编者按：美国全国经济研究所（NBER）是美国最大的经济学研究组织，其发布的工作论文代表着经济学研究最新的成果，每周一发布新论文。本周（10.21-10.27）共发布 14 篇新工作论文，本期将会推送最后 7 篇论文的题目与摘要，供读者学习。
学术财经研究团队翻译。
克服歧视：骚扰和歧视动态
Yi Chen, Adam Dearing, and Michael Waldman #33065
Abstract: A common feature of historical episodes in which integration was successful, as well as episodes where integration was unsuccessful, is the aggravated harassment of the early pathbreakers who put themselves at risk by violating the previous segregated norm. Examples abound including Jackie Robinson and Larry Doby in the case of Major League Baseball, Autherine Lucy who was the first Black student at the University of Alabama, and Jane Chastain and Melissa Ludtke who were early female sports reporters. In this paper, we explore from a theoretical perspective the role of harassment of what we refer to as integration pathbreakers in the success and speed with which integration occurs. In our model of labor market discrimination, harassment occurs because the harassers receive direct and immediate utility from harassing, but also because harassment has the potential to slow down or even stop integration. Our main result is that such a setting can exhibit path dependence, where the success or failure of the early integration pathbreakers can be pivotal for the success and speed of the subsequent integration process. That is, early success is more likely to be followed by successful and faster integration than early failure, even when the early success is not due to aspects of the environment that make integration easier. In addition to our formal theoretical analysis of the role of harassment in the success and speed of integration, we apply our results to various historical episodes.
摘要：在历史上，成功与不成功的融合事件中一个共同的特征是，早期的开拓者因打破原有的隔离规范而遭受加剧的骚扰。这类例子包括美国职业棒球大联盟的杰基·罗宾森和拉里·多比、阿拉巴马大学首位黑人学生奥瑟琳·露西，以及早期女性体育记者简·查斯坦和梅丽莎·路德克。在本文中，我们从理论角度探讨了我们称之为“融合开拓者”受到骚扰的作用，以及其对融合成功与速度的影响。在我们关于劳动力市场歧视的模型中，骚扰发生的原因不仅在于骚扰者通过骚扰获得直接和即时的满足感，还因为骚扰有可能延缓甚至阻止融合。我们的主要结论是，这种环境中可能会出现路径依赖现象，即早期融合开拓者的成功或失败对后续融合进程的成败和速度至关重要。也就是说，早期的成功更可能带来更成功和更快的融合，即使早期的成功并不是由于外部环境的因素使融合变得更容易。除了对骚扰在融合成功和速度中的作用进行正式的理论分析之外，我们还将我们的结果应用于各种历史事件。
信息外部性、搭便车与英国石油行业的最优勘探
Charles Hodgson #33067
Abstract: Information spillovers between firms can reduce R&D incentives if competitors can free ride on innovations. However, strong property rights may impede cumulative research and lead to inefficient duplication. These effects are particularly relevant in natural resource exploration, where discoveries are spatially correlated. Using UK offshore oil exploration data, I estimate a dynamic model that captures the trade-off between drilling now and waiting to learn from competitors. Removing free-riding incentives increases industry surplus by 52%, while perfect information flow raises it by 24%. Counterfactual policy simulations highlight a trade-off in property rights design: stronger property rights over exploration well data increase the rate of exploration, while weaker property rights increase the efficiency and speed of learning but reduce the rate of exploration. Spatial clustering of each firm's drilling licenses both reduces the incentive to free ride and increases the speed of learning.
摘要：如果竞争对手能够搭便车利用创新，企业之间的信息溢出可能会减少研发激励。然而，强有力的产权保护可能会阻碍累积性研究，并导致低效的重复工作。这些效应在自然资源勘探中尤为重要，因为发现具有空间相关性。利用英国近海石油勘探数据，我估计了一个动态模型，捕捉了现在钻探与等待从竞争对手那里获取信息之间的权衡。消除搭便车激励可以使行业总盈余增加52%，而完美的信息流可以使盈余增加24%。反事实政策模拟强调了产权设计中的一个权衡：对勘探井数据加强产权保护可以提高勘探率，而弱化产权保护可以提高学习的效率和速度，但会降低勘探率。各公司钻探许可证的空间聚集既减少了搭便车的动机，也加快了学习的速度。
喂养老虎：斯里兰卡的侨汇与冲突
Barthélémy Bonadio, Andrei A. Levchenko, Dominic Rohner, and Mathias Thoenig #33062
Abstract: This paper estimates and quantifies the impact of the diaspora remittance flows on the conflict intensity and outcomes in the Sri Lankan Civil War during the period 1996-2009. We develop an approach to infer which remittance inflows were likely to benefit the Tamil Tiger rebels relative to the central government based on Facebook connections data at the subnational level. Using shocks to source country remittance outflows, we show that exogenous increases in remittances accessible to the Tamil Tigers significantly increased their fighting strength. We then set up a quantitative model of two-sided armed conflict over many contested geographic locations, augmented with remittance flows that affect the fighting strengths of the two sides. We structurally estimate the key parameters using remittance and conflict data, and calibrate the model to the Sri Lankan subdistricts over the period of the conflict. Our main quantitative finding is that remittances had a significant impact on the timing of the central government victory, and were a substantially more important component of the military strength of the Tamil Tigers than of the government. Remittances that favored the Tamil Tiger rebels may have prolonged the war by as much as 14 years.
摘要：本文估计并量化了侨汇流入对1996年至2009年斯里兰卡内战冲突强度和结果的影响。我们开发了一种方法，基于次国家层面的Facebook连接数据，推断哪些侨汇流入可能更有利于泰米尔猛虎组织，而不是中央政府。通过利用侨汇来源国资金流出量的冲击，我们发现，泰米尔猛虎组织能够获取的侨汇外生增加显著增强了其战斗力。接着，我们构建了一个涉及多个争夺地理位置的双边武装冲突的定量模型，并将侨汇流入考虑在内，这些侨汇影响了双方的战斗实力。我们利用侨汇和冲突数据对关键参数进行了结构估计，并将模型校准至冲突期间的斯里兰卡次区。我们的主要定量发现是，侨汇对中央政府胜利的时间节点产生了显著影响，而且侨汇对泰米尔猛虎组织军事力量的贡献远大于对政府的贡献。支持泰米尔猛虎组织的侨汇可能使战争延长了多达14年。
收入分布中的退休激励与决策：来自加拿大的证据
Kevin S. Milligan and Tammy Schirle #33069
Abstract: We evaluate the retirement incentives embedded in Canada’s retirement income system with attention to where individuals are located in the income distribution. We find that larger social security benefits are available to individuals with lower earnings in their work history because of the benefit income tests, but those from the top of the income distribution tend to enjoy longer lives over which they may receive benefits. Overall, we see greater Social Security Wealth among individuals from lower deciles. The implicit tax rates on continued work tend to be higher for workers from lower-earning deciles. Considering changes to actuarial adjustments associated with early pension take up, these implicit tax rates on work at older ages fell substantially after 2011. Our regression estimates confirm the importance of incentives on retirement behavior, with substantially larger effects for individuals in lower deciles. These effects are greater for women than men. In simulations, we show that changes to the actuarial adjustment had some impact on retirement rates by lowering the implicit tax on work. The overall redistributive effect of these induced retirement changes was fairly small, however, as the actuarial adjustments brought the system closer to actuarial fairness.
摘要：我们评估了加拿大退休收入体系中嵌入的退休激励，特别关注个人在收入分布中的位置。研究发现，由于福利收入测试，工作历史中收入较低的个人可以获得更多的社会保障福利，而收入分布顶端的个人则往往享有更长的寿命，从而可以更长时间领取福利。总体而言，收入较低的个体拥有更高的社会保障财富。继续工作的隐性税率通常对收入较低的工人更高。考虑到与提前领取养老金相关的精算调整变化，这些老年工作者的隐性税率在2011年后显著下降。我们的回归估计证实了激励措施对退休行为的重要性，尤其对收入较低的个人影响更大，并且这种影响对女性的作用比男性更显著。在模拟中，我们展示了精算调整的变化通过降低工作的隐性税率对退休率产生了一定影响。然而，这些引发的退休变化的总体再分配效应相对较小，因为精算调整使该体系更接近精算公平。
美国同化与分配政策对美洲原住民死亡率的影响
Grant Miller, Jack Shane, and C. Matthew Snipp #33057
Abstract: In contrast to earlier United States policies of open war, forcible removal, and relocation to address the “Indian Problem,” the Dawes Act of 1887 focused on assimilation and land severalty — making American Indians citizens of the United States with individually-titled plots of land rather than members of collective tribes with communal land. Considerable scholarship shows that the consequences of the policy differed substantially from its stated goals, and by the time of its repeal in 1934, American Indians had lost two-thirds of all native land held in 1887 (86 million acres)—and nearly two-thirds of American Indians had become landless or unable to meet subsistence needs. Complementing rich qualitative history, this paper provides new quantitative evidence on the impact of the Dawes Act on mortality among American Indian children and adults. Using 1900 and 1910 U.S. population census data to study both household and tribe-level variation in allotment timing, we find that assimilation and allotment policy increased various measures of American Indian child and adult mortality from nearly 20% to as much as one third (implying a decline in life expectancy at birth of about 20%) — confirming contemporary critics’ adamant concerns about the Dawes Act.
摘要：与早期美国通过公开战争、强制迁移以及重新安置来解决“印第安人问题”的政策不同，1887年的《道斯法》侧重于同化和土地分配——将美洲原住民变为美国公民，并分配个人拥有的土地，而不是作为集体部落成员拥有公共土地。大量研究表明，该政策的后果与其声明的目标大相径庭。到1934年该法案废除时，美洲原住民已经失去了1887年所持有的三分之二的土地（8600万英亩），并且近三分之二的美洲原住民失去了土地，或无法满足生存需求。本文在丰富的定性历史研究基础上，提供了关于《道斯法》对美洲原住民儿童和成人死亡率影响的新量化证据。通过分析1900年和1910年美国人口普查数据，研究家庭和部落层面的土地分配时间差异，我们发现，同化和分配政策使美洲原住民儿童和成人的死亡率增加了约20%至三分之一（意味着出生时预期寿命下降约20%），这证实了当时批评者对《道斯法》的强烈担忧。
套利的统计极限
Rui Da, Stefan Nagel, and Dacheng Xiu #33070
Abstract: We investigate the economic consequences of statistical learning for arbitrage pricing in a high-dimensional setting. Arbitrageurs learn about alphas from historical data. When alphas are weak and rare, estimation errors hinder arbitrageurs—even those employing optimal machine learning techniques—from fully exploiting all true pricing errors. This statistical limit to arbitrage widens the equilibrium bounds of alphas beyond what traditional arbitrage pricing theory predicts, leading to a significant divergence between the feasible Sharpe ratio achievable by arbitrageurs and the unattainable theoretical maximum under perfect knowledge of alphas.
摘要：我们研究了统计学习对高维环境下套利定价的经济影响。套利者通过历史数据学习α（超额收益）。当α较弱且稀少时，即使是使用最优机器学习技术的套利者，估计误差也会阻碍他们完全利用所有真实的定价错误。这种套利的统计极限使α的均衡范围超出传统套利定价理论的预测，导致套利者能够实现的可行夏普比率与在完全掌握α的情况下无法实现的理论最大值之间出现显著差异。
估计大学生之间的同伴效应：来自STEM领域一对一配对实地实验的证据
Robert W. Fairlie, Daniel M. Oliver, Glenn Millhauser, and Randa Roland #33060
Abstract: An extensive literature in the social sciences analyzes peer effects among students, but estimation is complicated by several major problems some of which cannot be solved even with random assignment. We design a field experiment and propose a new estimation technique to address these estimation problems including the mechanical problems associated with repeated observations within peer groups noted by Angrist (2014). The field experiment randomly assigns students to one-to-one partnerships in an important gateway STEM course at a large public university. We find no evidence of peer effects from estimates of exogenous peer effect models. We push further and estimate outcome-on-outcome models which sometimes reveal peer effects when exogenous models do not provide good proxies for ability. We find some limited evidence of small, positive outcome-on-outcome peer effects (which would have been missed without our new estimation technique). Standard estimation methods fail to detect peer effects and even return negative estimates in our Monte Carlo simulations because of the downward bias due to mechanical problems. Simulations reveal additional advantages of our technique especially when peer group sizes are fixed. Estimates of non-linear effects, heterogeneous effects, and different measures of peer ability and outcomes reveal mostly null effects but we find some evidence that low-ability peers negatively affect low-ability and medium-ability students. The findings in this setting of long-term, intensive interactions with classroom random assignment and “throwing everything at it” provide evidence of, at most, small positive peer effects contrasting with the common finding of large peer effects in previous studies in education.
摘要：社会科学中有大量文献分析了学生之间的同伴效应，但由于多个主要问题的存在，估计这些效应变得复杂，其中一些问题即使通过随机分配也无法解决。我们设计了一项实地实验，并提出了一种新的估计技术，以解决这些估计问题，包括Angrist（2014）提到的与同伴群体中重复观察相关的机械性问题。该实地实验在一所大型公立大学的重要STEM入门课程中，随机分配学生进行一对一配对。我们通过对外生同伴效应模型的估计，未发现同伴效应的证据。进一步的分析中，我们估计了基于结果的模型（outcome-on-outcome），这些模型有时在外生模型无法有效预测能力时揭示出同伴效应。我们发现了一些有限的小规模、正向的基于结果的同伴效应（如果没有我们提出的新估计技术，这些效应可能会被忽视）。标准的估计方法未能检测到同伴效应，甚至在我们的蒙特卡罗模拟中返回了负向估计值，这是由于机械性问题导致的向下偏差。模拟结果揭示了我们技术的额外优势，尤其是在同伴组大小固定的情况下。非线性效应、异质性效应和不同的同伴能力及结果测量的估计大多显示无效应，但我们发现一些证据表明，低能力的同伴对低能力和中等能力学生产生了负面影响。该实验设定为长期、深入的课堂互动，并采用了随机分配等多种方法，总体结果显示同伴效应最多为小规模的正向效应，与以往教育研究中常见的大规模同伴效应的结论形成对比。
往期精选：
