A novel dataset on legal traditions, their determinants, and their economic role in 155 transplants

The law and the economy are deeply influenced by the legal tradition or origin, which is the bundle of institutions shaping lawmaking and dispute adjudication. The two principal legal traditions, common law and civil law, have been transplanted through colonization and occupation to the vast majority of the jurisdictions in the world by a group of European countries. Here, I illustrate a novel dataset recording the lawmaking institution employed by 155 of these jurisdictions at independence and in 2000 and four discretion-curbing adjudication institutions adopted by 99 of these “transplants” at the same two points in time. Contrary to the “legal origins” scholars׳ assumption, 25 transplants changed the transplanted lawmaking institution and 95 modified at least one of the transplanted lawmaking and adjudication rules. In “Endogenous Legal Traditions” (Guerriero, 2016a) [12], I document that these reforms are consistent with a model of the design of legal institutions by societies heterogeneous in their endowment of both the extent of cultural heterogeneity and the quality of the political process. In “Endogenous Legal Traditions and Economic Outcomes” (Guerriero, 2016b) [13] moreover, I show the relevance of considering legal evolution and the endogeneity between legal traditions and economics outcomes. The data illustrated here also include the proxies for the determinants of legal evolution I use in “Endogenous Legal Traditions” (Guerriero, 2016a) [12] and the novel measure of economic outcomes I employ in “Endogenous Legal Traditions and Economic Outcomes” (Guerriero, 2016b) [13].


a b s t r a c t
The law and the economy are deeply influenced by the legal tradition or origin, which is the bundle of institutions shaping lawmaking and dispute adjudication. The two principal legal traditions, common law and civil law, have been transplanted through colonization and occupation to the vast majority of the jurisdictions in the world by a group of European countries. Here, I illustrate a novel dataset recording the lawmaking institution employed by 155 of these jurisdictions at independence and in 2000 and four discretion-curbing adjudication institutions adopted by 99 of these "transplants" at the same two points in time. Contrary to the "legal origins" scholars' assumption, 25 transplants changed the transplanted lawmaking institution and 95 modified at least one of the transplanted lawmaking and adjudication rules. In "Endogenous Legal Traditions" (Guerriero, 2016a) [12], I document that these reforms are consistent with a model of the design of legal institutions by societies heterogeneous in their endowment of both the extent of cultural heterogeneity and the quality of the political process. In "Endogenous Legal Traditions and Economic Outcomes" (Guerriero, 2016b) [13] moreover, I show the relevance of considering legal evolution and the endogeneity between legal traditions and economics outcomes. The data illustrated here also include the proxies for the determinants of legal evolution I use in "Endogenous Legal Traditions" (Guerriero, Contents lists available at ScienceDirect journal homepage: www.elsevier.com/locate/dib 2016a) [12] and the novel measure of economic outcomes I employ in "Endogenous Legal Traditions and Economic Outcomes" (Guerriero, 2016b) [13]. &

Data accessibility
Data are with this article.

Value of the data
The data reveal that legal traditions are not fixed in the original transplanted form as assumed by the legal origins scholars [14], but they evolve. Hence, more theoretical and empirical research on the determinants of this evolution, as in [12], is necessary.
The data allow to study the impact of endogenous legal traditions on the economy as in [13]. The data are key to draw policy implications relevant for the current process of international legal harmonization.

Data
The dataset consists of cross-sectional observations on 155 countries that received their initial legal tradition exogenously mainly via colonization or occupation, i.e., transplants. For this sample, I report in the excel file "OIL_W" the lawmaking and adjudication institutions at independence and in 2000, the proxies for the determinants of their evolution I discuss in [12], and both the continuous measure of legal traditions and a measure of social welfare I employ in [13]. While the lawmaking institution determines the identity of the lawmakeri.e., the government, the legislature, or the president under statute law and appellate judges under case law, adjudication institutions modulate the discretion allowed by the legal system to lower adjudicating courts [5,10,20,22]. The drivers of the evolution of legal traditions are the extent of cultural diversity and the quality of the political process.

Legal traditions
The sources I use to code lawmaking institutions and identify the transplanted adjudication institutions are the first volume of the International Encyclopedia of Comparative Law [8] and the appendix accompanying [9], i.e., [1]. They illustrate in sufficient details the history of the legal order of the 155 transplants for which I also observe cultural diversity and/or the quality of political institutions. I exclude from this group Andorra, Guinea-Bissau, Mongolia, Mozambique, Netherlands Antilles, San Marino, and Suriname lacking sufficient information and Afghanistan, Austria, Bhutan, Denmark, France, Germany, Iraq, Maldives, Oman, Saudi Arabia, Soviet Union, Sweden, Switzerland, and the United Kingdom because origins, i.e., they developed their legal tradition internally. [3] treat also Finland, Norway, and the USA as origins but acknowledge that they could also be considered transplants  [5,10,20,22]. All in all, 10 transplants switched from statute law to case law, 15 abandoned case law to embrace statute law, 69 kept statute law, and the remainder maintained case law.
For 99 of the aforementioned 155 transplants, [9] analyze the procedural rules governing the adjudication of two ubiquitous legal disputes: the collection of a bounced check and the eviction of a non-paying tenant. From this information, I obtain four variables measured in the year 2000: 1.
Comprehensive-Appeal, which is one if issues of both law and fact can be reviewed in appeal and zero if only new evidence or issues of law can be reviewed, or if there is no appeal; 2. Judgment-Law, which is one if judgments must be on law only, and zero when they may be based on equity grounds; 3. Inquisitorial, which is 1 if judges can request or take evidence that has not been introduced by the parties and can refuse to collect or admit requested evidence, 0.5 if they have only one of the two prerogatives, and 0 otherwise; 4. Written-Evidence, which is one if the evidence is mostly submitted to the court in written form and zero otherwise. An important body of comparative law claims that all four variables equal one in a pure civil law jurisdiction and zero in a pure common law legal system [15,  To measure the evolution of legal traditions, I operate as follows. First, for each adjudication institution I stack one over the other the values for the collection of a bounced check and the eviction of a non-paying tenant. Second, I build on [5,7,10,20], and [22], and I obtain for the 99 transplants the four adjudication institutions at independence, i.e., Comprehensive-Appeal-I, Judgment-Law-I, Inquisitorial-I, and Written-Evidence-I. The United Kingdom and France transplanted respectively a pure common law and a pure civil law tradition, whereas Austria and the Soviet Union (Denmark, Germany, Sweden, and Switzerland) exported statute (case) law together with an intermediate level of discretion in adjudication. Finally, I construct: 1. Common, which is the sum of Comprehensive-Appeal-I, Judgment-Law-I, Inquisitorial-I, Written-Evidence-I, and Statute-I minus the sum of Comprehensive-Appeal, Judgment-Law, Inquisitorial, Written-Evidence, and Statute in the sample of transplants that received statute law; 2. Civil, which is the sum of Comprehensive-Appeal, Judgment-Law, Inquisitorial, Written-Evidence, and Statute minus the sum of Comprehensive-Appeal-I, Judgment-Law-I, Inquisitorial-I, Written-Evidence-I, and Statute-I in the sample of transplants that received case law. Thus, positive values of Common (Civil) correspond to a move toward a pure common (civil) law tradition with higher values implying deeper reforms. Estimating through a two-parameter logistic item response model the factor loadings of the five institutions characterizing a legal tradition reveals that they do not differ significantly from one whether measured at independence or in 2000, and thus they are equally important in capturing the latent "legal tradition" construct [18]. In [12], I also propose two alternative measures of the evolution of legal traditions: the dummy Common-D (Civil-D), which equals one when Common (Civil) is strictly greater than one, i.e., when relative to the bundle prevailing at independence at least one more institution typical of the pure common (civil) law tradition has been embraced. The pdf file "Measuring_LT" reports all the details of the codification exercise just explained.
In [13], I proxy the extent to which a transplant was nearer to a pure common law tradition in 2000 through the variable Common-Law, which is the first principal component normalized in order to range between 0 and 1 and extracted from Case-Law and 1 minus each of the following four aforementioned indicators for the eviction of a non-paying tenant, i.e., Comprehensive-Appeal, Judgment-Law, Inquisitorial, and Written. This choice is justified by the aforementioned evidence from two-parameter logistic item response model estimates [18].

Drivers of the evolution of legal traditions
In [12], I use two proxies for cultural heterogeneity. While one measures the cultural diversity in the transplant, the other gauges the cultural distance between the plurality ethnic group in the country that chose the legal tradition and the plurality ethnic group in the transplant. Cavalli-Sforza (1994) [6] suggest that the coancestry coefficient is an appropriate measure of the cultural distance between populations since it summarizes "the degree of genealogical relatedness of different populations over time. Thus, it can be interpreted as a general metrics for average differences in characteristics transmitted across generations" [19, p. 473] like preferences for punishment. When two populations split apart, their genes start to change due to random genetic drift, natural selection, and migration. The coancestry coefficient is the sum of the random drift-driven differences in the frequencies of DNA polymorphismsi.e., situations in which a DNA sequence exists in at least two different allelesbetween indigenous populations in place before 1500. Since [6] calculate the coancestry coefficient for macro-populations, I embrace the strategy proposed by Spolaore (2009) [19]. First, I identify the plurality ethnic group in each country by using ethnic composition data from [2] and information from [4]. Next, I match the plurality ethnic groups to the macro-populations reported in [6]. Finally, I normalize the resulting coancestry coefficients to make them range between 0 and 1, i.e., Genetic-Distance. Larger values of Genetic-Distance reveal a longer separation and, in turn, a larger cultural distance between populations. Cavalli-Sforza (1994) [6] also highlight the strong link between genetic pools and ethnolinguistic roots. This relationship, which is driven by the diverse cultural transmission instruments of populations that are increasingly distant from the common ancestors, makes the ethnolinguistic fractionalization in a country a natural proxy for the "within" cultural diversity. The measure of ethnolinguistic fractionalization ELF can be interpreted as the probability that two individuals randomly drawn from the population belong to different ethnolinguistic groups. ELF is based on data from the Narodov Mira atlas and [11]. Lower ethnolinguistic cohesion corresponds to higher values of ELF, and therefore a value of 0 (1) identifies a completely homogeneous (heterogeneous) country. In [12], I also experiment with the natural logarithm of the number of languages [16] and the ethnic fractionalization [2].
Following [17], in [12] I proxy the quality of the political process with the normalizedto range between 0 and 1 -first principal component extracted from the constraints on the executive authority score and Polity score collected from the POLITY IV dataset at http://www.systemicpeace. org/polity/ and averaged between independence and 2000, Pc-Institutions.

A measure of social welfare
In [13], I construct a proxy capturing simultaneously the long-run technological efficiency of the law and society's cultural satisfaction with legal rules building on data from the World Business Environment Survey [21]. This firm-level survey asked in 2000 managers to rate on a 6-point scale how much they believe the legal system will uphold contracts and property rights in a business dispute and how much they believe it is "fair and impartial." While the former aspect relates to the ability of the legal system to hit its efficiency targetsi.e., expanding trade and sheltering private investment from expropriation, the latter substantiates society's cultural satisfaction with the law. To capture both aspects, I consider the normalized first principal component extracted from the country average of each question, i.e., Social-Welfare.