- 1. See ILO Resolutions Concerning Statistics of Employment in the Informal Sector adopted by the 15th International Conference of Labour Statisticians (ICLS), January 1993, as well as the following 17th Conference in 2003.
- 2. It is hard to argue for reverse causality since it is more typical for remittances to be sent when households face hard times. In other words, the literature points to remittances being sent when unemployment increases (a positive relationship), as an insurance (see for instance Amuedo-Dorantes and Pozo, 2006b, on Mexican migrant households). However, it is clear that both variables may be influenced by economic reasons related to the crisis; that is, unemployment rose and remittances fell because of dire economic circumstances in both home and sending countries.
- 3. Examples in Eastern Europe are provided by Silasi and Simina (2007) for Romania and Thaut (2009) for Lithuania.
- 4. Several studies show that remittances decrease proportionally to the length of stay in the countries of immigration. Funkhouser (1995) argues that remittances between the United States and Nicaragua drop by three dollars for every month spent in the US. DeSipio (2002) estimates that a 1% increase in time spent by Mexicans in the United States lowers the probability of transferring money home by 2%. Filipski and Taylor (2011) show that the annual rate of decay of remittances to Mexico is about 3.5%.