第六季第40集：解读全球建筑业

对于许多关键建设投入品，低收入国家的价格更高，这非常引人注目，因为这些国家正是需要建设基础设施的地方。

For a lot of the key construction inputs, prices were higher in low income countries, which is very striking because these are the countries that are going to build up their infrastructure.

欢迎收听Vox Dev Talks，我是蒂姆·菲利普斯。政策制定者和政客们热衷于谈论投资国内基础设施——它将如何促进经济增长、创造就业、增强企业实力，以及他们的政策将如何推动这一进程。要实现这些目标，投资基础设施的最佳方式是什么？都柏林圣三一学院的玛蒂娜·基希贝格是这方面的专家。

Welcome to Vox Dev Talks. My name is Tim Phillips. Policymakers and politicians love to talk about investing in domestic infrastructure, why it's gonna grow the economy, also how it's gonna provide jobs and strengthen firms, and how their policies are going to encourage it. What's the best way to invest in infrastructure to realize these goals? Martina Kirch berger of Trinity College Dublin is an expert on this topic.

现在她与我连线。玛蒂娜，欢迎来到Vox Dev Talks。

She joins me now. Martina, welcome to Vox Dev Talks.

谢谢，蒂姆。很高兴来到这里。

Thanks, Tim. It's great to be here.

玛蒂娜，基础设施对增长有多重要？我想先问，我们所说的基础设施指的是什么？

Martina, how important is infrastructure for growth? I suppose, what do we mean by infrastructure?

让我先从定义基础设施开始。我通常指的是经济的核心动脉，比如道路、桥梁、高速公路、机场，以及支持性基础设施如电力、供水、排污系统。此外还有社会基础设施，如用作医院、学校的建筑物。

Let me start by defining what I mean by infrastructure. I typically have in mind the core arteries of an economy. So think of the roads, the bridges, the highways, airports, and supportive infrastructure, electricity, water, sewage systems. And then there is social infrastructure. Buildings that are used as hospitals, as schools.

所有这些都由建筑行业建造。基础设施对增长有多重要？这在许多方面是个更难回答的问题。擅长提供良好基础设施的国家在许多事情上做法不同。因此从识别角度来看，很难说一个国家增长更快是因为基础设施，还是因为其他以不同方式实施的政策。

And all of these are built by the construction sector. How important infrastructure is for growth? That's a harder question in many ways. Countries that are very good at providing good infrastructure do a lot of things differently. And so from an identification perspective, it's very hard to say country grew faster because of the infrastructure, because of all these other policies that are implemented in different ways.

根据过去几十年进行的众多细致微观评估研究，我们了解到，在考虑具体基础设施项目时，其效果高度依赖具体情境，且与基础设施类型密切相关。这是乡村道路还是城市交通系统？学界已投入大量严谨工作来识别这些微观层面的影响。但总体而言，自赫希曼时代起，基础设施与经济其他部门的互补性就已被广泛认知。基础设施和建筑行业商品具有特殊性，它们往往是其他生产流程的投入要素。

What we know from a host of studies that have been done over the past decades, very careful micro evaluations is that when we think about specific infrastructure projects, it's very context specific, and it depends on the type of infrastructure we're thinking about. Is this a rural road versus an urban transit system? I think there's very careful work that has been going into identifying these micro level effects. But I think at an aggregate level, the complementarities between infrastructure and other parts of the economy have long been recognised going back to Hirschman. And infrastructure and construction sector goods are very particular in the sense that there are inputs into other types of production processes.

因此它们处于核心地位。围绕它们会产生大量互补性的私人和公共投资。可以说它们是激活各类其他经济部门活动的关键。但与此同时，它们也是资源浪费的源头。历史上不乏修建通往荒芜之地的道路这样的案例。

So they're very central. There are a lot of complementary private and public investments. And so they're sort of key to enable economic activity in all kinds of other sectors. At the same time, they're also a source of wasteful spending. And so historically, we also know that there are roads built to nowhere.

所以我认为确保资金合理使用至关重要。

And so I think getting the spending right is really important.

好的，让我问个或许更简单、更偏向国民经济核算的问题。这些基础设施的建设，即建筑行业，在总投资中占比多少？

Okay, let me ask you maybe probably a slightly easier question, more national accounting question. The construction of this infrastructure, the construction sector, what proportion of investment is that?

建筑业约占总投资——即固定资本形成总额——的一半左右。建筑业内部又可划分为三大类：住宅建设、非住宅建设以及土木工程。平均而言，这三者的比例约为40:30:30。因此基础设施部分通常属于土木工程，是建筑业的一个子集。

The construction sector accounts for about half of investment, so gross fixed capital formation. Then within construction, we have three big groups. We have residential construction, non residential and then civil engineering works. And on average, that split is about forty-thirty-thirty. So the infrastructure piece would be typically in the civil works a fraction of construction.

但这个我们通常很少深入研究的领域，实际上占据了投资的极大比重。

But it's a huge part of investment that we typically don't spend a lot of time studying.

特别是考虑到你刚才阐述的规模。政策制定者常将建筑行业视为工具性部门，用以实现其他政策目标。他们具体考虑的是哪些政策目标呢？

Especially considering the scale of it that you just explained to me. Policymakers, they think about the construction sector often as instrumental, a way to achieve their other policy goals. What policy goals are they thinking of?

他们心中有几个目标。首先，当他们建设新的基础设施时，会创造大量建筑就业岗位。这些新增的工作机会被视为有价值的贡献。同时，你还可以计算为特定建设项目提供材料的上游企业所创造的就业。其次，以某种基础设施投资为例，比如高速列车，就会直接催生相关工作岗位。

They have a number of goals in mind. First, when they build a new piece of infrastructure, there is sort of the bodies employed construction. So the jobs that are being created are considered something that's valuable. And you could also count the employment created in upstream firms that supply to a particular construction project. Second, if you think about a particular type of infrastructure investment, for example, an express train, there are direct jobs that are being then created.

由于这类基础设施的存在，需要配备列车员。这里可能会直接创造就业机会。例如在站台上，可能会有小商铺，摊贩可以设立小型摊位。基础设施需要维护保养。当然还有我们认为通过基础设施间接促成的就业机会。

Because of this infrastructure, there are conductors that are needed. There might be employment opportunities that are directly created there. For example, on platforms, you might have little shops, so vendors can set up small stalls. You need to maintain the infrastructure. And then, of course, there are the jobs that we think are enabled through the infrastructure.

因此运输成本的降低能促进劳动者与企业之间更优匹配，从而提升生产效率。我认为政策制定者在投资基础设施时，通常希望实现这些多重目标。

So a decrease in transport costs enables better matching between workers and firms, and that improves productivity. So there's all these multiple objectives that I think often policymakers have in mind to reach when they invest in infrastructure.

我们将讨论你围绕这个主题完成的几项研究。这些非常有趣的研究成果，是我在发展经济学领域前所未见的。我们稍后会详细讨论。首先，Martina，这是《Look Steph访谈》首次探讨水泥话题，因为你对此有专门研究。那么第一个问题：每个国家都有自己的水泥产业吗？

We're going to talk about several pieces of work that you've done around this. Really interesting pieces of work and the sorts of things that I don't think I've ever seen in development economics before. So we'll talk about those in a second. So first of all, Martina, for the first time ever on Look Steph talks, we're gonna talk about cement because you have done a piece of work on this. Now, does every country, first of all, does every country have its own cement industry?

要回答这个问题，需要区分两种主要工厂类型——我保证不会过多深入讲解具体工艺细节...

To answer this question, it's useful to distinguish between two main types of plants, and I promise I won't go much further into how we-

我想了解具体细节。

I want to know the details.

水泥厂分为粉磨站和全能厂两种。全能厂将石灰石与黏土混合，放入约2000度高温的窑炉中煅烧，产出熟料。最后阶段将这些熟料研磨成灰色粉末，这就是水泥。全能厂能完成整个生产过程。

So there are grinding and then there are integrated cement plants. And so an integrated cement plant takes limestone and then adds some clay and this gets put into a very hot oven, a kiln at about 2,000 degrees. And out of that comes clinker. And so in the final stage that clinker gets ground. And so an integrated cement plant does all of these things and produces cement, which is the gray powder.

研磨厂仅完成最后一道工序。它们将熟料研磨成水泥。因此这两类工厂的技术差异很大。这种区分很重要，因为当地石灰岩的储量决定了各国拥有何种类型的工厂。许多国家会根据经济地理、水泥需求分布及地质资源情况，同时拥有这两类工厂。

Grinding plants only do that last step. They take clinker and they ground it into cement. So the technologies of these two plants are very different. The distinction is relevant because the availability of local limestone determines what type of plants countries have. And so many countries would have a combination of those two plants depending on the economic geography and where cement is needed and their geological resources.

有些国家可能只拥有研磨厂。例如科特迪瓦因缺乏天然石灰岩，仅有研磨厂。石灰岩不适合跨国运输，因其重量大而价值相对较低。孟加拉国主要以研磨厂为主，兼有少量综合型工厂。还有些国家根本没有水泥厂。

And then some countries would only have, for example, grinding plants. So Cote D'Ivoire only has grinding plants to lack of natural limestone. Limestone you wouldn't ship across countries because it's just very heavy and has low value compared to its weight. Bangladesh has mainly grinding plants with a few integrated plants. And then there are some countries that don't have any plants.

比如佛得角目前没有任何运营中的水泥厂。值得注意的是，全球水泥消费量中只有5%参与国际贸易。从总量来看，它仍属于非贸易商品。尽管对某些国家而言进口水泥很重要，但贸易比例仅占5%。

So for example, Capverde doesn't have any operating plants at the moment. And so if we think about cement trade, it's worthwhile noting that only 5% of global consumption is traded. So it's still largely a non tradable good at an aggregate level. While it's certainly important for certain countries to import cement, only 5% is traded.

我记得搬运时感觉它确实很重

Because it's pretty heavy stuff I remember from carrying It's it

而且它是水硬性粘结剂。想象一下遇到潮湿环境，运输就更加复杂。当然可以运输，但所有这些因素共同导致水泥贸易量相当有限。

and it's a hydraulic binder. So you imagine with humidity, it's also not straightforward to transport. Of course, you can do it, but I think all of these factors play into the fact that it's just trade is fairly small.

这很有教育意义。你们收集了水泥价格数据。覆盖了哪些国家？具体是如何收集的？

This is an education. You collected data on these cement prices. For which countries did you collect the data? How did you collect this data?

我们采用了国际比较项目（ICP）的数据。ICP在2011年改革了建筑业购买力平价的计算方法，从原先对特定建筑构件定价的产出法，转向了更注重明确建材数量与质量标准的投入法。我长期研究各国建筑成本差异，但由于产品异质性太强，跨国单位成本比较极为困难。

So we used data from the International Comparison Project that's housed by the ICP. And the ICP did a change in 2011 in the way they collect prices to compute the construction sector PPP. They had an output based approach where they would price certain construction components and they moved to an input based approach, which was great because they were very careful in specifying the quantity and quality of key construction sector inputs. And I've looked at this for a long time thinking about the cost of construction across countries, and it's just very difficult to compare because products are so heterogeneous. It's just very hard to compare unit costs across countries.

但对于原材料，你可以这样做。例如，水泥被定义为一吨波特兰水泥，这是一个精确的定义。对于混凝土，它是沙子、水泥和骨料的混合物，比例、骨料尺寸以及预拌混凝土的抗压强度都有明确规定。因此，当我们比较这些价格时，实际上是在比较相似的产品。当然，我们仍不能完全排除剩余的质量差异，但我们需要这样才能开始进行价格比较。

But for inputs, you can do this. For example, cement is defined as a tonne of Portland cement, which is a precise definition. For concrete, it's the mix between sand, cement and aggregates, and the proportions are specified, as well as the size of the aggregate and the compressive strength of that ready mixed concrete. And so when we compare these prices, we really compare similar products. Surely, there'll be still we can't rule out that there are any quality differences left, but we need this in order to be able to start comparing prices.

于是我们开始研究这些价格。获得这些数据让我们非常兴奋，同时也发现了巨大差异。我们看到许多关键建筑原材料在低收入国家的价格更高，这非常引人注目，因为这些国家将大规模建设基础设施。如果他们为这些关键原材料支付过高价格，那么这就是一个值得我们思考的重要瓶颈。

So we started looking at these prices. We were very excited to get access to these data and we saw large differences. We saw that for a lot of the key construction inputs, prices were higher in low income countries, which is very striking because these are the countries that are going to build up their infrastructure massively. And so if they pay very high prices for these key inputs, then that's an important bottleneck that we should think about.

认为价格应该更高完全违反直觉。我们稍后会分析原因。你们能获取哪些国家的数据？

And it's completely counterintuitive to think that prices should be higher. We'll work out why that is a little bit later. Which countries were you able to get the data on?

我们有一个庞大的数据集，涵盖100多个国家。这确实让我们能在全球层面很好地了解这些价格的分布情况。

We have a large set. We have over 100 countries. This is really we get a good distribution of these prices at a global level.

这些国家的水泥价格比美国等高收入国家或我所在的英国更贵。

Cement in these countries, it's more expensive than it is in a high income country in The United States or in The UK where I am.

是的，比如2011年的价格数据显示，撒哈拉以南非洲地区的平均价格为2.6美元，而北美为190美元，东亚及太平洋地区为115美元。但即使在撒哈拉以南非洲内部也存在巨大差异。我们查看了2011年价格最高的10个国家，其中9个位于撒哈拉以南非洲。例如，中非共和国的水泥价格接近每吨500美元，而美国仅为190美元。即便到2017年价格有所下降，仍比美国高出两倍。若考虑购买力平价（计入当地物价水平），则高达4.5倍。这种价格与发展水平之间的负相关关系持续存在。

Yeah, so if we look at 2011 prices, for example, in Sub Saharan Africa, the average price was $2.60 US dollars compared to North America where it was 190 and East Asia and The Pacific 115 US dollars But there's large differences even within, say, Sub Saharan Africa. So we looked at the 10 most expensive countries in twenty eleven, and nine of these 10 were in Sub Saharan Africa. And so, for example, in the Central African Republic price per tonne of cement was close to US500 dollars compared to the prices in The US190 dollars Even in 2017, prices have come down, but still the price is twice as high as it is in The US. And in PPP terms, so taking local price levels into account 4.5 times. And this negative relationship between prices and development persists.

因此，对于许多关键原材料，最贫困国家的价格反而更高。

So prices are higher in the poorest countries for many of these key inputs.

现在听到这个问题的所有人心中都有同一个疑问：为什么价格如此昂贵？

The question on the mind of absolutely everyone that's hearing this now, why so expensive?

我们将投入成本作为解释因素之一进行了研究。价格高昂可能是因为原材料成本高，也可能是因为需求低于最小有效规模，导致这些工厂无法实现高效生产规模。技术可能落后，或者利润加成过高。这促使我们开始思考这些国家的市场结构。

So we looked at the cost of inputs as one of the explanations. It could be expensive because the inputs are expensive. It could be because demand is below minimum efficient scale, so these plants just can't produce at a scale that's efficient. Technology might be low or markups could be high. And this is when we started thinking about what the market structure looks in these countries.

我们开始将全球水泥厂和企业的分布数据数字化，以了解竞争状况。论文中发现，利用石灰石矿床分布带来的外生变量（影响企业数量）显示，企业数量较少的国家水泥价格显著更高。

And we started digitising data on the global distribution of cement plants and firms to get a sense of what competition looks like. And what we find in the paper is that using the distribution of limestone deposits for some exogenous variation that affects the number of firms, prices are significantly higher in countries where there are fewer firms.

这想必会影响——因为水泥是重要原材料——这些国家的基础设施建设成本吧？

Does this affect I imagine it must do because cement is a very important input. This affects the price of the infrastructure in these countries.

水泥支出在建筑行业占比很大，尤其是最贫困国家。但在富裕国家并非如此，那里水泥使用量大且价格低廉。我们计算的中位数是8%，第75百分位数为17%。通过ICP建筑构件数据也发现，水泥价格高时，混凝土构件的成本更高。

While cement is a substantial fraction of construction sector expenditures, particularly for the poorest countries. It's not the case for the rich countries where it's used abundantly and it's cheap. The median that we computed was 8%. And at the seventy fifth percentile, it's 17%. So we've also looked at ICP data on construction components, and we see that construction components that use concrete are more expensive when prices are high.

这种价格传导效应非常明显。关键在于评估基础设施成本时，要考虑生产网络。水泥作为许多产品的核心投入品价格昂贵，其动态影响将远超静态影响。

So there seems to be this very clear pass through. What's really important is that when we think about the cost of this infrastructure, to keep in mind the production network. So here, an input into the production of lots of other goods that's very central in the production network is very expensive. And so the dynamic implications are going to be much larger than just the static implications.

既然存在明显的超额利润，为什么这些市场没有出现更多竞争者？

Seeing as there are these excess profits clearly being made, why is there not more competition in these markets?

自2011年我们开始研究这些数据以来，已有企业进入该市场且价格有所下降。但现行政策限制了竞争。例如，严格的产品标准、进口禁令或限制，以及石灰石矿床勘探和生产许可证发放过程缺乏透明度。我们在论文和我的研究中并未区分这些现象的主要驱动因素，但相关文献在讨论该行业竞争状况时都会提及这些。

Since we started looking at the data in 2011, there has been entry in this market and prices have come down. But there are policies in place that limit competition. So for example, restrictive product standards, import bans or restrictions on import, also lack of transparency in how licences for the exploration of limestone deposits and production is awarded. That's not something we distinguish in the paper and the work that I've done, what the main drivers are for this, but this is what's being mentioned when you read literature on the state of competition in this industry.

我认为政策解决方案应该是扭转这些限制性做法——根据我们之前的讨论，这对经济可能非常有利。这样理解对吗？

I would have thought the policy remedies to fix this, which would probably, from what we've discussed before, be very positive for the economy, would be to reverse some of these restrictive practices. Is that right?

是的，而且一些国家的竞争监管机构非常活跃，这需要它们发挥积极作用。但重要的是要记住这是个难题。水泥或混凝土行业的卡特尔在全球范围内、包括高收入国家都屡见不鲜。2019年英国就发现并查处了一个卡特尔组织。

Yeah, and there are competition authorities very active in some of these countries, and it calls for an active role of those. I think what's still important to keep in mind is this is a hard problem. Cement cartels or cartels in concrete are being discovered across the world in countries at very high levels of income. It's just a difficult problem to solve. 2019 in The UK, for example, you had discovery and finds of a cartoon.

所以我们应该牢记这一点。

So we should keep this in mind.

明白了。其次，你们还研究了建筑业与就业之间的联系。人们通常认为建筑业是正规部门就业的来源。这在发展中国家尤其显著吗？

Okay. So secondly, you've also investigated the links between construction and labor jobs. Now, it's always assumed that construction is a source of jobs in the formal sector. Is this particularly relevant in developing countries?

是的，因为创造就业的压力巨大。2018至2035年间非洲劳动年龄人口将增长40%。政策制定者面临创造就业的压力，而建筑业——特别是大型项目——具有创造正规部门就业的潜力，这类工作被认为更具价值。政策制定者认为投资这些项目能创造优质岗位，使工人有机会提升职业技能，获得人力资本，进而实现职业发展。

Yeah, because there's huge pressure to create jobs. So there's a 40% increase in the working age population Africa between twenty eighteen and 2,000 and 35. So policymakers are faced with this pressure to create employment. And the construction sector, particularly construction of large projects, has this potential of create jobs in the formal sector, which are believed to be jobs that are valued by people. Policymakers have this idea that when they invest in these projects, they create valuable jobs that allow workers potentially to climb the job ladder, acquire human capital, and then move on to a different trajectory.

这是许多人持有的先验认知。但我们是否有强有力的实证证据表明建筑业确实创造了这类就业？

That's a prior belief that many people would share. But do we have strong empirical evidence that this is true, that construction does create these jobs?

我认为我们还没有。这正是我们开展这项工作的部分动机。我们与政策制定者交流时，他们也对此发现感到兴奋，因为他们原先认为我们在这里创造了非常优质的就业机会，这对人们有益。而且我们甚至在其他情境下也没有证据表明，这些工作能在建筑工程完工后继续为工人创造机会，他们可以将获得的人力资本运用到不同工作中，从而逐步提升收入。

I don't think we have. And this was part of the motivation for why we started this work. We were talking with policymakers, and they were excited to find this out too, because their prior was that we're creating very good jobs here, and that benefits people. And we don't have evidence from even other contexts that these jobs create opportunities that workers beyond the completion of a construction project, they can take the human capital they've acquired to different jobs, and that allows them to grow their earnings over time.

所以在你的论文《构建非洲城市》中，你们设计了一个随机对照试验来验证这一点，即建筑行业正式工作机会的影响。你们在哪里进行的？针对什么建筑项目？

So in your paper, it's called Constructing Africa's Cities, you created an RCT to test this, the effect of a formal job offer in the construction sector. Where did you do this? For what construction project?

该项目是塞内加尔达喀尔正在建设的快速公交系统。现在已投入运营，效果非常好。我们与政府、捐助方和建筑公司合作，研究参与该基础设施建设的工人后续发展。我们设计的随机对照试验中，工人可以在社区或公司总部不同名单上登记，然后随机分配工作机会。我们在基线时进行访谈，之后进行12轮跟踪，最后进行终期调查。

The construction project was the bus rapid transit system that was being built in Dakar in Senegal. It's up and running now and fantastic. And we collaborated with the government, donors and the construction firm to understand what happens to workers who are employed in the construction of this infrastructure. So we designed an RCT where workers could sign up at different lists, either in the communities or at the company's headquarters, and then we randomised who would get a job offer. And we interview people at baseline and then for 12 rounds, and then we conduct an end line.

明白了。那些幸运获得工作的人具体得到什么待遇？

Right. And what offer did I suppose the lucky recipients get?

他们获得的是三个月正式合同，来自一家跨国企业，包含基本工资、交通补贴和加班费。理论上合同可续签，但仅限于这三个月期限。

So they get an offer for a three month contract. It's a formal sector contract with a multinational firm that has a base salary, their transport premium, and then there's paid overtime. The contract is in theory renewable, but it's limited to this three month contract.

那么获得合同的群体短期内有什么直接变化？

And so what was the immediate effect for the people who did get a contract?

在前三个月我们观察到直接效应：人们更可能处于就业状态。当我们联系他们时，过去七天里他们的工作日数更长、工作天数更多、工时也增加了。但在收入方面，例如，我们并未看到显著影响。

We see immediate effects in the first three months that people are more likely to be working. In the previous seven days, when we called them, they're working longer days, more days, and they're working more hours. But we don't see any effect on earning, for example.

有意思。为什么你认为对收入没有影响？

Interesting. Why do you not see an effect on earning?

可能是因为他们的外部选择提供的薪资与这份工作的报酬相差不大。

It could be because their outside option is not giving them a very different salary from what they get paid in this work.

好的，但我们确实看到了对他们就业可能性的影响。这种影响在三个月后还会持续吗？

Okay, but we did see those effects on the likelihood that they will be working. Does this persist after the three months?

在终点阶段我们看到持续性非常微弱。这些工人在许多结果指标上与对照组并无差异。我们仍在研究根据教育程度或匹配质量的异质性——比如他们之前是否有建筑行业经验。但目前的证据并不表明这些工人处于显著不同的发展轨道上。在社会保障或医疗保险方面我们看到了些微暗示性证据，但需要更深入研究过渡情况和异质性才能获得完整图景。

So we see very little persistence at end line. These workers in many outcomes don't look different from the control group. And we're still looking into heterogeneity according to education or match quality based on whether they had experience in the construction sector before. But the evidence so far does not suggest that these workers are in a substantially different trajectory. We see a little bit of suggestive evidence in terms of having social security or health insurance, but we need to look more into transitions and to heterogeneity to get a fuller picture.

这某种程度上削弱了先前认为建筑业正在创造就业机会的信念。为什么这没有奏效？政策能做些什么来为这些工人创造更多机会吗？

Now, somewhat undermines this prior belief that construction is creating these job opportunities. Why is this not working? Can policy do anything to make this more of an opportunity for these workers?

目前我只能推测。我认为部分原因可能是薪资相对较低。与这些工人通常从事的工作相比，这些合同的期限也相当短暂——很大比例工人从事日结劳动，三个月对他们来说已经算长期了。但就职业晋升可能性而言，这种固定期限合同相对较短。这是这些行业的特点，许多合同期限都与工程建设周期挂钩。

At the moment, I can only speculate. I think part of it is that the salaries were fairly low as a potential explanation. The duration of these contracts was also fairly short in comparison to what these workers are engaged in, where a large proportion engages in daily labor, where they get paid at a daily rate, three months is long. But in terms of the possibility to progress within a firm, having this fixed term duration is relatively short. And that's a characteristic of the sectors that a lot of these contracts would be for the duration of the construction.

举例来说，如果我们将其与空间固定的制造业企业比较，当工人持有短期合同甚至以建设周期为期限时，企业投资和内部晋升的激励就不同了。语言障碍也可能存在。我们在世界其他地区也看到过证据表明，语言障碍会阻碍人力资本溢出。在另一个项目中，我们与建筑公司合作实施了由心理学家开发的非认知技能培训，特别是责任心培养。

So if we compare this, for example, with a manufacturing firm that's sort of fixed in space, the incentives to invest and to progress within a firm are different here when workers hold a contract at short term or even for the construction duration. There could also be language barriers. And we see this in other parts of the world. There's evidence that there are language barriers that prevent human capital spillovers. What we do in another project is we also work with construction firm and we implemented a training that we developed with psychologists in non cognitive skills and particularly conscientiousness.

我们在此看到了对受训工人收入和留职率的显著影响。因此我认为，政策制定者需要思考如何通过配套干预措施来提升这些基础设施建设所创造岗位的价值。

And there we see big effects on earnings and retention of workers that were trained. So I think for policymakers, it is important to think about what complementary interventions could be implemented to increase the value of these jobs that are created in the construction of the infrastructure.

在我们建筑业考察之旅的第三站，我们来谈谈企业。从本质上看，建造房屋、道路或桥梁通常被视为本土活动，因为你无法建好后再搬运到位。在中低收入国家也是如此吗？

So in the third stop of our tour through the construction industry here, we talk about firms. By their nature, we assume that constructing buildings or roads or bridges is a domestic activity because you can't construct them and then move them into place. In low middle income countries, is it?

建筑业确实通常被视为典型的本土或非贸易行业，这也是我接受的认知。但在低收入国家实地观察后，情况截然不同。你会看到跨国企业承建各类基础设施和大型建筑，它们真正实现了全球化运作。

It is true that construction typically is considered as a classical domestic or non tradable industry. And this is how I was trained too. I think the picture is very different when you spend time in low income countries. You can see global firms constructing all kinds of infrastructure, large buildings. They're truly global.

这些企业来自世界各地。当你实地考察时会对此有清晰认知。但目前还缺乏系统性量化这些流动及其影响的可靠证据。

They come from everywhere. And so I think this is something that you see very clearly when you spend time there. But I don't think there is a lot of good evidence that quantifies these flows at a systematic level or thinks about the implications.

你正在着手整理这些证据。关于承接建设项目的企业区位分布，你的研究揭示了哪些发现？

Now you're working to compile that evidence. What do we know from your work about location of firms that are delivering these construction projects?

我们知之甚少，因为这属于测量难题。在早前的书章研究中，我分析了世界银行2018年2月前的招标合同，追踪土木工程承包商的注册地。以撒哈拉以南非洲为例，200万美元以上合同中约50%由区域外注册企业承建——这些企业不仅非本土，甚至不在非洲地区。5000万美元以上的合同比例更高达75%。

We know very little because it's a very hard measurement problem. In a previous book chapter, I looked at prior review contracts by the World Bank in a period of 02/2018, and looked at where the firms that carry out civil work contracts are registered. And you can see for Sub Saharan Africa, for example, for contracts that are higher than US2 million dollars about 50% of these contracts are carried out by firms that are registered outside the region. That doesn't mean they're domestic, they're outside Sub Saharan Africa. For contracts above US50 million dollars that's 75% of contracts carried out by firms registered outside the region.

新项目中我们试图系统研究这些国际承包商的身份及其业务版图。通过数字化全球225强承包商的历史业务数据，我们发现：2003年2月时前十强均来自高收入国家，而2023年其中40%已是中国企业。同期中国企业的营收份额增长五倍，美国份额则降至三分之一。这个现象与美国建筑业生产率增长乏力的研究发现形成了有趣呼应。

In a new project, we want to get systematic evidence on who these global contractors are and where they work. We digitized lists on the top two twenty five global contractors and where they worked over a long period of time. And what we see is that, for example, in 02/2003, the top 10 firms are all from high income countries, while in 2023, 40% of these top 10 firms are from China. China's revenue share increased by five times over those periods, whereas The US's share, for example, is a third. This is interesting because it also relates to some of the findings that we have from The US construction industry and the weak productivity growth.

因此，不仅国内生产率增长疲软，而且在我们考察的时期内，其全球重要性也确实下降了。再次可见，低收入国家的参与度极低——这份承包商名单上没有一家来自低收入国家。这表明区域一体化程度非常有限。

So not only is there weak productivity growth domestically, but also the importance at a global level has really decreased over the period that we're looking at. And again, we see there's very little participation of low income countries. So there are no contractors on that list that come from low income countries. So very little regional integration.

为什么中国变得如此活跃，特别是在那些低收入国家？

Why has China become so much more active, especially in those low income countries?

他们拥有丰富经验，能高效开展基础设施建设投标，且具备多重竞争优势。可能还存在其他关联因素，比如基础设施与自然资源或金融的联动。正是这些原因共同推动了中国在基建领域的参与度大幅提升。

They have a lot of experience. They can draw on building up infrastructure efficiently. They're very competitive bidders, and that's for a number of reasons. There might be other links, too, that infrastructure might be linked to natural resources or finance. So it's a number of reasons that has contributed to this large rise in China's participation of the construction of infrastructure.

令我惊讶的是，政策制定者似乎非常希望培育本土建筑企业。如果他们真想这么做，有哪些可行方案？

It strikes me the policymakers would really like to grow domestic construction firms if they could. If they wanted to do that, what options do they have to do it?

他们可以借鉴中国通过合资企业在多个领域培育本土产业的经验。在建筑领域，当外国企业执行项目时与本地公司合作，这种做法确实有效。政策制定者可规定雇佣本地工人——因为现有案例中，连低技能工人都被引入他国完成基建项目。还可强制要求使用本地供应商。不过这些政策选项虽被我列出，但建筑行业尚无确凿证据表明哪些措施真正有效。

They could look back at China and how China used joint ventures to grow their domestic industries in the number of sectors. And I've seen this work really construction sector too, where a foreign firm, when they carry out a project, they work with a local firm. They could stipulate the hiring of local workers because there are instances when even workers at lower levels of skills are brought into countries to complete infrastructure projects. Policymakers could stipulate using local suppliers. Now, these policy options, I'm listing them, but I don't think we have any evidence on what works well in the construction sector.

因此我认为这是非常值得与政府合作、积累更多实证数据的领域。现有关于本土化政策的证据主要来自自然资源勘探领域，且效果参差不齐。所以研究对建筑业有效的政策将极具价值。

So I think this is an area where it'd be very useful to generate more evidence, partnering up with governments. I think the evidence that we have from local content policies that were used predominantly in the exploration of natural resources is mixed. And so thinking about policies that are effective in construction would be very useful.

Martina，这些细节令人着迷，同时也发现解决某些问题确实困难。这是否必然面临两难：既要快速建设基础设施，又要实现培育市场竞争、促进就业、激励本土企业等更宏观的政策目标——后者显然难度更大且耗时更久？

Martina, it's fascinating to get the detail on this and to also find out just how difficult it is to resolve some of these problems. Is it inevitably a trade off between the need to create infrastructure and do it quickly and these broader policy goals to create competition and entry in these input markets, provide employment, stimulate domestic firms, which are difficult and take longer?

在某种程度上，确实如此。我认为政策制定者非常清楚这些权衡。他们经常发现自己身处高度拥堵、污染严重的环境中，需要建立交通基础设施来促进人员流动。但我认为不必非得做出这种取舍，因为这同时也是他们面临的巨大机遇。到2040年，非洲90%的建筑尚未建成。

To some extent, is. And I think policymakers are very aware of these trade offs. They often find themselves in highly congested environments, extremely polluted, and they need to put in place, say, transport infrastructure to allow the flow of people across locations. But I don't think it has to be that trade off because I think it's also a huge opportunity that they have. 90% of buildings to exist in Africa in 2040 are yet to be built.

因此未来将会有大量建设。麦肯锡本周发布的报告指出，基础设施融资需求高达10.6万亿美元。这意味着未来可能进行大量投资。因此，思考如何培育国内建筑业并实现项目建设的溢出效应，我认为是非常重要的考量。在这里，动态视野和长远思考至关重要。有哪些方式可以刺激建筑业发展，从而在本地创造更多价值？

So there's going to be a lot of construction. McKinsey came out with a report this week that said there is an infrastructure financing need of $106,000,000,000,000 And so there is a lot of investments likely to be made in the future. So thinking about how to create some domestic construction industry and some spillovers from the projects that are being built, I think, really an important consideration. And here, a dynamic horizon is really important thinking down the line. And what are the ways that construction industry could be stimulated so that some of this can be created locally?

还有很多开放性问题。玛蒂娜，感谢你探讨这些议题。

A lot of open questions. Martina, thank you for talking about them.

非常感谢邀请我参加。

Thank you so much for having me.

如果想了解这项研究，可以阅读关于混凝土的文献——《混凝土思维与发展》（Byrne和Kirchberger 2021年）。还有《建设非洲城市：基础设施建设的劳动力市场影响》（作者Donovan、Kirchberger、Newman、Militerschafer和Thorn），以及《跨境建筑流动及其对低收入国家的影响》（作者Bernard、Kirchberger和Peters）。玛蒂娜，你刚才提到的早期书籍章节不在我的列表中。

If you want to read about this research, you want to read about concrete. Concrete thinking about development. That papers Byrne and Kirchberger 2021. Then also constructing Africa's cities, labor market effects of infrastructure construction. Authors Donovan, Kirchberger, Newman, Militerschafer and Thorn cross border flows in construction and their impact on low income countries authors Bernard, Kirchberger and Peters and Martina, you did mention an earlier book chapter that I don't have in my list.

是哪本书？

What is that?

书名是《建筑业在发展中国家的作用：若干关键问题》。

It's called the role of the construction sector in developing countries, some key issues.

好的，非常感谢。这里是Vox Dev Talk。确保不错过任何一期节目的最佳方式是关注我们。你可以在任何收听播客的平台找到我们。

Okay. Thank you very much. This has been a Vox Dev Talk. The best way to make sure you don't miss an episode is follow us. You find us wherever you get your podcasts.

和往常一样，我们的往期节目可以在voxdev.org上找到，那里还有关于我们讨论话题的文章。如果你喜欢我们的内容，请推荐给其他人。也请给我们留下评价。Fox Dev Talks是Talk Normal制作的节目。助理制作人是Megan Bieber，我们的编辑是Andre Zargarian。

Our past episodes, as always, are at voxdev.org where you also find articles about the topics we discuss. And if you like what you're hearing, please tell someone else about us. Leave us a review as well. Fox Dev Talks is a Talk Normal production. The assistant producer is Megan Bieber, and our editor is Andre Zargarian.

感谢Boxstep的Oliver Hany和Iman Siddiv。

Thanks to Oliver Hany and Iman Siddiv at Boxstep.