Is There a Link Between Travel and Spread of Covid-19?

Does travel from China help predict the number of Covid-19 cases in other countries? We can address this question by looking at mobility data gathered by the Global Mobilities Project of the Migration Policy Centre at the European University institute, Florence, Italy.

We could assume that travel from China should help to predict spread of the disease as it reflects the risk of exposure to the main source of the virus, but our research shows that mobility alone is not a sufficient explanation .

Our data shows that country-to-country travel flows are quite similar year after year. We correlated the volume of incoming visitors from China in 2016 to different nations and the number of registered cases of Covid-19 as of 23 February 2020. We do find a positive correlation (r=0.18), but there are several and sizeable outliers. For instance, Russia received 3,264,000 incoming trips from China, but had only 2 registered cases. In contrast, Italy had far fewer incoming travels at 712,000, but had 270 cases as of 23 February. South Korea and Iran are also much more likely to have Covid-19 patients than travel from China would lead us to predict, whereas Hong Kong and Macao have far fewer.

Our analysis suggest that there are clearly many more determinants than mobility and travel alone to explain differences in the spread of Covid-19. These include the extensiveness of tests to detect the virus, the vulnerability of resident populations, types of border controls and the use of quarantine. All of these vary greatly across countries. 

Source: Global Transnational Mobility Dataset (GTMD), MPC/EUI and Johns Hopkins University, CSSE. Both axes in the figure are log-transformed.