Im­pact Se­ries: 01–21

Crumbling Economy, Booming Trade
The Surprising Resilience of World Trade in 2020


How can it be that we experience a historic recession, but international trade is doing just fine? In this Kühne Impact Series, we argue that the COVID-19 recession is one in which non-tradable services suffer but tradable goods thrive. Reasons include that (i) consumption of in-person services is restricted; (ii) demand for consumer durables is strong; (iii) demand for medical goods is high; and (iv) online shopping is popular.

Full Article

The COVID-19 re­ces­sion is the deep­est re­ces­sion since the end of the Sec­ond World War. For ex­am­ple, the GDP of OECD coun­tries con­tract­ed by a stag­ger­ing -12.2 % dur­ing the COVID-19 re­ces­sion be­tween the fourth quar­ter of 2019 and the sec­ond quar­ter of 2020 (the two quar­ters pre­ced­ing the trough). As a point of com­par­i­son, the GDP of OECD coun­tries con- tract­ed by a much small­er -4.6% dur­ing the Great Re­ces­sion of 2008-09 be­tween the sec­ond quar­ter of 2008 and the first quar­ter of 2009 (the three quar­ters pre­ced­ing the trough).

Yet, in­ter­na­tion­al trade has been sur­pris­ing­ly re­silient in 2020. The ex­port vol­umes of OECD coun­tries fell by 24% dur­ing the first two quar­ters of 2020 but were al­ready back to 95% of their pre-cri­sis lev­el at the end of the third quar­ter of 2020. Dur­ing the Great Re­ces­sion, in con­trast, the ex­port vol­umes of OECD coun­tries fell by 32% over the same time span (two quar­ters - 2008-Q4 and 2009-Q1) and took more than two years to re­cov­er (in 2010-Q1, so al­most a year af­ter the trough of the re­ces­sion, ex­port vol­umes were still only at 85% of their pre-cri­sis lev­el).

The COVID-19 recession is the deepest recession since the end of the Second World War. Yet, international trade has been surprisingly resilient in 2020.

Note: This fig­ure dis­plays the quar­ter­ly evo­lu­tion of GDP and mer­chan­dise trade in OECD coun­tries dur­ing the Great Re­ces­sion and the COVID-19 cri­sis so far. Data: OECD

In this Kühne Im­pact Se­ries, we pro­pose an ex­pla­na­tion for the sur­pris­ing re­silience of world trade in 2020 and the as­so­ci­at­ed de­cou­pling of glob­al trade and GDP dy­nam­ics. We base our ex­pla­na­tion on an in­tu­itive con­cep­tu­al frame­work in which trade flows de-pend on the (i) sup­ply of trad­able goods by firms, (ii) the de­mand for trad­able goods by firms and house-holds, and (iii) trade costs (cus­tom du­ties and tax­es, but also trans­porta­tion costs and ex­port cred­its).

The dramatic drop in world trade at the onset of the pandemic was the result of a “perfect storm”.

From perfect storm to isolated crisis

The dra­mat­ic drop in world trade at the on­set of the pan­dem­ic was the re­sult of a “per­fect storm”. In par­tic­u­lar, the sup­ply of trad­ed goods was in­ter­rupt­ed by lock­downs and plants clo­sures; the de­mand for trad­ed goods plum­met­ed due to in­creased in­come un­cer­tain­ty, rais­ing un­em­ploy­ment and so­cial dis­tanc­ing; and trade costs sky­rock­et­ed due to ex­port re­stric­tions, ground­ed planes, and closed bor­ders. Hence, the pan­dem­ic had an ad­verse ef­fect on all ma­jor de­ter­mi­nants of glob­al trade, which ex­plains its ini­tial col­lapse. The sub­se­quent re­cov­ery of glob­al trade is due to sev­er­al fac­tors:

First, glob­al sup­ply re­cov­ered quick­ly. The key dri­ver of this was the spec­tac­u­lar re­bound of the Chi­nese econ­o­my. De­spite a large drop in GDP in the first quar­ter of 2020, Chi­na only ex­pe­ri­enced a very short V-shaped re­ces­sion. Chi­nese GDP was al­most back to its pre-pan­dem­ic lev­el by the end of March 2020. Im­por­tant­ly, Chi­na’s man­u­fac­tur­ing plants were open and run­ning dur­ing the most se­vere part of the lock­downs in West­ern economies (when plants were closed in Eu­rope or the Unit­ed States), and there­fore main­tained a lev­el of pro­duc­tion that could com­pen­sate the ini­tial col­lapse of the sup­ply of goods world­wide.

Note: This fig­ure dis­plays the quar­ter­ly change rel­a­tive to the pre­vi­ous year in the GDP of Chi­na, the Eu­ro­zone, and the U.S. over the pe­ri­od 2019–2020. Data: OECD

Lat­er, Eu­rope and the Unit­ed States also learned to cope bet­ter with the pan­dem­ic but nev­er quite man- aged to repli­cate Chi­na’s suc­cess.

Sec­ond, trade costs fell again quick­ly. Ex­port re­stric­tions were lift­ed as ear­ly as April 2020, there­by re­duc­ing the ini­tial dif­fi­cul­ties as­so­ci­at­ed with closed bor­ders. On top of this, fuel prices hit a record low in 2020, re­duc­ing trans­porta­tion costs. The price of crude oil per bar­rel was around US$55 in 2019, where- as it was less than US$20 in the first quar­ter of 2020.

Third, glob­al de­mand for trad­able goods quick­ly re­cov­ered. We now take a clos­er look at this part of the sto­ry be­cause it also plays a cen­tral role in our broad­er ar­gu­ment.

A recession in services

Our key point is that the de­mand for trad­able goods quick­ly re­bound­ed while the de­mand for non-trad­able ser­vices re­mained low. This ex­plains the de­cou­pling of glob­al trade and GDP dy­nam­ics and there­fore also the sur­pris­ing re­silience of glob­al trade. Trade and GDP dy­nam­ics can de­cou­ple be­cause GDP has many non-trad­able com­po­nents. For ex­am­ple, non-trad­able goods and ser­vices rep­re­sent 65% of U.S. GDP.

In par­tic­u­lar, U.S. na­tion­al ac­counts data re­veals that ser­vices re­quir­ing phys­i­cal in-per­son in­ter­ac­tions were both the key dri­ver of the down­turn in the sec­ond quar­ter of 2020 as well as the main rea­son why GDP re­mained low in the third quar­ter of 2020. Con­sumer durables, on the oth­er hand, only took a small hit in the sec­ond quar­ter of 2020 and even boomed in the third quar­ter of 2020 rel­a­tive to the pre­vi­ous year. This points to a shift in con­sumer ex­pen­di­ture away from pre­dom­i­nant­ly non-trad­ed in-per­son ser­vices (such as restau­rants, hair­cut­ting sa­lons or clean­ing ser­vices) to­wards pre­dom­i­nant­ly trad­ed con­sumer durables (such as com­put­ers, fur­ni­ture, and oth­er home ap­pli­ances).

Note: This fig­ure dis­plays the quar­ter­ly change in val­ue added of se­lect­ed U.S. sec­tors in 2020 rel­a­tive to 2019. Ex­clud­ed sec­tors are: Min­ing, Ex­trac­tion, Con­struc­tion, and Re­tail Trade. In­dus­tri­al goods in­clude non-durable in­ter­me­di­ate in­puts and in­vest­ment goods. Con­sumer non-durable goods in­clude food, tex­tile and ap­par­el, and pa­per prod­ucts. Con­sumer durable goods in­clude elec­tron­ics, fur­ni­ture, and home ap­pli­ances. Not in-per­son ser­vices in­clude busi­ness ser­vices, fi­nance, and real es­tate, as well as the health in­dus­try. Data: BEA – In­dus­try Eco­nom­ic Ac­count Data

This in­ter­pre­ta­tion is also con­sis­tent with re­cent re­search by Raj Chet­ty (Har­vard Uni­ver­si­ty) and co- au­thors who show that con­sumers over­whelm­ing­ly spent the mon­ey they re­ceived through fis­cal stim­uli on durable goods.1

There are a num­ber of plau­si­ble ex­pla­na­tions for this sub­sti­tu­tion in con­sumer spend­ing away from non- trad­ed ser­vices to­wards trad­ed goods. First, the pre­vail­ing san­i­tary mea­sures se­vere­ly re­strict­ed the pos­si­bil­i­ty to con­sume in-per­son ser­vices. Sec­ond, the need to work from home has pushed con­sumers to in­vest in elec­tron­ics and oth­er con­sumer durables. Third, the pan­dem­ic has cre­at­ed its own de­mand for trad­ed goods such as phar­ma­ceu­ti­cals, med­ical ap­par­el (gog­gles, gloves), and ob­vi­ous­ly masks. For ex­am­ple, trade vol­umes of FFP2 and FFP3 masks – which are main­ly pro­duced in Chi­na – in­creased 15-fold be­tween Feb­ru­ary and May 2020. And fourth, there has been a surge in on­line shop­ping (with an av­er­age in­crease in turnover of 16% rel­a­tive to 2019), which like­ly also added to the de­mand of trad­able goods.

Our key point is that the demand for tradable goods quickly rebounded while the demand for non-tradable services remained low.

Source: This fig­ure dis­plays the month­ly change in turnover for re­tail trade over­all, and for re­tail trade through mail or­der and in­ter­net for 2019–2020. Data: OECD

This shift in spend­ing to­wards trad­able goods is also the key dif­fer­ence be­tween the COVID-19 re­ces­sion and the Great Re­ces­sion of 2008–2009. In par­tic­u­lar, Jonathan Eaton (PennState) and co-au­thors showed that the col­lapse in de­mand in 2008–2009 shift­ed fi­nal spend­ing away from trad­able goods, par­tic­u­lar­ly by re­duc­ing in­vest­ment in durable goods by firms.2 They es­ti­mate this shift to ac­count for al­most two- thirds of the col­lapse in glob­al trade rel­a­tive to GDP dur­ing that time. In to­day’s pan­dem­ic, trade in in­vest­ment goods and con­sumer durables were in­stead on the rise as ear­ly as May 2020.

Note: This fig­ure dis­plays the month­ly change in ag­gre­gate vol­umes of trad­ed goods in 2020. Con­sumer durable goods in­clude elec­tron­ics, ap­pli­ances, and fur­ni­ture. Con­sumer non-durable goods in­clude agri­cul­tur­al goods, tex­tile, and ap­par­els, as well as pa­per prod­ucts. Non-durable in­ter­me­di­ate in­puts in­clude raw ma­te­ri­als, chem­i­cals, and plas­tics. In­vest­ment goods in­clude man­u­fac­tured parts and wood prod­ucts. Data: COM­TRADE

Conclusion – A recession that favors demand for trade

So, how can it be that we ex­pe­ri­ence a his­toric re­ces­sion, but in­ter­na­tion­al trade is do­ing just fine? The rea­son is that, fol­low­ing an ini­tial col­lapse, the COVID-19 re­ces­sion de­vel­oped into one where non-trad­able ser­vices suf­fer and trad­able goods thrive. Plau­si­ble ex­pla­na­tions for this are that (i) the lock­down mea­sures have made it all but im­pos­si­ble to con­sume in-per­son ser­vices; (ii) the need to work from home has in­creased spend­ing on con­sumer durables; (iii) the pan­dem­ic has di­rect­ly cre­at­ed de­mand for trad­able med­ical goods; and (iv) a surge in on­line shop­ping has fa­vored trad­able goods.

An open ques­tion is whether this de­mand for trade is sus­tain­able. Large parts of the econ­o­my are cur­rent­ly on “life sup­port”, with gov­ern­ment pro­grams and busi­ness own­ers cov­er­ing huge loss­es to keep busi­ness afloat. With­out a com­pre­hen­sive re­cov­ery, these parts of the econ­o­my will even­tu­al­ly col­lapse with un­cer­tain im­pli­ca­tions for glob­al trade.

  • Chet­ty, R. and Fried­man, J. N., and Hen­dren, N. and Step­n­er, M. No­vem­ber 2021
    The Eco­nom­ic Im­pacts of COVID-19: Ev­i­dence from a New Pub­lic Data­base Built Us­ing Pri­vate Sec­tor Data
  • Eaton, J. and Ko­r­tum, S. and Neiman, B. and Ro­ma­lis, J. 2016
    Trade and the Glob­al Re­ces­sion


Mathilde Le Moigne

Senior Research Fellow at the Kühne Center for Sustainable Trade and Logistics at the University of Zurich


Ralph Ossa

Kühne Foundation Professor of International Trade


More Issues

Twitter logo