This paper uses household survey (Sakernas) data from the 1996 and 2004 to estimate the determinants of earnings in Indonesia. The Indonesian labour market is segmented, with a majority of workers engaged in informal-sector occupations, and earnings data are available only for formal-sector workers (salaried employees). This posed problems for the estimation of earnings equations, because selection into different labour market statuses is likely to be non-random. In order to describe selection into different labour market statuses we use the most general version of the method proposed by Dubin and McFadden (1984), which Bourguignon, Fournier and Gurgand (2007) proved to be preferable to other available multinomial selection methods. We also deal with reverse causality between education attainment and earnings by estimating the selection equations using an instrumental variable technique. Our findings cast doubt on the use of a binomial selection rule and suggest that workers with higher levels of educational attainment are most likely to find a job in the formal sector, and that the informal sector is perceived by those workers who cannot obtain a job in the formal sector as an alternative to inactivity. This Working Paper relates to the 2008 OECD Economic Assessment of Indonesia (www.oecd.org/eco/surveys/indonesia).