Macroeconomic Implications of Child-Labor Laws: A Theoretical Framework

本项目使用动态一般均衡框架研究童工法的确定和影响。在工业化的第一批国家，童工现象在工业革命初期一直很普遍。今天，所有工业化国家都有童工和义务教育法，童工基本上已经消失了。在许多发展中国家，童工今天仍然很普遍，关于是否应该引入更多的限制，人们进行了激烈的辩论。有许多作者研究了童工法对福利的影响。然而，目前还没有研究开发出能够解释这些法律在其历史背景下的演变和影响的政治经济模型。本文描述了一个具有异构主体的动态一般均衡模型，该模型在投票过程中决定童工限制（CLR）。这项研究是与斯德哥尔摩大学的法布里齐奥·齐里博蒂（Fabrizio Zilibotti）合作进行的，结果得出了采用CLR的积极理论，并允许详细分析CLR对经济和人口中不同群体的影响。该模型的关键机制是父母对子女数量的决定与他们对CLR的偏好之间的相互作用。特别是，孩子很少的父母从童工中得不到什么好处，因此可能赞成引入限制措施。另一方面，有许多孩子在工作的父母可能会反对CLR。第二个影响对CLR偏好的因素是（人力或物质）资本资产：拥有与童工互补因素的代理人将更有可能反对引入CLR。一个关键的结果是，如果允许投票，该模型可以显示出具有不同童工政策的多个稳定状态。其原因是政策鼓励行为的反馈，而这种反馈反过来又导致对相同政策的政治支持。多重稳定状态的可能性可以解释这样一个事实，即即使在处于同一发展阶段的国家之间，CLR和童工发生率也存在很大的跨国差异。此外，由于CLR构建了自己的选区，因此该模型解释了为什么CLR通常会随着时间的推移而变得更加严格。如果技术变革随着时间的推移提高了技能溢价，该模型可以产生从无监管到CLR的过渡。这种变化促使父母选择较小的家庭，并投资于人力资本，随着时间的推移，大多数人支持引入CLR。这种对向CLR过渡的描述与以下观察一致：在一些工业化国家，CLR是在工资不平等加剧的时期之后引入的，与生育率迅速下降和教育水平上升的时期相吻合。该项目需要详细记录一些国家的童工改革和之前的政治辩论，并模拟一个经过校准的定量模型，以评估该模型能否很好地再现改革的实际时间以及立法的后续后果。将基本分析扩展到引入附加政策，特别是义务教育法和上学补贴。初步结果表明，如果与许多工业化国家的教育改革同时进行，CLR就更有可能取得成功。另一个扩展是允许在选民预见未来政策的框架内进行动态投票。在这方面，预计未来将引入CLR，可能导致童工数量在限制措施生效之前就有所下降。这一机制可以解释一些实证研究发现，在引入CLR后，CLR对童工供给的影响相对较小。

This project studies the determination and effects of child labor laws using a dynamic general-equilibrium framework. In the first countries to industrialize, child labor was widespread up to the early stages of the industrial revolution. Today, child labor and compulsory schooling laws are in place in all industrialized countries, and child labor has mostly disappeared. In many developing countries, child labor is still common today, and there is a lively debate as to whether more restrictions should be introduced. There are a number of authors who have studied the welfare implications of child-labor laws. There are no investigations, however, which develop political-economy models that can account for the evolution and effects of these laws in their historical context. This proposal describes a dynamic general-equilibrium model with heterogeneous agents who determine child-labor restrictions (CLR) in a voting process. The research, which is joint with Fabrizio Zilibotti of the University of Stockholm, results in a positive theory of the adoption of CLR, and allows a detailed analysis of the effects of CLR on the economy and different groups in the population. The key mechanism in the model is an interaction between parental decisions on the number of children and their preferences for CLR. In particular, parents with few children have little to gain from child labor, and are therefore likely to favor the introduction of restrictions. Parents with many working children, on the other hand, would be expected to oppose CLR. A second factor influencing preferences over CLR are (human or physical) capital assets: agents who own factors which are complementary to child labor will be more likely to oppose the introduction of CLR. A key result is that if voting is allowed, the model can exhibit multiple steady states with different child-labor policies. The reason is a feedback in which policies encourage behavior, which in turn leads to political support for the same policies. The possibility of multiple steady states can account for the fact that there are large cross-country variations in CLR and the incidence of child labor, even across countries at the same stage of development. In addition, since CLR build their own constituency, the model explains why CLR generally tend to get more restrictive over time. The model can generate a transition from no regulation to CLR if technological change raises the skill premium over time. This change induces parents to choose smaller families and invest in human capital, which over time creates a majority in favor of introducing CLR. This account of the transition to CLR is consistent with the observation that in a number of industrialized countries CLR were introduced after a period of rising wage inequality, and coincided with a period of rapidly declining fertility and rising education levels. The project entails a detailed documentation of child-labor reforms and the preceding political debate in a number of countries, and simulations of a calibrated, quantitative version of the model to assess how well the model can reproduce the actual timing of reforms, as well as the ensuing consequences of the legislation. The basic analysis is extended to introduce additional policies, especially compulsory schooling laws and schooling subsidies. Preliminary results indicate that CLR are more likely to be successful if they are accompanied by education reforms, as they were in many industrialized countries. Another extension is to allow dynamic voting in a framework where voters foresee future policies. Here an anticipated future introduction of CLR can result in a decline in child labor even before the restrictions come into force. This mechanism can account for the fact that some empirical studies find only relatively small effects of CLR on child labor supply after they are introduced.