Global socio-economic losses and environmental gains 1

from the Coronavirus pandemic 2

3 Manfred Lenzen1, Mengyu Li1, Arunima Malik1,2*, Francesco Pomponi3, Ya-Yen Sun4, Thomas Wiedmann5, 4 Futu Faturay6, Jacob Fry1,7, Blanca Gallego8, Arne Geschke1, Jorge Gómez-Paredes9,10, Keiichiro Kanemoto7, 5 Steven Kenway11, Keisuke Nansai1,12, Mikhail Prokopenko13, Takako Wakiyama1, Yafei Wang14, Moslem 6 Yousefzadeh1 7 8 1Integrated Sustainability Analysis, School of Physics, The University of Sydney, NSW, Australia 9 2 Discipline of Accounting, School of Business, The University of Sydney, NSW 10 3 Resource Efficient Built Environment Lab, Edinburgh Napier University, Edinburgh UK 11 4 Business School, the University of Queensland, QLD, Australia 12 5 School of Civil and Environmental Engineering, UNSW Sydney, NSW, Australia 13 6 Fiscal Policy Agency, Ministry of Finance of the Republic of Indonesia, Jakarta, Indonesia 14 7 Research Institute for Humanity and Nature, Kyoto, Japan 15 8 Centre for Big Data Research in Health, UNSW Sydney, NSW, Australia 16 9 School of Earth Sciences, Energy and Environment, Yachay Tech University, Urcuquí, Ecuador 17 10 Nicholas School of the Environment, Duke University, Durham NC, USA 18 11 Advanced Water Management Centre, The University of Queensland, QLD, Australia 19 12 Center for Material Cycles and Waste Management Research, National Institute for Environmental Studies, 20 Tsukuba, Japan 21 13 Centre for Complex Systems, The University of Sydney, NSW, Australia 22 14 School of Statistics, Beijing Normal University, 19 Xinjiekouwai Street, Beijing, P.R.China 23 24 *Corresponding author: arunima.malik@sydney.edu.au; +61 2 93515451 25 26

27

28

29

2

Abstract 30

On 3 April 2020, the Director-General of the WHO stated: “[COVID-19] is much more than a 31

health crisis. We are all aware of the profound social and economic consequences of the 32

pandemic [1]”. Such consequences are the result of counter-measures such as lockdowns, and 33

world-wide reductions in production and consumption, amplified by cascading impacts 34

through international supply chains. Using a global multi-regional macro-economic model, we 35

capture direct and indirect spill-over effects in terms of social and economic losses, as well as 36

environmental effects of the pandemic. Based on information as of May 2020, we show that 37

global consumption losses amount to 3.8$tr, triggering significant job (147 million full-time 38

equivalent) and income (2.1$tr) losses. Global atmospheric emissions are reduced by 2.5Gt of 39

greenhouse gases, 0.6Mt of PM2.5, and 5.1Mt of SO2 and NOx. While Asia, Europe and the 40

USA have been the most directly impacted regions, and transport and tourism the immediately 41

hit sectors, the indirect effects transmitted along international supply chains are being felt 42

across the entire world economy. These ripple effects highlight the intrinsic link between socio-43

economic and environmental dimensions, and emphasise the challenge of addressing 44

unsustainable global patterns. How humanity reacts to this crisis will define the post-pandemic 45

world. 46

47

3

Introduction 48

On 9 January 2020, the World Health Organisation (WHO) first reported the outbreak of a 49

coronavirus disease (COVID-19), caused by the severe acute respiratory syndrome coronavirus 50

2 (SARS-CoV-2) [2], from the Chinese city of Wuhan. By the end of January, there were more 51

than 10,000 existing cases and more than 2,000 new confirmed cases daily, mostly in China’s 52

Hubei province, and more than 250 people had died. Initially, the spread of the virus beyond 53

China’s borders was slow, affecting a handful of nearby regions such as Japan, Hong Kong 54

and Singapore. However, by the end of February, when the number of daily new cases from 55

China decreased, infections accelerated again, this time across all continents, with major 56

outbreaks initially in South Korea and Iran, and then across Europe and the Americas (SI 1) 57

[3]. On 11 March, with more than 118,000 confirmed cases in 114 countries, and 4,291 deaths, 58

the WHO named COVID-19 a pandemic [4]. 59

Governments reacted to the outbreak by restricting people’s movements. By 3 April, with over 60

1 million confirmed cases worldwide [1], many countries implemented lockdown measures, 61

with close to 3 billion people asked to stay at home [5], more than 1 billion people alone in 62

India. These restrictions meant that people were unable to commute to their workplaces, and 63

as a result, offices and factories closed. Internationally, broad entry bans were applied, and 64

flight routes suspended. 65

Given the important role of large coronavirus-affected economies such as China, Europe and 66

the USA, in global manufacturing and trade, the slowdown in these countries’ production 67

inevitably leads to significant supply-chain interruptions, affecting especially businesses that 68

are heavily dependent on trade, such as specialised manufacturing [6] and health care supplies 69

[7]. Businesses may rely on inventories to bridge temporary supply shortfalls, generally for 70

4

two to five weeks [6], however after stocks are depleted, ensuing declines in production will 71

cascade throughout international supply-chain networks, affecting both downstream customers 72

and upstream suppliers. Options for switching to alternative inputs are limited wherever these 73

inputs are specialised and essential, such as parts for vehicles. Options for switching to 74

alternative supply locations are also limited when production is concentrated, or when the 75

output of many regions is reduced. This has been the case during the COVID-19 global 76

outbreak. It is therefore crucial to put in place preparedness measures that are aimed at 77

minimising disruptions for populations and economic losses for business [8]. 78

Apart from profound social and economic implications [1], there was however a silver lining: 79

the grounding of planes and shutdown of factories due to the implementation of travel bans 80

and lockdowns had a beneficial effect on air pollution. By March, the decline in coal use by 81

power plants, oil refining, steel manufacturing and air travel was estimated to have caused a 82

250 Mt decrease in CO2 emissions [9]. NASA and the European Space Agency reported a 83

dramatic fall in N2O pollution across North-Eastern China [10] and the lowest average level of 84

N2O ever recorded in India was a result of the nationwide curfew at the end of March [11]. The 85

quantification of these wide-ranging impacts at a global scale requires an ability to capture 86

supply-chain-driven spill-over impacts across regions and sectors for multiple indicators, 87

otherwise the assessment is incomplete [6, 12, 13]. This ability is provided by input-output 88

analysis [14, 15], more specifically global multi-region input-output (MRIO) analysis. 89

Prior work has primarily focussed on modelling the mechanisms for spread of diseases [16]. 90

Global MRIO-based supply-chain analyses of disasters exist [17], however not of a virus 91

pandemic. Preliminary input-output analyses have been carried out for estimating the possible 92

impacts of the COVID-19 outbreak within China [18] and Japan [19], both using national 93

input-output (IO) tables. A study on the economic and productivity risks resulting from the 94

5

2009 H1N1 epidemic [20] covers only a small region in the USA, and recommends examining 95

employment and income as loss categories, which is realised in this work. Initial assessments 96

based on MRIO tables and computable general equilibrium (CGE) analysis point to scenarios 97

of wide-ranging decline in global gross domestic product and consumption [21, 22]. There are 98

urgent calls to develop assessment frameworks for capturing interactions between regions and 99

sectors to enable timely quantification of actual economic impacts, such as for the COVID-19 100

pandemic [23], to help guide policy responses. The positive environmental effects of the 101

pandemic are yet to be quantified. 102

The aim of this study is to provide a comprehensive global estimate of how the 2020 COVID-103

19 pandemic in the most affected countries reduced economic activity and environmental 104

pressures in all other countries because of globalised trade links. Our study covers reduced 105

consumption, employment, income, emissions of greenhouse gases (GHG), PM2.5 and other air 106

pollutants (SO2 and NOx). Reductions in individual national economies have amplified one 107

another, leading to even larger economic losses and environmental gains at a global scale. Our 108

assessment differs from previous ones in that we are not offering scenarios, but an assessment 109

of the overall global effects based on assessments of impacts as of May 2020 (SI 4). 110

111

Materials and methods 112

Supply-chain analysis 113

The transmission of monetary or physical impacts between economic sectors and regions is 114

called spill-over. International spill-over effects have traditionally been investigated using 115

multi-region input-output (MRIO) analysis [24, 25]. This technique was conceived by Nobel 116

6

prize Laureate Wassily Leontief before WWII [26]. Since then, it has been used extensively 117

for tracing economic and environmental impact across complex supply-chain networks [27, 118

28], with high-level applications to carbon emissions, biodiversity loss, air pollution, and 119

public health [29, 30]. Input-output databases are regularly collected by more than 100 national 120

statistical agencies worldwide, all governed by United Nations standards [31]. The centrepiece 121

of global MRIO data is an NN intermediate demand matrix (T), mapping the connections 122

between all industry sectors in the global economy. Contemporary MRIO databases distinguish 123

in excess of 𝑁 ≥ 10,000 industries and products [32]. Seeing that industries can either supply 124

other industries or final consumers, summing intermediate and final demand (y; for example 125

households) yields total output 𝐱 = 𝐓𝟏 + 𝐲, where the vector 𝟏 = {1,1, … ,1} is a summation 126

operator. Defining a technical coefficient matrix 𝐀 ≔ 𝐓�̂�−1 , with the hat symbol denoting 127

vector diagonalization, lets us derive Leontief’s fundamental accounting identity 𝐱 = 𝐓�̂�−1 +128

𝐲 ⇔ 𝐱 = (𝐈 − 𝐀)−1𝐲, where I is an identity matrix [28]. The term (𝐈 − 𝐀)−1 is Leontief’s 129

famous inverse, providing information on the complex links between geographically distant 130

producers and consumers. 131

Disaster analysis 132

One particular strand of MRIO analysis concerned with the impacts of shocks to the economy 133

is disaster analysis [33]. Such type of analysis considers the direct and indirect supply-chain 134

effects of disasters resulting in loss of production and reduced business activity [34]. Several 135

variations of the input-output framework have been developed over the years for quantifying 136

disruptions ranging from natural events [35], extreme weather events [36] to terrorism [37] and 137

blackouts [38]. The variations have focussed on enhancing the capability of standard input-138

output analysis in capturing temporal frames, such as in the regional econometric input-output 139

model [39], and for capturing recovery times in a post-disaster world as in the inoperability 140

7

input-output model (IIM) [40-42], and its uses for quantifying negative economic impacts [43]. 141

A relatively new technique of hypothetical extraction (HEM) relies on assessing hypothetical 142

scenarios where industries cease to operate [44], as applied for the case of shutdown of IT 143

services in the UK [45]. Since in this work we are concerned with the quantification of actual 144

impacts of a global pandemic on people’s livelihoods, we build on a disaster analysis method 145

that focuses on post-disaster consumption possibilities [46]. This method uses a so-called event 146

matrix , where diagonal elements ii describe the relative loss of industries i=1,…,N as a 147

direct result of a disaster. Here, we minimise the departure (�̃� − 𝐱)2 of post-disaster output �̃� 148

from pre-disaster output 𝐱, subject to two conditions [47]: First, we ask that �̃� ≤ (𝐈 − 𝚪)𝐱. 149

Second, we require that post-disaster final demand �̃� = (𝐈 − 𝐀)�̃� ≥ min(0, 𝐲St). Here, 𝐲St ≤150

0 holds information on stocks, on which industries may draw for continuing sales despite their 151

production downturn. The condition therefore says that final demand may not be negative for 152

industries that do not hold any stocks. For those who do hold stocks, we ask that losses may 153

not drive final demand so far down as to exceed the value of these stocks. The solution �̃� of 154

this optimisation problem are the post-disaster consumption possibilities. We then truncate �̃� 155

at 𝐲 , using min(�̃�, 𝐲) , and thus count only consumption losses and not the relatively 156

insignificant increases in final consumption possibilities that come about through the decrease 157

of intermediate demand [46]. This is because we assume that increased availability of 158

commodities will not necessarily translate into increased demand, because that demand has not 159

existed in the pre-disaster economy. In contrast to , post-disaster consumption possibilities 160

include all spill-over effects, meaning that people in regions not directly hit by the disaster may 161

see their consumption curtailed because of supply-chain relationships. The optimisation was 162

carried out using the quadprog function in Matlab, which calls an interior-point algorithm. 163

Economic and environmental impacts 164

8

Reduced post-disaster consumption translates into reduced employment and family incomes, 165

but also into reduced emissions of greenhouse gases and air pollutants. Whilst some of the 166

direct environmental consequences of the COVID-19 outbreak could be measured [9], 167

quantifying the combined effects of all regional and sectoral spill-over cascades requires an 168

MRIO analysis. Leontief already proposed an extension of his approach to include physical 169

quantities [48]. This is accomplished by coupling the fundamental accounting identity with a 170

so-called satellite account (Q), holding data on physical quantities for every of the N industries 171

in the intermediate demand matrix T. Following Leontief’s calculus, the economic and 172

environmental impacts F of the disaster can be computed from consumption losses – the 173

difference of pre-and post-disaster consumption possibilities �̃� − 𝐲 – as 𝐹 = 𝐐�̂�−1(𝐈 −174

𝐀)−1(�̃� − 𝐲)[49]. We carry out an uncertainty analysis [49] to accompany our results with 175

reliability estimates (SI 5). 176

Data sources 177

Meaningful application of MRIO analysis to the spill-over effects of a global pandemic 178

requires providing rapid information on regions and sectors that will likely be affected. 179

However, many official input-output and national accounts are published with a multi-year 180

delay. We therefore apply a recent innovation to compiling MRIO accounts: a virtual 181

laboratory [50]. By employing collaborative cloud-computing environments, virtual MRIO 182

laboratories have led to significant gains in research efficiencies and timely data provision. 183

MRIO labs are currently operating in six countries [51]. Here we use the Global MRIO Lab 184

[52] for compiling a tailored global MRIO data set, distinguishing 38 regions with 26 sectors 185

each (SI 2). We use the most recent global data sources, for the monetary variables T and y 186

from the United Nations, and for the satellite Q from the EU’s Joint Research Centre for 187

emissions of greenhouse gases, PM2.5 and air pollutants, and from the International Labour 188

9

Organisation for employment (SI 3). To populate the event matrix 𝚪 for the COVID-19 189

pandemic, we carry out an extensive data collection exercise, including recent online accounts 190

of industry losses across the world (SI 4). This strategy, combined with the utilisation of the 191

Global MRIO Lab, allows us to rapidly respond to the outbreak with information on the likely 192

magnitude and spread of the economic and environmental repercussions. 193

194

Results 195

Our approach uses constrained non-linear optimisation (SI 5) to determine the maximum level 196

of global consumption that is possible under given (exogenous) reductions of economic output 197

by taking into account the effects of industry shutdowns, lockdowns and travel restrictions. 198

The difference between this maximum level and the pre-COVID-19 world economy are 199

consumption losses. In turn, these consumption losses set in motion supply-chain effects that 200

ripple across the world economy, and lead to global losses of income, employment, and 201

reductions of emissions. 202

On the basis of reported direct losses experienced by global businesses due to the COVID-19 203

pandemic, we estimate the total consumption loss, including all regional and sectoral spill-204

overs, to be about 3.8$tr, or 4.2% of global GDP, and comparable to the GDP of Germany. 205

This figure carries an uncertainty (due to stochastic errors in underlying data based on media 206

reports, published academic articles, official publications and expert opinion) of ≈ 10%, 207

however systematic uncertainties mean that it is likely an underestimate (SI 5). Similarly, jobs 208

of workers globally reduce by 147 million FTE (full-time equivalent), or 4.2% of the global 209

workforce, with associated losses of wages and salaries of 2.1$tr, or 6.0% of global income. 210

Global emissions of GHGs, PM2.5, and air pollutants also reduce, by 2.5 Gt, 0.6 Mt and 5.1 Mt, 211

10

or 4.6%, 3.8% and 2.9% of the global annual totals, respectively. Reductions in GHG emissions 212

are larger than any drop in anthropogenic emissions in human history (SI 6.5, Fig SI11), 213

including when fossil-fuel CO2 emissions dropped by 0.46 Gt CO2 in 2009 due to the global 214

financial crisis (GFC) [53], and when CO2 emissions from land use change dropped by 2.02 215

Gt CO2 in 1998 [53] . 216

Spill-over effects are significant: our production layer decompositions (SI 6) show that direct 217

impacts are magnified about 1.5-fold, as a consequence of the immediate impacts rippling 218

through the global supply-chain network and causing widespread indirect impacts. In the 219

following, we will unravel the overall results given above, and provide insights for individual 220

regions and industry sectors. 221

The global figures listed above can be broken down into estimates for the 38 regions 222

distinguished in our study (SI 2). Clearly, the more significant consumption and income losses 223

are borne out in large economies with either high numbers of coronavirus cases and/or stringent 224

countermeasures, i.e. China, USA, Italy, Spain, Germany, UK and France (Fig. 1). OPEC 225

nations lose income because of reduced oil extraction and refining activity as a result of the 226

reductions in transport, especially aviation. Low-wage countries such as China and India stand 227

out in terms of employment losses. Reductions in GHG emissions occur all over the globe, but 228

mainly across China and Northern America, and less across Europe’s less emissions-intensive 229

economies. Reductions of PM2.5 emissions are expectedly large in China and India, whilst 230

energy- and transport-related SO2 and NOx emissions also fall across the rest of Asia and the 231

Americas (Fig. 2). The results presented in this study are based on current information gathered 232

from various sources (SI 4), yet are broadly comparable with estimates produced by 233

international organisations (SI 7). 234

11

Fig. 1: Global impacts from the COVID-19 pandemic broken down by world region. 235 Accompanying data tables are in SI 6.1. 236

Fig. 2: Sectoral breakdown of global impacts from the COVID-19 pandemic, in indicator-237 specific units (US$bn for consumption and income, million FTE for employment, Mt for 238 greenhouse gas emissions, and kt for other emissions). The bands represent direct and 239 indirect impacts by purchased commodity. For example, the Yellow band refers to final 240 demand purchases of electricity, gas and water; however, utilities’ losses and reductions in 241 income, jobs and emissions are also included in the supply chain of other commodities, such 242 as Manufacturing (Blue). Accompanying data tables are in SI 6.2. 243

244

An assessment of total impacts of COVID-19 at a sector-level reveals that transport and tourism 245

are the economically worst-hit sectors (Fig. 2). This is unsurprising, with falling air travel 246

demands as people made cancellations, coupled with a suite of travel restrictions imposed by 247

countries worldwide to slow the spread of the virus, and airlines going bankrupt [54]. The 248

International Air Transport Association (IATA) estimated that global revenues could fall more 249

than 44% below 2019 figures [55] (SI 4.4). Significantly affected by lockdowns are retail and 250

wholesale, as well as service sectors, including business services in the supply chains of 251

tourism and transport, and entertainment and personal services. Manufacturing operations are 252

reduced in China, Europe, and across the OPEC, mining in Australia [56] (ores and gas) and 253

the OPEC (oil), because of supply-chain knock-on effects. 254

Manufacturing and Transport & tourism dominate reductions in GHG and SO2 emissions, 255

because of their intensive use of fuels (Fig. 2). A large part of reductions in PM2.5 emissions 256

globally is driven by reduced power, gas and water utilities output in Asia and the Americas. 257

It has been previously shown that developed nations outsource PM2.5-related impacts to Asia 258

[57]. The fact that utilities are not represented more uniformly across regions is either due to 259

reductions not having occurred, or missing information. This once more underlies that our 260

results are likely underestimates (SI 5). 261

12

Out of the total income losses of $2.1tr, $536bn or about 21% is lost because of a reduction in 262

international trade, demonstrating the importance of international spill-overs that cause the 263

effects of the COVID-19 pandemic to be felt in all countries across the globe (Fig. 3). Given 264

the reliance of many national economies on China, we observe significant losses in supply 265

chains that originate in Mainland China. The pandemic has exposed vulnerable businesses 266

whose supply chains are heavily concentrated in countries that are most directly impacted by 267

the crisis [58]. Countries that have significant trade relationships with the most coronavirus-268

affected countries also experience emission reductions (SI 6). 269

Fig. 3: Wage and salary income losses as a consequence of trade volume reductions in 270 international supply chains due to the global COVID-19 effects. Lines connect ultimate 271 origins and destinations of supply chains, both direct and multi-node. Line thickness represents 272 trade volume lost. 273

274

13

Discussion 275

276

The core principles underlying the mitigation and suppression strategies adopted to control the 277

pandemic – distancing of individuals and quarantining of communities – work directly against 278

the strengths of the global economy, built around connectivity and inter-dependence. Ironically, 279

the loss of connectivity imposed to prevent the spread of the COVID-19 has triggered an 280

economic “contagion”, with the crisis precipitated by the COVID-19 pandemic cascading 281

across socio-economic sectors, and causing major disruptions to trade, tourism, energy and 282

finance sectors. 283

284

The global economy is heading towards a recession “way worse than the GFC” [59]; global 285

consumption loss thus far, including both direct and indirect spill-over effects is significant. 286

The strongest effects are felt in China, other major Asian economies, Europe, and USA, and 287

largely in the transport and tourism sector, yet with world-wide ripple effects. These losses are 288

likely to increase and spread to the rest of the global economy, as lockdown measures continue. 289

Yet lifting restrictions too soon could result in more severe and prolonged economic impacts 290

[4]. Thus, governments are faced with the challenge of attempting to keep the global economy 291

afloat by spending the International Monetary Funds’ war chest [59] and other emergency 292

funds, while trying to find new ways of working (e.g. extending teleworking) [60]. Global 293

economic measures taken to address the economic impacts of COVID-19 have been rapidly 294

developed. To-date, most measures have aimed at broad-brush support for “business”, e.g. the 295

US, UK and Australian governments committing significant support for businesses and 296

households. Other measures include (a) additional bank lending (b) health system 297

improvements (largely intensive care facilities), support to State and Local government account 298

for a large percentage of packages. Measures which directly address impacts to sectors 299

14

impacted in Fig. 2 (e.g. Transport & Tourism, Trade and Services) are less explicit. It appears 300

that governments’ primary approach is to put a safety net under the economy, rather than to 301

promote those forms of business either most affected, or most able to grow and develop in the 302

anticipated post-COVID-19 economic landscape. 303

304

Socially, the pandemic has resulted in significant labour market shocks [60], which are poised 305

to grow as the pandemic persists. Moreover, subsequent economic shocks are likely to impact 306

even further the quantity and quality of jobs, as well as affect vulnerable groups [60], such as 307

migrant and unskilled workers who may not adapt to virtual-work arrangements. 308

309

Environmentally, the pandemic has brought a reduction in greenhouse gas and air pollution, 310

mainly from a fall in fossil fuel consumption as airplanes are grounded, transportation reduced, 311

trade hindered, and factories closed down. These bring important environmental gains as well 312

as social benefits. Reductions in PM2.5 alone are likely to save thousands of lives. The current 313

drop in GHG emissions is larger than anything the world has seen since humans started to use 314

fossil fuels (our estimates until the end of May 2020 are at the higher end of estimates of global 315

CO2 emission reductions projected until the end of 2020 by one other study [61]). None of the 316

attempts by any government or any international agreement in the 32-year history of 317

intergovernmental climate policy has had such a dramatic mitigation effect. Incidentally, the 318

drop of approximately 4.5% in global GHG emissions caused by reactions to COVID-19 still 319

falls short of what would be needed every year until 2050 to limit global warming to 1.5ºC (Fig 320

SI13). 321

322

Clearly, the social and environmental consequences of fighting the pandemic will be much 323

broader than negative impacts on jobs and income, and positive impacts on atmospheric 324

15

pollution (e.g. fighting the pandemic may bring about mass surveillance [62, 63], or negative 325

environmental impacts from continuous cleaning and disinfecting activities [64]). However, 326

the contrast between the socio-economic and the environmental variables that we have assessed 327

in this study reveals the dilemma of the global socio-economic system. 328

329

The short-term economic, social, and environmental impacts of the COVID-19 pandemic are 330

profound; and pose several challenges and dilemmas. The current crisis is likely to deepen 331

systemic socioeconomic vulnerabilities, widen income and wealth gaps, overburden, if not 332

decimate, healthcare systems in lower income countries [65], and proliferate the spread of 333

emerging zoonotic diseases [66]. When such systemic vulnerabilities are coupled with 334

significant climatic variations, their combined effects may lead to tipping points in socio-335

economic systems, e.g. via food systems failure and large-scale urban abandonment [67]. 336

337

To this point, it is worth noting that the COVID-19 crisis takes place in the shadow of other 338

more “silent”, yet much longer-lasting, ongoing global crises, such as climate change. The GFC 339

led to only a small dent in the continued upward trajectory of global GHG emissions [68]. 340

While the COVID-19 crisis will leave a larger impression, it will still not be enough to avoid 341

dangerous climate change, and may quickly be erased as we attempt to go back to business-as-342

usual and give way to “retaliatory pollution” [69]. This could potentially be averted by 343

implementing stimulus plans to boost clean energy technologies and facilitate ‘just transitions’, 344

to encourage investments in teleworking and teleconferences for reducing carbon-intensive 345

travel, and to devise policies for addressing rebound effects. Yet, it remains to be seen if such 346

an approach would fare differently than the G20 countries’ green economic stimulus after the 347

GFC [70], and if countries eager to get back to economic growth will avoid basing it on low-348

priced oil. For business-as-usual economic growth is still strongly coupled with the use of fossil 349

16

fuels and will therefore undermine any environmental gains. In contrast, the current 350

environmental improvements mean social and economic hardship since the unfolding global 351

recession is unplanned and unmanaged. Scientists have long argued that any deliberate 352

economic ‘de-growth’ intended to prevent global ecological crises needs to be well managed 353

and based on targeted investment and tax policies [71, 72][71]. 354

355

On one hand the pandemic has shown the risks and fragility of our highly interconnected and 356

interdependent economies and societies, which highlights the need for global cooperation and 357

solidarity, as no country will be “immune” to situations developing elsewhere. On the other 358

hand, it has shown that we can face crises with concerted and decisive interventions and 359

changes in behaviours and lifestyles, which can lead to significant environmental benefits 360

while protecting people's livelihoods at the same time [71]. Both are key factors to address the 361

wider environmental crises. 362

363

Future work could focus on examining the impact of lasting changes in behaviour and lifestyles 364

induced by the pandemic on wider global issues, such as greenhouse gas and air pollutant 365

emissions, resource use or biodiversity decline. Unequivocally, the COVID-19 pandemic has 366

had varying levels of effects on economy and environment of low-, middle- and high-income 367

countries, and those of islands. The future of the global ecosystem hinges on regional policy 368

decisions that will define the post-pandemic world. Government policies during the pandemic 369

have been reported to have caused a 17% drop in daily global emissions in April 2020, 370

compared to the 2019 average, and up to 20% in some countries [70][61]. Much of these 371

policies are temporary measures for slowing down the spread of the virus, which have had a 372

positive impact on environment and air quality. Future research focussed on analysing the post-373

pandemic approach of countries to transition to greener economies will shed light on the likely 374

Field Code Changed

17

challenges faced in ensuring such emission reductions are maintained or even increased with 375

the aim of meeting the Paris goals [73][72] and are economically viable and socially just. 376

The future of the global ecosystem hinges on regional policy decisions that will define the post-377

pandemic world. A sustainable post-pandemic world will be defined by the collective action 378

of all societal groups to seize this opportunity to ensure a transition to green economies instead 379

of a ‘business-as-usual’ pathway. 380

As indicated by the UN Secretary General [74][73], humanity faces a choice: attempting to 381

return to a business-as-usual path with more unnecessary crises, or developing a different 382

economy that is compatible with more sustainable and resilient human societies. Clearly, the 383

decision we take now and after the crisis will define our post-pandemic world. 384

385

386

387 388 389

390

Field Code Changed

Field Code Changed

18

References 391

1. WHO. WHO Director-General's opening remaks at the media briefing on COVID-19 - 392 3 April 2020 2020 [cited 2020 4 April]. Available from: 393 https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-394 media-briefing-on-covid-19--3-april-2020. 395 2. WHO. Naming the coronavirus disease (COVID-19) and the virus that causes it 2020 396 [cited 2020 1 April]. Available from: https://www.who.int/emergencies/diseases/novel-397 coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-398 the-virus-that-causes-it. 399 3. WHO. Coronavirus disease (COVID-2019) situation reports. Geneva, Switzerland: 400 World Health Organization, 2020. 401 4. WHO. WHO Director-General's opening remaks at the media briefing on COVID-19 - 402 11 March 2020 2020 [cited 2020 6 April]. Available from: 403 https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-404 media-briefing-on-covid-19---11-march-2020. 405 5. Business Insider. A third of the global population is on coronavirus lockdown - here's 406 our constantly updated list of countries and restrictions 2020 [cited 2020 6 April]. Available 407 from: https://www.businessinsider.com/countries-on-lockdown-coronavirus-italy-2020-408 3?r=DE&IR=T. 409 6. Haren P, Simchi-Levi D. How coronavirus could impact the global supply chain by 410 mid-March. Harvard Business Review. 2020;(28 Feb). 411 7. FDA. Coronavirus (Covid-19) Supply Chain Update. Silver Spring, USA: U.S. Food & 412 Drug Administration, 2020. 413 8. Reimold K, Geddes K. Global monitoring of disease outbreak preparedness. 414 Cambridge, USA: Harvard Global Health Institute, 2018. 415 9. Myllyvirta L. Coronavirus has temporarily reduced China’s CO2 emissions by a 416 quarter. Centre for Research on Energy and Clean Air and CarbonBrief, based on data in 417 WIND Information, 2020. 418 10. NASA. Airborne Nitrogen Dioxide Plummets Over China. earth observatory. 419 Washington D.C., USA: North American Space Agency, 2020 420 https://earthobservatory.nasa.gov/images/146362/airborne-nitrogen-dioxide-plummets-421 over-china. 422 11. Myllyvirta L, Dahiya S. India’s Coronavirus Curfew Resulted in the Lowest One-Day 423 Traffic Pollution Levels on Record. CREA. Centre for Research on Energy and Clean Air, 2020 424 https://energyandcleanair.org/janata-curfew-pollution-levels/. 425 12. Rice JB. Prepare your supply chain for Coronavirus. Harvard Business Review. 426 2020;(27 Feb). 427 13. GSGIR. The new coronavirus could have a lasting impact on global supply chains. The 428 Economist. 2020;(15 Feb). 429 14. Zhang L, Hu Q, Zhang F. Input-output modeling for urban energy consumption in 430 Beijing: dynamics and comparison. PloS one. 2014;9(3). 431 15. Eckelman MJ, Sherman J. Environmental impacts of the US health care system and 432 effects on public health. PloS one. 2016;11(6):e0157014. 433 16. Rivers C, Chretien J-P, Riley S, Pavlin JA, Woodward A, Brett-Major D, et al. Using 434 “outbreak science” to strengthen the use of models during epidemics. Nature 435 Communications. 2019;10(1):3102. doi: 10.1038/s41467-019-11067-2. 436

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

19

17. Koks E, Rozenberg J, Zorn C, Tariverdi M, Vousdoukas M, Fraser S, et al. A global 437 multi-hazard risk analysis of road and railway infrastructure assets. Nature Communications. 438 2019;10. doi: 10.1038/s41467-019-10442-3. 439 18. Duan H, Wang S, Yang C. Coronavirus: limit short-term economic damage. Nature. 440 2020;578:515. 441 19. Inoue H, Todo Y. The Propagation of the Economic Impact through Supply Chains: 442 The Case of a Mega-City Lockdown against the Spread of COVID-19. SSRN. 443 2020:http://dx.doi.org/10.2139/ssrn.3564898. 444 20. Santos JR, May L, Haimar AE. Risk-based input-output analysis of influenza epidemic 445 consequences on interdependent workforce sectors. Risk analysis : an official publication of 446 the Society for Risk Analysis. 2013;33(9):1620-35. Epub 12/24. doi: 10.1111/risa.12002. 447 PubMed PMID: 23278756. 448 21. pwc. The possible economic consequences of a novel coronavirus (COVID-19) 449 pandemic 2020 [cited 2020 1 April]. Available from: 450 https://www.pwc.com.au/publications/australia-matters/economic-consequences-451 coronavirus-COVID-19-pandemic.pdf. 452 22. McKibbin WJ, Fernando R. The global macroeconomic impacts of covid-19: Seven 453 scenarios. CAMA Working Paper, Technical Report Canberra, Australia: CAMA, Australian 454 National University, 2020 19/2020. 455 23. Yu KDS, Aviso KB. Modelling the Economic Impact and Ripple Effects of Disease 456 Outbreaks. Process Integration and Optimization for Sustainability. 2020:1-4. 457 24. Auer R, Levchenko AA, Sauré P. International inflation spillovers through input-458 output linkages. BIS Working Paper. Basel, Switzerland: Bank for International Settlements, 459 2017 No 623. 460 25. Alsamawi A, Murray J, Lenzen M, Kanemoto K, Moran D. A novel approach to 461 quantitative accounting of income inequality. PLoS ONE. 2014;9(10):e110881. 462 26. Leontief W. Quantitative input and output relations in the economic system of the 463 United States. Review of Economics and Statistics. 1936;18(3):105-25. 464 27. Dietzenbacher E, Lenzen M, Los B, Guan D, Lahr ML, Sancho F, et al. Input–output 465 analysis: The next 25 years. Economic Systems Research. 2013;25(4):369-89. doi: 466 10.1080/09535314.2013.846902. 467 28. Miller RE, Blair PD. Input-Output Analysis: Foundations and Extensions. Englewood 468 Cliffs, NJ, USA: Prentice-Hall; 2010. 469 29. Wiedmann T, Lenzen M. Environmental and social footprints of international trade. 470 Nature Geoscience. 2018:in press. 471 30. Malik A, Lenzen M, McAlister S, McGain F. The carbon footprint of Australian health 472 care. The Lancet Planetary Health. 2018;2(1):e27-e35. doi: 10.1016/S2542-5196(17)30180-473 8. 474 31. UNSD. System of Environmental- Economic Accounting— SEEA Revision. Internet 475 site. United Nations, 2017 https://unstats.un.org/unsd/envaccounting/seearev/. 476 32. Tukker A, Dietzenbacher E. Global multiregional input-output frameworks: An 477 introduction and outlook. Economic Systems Research. 2013;25(1):1-19. doi: 478 10.1080/09535314.2012.761179. 479 33. Okuyama Y, Santos JR. Disaster impact and input-output analysis. Economic Systems 480 Research. 2014;26(1):1-12. doi: 10.1080/09535314.2013.871505. 481

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

20

34. Rose A. Economic principles, issues, and research priorities in hazard loss estimation, 482 In: Y. Okuyama and S.E. Chang (Eds) Modeling Spatial and Economic Impacts of Disasters. 483 New York: Springer2004. 484 35. Faturay F, Sun Y-Y, Dietzenbacher E, Malik A, Geschke A, Lenzen M. Using virtual 485 laboratories for disaster analysis–a case study of Taiwan. Economic Systems Research. 486 2019:1-26. 487 36. Schulte in den Bäumen H, Többen J, Lenzen M. Labour forced impacts and 488 production losses due to the 2013 flood in Germany. Journal of Hydrology. 2015;527:142-489 50. 490 37. Lian C, Haimes YY. Managing the risk of terrorism to interdependent infrastructure 491 systems through the dynamic inoperability input–output model. Systems Engineering. 492 2006;9(3):241-58. 493 38. Anderson CW, Santos JR, Haimes YY. A risk-based input–output methodology for 494 measuring the effects of the August 2003 northeast blackout. Economic Systems Research. 495 2007;19(2):183-204. 496 39. Donaghy KP, Balta-Ozkan N, Hewings GJ. Modeling unexpected events in temporally 497 disaggregated econometric input–output models of regional economies. Economic Systems 498 Research. 2007;19(2):125-45. 499 40. Haimes YY, Horowitz BM, Lambert JH, Santos JR, Lian C, Crowther KG. Inoperability 500 input-output model for interdependent infrastructure sectors. I: Theory and methodology. 501 Journal of Infrastructure Systems. 2005;11(2):67-79. 502

41. Santos JR, Haimes YY. Modeling the Demand Reduction Input‐Output (I‐O) 503

Inoperability Due to Terrorism of Interconnected Infrastructures. Risk Analysis. 504 2004;24(6):1437-51. 505 42. Dietzenbacher E, Miller RE. Reflections on the inoperability input–output model. 506 Economic Systems Research. 2015;27(4):478-86. 507 43. Oosterhaven J. On the limited usability of the inoperability IO model. Economic 508 Systems Research. 2017;29(3):452-61. doi: 10.1080/09535314.2017.1301395. 509 44. Dietzenbacher E, van Burken B, Kondo Y. Hypothetical extractions from a global 510 perspective. Economic Systems Research. 2019;31(4):505-19. doi: 511 10.1080/09535314.2018.1564135. 512 45. Xia Y, Guan D, Steenge AE, Dietzenbacher E, Meng J, Mendoza Tinoco D. Assessing 513 the economic impacts of IT service shutdown during the York flood of 2015 in the UK. 514 Proceedings of the Royal Society A. 2019;475(2224):20180871. 515 46. Steenge AE, Bočkarjova M. Thinking about imbalances in post-catastrophe 516 economies: An input–output based proposition. Economic Systems Research. 517 2007;19(2):205-23. doi: 10.1080/09535310701330308. 518 47. Faturay F, Sun Y-Y, Dietzenbacher E, Malik A, Geschke A, Lenzen M. Using Virtual 519 Laboratories for disaster analysis – A case study of Taiwan. Economic Systems Research. 520 2020;32(1):58-83. 521 48. Leontief W, Ford D. Environmental repercussions and the economic structure: an 522 input-output approach. Review of Economics and Statistics. 1970;52(3):262-71. 523 49. Lenzen M, Wood R, Wiedmann T. Uncertainty analysis for Multi-Region Input-Output 524 models – a case study of the UK’s carbon footprint. Economic Systems Research. 525 2010;22(1):43-63. 526 50. Lenzen M, Geschke A, Wiedmann T, Lane J, Anderson N, Baynes T, et al. Compiling 527 and using input–output frameworks through collaborative virtual laboratories. Science of 528

21

The Total Environment. 2014;485–486:241-51. doi: 529 http://dx.doi.org/10.1016/j.scitotenv.2014.03.062. 530 51. Geschke A, Hadjikakou M. Virtual laboratories and MRIO analysis – an introduction. 531 Economic Systems Research. 2017;29(2):143-57. doi: 10.1080/09535314.2017.1318828. 532 52. Lenzen M, Geschke A, Abd Rahman MD, Xiao Y, Fry J, Reyes R, et al. The Global MRIO 533 Lab - charting the world economy. Economic Systems Research. 2017;29(2):158-86. 534 53. Friedlingstein P, Jones MW, O'Sullivan M, Andrew RM, Hauck J, Peters GP, et al. 535 Global carbon budget 2019. Earth System Science Data. 2019;11(4). 536 54. Business Insider. Many of the world's airlines could be bankrupt by May because of 537 the COVID-19 crisis, according to an aviation consultancy. These airlines have already 538 collapsed because of the pandemic. 2020 [cited 2020 1 April]. Available from: 539 https://www.businessinsider.com/coronavirus-airlines-that-failed-bankrupt-covid19-540 pandemic-2020-3?r=AU&IR=T. 541 55. IATA. Deeper Revenue Hit from COVID-19 2020 [cited 2020 1 April]. Available from: 542 https://www.iata.org/en/pressroom/pr/2020-03-24-01/. 543 56. Claughton D, Fowler C, Fitzgerald D. Mining exploration and service companies hit by 544 coronavirus restrictions 2020. Available from: https://www.abc.net.au/news/rural/2020-04-545 07/covid19-mine-explorers-and-service-sector/12125518. 546 57. Liang S, Stylianou KS, Jolliet O, Supekar S, Qu S, Skerlos SJ, et al. Consumption-based 547 human health impacts of primary PM2. 5: The hidden burden of international trade. J Clean 548 Prod. 2017;167:133-9. 549 58. Thomas L. Coronavirus wreaks havoc on retail supply chains globally, even as China’s 550 factories come back online. CNBC, 2020 https://www.cnbc.com/2020/03/16/coronavirus-551 wreaks-havoc-on-retail-supply-chains-globally.html. 552 59. International Monetary Fund. Transcript of Kristalina Georgieva's Participation in the 553 World Health Organization Press Briefing 2020 [cited 6 April]. Available from: 554 https://www.imf.org/en/News/Articles/2020/04/03/tr040320-transcript-kristalina-555 georgieva-participation-world-health-organization-press-briefing. 556 60. ILO. COVID-19 and world of work: impacts and responses 2020 [cited 2020 5 April]. 557 Available from: https://www.ilo.org/wcmsp5/groups/public/---dgreports/---558 dcomm/documents/briefingnote/wcms_738753.pdf. 559 61. Le Quéré C, Jackson RB, Jones MW, Smith AJ, Abernethy S, Andrew RM, et al. 560 Temporary reduction in daily global CO 2 emissions during the COVID-19 forced 561 confinement. Nature Climate Change. 2020:1-7. 562 62. Mozur P, Zhong R, Krolik A. In Coronavirus Fight, China Gives Citizens a Color Code, 563 With Red Flags: The New York Times; 2020 [cited 2020 6 April]. Available from: 564 https://www.nytimes.com/2020/03/01/business/china-coronavirus-surveillance.html. 565 63. Harari YN. The world after coronavirus: Financial Times; 2020 [cited 2020 6 April]. 566 Available from: https://www.ft.com/content/19d90308-6858-11ea-a3c9-1fe6fedcca75. 567 64. Mahbubani R. More than 100 wild animals in China died from poisoning in a mass 568 die-off seemingly triggered by coronavirus disinfectant.: Business Insider; 2020 [cited 2020 6 569 April]. Available from: https://www.businessinsider.com/disinfectant-control-coronavirus-570 triggers-mass-die-off-animals-china-2020-2. 571 65. Walker PG, Whittaker C, Watson O, Baguelin M, Ainslie K, Bhatia S, et al. The Global 572 Impact of COVID-19 and Strategies for Mitigation and Suppression. On behalf of the imperial 573 college covid-19 response team, Imperial College of London. 2020. 574

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted ...

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

22

66. Salata C, Calistri A, Parolin C, Palù G. Coronaviruses: a paradigm of new emerging 575 zoonotic diseases. Pathogens and Disease. 2019;77(9):ftaa006. 576 67. Scheffer M, Bascompte J, Brock WA, Brovkin V, Carpenter SR, Dakos V, et al. Early-577 warning signals for critical transitions. Nature. 2009;461(7260):53-9. 578 68. Peters GP, Marland G, Le Quéré C, Boden T, Canadell JG, Raupach MR. Rapid growth 579 in CO2 emissions after the 2008–2009 global financial crisis. Nature Climate Change. 580 2012;2(1):2-4. doi: 10.1038/nclimate1332. 581 69. Plume B, Popovich N, Lawal S. The Coronavirus and Carbon Emissions: The New York 582 Times; 2020 [cited 2020 6 April]. Available from: 583 https://www.nytimes.com/2020/02/26/climate/nyt-climate-newsletter-coronavirus.html. 584 70. Barbier E. How is the global green new deal going? Nature. 2010;464(7290):832-3. 585 doi: https://doi.org/10.1038/464832a. 586 71. D’Alessandro S, Cieplinski A, Distefano T, Dittmer K. Feasible alternatives to green 587 growth. Nature Sustainability. 2020:1-7. doi: https://doi.org/10.1038/s41893-020-0484-y. 588 72. Wiedmann T, Lenzen M, Keyßer LT, Steinberger JK. Scientists’ warning on affluence. 589 Nature communications. 2020;11(1):1-10. 590 73. IPCC. Special Report: Global Warming of 1.5 ºC. Summary for Policy Makers: 591 Intergovernmental Panel on Climate Change; 2018. Available from: 592 https://www.ipcc.ch/sr15/. 593 74. UN Secretary-General. ‘We Are Only as Strong as the Weakest’, Secretary-General 594 Stresses, at Launch of Economic Report on COVID-19 Pandemic 2020 [cited 2020 6 April]. 595 Available from: https://reliefweb.int/report/world/we-are-only-strong-weakest-secretary-596 general-stresses-launch-economic-report-covid-19. 597 598 1. WHO. WHO Director-General's opening remaks at the media briefing on COVID-19 - 599 3 April 2020 2020 [cited 2020 4 April]. Available from: 600 https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-601 media-briefing-on-covid-19--3-april-2020. 602 2. WHO. Naming the coronavirus disease (COVID-19) and the virus that causes it 2020 603 [cited 2020 1 April]. Available from: https://www.who.int/emergencies/diseases/novel-604 coronavirus-2019/technical-guidance/naming-the-coronavirus-disease-(covid-2019)-and-605 the-virus-that-causes-it. 606 3. WHO. Coronavirus disease (COVID-2019) situation reports. Geneva, Switzerland: 607 World Health Organization, 2020. 608 4. WHO. WHO Director-General's opening remaks at the media briefing on COVID-19 - 609 11 March 2020 2020 [cited 2020 6 April]. Available from: 610 https://www.who.int/dg/speeches/detail/who-director-general-s-opening-remarks-at-the-611 media-briefing-on-covid-19---11-march-2020. 612 5. Business Insider. A third of the global population is on coronavirus lockdown - here's 613 our constantly updated list of countries and restrictions 2020 [cited 2020 6 April]. Available 614 from: https://www.businessinsider.com/countries-on-lockdown-coronavirus-italy-2020-615 3?r=DE&IR=T. 616 6. Haren P, Simchi-Levi D. How coronavirus could impact the global supply chain by 617 mid-March. Harvard Business Review. 2020;(28 Feb). 618 7. FDA. Coronavirus (Covid-19) Supply Chain Update. Silver Spring, USA: U.S. Food & 619 Drug Administration, 2020. 620

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Hyperlink, Font: (Default) Calibri, Font color:

Auto, (Asian) Japanese, (Other) English (United

Kingdom), Check spelling and grammar

Formatted: Font: (Default) Calibri, Font color: Auto,

(Asian) Japanese, (Other) English (United Kingdom)

Formatted: Left, Space After: 0 pt

23

8. Reimold K, Geddes K. Global monitoring of disease outbreak preparedness. 621 Cambridge, USA: Harvard Global Health Institute, 2018. 622 9. Myllyvirta L. Coronavirus has temporarily reduced China’s CO2 emissions by a 623 quarter. Centre for Research on Energy and Clean Air and CarbonBrief, based on data in 624 WIND Information, 2020. 625 10. NASA. Airborne Nitrogen Dioxide Plummets Over China. earth observatory. 626 Washington D.C., USA: North American Space Agency, 2020 627 https://earthobservatory.nasa.gov/images/146362/airborne-nitrogen-dioxide-plummets-628 over-china. 629 11. Myllyvirta L, Dahiya S. India’s Coronavirus Curfew Resulted in the Lowest One-Day 630 Traffic Pollution Levels on Record. CREA. Centre for Research on Energy and Clean Air, 2020 631 https://energyandcleanair.org/janata-curfew-pollution-levels/. 632 12. Rice JB. Prepare your supply chain for Coronavirus. Harvard Business Review. 633 2020;(27 Feb). 634 13. GSGIR. The new coronavirus could have a lasting impact on global supply chains. The 635 Economist. 2020;(15 Feb). 636 14. Zhang L, Hu Q, Zhang F. Input-output modeling for urban energy consumption in 637 Beijing: dynamics and comparison. PloS one. 2014;9(3). 638 15. Eckelman MJ, Sherman J. Environmental impacts of the US health care system and 639 effects on public health. PloS one. 2016;11(6):e0157014. 640 16. Rivers C, Chretien J-P, Riley S, Pavlin JA, Woodward A, Brett-Major D, et al. Using 641 “outbreak science” to strengthen the use of models during epidemics. Nature 642 Communications. 2019;10(1):3102. doi: 10.1038/s41467-019-11067-2. 643 17. Koks E, Rozenberg J, Zorn C, Tariverdi M, Vousdoukas M, Fraser S, et al. A global 644 multi-hazard risk analysis of road and railway infrastructure assets. Nature Communications. 645 2019;10. doi: 10.1038/s41467-019-10442-3. 646 18. Duan H, Wang S, Yang C. Coronavirus: limit short-term economic damage. Nature. 647 2020;578:515. 648 19. Inoue H, Todo Y. The Propagation of the Economic Impact through Supply Chains: 649 The Case of a Mega-City Lockdown against the Spread of COVID-19. SSRN. 650 2020:http://dx.doi.org/10.2139/ssrn.3564898. 651 20. Santos JR, May L, Haimar AE. Risk-based input-output analysis of influenza epidemic 652 consequences on interdependent workforce sectors. Risk analysis : an official publication of 653 the Society for Risk Analysis. 2013;33(9):1620-35. Epub 12/24. doi: 10.1111/risa.12002. 654 PubMed PMID: 23278756. 655 21. pwc. The possible economic consequences of a novel coronavirus (COVID-19) 656 pandemic 2020 [cited 2020 1 April]. Available from: 657 https://www.pwc.com.au/publications/australia-matters/economic-consequences-658 coronavirus-COVID-19-pandemic.pdf. 659 22. McKibbin WJ, Fernando R. The global macroeconomic impacts of covid-19: Seven 660 scenarios. CAMA Working Paper, Technical Report Canberra, Australia: CAMA, Australian 661 National University, 2020 19/2020. 662 23. Yu KDS, Aviso KB. Modelling the Economic Impact and Ripple Effects of Disease 663 Outbreaks. Process Integration and Optimization for Sustainability. 2020:1-4. 664 24. Auer R, Levchenko AA, Sauré P. International inflation spillovers through input-665 output linkages. BIS Working Paper. Basel, Switzerland: Bank for International Settlements, 666 2017 No 623. 667

24

25. Alsamawi A, Murray J, Lenzen M, Kanemoto K, Moran D. A novel approach to 668 quantitative accounting of income inequality. PLoS ONE. 2014;9(10):e110881. 669 26. Leontief W. Quantitative input and output relations in the economic system of the 670 United States. Review of Economics and Statistics. 1936;18(3):105-25. 671 27. Dietzenbacher E, Lenzen M, Los B, Guan D, Lahr ML, Sancho F, et al. Input–output 672 analysis: The next 25 years. Economic Systems Research. 2013;25(4):369-89. doi: 673 10.1080/09535314.2013.846902. 674 28. Miller RE, Blair PD. Input-Output Analysis: Foundations and Extensions. Englewood 675 Cliffs, NJ, USA: Prentice-Hall; 2010. 676 29. Wiedmann T, Lenzen M. Environmental and social footprints of international trade. 677 Nature Geoscience. 2018:in press. 678 30. Malik A, Lenzen M, McAlister S, McGain F. The carbon footprint of Australian health 679 care. The Lancet Planetary Health. 2018;2(1):e27-e35. doi: 10.1016/S2542-5196(17)30180-680 8. 681 31. UNSD. System of Environmental- Economic Accounting— SEEA Revision. Internet 682 site. United Nations, 2017 https://unstats.un.org/unsd/envaccounting/seearev/. 683 32. Tukker A, Dietzenbacher E. Global multiregional input-output frameworks: An 684 introduction and outlook. Economic Systems Research. 2013;25(1):1-19. doi: 685 10.1080/09535314.2012.761179. 686 33. Okuyama Y, Santos JR. Disaster impact and input-output analysis. Economic Systems 687 Research. 2014;26(1):1-12. doi: 10.1080/09535314.2013.871505. 688 34. Rose A. Economic principles, issues, and research priorities in hazard loss estimation, 689 In: Y. Okuyama and S.E. Chang (Eds) Modeling Spatial and Economic Impacts of Disasters. 690 New York: Springer2004. 691 35. Faturay F, Sun Y-Y, Dietzenbacher E, Malik A, Geschke A, Lenzen M. Using virtual 692 laboratories for disaster analysis–a case study of Taiwan. Economic Systems Research. 693 2019:1-26. 694 36. Schulte in den Bäumen H, Többen J, Lenzen M. Labour forced impacts and 695 production losses due to the 2013 flood in Germany. Journal of Hydrology. 2015;527:142-696 50. 697 37. Lian C, Haimes YY. Managing the risk of terrorism to interdependent infrastructure 698 systems through the dynamic inoperability input–output model. Systems Engineering. 699 2006;9(3):241-58. 700 38. Anderson CW, Santos JR, Haimes YY. A risk-based input–output methodology for 701 measuring the effects of the August 2003 northeast blackout. Economic Systems Research. 702 2007;19(2):183-204. 703 39. Donaghy KP, Balta-Ozkan N, Hewings GJ. Modeling unexpected events in temporally 704 disaggregated econometric input–output models of regional economies. Economic Systems 705 Research. 2007;19(2):125-45. 706 40. Haimes YY, Horowitz BM, Lambert JH, Santos JR, Lian C, Crowther KG. Inoperability 707 input-output model for interdependent infrastructure sectors. I: Theory and methodology. 708 Journal of Infrastructure Systems. 2005;11(2):67-79. 709

41. Santos JR, Haimes YY. Modeling the Demand Reduction Input‐Output (I‐O) 710

Inoperability Due to Terrorism of Interconnected Infrastructures. Risk Analysis. 711 2004;24(6):1437-51. 712 42. Dietzenbacher E, Miller RE. Reflections on the inoperability input–output model. 713 Economic Systems Research. 2015;27(4):478-86. 714

25

43. Oosterhaven J. On the limited usability of the inoperability IO model. Economic 715 Systems Research. 2017;29(3):452-61. doi: 10.1080/09535314.2017.1301395. 716 44. Dietzenbacher E, van Burken B, Kondo Y. Hypothetical extractions from a global 717 perspective. Economic Systems Research. 2019;31(4):505-19. doi: 718 10.1080/09535314.2018.1564135. 719 45. Xia Y, Guan D, Steenge AE, Dietzenbacher E, Meng J, Mendoza Tinoco D. Assessing 720 the economic impacts of IT service shutdown during the York flood of 2015 in the UK. 721 Proceedings of the Royal Society A. 2019;475(2224):20180871. 722 46. Steenge AE, Bočkarjova M. Thinking about imbalances in post-catastrophe 723 economies: an input–output based proposition. Economic Systems Research. 724 2007;19(2):205-23. 725 47. Faturay F, Sun Y-Y, Dietzenbacher E, Malik A, Geschke A, Lenzen M. Using Virtual 726 Laboratories for disaster analysis – A case study of Taiwan. Economic Systems Research. 727 2020;32(1):58-83. 728 48. Leontief W, Ford D. Environmental repercussions and the economic structure: an 729 input-output approach. Review of Economics and Statistics. 1970;52(3):262-71. 730 49. Lenzen M, Wood R, Wiedmann T. Uncertainty analysis for Multi-Region Input-Output 731 models – a case study of the UK’s carbon footprint. Economic Systems Research. 732 2010;22(1):43-63. 733 50. Lenzen M, Geschke A, Wiedmann T, Lane J, Anderson N, Baynes T, et al. Compiling 734 and using input–output frameworks through collaborative virtual laboratories. Science of 735 The Total Environment. 2014;485–486:241-51. doi: 736 http://dx.doi.org/10.1016/j.scitotenv.2014.03.062. 737 51. Geschke A, Hadjikakou M. Virtual laboratories and MRIO analysis – an introduction. 738 Economic Systems Research. 2017;29(2):143-57. doi: 10.1080/09535314.2017.1318828. 739 52. Lenzen M, Geschke A, Abd Rahman MD, Xiao Y, Fry J, Reyes R, et al. The Global MRIO 740 Lab - charting the world economy. Economic Systems Research. 2017;29(2):158-86. 741 53. Friedlingstein P, Jones MW, O'Sullivan M, Andrew RM, Hauck J, Peters GP, et al. 742 Global carbon budget 2019. Earth System Science Data. 2019;11(4). 743 54. Business Insider. Many of the world's airlines could be bankrupt by May because of 744 the COVID-19 crisis, according to an aviation consultancy. These airlines have already 745 collapsed because of the pandemic. 2020 [cited 2020 1 April]. Available from: 746 https://www.businessinsider.com/coronavirus-airlines-that-failed-bankrupt-covid19-747 pandemic-2020-3?r=AU&IR=T. 748 55. IATA. Deeper Revenue Hit from COVID-19 2020 [cited 2020 1 April]. Available from: 749 https://www.iata.org/en/pressroom/pr/2020-03-24-01/. 750 56. Claughton D, Fowler C, Fitzgerald D. Mining exploration and service companies hit by 751 coronavirus restrictions 2020. Available from: https://www.abc.net.au/news/rural/2020-04-752 07/covid19-mine-explorers-and-service-sector/12125518. 753 57. Liang S, Stylianou KS, Jolliet O, Supekar S, Qu S, Skerlos SJ, et al. Consumption-based 754 human health impacts of primary PM2. 5: The hidden burden of international trade. J Clean 755 Prod. 2017;167:133-9. 756 58. Thomas L. Coronavirus wreaks havoc on retail supply chains globally, even as China’s 757 factories come back online. CNBC, 2020 https://www.cnbc.com/2020/03/16/coronavirus-758 wreaks-havoc-on-retail-supply-chains-globally.html. 759 59. International Monetary Fund. Transcript of Kristalina Georgieva's Participation in the 760 World Health Organization Press Briefing 2020 [cited 6 April]. Available from: 761

26

https://www.imf.org/en/News/Articles/2020/04/03/tr040320-transcript-kristalina-762 georgieva-participation-world-health-organization-press-briefing. 763 60. ILO. COVID-19 and world of work: impacts and responses 2020 [cited 2020 5 April]. 764 Available from: https://www.ilo.org/wcmsp5/groups/public/---dgreports/---765 dcomm/documents/briefingnote/wcms_738753.pdf. 766 61. Le Quéré C, Jackson RB, Jones MW, Smith AJ, Abernethy S, Andrew RM, et al. 767 Temporary reduction in daily global CO 2 emissions during the COVID-19 forced 768 confinement. Nature Climate Change. 2020:1-7. 769 62. Mozur P, Zhong R, Krolik A. In Coronavirus Fight, China Gives Citizens a Color Code, 770 With Red Flags: The New York Times; 2020 [cited 2020 6 April]. Available from: 771 https://www.nytimes.com/2020/03/01/business/china-coronavirus-surveillance.html. 772 63. Harari YN. The world after coronavirus: Financial Times; 2020 [cited 2020 6 April]. 773 Available from: https://www.ft.com/content/19d90308-6858-11ea-a3c9-1fe6fedcca75. 774 64. Mahbubani R. More than 100 wild animals in China died from poisoning in a mass 775 die-off seemingly triggered by coronavirus disinfectant.: Business Insider; 2020 [cited 2020 6 776 April]. Available from: https://www.businessinsider.com/disinfectant-control-coronavirus-777 triggers-mass-die-off-animals-china-2020-2. 778 65. Walker PG, Whittaker C, Watson O, Baguelin M, Ainslie K, Bhatia S, et al. The Global 779 Impact of COVID-19 and Strategies for Mitigation and Suppression. On behalf of the imperial 780 college covid-19 response team, Imperial College of London. 2020. 781 66. Salata C, Calistri A, Parolin C, Palù G. Coronaviruses: a paradigm of new emerging 782 zoonotic diseases. Pathogens and Disease. 2019;77(9):ftaa006. 783 67. Scheffer M, Bascompte J, Brock WA, Brovkin V, Carpenter SR, Dakos V, et al. Early-784 warning signals for critical transitions. Nature. 2009;461(7260):53-9. 785 68. Peters GP, Marland G, Le Quéré C, Boden T, Canadell JG, Raupach MR. Rapid growth 786 in CO2 emissions after the 2008–2009 global financial crisis. Nature Climate Change. 787 2012;2(1):2-4. doi: 10.1038/nclimate1332. 788 69. Plume B, Popovich N, Lawal S. The Coronavirus and Carbon Emissions: The New York 789 Times; 2020 [cited 2020 6 April]. Available from: 790 https://www.nytimes.com/2020/02/26/climate/nyt-climate-newsletter-coronavirus.html. 791 70. Barbier E. How is the global green new deal going? Nature. 2010;464(7290):832-3. 792 doi: https://doi.org/10.1038/464832a. 793 71. D’Alessandro S, Cieplinski A, Distefano T, Dittmer K. Feasible alternatives to green 794 growth. Nature Sustainability. 2020:1-7. doi: https://doi.org/10.1038/s41893-020-0484-y. 795 72. IPCC. Special Report: Global Warming of 1.5 ºC. Summary for Policy Makers: 796 Intergovernmental Panel on Climate Change; 2018. Available from: 797 https://www.ipcc.ch/sr15/. 798 73. UN Secretary-General. ‘We Are Only as Strong as the Weakest’, Secretary-General 799 Stresses, at Launch of Economic Report on COVID-19 Pandemic 2020 [cited 2020 6 April]. 800 Available from: https://reliefweb.int/report/world/we-are-only-strong-weakest-secretary-801 general-stresses-launch-economic-report-covid-19. 802 803 804

805

806

807

27

808

809

810

811

812

813

814

815

816

817

818

819

820

Supporting Information 821

822

Fig. SI1 Graph showing total confirmed COVID-19 cases for 12 aggregated world regions, 823 starting from 21 January 2020 to 25 May 2020. 824 825 Fig. SI2 Graph showing total deaths from COVID-19 for 12 aggregated world regions, 826 starting from 21 January 2020 to 25 May 2020. 827 828 Fig. SI3 Graph showing total confirmed cases of COVID-19 and total deaths from the 829 disease for about 210 countries and regions, as of 8 April 2020. Data taken from the 830 Worldometer database 831 832 Fig. SI4 Regional detail captured in this study, shown as 38 regions. The 38 regions are 833 aggregated into 12 categories, listed in the legend on the right of the figure. China (Mainland): 834 China (excludes Hong Kong and Taiwan). 835 836 Fig. SI5: Data collection template. 837 838 Fig. SI6: Probability distributions resulting out of Monte-Carlo simulations of our MRIO 839 disaster analysis. The stochastic standard deviations are derived by fitting a normal 840 distribution. Mode: value that appears most often; IQR: interquartile range, being equal to 841 the difference between 75th and 25th percentiles. IQR/M: ratio of IQR and mode, measuring 842 the variability. 843 844 Fig. SI7: Economic and emission losses from two sensitivity scenarios. Scenario 1 uses the 845 IATA third assessment report, and Scenario 2 adopts the IATA fourth assessment report, 846 ceteris paribus. 847 848

28

Fig. SI8: Sensitivity (in % relative deviation from unperturbed result) of Covid-19-849 related reductions in global GHG emissions as a result of a 25% perturbation of 850

individual entries in the matrix. 851

852 Fig. SI9: Production layer decomposition of COVID-19 impacts, by industry sector. The 853 horizontal axis depicts production layers, with ‘1’ being the sectors immediately hit by losses 854 (lockdowns, closures, bans), ‘2’ their suppliers, ‘3’ the suppliers of their suppliers, etc. 855 Transport and tourism are mostly affected. 856 857 Fig. SI10: Production layer decomposition of COVID-19 impacts, by region. The 858 horizontal axis depicts production layers, with ‘1’ being the regions immediately hit by losses 859 (lockdowns, closures, bans), ‘2’ their immediate trade partners, ‘3’ the partners of trade 860 partners, etc. China and East Asia are mostly affected. 861 862 Fig. SI11: Time series of total global CO2 emissions from 1750 to May 2020. Data from 863 1750 to 2018 are from the Global Carbon Budget, 2019 is estimated using projection ratios 864 from and May 2020 data are from this study (which include non-CO2 emissions). 865 866 Fig. SI12: Time series of total global GHG emissions from 1970 to May 2020 (including 867 CO2, CH4, N2O, F-gases and emissions from land use change and forestry). Data from 868 1970 to 2017 are from the EDGAR database, 2018 and 2019 are estimated using projection 869 ratios from Peters et al. and May 2020 data are from this study. 870 871 Fig. SI13: Left panel: Same as Fig. SI12 but with a continuous drop of 9% per year after 872 2020 visualised in the red, dotted line (i.e. twice the drop of 4.5% induced by COVID-19 873 until May 2020 estimated in this study). In order to limit global warming to 1.5ºC without 874 the use of technology that removes carbon dioxide from the atmosphere, a drop of about 9-875 10% in total global GHG emissions would be required every year between 2020 and 2050. 876 Right panel: Figure taken from the IPCC 2018 Special Report on Global Warming of 1.5ºC 877 (Summary for Policy Makers), showing global net anthropogenic CO2 emissions in modelled 878 pathways limiting global warming to 1.5°. 879 880 Fig. SI14: Reductions in emissions as a result of a decline in trade volume in international 881 supply chains due to the global COVID-19 effects. Lines connect ultimate origins and 882 destinations of supply chains, both direct and multi-node. Line thickness represents trade 883 volume lost. 884 885 Table SI4.1 Shares of global GDP, tourism GDP and SARS-Cov-2 reported cases for 886 selected countries/regions (shares for last column valid as of 22th May 2020) 887 888 Table SI4.2 Data collection sources 889 890 Tab. SI5.1: Summary of deviations in main results from varying the main data source for 891 compiling the underlying MRIO table. 892

Table SI6.1 Data table for Fig.1 893

Table SI6.2 Data table for Fig.2 894

Tab. SI6.3: Average wages of lost employment 895

29

896 Table SI7.1: Comparisons of the impacts of the COVID-19 pandemic, for loss of GDP. 897 The Bloomberg, World Bank and CAMA scenarios are all of global pandemics. 898 899 Table SI 7.3: Comparisons of the impacts of the COVID-19 pandemic, for employment. 900 901 902 903 904 905

906