) | employment outcomes (J68) |