Leveraging Global Talent

Which Countries Are More Open to Immigration? The answer to this simple question is far from straightforward. Consider this scenario: Is a country where 30% of the population is foreign-born—but all from a single neighboring nation—more open than a country where 20% of the population is foreign-born, yet from geographically, economically, and culturally diverse backgrounds? Many factors influence a country's openness to immigration. Wealthier, larger, and geographically well-positioned nations tend to attract more migrants. Historical ties, shared languages, and proximity to conflict zones or regions affected by natural disasters also play significant roles.

To isolate a country’s openness to immigration from these external influences, we apply econometric techniques that partial out the effect of these factors from the immigration rate of a place. Specifically, we use gravity models of migration, which consider bilateral migration flows as a function of geographic distance, linguistic and historical connections, shared borders, and country-level characteristics such as income, population, and land area. The key metric in our analysis are the residuals of the gravity regressions—the difference between a country’s actual migration levels and those predicted by the model. Host countries with a larger number of positive residuals are considered more open. Intuitively, we say that a place A is open to place B, if A receives more migrants than expected from B, based on their distance, population, wealth, and cultural ties.

Data

Our data comes from four sources. The data on migration stocks comes from the United Nations' global matrices of bilateral migrant stocks for 2000, 2010, and 2020. The information about the share of tertiary and non-tertiary immigrants in the total migrant stock is based on estimates by the World Bank as used in the World Development Report 2023. Most explanatory variables for the Gravity model come from the CEPII Gravity database (Conte et al. 2022), and the GDP p.c. PPP, total population and the land area variables come from the World Bank Development Indicators.

Missing values. In order to increase the reliability of our measures, we limit the estimates to places whose population in all of these years was greater than 1.2 millions. Because of data reliability concerns data for Syria and Cuba, as places of destination, was not included in the analysis. The estimates are available for 148 places. Additionally, data was unavailable for the following places and years: Afghanistan in 2000, Eritrea in 2020, Somalia in 2000 and 2010, South Sudan, Turkmenistan, Venezuela and Yemen in 2020. Bilateral data is unavailable between China and Hong Kong and between Israel and Palestine.

Definition of skills. Any skills refers to migrants with any level of education, including those without education. High skills refers to migrants with tertiary education, and low skills to those with less than tertiary education.

Rankings. To get to the rankings, we estimate the diversity-based openness measure using 6 different thresholds for what is considered a positive residual, take the average of these, round up the average to the closest (larger) integer, and take the rank of the rounded average. The thresholds for the size of the residuals are: 5, 6, 7, 8, 9 and 10 in a 1 million inhabitants of the host country. This renders a more robust measure of a country's rank, as opposed to rank based on a single estimate.

Data access. Visit GitHub repository https://github.com/LjubicaN/Openness.

Preprint

Nedelkoska, L., Martin, D., Lochmann, A., Hausmann, R., Bahar, D., & Yildirim, M. A. (2025). De facto Openness to Immigration. arXiv preprint arXiv:2502.16407.

Templeton Project

This publication is based on research supported by the Templeton World Charity Foundation, Inc. (DOI: 501100011730) under the grant Leveraging the Global Talent Pool to Jumpstart Prosperity in Emerging Economies awarded to the Growth Lab at Harvard University (Grant No. TWCF-2022-30478). Ricardo Hausmann is the Principal Investigator.