Macroeconomic effects of Covid-19： a mid-term review -bis.ppt

《Macroeconomic effects of Covid-19： a mid-term review -bis.ppt》由会员分享，可在线阅读，更多相关《Macroeconomic effects of Covid-19： a mid-term review -bis.ppt（22页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、BIS Working PapersNo 959Macroeconomic effects ofCovid-19:a mid-term reviewby Phurichai RungcharoenkitkulMonetary and Economic DepartmentAugust 2021JEL classification:E00,I18.Keywords:Covid-19 pandemic,health-economictradeoffsBIS Working Papers are written by members of the Monetary and EconomicDepar

2、tment of the Bank for International Settlements,and from time to time by othereconomists,and are published by the Bank.The papers are on subjects of topicalinterest and are technical in character.The views expressed in them are those of theirauthors and not necessarily the views of the BIS.This publ

3、ication is available on the BIS website(www.bis.org).Bank for International Settlements 2021.All rights reserved.Brief excerpts may bereproduced or translated provided the source is stated.ISSN 1020-0959(print)ISSN 1682-7678(online)Macroeconomic effects of Covid-19:a mid-term reviewPhurichai Rungcha

4、roenkitkulAugust 10,2021Forthcoming in the Pacific Economic ReviewAbstractThis article provides an interim assessment of the macroeconomic consequences of the Covid-19 pandemic.Estimates suggest a median output loss of about 6.5 percent in 2020,a gapthat is expected to narrow to around 4 percent of

5、the pre-pandemic trend by the end of2021.There is however a high dispersion of economic losses across economies,reflectingvarying exposures to the pandemic and societies responses.High-frequency indicators andepidemiological models provide some insights into the interactions between the pandemicevol

6、ution and societies strategies of combating it,including the role of vaccination.Thearticle draws lessons from experiences thus far and discusses challenges ahead.JEL Classification:E00,I18Keywords:Covid-19 pandemic,health-economic tradeoffsI am grateful to Frederic Boissay,a joint work with whom la

7、ys the basis for some materials being used here.I thank Claudio Borio,Benoit Mojon,Daniel Rees,Hyun Song Shin,Christian Upper,and colleagues at the BISfor helpful discussion and comments.I also thank the editor Tai-kuang Ho and an anonymous referee for usefulsuggestions.All remaining errors are mine

8、.The views expressed are mine and do not necessarily represent thoseof the BIS.Bank for International Settlements1 IntroductionThe Covid-19 pandemic is a twin crisis of truly global proportion.It is first and foremost aglobal health emergency,caused by a novel coronavirus that,absent policy actions,

9、could claimtens of millions of lives.1 The epidemiological root of the crisis makes it particularly perilous:a virus cannot be negotiated with and does not respect borders.Indeed,no country,rich orpoor,large or small,managed to escape from the pandemics disruptive force.And as the worldwas galvanise

10、d into action in a bid to save lives,the health crisis escalated into an economicone.The historic large-scale lockdowns designed to contain the virus prompted a collapse ineconomic activity never before seen.The economic significance of the Covid-19 cannot be overstated.The massive scale ofeconomic

11、losses alone makes the crisis one of the most pressing and consequential problems,for many economists in their lifetimes.The unusual character of the crisis,a mixture betweenepidemiology and economics,makes the problem a challenging one intellectually.At the sametime,in terms of policy prescription,

12、the stake cannot be higher for economists to get it right.How costly have the lockdowns been in output terms?How much livelihood is worth sacrificingto save lives?Does vaccine arrival mean economies can reopen again?These are but somequestions that economists can help shed light on.To be sure,the ec

13、onomics of epidemics is nota new field.But the Covid-19 has uncertainly made it mainstream.It is impossible to do a full justice to a crisis that has spawned a whole new literature,a book collection and a new journal in a matter of few months.2 The modest objectiveof this article is two-fold.The fir

14、st goal is to gain a quantitative sense of economic lossesand the channels through which they came about,in the process drawing comparisons withprevious pandemics/epidemics.The second part is devoted to discussing the policy challengesin managing the Covid-19 pandemic.Of particular interest is the c

15、omplex two-way interactionsbetween economic activity and the spread of the virus,and how one could make an informeddata-driven decision in striking a balance between lives and livelihood.I will then briefly discussremaining challenges in the next stage of the crisis,with large-scale vaccination in s

16、ight but thevirus also mutating continuously.2 Lessons from past epidemicsExperiences with past epidemics hold a number of lessons that resonate with the Covid-19crisis.Losses of human capital can cause severe and long-lasting economic damages,especiallyin pandemics.3 Epidemics with relatively low f

17、atality rate can also significantly harm economicactivity in affected areas.Economic losses can arise through various channels.Fears and healthconcerns can harm sentiments and hold back demand.Public health measures designed tocontain the outbreak,somewhat paradoxically,can be detrimental to economi

18、c activity.As1As of June 2021,there have already been 4 million confirmed deaths from the Covid-19.An uncontrolledpandemic could have claimed many more,e.g.applying age-specific infection fatality rates to the globaldemographic structure implies 55 million total fatalities.Surges in demand for ICU b

19、eds would likely leadto additional deaths.2See Brodeur et al.(2021)and Bloom et al.(2021)for literature reviews on the economic impact and comparisonswith previous epidemics,echoing some of the same themes in this article.3This section draws in part from Boissay and Rungcharoenkitkul(2020).1Table 1:

20、Studies on past epidemicsEpidemics FatalitiesInfluenzapandemic1918-19Studies and methodsBarro et al.(2020)Cross-country panel regressionsBrainerd and Siegler(2003)US states dataEconomic losses6 ppt lower GDP growth8 ppt lower consumption growthMortality significantly lowers growth infollowing decade

21、Up to 50millionCorreia et al.(2020)US states data18%decline in manufacturing activityper year;prompter and more aggressivecontainment helped reduce the impact4.8%loss in annual global GDPGlobal Preparedness MonitoringBoard(2019)1918-type pandemicHypotheticalpandemicsFan et al.(2016)1918-type pandemi

22、cInclude intrinsic cost of mortality0.4-1%of GDP loss per year due to ex anteprospects of a pandemic,86%of whichis due to mortality and 14%to incomeloss.For moderate pandemics,the shareof income loss is larger at 40%Keogh-Brown et al.(2010)H1N1 pandemicMultisector CGE,Europe1.4-6%loss in annual GDP0

23、.5-2 ppt of which due to mortality0.9-4 ppt from school closures and absenceBurns et al.(2006)1918-type pandemic3.1%loss in annual global GDP0.4 ppt of which due to mortality0.9 ppt from illness and absenteeism1.9 ppt from efforts to avoid infection4.25%loss in annual GDP2.25 ppt from supply side2 p

24、pt from demand side0.1%loss in global GDP in 2003Arnold et al.(2006)1918-type pandemicSARS2003Lee and McKibbin(2004)CGE model774Hai et al.(2004)Chinese surveys1-2 ppt lower GDP growth in ChinaH5N1455Burns et al.(2006)World Bank estimateWorld Bank(2014)CGE model0.1%loss in annual global GDP0.4%for As

25、ia2003-19Ebola2014-1611,3232.1-3.4 ppt lower GDP growth in affectedcountries in 1st year of epidemicin any recession,financial and credit market procyclical developments can amplify and worsenoutput losses.The disproportionate effect of an epidemic on the service sector,which dependsparticularly on

26、social interactions,is also well-recognised from prior experiences.A key issue addressed by previous research is the magnitude of economic losses in thewake of an epidemic.Table 1 summarises selected studies on notable epidemics and estimatesof output losses.The influenza pandemic of 1918 in particu

27、lar has received much attention inthe literature,being one of deadliest pandemics in modern history.4 Hypothetical studies priorto the Covid-19 indeed often model a pandemic scenario around the 1918 influenza.At thesame time,there have also been numerous works on more recent epidemics,such as SARS i

28、n2003 and Ebola during 2014-2016,which serve as a basis for comparison.It is evident from this research that severe pandemics brought about not only extremelyhigh fatality,but also economics costs that were an order of magnitude larger than those of4The world has seen deadlier epidemics in the more

29、distant past.The bubonic plague of the 14th century claimedup to 200 million lives,while the European viruses in Mexico effectively decimated an entire civilisation.Butmuch poorer sanitary standards during these times arguably make these episodes less useful as a benchmark forthe current situation.2

30、typical epidemics.The 1918 influenza led to an output loss of around 6-8%of GDP accordingto Barro et al.(2020)and a 20%decline in manufacturing employment according to Correiaet al.(2020).Hypothetical studies suggest a present-day repeat of the 1918 pandemic wouldlead to an output loss of 3-6%of ann

31、ual global GDP,(see Global Preparedness Monitoring Board(2019),Burns et al.(2006)and Keogh-Brown et al.(2010).Meanwhile,outbreaks of SARS,H5N1 or Ebola were highly disruptive for the affected countries,but entailed only a negligibleoutput loss as a percentage of global GDP(See Lee and McKibbin(2004)

32、and Hai et al.(2004)for SARS,Burns et al.(2006)for H5N1 and World Bank(2014)for Ebola outbreaks).While most agree on the high cost of the 1918 pandemic,the precise mechanism generatingeconomic losses remains a subject of debate.One hypothesis is that the pandemic harmedeconomies by destroying human

33、lives and associated productive capacity.Illnesses and mortalityrisks could also disrupt economic activity.5 An alternative view is that much of output lossesresulted from governments non-pharmaceutical intervention(NPIs)such as enforced lockdownsand closures,introduced to mitigate fatality.Sorting

34、out which of these two factors is moreimportant requires a natural experiment.For example,Correia et al.(2020)exploit differences inNPI intensities across states in the US,and find those that implemented NPIs for longer tendedto experience lower mortality and higher manufacturing employment growth.T

35、hey concludethat output loss was a direct consequence of the pandemic rather than policies designed to quellit.Lilley et al.(2020)dispute the finding however,arguing that part of the observed employmentexpansion were driven by pre-1918 population growth trend.This debate of whether the curemay be wo

36、rse than the disease and whether there is a health-economic trade-off echoes some ofpresent-day discussion(see next section).Previous research sometimes divides the transmission mechanism of an epidemic intodemand versus supply channels,recognising that both can play a role.Supply effects mayinclude

37、 a retrenchment in labour supply,by workers who fall ill or refrain from participationin high-contact jobs due to health concerns.Demand effects include cutbacks in spending onservices,or other goods due to lost income.For example,Arnold et al.(2006)examine thesupply channel in a 1918-like pandemic,

38、by combining an estimated loss of work days with anestimated productivity per worker.They conclude that,in the first year,the pandemic wouldreduce GDP by 2.3%.The same study finds the effect through the demand side to be about2%of GDP,assuming a larger effect in industries reliant on social interact

39、ions.In other words,both demand and supply channels appear to be similarly important in quantitative terms.Another important lesson from previous epidemics is the potential long-lasting effects onthe economy.On the supply side,Fan et al.(2016)find that mortality and the reduction of thelabour force

40、made up the largest part of the cost associated with the 1918 influenza pandemic,due to its persistent effect.A one-time reduction in labour force raised the capital-to-labourratio and lowered the rate of return to capital,thus slowing the pace of capital accumulationand GDP growth for many years.6

41、Aggregate demand itself could also be persistently depressed.5Barro et al.(2020)note that the illness of US President Woodrow Wilson may have contributed to the harshterms of the Versailles Treaty in 1919,with tragic ramifications in the following decades.6Jorda et al.(2020)also find that real wages

42、 remained elevated over more than three decades after majorpandemics.The concurrent World War I presented an identification challenge in quantifying the pandemicscontribution to mortality,however(see Barro et al.(2020)3Table 2:Now and thenEpidemicsDeath toll ContainmentmeasuresFinancialamplification

43、LittleReal amplificationContext1918 pandemic50 millionSocial distancingLittleLittleWWI;high share ofmanufacturing in GDP(AEs)SARS774Social distancingLittleChineseacceleratinggrowthCovid-19,1 MarCovid-19,8 Apr299682,220Wuhan/Lombardy Market sellofflockdownsSupplydisruptionschain Globalisedeconomies;I

44、ntegratedGlobal lockdownTighteningconditionsfinancial Supplychain supply/creditchains;Sudden High share of servicesin GDP for many;Highdisruptions;stop in demandCovid-19,9 Oct1.06 million Global lockdown Stableeasedfinancial HH/firm balance sheets leverage in parts of realsectormarkets,but banks und

45、er duressunder pressureJorda et al.(2020)examine 12 major pandemics in Europe since the 14th century,finding thatthey were followed by multiple decades of low interest rates.Their interpretation is that higherprecautionary saving and scarce investment opportunities kept the natural interest rates lo

46、w.Unlike wars,pandemics do not destroy physical capital and typically lead to a long period ofexcess capital relative to workers.3 Covid-19:a unique crisisWhile many lessons from past pandemics resonate today,the Covid-19 crisis also has severaldistinctive characteristics that set it apart from hist

47、orical benchmarks.Table 2 provides forsome bases for comparison.The context of highly globalised and integrated economies alonemakes a modern-day pandemic more likely and potentially more destructive to begin with.7 Butthe most consequential and defining distinction is the large-scale lockdown polic

48、ies introducedacross the world.The unprecedented health policy response was designed to avert the extremelyhigh death tolls associated with previous major pandemics.At the same time,the collateraldamage of putting the economy in an induced coma has been staggering.Indeed,the proximatecause of the ec

49、onomic crisis this time is a conscious policy decision to save lives,rather than anunforeseen shock as typically the case in past recessions.The unique character of the current crisis means the duration of economic duress is directlytied to pandemic developments.The latter are uncertain and highly n

50、on-linear:small tweaksto policy could be a difference between infections stabilising or rising exponentially.As Table2 shows,snapshots from March,April and October painted very different pictures,as thesituation deteriorated quickly despite strong policy responses.This section reviews some ofthese c