The asymmetric impact of the spread of COVID‑19 and its consequences between countries and regions within countriesFootnote 3 is associated with many factors of various nature, each contributing to the stochastic process. Complex nonlinear relationships create a special situation for each location. The origins of the spatial asymmetry of regional development are diverse, associated with factors of both the first (geography and climate) and second (population density and concentration of economic activity) type (Bailey et al., 2020; Kolomak, 2013). In a pandemic, factors that have had clear benefits in the past (such as agglomeration effects and density of interactions) can contribute to the spread of COVID-19 and its consequences.
Different regional characteristics are significant for different stages of the spread of a pandemic. The hardest hit regions in the first wave were major metropolitan areas such as New York and London, with vibrant social life, diverse populations, and densely populated residential areas. In second place in terms of the spread of infection were large industrial centers connected by global supply chains (Azzolina et al., 2020).
The first wave of the pandemic in Russia (March–May 2020) mostly affected the largest agglomerations with a high population density and high level of contacts: international centers of transport, commodity and financial flows, and border and coastal regions, with a predominance of random factors (Kalabikhina and Panin, 2020; Zemtsov and Baburin, 2020a). An OECD report and an article by Russian researchers (Puzanov and Alov, 2020) emphasize that a direct relationship between population density and morbidity has not been established. The incidence rate with respect to the population of a region in the first wave of the pandemic was higher not in the most populated regions, but in smaller populated areas, where outbreaks of infections associated with “supercarriers” occurred (Bailey et al, 2020), which indicates the strong influence of nonsystemic (random) factors on the epidemiological situation. Perhaps the most famous example is a conference in Boston (United States) in February 2020, which brought together leaders in biotechnology from around the world: 90 people were infected,Footnote 4 who then dispersed to cities and countries.
The second wave began in September 2020, and we believe that the dynamics and scale of infection were influenced by more fundamental characteristics of the regions, which is confirmed by the stable ranking (preservation of order) of the most affected regions.
For this study, we divided the numerous features of the regions that can affect the spread of coronavirus into two groups of characteristics: (1) those that reflect the vulnerability of the region to the spread of infection and (2) those that develop the ability to resist the spread of the disease.
A region’s vulnerability to the spread of infection depends on a wide range of factors: climatic (in warmer and drier regions, spreading is less), geographic, demographic, economic, social, political (in the United States, red states with Republican leadership introduced less stringent restrictions and experienced a greater increase in morbidity than blue states (Hallas et al., 2021)) and others, which directly or indirectly affect the incidence of coronavirus infection.
Resistance to the spread of a new, unknown infection transmitted by contact in the absence of specific vaccines and medicines depends on the speed and rigidity of implemented measures to limit mobility and contacts between people (closing of external borders, borders between regions within countries, stopping business activity, introduction of quarantine measures, social distancing, etc.). Restrictions on economic and social contacts and their observance have been and remain the main nonpharmaceutical actions aimed at inhibiting infection processes.
We have focused on several regional characteristics associated with the spread of the coronavirus, as demonstrated by the research results.
Wealth of regions. The indicator of gross income per capita is conventionally used in regional studies. The relationship between gross income per capita and infection and mortality rates can be both negative and positive. Regions with high GRP per capita are more urbanized, embedded in global trade and transport flows, large companies are located in them, and they are characterized by a rich social life. As a result of the concentration of economic and social activity in wealthy regions, higher infection and mortality rates can be expected. This assumption is confirmed, e.g., by the results of (Kapitsinis, 2020) across 119 regions in 9 European Union countries: wealthier regions have higher infection and mortality rates. However, wealthy regions have not only a dense network of contacts, but also more developed infrastructure and healthcare systems, and there are also more opportunities for telecommuting.
At the same time, in poor regions, many people cannot stop economic activity due to low income; in addition, they have fewer opportunities to work remotely due to the nature of work or lack of access to appropriate infrastructure. In such regions, provision of health services is also lower. Woods (2020) argues the existence of a feedback loop between the wealth of the region and the prevalence of infection: both COVID-19 and measures taken to control its spread predominantly affect poor strata and impoverished districts. Factors that contribute to higher mortality in poor areas include a high prevalence of chronic diseases, limited access to health services, and demographic and occupational characteristics.
Population density. Among the factors contributing to the spread of infection, researchers consider the population density and size, including the presence of large urban agglomerations in a region. However, the results obtained are ambiguous.
Large cities are characterized by a high density of businesses, a developed communications network, including public transport, the presence of large shops and other places where people concentrate, which makes it possible to consider COVID-19 a megalopolitan pandemic. Conversely, rural areas have suffered the least (Woods, 2020).
Yu. Ponomarev and D. Radchenko (2020) confirm that the populations of million-plus cities are most at risk of contracting coronavirus. S.P. Zemtsov and V.L. Baburin (2020a, 2020b) hew to a similar viewpoint, noting that regions with a high proportion of urban population are most susceptible to the spread of a pandemic, since in cities not only is the sanitation intensity higher, but natural and ecological conditions also deteriorate, which negatively affects people’s health.
However, there is evidence that refutes this viewpoint. For example, recent US data show that regions with a higher proportion of rural populations have been hit harder. A.S. Puzanov and I.N. Alov (2020), based on a review of international publications, explain the absence of a direct relationship between population density and the spread of infection by the fact that, in developed countries, densely populated regions with a high level of prosperity, education of residents, and better access to broadband Internet have greater potential for social distancing and teleworking, which makes it possible to more successfully counter the spread of infection.
The environment, primarily, air quality, has a significant impact on morbidity (infection and lethality). Air pollution levels can affect the incidence of COVID-19 in several ways. Among them, the most significant are the higher susceptibility to lung disease in regions with high levels of air pollution with particulate matter, as well as the possibility of long-distance viral transmission via these particles. Moreover, the results of some studies (Accarino et al., 2020; Azzolina et al., 2020; Becchetti et al., 2020; Perone, 2021) suggest a causal relationship between air pollution and the effects of infection.
Resistance to the pandemic. Up to early 2021, “nonpharmaceutical interventions” were the main means of countering the spread of infection. Social distancing and stopping or limiting contact-related activity (lockdowns) remain the main ways of combating the spread of infection. Most countries have adopted various restrictive measures, but it is important not only to legislate norms and rules, but also to comply with them.
The reaction of people to restrictions and the need to follow them over, as it turned out, a long period is associated with many intertwining circumstances. Among them, not only economic losses (e.g., loss of livelihoods) are important, but also an acute change in everyday life as a result of forced isolation and breakdown of social ties, accompanied by increased anxiety, fear, and depression. Coerced social distancing triggers resistance, as evidenced by massive fines for violations, as well as abrogation of the severest restrictions (in particular, the rapid repeal of electronic passes in several Russian regions).
Compliance with constraints is not easy to assess directly. Whereas people’s movements can be tracked (to some extent) using data from mobile devices,Footnote 5 violations of compliance with social distancing, as well as sanitary standards and the use of personal protective equipment, are much more difficult to assess. An indirect characteristic may be the number of penalties imposed for violation of restrictions. At the end of November 2020, Russia faced the administrative responsibility of penalities for violating pandemic-related restrictions, more than 1.1 mln people.Footnote 6
Maloney and Taskin (2020) show that the decline in mobility in developed countries is largely voluntary, dependent on awareness, fear, or social responsibility and occurs independently of the demands of government agencies; i.e., it reflects more deliberate and informed behavior rather than a reaction to repressive measures. However, for the poorest countries, such an effect did not appear, which is interpreted as forced refusal to comply with mobility restrictions due to lack of livelihood.
We believe that in the long term, compliance with the restrictions is more related to the social capital accumulated in a regional or local community. Social capital is viewed as a set of norms, values, and interactions adopted in a community that promote cooperation within or between groups and ensure the organization of collective actions to achieve social well-being (Grootaert, 2004; Kosarev et al., 2019; Polishchuk and Menyashev, 2011). Social capital includes trust and solidarity, collective action and cooperation, civic responsibility, social cohesion, and other categories.
We assume that in communities with a higher level of social capital, the degree of compliance with restrictive requirements will be higher. This will have a positive impact on reducing the risk of infection. There are studies that support our hypothesisFootnote 7 according to which there is a high coronavirus mortality rate in countries where the level of trust in the government is low. Trust in government helps people overcome their pent-up “restriction fatigue” and continue to comply with them. During the first wave of the pandemic, trust in government increased in many countries; trust in regional and local authorities increased more often.
Measuring social capital is a difficult task. While there are several common indicators for the national level, there are very few empirical tools for assessing social capital for Russian regions. Scientists use various indicators that indirectly reflect the level of social capital.Footnote 8 We used data on voter turnout as an indicator of civic responsibility, cohesion, and trust in the authorities (note that this is an indicator used in international assessments of social capital (Grootaert, 2004)).