Tax compliance: when do employees behave like the self-employed?

Abstract

Previous studies of the underreporting of income for tax purposes have used private employees as the benchmark to which other groups’ compliance was measured. In this paper it is suggested that there are a number of circumstances when there will be an incentive for private employees and their employers to collude to understate employee wages and salaries for purposes of taxation. The existence of high marginal tax rates of income tax combined with high social security payroll taxes are the typical conditions that stimulate this behaviour. These conditions are present in North Cyprus. This paper examines a rich source of household consumption expenditure and income data for North Cyprus that allows one to separate out the consumption expenditures made by the self-employed, private employees and civil servants over specific periods of time. From the comparison of consumption expenditures on food by these three groups it is possible to estimate how much self-employed and the private employees understated their incomes as compared to the civil servants. It is found that in North Cyprus private employees understate their incomes by approximately the same proportion of their incomes as do the self-employed.

Acknowledgement

The authors wish to thank Pradip Ghimire for his assistance in the completion of this paper.

Notes

Often the motivation to declare some income is to obtain the length of source credits for social security and perhaps to avoid harsh treatment by the tax administration if caught.

By way of comparison South Cyprus exempts the first US$12 000 of income from personal taxation. Belize another microstate and former British Colony with a per capita income of about $3500 exempts the first US$10 000 from personal income tax. Both of these countries collect about 25% of GDP through all their tax combined.

There are certain limitations due to our data. The study will base its findings on the households that have taken part in the HCES and have to base its estimates on what they have chosen to tell the survey. In both respects, it is reasonable to suspect that we may not be getting the complete picture of the extent of the undeclared income. People who are heavily involved in tax evasion may simply fail to respond to the HCES, perhaps because they do not believe the confidentiality assurances that are given, and fear that the information they give to the survey may somehow find its way back to tax authorities. Other people, whilst responding to the survey, might nonetheless be aware of the implications of appearing to live beyond their means, and might adjust the amount of expenditure they report accordingly.

The methodology used here is similar to that employed by British, Swedish and Canadian data (Pissarides and Weber, 1989; Apel, 1994 and Mirus and Smith, 1997) on British, Swedish and Canadian data respectively.

For the HCES survey data to be fully applicable for addressing questions concerning tax evasion the respondents need to report to the Survey the same level of income they have reported on their tax return, and the same expenditure patterns exist for food (ceteris paribus), whether one is self-employed or a private sector employee. Given the fear of self incrimination, it is likely that most people will report incomes similar to the level they reported to the tax authorities.

This well established empirical observation was initially developed by, Milton Friedman (1957).