Luxembourg Income Study

views updated

Luxembourg Income Study


The Luxembourg Income Study (LIS), a research center and microdata archive, was founded in 1983 by Timothy Smeeding, Lee Rainwater, Gaston Schaber, and a team of multidisciplinary researchers in Europe. With support from the government of Luxembourg, LIS and its staff became an independent nonprofit institution in 2002. LIS is organized as a consortium of countries with financing from the national science foundations and other funders in the participating countries, as well as from the Luxembourgian government. LIS operations are governed by a board whose members represent the countries that provide data and financing. Janet Gornick, a political economist and sociologist based in the United States, became overall director of LIS in September 2006. Markus Jäntti, an economist in Finland, took up the post of LIS Research Director in 2005.

LIS has four goals:

  1. To harmonize cross-national data, thus relieving researchers of this task, relying on an expert staff that carries out the harmonization work and provides support services for users.
  2. To provide a method allowing researchers to access these data under privacy restrictions required by the countries providing the data.
  3. To create a system that allows research requests to be received and results returned quickly to users at remote locations.
  4. To promote comparative research on the economic and social well-being of populations across countries.

In 2006, LIS included data from thirty countries, mostly in Europe and North America, but also including Australia, Israel, and Taiwan. The database contained over 150 datasets, organized into five time periods (known as waves ) spanning the years 1968 to 2002. Wave six is due to come online in 2007. The data can be accessed in multiple ways. Researchers can write programs (in SPSS, SAS, or STATA) and send them via electronic mail directly to the LIS server; results are returned to the researcher, with average processing time under two minutes. There is also a web-based tabulator that allows users to construct tables using keywords. The LIS website provides a set of country-level indicators (known as Key Figures ), including measures of inequality and poverty for each LIS dataset.

Extensive documentation for each dataset details technical aspects of the original survey, a record of the harmonization process, and institutional information on tax and transfer programs corresponding to the microdata variables. The LIS website also houses a comparative welfare states database and a family policy database; both contain an array of country-level policy indicators. These policy databases are widely used by LIS microdata researchers, who often seek to link policy variables to microlevel outcomes.

Reports based on LIS data have appeared in books, journal articles, and dissertations, and are often featured in the popular media. Each completed study is made available in the LIS Working Paper series, which numbered more than 450 papers by Fall 2006. The LIS website offers a Working Papers search engine, a complete set of abstracts, and most of the papers in full text.

Beginning in 2005, LIS expanded by adding a wealth data project. The Luxembourg Wealth Study (LWS) established a network of producers of microdata on household wealth and has, like LIS, harmonized the country-specific data into a common template, including comparable measures of net worth and its components. The LWS project will help to set guidelines for wealth data producers, as the LIS project has with income data.

LIS conducts annual training workshops that introduce researchers to the database and to cross-national research on wages, income, employment, and social policy. Between 1988 and 2006, more than five hundred scholars attended the workshops. LIS publishes a newsletter twice yearly, which is mailed to over 1,400 scholars in thirty-five countries.

The LIS microdatasets include income, employment, and demographic variables at the person and household level. Since LISs inception, these microdatasets have been used by more than a thousand researchers in many countries to analyze economic and social policies and their effects on outcomes including poverty, income inequality, employment status, wage patterns, gender inequality, family formation, child well-being, health status, immigration, political behavior, and public opinion.

One of the most fruitful uses of LIS is for the study of income distributions across the richest countries of the world. Figure 1, derived from LISs Key Figures, summarizes income distributions in the LIS countries, using four measures of inequality (ratios of tenth and ninetieth centiles to the median; the 9010 decile ratio, and the Gini coefficient). This figure shows clearly that income distributions vary dramatically across countries, with variation seen at both the bottom and the top of the distributions. The figure also reveals loose clusters of countries, with lower levels of inequality in the Nordic countries, moderate levels in most of the continental European countries, and higher levels in southern Europe and in the English-speaking countries of Australia, Canada, Ireland, the United Kingdom, and, most especially, the United States. Notably, the former communist countries of eastern Europe report remarkably varied levels of income inequality.

LIS-based research has catalyzed changes in national policiesfor example, British policy toward children, based on the work of Jonathan Bradshaw (Bradshaw and Chen 1997)and has informed the United Nations, the Organization for Economic Cooperation and Development (OECD), and other major bodies about poverty, inequality, and employment outcomes across countries. Results based on LIS have been published in and lauded by Science (Butz and Torrey 2006), The Lancet (Lynch et al. 2001), and major academic journals in the fields of economics, political science, sociology, comparative public policy, and social measurement. A twenty-year anniversary volume, published by The Socio-Economic Review in 2004, further summarizes and explains the accomplishments of LIS (Smeeding 2004).

In 2006 LIS completed a comprehensive internal review of its data template and harmonization rules in order to improve the quality of the LIS data and to identify ways to increase cross-country comparability in response to changes in the previous two decades in the participating countries social policies and survey content. This review also led to a restructuring of the pension

and family benefits data, an expansion of the person-level data, and a substantial increase in the number of labor market variables included in LIS. These revisions enabled the many researchers who use LIS primarily for employment research to go further in their comparative analyses.

Thereafter LIS anticipated adding new datasets for all of the participating LIS countries; the newest wave of data (LISs sixth wave) was to include datasets from approximately 2004. In addition, LIS anticipated adding two new countries in 2007: South Korea and Japan. LIS also continued to work to bring in datasets from Portugal, New Zealand, and Turkey.

LISs income surveys have mostly come from high-income countries, as classified by the World Bank. Of the thirty countries participating as of 2006, twenty-one are high-income and nine are upper-middle-income (Czech Republic, Estonia, Hungary, Mexico, Poland, Romania, Russia, Slovak Republic, and Taiwan). One of LISs main priorities is to substantially increase the inclusion of middle-income countries. LIS intends to add microdata, at multiple points in time, from ten middle-income countries, including, for example, Brazil, Bulgaria, Chile, China, Indonesia, and South Africa.

SEE ALSO Discrimination; Gini Coefficient; Inequality, Income; Poverty; Social Exclusion


Bradshaw, Jonathan R., and Jun-Rong Chen. 1997. Poverty in the UK: A Comparison with Nineteen Other Countries. Benefits 18: 1317.

Butz, William P., and Barbara Boyle Torrey. 2006. Some Frontiers in Social Science. Science 312 (5782): 18981900.

Gottschalk, Peter, and Timothy M. Smeeding. 2000. Empirical Evidence on Income Inequality in Industrialized Countries. In Handbook of Income Distribution, Vol. 1, eds. Anthony B. Atkinson and François Bourguignon, 261307. New York: Elsevier.

Luxembourg Income Study.

Lynch, John, George Davey Smith, Marianne Hillemeier, et al.2001. Income Inequality, the Psychosocial Environment, and Health: Comparisons of Wealthy Nations. The Lancet 358 (9277): 194200.

Smeeding, Timothy M., guest ed. 2004. Socio-Economic Review 2 (2): 149339.

Janet C. Gornick

Timothy M. Smeeding