In this paper we look at the costs and benefits for Iceland from joining the EMU from a labour-market perspective. We conclude that none of Mundell's three criteria for an optimal currency area are at present fulfilled for Iceland and the initial Euro zone. Shocks to the Icelandic economy are found to be asymmetric with those experienced in other countries, nominal wages rigid, and migration limited. The painful adjustment of the Faeroese economy to macroeconomic shocks in the early 1990s suggests that the disadvantages of not having a separate currency can be substantial if nominal wages are rigid. Substantial variation in labour market participation and frequent adjustments of the exchange rate seem to have held unemployment in check in Iceland, at least until around 1988.