Research Topics

This paper presents a description of the quality of match of the statistical matches used in the Levy Institute Measure of Time and Consumption Poverty (LIMTCP) estimates prepared for Ethiopia and South Africa. For Ethiopia, the statistical match combines the Ethiopian Socio-economic Survey—Wave 3—2015/2016 (ESS) with the Ethiopian Time Use Survey (ETUS) 2013. For South Africa it combines the October Household Survey (OHS) 1998 with the time use data obtained from the SA-Time Use Survey (SATUS) 2000, and the South African Living Conditions Survey (SALCS) 2014/2015 with the SATUS 2010. In all cases, the alignment of the two datasets is examined, after which various aspects of the match quality are described. Despite the differences in the survey years, the quality of match for South Africa is high and the synthetic dataset appropriate for the time poverty analysis. For Ethiopia, due to data quality differences, we restrict the analysis to married couple households with an employed spouse and young children. Conditioning on the restriction and sample reweighting, the Ethiopian synthetic dataset seems appropriate for the time poverty analysis.

This paper assesses the quality of the statistical matching used in the LIMTIP estimates for Italy for 2008 and 2014. The match combines the 2008–9 and 2013–14 Italian Time Use Survey (IT-TUS) with the Italian data collected for the European Survey on Income and Living Conditions (IT-SILC) in 2009 and 2015. After the matching, the analysis of the joint distributions of the variables shows that the quality is good.
 
The preliminary results of the LIMTIP estimates in Italy display widespread time poverty, which translates into significant hidden poverty. The LIMTIP also reveals that the increase in the poverty rate between 2008 and 2014 was even higher that what standard poverty measures report.

This paper describes the quality of the statistical matching between the March 2014 supplement to the Current Population Survey (CPS) and the 2013 American Time Use Survey (ATUS) and Survey of Consumer Finances (SCF), which are used as the basis for the 2013 Levy Institute Measure of Economic Well-Being (LIMEW) estimates for the United States. In the first part of the paper, the alignment of the datasets is examined. In the second, various aspects of the match quality are described. The results indicate that the matches are of high quality, with some indication of bias in specific cases.

This paper documents the sources of data used in the construction of the estimates of the Levy Institute Measure of Economic Wellbeing (LIMEW) for the years 1959, 1972, 1982, 1989, 1992, 1995, 2000, 2001, 2004, 2007, 2010, and 2013. It also documents the methods used to combine the various sources of data into the synthetic dataset used to produce each year’s LIMEW estimates.

This paper presents the quality analysis of the statistical matching conducted for a research study on household consumption behavior, household indebtedness, and inequality for Turkey. The match has been done for four years (2005, 2008, 2009, and 2012) of Household Budget Surveys (HBS) and the Survey on Income and Living Conditions (SILC). The aim of the statistical matching is to transfer household expenditure data from the HBS to the SILC to create synthetic data sets that have information on household consumption expenditures as well as household income and indebtedness. We are following the methodology of constrained statistical matching, using estimated propensity scores developed in Kum and Masterson (2010) to produce the synthetic data sets that we need. The analysis shows that the match is of high quality.

This document presents a description of the quality of match of the statistical matches used in the LIMTCP estimates prepared for Ghana and Tanzania. For Ghana, the statistical match combines the Living Standards Survey Round 6 (GLSS6) with the Ghana Time Use Survey (GTUS) 2009, and for Tanzania it combines the Household Budget Survey (THBS) 2012 with the time-use data obtained from the Integrated Labor Survey Module (ILFS) 2006. In both cases, the alignment of the two datasets is examined, after which various aspects of the match quality are described. Despite the differences in the survey years, the quality of match is high and the synthetic dataset appropriate for the time poverty analysis.

This paper evaluates the presence of heterogeneity, by household type, in the elasticity of substitution between food expenditures and time and in the goods intensity parameter in the household food and eating production functions. We use a synthetic dataset constructed by statistically matching the American Time Use Survey and the Consumer Expenditure Survey. We establish the presence of heterogeneity in the elasticity of substitution and in the intensity parameter. We find that the elasticity of substitution is low for all household types.

This paper describes the quality of the statistical match between the Current Population Survey (CPS) March 2011 supplement and the Consumer Expenditure Survey (CEX) 2011, which are used for the integrated inequality assessment model for the United States. In the first part of this paper, the alignment of the datasets is examined. In the second, various aspects of the match quality are described. The results show appropriate balance across different characteristics, with some imbalances within narrow characteristics.

This paper describes the quality of the statistical matching between the March 2011 supplement to the Current Population Survey and the 2010 American Time Use Survey and Survey of Consumer Finances, which are used as the basis for the 2010 LIMEW estimates for the United States. In the first part of the paper, the alignment of the datasets is examined. In the second, various aspects of the match quality are described. The results indicate that the matches are of high quality, with some indication of bias in specific cases.

The quality of match of the statistical match used in the LIMTIP estimates for South Korea in 2009 is described. The match combines the 2009 Korean Time Use Survey (KTUS 2009) with the 2009 Korean Welfare Panel Study (KWPS 2009). The alignment of the two datasets is examined, after which various aspects of the match quality are described. The match is of high quality, given the nature of the source datasets. The method used to simulate employment response to availability of jobs in the situation in which child-care subsidies are available is described. Comparisons of the donor and recipient groups for each of three stages of hot-deck statistical matching are presented. The resulting distribution of jobs, earnings, usual hours of paid employment, household production hours, and use of child-care services are compared to the distribution in the donor pools. The results do not appear to be anomalous, which is the best that can be said of the results of such a procedure.

The quality of match of three statistical matches used in the LIMTIP estimates for Argentina, Chile, and Mexico is described. The first match combines the 2005 Uso del Tiempo (UT 2005) with the 2006 Encuesto Annual de Hogares (EAH) for Argentina. The second match combines the 2007 Encuesta Experimental sobre Uso del Tiempo en el Gran Santiago (EUT 2007) with the 2006 Encuesta Caracteristización Socioeconómica Nacional (CASEN 2006) for Chile. The third match combines the 2008 Encuesta Nacional de Ingresos y Gastos de los Hogares (ENIGH 2008) with the 2009 Encuesta Nacional sobre Uso del Tiempo (ENUT 2009) for Mexico. In each case, the alignment of the two datasets is examined, after which various aspects of the match quality are described. In each case, the matches are of high quality, given the nature of the source datasets.

The quality of match for each of four statistical matches used in the LIMEW estimates for France for 1989 and 2000 is described. The first match combines the 1992 Enquête sur les Actifs Financiers with the 1989–90 Enquête Budget de Famille (BDF). The second match combines the 1998 General Social Survey (EDT) with the 1989–90 BDF. The third match combines the 2003–04 Enquête Patrimoine with the 2000–01 BDF. The fourth match combines the 1999 EDT with the 2000 BDF. In each case, the alignment of the two datasets is examined, after which various aspects of the match quality are described. In each case, the matches are of high quality, given the nature of the source datasets.

The quality of match of four statistical matches used in the LIMEW estimates for Great Britain for 1995 and 2005 is described. The first match combines the fifth (1995) wave of the British Household Panel Survey (BHPS) with the 1995–96 Family Resources Survey (FRS). The second match combines the 1995 time-use module of the Office of Population Censuses and Surveys Omnibus Survey with the 1995–96 FRS. The third match combines the 15th wave (2005) of the BHPS with the 2005 FRS. The fourth match combines the 2000 United Kingdom Time Use Survey with the 2005 FRS. In each case, the alignment of the two datasets is examined, after which various aspects of the match quality are described. In each case, the matches are of high quality, given the nature of the source datasets.

The quality of match of four statistical matches used in the LIMEW estimates for the United States for 1992 and 2007 is described. The first match combines the 1992 Survey of Consumer Finances (SCF) with the 1993 March Supplement to the Current Population Survey, or Annual Demographic Supplement (ADS). The second match combines the 1985 American Use of Time Project survey (AUTP) with the 1993 ADS. The third match combines the 2007 SCF with the 2008 March Supplement to the CPS, now called the Annual Social and Economics Supplement (ASEC). The fourth match combines the 2007 American Time Use Survey with the 2008 ASEC. In each case, the alignment of the two datasets is examined, after which various aspects of the match quality are described. Also in each case, the matches are of high quality, given the nature of the source datasets.

The quality of match of four statistical matches used in the LIMEW estimates for Canada for 1999 and 2005 is described. The first match combines the 1999 Survey of Financial Security (SFS) with the 1999 Survey of Labour and Income Dynamics (SLID). The second match combines the 1998 General Social Survey (GSS) with the 1999 SLID. The third match combines the 2005 SFS with the 2005 SLID. The fourth match combines the 2005 GSS with the 2005 SLID. In each case, the alignment of the two datasets is examined, after which various aspects of the match quality are described. Also in each case, the matches are of high quality, given the nature of the source datasets.