2.1. Sample
The sample comprised 167 countries, excluding 11 countries that produce authorized vaccines, namely the United Kingdom, United States, Germany, China, Russia, Cuba, India, Kazakhstan, Taiwan, Uzbekistan, and the Netherlands. The origin of the vaccine manufacturer is elicited from the COVID-19 Vaccine Tracker website, except for Pfizer/BioNTech, which does not report any specific country name. We alternatively defined vaccine origin based on the location of the headquarters.
2.2. Measures
Vaccination Outcomes
Three types of vaccination outcomes were used as dependent variables: the number of days until the administration of the first vaccine dose in the country since December 31, 2019 (when the WHO China Country Office was informed of cases of pneumonia of unknown etiology detected in Wuhan City, China), the number of doses implemented per hundred citizens as of July 30, 2021; these data come from Our World in Data [5]. The third outcome was seven binary indicators which take unity if each of the major vaccines are approved and used in the country, and zero otherwise. The major vaccines include Oxford/AstraZeneca, Pfizer/BioNTech, Moderna, Johnson&Johnson, Sputnik V, Sinopharm/Beijing, and Sinovac. Data on the third outcome were obtained from the COVID-19 Vaccine Tracker website [10].
Governance Index
The national governance data was obtained from the Worldwide Governance Indicators, which consist of six indicators in 204 countries: 1) voice and accountability; 2) political stability and absence of violence; 3) government effectiveness; 4) regulatory quality; 5) rule of law; and 6) control of corruption [11]. These are frequently employed in the literature on national governance [12]. We elicited the indices as of 2019—the latest values in the dataset. Standardized indices were used for the empirical analyses (mean = 0, SD = 1).
Controls
This study also used the cumulative number of confirmed cases as of July 30, 2021, the population size as of 2020, GDP per capita in the most recent year available, and binary indicators of member countries of the OECD and the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) as of 2021. The first three variables were obtained from Our World in Data, while the data on OECD and ICH membership come from the official website of each organization.
2.3. Statistical Analysis
First, for data reduction, we performed a principal component analysis using the items for national governance. Following the Kaiser-Harris criterion, we retained components with eigenvalues greater than one. The obtained composite index was standardized (mean = 0; SD = 1).
Second, an ordinary least squares (OLS) regression was conducted to examine the cross-country association between national governance and two outcome variables on vaccination—days until the administration of the first dose in the country and the number of doses per 100 citizens. Specifically, these outcomes were regressed on the composite governance index, GDP per capita, population size, indicator of ICH member countries, and the cumulative number of confirmed cases. Results are reported as OLS coefficients with 95% confidence intervals (95% CIs).
Third, to explore which component of governance plays a central role, we regressed the outcomes on individual governance indices separately.
Fourth, to test the heterogeneous patterns across vaccine manufacturers, multivariate logistic regression analysis was conducted for the seven major vaccines. The dependent variables were binary indicators of whether the country approved and used the vaccine. Results from regression analyses were reported as odds ratios (ORs) with 95% CIs.
We compared the results from the model using all countries (N = 167) and the model including only non-OECD countries (N = 133) throughout the analyses. All analyses were performed using the Stata 17.