关于工具变量的材料包, 标题,模型,内生变量,工具变量

凡是搞计量经济的，都关注这个号了

邮箱：econometrics666@sina.cn

所有计量经济圈方法论丛的code程序, 宏微观数据库和各种软件都放在社群里.欢迎到计量经济圈社群交流访问.

@因果推断研究小组撰写

一个关于工具变量的材料。包括：论文标题、模型、内生变量是什么（为什么有内生性）、工具变量是什么（为什么恰当）。在学术这件事情上，因果推断研究小组绝对是认真的，也必定会集合海内外学者推进前沿研究。

1. Angrist, J., & Krueger, A. (1991). Does Compulsory School Attendance Affect Schooling and Earnings? The Quarterly Journal of Economics, 106(4), 979-1014. Retrieved from http://www.jstor.org/stable/2937954

Abstract: We establish that season of birth is related to educational attainment because of school start age policy and compulsory school attendance laws. Individuals born in the beginning of the year start school at an older age, and can therefore drop out after completing less schooling than individuals born near the end of the year. Roughly 25 percent of potential dropouts remain in school because of compulsory schooling laws. We estimate the impact of compulsory schooling on earnings by using quarter of birth as an instrument for education. The instrumental variables estimate of the return to education is close to the ordinary least squares estimate, suggesting that there is little bias in conventional estimates.

Model:

Angrist & Krueger（1991）利用美国的男性人口调查数据，提出了educ的一个巧妙的二值工具变量。如果该男性是在第一季度出生的，令frstqrt等于1，否则为0。(2) 式中的误差项——特别是能力——与出生季度不相关，即出生季度只通过影响教育从而影响工资。但是，frstqrt还要与educ相关。事实表明，在基于出生季度的总体中，教育年数确实有系统性差异。Angrist和Krueger认为是缘于在各州实行的义务就学法，这很有说服力。简单地说，年初出生的学生往往入学较晚。因此，他们在达到义务教育年龄时（大部分州定为16岁），所受的教育略少于入学较早的学生。

2. Angrist, J. (1990). Lifetime Earnings and the Vietnam Era Draft Lottery: Evidence from Social Security Administrative Records. The American Economic Review, 80(3), 313-336. Retrieved from http://www.jstor.org/stable/2006669

Abstract: The randomly assigned risk of induction generated by the draft lottery is used to construct estimates of the effect of veteran status on civilian earnings. These estimates are not biased by the fact that certain types of men are more likely than others to service in the military. Social Security administrative records indicate that in the early 1980s, long after their service in Vietnam was ended, the earnings of white veterans were approximately 15 percent less than the earnings of comparable nonveterans.

Model:

Angrist（1990）研究了，参加越南战争的老兵，其终身收入因参加越战而受到的影响。一个简单模型为：income= veteran + others，其中veteran是二值变量。不过可能存在一个自我选择（self-selection）的问题：也许人们因为能从军队中得到最多的收入而选择参军，或者参军的决策与其他对收入有影响的特征相关。这些问题将导致veteran与u相关。

Angrist指出，越南战争的征兵抽签提供了一个自然试验（natural experiment），从而产生了veteran的一个工具变量。年轻人被分给的征兵抽签号决定了他们是否会被征召去服役于越南战争。因为所分给的号码（毕竟）是随机分配的，征兵抽签号与误差项u不相关似乎是可信的。而得到号码足够小（指号码小于某个数）的人必须服役于越南战争，使得成为老兵的概率与抽签号相关。如果以上两点都是正确的，征兵抽签号是veteran的一个好的IV候选者。

3. Evans, W., Oates, W., & Schwab, R. (1992). Measuring Peer Group Effects: A Study of Teenage Behavior. Journal of Political Economy, 100(5), 966-991. Retrieved from http://www.jstor.org/stable/2138631

Abstract: Individuals or households often have some scope for choice of peer groups, whether through the selection of neighborhood of residence, school, or friends. This study addresses the estimation of peer group effects in cases in which measures of peer group influence are potentially endogenous variables. Using a rich data set on individual behavior, the paper explores teenage pregnancy and school dropout behavior. For both cases, the estimation of a straightforward single-equation model yields statistically significant peer group effects; however, these effects disappear under simultaneous equation estimation. The results are robust and suggest the need for careful modeling of the choice of peer groups.

Model:

Our measure of the peer group effect , LOG(DISADVANTAGED), is the natural log of the percentage of students in the respondent's school who were classified as economically disadvantaged.

埃文斯等( Evans et al., 1992) 试图验证学校中的贫困生比例对学生怀孕或辍学行为是否有显著影响。他们运用大都会地区的失业率、家庭收入中位数和贫困率作为学校中贫困学生比例的工具变量。其理由是: 以都会为单位的失业率和贫困率必然和辖区内学校的贫困生比例有关，但又不直接影响学生的怀孕或辍学等行为。

4. BENTOLILA, S., MICHELACCI, C. and SUAREZ, J. (2010), Social Contacts and Occupational Choice. Economica, 77: 20-45. doi:10.1111/j.1468-0335.2008.00717.x

Abstract: Social contacts help to find jobs, but not necessarily in the occupations where workers are most productive. Hence social contacts can generate mismatch between workers' occupational choices and their productive advantage. Accordingly, social networks can lead to low labour force quality, low returns to firms' investment and depressed aggregate productivity. We analyse surveys from both the US and Europe including information on job finding through contacts. Consistent with our predictions, contacts reduce unemployment duration by 1–3 months on average, but they are associated with wage discounts of at least 2.5%. We also find some evidence of negative externalities on aggregate productivity.

Model:

邦托利阿等( Bentolila et al., 2010) 使用联邦就业率作为工具变量来分析“使用社会关系”对个人收入以及失业期限是否具有影响。联邦就业率与个人收入没有直接关系。但就业率高，则在联邦内使用关系求职的必要性就低。

5. Hoxby, Caroline, M. 2000. "Does Competition among Public Schools Benefit Students and Taxpayers?" American Economic Review, 90 (5): 1209-1238.

Abstract: Tiebout choice among districts is the most powerful marketforce in American public education. Naive estimates of its effects are biased by endogenous district formation. I derive instruments from the natural boundaries in a metropolitan area. My results suggest that metropolitan areas with greater Tiebout choice have more productive public schools and less private schooling. Little of the effect of Tiebout choice works through its effect on household sorting. This finding may be explained by anotherfinding: students are equally segregated by school in metropolitan areas with greater and lesser degrees of Tiebout choice among districts.

Model:

霍克斯比( Hoxby, 2000)在一篇经典研究中，采用区域内河流数量作为该区域学校数量的工具变量，以此来验证学区内的学校竞争是否可以提高教学质量。区域内学校数量之所以是内生的，是因为它可能是该区域长期历史积累下的某种特征的结果。而使用河流数作为工具变量则具有很强的说服力:河流数量越多，就会因交通问题导致更多学校的设立; 但河流数是天然形成的，本身和教学质量无直接关系。

6. Cipollone, Piero, and Alfonso Rosolia. 2007. "Social Interactions in High School: Lessons from an Earthquake." American Economic Review, 97 (3): 948-965.

Abstract: After an earthquake hit Southern Italy in 1980, young men from certain towns were exempted from compulsory military service. We show that the exemption raised high-school-graduation rates of boys by more than 2 percentage points. We do this by comparing high-school-graduation rates of young exempt men and older nonexempt men from the least damaged areas and men of the same age groups from nearby towns that were not hit by the quake. Similar comparisons show that graduation rates of young women in the affected areas also increased. Since in Italy women are not subject to the draft, the findings suggest the presence of spillover effects.

Model:

在班级效应( classeffect) 研究中，西波隆和罗索利亚( Cipollone ＆ Rosolia， 2007) 以地震导致的男性免征兵政策作为高中班级性别构成的工具变量，以分析意大利学生中班级性别构成对女生成绩的影响。地震作为一种天象，显然是随机和外生的。

7. Kaivan Munshi, Networks in the Modern Economy: Mexican Migrants in the U. S. Labor Market, The Quarterly Journal of Economics, Volume 118, Issue 2, May 2003, Pages 549–599, https://doi.org/10.1162/003355303321675455

Abstract: This paper attempts to identify job networks among Mexican migrants in the U. S. labor market. The empirical analysis uses data on migration patterns and labor market outcomes, based on a sample of individuals belonging to multiple origin-communities in Mexico, over a long period of time. Each community's network is measured by the proportion of the sampled individuals who are located at the destination (the United States) in any year. We verify that the same individual is more likely to be employed and to hold a higher paying nonagricultural job when his network is exogenously larger, by including individual fixed effects in the employment and occupation regressions and by using rainfall in the origin-community as an instrument for the size of the network at the destination.

Model:

孟希( Munshi，2003)使用墨西哥移民来源地区的降水量作为移民数量的工具变量，证明了乡的移民越多，他们在美国获得工作的几率和收入会越高。移民来源社区的降雨量作为工具变量的理由是: 墨西哥某社区的降水量和美国的劳动力市场显然没有任何关联; 但降水量和社区的农业收入有关，并通过影响农业预期收入而影响到移民美国的决策。

8. Acemoglu, Daron, Simon Johnson, and James A. Robinson. 2001. "The Colonial Origins of Comparative Development: An Empirical Investigation." American Economic Review, 91 (5): 1369-1401.

Abstract: We exploit differences in European mortality rates to estimate the effect of institutions on economic performance. Europeans adopted very different colonization policies in different colonies, with different associated institutions. In places where Europeans faced high mortality rates, they could not settle and were more likely to set up extractive institutions. These institutions persisted to the present. Exploiting differences in European mortality rates as an instrument for current institutions, we estimate large effects of institutions on income per capita. Once the effect of institutions is controlled for, countries in Africa or those closer to the equator do not have lower incomes.

Model：

阿西莫格鲁等在一项经典研究( Acemogluet al., 2001) 中，把殖民地时代一个国家的自然死亡率作为该国当今制度的工具变量。其理由非常巧妙: 如果该地区当年的死亡率高，那么欧洲殖民者就相对不愿定居下来，从而在当地建立起更具掠夺性的“坏”制度。由于制度的“路径依赖”，殖民时代的制度显然和现在的制度关系密切。因此，历史上的死亡率作为工具变量，应该和当今制度紧密相关，而一百年前的死亡率作为一种自然生理现象，又和目前的人均收入没有直接关系。

9. Angrist, J., & Evans, W. (1998). Children and Their Parents' Labor Supply: Evidence from Exogenous Variation in Family Size. The American Economic Review, 88(3), 450-477. Retrieved from http://www.jstor.org/stable/116844

Abstract: Research on the labor-supply consequences of childbearing is complicated by the endogeneity of fertility. This study uses parental preferences for a mixed sibling-sex composition to construct instrumental variables (IV) estimates of the effect of childbearing on labor supply. IV estimates for women are significant but smaller than ordinary least-squares estimates. The IV are also smaller for more educated women and show no impact offamily size on husbands' labor supply. A comparison of estimates using sibling-sex composition and twins instruments implies that the impact of a third child disappears when the child reaches age 13.

Model:

安古瑞斯特和伊凡斯( Angrist & Evans, 1998) 试图分析家庭中的孩子数是否影响母亲的就业。由于生育孩子数量是可以被选择的，因此解释变量显然是内生的。为解决这一问题，他们巧妙地挖掘了人类生育行为中偏好有儿有女的特征，将子女“老大”和“老二”的性别组合情况作为工具变量。理由是: 头两胎如果是双子或双女，那么生育第三胎的可能性大大增加，进而增加子女数。而子女性别是完全随机的。

10. Nancy Qian, Missing Women and the Price of Tea in China: The Effect of Sex-Specific Earnings on Sex Imbalance, The Quarterly Journal of Economics, Volume 123, Issue 3, August 2008, Pages 1251–1285, https://doi.org/10.1162/qjec.2008.123.3.1251

Abstract: Economists have long argued that the sex imbalance in developing countries is caused by underlying economic conditions. This paper uses exogenous increases in sex-specific agricultural income caused by post-Mao reforms in China to estimate the effects of total income and sex-specific income on sex-differential survival of children. Increasing female income, holding male income constant, improves survival rates for girls, whereas increasing male income, holding female income constant, worsens survival rates for girls. Increasing female income increases educational attainment of all children, whereas increasing male income decreases educational attainment for girls and has no effect on boys' educational attainment.

Model:

钱楠筠(Qian, 2008) 巧妙地用茶叶的价格作为中国家庭中男性收入和女性收入之比的工具变量，证明了家庭收入的性别结构最终影响了中国家庭男女出生性别比例( 传统上我们相信家庭的总收入影响男女出生比例) 。用茶叶价格作为工具变量，是因为茶叶产业链的特性决定了从业人员以女性为主，茶叶价格提高就意味着女性在家庭中的经济地位提高，从而导致家庭女性胎儿被流产的几率降低。而茶叶的价格显然和家庭男女出生性别比例没有其他任何的因果逻辑联系。

11. Robert E. Hall, Charles I. Jones, Why do Some Countries Produce So Much More Output Per Worker than Others?, The Quarterly Journal of Economics, Volume 114, Issue 1, February 1999, Pages 83–116, https://doi.org/10.1162/003355399555954

Abstract: Output per worker varies enormously across countries. Why? On an accounting basis our analysis shows that differences in physical capital and educational attainment can only partially explain the variation in output per worker—we find a large amount of variation in the level of the Solow residual across countries. At a deeper level, we document that the differences in capital accumulation, productivity, and therefore output per worker are driven by differences in institutions and government policies, which we call social infrastructure. We treat social infrastructure as endogenous, determined historically by location and other factors captured in part by language.

Model:

豪尔和琼斯( Hall & Jones, 1999)非常具有想象力地用各国到赤道的距离作为工具变量。理由是，到赤道的距离大致反映了各国受西方制度影响的深浅程度，而这一距离显然是外生的。

12. Card, David. (1993). Using Geographic Variation in College Proximity to Estimate the Return to Schooling. NBER working papers.

Abstract: A convincing analysis of the causal link between schooling and earnings requires an exogenous source of variation in education outcomes. This paper explores the use of college proximity as an exogenous determinant of schooling. When college proximity is taken as an exogenous determinant of schooling the implied instrumental variables estimates of the return to schooling are 25-60% higher than conventional ordinary least squares estimates. Since the effect of a nearby college on schooling attainment varies by family background it is possible to test whether college proximity is a legitimately exogenous determinant of schooling. The results affirm that marginal returns to education among children of less-educated parents are as high and perhaps much higher than the rates of return estimated by conventional methods.

Model:

卡德( Card, 1993) 使用被访者的家到最近的大学的距离作为教育的工具变量，以此来分析教育是否能增加个人的收入和地位。我们知道，教育作为解释变量之所以是内生的，是因为人们会选择上或不上大学。而从家到大学的距离，会影响到是否上大学这个理性选择。

13. Joshua D. Angrist, Victor Lavy, Using Maimonides' Rule to Estimate the Effect of Class Size on Scholastic Achievement, The Quarterly Journal of Economics, Volume 114, Issue 2, May 1999, Pages 533–575, https://doi.org/10.1162/003355399556061

Abstract: The twelfth century rabbinic scholar Maimonides proposed a maximum class size of 40. This same maximum induces a nonlinear and nonmonotonic relationship between grade enrollment and class size in Israeli public schools today. Maimonides’ rule of 40 is used here to construct instrumental variables estimates of effects of class size on test scores. The resulting identification strategy can be viewed as an application of Donald Campbell’s regression-discontinuity design to the class-size question. The estimates show that reducing class size induces a significant and substantial increase in test scores for fourth and fifth graders, although not for third graders.

Model:

a model for individual pupils’ test scores is used to describe the causal relationships to be estimated. For the i^th student in class c and school s, we can write

安古瑞斯特和拉维关于以色列班级规模效应的经典研究( Angrist & Lavy, 1999): 小班是否有利于学生提高成绩。他们采用的工具变量是所谓迈蒙尼德( Maimonides) 规则下的本地招生人数。根据迈蒙尼德规则，凡是班级人数超过40 的就会被分裂为两个小班。因此，招生人数和班级规模之间就存在一个紧密的非线性关系。

14.方颖, 赵扬. 寻找制度的工具变量:估计产权保护对中国经济增长的贡献[J].经济研究,2011,46(05):138-148.

摘要：本文旨在建立制度的合适工具变量,从而能够正确估计制度对中国经济增长的贡献。文章提出以中国1919年基督教教会初级小学的注册学生人数作为制度的工具变量,并详细论述工具变量的相关性和外生性诸条件。通过两阶段最小二乘法,本文发现制度对中国经济的贡献显著为正。在控制了地理因素和政府政策效应等变量以后,制度对经济增长的效应仍然最为显著。 

方颖、赵扬( 2011) 为估算各地区产权保护制度对经济增长的贡献，采用的工具变量是1919 年中国不同城市基督教初级教会小学注册人数在当地人口中的比例。因为这一比例代表了该地区历史上受西方影响的程度: 入读教会小学的人口比例越大，该地区受西方影响越大，也就越有可能在今天建立起较好的产权保护制度。而教会小学建立的初衷在于布道，这一变量和当年以及现在各地区的经济水平并不直接相关。

15. 章元, 陆铭. 社会网络是否有助于提高农民工的工资水平?[J].管理世界,2009(03):45-54.

摘要：本文基于中国22个省份的农户调查数据研究了社会网络在城市劳动力市场上对于农民工工资水平的影响,并利用农民工祖辈的社会背景及是否来自革命老区作为工具变量考察了社会网络可能具有的内生性是否会导致估计偏误。本文发现:只有非常微弱的证据表明拥有更多的社会网络能够直接提高农民工在城市劳动力市场上的工资水平,它在具有较高竞争性的城市劳动力市场上的主要作用是配给工作,但并不能直接改变劳动力市场的均衡价格水平,它只能通过影响农民工的工作类型而间接地影响他们的工资水平。

章元、陆铭( 2009) 在分析农民工的家庭网、亲友网和收入之间的关系时，用农民工祖辈的社会背景及是否来自革命老区作为工具变量。其理由是，祖辈社会背景和是否来自老区会影响到农民工的社会网络规模，但这些历史因素和今天农民工在异地的收入没有其他直接联系。

还有一些其他文献，下面是对Explained variable, explanatory variable, IV和Literature进行整理得到的表格。

下面这些短链接文章属于合集，可以收藏起来阅读，不然以后都找不到了。