研究应用隐私增强技术的理由

Claire M. Bowen, Joshua Snoke, Aaron R. Williams, and Andrés F. Barrientos #32909

Abstract: Research on privacy enhancing approaches for sharing data has grown significantly over the past two decades. This increased interest has led to extensive theoretical and methodological research, but the number of practical applications of privacy enhancing technologies has lagged far behind. This paper provides an overview of the Safe Data Technologies Project and the approach we have taken to conducting privacy research with the specific aim of putting theory into practice and incorporating user input. We provide an overview of the broader project goals, which aim to safely expand access to administrative tax data that is currently highly restricted. We highlight how understanding user interactions with the privacy enhancing methods has driven our research path and challenged the often unrealistic assumptions underlying much of the theoretical work. We review the primary findings from our research, discuss our plans for future directions, and make the case for researchers to pursue similar lines of applied inquiry.

摘要：过去二十年来，关于数据共享的隐私增强方法的研究显著增长。这种兴趣的增加带来了大量的理论和方法学研究，但隐私增强技术的实际应用数量却远远落后于理论发展。本文概述了“安全数据技术项目”以及我们为将理论付诸实践并结合用户反馈所采取的隐私研究方法。我们介绍了该项目的总体目标，即安全扩大目前高度受限的行政税务数据的访问权限。我们强调了对用户与隐私增强方法的互动理解如何引导我们的研究方向，并挑战了许多理论工作中常见的不切实际的假设。我们回顾了研究中的主要发现，讨论了未来的研究计划，并为研究人员探讨类似的应用研究方向提出了理由。

（虚）假信息战争

Adrian Casillas, Maryam Farboodi, Layla Hashemi, Maryam Saeedi, and Steven Wilson #32896

Abstract: Over the past decade, social media platforms have emerged as prominent vehicles for displaying dissent. In response, various actors have increasingly spread fake news on these platforms to impair the opposition—the (dis)information war. We analyze a methodology to identify disinformation using network-based characteristics of the news initiators, and use data from Twitter (now X) to assess the effectiveness of this method in limiting the spread of disinformation. We find that it detects at least 85% of verified instances of disinformation without misidentifying any true news, and reduces both account engagement and lifespan of disinformation by at least a factor of two, highlighting the importance of swift discovery of disinformation to interrupt its exponential spread.

摘要：在过去的十年中，社交媒体平台已成为展示异议的主要渠道。对此，各方势力越来越多地在这些平台上传播虚假新闻，以削弱反对派的力量——这就是所谓的（虚）假信息战争。我们分析了一种基于网络特征来识别虚假信息的研究方法，并使用来自Twitter（现称X）的数据评估该方法在限制虚假信息传播方面的有效性。我们发现该方法能够检测到至少85%的已验证虚假信息实例，且不会错误识别任何真实新闻。同时，它能将虚假信息的账户互动和寿命至少减少一半，强调了迅速发现虚假信息对于打断其指数传播的重要性。

更广泛的覆盖范围总是更好的吗？既有扭曲、排放弹性与排放定价范围

Lawrence H. Goulder, Marc A.C. Hafstead, and Roberton C. Williams III #32915

Abstract: Economists often regard broad-based carbon pricing (whether in the form of a carbon tax or cap and trade) as the most efficient policy to reduce carbon dioxide emissions. Relative to a narrower policy that exempts some emissions sources, a broader policy is often favored because it can exploit more low-cost emissions reduction opportunities and cause less emissions leakage to uncovered sources. Yet narrower approaches have gained considerable political support, partly because they avoid price increases for outputs (such as gasoline) regarded as especially critical to household budgets. Some analysts might lament any departure from broad carbon pricing, citing efficiency costs. This paper offers theory and numerical simulations revealing that such a shift need not sacrifice efficiency. This result reflects differences across sectors in distortions from preexisting taxes and in the elasticity of emissions with respect to the carbon price. Our analytical model reveals that a narrower policy that exploits these differences can be more cost-effective than a policy with a broad, economy-wide tax base. Our numerical model of the US economy compares quantitatively the effects of an economy-wide carbon price with those of several narrower policies, including one that applies only to the power sector, one that exempts gasoline, and one that exempts energy-intensive trade-exposed industries. We compare policies under alternative specifications for policy stringency and find that the broader policy always becomes more cost-effective at sufficiently high stringency.

摘要：经济学家通常认为广泛的碳定价（无论是碳税还是排放交易）是减少二氧化碳排放的最有效政策。相对于排除某些排放源的更窄政策，广泛政策通常更受青睐，因为它可以利用更多低成本的减排机会，并减少未覆盖排放源的排放泄漏。然而，较窄的政策获得了相当大的政治支持，部分原因是它避免了对一些对家庭预算尤其重要的产品（如汽油）价格上涨。一些分析人士可能会对偏离广泛的碳定价表示遗憾，认为这会带来效率损失。本文通过理论分析和数值模拟表明，这种转变不一定会牺牲效率。这个结果反映了各部门在既有税收扭曲和排放对碳价弹性方面的差异。我们的分析模型表明，利用这些差异的较窄政策可能比覆盖全经济税基的广泛政策更具成本效益。我们基于美国经济的数值模型定量比较了全经济碳定价与几种较窄政策的影响，包括仅适用于电力行业的政策、豁免汽油的政策以及豁免能源密集型贸易暴露行业的政策。我们在不同政策严格性设定下比较这些政策，发现当政策严格性足够高时，广泛政策总是变得更具成本效益。

贸易、创新与企业融资

Paul Bergin, Ling Feng, and Ching-Yi Lin #32904

Abstract: While the trade literature has tended to view export activity and innovation as complementary activities, we present evidence that financial constraints are a reason the two activities can act as substitutes for small exporters. In particular, we find that small exporters have lower expenditure on R&D than comparable non-exporters, and we find a corresponding pattern in the leverage ratio of the capital structure of small firms. A model that combines firm decisions regarding the amount of innovation, exporting, and endogenous financial capital structure is able to account for these empirical findings. The model implies that small firms are unable to fully reap the gains from exporting due to financial constraints, as they reduce R&D to finance the costs of export participation.

摘要：虽然贸易文献倾向于将出口活动和创新视为互补活动，但我们提供的证据表明，金融约束是导致这两项活动对小型出口商而言可以相互替代的原因。具体而言，我们发现小型出口商在研发上的支出低于相似的非出口商，且在小型企业资本结构的杠杆率中也呈现出相似的模式。一个结合企业在创新、出口和内生金融资本结构方面决策的模型能够解释这些实证发现。该模型表明，由于金融约束，小型企业无法充分获得出口带来的收益，因为它们减少了研发支出来为出口参与成本融资。

分散经济中的最优货币和财政政策

Lydia Cox, Jiacheng Feng, Gernot Müller, Ernesto Pastén, Raphael Schoenle, and Michael Weber #32914

Abstract: The jointly optimal monetary and fiscal policy mix in a multi-sector New Keynesian model with sectoral government spending and productivity shocks entails a separation of roles: Sectoral government spending optimally adjusts to sectoral output gaps and inflation rates---a policy supported by evidence from sectoral federal procurement data. Monetary policy optimally focuses on aggregate stabilization, but deviates from a zero-inflation target; in a model calibration to the U.S., however, it effectively approximates a zero-inflation target. Because monetary policy is a blunt instrument and government spending trades off stabilization against the optimal-level public good provision, the first best is not achieved.

摘要：在一个具有部门政府支出和生产率冲击的多部门新凯恩斯主义模型中，最优的货币和财政政策组合涉及角色分工：部门政府支出最优地根据部门产出缺口和通胀率进行调整——这一政策得到了部门联邦采购数据的支持。货币政策最优地专注于整体经济的稳定，但偏离零通胀目标；然而，在对美国的模型校准中，货币政策有效地接近于零通胀目标。由于货币政策是一种钝性工具，且政府支出需要在稳定性与最优公共产品供给水平之间进行权衡，无法实现“第一最佳”状态。

从主观预期数据估计灵活的收入过程：来自印度和哥伦比亚的证据

Manuel Arellano, Orazio Attanasio, Samuel Crossman, and Víctor Sancibrián #32922

Abstract: We develop a methodology for modeling household income processes when subjective probabilistic assessments of future income are available. This allows us to flexibly estimate conditional cdfs directly using elicited individual subjective probabilities, and to obtain empirical measurements of subjective risk and persistence. We then use two longitudinal surveys collected in rural India and rural Colombia to explore the nature of income dynamics in those contexts. Our results suggest linear income processes are rejected in favor of more flexible versions in both cases; subjective income distributions feature heteroskedasticity, conditional skewness and nonlinear persistence.

摘要：我们开发了一种方法，用于在可以获得未来收入的主观概率评估时对家庭收入过程进行建模。这使我们能够通过个人主观概率的提取，灵活地直接估计条件累积分布函数（cdf），并获得主观风险和持续性的实证测量。随后，我们利用在印度农村和哥伦比亚农村收集的两项纵向调查，探讨这些背景下的收入动态。我们的研究结果表明，线性收入过程在两种情况下都被更灵活的模型所取代；主观收入分布表现出异方差性、条件偏度和非线性持续性。

信用评分：表现与公平

Stefania Albanesi and Domonkos F. Vamossy #32917

Abstract: Credit scores are critical for allocating consumer debt in the United States, yet little evidence is available on their performance. We benchmark a widely used credit score against a machine learning model of consumer default and find significant misclassification of borrowers, especially those with low scores. Our model improves predictive accuracy for young, low-income, and minority groups due to its superior performance with low quality data, resulting in a gain in standing for these populations. Our findings suggest that improving credit scoring performance could lead to more equitable access to credit.

摘要：信用评分在美国消费者债务分配中起着关键作用，但关于其表现的证据却很少。我们将一种广泛使用的信用评分与一个基于机器学习的消费者违约模型进行比较，发现借款人，尤其是低信用评分人群中存在显著的错误分类。由于我们的模型在处理低质量数据时表现更优，它提升了对年轻人、低收入群体和少数族裔群体的预测准确性，从而改善了这些群体的信用状况。我们的研究结果表明，提升信用评分的表现可能有助于实现更公平的信贷获取。

法官评判法官：联邦巡回上诉法院中的党派性与政治

Alma Cohen and Rajeev H. Dehejia #32920

Abstract: We examine how politicization and polarization influence judicial review within U.S. Federal appellate courts. Analyzing over 400,000 cases from 1985 to 2020, we find that judges' political alignment or misalignment with trial judges increasingly affect their decisions, particularly in the last two decades. This trend is significant in precedential cases: panels of Democratic judges are 6.9 percentage points more likely to reverse Republican trial judges compared to Democratic ones, whereas Republican panels are 3.6 percentage points less likely to reverse fellow Republican judges. This effect persists across ideological and non-ideological cases and even among judges appointed before 2000.

摘要：我们研究了政治化和极化如何影响美国联邦上诉法院的司法审查。通过分析1985年至2020年间的40多万起案件，我们发现法官与初审法官的政治立场一致或不一致，越来越多地影响他们的裁决，特别是在过去二十年中。这一趋势在具有先例意义的案件中尤为显著：由民主党任命的法官组成的小组推翻共和党初审法官裁决的可能性比推翻民主党法官裁决的可能性高出6.9个百分点，而共和党法官小组推翻共和党初审法官裁决的可能性则低3.6个百分点。这种效应在意识形态相关和非意识形态案件中都存在，甚至在2000年之前任命的法官中也有所体现。

民主、资本主义与平等：非个人规则的重要性

Naomi R. Lamoreaux and John Joseph Wallis #32930

Abstract: We usually consider it progress when a country begins to shift from an autocratic to a democratic form of government. However, the introduction of elections and other early trappings of democracy often has the perverse effect of exacerbating political instability. It also increases the incentives for those in power to manipulate the economy for political ends and thus often negatively affects economic growth. We argue that the key to getting beyond these pernicious effects—to reconciling democracy and capitalism—is to move to a governance structure based on impersonal rules that apply in the same way to everyone (or at least to broad categories of everyone). We lay out the theoretical basis for this argument and illustrate it with evidence about how the transformation worked (or not) in the case of the United States, United Kingdom, and Germany.

摘要：我们通常认为，一个国家从独裁政体转向民主政体是进步的表现。然而，引入选举和其他早期民主形式往往会产生相反的效果，加剧政治不稳定。这也增加了当权者出于政治目的操纵经济的动机，因此往往对经济增长产生负面影响。我们认为，克服这些有害影响、调和民主与资本主义的关键在于建立一个基于非个人规则的治理结构，这些规则以同样的方式适用于每个人（或至少适用于广泛的群体）。我们阐述了这一论点的理论基础，并通过美国、英国和德国的转型案例来说明这一理论是如何运作（或未能运作）的。

自工资弹性：量化最低工资对就业的影响

Arindrajit Dube and Ben Zipperer #32925

Abstract: The own-wage elasticity (OWE) of employment estimated using minimum wage increases provides an economically meaningful measure of the policy on jobs. We discuss how to interpret the magnitude of the OWE, including in terms of welfare and under alternative models of the labor market. We present a comprehensive set of OWE estimates from 88 studies and introduce an regularly updated repository of the estimates---https://economic.github.io/owe---an up-to-date snapshot of the existing literature for scholars and policymakers. We find that most studies to date suggest a fairly modest impact of minimum wages on jobs: the median OWE estimate of 72 studies published in academic journals is -0.13, which suggests that only around 13 percent of the potential earnings gains from minimum wage increases are offset due to associated job losses. Estimates published since 2010 tend to be closer to zero.

摘要：通过最低工资上调估算的就业自工资弹性（OWE）提供了一种对政策影响就业的有意义的经济衡量标准。我们讨论了如何解释OWE的幅度，包括在福利方面以及不同的劳动市场模型下的解释。我们展示了来自88项研究的全面OWE估算结果，并引入了一个定期更新的估算库——https://economic.github.io/owe——为学者和政策制定者提供现有文献的最新快照。我们发现，大多数研究表明最低工资对就业的影响相对温和：72篇发表在学术期刊中的研究的OWE估算中值为-0.13，这意味着只有大约13%的潜在工资增长因相关的失业而被抵消。自2010年以来发布的估算结果则更接近于零。

理解低成本贷款对强迫劳动的影响

Anisha Sharma, Manisha Shah, and Beata Łuczywek #32912

Abstract: Approximately 27.5 million individuals fell victim to forced labor in 2021. The Indian construction industry is particularly vulnerable to forced labor as workers experience excessive work hours, required work on rest days, and unpaid wages. Micro-contractors (MCs), who oversee worker environments, frequently struggle with their own financial constraints due to limited access to working capital. This study investigates whether alleviating MC liquidity constraints improves labor conditions for their workers in Bengaluru and Delhi by offering randomly selected MCs access to low-cost loans. Our findings reveal this intervention does not improve working conditions overall; in fact, some outcomes slightly worsen. However, workers employed by more educated and non-migrant treatment MCs experience significantly better labor conditions, underscoring important heterogeneity among MCs. This research offers new causal insights into efforts to combat forced labor.

摘要：2021年，约有2750万人沦为强迫劳动的受害者。印度建筑业尤其易受强迫劳动的侵害，工人面临超长工作时间、休息日被强制工作以及未付工资等问题。负责工人工作环境的微型承包商（MCs）由于缺乏工作资金，常常面临财务困境。该研究通过为班加罗尔和德里的随机选择的微型承包商提供低成本贷款，调查缓解MC的流动性约束是否能改善其工人的劳动条件。研究结果表明，这一干预总体上并未改善工作条件，某些结果甚至略有恶化。然而，受教育程度较高的MC和非移民MC雇佣的工人确实经历了显著更好的劳动条件，突显了MC之间的重要差异性。这项研究为打击强迫劳动的努力提供了新的因果见解。

餐饮业就业、最低工资与边界不连续性

Arindrajit Dube, Michael Reich, Akash Bhatt, and Denis Sosinskiy #32902

Abstract: Dube, Lester and Reich (2010, DLR), using state minimum wage discontinuities across bordering counties and Quarterly Census of Employment and Wages data, did not detect negative minimum wage effects on restaurant employment. Jha, Neumark and Rodriguez-Lopez (2024, JNR) claim that looking within multi-state commuting zones and using County Business Patterns data provides a superior approach to DLR and does find disemployment effects. We show that JNR’s results are confounded by parallel trends violations in the 1990s, when minimum wage events were rare and small in magnitude; JNR’s outmoded two-way-fixed-effects model amplifies the biases introduced by these violations. Our estimates using their specifications and data on only post-2000 data fail to detect disemployment effects. The same results hold using QCEW and ACS datasets. Our preferred event study difference-in-differences approach, which analyzes only data that fall clearly within an event’s window, also does not detect negative employment effects. This result holds whether we compare across all states, look within commuting zones or within border county pairs, and regardless of the data set or time period.

摘要：Dube、Lester 和 Reich（2010，DLR）使用跨州边界县的最低工资不连续性以及《就业与工资季度普查》数据，未检测到最低工资对餐饮业就业的负面影响。Jha、Neumark 和 Rodriguez-Lopez（2024，JNR）声称，基于多州通勤区内的分析并使用《县商业模式》数据，比DLR的方法更优，并发现了负面就业影响。我们表明，JNR的结果受到1990年代平行趋势违反的影响，当时的最低工资调整事件罕见且幅度较小；JNR的过时的双向固定效应模型放大了这些偏差。使用他们的规范和仅限2000年后的数据进行估算，我们未能检测到就业减少效应。使用《就业与工资季度普查》和《美国社区调查》数据集得出的结果相同。我们首选的事件研究双重差分方法，仅分析明确落在事件窗口内的数据，也未检测到负面就业影响。这一结果在跨州比较、通勤区内比较或边界县对比较中均成立，且与所用数据集或时间段无关。

往期精选：

      重磅：学术财经•学术笔记全汇总

重磅：学术财经全球价值链专题学术笔记大汇总（附下载链接失效学术笔记推文）

重磅| 基金申请内部学习和交流

【学术笔记·第37篇】风险和全球供应链：我们知道什么和我们需要知道什么

【学术笔记·第36篇】各有所长：全球供应链风险敞口（Exposure）测度

【学术笔记·第35篇】日本生产者价格指数理论与实证：最终需求-中间需求加总系统

【学术笔记·第34篇】北美自由贸易的贸易和福利影响估计

【学术笔记 第33篇】 英属印度时期铁路：交通基础设施的影响评估（AER）

【学术笔记 第32篇】 技术、地理和贸易(经典EK模型）

【学术笔记 第31篇】2018年贸易战对美国物价和福利的影响

【学术笔记 第30篇】全球价值链和国家部门层面实际有效汇率