Database Specific
Database Specific Abstract AbstractGross replacement rates (GRRs) express gross unemployment benefit levels as a percentage of the previous gross earnings. The index available here is defined as the average of the gross unemployment benefit replacement rates for two earnings levels (67 and 100% of the Average Production Worker wages), three family situations (single person, one-earner couple and two-earner couple), and three durations of unemployment (1 year, 2 to 3 years, and 4 to 5 years). For further details, see OECD (1994), The OECD Jobs Study (chapter 8) and Martin J. (1996), “Measures of Replacement Rates for the Purpose of International Comparisons: A Note”, OECD Economic Studies, No. 26. Pre-2003 data have been revised. Please note: Net Replacement Rates (NRRs) provide a more complete measure of benefit generosity and income maintenance, especially when compared over longer periods of unemployment (please see below for the relationship between GRRs and NRRs). Updating and maintainenance of the gross replacement rate (GRR) index, originally constructed as part of the OECD Jobs Study (1994), has been reliant on access to Average Production Worker (APW) wages. These data have not been collected by the OECD since 2005 and the Historical GRR series has been dismissed. Instead, the OECD provides annually since 2001 the NRR indicator based on a slightly different definition of the reference wage, the Average Wage (AW), for most OECD and EU countries.
Data Characteristics Data Characteristics Other data characteristics Other data characteristicsRelationship of the Historical GRR index to NRRs. The Historical GRR index differ from NRR indicator in the following ways:
1. GRRs do not take tax and social security contributions on earnings and on benefits into account. If tax systems are progressive, then taxes paid while in work will be a greater percentage of income than during unemployment. This decrease of in-work income in relation to out-of-work income is captured by NRRs, which will therefore tend to be higher than GRRs. Furthermore, changes in the tax treatment of benefits will mean that the time series of GRRs will appear different to that of NRRs.
2. No children are included in the household types considered in the GRR index. Consequently, they do not capture the effects of changes in family-related benefits. The absence of such benefits will generally lead to lower measures of benefit generosity.
3. No housing benefits are included. These benefits can provide a significant part of income for households without earnings. GRRs will again be lower than NRRs which include housing benefits.
4. Social assistance is not included in most countries, unless it consists of a general income guarantee at nationally determined level. In the part of the GRR indices reflecting incomes in years 4 and 5 (and even years 2 and 3), benefit income is therefore assumed to be zero in many countries. Were it to be assumed that social assistance was paid, average GRRs would be higher.
5. In-work benefits are not included in the GRR indices. In countries where they exist, the exclusion will, at certain earnings levels, tend to reduce NRRs in relation to GRRs since in-work incomes are higher when in-work benefits are taken into account.
6. For more information on the OECD tax-benefit database, please visit the project webpage or contact the OECD tax-benefit team.
Reference period Reference periodOdd-numbered years from 1961 to 2005.
Gross replacement rates (GRRs) express gross unemployment benefit levels as a percentage of the previous gross earnings. The index available here is defined as the average of the gross unemployment benefit replacement rates for two earnings levels (67 and 100% of the Average Production Worker wages), three family situations (single person, one-earner couple and two-earner couple), and three durations of unemployment (1 year, 2 to 3 years, and 4 to 5 years). For further details, see OECD (1994), The OECD Jobs Study (chapter 8) and Martin J. (1996), “Measures of Replacement Rates for the Purpose of International Comparisons: A Note”, OECD Economic Studies, No. 26. Pre-2003 data have been revised. Please note: Net Replacement Rates (NRRs) provide a more complete measure of benefit generosity and income maintenance, especially when compared over longer periods of unemployment (please see below for the relationship between GRRs and NRRs). Updating and maintainenance of the gross replacement rate (GRR) index, originally constructed as part of the OECD Jobs Study (1994), has been reliant on access to Average Production Worker (APW) wages. These data have not been collected by the OECD since 2005 and the Historical GRR series has been dismissed. Instead, the OECD provides annually since 2001 the NRR indicator based on a slightly different definition of the reference wage, the Average Wage (AW), for most OECD and EU countries.
Odd-numbered years from 1961 to 2005.
Relationship of the Historical GRR index to NRRs. The Historical GRR index differ from NRR indicator in the following ways:
1. GRRs do not take tax and social security contributions on earnings and on benefits into account. If tax systems are progressive, then taxes paid while in work will be a greater percentage of income than during unemployment. This decrease of in-work income in relation to out-of-work income is captured by NRRs, which will therefore tend to be higher than GRRs. Furthermore, changes in the tax treatment of benefits will mean that the time series of GRRs will appear different to that of NRRs.
2. No children are included in the household types considered in the GRR index. Consequently, they do not capture the effects of changes in family-related benefits. The absence of such benefits will generally lead to lower measures of benefit generosity.
3. No housing benefits are included. These benefits can provide a significant part of income for households without earnings. GRRs will again be lower than NRRs which include housing benefits.
4. Social assistance is not included in most countries, unless it consists of a general income guarantee at nationally determined level. In the part of the GRR indices reflecting incomes in years 4 and 5 (and even years 2 and 3), benefit income is therefore assumed to be zero in many countries. Were it to be assumed that social assistance was paid, average GRRs would be higher.
5. In-work benefits are not included in the GRR indices. In countries where they exist, the exclusion will, at certain earnings levels, tend to reduce NRRs in relation to GRRs since in-work incomes are higher when in-work benefits are taken into account.
6. For more information on the OECD tax-benefit database, please visit the project webpage or contact the OECD tax-benefit team.