The persistent myth of high expatriate failure rates

Research output: Contribution to journalArticleScientificpeer-review

247 Citations (Scopus)


This paper provides a critical analysis of research and notably quotations in the field of expatriate failure rates. Over the last three decades it has become almost ‘traditional’ to open an article on expatriate management by stating that expatriate failure rates are (very) high. Virtually every publication on the topic defines and measures expatriate failure as the percentage of expatriates returning home before their assignment contract expires. Of course, premature re-entry might be a very inadequate way to measure expatriate failure. One can easily argue that those expatriates who stay on their assignment but who fail to perform adequately are (potentially) more damaging to the company than the ones who return prematurely. Furthermore, successful completion of a foreign assignment does not mean that the possibility of expatriate failure has been avoided. Sometimes, returning home poses even larger problems than the foreign assignment itself. The repatriate must face re-establishing himself within the home organization and readjusting to the home culture. Failure to do so, for whatever reason, can also be regarded as expatriate failure. These reservations being made, in this paper it is argued that there is almost no empirical foundation for the existence of high failure rates when measured as premature re-entry. The persistent myth of high expatriate failure rates seems to have been created by massive (mis)quotations of three articles. Only one of these articles contained solid empirical evidence on expatriate failure rates and in fact showed them to be rather low.
Original languageEnglish
Pages (from-to)457-474
JournalInternational Journal of Human Resource Management
Issue number2
Publication statusPublished - 1 May 1995
Externally publishedYes


Dive into the research topics of 'The persistent myth of high expatriate failure rates'. Together they form a unique fingerprint.

Cite this