Lessons for the Age of Consequences: COVID-19 and the Macroeconomy

Figures & data

Figure 1. Cumulative confirmed COVID-19 deaths per 100,000 population (data up to February 23, 2021 and up to September 21, 2021).

Source: Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality). NL = The Netherlands; NZ = New Zealand; Switz = Switzerland. The same in all subsequent figures. See Table A1 (Appendix).

Figure 2. Scatterplot of additional public spending versus cumulative confirmed COVID-19 deaths per 100,000 population.

Sources: Data on additional (discretionary) public spending (as percentage of GDP, until January 2021) are from the IMF (2020) October 2020 Fiscal Monitor Database of Fiscal Measures in response to COVID-19; data on COVID-19 mortality are from Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021. Notes: (1) average additional public spending (as percentage of GDP) is 7.2% for the panel of 22 OECD countries; the figure reports country-wise deviations from this average; (2) the unweighted average cumulative confirmed COVID-19 mortality is 86.4 deaths per 100,000 population for the panel of 22 OECD economies; the figure reports deviations from this average. (3) the estimated linear relationship is negative and statistically significant at 2.5% (when I exclude the observations for the U.K. and the U.S. from the regression).

Figure 3. Scatterplot of change in real GDP during 2019–2020 versus cumulative confirmed COVID-19 deaths per 100,000 population.

Sources: data on change in real GDP during 2019–2020 are from AMECO Database; data on COVID-19 mortality are from Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021. Notes: (1) the (unweighted) average decline in real GDP for the panel of 22 OECD economies is 6.2%; the figure reports country-wise deviations from this average; (2) the unweighted average cumulative confirmed COVID-19 mortality is 86.4 deaths per 100,000 population for the panel of 22 OECD economies. (3) the estimated linear relationship is negative and statistically significant at 1%.

Figure 4. Scatterplot of percentage change in life expectancy (2014-2018) versus cumulative confirmed COVID-19 deaths per 100,000 population.

Sources: data on percentage change in life expectancy at birth (total population) during 2014–2018 are from World Bank’s World Development Indicators (https://data.worldbank.org/indicator/SP.DYN.LE00.IN?name_desc=false); data on COVID-19 mortality are from Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021. Notes: (1) the average percentage change in life expectancy at birth (total population) during 2014–2018 was 0.38% for the 22 OECD countries; (2) the unweighted average cumulative confirmed COVID-19 mortality is 86.4 deaths per 100,000 population for the panel of 22 OECD economies; (3) the estimated linear relationship is negative and statistically significant at 1%.

Figure 5. Scatterplot of (after-tax-and-transfer) inequality (Gini) versus cumulative confirmed COVID-19 deaths per 100,000 population.

Sources: Gini-coefficients are from OECD Data (https://data.oecd.org/inequality/income-inequality.htm); data on COVID-19 mortality are from Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021. Notes: (1) the average (after-tax-and-transfer) Gini coefficient for the panel of 22 OECD economies is 30.9; the figure reports country-wise deviations from this average; (2) the unweighted average cumulative confirmed COVID-19 mortality is 86.4 deaths per 100,000 population for the panel of 22 OECD economies; the figure reports deviations from this average. (3) the estimated linear relationship is positive and statistically significant at 2.5% (when I exclude the observations for Belgium from the regression).

Figure 6. Additional public spending on COVID-19 relief (in euro’s & until January 2021).

Sources: Data on additional public spending, until (January 2021) are from the IMF (2020) October 2020 Fiscal Monitor Database of Fiscal Measures in response to COVID-19; data on GDP and population are from AMECO database.

Figure 7. Scatterplot of public debt in 2019 (% of GDP) versus additional public spending on COVID-19 relief (euros per person).

Sources: Data on public debt in 2019 (as percentage of GDP) are from AMECO Database; data on additional public spending, until (January 2021) are from the IMF (2020) October 2020 Fiscal Monitor Database of Fiscal Measures in response to COVID-19; data on GDP and population are from AMECO database. Note: the estimated linear relationship is negative and statistically significant at 1% (when I exclude the observations for the U.S.A. from the regression).

Figure 8. Scatterplot of additional pubic spending on COVID-19 relief (euros per person) versus cumulative confirmed COVID-19 deaths per 100,000 population.

Sources: Data on additional public spending, until (January 2021) are from the IMF (2020) October 2020 Fiscal Monitor Database of Fiscal Measures in response to COVID-19; data on GDP and population are from AMECO database. Data on COVID-19 mortality are from Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021. Note: the estimated linear relationship is negative and statistically significant at 2.5% (when I exclude the observations for the U.S.A. from the regression).

Box 1. Austerity and public health policy: the U.K. and the U.S.

the U.K.

the U.S.A.

Between 2009–10 and 2018–19 spending on the National Health Service declined from 7.6% to 7.2% of GDP. In terms of growth as a percentage of GDP, the nine-year period since 2009–10 is the lowest since the first decade (1948–58) of the National Health Service (NHS). Government spending declined from 42% of GDP in 2009–10 to 35% of GDP in 2018–19. Social protection and education spending each declined by 1.5 percentage points of GDP. In 2007–08, 45% of all government spending on the early years and childcare support was targeted explicitly at low-income families. By 2018, the share of public spending on low-income families had decreased to 27%. The number of hospital beds has been negatively affected by Private Equity buyouts of hospitals British government spending has not only declined in key social determinants of health, but that is now also allocated in a regressive manner — meaning that spending allocations are less weighted towards deprived areas and communities than previously. Source: Marmot (2020a, 2020b).

Between 2002 and 2019, the share of US health spending devoted to public health fell from 3.21% to 2.45% of GDP. Funding of the Public Health Emergency Preparedness Program-me (the main source of federal support for state and local public health emergency capacity) fell by one-third. As a consequence of funding shortfalls, state and local public health agencies lost 50,000 positions, a 20% decrease in the front-line workforce for fighting epidemics. Access to medical care, always compromised in the U.S., contracted further during the Trump administration. One million health-care workers and many immigrant workers at high risk of coronavirus exposure were uninsured at the beginning of the COVID-19 health emergency. The number of people without health insurance had increased by 2.3 million during Trump’s presidency, even before COVID-19-crisis-driven losses of employment-based coverage increased the number of uninsured people by millions. The number of hospital beds has been negatively affected by Private Equity buyouts of hospitals (in favour of outpatient care); see Appelbaum and Batt (2020). During 2010–20, the budget of Centres for Disease Control and Prevention (CDC) was lowered by 10% (inflation-adjusted). A hiring freeze in 2017 left hundreds of CDC positions for researchers and officials vacant. Source: The Lancet commission on ‘Public Policy and Health in the Trump Era’ (Woolhandler et al. 2021, p. 708).

Figure 9. Scatterplot of (after-tax-and-transfer) Gini coefficients versus change in real GDP during 2019–2020.

Sources: Data on real GDP are from AMECO Database. Data on inequality (Gini coefficients) are from the OECD (https://data.oecd.org/inequality/income-inequality.htm). Note: the estimated linear relationship is negative and statistically significant at 1% (when I exclude the observations for the U.S.A. from the regression).

Figure 10. S&P500: stock-market ‘panic’, followed by ‘mania’ (February 2020–February 2021).

Source: S&P Dow Jones Indices, FRED Database. Notes: The index is shown in thousands; rates of growth indicated in the figure are for the respective phase.

Figure 11. M3/GDP (percentage) and inflation: U.S.A (1970–2020). Panel a: M3/GDP (1970–2020). Panel b: scatterplot of annual percentage change in the M3/GDP ratio and annual inflation (1970–2020).

Sources: Data on M3 and GDP are from OECD Statistics. Data on inflation (measured by annual changes in the Price Index of Personal Consumption Expenditures) are from the Federal Reserve. The estimated linear relationship is negative and statistically significant at 1%.

Figure 12. M3/GDP (percentage) and inflation: U.K. (1986–2020). Panel a: M3/GDP (1986–2020). Panel b: scatterplot of annual percentage change in the M3/GDP ratio and annual inflation (1986–2020).

Sources: Data on M3 and GDP are from OECD Statistics. Data on inflation (measured by annual changes in the Consumer Price Index) are from the Federal Reserve. The estimated linear relationship is not statistically significant.

Figure 13. Scatterplot of tax-to-GDP ratio (in 2019) versus the income share of the top 1%.

Sources: Data on the tax-to-GDP ratio in 2019 are from OECD Statistics; data on the income share of the top 1% are from the World Inequality Database. Notes: (1) the (unweighted) average tax-to-GDP ratio is 36.1% for the panel of 22 OECD countries; (2) the unweighted average income share of the top 1% is 11.5% for the panel of 22 OECD economies. (3) the estimated linear relationship is negative and statistically significant at 1%.

Figure 14. Scatterplot of tax-to-GDP ratio (in 2019) versus the Gini coefficient of (after-tax-and-transfer) income inequality.

Sources: Gini-coefficients are from OECD Data (https://data.oecd.org/inequality/income-inequality.htm); data on the income share of the top 1% are from the World Inequality Database. Notes: (1) the (unweighted) average tax-to-GDP ratio is 36.1% for the panel of 22 OECD countries; (2) the average (after-tax-and-transfer) Gini coefficient for the panel of 22 OECD economies is 30.9; (3) the estimated linear relationship is negative and statistically significant at 1%.

Table

	1	2	3	4	5	6	7	8	9
Australia	3.64	−4.9	11.7	5490	2.4	34	12.44	0.55	4.60
Austria	94.91	−7.1	8.5	3595	1	28	9.23	0.25	122.73
Belgium	191.94	−8.4	4	1538	0.5	27	8.49	0.38	222.02
Canada	58.61	−5.8	12.5	4692	3.7	32	14.30	0.18	73.01
Denmark	40.43	−3.9	5.9	3061	1.1	25	11.23	0.31	45.20
Finland	13.28	−4.3	2.6	1099	1.2	26	9.95	0.68	19.15
France	126.54	−9.4	5.2	1739	0	29	9.99	0.01	174.02
Italy	158.84	−9.9	4.9	1327	−0.1	33	8.82	0.31	216.11
Germany	82.44	−5.6	8.3	3334	0.9	29	12.96	−0.24	111.84
Greece	58.92	−9	6.8	1050	0.7	34	12.88	0.49	134.99
Ireland	85.24	−2.4	5.9	4152	0.3	31	11.57	1.12	104.81
Japan	5.96	−5.5	11.3	3865	0.8	33	12.44	0.75	13.64
Korea	3.05	−1.5	3.5	952	0.8	30	14.14	1.11	4.66
NL	89.21	−5.3	4.6	2063	1	28	6.97	0.13	106.77
NZ	0.53	−6.2	19.5	6855	n.a.	33	11.42	0.56	0.55
Norway	11.44	−3.6	5.4	3225	1.8	25	10.44	0.80	15.73
Portugal	155.84	−9.3	3.2	615	1.5	34	11.59	0.25	174.37
Spain	144.76	−12.4	3.5	815	2.6	35	12.20	0.24	182.22
Sweden	124.21	−3.4	5.2	2360	2	28	8.96	0.37	143.65
Switzerl.	116.33	−4	4.8	3557	0.5	30	10.91	0.67	128.28
UK	181.97	−10.3	9.2	3198	1.3	36	12.74	−0.06	202.80
USA	152.90	−4.6	11.8	6525	4.2	39	18.72	−0.38	205.27
average	83.24	−6.30	7.0	2790	1	30.5	11.13	0.42	109.38

Figure A1. Cumulative confirmed COVID-19 deaths per 100,000 population: February 23 (horizontal axis) versus September 21 (vertical axis), 2021.Sources: see Table A1. The estimated linear relationship is positive and statistically significant at 0.1%. The slope coefficient is 1.24.

Table

	Cumulative excess deaths per 100,000	Cumulative confirmed COVID-19 deaths per 100,000 (until February 23, 2021)	Difference (%)	Date cumulative excess deaths
Austria	110.3	94.91	16	Jan 31, 2021
Belgium	174.8	191.94	−9	Jan 31, 2021
Denmark	22.3	40.43	−45	Feb 7, 2021
France	94.4	126.54	−25	Jan 31, 2021
Italy	138.4	158.84	−13	Nov 25, 2020
Germany	58.7	82.44	−29	Feb 7, 2021
Korea	1.1	3.05	−62	Dec 6, 2020
NL	112.5	89.21	26	Feb 7, 2021
Norway	6.5	11.44	−43	Jan 3, 2021
Portugal	217.6	155.84	40	Feb 14, 2021
Spain	170.8	144.76	18	Jan 31, 2021
Sweden	114.0	124.21	−8	Jan 31, 2021
Switz	116.5	116.33	0	Jan 31, 2021
UK	173.9	181.97	−4	Feb 5, 2021
USA	142.3	152.90	−7	Jan 16 2021

Sources: Cumulative confirmed COVID-19 deaths per 100,000 population: Johns Hopkins University & Medicine, Coronavirus Resource Centre (https://coronavirus.jhu.edu/data/mortality); data up to February 23, 2021; cumulative excess deaths: Financial Times Coronavirus Tracker (accessed March 7, 2021); https://www.ft.com/content/a2901ce8-5eb7-4633-b89c-cbdf5b386938

Figure A2. Cumulative confirmed COVID-19 deaths per 100,000 population versus cumulative excess deaths per 100,000 population. Sources: see Table A2. The estimated linear relationship is positive and statistically significant at 1%.

Figure A3. After-tax-and-transfer Gini coefficients versus the proportion of men suffering from overweight and obesity. Sources: see Table A1. Data on the proportion of men suffering from overweight and obesity are from Global Obesity Observatory https://data.worldobesity.org/. Note: The estimated linear relationship is positive and statistically significant at 2.5% (when I exclude Japan from the regression).

IMF. 2020. Fiscal Monitor: Policies for the Recovery. October. Washington, DC: IMF.

 Google Scholar

Marmot, M. 2020a. Health Equity in England: The Marmot Review Ten Years On. Institute of Health Equity. https://www.health.org.uk/publications/reports/the-marmot-review-10-years-on.

 Google Scholar

Marmot, M., J. Allen, and P. Goldblatt. 2020b. Build Back Fairer: The COVID-19 Marmot Review. The Pandemic, Socioeconomic and Health Inequalities in England. Institute of Health Equity. http://www.instituteofhealthequity.org/resources-reports/build-back-fairer-the-COVID-19-marmot-review/build-back-fairer-the-COVID-19-marmot-review-executive-summary.pdf.

 Google Scholar

Appelbaum, E., and R. Batt. 2020. ‘Private Equity Buyouts in Healthcare: Who Wins, Who Loses?’ INET Working Paper No. 118. New York: Institute for New Economic Thinking.

 Google Scholar

Woolhandler, S., D. U. Himmelstein, S. Ahmed, Z. Bailey, M. T. Bassett, M. Bird, J. Bor, et al. 2021. ‘Public Policy and Health in the Trump era.’ The Lancet 397 (10275): 705–753.

 PubMed Web of Science ®Google Scholar