Continuous Erosion: The Israeli Public-Private wage gap, and an International Comparison
The State of Israeli Workers
Ido Lan and Yoni Ben-Bassat
Translation: Adi Marcus September 2023
Key Findings:
1. As of 2021, Israeli public sector wages are the lowest amongst studied countries, compared to local private sector wages.
2. This has not always been the case. A continuous erosion of the public sector pay has been taking place in the last three decades – during these years the public-private wage gap increased by about 30 percentage points.
3. Erosion of public sector wages in Israel is very high– the decline in the public sector pay in Israel is very high compared to all the developed countries studied.
4. Large public-private wage gaps may harm the quality of Israeli public services.
1. Public Sector Pay in Israel – the lowest compared to Private Sector Pay
Figure 1 depicts the public-private wage gap (ratio) per employee in Israel (light blue) and in 23 other OECD countries[1] in 2021 (column), compared to 2001 (diamond) and 1995 (square). The data reveals that in 2021, Israeli public sector employees received the lowest relative pay compared to other countries, with a wage gap between the public and private sectors of negative 29%.
The wider the (negative) gap between public sector pay and private sector pay, the more challenging it becomes to attract and retain a high-quality workforce into the public sector. The significant gap found in Israel sheds light on Israel’s problematic wage-policy for public servants.
It is noteworthy that the margins of the gap between Israel and other examined countries are partly affected by the definition of the public sector. As detailed in the appendix, different definitions can lead to different gap measurements.
2. Three decades of continuous erosion of public sector pay
Figure 2 shows the evolution of the public-private wage gap (ratio) per employee from 1995 to 2021.
In 1995, the wage gap was low, positive, and very close to the OECD average. However, since the mid-1990s the gap steadily increased, turning negative, with public sector workers’ pay eroding substantially compared with their private-sector counterparts.
Although the 2008 financial crisis resulted in a slight alleviation of the wage gap, the widening trend resumed in 2011 and persisted through to 2021, while the OECD average remained relatively stable. Specifically, we can see that from 2015 through to 2021, Israel had the largest negative wage gap among studied countries (Israel’s line falls below the grey area).
It is important to note that prior to the timeframe of the data presented in this report, collective agreements signed in the public sector between 1993 and 1995 led to significant pay increases, which were atypical to the long-term trend in Israel (Mazar, 2022). Accordingly, in this paper, we also use 2001[2] as a reference, since the wage gap in 2001 is more typical and indicative of the long-term trend in Israel (see Figure 3). In total, over 26 years, the public-private wage gap increased by 30 percentage points (from 1.9% to -29.4%) and since 2001, it has increased by 20 percentage points (from -12.3%).
In addition, unlike other developed countries, Israel has a large standing army whose conscripts are paid much less than the statutory minimum wage for employees. Therefore, we added to Figure 2 an estimate[3] of the wage gap as of 2005, excluding the IDF military forces (both conscripts and career officers). As a result, Israel’s relative position somewhat improves. However, for 2021 it still has the highest negative wage gap among the countries studied.
3. The change in the public and private sectors' wage gap is highly anomalous
Figure 3 presents the changes in the per-employee public-private wage gap between 2001 and 2021 (column), and between 1995 and 2021 (diamond).
As can be seen, the per-employee wage gap between the sectors in Israel declined by 17 percentage points between 2001 and 2021 (31 percentage points compared to 1995). Of the six countries in which there was a large decline in their wage gap between 2001 and 2021, only Israel and Estonia had a negative public-private wage gap in 2001, although Estonia’s gap was significantly smaller than Israel’s (see figure 1). Moreover, the three countries which experienced a stronger decline in relative public sector wages than that of Israel are the same countries, in which wage ratio was the highest among the studied countries in 1995 (indicating a high relative public sector wage). In other words, the increase in the wage gap in Israel is anomalous, particularly when taking into account that the wage gap in Israel was initially low.
These data indicate that the erosion in the public sector wage compared to that of the private sector in Israel is uncommon and irregular compared with the other developed countries and could be an obstacle for ensuring the public sector is of high quality. If the public sector pay continues to erode compared to that of the private sector, it will be difficult to recruit high quality personnel for public services.
Appendix - Sources of Data and Definitions
About the Data
Three types of definitions are used in research literature to define the public sector: (1) A sectoral definition - based on public services in the industry classification (ISIC/NACE codes); (2) An institutional definition - based on the national accounts system as part of a framework which divides output to institutional units, and; (3) A subjective definition - based on how employees define their workplace in surveys.
The OECD data that we use[4] are based on the sectoral economic definition, which follows some previous studies on public sector pay.[5] The main disadvantage of this approach is that the sectoral definition for defining the public sector also includes employees who, in practice, belong to the private sector and vice versa. However, as aforesaid, this is the generally accepted definition that allows the international comparison we conducted in this study.
The sectors of the economy which are included in the sectoral definition under the public sector are:
- Public administration and security (ISIC section O): This is a distinctly public sector that includes employees of the government ministries, local authorities, the National Insurance Institute, and security forces.
- Education (ISIC section P): This is a public sector that mainly includes staff of kindergartens, schools, public universities, and colleges, as well as private educational institutions.
- Health and welfare (ISIC section Q)[6]: This section includes the public health system (in Israel, it includes health insurance funds and hospitals) and the welfare system, but it also includes private healthcare and welfare institutions (such as homes for the elderly, dentistry, private mental health services).
The private sector sections included in this definition are B-N, which comprise the accepted definition of the business sector in various studies. It should be noted that Section A (agriculture, forestry, and fishing) and Sections R-T (arts, entertainment and recreation; other service activities; and households as employers) are not included under the public or private sectors under this definition, and therefore are omitted from the calculations.
As previously mentioned, there is also an institutional definition of the public sector. This definition is internationally recognized under the System of National Accounts, and it is used to distinguish public sector entities from other institutional units (UN et al., 2009). According to the international definition, any organization that produces outputs that are not traded on the market and is controlled by the government in the broadest sense (the government or various entities under its control, directly or indirectly) should be included under the definition of the public sector. In addition, government corporations are companies that are controlled by the government in the broadest sense but provide outputs that are traded on the market. Therefore, the public sector can be defined according to the institutional definition as the government sector or as the government and government corporation sector.
Why the sectoral definition is preferred over the institutional one
Of the two foregoing definitions, the more common definition in international comparisons between public and private sector pay is the sectoral definition or the subjective definition (determined in survey data) and not the institutional definition. This is mainly due to the limited availability of data (Geordano et al., 2011). Furthermore, in recent years, the discourse regarding comparability between countries as per the institutional definition has increased (see: Perez et al., 2016). We will note below several problems with the institutional definition that limit its reliability in international comparison:
Technical aspects:
1. Long-term international comparisons are challenging due to the lack of available data. For example, data from Israel regarding employees and work compensation as per the institutional definition only exists since 2012,[7] and does not allow us to review the trend over 3 decades as we have done in this study.
2. The institutional division in the OECD data includes only the government sector and does not allow the separation of government corporations from the private sector. For example, the government hospitals in Israel employ tens of thousands of employees, but they are defined as government corporations[8] and therefore are counted under the private sector in the OECD data.
Fundamental aspects:
1. The institutional approach, as aforesaid, includes, under the definition of the 'public sector', only the wages paid by an entity (that is not a corporation) that is controlled by the state. But privatization and outsourcing in recent decades have led to the inclusion of a reduced number of employees under this definition. Many low-wage jobs are outsourced, and those employees are employed through external contractors (ibid).
2. Furthermore, even after privatization, the wages of the employees are sometimes set indirectly and sometimes directly by the state. For example, the wages and number of positions for employees under the private welfare sector in Israel are set in the terms of the tender (Houminer-Rosenblum, 2023). In general, in developed countries, the state provides public services. Therefore, their prices are not set by market equilibrium, rather, they are set by the state. In this context, it is customary to perceive the state as a market maker in social services, which dictates the working conditions, also when these are outsourced to private companies (Flecker et al., 2009).
3. Thus, for example, although in both Denmark and Belgium, there is state health insurance that provides extensive health services to citizens free of charge, in the former, most health sector employees are defined as public servants, whilst in the latter, the state insurance services are provided by physicians employed by private hospitals (Perez et al., 2016). It is clear that at the end of the day, both in Belgium and in Denmark, the medical staff wages are set by the state, directly in an employment agreement or indirectly by fixing prices for various operations in the healthcare market.
To fully comprehend the wage level in public sector services, it is crucial to consider the industry definition, especially in cases where the services have been privatized. Relying solely on the institutional definition may result in an incomplete understanding of the wages paid within the industry.
On international comparison
Characteristics that vary between countries (but remain constant within each country) could affect the wage disparity between the public and private sectors in international comparison. For instance, countries that have a large standing army or high birth rate may exhibit a different wage gap than those without these characteristics. However, comparison over a long period cancels out the effects of each country’s unique fixed characteristics (in Israel, for example, a large standing army existed both in 1995 and 2021). In other words, the growth in the wage gap between sectors in Israel cannot be explained by specific characteristics since these characteristics have existed throughout the periods examined.
[1] The countries were selected according to data availability, presenting all the OECD countries for which continuous data from 1995 through 2021 was available. A review of the data for a larger group of developed countries for which data was available through 2019 provides a similar picture.
[2] In 2001, private sector pay peaked (see: Mazar, 2022), and the employees’ share of the business sector GDP peaked (see Ben Bassat, 2022). Consequently, this is the most exacting point of comparison compared with 1995. [3] The wage data without the IDF is taken from the National Insurance Institute - Wages and Employment Statistics, which provides wage data per employee job , while all the other wage series use total compensation per worker according to the National Accounts. For a review of the differences between these data sets, see: Itzhaky and Peleg, 2011. Furthermore, there is a break in the National Insurance series in 2012 described in the diagram.
[4] All the data is this study are based on the OECD National Accounts data (employee compensation and number of employees) and relates to salaried employees only (excluding self-employed, whose wages are imputed in national accounts). [5] See: Mazar, 2022. Campos et al. 2017; Muller & Schlutten, 2015; Christofides & Michael, 2013, 2020; Naddeo, 2016; Sławińska, 2021 Giordano et al., 2011; [6] For a comprehensive breakdown of which sub-sections are included in each branch section see: CBS, 2015. [7] Although OECD data exists since 2000, there is a significant break in 2012. At the same time, discussions held with the CBS indicate that, indeed, the existing data from before 2012 are less reliable compared to the existing data from then onwards. [8] See: Wages and Employment Statistics, National Insurance Institute, Table 1, Employment according to Sector and Sub-Sector.