SMEs responsible for public–private sector pay gap

A Banca d’Italia working paper based on data from its biennial survey of household incomes and wealth highlights a significant gap in pay between public and private sector workers in Italy. By controlling sociodemographic variables, this gap is shown to be larger for women (especially blue-collar workers). The results of a quantile regression technique suggest that the lower wages paid by small companies to their employees plays a major role in explaining the wage gap.

Background

The wage gap between the public and private sector is a recurrent issue in debates about the Italian economy and labour market. This gap is often said to be caused by the more ‘generous’ pay increases agreed in collective bargaining in the public sector because of its lower exposure to competitive pressures. This seems to be the rationale behind the 2010–2013 freeze on public sector pay with the aim of containing public expenditure even though a dossier presented to the Italian parliament (in Italian, 80Kb PDF) by the National Institute of Statistics (Istat) does not support such a view.

A second issue of interest is the relatively larger number of women employed in the public sector than in the private sector, primarily because of the public sector’s high demand for occupations traditionally dominated by women such as clerical work, teaching, healthcare and social services.

Findings of national bank survey

A working paper, The public-private pay gap: a robust quantile approach (in Italian) published by the national bank Banca d’Italia in July 2011, investigates the issue by making use of data from its Survey on Households’ Income and Wealth (SHIW) survey that has been carried out every two years since 1965. This survey examines the wealth and income of Italian households by collecting a wide set of information, including the occupational status of their members, work seniority, living conditions, and reported earnings and wages. The sample used in the most recent surveys is made up of about 8,000 households (24,000 individuals) distributed over about 300 Italian municipalities.

According to the working paper, employees in the public sector have higher pay than those in the private sector by 34% for women and 25% for men. Such a gap tends to increase with pay levels, as shown by analysis by quartiles; it increases from 30% at the 25th quantile to 36% at the 75th quantile for women, and from 21% at the 25th quantile to 26% at the 75th quantile for men.

However, the wage gap declines with regional employment rates from 30% in the northern regions of Italy, where the employment rate is above 60%, to 55% in the southern regions where the employment rate falls below 50%.

Several factors contribute to this gap. The vast majority of men and women in the public sector are white-collar workers (especially in the less developed southern regions), while there are more blue-collar workers than white-collar workers in the private sector. Similarly, the share of managerial positions is higher in the public sector than in the private one for both men (7% as against 2%) and women (3% in the public sector and negligible in the private sector). This reflects further differences between the sectors such as educational qualification and the social background of employees.

Effect of sociodemographic factors

Sectoral wage gaps are considerably lower after controlling for personal and work characteristics. This is achieved using a quantile regression technique which allows comparison of both personal and work characteristics and their distribution. This technique produces a mean public–private sector wage gap of 14.1% for women and 4.2% for men (see table).

The sectoral gap among women displays an irregular profile across the deciles, while it declines considerably among men as income increases.

Blue collar workers have lower wage gaps than white-collar workers for both sexes but the pattern is reversed. Women display a wider sectoral gap among white-collar workers than among blue-collar workers (16.1% and 9.2% respectively), while for men the reverse is true (7.7% among blue-collar workers and 4.4% among white-collar workers).

Notes: * significance at 90%; ** significance at 95%; ***significance at 99%

Coefficient estimates behave when the regression specification is modified by adding or removing regressors. If the coefficients are plausible and robust, this is commonly interpreted as evidence of structural validity.

Source: Depalo and Giordano (2011)

Effect of company size

The authors of the working paper carried out several robustness tests in order to check whether regression coefficients were plausible, although this exercise was limited to the 2004 wave of the survey such as differentiating between wages in large companies and in small and medium enterprises (SMEs) with fewer than 100 employees. Wages are lower in SMEs with fewer than 100 employees due to:

• more instances of violation of contractual rules such as gender discrimination and use of undeclared employment;
• less likelihood of company-level bargaining centred on performance-related pay than in both larger companies and the public sector.

In this exercise, the sectoral pay gap becomes bigger among women irrespective of company size, while for men the pay gap is larger only for those working for small companies. Men working in large private companies are no worse off than those working in the public sector.

Thus the public–private sector wage gap in Italy is primarily due to the weight of small companies in the private sector, with their lower pay and wider gender segregation.

Commentary

This paper is an important contribution to the Italian debate on wages, especially the arguments about the private–public sector wage gap for which no official statistics are available, because it shows that such a gap is found mainly among small and micro companies which tend to have lower productivity than large ones. However, the SHIW data are less reliable than those from the Istat source; the former relies on reported wages while the latter is based on data on social security contributions (in Italian) reported by the Italian National Social Security Institute (INPS). This implies some underreporting, which can vary between sectors and is supposedly more frequent in small companies.

Reference

Depalo, D. and Giordano, R. (2001), Il differenziale salariale tra settori pubblico e privato: una analisi robusta sui quantili [The public–private pay gap: a robust quantile approach], Discussion Paper No. 824, Banca d’Italia, Rome.

Mario Giaccone, Ires