About Us
The International Demographic Inequality Lab (IDIL.LI) aims to contribute to your factfulness about the historical and recent trend of various dimensions of inequality by providing open access to demographic data processed by carefully selected methods.
The most important input data we use are census data sourced from IPUMS and other population data. These input data are about who marries whom, who cohabits with whom, how long people live, what is the education level of individuals and their parents.
We transform data into evidence about inequality by applying rigorously selected models and methods. In particular, our models and methods make the data speak about how permissive is a given generation about marrying out of their own educational group or racial group relative to a previous generation; what is the life expectancy gap between people in different educational and racial groups; whether the (lack of) high school degree or (lack of) college degree is more heritable from parents to children.
Our set of indicators covers approximately 80 countries including both developed and developing countries. In case of the US, some of our indicators are computed at the sub-national level, i.e., for each state in the United States.
To allow you to study both the historical trend and the recent evolution of inequality, we computed some of our indicators for the early Silent generation (born between 1928 and 1936), as well as for the next generations up until the Generation Z (born between 1997 and 2012). We keep updating our indicators with new observations as new input data are released.
BACKGROUND
Inequality has became a prominent part of the political discourse since the publication of the seminal book Capital in the Twenty-First Century in 2014 by Thomas Piketty. However, there is a substantial knowledge gap about global trends – including the trend of various dimensions of inequality as it was documented in the book Factfulness by Hans Rosling, Anna Rosling Rönnlund and Ola Rosling.
IDIL.LI promotes an evidence-based view on the historical and recent trends of inequality. Thereby, it aims to support evidence-informed policy-making.
While these objectives are largely overlapping with the objectives of WID.WORLD (i.e., Piketty’s inequality lab), and GAPMINEDER.ORG (co-founded by Hans Rosling, Anna Rosling Rönnlund and Ola Rosling), there are some important differences as well.
For instance, while WID.WORLD focuses on the evolution of the monetary dimensions of inequality (i.e., income inequality and wealth inequality), IDIL.LI reports the trends mainly in the non-monetary dimensions, such as health disparity signaled by the life expectancy gap; inequality of opportunity documented by studying intergenerational mobility; and the overall level of inequality covering all those dimensions of inequality (e.g., health, wealth, income, opportunity to be employed) that are relevant for marital choice in an aggregated form by letting the marriage market performing the aggregation.
However, IDIL.LI‘s most important distinctive feature is not the scope of dimensions covered. Similarly, it is neither the scope of countries analyzed, nor the scope of questions addressed that the WID.WORLD and the GAPMINEDER.ORG are champions at, respectively.
IDIL.LI is outstanding, because of the genuine approaches we apply to transform publically available data into evidence. The NM-method and the GNM-method are prominent examples. Both methods serve the purpose of performing counterfactual decompositions by constructing counterfactuals. The counterfactual decompositions are used to study the trend in the directly unobservable non-structural determinant of the prevalence of marital homogamy (e.g. marital preferences and norms shaped by the overall level of inequality) by controlling for the observable structural determinant, such as the trait distributions of marriageable men and women.
The NM-method was developed by Anna Naszodi and Francisco Mendonca (2021). It was first applied in a policy brief of the European Commission by the same authors in 2019, and used in a comprehensive policy report published in 2019.
Additional applications of the NM-method, the analysis of its analytical properties and empirical performance, its validation with dating data and survey data, as well as its comparison with a comprehensive set of alternative methods can be found in the papers by Naszodi (2021b), Naszodi and Mendonca (2022), and Naszodi (2023 a, b, c, d).
The generalized version of the NM-method, the GNM-method, was conceptualized by Naszodi (2021a), while being implemented and applied by Naszodi and Mendonca (2023). It is applicable to analyze marital sorting along multiple dimensions, such as race and education level. Recently, these works were featured in the New York Times.
STATE OF THE PROJECT (2023 April)
Some parts of IDIL.LI are still under construction. Progress is expected to speed up after hiring research assistants to be financed by grants.
FUTURE PLANS
We plan to achieve full transparency about the input data used and the methods applied. Also, it is on our agenda to show case how our indicators can be used for policy impact assessments.