海外之声丨疫情期间的供给瓶颈，劳动力市场和通胀

导读

新冠疫情爆发之初，供应链受到干扰，供给瓶颈迹象开始出现，并持续至今。其中，两个因素值得关注。其一，需求结构从服务转向产成品，而产成品相较于服务更加依赖于稳定的供应链。其二，供应链参与者的行为改变。面对可预见的供应短缺，供应链参与者提前订购囤积产成品加剧了供给瓶颈，即所谓的“长鞭效应”。

受供给瓶颈影响最严重的是大宗商品和货运业。由于出现供应短缺，大宗商品价格急剧上涨；随后，需求减弱，价格又大幅下跌。此外，由于需求强劲而供应乏力，亚洲与美国之间的运输成本迅速飙升。

供给瓶颈导致通胀率升高。欧美和拉丁美洲的通胀增幅明显，亚洲则变化不大。耐用品价格迅速上涨，服务品价格则增长缓慢。如果现在的通胀飙升使得工人工资和物价呈螺旋式上涨，那控制通胀率就比较困难了。也就是说，劳动力市场的变化是影响通胀走向的关键因素。

到2020年中，全球总的工作小时减少了10%到20%，主要体现在三方面：第一，失业率上升；第二，工人平均工作时间减少；第三，劳动参与率下降。这三种情况对经济复苏有不同程度的影响。

在美国，失业率上升的同时，也伴随着较高的岗位空缺率，这偏离了贝弗里奇曲线（在该曲线中，失业率与岗位空缺率成反比）。但在日本和欧元区，雇员-雇主关系维持稳定，则没有出现工作与技能不匹配的情况。这种劳动市场的不同表现会怎样影响工人的工资对未来的通胀走向有重要意义。

作者 | Hyun Song Shin，国际清算银行研究部主管

英文原文如下：

Bottlenecks, labour markets and inflation in the wake of the pandemic[1]

Hyun Song Shin

Economic Adviser and Head of Research, Bank for International Settlements

G20 International Seminar “Recover together, recover stronger”

Bali, 9 December 2021

It is a pleasure to join this session. Supply bottlenecks have grabbed all the headlines recently, but the theme chosen by the Indonesian G20 Presidency (“Recover together, recover stronger”) also prompts us to consider the longer-term structural changes brought about by the pandemic and the policy measures deployed in response. This is particularly apt now, as we look ahead to future inflation developments. I will argue that longer-term structural issues, especially in the labour market, are crucially important in plotting the course ahead. For my remarks, I will draw on two recent BIS Bulletins that have addressed recent labour market developments and supply bottlenecks[2].

Bottlenecks 

Let me start with a recap of where we stand on bottlenecks. Bottlenecks that started as supply disruptions due to the initial Covid shock in 2020 have this year morphed into something more persistent. Widespread reports of shortages have gone hand in hand with supply that appears to be running at full speed. Production of key manufactured inputs, such as semiconductors, now comfortably exceeds pre-pandemic levels. The same is true for many raw materials and freight volumes on key shipping routes. 

As well as supply, two additional factors are key in understanding where we are. The first is the shift in the composition of demand away from services to manufactured goods, which tend to be more dependent on the smooth functioning of supply chains than are services. The second is the endogenous behavioural changes that have given rise to so-called bullwhip effects[3], whereby supply chain participants react to perceived shortages by ordering more, ordering earlier and by hoarding inputs. This kind of reaction is prudent and rational when considered in isolation but can lead to aggregate outcomes that are ultimately self-defeating. 

A brief review of the main bottleneck-affected industries highlights the complex dynamics at work. The sharp swings in the prices of commodities such as lumber, iron ore and coal over the past year illustrate well how the current bottlenecks are not simply a uniform squeeze everywhere along the supply chain. Instead, we have seen commodities demand whipsaw as pressures emerged at different points in the supply chain, resulting in large price fluctuations (Graph 1, left-hand panel). For lumber, iron ore and coal, prices are well below peaks reached earlier in the year. Anecdotal evidence also points to localised gluts where the lack of storage capacity has depressed prices at key points.

The case of shipping freight could be considered a more straightforward example of strong demand interacting with constrained supply, especially for trade between Asia and the United States. But here too prices have exhibited large fluctuations (Graph 1, right-hand panel). Shipping costs for bulk commodities rose sharply, only to retrace much of the gains later (blue line). More recently, container shipping prices have also started to decline (red line). One sign that behavioural responses could be important comes from the pattern of inventories. The motor vehicle industry provides a good case study. Focusing on the United States, where we have the most comprehensive data, manufacturers’ inventories are actually up slightly since the start of the pandemic, even as retail inventories have fallen significantly (Graph 2, right-handpanel). These contrasting trends may be symptomatic of bullwhip effects, as supply chain participants adjust in an uncoordinated manner, creating moving targets for others.

The case of semiconductors is different again. Despite reports of disruptions at some specific factories, it is hard to find evidence of generalised contractions in supply. Changes in production relationships may also have exacerbated the behavioural responses, as firms that cancelled semiconductor orders early in the pandemic – automobile producers are the most cited example – found themselves at the back of the queue when demand returned. Indeed, available sales and billing data point to a strong increase in supply since the start of the pandemic (Graph 2, left-hand panel). 

When taken as a whole, the signs point to strong demand that has outpaced supply capacity that is growing, but not growing fast enough. However, the key question is how much of this stronger demand can be attributed to the bullwhip effect. To what extent will the behavioural responses that gave rise to bottlenecks work in reverse to clear up backlogs once supply chain problems begin to ease? Depending on the answer, we may find that supply bottlenecks may be resolved faster than currently feared, just as they have persisted longer than initially expected.

Inflation 

Let me now turn to inflation. Inflation has risen in many countries and bottlenecks have clearly played a significant role. But inflation has not risen uniformly (Graph 3, left-hand panel). Among advanced economies, the pickup in inflation has been most pronounced in the United States and the euro area, as well as in some emerging market economies, especially in Latin America. However, the increase in inflation has not been as evident in Asia. In Japan, consumer prices have barely grown over the past year. And inflation remains low in China. 

As well as the differences across countries, the differences across product categories are also notable. Durable goods prices have risen especially sharply since the start of the year (Graph 3, centre panel), breaking the trend of low or negative price changes over much of the past few decades. The rise in the price of services has been much more modest, although it has picked up in recent months in some economies (Graph 3, right-hand panel), although not in Japan.

We should be wary of disregarding inconvenient price increases. Nonetheless, it is worth emphasizing how unusual the recent rise in durable goods prices is when set against the experience of well-established long-term price trends. As the left-hand panel of Graph 4 shows, the rise in durable goods prices in the United States in recent months is much larger than any other seen over at least the past two decades. Although the rise in durables prices in other countries has not been quite as large, a prolonged period in which durable goods prices rise faster than those of services represents a sharp break from the trend over recent decades (right-hand panel).

However, there is something of a race against time. If the current inflation surge feeds a wage-price spiral reinforced by a possible unmooring of inflation expectations, bringing inflation under control will be a much more difficult task. That’s why the labour market will be the key determinant of the course of inflation. This is where I would like to turn now.

Labour markets

Labour market conditions across countries have shared some similarities as well as some striking differences. 

The overall trajectory in total hours worked has been quite similar since the start of the pandemic. As you see in Graph 5, they fell by 10–20% in the middle of 2020. Following these sharp declines, total hours worked have rebounded much more quickly than after recent recessions, although they remain 2–5% below their pre-pandemic levels. 

However, these similar declines in hours came about in strikingly different ways. There are three ways in which hours worked could fall. The first is an increase in unemployment. The second is a drop in average hours worked per employee. The third is a decline in the participation rate of the labour force. The respective roles of these three factors differ widely across economies. In the United States, the initial decline largely reflected a shift from employment into unemployment and, to a lesser extent, a move of workers out of the labour force. Although much of the rise in unemployment has been reversed, participation rates remain substantially below where they were before the pandemic, particularly for older workers.

In most other advanced economies, the decline in hours worked reflected mainly a fall in average hours worked per employee, due largely to furlough schemes that kept worker-firm relationships intact.

In many emerging market economies, large numbers of workers left the labour force. Some of this could reflect a shift into the informal sector, although informal work was also hard hit by the pandemic.

However, I would like to draw your attention to one key observation. The differences in the way that hours worked declined have also influenced the shape of the recovery. Specifically, in countries that saw large rises in unemployment at the height of the crisis (such as the United States), the labour market has been slower to spring back to its pre-pandemic setting. 

One way to see this is through the Beveridge curve, which shows the relationship between the unemployment rate and job vacancies. Normally, changes in economic activity would show up as shifts along the Beveridge curve, with stronger labour demand going hand in hand with lower unemployment and higher job vacancies. However, in the United States, the Beveridge curve has shifted out since the start of the pandemic (Graph 6, top left-hand panel). This means that many more job openings are on offer than previously for the same level of unemployment. The UK Beveridge curve has also started to drift out in recent months. 

Importantly, no such rightward shift in the Beveridge curve is evident in jurisdictions such as Japan and the euro area, where worker-firm relationships remained largely intact.

The conventional interpretation of the rightward shift in the Beveridge curve is a labour market mismatch between jobs and skills – say, due to the reallocation from the real estate sector after the Great Financial Crisis. To some extent, that may be true at present. At the same time, it is worth noting that job vacancies have risen most in some of the service industries that saw the largest job losses since the start of the pandemic. So, a simple sectoral reallocation story seems inadequate. 

In any case, the contrast between the Beveridge curves across economies is very striking. In Japan and the euro area, there is no evidence of any deterioration of the jobs-skills match. Preserving the employment relationship appears to have kept the economy on a path where the recovery is closer to bringing the economy to its pre-pandemic state, at least in terms of the Beveridge curve relationship.

Getting to the bottom of the reasons for this contrast in labour market outcomes is an important task ahead for policy makers. Perhaps one way to approach this question is to start from the premise that firms and workers are part of the intricate web of relationships in the economy with relationship-specific capital that acts as the “glue” for the economy as a whole. Barry Eichengreen has written eloquently on this issue[4].

The ties that bind all of us as colleagues, neighbours, workers and employers arguably go beyond the transactional nature of the weekly payslip. Once these relationships are broken, attempts to put the pieces back together will not be able to draw on the same reservoir of relationship-specific capital that was in place previously. A more systematic analysis that draws on more than simple, atomistic optimisation models beckons. However, the recent upward drift in the UK Beveridge curve from the middle of 2021 suggests that any simplistic explanation will be found wanting, as the UK had also implemented furlough schemes similar to euro area economies. 

More immediately, how these differences in labour market functioning will translate into differences in wage growth will be important for inflation developments. Unfortunately, the underlying pace of wage growth is particularly hard to read at the moment because of pandemic-related shifts in the composition of employment and the effect of furlough schemes[5].

Nevertheless, across most advanced economies, average wage growth since the start of the pandemic looks to have been in line with its pre-pandemic trends, or a little below, although this data will not capture the effect of recent inflation increases on wage negotiations. It is notable, however, that in the United States, where labour market changes are most apparent, wage growth has picked up despite labour market conditions that appear weaker than before the pandemic (Graph 7). 

The theme chosen by the Indonesian G20 Presidency (“Recover together, recover stronger”) is especially apt for today’s discussion because it prompts us to consider both short-term and longer-term structural changes. The supply bottlenecks are a short-term issue, especially if the bullwhip effect goes into reverse. In this case, just as supply bottlenecks have persisted longer than expected, they may be resolved faster than currently feared.

However, the key to gauging where global inflation is headed is in the labour market, and whether the reduced efficiency of matches exhibited in the Beveridge curves of some economies translates into a more sustained wage-price spiral. In this respect, longer-term structural issues are more important in understanding the current state of the global economy, especially when we consider future inflation developments.

Reference

1 The views expressed here are mine, and not necessarily those of the BIS. I am grateful to Daniel Rees and Phurichai Rungcharoenkitkul for their efforts in putting together the analysis for this speech; to Nicolas Lemercier and Alessandro Barbera for research assistance and to Egon Zakrajšek and helpful comments.

2 See F Boissay, E Kohlscheen, R Moessner and D Rees, “Labour markets and inflation in the wake of the pandemic”, BIS Bulletin, no 47, October 2021 and D Rees and P Rungcharoenkitkul, “Bottlenecks: causes and macroeconomic implications”, BIS Bulletin, no 48, November 2021.

3 https://en.wikipedia.org/wiki/Bullwhip_effect

4 Barry Eichengreen, “The human capital costs of the crisis”, Project Syndicate, April 2020.

5 See Bank of England, “How strong is pay growth?”, Bank Overground, 31 August 2021, and G Koester and E Hahn, “Developments in compensation per hour and per employee since the start of the COVID-19 pandemic”, ECB Bulletin, no 8, August 2020.

编译  岑晓敏

编辑  刘嘉璐

来源  BIS

责编  李锦璇、蒋旭

监制  商倩、董熙君

点击查看近期热文

欢迎加入群聊

为了增进与粉丝们的互动，IMI财经观察建立了微信交流群，欢迎大家参与。

入群方法：加群主为微信好友（微信号：imi605）,添加时备注个人姓名（实名认证）、单位、职务等信息，经群主审核后，即可被拉进群。

欢迎读者朋友多多留言与我们交流互动，留言可换奖品：每月累积留言点赞数最多的读者将得到我们寄送的最新研究成果一份。

关于我们

中国人民大学国际货币研究所（IMI）成立于2009年12月20日，是专注于货币金融理论、政策与战略研究的非营利性学术研究机构和新型专业智库。研究所聘请了来自国内外科研院所、政府部门或金融机构的90余位著名专家学者担任顾问委员、学术委员和国际委员，80余位中青年专家担任研究员。

研究所长期聚焦国际金融、货币银行、宏观经济、金融监管、金融科技、地方金融等领域，定期举办国际货币论坛、货币金融（青年）圆桌会议、大金融思想沙龙、麦金农大讲坛、陶湘国际金融讲堂、IMF经济展望报告发布会、金融科技公开课等高层次系列论坛或讲座，形成了《人民币国际化报告》《天府金融指数报告》《金融机构国际化报告》《宏观经济月度分析报告》等一大批具有重要理论和政策影响力的学术成果。

2018年，研究所荣获中国人民大学优秀院属研究机构奖，在182家参评机构中排名第一；在《智库大数据报告（2018）》中获评A等级，在参评的1065个中国智库中排名前5%。2019年，入选智库头条号指数（前50名），成为第一象限28家智库之一。

国际货币网：http://www.imi.ruc.edu.cn

微信号：IMI财经观察

（点击识别下方二维码关注我们）

理事单位申请、

学术研究和会议合作

联系方式:  

010-62516755 

[e-mail]

只分享最有价值的财经视点

We only share the most valuable financial insights.