Shadow price of working in the shadows: services industry evidence

ABSTRACT

In this paper, we use an exogenous policy variation in the labour market to determine the wage gap between formally and informally employed workers. For our purposes, ‘informal employment’ describes employees who are not officially registered with any social security scheme. We use self-reported employee registration status to identify such workers, but the choice of working unregistered is not exogenous. Nevertheless, through an amnesty that was extended to only some workers in the labour market, we reduce the endogeneity problem, enabling estimation of the wage gap between these two groups. Our two-stage least square estimates reveal that the hourly wage penalty of working in the shadows is as high as 59%, and the monthly salary penalty is around 66%. Moreover, the wage gap is higher (as high as 70%) for those working in the services sector, as unregistered workers in this sector tend to be low skilled and low educated, and the monitoring of this sector is more difficult. Our analysis contributes to the literature by using an instrumental variable to treat the endogeneity of workers’ registration status. In addition, it shows that people working informally in the services industry receive a higher average wage penalty than other informally employed workers.

KEYWORDS:

• Informal employment
• migration
• services sector
• shadow economy
• wage differentials

Acknowledgements

The authors would like to thank to Murat Kirdar, two anonymous referees and participants at the 5th International Conference on ‘The Shadow Economy, Tax Evasion and Informal Labour’ in July 2017 for their valuable feedback on earlier drafts of this paper. All remaining errors are our own.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. According to the Household Labour Survey results, informal employment amounted to as much as 25% of total employment in 2004 (Figure 1).

2. Although the cost increases, if a firm does not register an employee, it will avoid paying social security contributions for him or her, and the employee will avoid paying taxes; thus, they can negotiate a higher salary than if the same employee were registered.

3. For the main analysis, we are not using 2005 data, as we explained in the previous paragraph. Thus, when we discuss the sample size and summary statistics of the variables used, they refer to this pool. However, we will use 2004–2005 pooled data to compare our results, which will be discussed later.

4. There are other classifications, such as ‘Other Services’, that could be included in this category, but we chose to use only these two main activity areas.

5. This number might look inconsistent with the number in Table 1, but keeping in mind that the numbers in Table 1 also include self-employed, public services and female workers, who are more likely to be working in the services sector.

6. Monthly hours of work were obtained by multiplying weekly hours by 4.25.

7. Economic growth reached to double digits in 2004, 2005 and 2006, being 15.4, 13.5 and 13.2, respectively, which in effect pushed wages up in the labour market including unregistered employees.

8. For comparison purposes, we conducted the same analysis using 2004 and 2005 data. Due to possible complications from taking up the amnesty offer, we choose not to discuss the results at this point. The results are available upon request.

9. The exclusion of the interpretations of OLS estimates are suggested by an anonymous referee.

10. The first-stage results of the estimations are not presented here but are available upon request. The coefficient of the ‘services’ dummy in the first stage was statistically insignificant.

11. We cannot have services interact with registration status in these models; since registration status is estimated in another step, their interaction would require a more advanced statistical approach, which is beyond the scope of this paper.

12. In fact, we had annual data until 2011 and found that the percentage of unregistered employment in Northern Cyprus started to increase after 2010. The government had another amnesty in 2012, which is a major problem: since people anticipate that a new government will extend an amnesty in the near future, they are less inclined to take up amnesty offers immediately.

13. We conducted the same analysis for the construction sector, which is also a problematic sector in Northern Cyprus in terms of having unregistered employment. That sector also has a significant wage penalty for working in the shadows, but it is less than the one in the services sector.