Search

This paper proposes a method to analyze interval-censored data, using multiple imputation based on a heteroskedastic interval regression approach. The proposed model aims to obtain a synthetic data set that can be used for standard analysis, including standard linear regression, quantile regression, or poverty and inequality estimation. The paper presents two applications to show the performance of the method. First, it runs a Monte Carlo simulation to show the method's performance under the assumption of multiplicative heteroskedasticity, with and without conditional normality. Second, it uses the proposed methodology to analyze labor income data in Grenada for 2013–20, where the salary data are interval-censored according to the salary intervals prespecified in the survey questionnaire. The results obtained are consistent across both exercises.

We contribute new UK evidence about measurement errors and employment earnings to a field dominated by findings about the USA. We develop and apply new econometric models for linked survey and administrative data that generalize those of Kapteyn and Ypma (Journal of Labor Economics, 2007). Our models incorporate mean-reverting measurement error in administrative data in addition to linkage mismatch and mean-reverting survey measurement error and 'reference period' error, while also allowing error distributions to vary across individuals. Annualised survey earnings underestimate true annual earnings on average. Mean-reversion in survey measurement errors is absent. Both earnings sources underestimate true earnings inequality. The survey earning measure is more reliable than the administrative data earnings measure, but hybrid earnings predictors based on both sources are distinctly more reliable than either source-specific measure. The models with heterogeneous measurement error distributions indicate how data quality may be improved. For example, for survey quality, our results highlight how respondents showing payslips to interviewers have smaller survey error variances. For administrative data, our results suggest that greater error variances are associated with non-standard jobs, private sector jobs, and employers without good payroll systems.

Researchers use finite mixture models to analyze linked survey and administrative data on labour earnings (or similar variables), taking account of various types of measurement error in each data source. Different combinations of error-ridden and/or error-free observations characterize latent classes. Latent class probabilities depend on the probabilities of the different types of error. We introduce a set of Stata commands to fit a general class of finite mixture models to fit to linked survey-administrative data We also provide post-estimation commands for assessment of reliability, marginal effects, data simulation, and prediction of hybrid earnings variables that combine information from both data sources.

Meijer, Rohwedder, and Wansbeek (MRW, Journal of Business & Economic Statistics, 2012) develop methods for prediction of a single earnings figure per worker from mixture factor models fitted using earnings data from multiple linked data sources. MRW apply their method using parameter estimates of Kapteyn and Ypma's mixture factor model (KY, Journal of Labour Economics 2007) fitted to earnings data for Swedish workers aged 50+. First, we replicate MRW's empirical analysis using the Swedish model estimates. Second, we check the generality of their empirical finding with a new application. Using estimates of a KY model fit to a linked dataset on earnings for UK employees of all ages, we confirm that MRW's principal findings about the performance of their various predictors of true earnings also hold in this different setting.

In this paper, we argue that structural inequalities by race, class and gender are responsible for asymmetries in the costs of COVID-19 pandemic borne by individuals in the U.S., both in terms of health outcomes and economic well-being. Minority and low-income individuals are more exposed to contamination and more likely to develop severe infections that can lead to hospitalization and death due to COVID-19. Likewise, women and racialized populations are experiencing higher economic impacts in the form of job losses and fall in their well being. Policies designed to combat the epidemic have been failing in closing the racial and class gap. The coronavirus outbreak is exacerbating biases and increasing social, gender, and racial gaps—and consequently increasing the length and severity of the crisis.

In this working paper, we analyze factors that may explain gender differences in the allocation of time to household production in sub-Saharan Africa. The study uses time use survey data to analyze the determinants of time spent on household production by husbands and wives in nuclear families in Ethiopia, Ghana, Tanzania, and South Africa. We assume that the time spent by each spouse is a function of personal and household characteristics. A bivariate Tobit model is used to estimate the marginal impact of a set of key variables that figure recurrently in the literature on time allocation. We observe a high degree of variability in the results for the set of countries, which does not allow us to draw hard general conclusions. We do find some weak evidence that supports time availability and gender ideology theory as well as for the hypothesis that bargaining power plays a role in explaining the intrahousehold allocation of household production.

This paper presents a description of the quality of match of the statistical matches used in the 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.

In this paper we examine the education and occupation mismatch for Hispanics in the US using a novel objective continuous mismatch index and explore the role of immigrants' social networks on this mismatch. We explore whether having a larger social network helps Hispanics in finding jobs that better match with their skill and education levels or whether living in areas with larger concentration of Hispanics leads to more competition for the same jobs in the labor market. Given that the legal status of immigrants influence how the social networks are leveraged and their impact on labor market outcomes, we focus on the citizenship status for Hispanics. The quality of match between Hispanic's college degree major and occupation is measured using one of the continuous indices proposed in Rios-Avila and Saavedra-Caballero (2019) and calculated using pooled data for all college graduates in the US from 2010 to 2017. The Hispanic networks measures are constructed as the share of Hispanic population who are 25 years or older with respect to the total population of the same age and the second measure only includes Hispanics with at least a bachelor's degree using the weighted pooled data from 2010 to 2015. We find that networks have a positive impact on the job-match quality, but mostly for Hispanic citizens and this effect is stronger when the networks constitutes of at least a college degree. This shows that Hispanic citizens living in higher concentration of Hispanic college graduates are better able to leverage their networks or their networks are better able to match them with jobs closer to their field of specialization and skill set.

Whether China's low fertility rate is the consequence of the country's strict population control policy is a puzzling question. This paper attempts to disentangle the Chinese population control policy's impacts on the fertility rate from socioeconomic factors using the synthetic control method proposed by Abadie and Gardeazabal (2003). The results indicate that the population control policy significantly decreased China's birth rate after the "Later, Longer, and Fewer" policy came into force, but had little effect on the birth rate in the long run. We estimate that between 164.2 million and 268.3 million prevented births from 1971 to 2016 can be attributed to the Chinese population control policy. In addition, we implement a placebo study to check the validity of the method and confirm the robustness of the paper's conclusions.

The costs of the COVID-19 pandemic - in terms of both the health risks and economic burdens - will be borne disproportionately by the most vulnerable segments of US society. In this public policy brief, Luiza Nassif-Pires, Laura de Lima Xavier, Thomas Masterson, Michalis Nikiforos, and Fernando Rios-Avila demonstrate that the COVID-19 crisis is likely to widen already-worrisome levels of income, racial, and gender inequality in the United States. Minority and low-income populations are more likely to develop severe infections that can lead to hospitalization and death due to COVID-19; they are also more likely to experience job losses and declines in their well-being. The authors argue that our policy response to the COVID-19 crisis must target these unequally shared burdens - and that a failure to mitigate the regressive impact of the crisis will not only be unjust, it will prolong the pandemic and undermine any ensuing economic recovery efforts. As the authors note, we are in danger of falling victim to a vicious cycle: the pandemic and economic lockdown will worsen inequality; and these inequalities exacerbate the spread of the virus, not to mention further weaken the structure of the US economy.