Is there an informal employment wage penalty in Egypt? Evidence from quantile regression on panel data

The US dollar exchange rate was equal to 6.04 EP in May 2012, and it was an average of 7.05 in 2014. Thus, 700 EP was 116 US dollars and 1200 EP was 170 US dollars.

For an analysis of the public versus private formal wage gap in Egypt, see Tansel et al. (2018).

In the ELMPS the computation of the hourly wage depends on the type of worker. For regular workers, there are questions about the hours per day and the days per week that are multiplied to get hours per week. This is then multiplied by the number of weeks worked in the past 3 months to get the hours worked in the past 3 months. The regular workers are also asked to report the amount and the frequency of different types of wages (basic, overtime, primary job, secondary job, etc.) which are used to calculate the wages in the past 3 months. This is divided by the number of hours worked in the past 3 months to get hourly wages. As the last step, the hourly wage is deflated by CPI base 2012 to get the real hourly wage for the regular workers. For irregular workers, there are questions about the number of weeks worked in each of the last 3 months and the number of hours per week in each of those months to estimate the number of hours worked in the last 3 months. The number of hours per week in the past 3 months is added up and divided by the number of weeks worked to get the number of hours per week. There are data on the daily wage (it is the usual type of wage) for irregular workers and their usual hours per day (asked in a question). This is used to calculate the hourly wage. There are do files that do the computations for both wages and hours in the Open Access Micro Data Initiative (OAMDI) of ERF. We have benefited from these files as well as from personal correspondence with Caroline Krafft.

We report the estimates of the female sample in Tansel et al. (2015). However, the labor force participation of women in Egypt is very low and most women are either inactive or work as UFW (Tansel and Ozdemir 2014). Issue of women’s selection into employment is not addressed since selection within the QR framework is a nonstandard econometric procedure. Further, the number of observations is small in the female sample. The selection into labor force is less of an issue in the male sample (which is the focus of this paper) due to men’s high participation rate.

Full estimation results reported in Tansel et al. (2015) indicate the following. The wage returns to experience (as proxied by age) are positive and exhibit a quadratic relationship. The wage returns to age decreases as one moves to higher quantiles in the QR estimation, while there is no discernable pattern across the quantiles in the FEQR estimation. The returns to education is quite low, 1.5% at the mean, and increases smoothly across quantiles and about 2% at the highest quantiles (insignificant at the lowest quantiles). Largest returns are attained in the construction sector (which is the sector with most concentration of informality) compared to manufacturing. The wage returns in the construction sector are highest at the lowest quantile and decrease across quantiles. Assaad (1997) finds segmentation within the construction sector itself. The wage returns are higher both at the mean and across quantiles when unobserved heterogeneity is taken into account. The wage returns are lower in the trade and service sectors of economic activity compared to manufacturing. The wage returns are higher in Greater Cairo than in all of the other regions. Finally, the wage returns are significantly higher in 2006 and 2012 than in 1998.

These estimates are available from the authors upon request.