The comovement between epidemics and atmospheric quality in emerging countries

ABSTRACT

This research examines the short- or long-term relationship between epidemics and atmospheric quality via panel data of 69 countries over the period 1990–2019. By employing the panel univariate LM unit root test, panel cointegration tests with multiple structural breaks, and FMOLS estimations as well as the panel vector error correction model (VECM), we find that a bi-directional relationship among variables exists in the full sample. More importantly, from a long-term perspective we also note that the impact of epidemics on atmospheric quality is negative. Therefore, we hypothesize that this may be related to the retaliatory emissions of companies after epidemics and poor government supervision. For a more in-depth investigation, we take CO₂ emissions of the industrial and transportation sectors as the proxy variables and see that the more developed an economy is, the greater is the cointegration between epidemics and atmospheric quality. Our research offers implications for policy makers, such that improving atmospheric quality is an important way to prevent epidemics, and in order to alleviate and eliminate the spread of epidemics governments should pay more attention to environmental control.

KEYWORDS:

• Epidemics
• atmospheric quality
• co₂ emissions
• panel data
• panel cointegration

JEL:

• Q53
• I18

Notes

1. As of December 18, 2020, COVID-19 has caused more than 72 million confirmed cases and over 1,600,000 deaths worldwide, making it the most serious public health event of the 21^st century.

2. At present, the third generation of the air quality model “Models-3/CMA_Q” developed by the U.S. EPA, based on the concept of “One Atmosphere”, is the most representative.

3. The definition of Number in EM-DAT is the abnormal increase in the number of infectious disease cases already existing in the population of a country; or the emergence of new cases of infectious diseases in a country; Death refers to the number of deaths of these epidemic infection cases within a certain period (one year); Effect refers to the value of property loss caused by the epidemic. The estimated loss is expressed in US dollars (‘000), and the figures are shown true to the year of the event.

4. The results of Model AA are all available upon request. We also conduct the NP (2010) unit root test with two breaks, which are provided by Narayan and Popp (2010). Since the LM test can also report the panel results, we only list the results of the LM test. The NP (2010) results are available upon request.

5. To conserve space, we just report the LM unit root test of EE. The results of EN and ED are available upon request.

6. Population-weighted exposure to ambient PM_2.5 pollution is defined as the average level of exposure of a nation’s population to concentrations of suspended particles measuring less than 2.5 microns in aerodynamic diameter that are capable of penetrating deep into the respiratory tract and causing severe health damage. Exposure is calculated by weighting mean annual concentrations of PM_2.5 by population in both urban and rural areas.

7. Incidence of tuberculosis is the estimated number of new and relapsed tuberculosis cases arising in a given year, expressed as the rate per 100,000 population. All forms of TB are included, including cases in people living with HIV. Estimates for all years are recalculated as new information becomes available and techniques are refined, and so they may differ from those published previously. The definition of emissions from the CO_2industry and CO_2transport: https://www.iea.org/data-and-statistics/

8. Because the parameters $\lambda$ are the error-correction items for examining the long-run relationship between CO₂ emission and epidemics, the lag terms of the other variables help test for short term causality. Therefore, the joint test with ($\lambda$/$\mathrm{\Delta }$CO2) and ($\lambda$/$\mathrm{\Delta }$Epidemic) can comprehensively analyze the causality of two variables in the long run. In other words, if there exists an equilibrium in the long run, then the parameters $\lambda$ are statistically significant, and we thus present the bi-directional relationships of CO₂ emission and epidemics by the parameters ($\lambda$/$\mathrm{\Delta }$CO2) and ($\lambda$/$\mathrm{\Delta }$Epidemic), respectively.