第六季第36集：文化与经济发展

人们应该从项目一开始就将文化因素纳入考量。最不愿看到的情况是赢得了项目却遭遇失败，然后才后知后觉地发现——哦，这会不会是文化差异导致的？到那时再想深入调查就非常困难了。

People should take culture into account from the beginning of the project. What you don't want to do is you won the project and it fails and then you kind of expose like, oh, was that because of culture? Then it's very difficult at that stage to really investigate that.

欢迎收听Vox Dev Talks，我是蒂姆·菲利普斯。文化如何影响发展政策？发展政策又如何反作用于文化？如果我们忽视文化规范，或未能理解它们与善意政策间的相互作用，这种认知鸿沟可能会破坏原本结构完善的发展项目。一篇新论文探讨了我们对文化、政策与经济发展及其相互作用的现有认知。

Welcome to Vox Dev Talks. My name is Tim Phillips. How does culture affect development policy, and how does development policy affect culture? If we don't take account of cultural norms or we fail to learn about how they interact with well intentioned policies, then this gap in our knowledge, well, it might be undermining otherwise well structured development projects. A new paper investigates what we know about culture, policy, and economic development and their interaction.

论文作者包括加州大学洛杉矶分校的娜塔莉·鲍、加州大学圣地亚哥分校的莎拉·洛斯以及芝加哥大学的爱德华多·蒙特罗。今天我们很荣幸邀请到三位齐聚于此，阵容强大。娜塔莉，欢迎再次做客VoxDevTalks。

Natalie Bau of UCLA, Sarah Lowes of UC San Diego, and Eduardo Montero of the University of Chicago are the authors. And we have all three of them joining me now, full house. Natalie, welcome back to VoxDevTalks.

很高兴再次见到你，蒂姆。

Nice to see you again, Tim.

莎拉，欢迎你。

Sarah, welcome.

谢谢，很荣幸参与。

Thank you. Good to be here.

还有爱德华多，同样欢迎你。

And Eduardo as well. Welcome.

谢谢。感谢邀请。

Thank you. Thanks for the invitation.

娜塔莉，你为什么写这篇评论？

Natalie, why did you write this review?

嗯，有一本新的文化经济学手册即将出版，萨拉、埃德瓦尔多和我被邀请撰写这一章，所以我们几乎别无选择。

Well, there's a new handbook of cultural economics coming out, and Sarah, Edwaldo, and I were asked to write this chapter, so we almost didn't have a choice.

这也算是一种动力，不是吗？在这种语境下谈论文化时，我们实际指的是什么？这个词的含义是什么？

That's a motivation of a sort, isn't it? When we talk about culture in this context, what do we actually mean? What does the word mean?

这是个棘手的问题。我觉得应该让萨拉或爱德华多来回答。

This is a tricky question. I feel like I should make Sarah or Eduardo answer this one.

是的，说到文化，我们通常指的是代际间社会传递的价值观和信仰。但举些更具体的例子，比如娜塔莉研究中的案例，有时我指的是诸如婚礼支付方式这类习俗——比如新郎家庭是否要向新娘家庭转移钱财。娜塔莉研究过彩礼就是其中一例。类似的例子还有很多，比如广义信任度，你是否信任他人？

Yeah, so by culture, we generally mean the values and beliefs that are socially transmitted across generations. But to give more concrete examples, say from example from Natalie's work, sometimes I don't mean practices like what types of payments are made when people get married, like are there transfers of money between say the groom's family and the bride's family. So that's one Natalie studied bride price, for example. But there's lots of examples we can think about. Like generalize trust, do you trust other individuals?

你觉得这样做对吗？

Do you think it's the right thing to do?

经济学家可以从多种角度思考文化。它可能是决定游戏规则和人际互动方式的因素，因为如果你不遵循社会规范，可能会面临社会制裁。它也可能像人们在决策时使用的启发法，对吧？毕竟每次都要在信息不全的情况下计算出最佳选择是很耗费认知的，而随着时间的推移，某种启发法已经演化并传承下来，可供你决策时使用。它还可能与你的信仰或态度有关。

Economists can think about culture in a lot of different ways. It could be something that determines the rules of the game and how people interact because you might face social sanctions if you don't follow Sultan norms. It could be like a heuristic people use in decision making, right? Whereas it's, you know, cognitively complex to like work out the best option every time where you may not have all the information, but over time, some kind of heuristic has evolved and been passed down that you could use to make decisions. It could be something that involves your beliefs or attitudes.

所以泰勒给出的定义确实非常宽泛。它涵盖了文化影响决策制定的所有丰富可能性。

So, really like about the definition Taylor gave is it's pretty broad. It allows for kind of all the richness of how culture can affect decision making.

在发展政策的语境下，当我们思考文化时，真正需要考虑的是文化如何塑造激励机制。也就是说，个体做出不同类型的选择时会获得怎样的回报？

In the context of development policy, when we think about culture, we really want to think about how culture is shaping incentives. So what are the payoffs associated with different types of choices individuals are making?

是的，完全正确。

Yeah, absolutely.

既然这是个相当宽泛的定义，你们如何识别什么是文化？我们该如何测量文化？

Given the fact that it is a fairly broad definition, how do you recognise what is culture? How can we measure culture?

过去很多经济学家认为文化是非常软性的概念，就像模糊不清的东西。

A lot of economists in the past used to think about culture being very soft. It's like fuzzy.

嗯。

Mhmm.

我们如何定义它？作为经济学家，我们偏好那些可以精确测量的严谨事物。过去几十年里，经济学家在衡量文化及获取相关微观数据方面已取得长足进步。比如通过问卷调查询问人们是否信任他人，当然这种方法也存在局限性。

What do we know what it is? Whereas economists, we like, like, rigorous things that we can measure. I think in the last few decades, economists have gotten much better about measuring culture and thinking about how to get micro data on it. So there's things like survey questions where you ask people, do you trust others? There's downsides to that.

莎拉通过大量实验室实地实验，运用行为博弈来测量人们在特定选择情境下的反应。同时，人类学家提供的优质民族志数据也被经济学家广泛采用。近年来，经济学家在文化量化研究方面显著提升，更懂得如何将文化因素纳入计量经济模型。

So Sarah has done a lot of work with lab in the field experiments trying to use behavioral games to measure people in the field, how do they react under different choice sets. And then there's also now, like, extremely good ethnographic data as well about societies from anthropologists that economists have started using a lot more. And so I think in the last few years, economists have gotten much better about rigorously measuring culture and thinking about how do you, you know, put culture into data, fit it into our our econometric models.

我要补充的是，我们很幸运能借鉴人类学和社会学长期的研究积累。嗯。经济学往往带有学科扩张性，但这次我们无需从头开始，完全可以从其他领域关于文化测量的既有成果中汲取养分。

And I would add that we're lucky that anthropologists and sociologists have been thinking about this for a pretty long time. Mhmm. And so economics tends to be quite imperialistic. We don't have to reinvent the wheel here and we can learn a lot from what these other fields have thought about in terms of how to measure and think about culture.

这种文化领域的学科扩张是否意味着经济学家过去忽视或低估了文化的影响力？

Does that cultural imperialism mean that economists have either ignored or undervalued the influence of culture in the past?

我认为过去二十年里，学界对文化作用的认知有了显著发展。历史上文化常被视为难以量化的因素，甚至因具有残余属性而被认为不够严谨——当无法解释某种现象时，人们就会归因于文化。

I think like the past twenty years, there's been a lot of growth and excitement about understanding the role of culture. Cultural historically was something that was called to measure. And also, was seen as, like, maybe not that rigorous because it was a little bit residual. It's like, well, we can't explain why something is happening. So let's blame it on culture.

经济学家反感这种解释方式。没人愿意接受这种可以解释一切的残余变量。正因如此，学界长期回避文化研究。但越来越精细的微观数据，以及人类学测量方法的影响，确实推动了该领域的复兴。

Economists don't like that. Like, you don't want an explanation that's it's like it's the residual you just go to. It can explain everything. And so I think for a long time, shied away from culture because of that. But I think the existence of better and better micro data and more influence from measurement in anthropology has really led to this resurgence.

你们论证文化认知对政策效能重要性的方法颇具独创性。能否详细说明这项研究的设计初衷和具体操作？

You know, quite an ingenious way to demonstrate that the understanding of culture is important for policy effectiveness. Tell me a little bit about this and explain what you did and why you did it.

我们开篇试图阐明为何应关注文化与政策的交汇点。通过分析世界银行的项目数据、实施情况及其成效评估，我们试图用这些数据揭示文化因素可能影响政策效能的典型事实。在此过程中，我们特别关注项目负责人——若某项目的负责人来自项目实施地，该项目是否会更成功。

We begin the chapter by trying to motivate why we should care about the intersection of culture and policy. And we use some data on the World Bank and projects that are implemented by the World Bank and then evaluations that the World Bank does of these projects and how effective they are. The logic was, can we use this data to provide some stylized facts about culture potentially mattering for the efficacy of policy? As part of this exercise, we identify project team leads. So the individuals who are in charge of these World Bank projects, you have an individual who's from the place where a project is being implemented, is that project more successful.

确实发现一些证据表明：当负责人来自项目实施地时，项目成果评分更高。当然这只是相关性证据，既无识别策略也未经随机分配，需谨慎解读。我们的理解是，这些负责人能带来本土特有的知识，可能促进项目成效。不过需要说明，世界银行的管理策略中不委派本地负责人也有其合理考量。

And we do find some evidence that if individuals happen to be from the place where the project is being implemented, the projects get higher outcome ratings. Now, of course, this is purely correlational evidence, there's no identification strategy, there's no randomization of task team leads to the projects that are being implemented, you have to interpret these data with a bit of caution. But the way we were thinking about it is that individuals bring location specific knowledge to projects when they are being implemented in their country of origin. And this could potentially facilitate better project outcomes. Now, just in defense of the World Bank and the types of management strategies they have, there's obviously really good reasons why you might not assign someone who's from the place where a project is being implemented.

确实如此。读完你们的研究后我查证过，这些理由很充分。他们...

Well, exactly. Because after I read what you've done, I went and looked this up. Now there are good reasons. They?

首要原因是避免任何可能滋生腐败或非客观决策的表象与可能性。

So the primary reason is that you might just wanna prevent any sort of appearance of or ability to have any corrupt or non objective types of decision making.

这背后存在逻辑：世界银行警惕地方俘获风险。想象一下，本国负责人可能偏袒同胞，助长裙带关系，这确实值得警惕。

So there's, like, logic behind it too, which is the World Bank was afraid of the potential of local capture. You can think about it. If you're from the country, maybe you'll favor certain people from your country, home country. There'll be more potential for nepotism. This is definitely something that we should be worried about.

政治经济学与发展领域历来重视此类问题，其合理性毋庸置疑。此外还有其他原因，比如跨地区轮岗能促进对不同背景的认知——我们都有过在不同环境中工作的学习经历。

A lot of work has been going into things like this in the past historically in political economy and development. So there's very valid reasons for it. I think there are other reasons too. Like, there's a potential for learning about different contexts if you're rotated across different places. I think we've all learned from working in different contexts as well.

因此需要权衡这两种价值。例如若让我们首年就在陌生环境工作，必然面临适应本地规范、文化、合作对象等挑战。就像经济学家在不同语境下工作，既收获经验也面临困难，世界银行团队负责人同样如此。

So there's a value, but you wanna balance these two potentially. For example, if you made us all work in a new place, a totally different place, the first year, we'd probably struggle because we'd have to really understand the local norms, the culture, who to work with, who we can trust. So there's pros and cons just like as economists working in different contexts. We learn a lot, but there's also challenges. The same thing could be true for these World Bank team leaders.

了解一个地方、熟悉其文化具有重要价值。或许你能因此更好地实施项目、优化设计。就像经济学中的许多现象一样，这里存在着各种权衡取舍。

There's value of knowing a place, being familiar with the culture. Maybe you can implement projects a bit better, design them better. So like a lot of things in economics, there's these trade offs.

这种情况并非世界银行独有。甚至在许多国家的内部公务员体系中，人员往往被刻意不分配至家乡所在地，或可能每隔几年轮调一次。正如爱德华多所说，这确实体现了某种权衡——将人员派驻异地可能是顾虑现有关系网的影响，而定期轮调又导致他们刚适应当地文化就被调往他处。

And this is not exclusive to the World Bank. Many even in internally to countries, many civil services are designed this way where people are adorably not assigned to their home location, or maybe even rotate every few years. And I think so as Eduardo says, this really points to kind of the trade offs, maybe you assign someone somewhere else because you're worried about the existing networks. Now you rotate them every few years, it's just as soon as they begin to warm the local culture, they'll rotate it out to somewhere else.

从你的综述中可以明显看出，现有大量证据表明政策失效往往源于文化影响或对文化因素的忽视。能否举例说明这种情况？

Reading your review, it's very obvious that now we do have quite a few pieces of evidence that policy has been less effective often because of the influence of culture or because it did not take account of the influence of culture. Can you give me some examples of this?

我认为有很多政策因未考虑文化因素而产生意外后果的案例。这未必意味着政策在主要目标上失效，但可能同时引发非预期的、甚至负面的影响。以我在低收入国家养老金计划的研究为例，我曾考察印尼和加纳的养老金推广。我们通常将养老金视为重要的现代化政策，特别是在人们逐渐减少依赖子女养老的背景下。

Well, I think there have been a lot of cases where policies have had unintended consequences because of failure to think about culture. And so that doesn't necessarily mean it's not effective at one goal, but it may have a consequence that isn't intended at the same time and could even be negative. An example from my own work comes from pension plans in low income countries. I've studied the rollout of pension plans in Indonesia and Ghana. We think of pension plans as kind of a fairly important modernization policy, especially as people are starting to wire west on their kids for old age support.

通常需要福利国家介入提供保障。但我的研究表明，在现存亲属赡养传统（如母系社会中女儿赡养父母/父系社会中儿子赡养父母）的文化背景下，这些传统原本激励父母对特定子女进行教育投资，因为他们能更直接看到回报。而养老金计划的实施降低了子女留守赡养的可能性——虽然这种后果的好坏见仁见智——但更重要的是，它同时削弱了教育投资的动机。

Often the welfare state needs to kind of step in to provide that. But what I've shown in my work is that when you had these kind of existing kinship traditions where Sultan Sjodwin would take care of their parents in the old age, these the traditions I work out, matricality with daughters take care of parents and patricality with sons take care of parents, those traditions actually were incentivizing parents to educate those kids because they saw more the returns to those educational investments. When you put in the pension plans, you know, actually reduced the likelihood that kids would stay with their parents and take care of them. You can kind of argue about whether this was kind of a good or bad consequence, this is complicated effect. But on top of that, it also reduced the incentive to educate those kids.

在传统由女儿赡养的文化群体中，女童相对受教育率下降；由儿子赡养的群体中，男童教育率同样下滑。该政策在提供养老保障方面或许是有效的，但既挤占了原有的子女赡养体系，又可能减少了教育投入——而这恰恰是这些发展中经济体最需要提升的领域。萨拉想谈谈保险箱项目吗？这算是我最钟爱的政策失效案例之一。

The relative education rates of the toddlers fell in the cultural groups where toddlers took care of parents and the relative education rates of sons fell in the cultural groups where sons took care parents. The policy may have been effective, it did provide more old age support. At the same time, it clouded out a lot of the old age support that was previously going on from children and it also maybe reduced educational investment at a time when probably these growing countries really wanted to increase educational investment. You want to talk about lockboxes, Sala? I think this is like kind of one of my favorite ineffective policies.

是的，我很乐意讨论。这个随机对照试验实施于一个相对偏远、自给自足且市场参与度低的群体——奇曼内人。项目旨在提升储蓄率，通过为村民提供存放积蓄的保险箱来实现。

Yes, I'm happy to talk about this. So this is an RCT that was implemented among a group that sort of was pretty remote, autarkic, so not well integrated into markets. So the Tsimane. And the goal was to increase savings. So they provide these individuals in these villages with access to lockboxes to put their savings in.

因此或许令人惊讶的是，他们发现当个人拥有这些保险箱时，实际效果反而是酒精消费量增加了。我认为这让研究人员感到困惑，因为这显然不是他们实施此措施的初衷。

And so perhaps surprisingly, they find that when individuals have these lockboxes, the effect is that they actually have an increase in alcohol consumption. And that was, I think, puzzling to the researchers that was not obviously the reason why they implemented this

我也觉得困惑。是啊。

puzzling to me. Yeah.

于是他们请来了一位对该地区有深入了解的人类学家。在这个人们尚未融入市场经济的环境中，金钱的唯一用途就是购买酒精。考虑到这种文化背景，这个结果或许并不意外。人类学家向经济学家们解释了为何会出现这些看似反常的现象。

So they brought on an anthropologist who had a, know, a pretty deep knowledge of the setting. And in this setting where people aren't integrated into markets, the only use that people had for money was to purchase alcohol. So it was a pretty, perhaps expected outcome given the cultural setting, The anthropologist was able to explain to the economists why they are finding these otherwise unexpected outcomes.

难以置信。爱德华多，你能给我举个实例吗？

Incredible. Eduardo, do you have an example for me?

我和莎拉有篇关于殖民地医疗健康运动及其长期影响的论文。在法属中非殖民地，政府曾大力防治昏睡病。基本上，村民被迫接受血液检测，并经常被迫使用疗效极差的药物治疗。这段历史从1920年持续到1956年左右。长期来看，我们发现这些地区对医学的信任度更低。

Sarah and I have this paper about colonial medical health campaigns and their long term consequences. In former French colonies in Central Africa, the government went around trying to control sleeping sickness. Basically, people in villages were forced to submit to blood test and often forced treatments using medicine that was very ineffective. This is from 1920 to around 1956. And so in the long run, we find evidence that these places mistrust medicine more.

但我想，如果不了解一个地方的历史文化，很容易简单归结为'这里的人对健康或疫苗不感兴趣'。我们的研究甚至显示，世界银行的健康项目（再次提到它们）在这些地区效果较差——特指健康类项目，其他世行项目影响不大。这个例子说明，由于历史造成的对医学的不信任文化，导致当今政策效果打折。但如果我们理解这段历史和文化，或许能设计出不同的健康干预措施，或制定不同的成功干预标准。

But I think if you hadn't had this understanding of the history of a place, the culture, it'd be very easy to be like people here aren't interested in health or vaccines or other things like that. And we even show that World Bank health projects, speaking of them again, are less effective in these areas. Health ones specifically, not so much other World Bank projects. That's an example of policies today are less effective because of a history that created a culture of mistrust in medicine. But maybe once we understand the history and the culture, we might design different health interventions or have different standards for what's a successful intervention or not.

人类学家早就指出...嗯...这些殖民时期的医疗运动确实深刻影响了当今世界部分地区人们对医学的看法。我们在论文中做的，就是收集这些运动发生地的档案数据，研究其对现状的影响——比如当DHS数据中显示有人被提供免费血液检测时，他们是否拒绝？我们发现那些有较多殖民医疗历史的地区，如今更可能拒绝免费血液检测，疫苗接种率也更低。

Anthropologists had highlighted Mhmm. These colonial campaigns just really affected people's views on medicine today in parts of the world. And so what we did in our paper is gather some archival data from where these campaigns happened and study how did that impact things like if you're offered a free blood test from the DHS data, do you refuse it or not? And we find places that had more of a history of these colonial campaigns. We find these places are more likely to refuse free blood tests and have fewer vaccinations today.

在尼日利亚等地，疫苗接种运动曾出现过一些相当极端的例子——人们真的会藏起孩子或逃离以避免接种，因为他们深信这些疫苗会导致不育或其他健康问题。这确实是个非常严重的公共卫生问题。

And there've been some pretty extreme examples of vaccine campaigns in places like Nigeria, where people have literally hidden their dawdles or people have fled to avoid the vaccines because they're so convinced that these vaccines, know, for example, cause sterility or have other negative health consequences. So this is a very serious health problem.

关于这点我必须强调，我在多期Vox Dev Talks节目中与众多人士探讨过：当随机对照试验成功后尝试推广时，若仔细观察就会发现，目标推广地区的文化差异会造成巨大影响。能举些具体例子吗？

Now, aspect of this that I must have talked to many, many people, many episodes of Vox Dev Talks have been concerned with this, where you have a successful RCT, you try and scale it up. It is unsuccessful when you look more closely at it than some cultural differences in the areas to which you're trying to scale it up make a big difference. What are the examples of this?

回到Sarah提到的保险箱案例，这非常有趣——在其他情境中，这种简单政策（提供保险箱）不仅有效提升储蓄率，还促进了后续资产投资和收入增长。但在玻利维亚农村实施时，效果却截然不同。

Just going back to the example Sarah brought up about the lockboxes, that's a really interesting example because in other contexts, it had been found to be a very effective way of first increasing savings, of course, through the lockboxes, but then on subsequent other investment in assets and incomes in other contexts. So that's an example where the same policy, such a simple policy like giving people lock boxes, had been shown to be effective in other contexts. But when you do it in a context like in rural parts of Bolivia, you find very different effects.

另一个典型案例是无条件现金转移支付。大量证据表明其短期扶贫效果显著。但我们在肯尼亚发现，当地社区竟主动抵制接收免费资金——这对多数人来说难以理解。毕竟若问'想要一千美元吗'...

Another good example, I think, are unconditional cash transfers. So there's a lot of really excellent evidence that they're really effective at alleviating poverty, especially in the short run. Mhmm. But in our chapter, we cover some examples from Kenya where actually communities actively resisted receiving free money, which I think is puzzling for many people. If you were to just hear like, hey, would you like a thousand dollars?

通常回答会是'当然要'。但肯尼亚人却说'不参与'。证据表明这种抵制源于当地对互惠原则的理解：免费获取不符合文化认知，反而引发强烈怀疑。与其参与这种可疑且可能有害（因文化冲突）的项目...

I think the answer is often yes, please. Sure. But in this case in Kenya, people were actually like, no, we don't wanna participate in this. And the evidence suggests that part of that resistance comes from the way reciprocity is understood in this context that you can't receive things for free that just didn't culturally resonate with people and in fact was actually really suspicious. And so rather than participate in something that was suspicious to them and potentially like malevolent because it didn't meld well with the culture.

人们拒绝了白给的钱。

People refused free money.

能否视此为积极启示？当深刻理解文化后，干预政策不仅能更有效快速实施，还可能产生原始设计中未预见的溢出效应？

Can we think about this as a positive, the understanding of culture and how with a deep understanding of culture then interventions or policies can become even more effective or work quicker or have spillover effects that maybe wouldn't have been thought of in the original design?

因此我认为这更像是文化经济学的新前沿领域。作为经济学家，我们越来越需要思考如何在政策设计中考虑文化因素，或如何利用现有的文化实践。回顾爱德华多和莎拉提到的例子，撒哈拉以南非洲地区存在严重的医疗不信任现象。实际上，这种不信任表现为实地干预会带来稳定性。我的一些研究探讨了为什么撒哈拉以南非洲的年轻女性不愿采取避孕措施，尽管她们经常进行无保护的性行为。即使她们正积极参与教育，未计划的怀孕仍可能迫使她们辍学或打乱人生轨迹。

So I think this is more of the new frontiers in cultural economics. And so, I think as economists are increasingly thinking about how do you like take culture into account in policy design or leverage existing cultural practices. Going back to the example Eduardo and Sarah were talking about of there being huge amounts of medical distrust in Sub Saharan Africa. And actually, that kind of manifests in a field that interventions will cause stability, I have some work looking at the why young women don't take up contraceptives in Sub Saharan Africa, even though they are often having lot of unprotected sex. And even if they're actively engaged in no education, and so they may be forced to drop out or kind of have their lives dewailed by unplanned pregnancies.

我们在焦点小组中发现，人们普遍存在一种强烈担忧，认为现代避孕方法会导致不孕。为此我们专门针对这种医疗不信任设计了治疗方案，成功将该人群的避孕采用率提高了约30%至40%。

And what we saw in focus groups is that there was this tremendous feel that these modern contraceptive methods cause infertility. And so we actually designed a treatment to exactly target that medical distrust, and we were able to increase contraceptive take up weights by about thirty to forty percent in that population.

那么娜塔莉，让我们通过你所说的新前沿来总结。一个积极的方面是，我们可以从这些所谓的失败中学习——那些未达预期的结果。但如何确保我们确实从中吸取教训？因为这些案例往往既未发表也未公开。

So let's finish up by looking at what you are calling the new frontier, Natalie. And one positive thing is that we are learning from, we can call them these failures, the results that aren't as good as we hoped that they would be. But how can we make sure we do learn from them? Because very often they don't get published and they don't get publicized.

蒂姆，你提出了一个重要问题。我们无法从未知的事物中学习。我认为首要的是，当项目失败或遇到问题时，这些情况应该被公开记录或提供给政策制定者。缺乏这些信息和数据，我们就错失了了解问题所在、总结经验以及未来改进的机会。这个问题比我们今天能解决的更大，但确实表明前进的方向之一是确保相关数据对感兴趣的人开放。

Tim, you raise an important problem. We can't learn from things we don't know about. So I think the first order thing is that when projects are unsuccessful, when they face problems, it should be something that's publicly documented or available to policymakers, because without that information, that data, it's just a missed opportunity to learn what went wrong, what went well, how do we improve in the future. And that's a bigger problem than I think we can solve today. But it certainly suggests that, you know, one path forward is making sure this data and information are available to people who are interested.

正如你提到的蒂姆，文件抽屉问题非常严重。有人认为进行LCT研究时，由于投入的时间和金钱，结果被埋没的可能性较小。但我们明确建议研究者：所有项目——即使是零结果——都应该发表，理想情况下期刊应该有兴趣刊登有意义的零结果。人们应该从项目伊始就考虑文化因素。最糟糕的情况是项目失败后才后知后觉地追问：是不是文化因素导致的？

As you were mentioning, Tim, the file draw problem is a really serious issue. There's been some argument that when people do LCTs, maybe it's less likely to occult simply because of the time and money involved. But, you know, our obvious recommendation to researchers is that all projects, even if the null results should be published, ideally journals should be interested in publishing interesting null results. People should take culture into account from the beginning of the project. What you don't want to do is you won the project and it fails and then you kind of expose like, Oh, was that because of culture?

到那时再调查就非常困难了。因此我们提出系列建议：一是尽早咨询人类学家，思考哪些文化背景因素可能产生影响；二是开展自己的焦点小组并深入当地语境非常重要；三是如果你正在做RCT，在设计干预方案时就要考虑文化重要性，这可能意味着按某些文化特征进行分层设计。

Then it's really difficult at that stage to really investigate that. And so, we have a series of recommendations. One is to talk to anthropologists really early on to think about what elements of the cultural context might matter. I think also doing your own focus groups and embedding yourself in the context are really important. If you're doing, for example, your own RCT, and that was to kind of design the intervention itself with the idea that culture may be important, that could mean stratifying along certain cultural characteristics.

比如莎拉在研究年龄组时就有这样的实践。还包括确保收集相关文化实践的数据，这样后期就不会陷入「是不是文化原因」的困惑，而能实际验证这些假设。我认为这些文化导向的评估方法能实现更具文化意识的设计。同时需要明确：文化意识设计并非简单地将干预措施与某些现有文化传统结合，然后看到统计显著结果就万事大吉。

And, you know, Sarah has some work where they do that when they were looking at age sets. It could also mean making sure that you collect the data on the relevant cultural practices so that you're not like, Oh, was it cultural? Later on, you actually can test those hypotheses. So, think those COEs are way to have more cultural conscious design. And I think also it's important to recognize that cultural conscious design does not mean just interacting in intervention with some existing cultural traditions, saying, Oh, I see some statistically significant results.

我们想了解为何该干预措施会与这些特征相互作用，以及这是否可能对其他环境具有外部有效性。因此，重要的是思考这些特征如何改变激励机制，如何将吸收的文化特征与既定变量的影响区分开来，或许还要考虑COVID与文化特征的关联。我认为所有这些都需要从项目初期就进行极其周密的设计思考。

We want to understand why that intervention interacts with those characteristics and whether that's likely to be externally valid to other settings. And so, it's important to think about why those characteristics are changing incentives, how we can separate the effects of whatever absorbed cultural characteristic from admitted variables and maybe COVID with that cultural characteristic. And so I think all of this requires very careful design thinking from early on in the project.

针对这一章节，我常听到的回应是：'哦，你指望我们考虑环境中所有潜在的文化特质，并基于这些极其本地化的知识制定政策？这难道不是天方夜谭吗？'其实开始这个过程有个简单方法——许多发展研究者和政策制定者已经在做了——就是寻求本地合作者。那些真正深刻理解当地情况的人。

A common response I get to this chapter is that, oh, you expect us to take into account every potential cultural trait in a setting and build policy with all of this deeply, deeply local knowledge. Is doesn't that seem impossible? An easy way to begin this process, a lot of development researchers and policy makers already do this, is having local collaboration. People from the context who really understand well what's happening.

跨学科寻找合作者容易吗？具体怎么做？正如你开头提到的娜塔莉，经济学可能历来不太擅长借鉴其他学科。该咨询哪些专家是否显而易见？

Is it easy to find people to work with from other disciplines? How do you do that? Economics, as you mentioned at the beginning, Natalie, maybe hasn't had the best record of looking outside its own discipline. Is it obvious who to consult with?

跨学科合作其实相当具有挑战性，特别是因为不同领域的发表要求和标准差异巨大。即使你找到合作默契、能产出卓越成果的人，对方可能也不愿花费六七年时间试图在顶级经济学期刊发表论文——而这在经济学界是常态。比如'搬迁机会'项目，他们研究当给予极贫困家庭搬入高收入社区的凭证时会发生什么。结果发现对孩子有显著正面影响，但最初对父母效果不明显。为此他们引入了定性社会学家进行深度家访，最终解释了这种现象。

Well, think interdisciplinary collaborations can be pretty challenging, particularly because the publication like expectations and standards are really different across fields. So, even if you find somebody who, you know, you work with very well and you could do amazing work with, you know, they may not be interested in trying to spend six years or seven years trying to publish in a top economics journal, right, which is pretty standard within economics. One example is this project called Moving to Opportunity, where they looked at what happened when you gave households very poor households vouchers to move to higher income neighborhoods. It turns out there were very big positive all one effects on the kids, but they initially didn't see much positive effect on their parents. And so to try and understand that, they actually brought in some qualitative sociologists who did like very in-depth interviews with the households and were able to provide some explanations, being able to understand why this was happening.

即便不进行正式合作，向跨学科听众展示研究成果、咨询特定领域专家、阅读相关文献，都能显著提升研究质量。而且人们通常很乐意交流观点。我最近研究DALL·E时直接给人类学家发邮件说：'我们在做这个，不知您是否觉得荒谬，但我想听听意见。能否通话半小时？'

But I think even short of that, just presenting to interdisciplinary audiences, like talking to people who are experts in certain areas, reading the relevant literature, I think all of this can lead to like much more informed research. And I think often people are very happy to talk about things. I've recently been doing some work on DALL E and I cold emailed some anthropologists and I said, this is what we're working on. I don't know if you think it's crazy, but I would like to know if you think it's crazy. Can we just talk for like half an hour or an hour on the phone, you know?

这种交流极其宝贵，长期来看还可能发展为合作。其实合作契机往往自然产生：经济学常做焦点小组这类定性研究，虽然从不写入论文，但能指导干预措施和数据收集。我们只是没想过发表这些。

That's really valuable. And then also those conversations can turn into collaborations, you know, in the long run. And I think there's also some like kind of natural ways of collaborations to occur. In economics, we do a lot of focus groups, all version of qualitative research, and that never is in the papers. We do it to inform our intervention, to inform what data to collect, but we never think about trying to publish that.

实际上，有位在赞比亚合作的社会学家告诉我：'你们做了这么多精彩访谈，我是民族志学者，何不再做几场访谈让我写成论文？'可见在我们现有研究流程中，完全可以嵌入更多跨学科工作，并受益于更擅长定性研究的专家。

And actually, I was talking to sociologist we're working with in Zambia and he said, you did all these really interesting interviews. I'm an ethnographer. Why don't we just do a couple more interviews and I can write this up as a paper? So I think there's already ways that we could nest more interdisciplinary work and also gain a lot from people who know a lot more about how to do that qualitative work, just within our natural process that already exists.

简单总结一下，我们还有多少未知之处？为了在这个领域帮助我们的实践，我们需要找出哪些信息？

Just to sum this one up, how much do we still not know? What do we need to find out to be able to help our practice in this area?

简而言之，我们不知道的还有很多。这是我写过的最短的章节之一，因为事实证明还有很多领域有待探索。我先提出一个建议，我想娜塔莉已经多次暗示过，那就是我们需要关于文化实践的更好数据。目前很多关于文化的研究依赖于那些非常优秀但可能已经过时的民族志资料，这些资料大多是在殖民时期收集的，我们现在完全可以做得更好。因此，我们需要思考如何系统地记录文化实践，不仅是在某个时间点，还要跨越时间，因为文化显然是在不断变化的。

The short answer is there's a lot we don't know. This was one of the shorter chapters I've written, because it turns out there's a lot still left to explore. I'll start with one recommendation that I think Natalie's already hinted at a few times, which is that we need better data on cultural practice. So a lot of the work on culture relies on these really excellent, but now maybe perhaps dated ethnographies that were collected during the colonial era, we can probably do better than that now. So thinking about how we can systematically document cultural practices, both at a point in time, but also across time, because obviously culture is changing.

而且文化不是静态的。而我们目前的数据在捕捉这种变化方面表现不佳。

And it's not a static thing. And our data so far are not great at capturing that.

我们目前所做的‘水行之路’研究基于民族志地图集，这基本上是某个静态时间点的数据，而且对国家来说并非同一时间点。我们大多忽略了这一点。关于文化动态的研究非常少，因为我们通常没有大多数文化变量随时间变化的多次测量数据。很多这样的信息可能存在于人类学中，只是需要被编码整理。我想已经有一些人可能在考虑这方面的工作了，但这将成为一个令人兴奋的方向，同时也可以考虑在现有的、随时间重复收集的数据中嵌套更多文化变量。

The Water to Walk we do lies on the ethnographic atlas, which is kind of a static point in time, and not actually the same point in time for the country. We mostly ignore that. We've gotten very little work on the dynamics of culture because we often don't have multiple measures over time for most cultural variables. And a lot of that information does potentially exist in anthropology but would just need to be codified. I think there are some people maybe think about working on this already, but I think that would become an exciting direction and also thinking about nesting more cultural variables on existing data that's being collected and kind of at repeated points over time.

我认为我们之所以在诸如一夫多妻制这类主题上有相对较多的研究，是因为这些文化变量已经包含在人口与健康调查中。因此人们实际上可以利用现有数据来研究这些。不幸的是，从发展趋势来看，我们可能实际上会减少这类数据集的收集。希望未来在继续收集的数据集中，更广泛的文化变量能够嵌套到现有的数据收集工作中。

I think probably the reason we have relatively more work on sort of things like polygyny is because those cultural variables were in the demographic and health surveys. And so people actually could use existing data to look at that. Unfortunately, the direction things are moving in, we may actually be getting less of those types of data sets. Hopefully in the future for the data sets that continue to be collected, the broader set of cultural variables could be nested in those existing data collection efforts.

祝你们接下来的研究顺利。娜塔莉，非常感谢你。谢谢。莎拉也是。

Good luck doing the rest of the research. Natalie, thank you very much. Thanks. Sarah, as well.

非常感谢。

Thank you very much.

这篇论文名为《文化政策与经济发展》。作者们，你们已经听过了他们三位的介绍，分别是娜塔莉·鲍、莎拉·洛斯和爱德华多·蒙特罗。这里是Vox Dev Talk节目。确保不错过任何一期的最佳方式，就是关注或订阅我们。你可以在任何获取播客的地方找到我们。

The paper is called Culture Policy and Economic Development. Authors, you heard from all three of them, Natalie Bao, Sarah Loews, and Eduardo Montero. This has been a Vox Dev Talk. The best way to make sure you don't miss an episode, follow us or subscribe. You find us wherever you get your podcasts.

我们过往的所有节目都可以在voxdev.talk上找到。你还会看到voxdev.talk上关于我们所有专题话题的文章。相关内容相当丰富，这点我很清楚。记住，如果你喜欢听到的内容，请向他人推荐我们。

All our past episodes are voxdev.talk. You will also find articles of voxdev.talk about all the topics that we feature. There's quite a lot about this. I know that. And remember, if you like what you're hearing, tell someone else about us.

行动起来吧。也请给我们留下评论。VOXDEF Talks是Talk Normal制作公司出品。助理制作人是梅根·比伯，我们的编辑是安德烈·扎尔加良。感谢VOXDEF的奥利弗·哈尼和伊曼·西迪克。

Go on. Leave us a review as well. VOXDEF Talks is a talk normal production. The assistant producer is Megan Bieber, and our editor is Andre Zargarian. Thanks to Oliver Hany and Iman Siddik at VOXDEF.