Union wage effects in Australia

an endogenous switching approach

Daehoon Nahm, Michael Dobbie*, Craig MacMillan

*Corresponding author for this work

Research output: Contribution to journalArticle

2 Citations (Scopus)


Using data from the Household and Labour Income Dynamics Australia (HILDA), an endogenous switching model is employed to analyse union wage effects in Australia between 2001 and 2013. An advantage of this approach is that the decision to join a union is treated as potentially endogenous, a function of the wage differential between union and non-union workers, rather than exogenous as is the case in virtually all previous Australian studies. The article finds that the decision to join a union is highly sensitive to the wage differential between union and non-union workers. The article also finds that male (female) union workers with average union characteristics earn 12% (18%) more than male (female) non-union workers with average non-union characteristics. However, a decomposition analysis finds that this difference is due to union workers having better human capital endowments than their non-union counterparts. In addition, they also receive a lower return for those endowments. These decomposition results suggest that union wage effects in Australia may be negative, rather than the small positive effects typically found in the Australian literature.

Original languageEnglish
Pages (from-to)3927-3942
Number of pages16
JournalApplied Economics
Issue number39
Publication statusPublished - 1 May 2017


  • endogenous switching
  • panel data
  • Union wage effects

Fingerprint Dive into the research topics of 'Union wage effects in Australia: an endogenous switching approach'. Together they form a unique fingerprint.

Cite this