Annex D. Methodology and Additional Notes on Compensation of Government Employees

Occupations

In order to build a comparable database, this survey is based on a classification of occupations typical in most ministries of OECD countries, and of service delivery agents.

This survey collects data on compensation levels of public servants in central/federal/national government in the OECD countries, and accession countries. The survey focuses on the central/federal government level and excludes states, regional and local levels and social security institutions. The survey excludes all public and quasi-public corporations at all government levels1 . The survey does not cover the subordinated offices/organisations of central government ministries, often referred to as “agencies”, “executive agencies”, except for D1 and D2 positions (Box D.2) and service delivery agents (Box D.3).

The questionnaire asks for information concerning a number of occupations within central/federal/national government grouped under four basic headings: top managers, middle managers, professionals, and secretaries. The selected occupations are considered relatively representative and comparable across countries. Information for those positions is collected from three core ministries (Interior, Finance, and Justice) and three sectoral ministries (Education, Health, and Environment) (Box D.1).

Moreover, countries are requested to provide information concerning some frontline service delivery agents such as detectives/inspectors, police officers, immigration officers, customs inspectors, and tax inspectors.

The classification and the definition of the occupations are an adaptation of the International Standard Classification of Occupations (ISCO-08) developed by the International Labour Organisation (ILO). Few countries follow the ISCO model to classify their occupations in government. In the questionnaire countries were asked to identify and describe the jobs that may correspond to the identified occupations, including a submission of a job description and examples of key responsibilities in the most senior and less senior positions for each occupation. The survey focuses on employees under the general employment framework or statute and not on consultants.

Box D.2 contains the classification and definitions of the occupations covered in this survey and which are considered to be relatively typical in every government. There is a large focus on managers in general as the criteria for considering an official to be a manager is to supervise and lead the work of at least three people. Because it is extremely difficult to provide for more detailed descriptions of responsibilities that differentiate across the different layers of management, the option has been chosen to focus on hierarchical differentiation rather than a more detailed description of functions. Since there is no common definition of managerial positions and the number of managerial levels varies across countries and ministries, for the purpose of this survey, D1 will denote the highest managerial level below the minister/secretary of state (who are designated by the President/Prime Minister) and appointed by the minister (sometimes designated by the President/Prime Minister). This survey will cover until D4 managerial level positions but D5 and D6 levels will be considered only if they are reported by participating countries. In the particular case of managerial positions countries will be asked whether data can be published considering the confidentiality of information.

The category of “professionals” has been divided between junior and senior positions. This group corresponds to the least identifiable group and involves staff with a large variety of experience.

This survey also includes a section on compensation of service delivery agents including police detectives and inspectors, police officers, immigration officers, customs inspectors, and tax inspectors. These occupations are defined using job descriptions taken and adapted from ISCO-08 (Box D.3). The intention is to have an understanding of the compensations of service delivery agents for some functions that are relatively commonly provided by national governments such as tax administration, immigration services, and policing which are not covered in other surveys (such as those related to health and education services which typically involve sub-national governments).

These functions are organised in central government, and can be located in either ministries or agencies. It should be noted that in some countries functions like immigration officers do not exist as these activities are carried out by the police. In other countries, some of the functions mentioned above are carried out by states and/or local governments.

Compensation

The survey focuses on total compensation, which has two main components: 1) wages and salaries, and 2) employer’s social contributions. Data on remuneration levels were asked for full time jobs.

1. Gross wages and salaries which include the values of any social contributions, income taxes, etc., payable by the employee even if they are actually withheld by the employer for administrative convenience or other reasons and paid directly to social insurance schemes, tax authorities etc., on behalf of the employee. Employer’s social contributions are not included in gross wages and salaries. In kind compensation is excluded from the survey (unless a government cannot exclude them, in which case, a note needs to explain the situation). Gross wages and salaries include:

• Basic wages and salaries (as laid down in the salary scales) refer to the regular annual payments to employees for their time worked and services delivered to government. Although salaries and wages are paid at regular weekly, monthly or other interval, for the purposes of this survey the annual salary is requested. Overtime payments are excluded from the data.

• Additional payments – because of the difficulties in getting exhaustive data and ensuring comparability across countries, additional payments have been limited to its most significant categories including:

  • Compensations for time not worked make reference to annual leave and bank holidays only.

  • Bonuses and gratuities regularly paid refer to year-end and seasonal bonuses; profit-sharing bonuses; and additional payments in respect of vacation, supplementary to normal vacation pay and other bonuses and gratuities.

  • Bonuses and gratuities not paid in a regular fashion (performance-related pay) refer to ad hoc bonuses or other exceptional payments linked to the overall performance of the employee to which he/she may be entitled.

2. Employers’ social contributions are social contributions payable by employers to social security funds or other employment-related social insurance schemes to secure social benefits (health insurance, pensions) for their employees. Employers’ social contributions can be divided into:

• Employer’s contribution to statutory social security schemes or to private funded social insurance schemes for covering old age, pension, sickness and health. Employer’s social contributions represent social contributions payable by employers to social security funds or other employment-related social insurance schemes to secure social benefits (health insurance, pensions) for their employees. In some countries, these social contributions pay for public schemes, while in others for private schemes. Employer’s social contributions sometimes also include specific funds created for example in social agreements. Data collected on employer’s social contributions have been limited to health and pension plans, which represent the majority of employer’s social contributions.

• Unfunded2 employees social benefits paid by employers limited to health and pension benefits. They represent the counterpart to social benefits paid directly by general government institutions without participating in, or establishing a fund, reserve or other special scheme for this purpose. Since these contributions do not involve actual cash flows, they have to be imputed. These unfunded pension or health schemes exist in many countries.

Not all countries have been able to include the social contribution element in their survey responses (mainly for unfunded pension schemes). As a consequence, it has been necessary to estimate this component using other data sources for those countries. In the National Accounts, imputations for unfunded pension’s schemes are made conceptually consistent across countries. Therefore, by using the National Accounts data it was possible to estimate the overall rate of employer’s social contributions that was reported in the different existing databases regarding government compensation of employees. The rate chosen to calculate compensation costs in the data for this publication has been chosen after investigation and discussion with the countries. The source of National Accounts for this share was selected in the following countries: Germany, Greece, Japan, Norway, Portugal, Spain and Lithuania. Moreover, for Belgium this share was estimated using a combination of information from the compensation survey and National Accounts data.

We should note that, contrary to the compensation survey where employers’ contributions are restricted to health and pensions, data under the National Accounts framework consider all employer’s social contributions. By consequence, the resulting share, to a certain extent, was overestimated when this source was taken into account. Moreover, National Accounts data provide ratios of employer’s social contributions for all government employees. Using this ratio hence doesn’t accommodate any differences that may exist for instance in ratios of social contributions across occupations. For the countries which have provided data for employer’s social contributions in the survey, the exact data for social contributions (that may vary across occupation) have been used.

The level of social contributions is only a proxy. The quantity and quality of benefits that employees receive through the employers’ and employees’ social contributions depend on many variables such as the quality and efficiency of the management of the funds and services in each country.

Use of comparators

Calculations have been made converting compensation data in USD using the PPP methodology. This compensates for differences in exchange rates and in relative price levels. The PPP does not take into account the relatively different costs of living in capital cities within and across countries. In many countries, the majority of central government employees are employed in capital cities. Wages can tend to make up for the relative difference in the costs of living in capital cities. PPPs are calculated for various levels of aggregation up to and including GDP. The PPP for GDP covers both final consumption expenditure and gross capital formation. The PPP for actual individual consumption covers all households consumption expenditures - which represent the private consumption component - and that part of government final expenditure supplied to individual households (e.g. health, education etc.).

The PPP used for the conversion of compensation in national currency of government employees by different positions was the PPP for private consumption. Prior to the 2013 edition of Government at a Glance (OECD, 2013), compensations were converted using PPP for GDP. As consequence, average compensations by positions in USD published in this edition are not directly comparable with the figures published prior to the 2013 edition.

The OECD also compared countries with data normalised with GDP per capita data available through the OECD National Accounts Statistics database. This normalisation is a way to remove for differences in levels of average wealth in the country.

The ratio of compensation of employees relative to GDP per capita has not been corrected for working time. This approach was followed in order to maintain consistency between the measures compared.

Computations for comparing annual compensation including adjustment for working time

Average comparative annual compensation is calculated as:

where:

= Average annual compensation of employees in country c within occupational group o in PPP corrected for working time

= Average annual compensation in domestic currency in country c within occupational group o in national currency

= Purchasing power parity of country c

= Ratio of average working time in country c. This corresponds to average annual working hours in country c (from survey data) divided to 2088. The number 2088 equals the theoretical working hours in year with 40 hours of work per week, no holidays or leave of any kind. This also results in an average of 261 working days per year with each working day including 8 hours of work.

The differences between the time people actually work and the annual average compensation (annual average gross salary plus employer’s social contributions) is calculated so as to obtain an adjusted annual average compensation. Indeed, to put the compensation of employees reported on a comparable basis across countries, the differences in the working time (number of hours worked per week in the civil service, the legal or average holiday entitlement as specified in the work contract, and the number of public holidays per year that apply to the civil service) are used for the calculation of the adjusted annual average compensation. For all managers (namely D1, D2, D3 and D4 positions), since weekly working times apply very unevenly to this category of employees data was adjusted only for holidays.

The working time corrections are reported in Table D.1.