I. INTRODUCTION

Climate change has severe impacts on developing countries, affecting physical systems (rivers, lakes, floods, droughts, coastal erosion, and sea-level effects), biological systems (terrestrial and marine ecosystems), and human systems (health and livelihoods, and food production) (IPCC 2014). The literature explaining the pattern of policies addressing climate change and other environmental problems tends to emphasize relatively modern determinants, including the current levels of corruption, democracy, recent political instability, various political institutions, and trade policy. (1) However, history should not be ignored when we attempt to understand present-day policymaking. Is it possible that history going as far back as 1 AD still matters significantly for today's environmental policy outcomes? In this article, we study the joint effect of long-term statehood experience and legal heritage on climate change policy outcomes in former colonies. To the best of our knowledge, this is the first study to merge these two strands of the literature. Moreover, it is the first to investigate the role of statehood history or state capacity for any type of environmental policymaking.

Long-term historical processes, including the development of agriculture, urbanization, the use of money as a medium of exchange, taxation, and experience with government administration all build stocks of human capital and experience through learning-by-doing (Acemoglu. GarciaJimeno, and Robinson 2015; Burkett. Humblet, and Putterman 1999; Putterman 2000). (2) The accumulation of statehood experience influences a country's ability to consolidate and centralize power, and creates a capable bureaucracy which provides overall "state capacity" (Besley, Ilzetzki, and Persson 2013; Besley and Persson 2009, 2011; Gennaioli and Rainer 2007; Michalopoulos and Papaioannou 2014). Besley and Persson (2011) present a theory predicting that state (fiscal) capacity and good institutions (legal capacity) are complements, which is supported by empirical evidence provided by Michalopoulos and Papaioannou (2014).

The form of legal system utilized has important implications for regulatory, judicial, financial, and economic outcomes; the literature has primarily focused on the implications of French civil law versus British common law for financial development and regulation (see. e.g.. Botero et al. 2004; La Porta et al. 1997, 1998; La Porta, Lopez-de-Silanes, and Shleifer 2008). The resulting Legal Origins Theory (LOT) developed by La Porta, Lopez-de-Silanes, and Shleifer (2008) argues that British common law allocates more weight to private markets, the courts, and decentralized policymaking, while French civil law gives a strong role to the centralized state. Recent empirical work based on LOT finds that common law countries set weaker climate change policies than civil law countries (Fredriksson and Wollscheid 2015).

So far, the literature on statehood experience has developed parallel to the literature on legal origins. We remedy this gap in the literature. We argue that the implications of receiving common law versus civil law should be particularly important in countries with a greater accumulated history of statehood at the time of colonization. In those countries, state (administrative) capacity was further developed, political power tended to be more centralized, the preexisting accountability of local leaders was greater, and the ability to receive and implement the transplanted legal framework from the colonizer was therefore enhanced. In contrast, areas without much statehood experience did not have a strong bureaucracy or state capacity, and the transfer of the colonizer's legal approach was less complete. Our hypothesis is consequently that a longer history of statehood at the time of colonization serves to increase the differential impact of common and civil law legal origins on the stringency of modern climate change policies, as well as on other environmental policies.

Using a cross-section of up to 78 former colonies, our results lend support to this hypothesis. Our estimates suggest that the difference between common law and civil law countries in modern climate change policies is conditional on statehood experience, measured by the length of state history from 1 to 1800 AD. In particular, the divergence in modern climate change policy stringency due to legal origin (common law vs. civil law) is inflated by the length of statehood experience. Moreover, legal origin has no effect in countries which, at or around the time of colonization, had no statehood experience. We believe these are novel findings in the literature. Moreover, our findings apply more generally to several other measures of environmental policy outcomes.

These findings give a more fine-tuned picture of the role of legal origin compared to the previous literature on regulation in previous colonies. Our analysis helps improve our understanding of the long-term determinants of climate change policies and other environmental policies, as well as regulatory outcomes in general. Our result suggests that it is important to take the accumulated statehood experience into account when discussing the role of legal origin. The analysis may facilitate the prediction of which countries will tend to become leaders and laggards in the climate change policy formation process. An improved understanding of the institutional barriers created by legal and state history may also improve capacity building efforts, as well as the formulation of mechanisms and designs within international climate change treaties. Resource transfers may more easily be targeted toward their most productive uses.

The article proceeds as follows. The next section reviews some of the related literature. Section III provides a discussion on the empirical specification, estimation issues, and data. The empirical estimates are presented and analyzed in Section IV. Several robustness checks are also performed, including the use of an alternative dependent variable. The last section summarizes and concludes.

II. RELATED LITERATURE AND HYPOTHESIS SPECIFICATION

The state is one of the most important forms of institutional development. The creation of nation states has led to a number of fundamental and far-reaching changes in human history. In recent years, statehood experience (sometimes denoted "state antiquity" in the literature) has gained considerable attention from the literature on long-run comparative economic development, helping to uncover the reasons for low income levels (Putterman 2008; Putterman and Weil 2010), bad institutions (Ang 2013b), unequal distribution of income (Putterman and Weil 2010), and financial underdevelopment (Ang 2013a). The results of these studies generally suggest that a longer state history is associated with more favorable economic outcomes. A longer history of statehood often implies a stronger state capacity in the form of legal and fiscal capabilities (Becerra, Cavallo, and Scartascini 2012; Besley and Persson 2009, 2011). In particular, Michalopoulos and Papaioannou (2013) show that precolonial ethnic political centralization is strongly related to contemporary regional development in Africa. However, despite the above contributions, how statehood experience is related to environmental regulations has so far not been considered in the literature.

A large literature has investigated the implications of common law and civil law for modern economic and legal institutions, for example, for the regulation of firm entry and labor; degree of judicial independence; formalism of judicial procedures; securities, company, and bankruptcy laws; and government ownership of banks (Botero et al. 2004; Djankov et al. 2002, 2003; La Porta et al. 1997, 1998; La Porta, Lopez-de-Silanes, and Shleifer 2008). These institutions in turn have important implications for economic outcomes, for example, unemployment and labor participation rates, corruption and the share of the unofficial economy, property rights, stock market development, and private credit. While legal systems were transplanted in a nonvoluntary fashion to the colonies (David 1985; McNeill and McNeill 2003), legal philosophies are highly persistent even if domestic laws subsequently evolved in the recipient countries.

La Porta, Lopez-de-Silanes, and Shleifer (2008) present LOT which argues that British common law gives more weight to private markets and the courts, while civil law assigns an important role to the state. (3) Botero et al. (2004) suggest that countries with different legal origins essentially employ alternative institutional technologies for the social control of business activity (see also Glaeser and Shleifer 2002). The legacies of legal origins exhibit a high degree of institutional persistence and path dependency where historical, economic, political, social, and cultural features have a lasting impact (Acemoglu, Johnson, and Robinson 2001; Greif 1998; Hodgson 1998; Marchand 2016; North 2006).

Four implications of LOT are relevant for climate change policies. First, common law favors markets and judicial resolutions, not government regulation as emphasized by civil law. (4) Second, decentralized systems used in common countries are less prone to take transboundary pollution externalities into account compared to centralized systems emerging in civil law countries (see Sigman 2014). Third, as common law respects private property relatively more, climate policies affecting capital owners are likely to be weaker in those countries. Fourth, while common law may be more adaptable to new situations as jurisprudence is more important as a source of law (Levine 2005; Ponzetto and Fernandez 2008), civil law country governments can more easily address emerging social problems due to a larger local bureaucracy (Mulligan and Shleifer 2005). Overall, most of the above arguments support the hypothesis that common law should yield weaker climate change policies.

Fredriksson and Wollscheid (2015) report that former colonies with a common law legal heritage indeed set weaker climate change policies than civil law countries. Oto-Peralias and Romero-Avila (2014) show that the effect of legal origins on several financial outcomes and business regulations is conditional on initial conditions (population density in year 1500 AD). In particular, the effect of common law is negatively related to precolonial population density, that is, the outcome is closer to civil law countries when density is high. Berkowitz. Pistor, and Richard (2003) argue that it does not matter significantly which form of law a country uses for it to be effective. According to them, what matters is whether the transplanted law was developed domestically, if it was received through colonization but adapted to local conditions, or whether the local population already had some familiarity with its legal principles. Under these circumstances, the demand for the law would be high and it would be used effectively to enhance economic development.

Gennaioli and Rainer (2007) discuss the importance of precolonial institutions in Africa, especially political centralization, for modernization programs during the colonial and the postcolonial periods. Policy coordination and implementation were facilitated by the existence of a precolonial hierarchy of chiefs in areas with centralized ethnic groups (see Schapera 1970). Bargaining with senior traditional chiefs in these areas resulted in faster adoption of European policies than in areas with high political fragmentation. Preexisting accountability of local leaders in traditionally centralized systems was the mechanism which facilitated reforms. In fragmented areas, local chiefs pursued their own distorted policies, leading to lower rates of modernization and even tyranny (Tosh 1978). Gennaioli and Rainer (2007) argue that institutions emerging in the African colonies were heavily influenced by the existing institutions found by the colonizers (see also Roberts and Mann 1991).

Michalopoulos and Papaioannou (2013) document a strong association between African precolonial ethnic political centralization and modern regional development. Michalopoulos and Papaioannou (2013) argue that state capacity and legal capacity are complements in Africa, supporting the prediction of Besley and Persson (2011). In particular, they find that differences in national institutions (rule of law and corruption) yield differences in regional development among (Southern African) countries with high levels of state capacity. However, in (Western African) countries with low state capacity, no correlation is detected between national institutions and regional development. (5) These findings indicate that differences in the adoption of a colonizer's legal system across centralized and fragmented areas should similarly be determined by the level of statehood at the time of colonization. Path dependency and institutional persistence (see, e.g., Greif 1998; Marchand 2016; North 2006) have preserved or inflated the resulting differences across legal origins over time.

Our hypothesis is thus that the transfer of legal origin should be more thorough and deep-rooted, with a greater impact today, in countries which at the time of colonization had a longer history of statehood experience. These colonies had a more developed legal and administrative bureaucracy, and hence were in a better position to facilitate the absorption of the legal philosophies. Therefore, the difference in the stringency of climate change policies between common and civil law countries should be greater in former colonies with a more extensive statehood experience. The divergence will increase due to a longer statehood history both because the antiregulation stance of common law will be more deeply embedded, and because the pro-regulation perspective of civil law becomes more deeply rooted in the society. Our hypothesis appears novel, as the previous literature has not investigated the regulatory effects of the absorption of legal origins conditional on state history, political centralization, or/and other forms of state capacity.

III. EMPIRICAL APPROACH

A. Regression Model

The following model is regressed to investigate how climate change policy is related to statehood history and the legal tradition of a country:

(1) C[C.sup.policy.sub.i] = [alpha] + [beta][State.sub.i] + [[gamma].sub.1] Common Law L[O.sub.i] + [[gamma].sub.2]Mixed Law L[O.sub.i] + [[delta].sub.1] State i x Common Law L[O.sub.i] + [[delta].sub.2][State.sub.i] x Mixed Law L[O.sub.i] + [cv'.sub.i][theta] + [[epsilon].sub.i]

Where C[C.sup.policy] is an index of climate change policy, State is a measure of statehood experience covering the period 1-1800 AD. Common Law LO is a dummy variable for countries classified as having a British common law tradition. Mixed Law LO is a dummy variable for countries classified as having a mixed (i.e.. a combination of common and civil) law traditions, cv' is a set of control variables included in regressions to allow for the influence of some geographic and other effects, and e is an unobserved error term. "Other legal origins" is the excluded group, which includes 39 French and 3 German civil law countries.

Mixed legal tradition refers to those jurisdictions that have inherited elements of both common law and civil law. For instance. South Africa and Sri Lanka are coded as jurisdictions having a mixed legal tradition as they were initially colonized by the Netherlands and hence had inherited some form of civil law tradition. However, this legal system was partially replaced with common law when they were subsequently conquered by England (Klerman et al. 2011). Importantly, the mixed legal heritage arose due to exogenously determined events from the colonies' perspectives.

As discussed in the previous section, our main variable of interest is the interaction term State X Common Law LO. The interaction term is expected to carry a negative sign due to the proposition that, relative to civil law countries, common law countries equipped with stronger state capacity impose less stringent climate change rules and regulations.

It is important to highlight that our estimations include only former colonies, which are not Organization for Economic Cooperation and Development members. This consideration is relevant because colonizers typically imposed a system of law upon their colonies. The legal structures imposed on former colonies under colonialism are therefore exogenously determined. By contrast, development of the legal systems in the origin countries, such as England, France, and Germany, was influenced by their own economic, cultural, and political factors, and hence is endogenously determined.

Countries which adopted foreign legal systems voluntarily, such as Ethiopia, Japan, Thailand, and Turkey, present similar endogeneity issues as they adopted a legal system, primarily French and German, which was deemed favorable for improving their economic conditions. Hence, a major advantage associated with the use of only former colonies in our analysis is that the exogenous nature of legal systems transplanted by colonialism enables us to interpret our estimates as reflecting the causal impact of legal tradition on climate policies (Klerman et al. 2011; La Porta, Lopez-de-Silanes, and Shleifer 2008).

B. Data

Climate Change Policy Index (C[C.sup.policy]). We use the Climate Change Cooperation Index (C3-I) of Bernauer and Bohmelt (2013) as our measure of climate change policy strictness. The C3-I provides times-series cross-section data for up to 172 countries over the period 1996-2008. The index consists of a policy and an emissions component. The policy component, which is used as our main measure of climate change policy, is the aggregate of the following subcomponents: (1) whether a country signed the United Nations Framework Convention on Climate Change (UNFCCC) and how soon this was done; (2) whether a country signed the Kyoto Protocol and how soon this was done; (3) whether a country submitted the latest national climate report and if this was done on time; and (4) how often a country made financial contributions to the UNFCCC on time over the period 1996-2008. These four subcomponents are summed and an average value of the climate policy index (C[C.sup.policy]) is obtained for each country.

State History (State). We use the latest version (version 3.1) of the state history data assembled by Putterman (2004), who provides state antiquity data covering 39 half centuries from 1 AD to 1950 AD for 151 countries. This index of state history gives a score from 0 to 50, reflecting: (1) the presence of a government above the tribal level (1 point if yes, 0 points if no); (2) whether this government is foreign or locally based (1 point if locally based, 0.5 points if foreign [i.e., the country is a colony], 0.75 if in between [a local government with substantial foreign oversight]); and (3) the proportion of the current territory covered by this government (1 point if over 50%, 0.75 points if between 25% and 50%, 0.5 points if between 10% and 25%, and 0.3 points if less than 10%).

To illustrate, state history (State) for the 18 centuries to 1800 AD is calculated as follows:

(2) [State.sub.i] = ][[SIGMA].sup.36.sub.t=1] [(1.05).sup.1-t] x [S.sub.i,t]]/[[[SIGMA].sup.36.sub.t=1] [(1.05).sup.1- t] x 50]

where [S.sub.i,t] is the state presence for country i for the 50-year period t (see Putterman and Weil 2010). The scores on the three questions were multiplied by one another and by 50, so that for a given 50-year period, a country has a score of 50 if it was an autonomous nation, 0 if it had no government above the tribal level, 25 if the entire territory was ruled by a foreign country, and so on. Appendix B provides more details on the construction of State and several examples. A 5% discount rate is applied to each of the half centuries so that less importance is attached to states formed in the more distant past. Our estimates are not sensitive to the use of alternative depreciation rates ranging from 0% to 20%. This approach of measuring state antiquity is broadly consistent with Putterman (2008), Putterman and Weil (2010), and more recently, Ang (2013a, 2013b). The index is converted to a scale from 0 to 1 where higher values reflect the presence of a longer state history. Figure 1 shows the dispersion of State across the world for all available countries in the Putterman dataset.

Although data on state history are available up to 1950 AD, we measure it only up to 1800 AD--the period in which colonialism reached its peak. Doing so not only preempts the issue of reverse causality from climate policy to state capacity, but the period considered is also more in tandem with the timing when legal systems were imposed on colonies. In the robustness checks, state experiences accumulated up to 1500 and 1650 AD are also considered in order to check if the results are driven by the period chosen.

Legal Origins (LO). The legal tradition of company law or the commercial code for each country is classified into British common (Common Law LO) or Mixed (Mixed Law LO), with civil law as the excluded category, using binary variables. To do this, we follow the legal tradition classification of Klerman et al. (2011). The advantage of using this classification, as opposed to the more traditional classification approach of La Porta et al. (1998), is that it enables us to identify colonies which were influenced by both common law and civil law legal structures (Mixed), which may have some bearing on the results. Although Klerman et al. (2011) are legal scholars, who are expected to have detailed insights into countries' legal histories, we also consider the more widely adopted legal system classification of La Porta, Lopez-de-Silanes, and Shleifer (2008) in the robustness analysis.

After assembling the various sources of data, 78 observations remain. The sample size is mainly constrained by the inclusion only of former colonies in the analysis. Table A2 provides a list of these countries, including legal origin and values of the State measure. Table 1 provides the summary statistics for the key variables used in the estimations.

A simple comparison between common law and civil law countries with similar levels of State, such as Malaysia and Nigeria (common law; State = 0.585 and 0.544; C[C.sup.policy] = 36.923 and 30.082, respectively) versus Armenia and Tajikistan (civil law; State = 0.561 and 0.524; C[C.sup.policy] = 49.279 and 53.267, respectively) indicates that common law countries tend to have weaker climate change policies. However, countries with no statehood experience in 1800 AD (State = 0.00) such as Zambia (common law; C[C.sup.policy] = 25.521) and Gabon (civil law; C[C.sup.policy] = 28.869) exhibit only a small difference in C[C.sup.policy].

Figure 2 shows the overall association between climate change policy index and statehood experience. Figure 3 exhibits the various relationships between our climate change policy index and statehood experience across different types of legal tradition. Consistent with our prediction, these diagrams provide preliminary evidence that State and C[C.sup.policy] are negatively correlated in countries that have the common law legal system. There is, however, no clear relationship between these variables for countries with "mixed" legal traditions.

However, it is necessary to check whether there is any difference in outcomes in colonies that did not have any statehood experience in 1800 AD. In this connection, it is worth highlighting that the intercept values of the regression models are quite similar in all three subsamples. This implies that countries which lacked any statehood experience in 1800 AD are likely to end up adopting more or less the same level of stringency in climate change policies. This empirical observation reinforces our proposition that how each type of legal tradition influences climate change policies depends crucially on statehood experience. To confirm this visual inspection, we perform an F test on the equality of the intercept coefficients for the models that include common law countries only and civil law countries only (their respective intercept coefficients are 36.288 and 30.102). The test-statistic obtained from the nested regression model is 2.590 (p = 0.113), suggesting that the intercept values are not statistically different.

IV. RESULTS

A. Main Findings

The estimation results of Equation (1) are presented in Table 2. We consider several alternative specifications in which the main covariates are entered with different combinations in the regressions to ensure that the results are not driven by any particular model specification. In particular, column 1 considers only the effect of statehood whereas column 2 includes only the legal origin dummies. We do not find any evidence in support of the notion that the variation in climate change policy across countries can be solely accounted for by differences in statehood or legal traditions.

Finally, the last column in Table 2 provides the full specification by adding common and mixed legal tradition and their interactions with state history. This complete specification will be used as our baseline model for robustness checks of the results. The results indicate that the interaction between statehood and the common law dummy is significantly correlated with climate change policy with the expected negative sign. This correlation is found to be statistically significant at the 1% level. Thus, the results suggest that a longer history of statehood experience at the time of colonization inflates the difference between civil law and common law countries' stringency of climate change policy. The effects are economically meaningful.

The results in column 3 suggest that at the mean value of State, the presence of common law legal origin is associated with a change in C[C.sup.policy] by (6.186-0.32x28.069=) -2.796, equal to a decline by 0.26 standard deviation (SO). At 1 SD above the mean of State (0.58), the marginal change in C[C.sup.policy] equals -10.09, equivalent to 0.95 SD. All other things being equal, this is equivalent to switching policy from the one in Gambia (C[C.sup.policy] = 33.53) to the one in Bangladesh (C[C.sup.policy] = 23.44). This may help explain why common law country Sudan, which had a relatively extensive state experience in 1800 AD (State = 0.703; 2.20 S D above the mean), has C[C.sup.policy] = 21.43 (1.31 SD below the mean).

Statehood experience is found to be significantly correlated with climate change policy, with a positive sign consistent with our prediction. We also find that while the mixed legal system dummy variable is statistically insignificant, its interaction with statehood is weakly significant in differentiating the variation in climate change policy observed across countries. On the whole, the results presented in Table 2 lend some initial (strong) support to our hypothesis that the stringency of climate change policy can be predicted by the strength of the interaction between statehood and common law legal heritage.

Figure 4 shows the partial regression line for the correlation between climate change policy and the interaction between state experience and common law system, while controlling for the influence of other variations stated in Equation (1). Evidently, the partial regression line shows a strong negative correlation between the variables, thus reinforcing the findings in Table 2.

B. Controlling for Other Effects

To ensure that the above results are not confounded by the influence of some other effects and that the early advantage conferred by statehood is not proxying some forms of geographic characteristics, we include several control variables in the regressions and report the results in Table 3. The nongeographic variables include manufacturing output ratio, per capita real income, democratic capital, two indices of institutional quality, and an index of ethnolinguistic fractionalization. Controlling for these effects is necessary because they may be correlated with statehood, legal origins, and the outcome variable. (6) Geographic measures that have been argued to be potentially influential for economic outcomes including latitude, landlockedness, mean elevation, distance to the nearest coast or river, and precipitation. These are also controlled for in the regressions. Definitions and sources of these variables are given in Appendix A.

Columns 1 and 2 add the contemporary measures individually. Column 3 includes democratic capital stock. Column 4 controls for the effects of institutions (control of corruption and political stability), column 5 controls for ethnolinguistic fractionalization, and column 6 includes some additional effects of geography. The last column includes all control variables simultaneously. It is evident that the coefficients of our core variable of interest, State X Common Law LO, vary only slightly from the baseline estimate reported in the last column of Table 2 when the effects of manufacturing output, democratic capital (discounted history of democracy), ethnolinguistic fragmentation, and geography are allowed for in the regressions (columns 1, 3, 5, and 6). In these models, the effect of State X Common Law LO remains precisely estimated and is at least significant at the 5% significance level.

In absolute terms, the coefficient of State X Common Law LO falls to 20.88 when per capita income is included as a control variable (column 2). This result is not surprising given that prior studies have shown that state history and climate change policy are both significantly correlated with economic development. Similar findings are obtained when two measures of institutional quality are included as additional control variables in column 4. Nevertheless, the significant effect of State X Common Law LO prevails even after we control for the effect of income level or institutions.

In column 7, we add all control variables jointly into the specification. We find that State X Common Law LO continue to exert a strong explanatory power on climate change policy. This effect is found to be significant at the 5% level.

C. Alternative Statehood Periods

Next, we consider several alternative statehood periods to shed further light on the results. The estimations here include all the geographic controls used earlier, and are reported in Table 4. Columns 1 and 3 consider statehood only up to 1500 and 1650 AD, respectively. This is done to ensure that our results are not driven by states formed during the European colonial period since the sixteenth century. Interestingly, the results indicate that the absolute size of the coefficient on State X Common Law LO declines when the alternative dates are used, rather than the period 1-1800 AD as done previously in Tables 2 and 3. In particular, the coefficient sizes decline by 4% when statehood up to 1650 AD is considered, and by 8% when statehood up to 1500 AD is used (the estimates are benchmarked against column 6 of Table 3, which included the geographic controls). These findings imply that a longer statehood confers a stronger state capacity, which enables legal philosophies to develop deeper roots in society and the legal approach used has a greater policy impact. Importantly, the statistical significance of the State X Common Law LO interaction remains largely intact, suggesting that our main findings are unlikely to be driven by the formation of states due to the Western colonization.

To ensure that the results are not driven by the exclusion of the remaining statehood period, we also add statehood 1501-1800 AD to the model in column 1 and statehood 1651-1800 AD to the specification in model 3. As shown in the results reported in columns 2 and 4, respectively, we obtain very similar findings.

In column 5, we measure statehood experience by considering only the presence of states for the period 175 1-1800 AD. Doing so enables us to check whether it is the long-term statehood experience or the state presence at the peak of colonialism that matters. Interestingly, the estimates show that climate change policy strictness is unrelated to the state presence during this period. This finding therefore provides further credence to our proposition that the ability of a country to implement the legal philosophies transplanted by its colonial powers depends on the length of its statehood experience.

The results in column 6, which uses statehood from 1 to 1750 AD, confirm our previous finding that the length of statehood matters. In column 7, we show the results in column 5 are unchanged when state history up to 1750 AD is controlled for.

D. Further Robustness Checks

Some further sensitivity checks are in order, and the results are reported in Table 5. First, we use the legal tradition classification of La Porta, Lopez-de-Silanes, and Shleifer (2008) to check if our results are robust to a coding method that is more widely used in the literature. The key difference between these two coding approaches relates to the classification of some jurisdictions as "mixed" by Klerman et al. (201 1), which is not done by La Porta, Lopez-de-Silanes, and Shleifer (2008). Consequently, Mixed Law LO and State X Mixed Law LO are excluded from the regression for this sensitivity check. In our sample of 78 countries, only nine classifications are different. In the study by Klerman et al. (201 1), eight of these countries (i.e., Botswana, Cyprus, Guyana, Israel, Lesotho, South Africa, Sri Lanka, and Zimbabwe) are classified as having a mixed system and one (i.e., Swaziland) as having adopted the French legal system, whereas La Porta, Lopez-de-Silanes, and Shleifer (2008) treat all of them as having British common law tradition. Hence, in principle, we should not expect much variation in the results as the common law dummy variables based on these two different coding procedures yield a correlation coefficient of 0.78. This conjecture is confirmed in the estimate reported in column 1.

We include continent fixed effects in column 2 in order to ensure that the results are not being spuriously driven by unobserved time-invariant region-specific characteristics. Moreover, countries within the same continents tend to have similar colonization history and state performance, and these arbitrary correlations may bias our results. To address this concern, standard errors are clustered by continent to allow for these patterns within but not across continents. That is, the observations are assumed to be independent across continents but not within continents. The estimates in column 2, however, remain largely insensitive to this consideration.

Next, our results may be influenced by the presence of some influential observations such as Fiji, Korea, and Uzbekistan. As there is no compelling reason to exclude them from the estimations, we perform a robust regression analysis, which is a form of weighted regression technique that takes into account the presence of extreme data points (see, e.g.. Andersen 2008). The results reported in column 3 indicate that our main findings prevail. We have also tried to separately exclude the low statehood experience outlier (i.e., Fiji) and the two high statehood experience outliers (i.e., South Korea and Uzbekistan) from the estimations, and the coefficients of interest remain significant at the 5% level. When all three influential observations are removed, the coefficient of interest is still significant at the 10% level. The additional results are not reported to conserve space (available upon request).

Additionally, our results may potentially be driven by the failure to control for the initial conditions of colonization. In particular, civil law countries were previously colonized mainly by France, Portugal, and Spain whereas common law countries by Great Britain. These countries developed different colonial models, based on differentials in endowment factors in the colonies, which had significant bearing on settlement decisions, institutional building (inclusive or exclusive), and consequently the legacies they left in the postcolonialism era (see also Acemoglu, Johnson, and Robinson 2001; Sokoloff and Engerman 2000).

For instance, the Spanish colonizers mostly settled in the most politically and economically advanced regions (i.e., those with older states), whereas the British colonizers did largely the opposite. Consequently, the civil law variable may be a proxy for endowment factors in precolonial areas such as the statehood experience. In this connection, we do not find this conjecture supported by our data, given that statehood experience and legal origins are only weakly related (the correlation is only 0.15 between State and French civil law legal origin dummy).

Nevertheless, we cannot rule out the possibility that significantly different mean values of State exist between civil law and common law countries, thus reflecting the fact that endowments shape colonial institutional building. That is, if countries with a civil law tradition indeed have longer statehood experience than those with a common law origin due to the pursuit of different colonization strategy, then what we find would not reflect the effect of legal origins conditional upon the level of statehood experience. but rather the fact that statehood experience is longer in civil law countries due to different colonization strategy. We address this concern in the following ways.

First, in order to compare the mean values of State between civil law and common law countries, we perform the two-group mean-comparison test. This is essentially a t test designed to compare means of the same variable between two groups. When the independent variable is chosen to be common law so that the two groups are common law (N = 24) and non-common law (N = 54) group, the t statistic is 1.72 and the corresponding p value is 0.09. When a comparison is made between French civil law countries versus others, the p value is 0.189. Clearly, whether the difference of means in State between civil law and common law countries is statistically different from zero depends on the decision rule we adopt. The evidence is therefore not conclusive.

Second, we directly control for endowment effects using the settler mortality measure of Acemoglu, Johnson, and Robinson (2001). Doing so, however, does not change the results in any significant manner (see column 4). Note that the results are almost unchanged when the improved measure of settler mortality rate provided by Albouy (2012) is used (results unreported to conserve space; available upon request). Third, in order to account for the possibility that legal origins are shaped by colonial strategy, which may confound our results, we include four colonial origin dummies (British, French, Portuguese, and Spanish with other European as the excluded group). Our main results prevail (see column 5).

We also provide some additional estimation by considering the individual components of the aggregate statehood experience index. Does one component of the index drive our results, or do all contribute? The overall index considered so far is based on three subcomponents, which capture the presence of a government above the tribal level ([ST.sup.presence]), whether this government is foreign or locally based ([ST.sup.autonomy]), and the proportion of the current territory covered by this government ([ST.sup.C0VERAGE]). The presence of a state above the tribal level appears highly relevant as a measure of the administrative capacity to adopt the colonizer's legal philosophy, for example. Appendix B provides the details of the individual components.

The results provided in the last three columns show that coefficients of the interaction between State and common law dummy are significant in all cases, suggesting that all dimensions of statehood experience matter. The interaction between [ST.sup.COVERAGE] (which captures the extent to which the territory of the modern country under consideration was ruled by a particular government) and common law dummy is found to have the most significant correlation. (7)

E. Endogeneity Issues

As is common in all empirical studies, it is necessary to correct for the endogeneity of statehood experience due to measurement errors or omitted variable bias. In this case, it is expected that the interaction between statehood experience and the legal origin dummies are also endogenous regressors. In order to isolate the exogenous variation in statehood experience, we use the timing of agricultural transition as the instrument. The data are obtained from the study by Putterman (2006). (8)

The instrumental variable results and the firststage regressions are presented in Table 6. It is evident that, in line with its ordinary least squares counterpart, the interaction term between statehood experience and common law legal origin is found to have a significant effect (columns 4). In the first-stage regressions, the timing of agricultural transition is found to be a significant determinant of statehood experience, consistent with our prediction (column 1). Similarly, the interaction of agricultural transition and legal origins significantly explains the interaction of statehood experience and legal origins (columns 2 and 3).

The first-stage F statistics provide supportive evidence that agricultural transition is a suitable instrument. Given that multiple endogenous regressors are used, we also adopt the procedure of Shea (1997) for testing instrument relevance. This approach tests for the strength of the relationship between every endogenous variable and the excluded instruments, after partialing out the included instruments and other endogenous variables. A small value of the Shea partial R-squared would indicate that the instruments lack sufficient relevance to explain all the endogenous regressors. The Shea's partial R-squared statistics range from 0.114 to 0.455, providing evidence that the instruments are quite strongly correlated with the endogenous variables, and hence are sufficiently relevant to explain the endogenous regressors.

Additionally, we also use the Anderson-Rubin (Anderson and Rubin 1949) method for testing the significance of the endogenous regressors in the structural equation. This approach is robust to weak instruments. The test rejects the null hypothesis that the coefficients of the excluded instruments are jointly equal to zero at the 5% level of significance, thus providing evidence that the endogenous regressors are relevant even in the presence of weak instruments.

To further increase the credibility of the results, we also perform the fractionally resampled Anderson-Rubin (FAR) test, using the procedure developed by Berkowitz, Caner, and Fang (2012), to test the joint significance of the endogenous regressors in our IV estimations. This weak exogeneity test is performed under the assumption that our instrument nearly, but not completely, satisfies the exclusion restriction. A rejection of the null implies that our instrument is endogenous. Clearly, the results indicate that the exclusion restriction assumption is not violated.

F. Alternative Environmental Policy Measures

Our analysis has so far focused only on testing the interaction effect of statehood experience and legal origins on climate change policy. Our results indicate that common law countries with experienced statehood are less inclined to adopt stringent climate change policies. Can the same effect be observed on additional environmental policy outcomes, including those that address more local or regional pollution problems? We adopt the same regression framework and consider four other environmental protection variables.

First, in addition to the C[C.sup.policy] policy index used above, Bernauer and Bohmelt (2013) also provide a more encompassing index of climate change policy outcomes, C3-I, which takes into consideration the level and trend effects of emissions. We use this overall index as the dependent variable in column 1 of Table 7. Note, however, that the C[C.sup.policy] is our preferred measure because emissions are not necessarily policy-determined and are often affected by other hard-to-control-for socioeconomic and political factors, and hence are more likely to subject our results to endogeneity bias. As a second measure of environmental policy stringency, we use the (inverse) maximum lead content per gallon of gasoline in 1996 collected from Lovei (1998) in column 2; Reyes (2015) discusses some health risks associated with lead exposure. Next, in column 3 we use the comprehensive environmental performance index of Hsu et al. (2014), which uses a broad spectrum of indicators to measure countries' protection of human health from environmental harm and the protection of ecosystems in year 2012. Finally, we also use a measure of the stringency of environmental regulations in the agricultural sector for year 1990 from Eliste and Fredriksson (2002) in the last column. This seeks to capture the overall regulatory framework (policies, legislation, and enforcement) addressing environmental problems associated with the agricultural sector.

The results in Table 7 do not alter our main findings markedly given the fact that the coefficient of State X Common Law LO remains significant at conventional levels with a negative sign. These results suggest that our findings may apply also to more local or regional environmental problems.

G. Effects on Regulation of Labor Markets

Finally, we consider whether the interaction between statehood experience and legal origins also affects other outcomes more broadly. La Porta, Lopez-de-Silanes, and Shleifer (2008) note that a number of studies have established that several outcomes such as investor protection, government ownership of banks, entry regulations, and regulation of labor markets vary among legal traditions. It is therefore plausible that how legal origins determine these outcomes also depends on the length of state history.

In the following analysis, we test our hypothesis by focusing on labor regulation as, similar to our environmental outcome measure, this outcome is also designed to address social problems and hence is conceptually closely related to our underlying hypothesis (see Botero et al. 2004). Moreover, understanding the sources of labor regulation is important because, as summarized in the review article of La Porta, Lopez-de-Silanes, and Shleifer (2008), it has implications for resource allocation through affecting productivity growth (Caballero et al. 2013), replacing permanent with temporary employment (Pierre and Scarpetta 2013), labor turnover (Haltiwanger, Scarpetta, and Schweiger 2006), labor misallocation (Lafontaine and Sivadasan 2007), and firm efficiency (Mamatzakiset al. 2015).

In particular, we analyze the following measures of labor market regulation: (1) the social security law index of Botero et al. (2004), (2) the employment protection legislation in terminating regular contracts (also known as EPLEX) of ILO (2015), (3) the ratio of unemployment insurance benefits to a worker's last earnings (Schindler and Aleksynska 2011), and (4) the extent of unemployment benefits coverage (Schindler and Aleksynska 2011). A larger value of these measures generally indicates more rigid or stringent labor market regulations. Appendix A provides the detailed definition of these variables.

The results reported in Table 8 suggest that our results indeed have some broader implications in which experienced common law states generally tend to adopt less stringent policies on labor markets than their civil law counterparts. These findings therefore suggest that our empirical framework can be extended to understand how the regulation of labor markets depends on legal tradition and state history.

V. SUMMARY AND CONCLUSIONS

A country's history of, for example, agricultural development, urbanization, money, taxation, and government administration, all build stocks of human capital and experience. The cumulative experience with statehood influences a country's ability to consolidate power and create a capable bureaucracy, which may be summed up as "state capacity." We argue that the level of long-term statehood experience should influence the propensity of former colonies to absorb and implement the legal frameworks transplanted by the colonial powers.

This study tests the hypothesis that the influence of former colonies' legal heritage on modern climate change policies is conditional on their historical experience with statehood (state capacity). Our empirical work provides support for this hypothesis. The positive difference in the climate change policy stringency in former colonies with common law and civil law systems is enlarged by a longer history of statehood experience. Consistent with this finding, we also find that legal origin has no effect on countries which had no statehood experience around the time of colonization. We obtain similar results also for several alternative environmental policies. In addition, the identified mechanism is found to also hold for several labor market regulations. Our results suggest that long-term historical processes should not be ignored when studying and designing modern policymaking, in particular those addressing climate change.

APPENDIX A: DATA

TABLE A1
Definition of Variables and Data Sources

Variable             Description                      Source

Main variables

C[C.sup.policy]      A composite index capturing      Bernauer and
                     several key components of        Bohmelt (2013)
                     climate change policy,
                     including the ratification
                     of the UNFCCC and the Kyoto
                     Protocol, the submission of
                     national climate report, and
                     financial contributions to
                     the UNFCCC secretariat. The
                     index also considers whether
                     these activities were
                     accomplished in a timely
                     manner

Slate                An index of state history        Putterman
                     covering the period from 1       (2004)
                     to 1800 AD, scaled to take
                     values between 0 and 1. The
                     latest version. v3.1, is
                     used. In the robustness
                     checks, two alternative
                     periods, 1 -1500 and 1 -1650
                     AD, are used

Caminan Law LO       A dummy variable that            Klerman et
Civil Law LO         identifies the legal             al. (2011);
Mixed Law LO         tradition of the company law     La Porta,
                     or commercial code of each       Lopez-de-
                     country as British common        Silanes, and
                     law. French civil law, German    Shleifer (2008)
                     civil law, or Mixed law.
                     An alternative classification
                     that excludes the "Mixed"
                     category, by La Porta.
                     Lopez-de-Silanes, and
                     Shleifer (2008), is used in
                     the robustness checks

Control variables

Manufacturing        Manufacturing value added as     World Bank
(% GDP)              a percentage of GDP (averaged    (2015a)
                     over the period 1996-2008)

Income per           Logged GDP per capita            World Bank
capita               converted to constant 2005       (2015a)
(logged)             international dollars using
                     purchasing power parity rates
                     (averaged over the period
                     1996-2008)

Democratic           Measures of a country's          Fredriksson
capital              discounted stock of democracy    and Neumayer
                     from 1800 to 2010                (2013)

Control of           An index reflecting              World Bank
corruption           perceptions of the extent        (2015b)
                     to which state power is used
                     to obtain private gains
                     (averaged over the
                     period 1996-2008)

Political            An index reflecting              World Bank
stability            perceptions of the likelihood    (2015b)
                     that the government will be
                     destabilized or toppled
                     (averaged over the period
                     1996-2008)

Latitude             Value of the latitude of         CIA(2015)
                     each country

Ethnolinguistic      An index capturing the           Alesina et
fractionalization    probability that two randomly    al. (2003)
                     selected individuals from a
                     country's population would
                     belong to different ethnic
                     groups. The primary source
                     used by Alesina et al. (2003)
                     is Encyclopedia Britannica
                     (2000). The dates are
                     generally from the early to
                     mid-1990s (the range is
                     1979-2001. and the median
                     year is 1994)

Landlockedness       A dummy variable that equals     CIA(2015)
                     1 if a country is fully
                     enclosed by land and 0
                     otherwise

Distance to coast    The mean distance of a           Gallup,
                     country to the nearest           Mellinger, and
                     coastline or sea-navigable       Sachs(2010)
                     river (in km)

Elevation            The mean elevation of a          G-Econ(2008)
                     country above sea level
                     (in km)

Precipitation        The average monthly              G-Econ(2008)
                     precipitation of a country
                     over the period 1961-1990
                     (in mm)

Years since          The number of years elapsed,     Putterman
agricultural         in 2000 AD, since the            (2006)
transition           transition to agriculture was
                     estimated to occur
                     (in thousand years)

Colonial origins     A dummy variable that            Nunn and
                     identifies the colonial          Puga(2012)
                     origin of a country as
                     Spanish, British, French.
                     Portuguese, or other European

Settler              European settler mortality       Acemoglu,
mortality rate       rates in the nineteenth          Johnson, and
                     century. Revised estimates       Robinson
                     of Albouy (2012) are also        (2001); Albouy
                     used as a check                  (2012)

Other environmental policy variables

C3-I                 The aggregate C3-I of            Bernauer and
                     Bernauer and Bohmelt (2013).     Bohmelt (2013)
                     It combines C[C.sup.policy]
                     (our baseline measure of
                     climate change policy) with
                     data on the emission level
                     and trend

Maximum lead         The maximum lead content of      Lovei (1998)
content              lease gasoline in 1996 (g/L)
                     permitted. The data are
                     rescaled so that a higher
                     value indicates more
                     environmental protection

Environmental        An environmental performance     Hsu et al.
performance index    index for year 2012. Its         (2014)
                     computation covers the
                     following nine policy
                     dimensions: health impacts,
                     air quality, water and
                     sanitation, water resources,
                     agriculture, forests,
                     fisheries, biodiversity and
                     habitat, and climate and
                     energy

Environmental        An index measuring the           Eliste and
regulation in        stringency of environmental      Fredriksson
agriculture          regulation in agriculture in     (2002)
                     year 1990. Covers water
                     pollution, air pollution,
                     land use, and biodiversity

Indicators of labor market regulations

Social security      A composite index of social      Botero et
laws                 security laws covering the       al. (2004)
                     following three areas: (1)
                     old-age, disability, and
                     death benefits: (2) sickness
                     and health benefits; and (3)
                     unemployment benefits

EPLEX                A composite indicator of         ILO (2015)
                     EPLEX governing the
                     termination of regular
                     contracts

Unemployment         The ratio of unemployment        Schindler and
insurance            insurance benefits a worker      Aleksynska
benefits ratio       receives relative to the         (2011)
                     worker's last gross earning

Unemployment         The fraction of unemployed       Schindler and
benefits coverage    individuals who collect          Aleksynska
                     unemployment insurance           (2011)
                     benefits

TABLE A2
List of Countries Included in the Estimations (by Legal Origin)

Common Law LO               (Avg = 0.243)

Australia                       0.000
Bangladesh                      0.507
Barbados                        0.094
Canada                          0.056
Fiji                            0.000
Gambia                          0.216
Ghana                           0.366
Grenada                         0.082
India                           0.736
Jamaica                         0.153
Kenya                           0.000
Malawi                          0.270
Malaysia                        0.585
New Zealand                     0.000
Nigeria                         0.544
Pakistan                        0.806
Papua New Guinea                0.000
Sierra Leone                    0.003
Singapore                       0.322
Sudan                           0.703
Trinidad and Tobago             0.107
Uganda                          0.196
United States                   0.086
Zambia                          0.000

French Civil Law LO         (Avg = 0.358)

Algeria                         0.603
Angola                          0.266
Armenia                         0.561
Azerbaijan                      0.462
Benin                           0.112
Burkina Faso                    0.262
Burundi                         0.123
Cambodia                        0.882
Cameroon                        0.395
Cape Verde                      0.175
Central African Republic        0.000
Chad                            0.219
Congo. Dem. Rep.                0.262
Congo. Rep.                     0.205
Cote d'Ivoire                   0.223
Dominican Republic              0.153
Egypt                           0.688
Gabon                           0.000
Guinea                          0.215
Indonesia                       0.590
Kazakhstan                      0.373
Kyrgyzstan                      0.265
Laos                            0.662
Madagascar                      0.230
Mali                            0.462
Mauritania                      0.390
Morocco                         0.829
Mozambique                      0.173
Niger                           0.350
Rwanda                          0.262
Sao Tome and Principe           0.162
Senegal                         0.448
Swaziland                       0.035
Syria                           0.569
Tajikistan                      0.524
Togo                            0.029
Tunisia                         0.751
Turkmenistan                    0.267
Uzbekistan                      0.794

Mixed Law LO                (Avg = 0.256)

Botswana                        0.299
Cyprus                          0.583
Guyana                          0.107
Israel                          0.501
Jordan                          0.505
Lesotho                         0.000
Mauritius                       0.045
Philippines                     0.168
Seychelles                      0.025
South Africa                    0.046
Sri Lanka                       0.795
Zimbabwe                        0.000

German Civil Law LO         (Avg = 0.667)

Georgia                         0.561
Korea. Rep.                     0.933
Mongolia                        0.508

Notes: The estimations include up to 78 countries listed in the table.
Figures in the table indicate statehood experience
accumulated over the period 1-1800 AD (State).

APPENDIX B: CONSTRUCTION OF STATE HISTORY VARIABLE (State)

The state history index of Putterman (2004) includes 39 periods of 50 years spanning from 1 to 1950 AD. In our study, we restrict the period to 1-1800 AD (i.e., 36 periods). The index is made up of the following three components:

[ST.sup.PRESENCE]:   Is there a government above the tribal level?

[ST.sup.AUTONOMY]:   Is there a government foreign or locally based?
                     [Local = 1; In between = 0.75: Foreign = 0.5]

[ST.sup.COVERAGE]:   How much 0f the territory of the modern
                     country was ruled by this government?
                     [>50% = 1 ; 25-50% = 0.75; 10-25% = 0.5;
                     <10% = 0.3]

The extent of state presence (S[A.sub.t]) in any particular 50 years period (t) is measured as the product of the scores on these components and 50. Consequently, a score of 0 indicates no presence of state. 25 reflects that a country has a supratribal authority but its entire territory is ruled by a foreign authority, and 50 indicates the presence of an autonomous nation, and so on.

SA, = [ST.sup.PRESENCE] X [ST.sup.AUTONOMY] X [ST.sup.C0VERAGE] X 50

(A1) 0 [less than or equal to] SA, [less than or equal to] 50, t = 1, 2, ..., 36

The length of state history, or state antiquity (State), is measured as the cumulative presence of state by combining data over the entire 36 periods. A 5% discount rate is applied to allow for the fact that states formed in the more distant past have relatively less influence on today's economic conditions. To ease interpretation, the series is scaled into 0 and 1 using its maximum possible value. Accordingly, state history for a particular country over 18 centuries (1-1800 AD) is calculated as follows:

State = [[SIGMA].sub.36.sub.t=1] [(1.05).sup.1-t] * S[A.sub.t]/ [[SIGMA].sub.36.sub.t=1] [(1.05).sup.1-t] * 50

The distribution of State and its components for all available 151 countries is given in Table Bl. In general, states in Asia and Europe are at least twice as old as those in Africa and America. Cambodia, China, Ethiopia, Japan. South Korea, Morocco, Myanmar, Nepal, and Turkey were among the nations with the highest levels of state experience (average State = 0.883) by 1800 AD. Countries in Oceania, including Australia, New Zealand, Fiji, and Papua New Guinea had zero state values as of 1800 AD. African nations such as Central African Republic, Gabon, Kenya, Lesotho, Liberia. Seychelles, Sierra Leone, Togo, Zambia, and Zimbabwe (average State = 0.006) are among those countries that have the youngest states, many of which also had no state experience at that time. It is interesting to observe that the emergence of states in several sub-Saharan African countries, such as the Central African Republic, Gabon, Kenya, and Zimbabwe, was in fact the outcome of the nineteenth-century colonization by the Western powers, either the British or the French, without which these nations may not have existed.

Tables B2 and B3 illustrate how the state history index is constructed for older and younger states, respectively, in each continent by giving a country example for each continent. Although our measure covers only 1-1800 AD, for completeness we provide information for all time periods available (1-1950 AD). The relatively experienced states considered are Egypt (Africa), China (Asia), Peru (America), France (Europe), and Australia (Oceania) whereas the relatively inexperienced states are Kenya (Africa), the Philippines (Asia), Uruguay (America), Latvia (Europe), and Papua New Guinea (Oceania).

TABLE B1
Average Values of State with Subindices across Continents

                 PRESENCE   AUTONOMY   COVERAGE   State      No.
                 of State   of State   of State           Countries

Africa            0.419      0.371      0.332     0.282      46
Asia              0.866      0.688      0.786     0.619      37
America           0.364      0.233      0.325     0.193      27
Europe            0.829      0.609      0.770     0.551      37
Oceania           0.000      0.000      0.000     0.000       4
All countries     0.608      0.473      0.541     0.407      151

Notes: PRESENCE of State is measured as: [36.summation over (t=1)]
[(1.05).sup.1-t] * [ST.sup.PRESENCE.sub.t] * 50)/([36.summation
over (t=1)] [(1.05).sup.1-t] * 50) and so on. [ST.sup.PRESENCE]
captures the presence of a government above the tribal level;
[ST.sup.AUTONOMY] reflects whether this government is foreign or
locally based; and [ST.sup.COVERAGE] measures the proportion of the
current territory covered by this government. State is given as
([36.summation over (t=1)] [(1.05).sup.1-t] * S[A.sub.t]) /
([36.summation over (t=1)] [(1.05).sup.1-t] * 50), where S[A.sub.t]
= [ST.sup.PRESENCE] x [ST.sup.AUTONOMY] x [ST.sup.COVERAGE] x 50.
Data for all countries available are shown.

TABLE B2
Examples of State History Construction for Experienced
States in Each Continent

Region          Africa              Asia                America
Country         Egypt               China               Peru

1-50 AD         1, 0.5, 1 (under    1,1,1 (ruled by     0, 0, 0
51-100 AD       Roman occupation)   the Han dynasty)    (nonexistence
                                                        of states)
101-150 AD
151-200 AD
201-250 AD                          1.1, 0.75 (the
251-300 AD                          Han empire was
301-350 AD                          split into
                                    several warring
                                    states)
351-400 AD
401-450 AD
451-500 AD
501-550 AD

551-600 AD
601-650 AD                          1,1,1 (unified      1.1, 0.5 (the
651-700 AD                          under the Sui       Huari state
                                    and, later, the     existed in the
                                    T'ang dynasties)    southern part)
701-750 AD                                              1.1.1 (Huari
                                                        grew to become
                                                        an empire)
751-800 AD                          1,1, 0.75
                                    (central regime
                                    broke down,
                                    followed by some
                                    political chaos)
801-850 AD                          1, 1, 1 (the        1.1, 0.75 (the
                                    T'ang power was     Huari empire
                                    restored)           collapsed and
                                                        the area was
                                                        ruled by a
                                                        number of
                                                        smaller
                                                        states)
851-900 AD      1, 1, 1
                (establishment
                of the
                Tulunid dynasty)
901-950 AD      1, 0.5, 1 (became   1.1, 0.75
                a foreign-based     (centralized
                caliphal            order collapsed
                province)           again and
                                    multiple kingdoms
                                    emerged)
951-1000 AD     1, 0.77, 1 (the
                Fatimid Caliphate
                was established
                in 973)

1001-1050 AD    1, 1, 1 (under      1, 1, 1 (power was
                the rule of the     united under the
                Egypt-based         Sung dynasty)
                Fatimids)
1051-1100 AD
1101-1150 AD
1151-1200 AD    1, 0.855, 1 (the
                Ayyubid dynasty
                became allegiant
                to the foreign-
                based Abbasid
                Caliphate in
                1171)

1201-1250 AD    1, 0.75, 1
                (became quasi-
                independent
                under the
                Ayyubids)
1251-1300 AD    1, 1,1 (under       1, 0.895, 1 (the
                the autonomy        Yuan dynasty was
                of the Mamluk       established
                dynasty)            in 1279)
1301-1350 AD                        1, 0.75, 1 (under
                                    the quasi-local
                                    Mongol rule)
1351-1400 AD                        1, 0.91, 1
                                    (the Yuan dynasty
                                    collapsed and was
                                    replaced by the
                                    Ming dynasty in
                                    1368)
1401-1450 AD                        1, 1, 1 (the
                                    Chinese rule was
                                    integrated under
                                    the Ming and
                                    Ch'ing dynasties)
1451-1500 AD                                            1,1,1 (the
                                                        Inca empire
                                                        united the
                                                        entire area)
1501-1550 AD    1, 0.67, 1                              1, 0.82, 1
                (conquered by                           (conquered
                the Ottomans                            by the
                in 1517)                                Spanish in
                                                        1532)
1551-1600 AD    1, 0.5, 1                               1,0.5. 1
                (under the                              (under the
                Ottomans' rule)                         Spanish
                                                        colonial
                                                        rule)
1601-1650 AD
1651-1700 AD
1701-1750 AD
1751-1800 AD
1801-1850 AD    1, 0.75, 1                              1, 0.79, 1
                (allowed partial                        (gained
                autonomy under                          independence
                the Ottomans                            in 1821)
                and the French)
1851-1900 AD    1, 0.66, 1                              1,1,1 (under
                (under the                              independent
                British                                 rule)
                occupation
                in 1882)
1901-1950 AD    1, 0.78, 1          1, 1, 0.875 (the
                (independent        Ch'ing rule
                from Britain        ceased in 1911
                in 1922)            and was replaced
                                    by the Republic
                                    of China, but
                                    the nation was
                                    politically
                                    fragmented)

Region          Europe                    Oceania
Country         France                    Australia

1-50 AD         1, 0.5, 1                 0, 0, 0
51-100 AD       (Gaul was under           (nonexistence
                the Romans' rule)         of states)
101-150 AD
151-200 AD
201-250 AD
251-300 AD
301-350 AD
351-400 AD
401-450 AD
451-500 AD      1, 1, 0.75
501-550 AD      (controlled by
                multiple Germanic
                kingdoms)
551-600 AD
601-650 AD
651-700 AD
701-750 AD
751-800 AD      1, 1, 0.895 (rule
                was unified by
                Charlemagne in
                771)
801-850 AD      1, 1, 1 (under
                unified
                domestic-rule)
851-900 AD      1, 1, 0.75 (rule
                was divided among
                several Frankish
                kingdoms)
901-950 AD
951-1000 AD
1001-1050 AD
1051-1100 AD
1101-1150 AD
1151-1200 AD
1201-1250 AD
1251-1300 AD    1, 1, 1 (under
                centralized rule and
                area expanded to nearly
                its current size)
1301-1350 AD
1351-1400 AD    1, 0.75, 0.75 (a large
                part of the area was
                held by the English
                during the Hundred
                Years War)
1401-1450 AD
1451-1500 AD    1.1,1 (the Hundred
                Years War ended and
                territory was regained)
1501-1550 AD
1551-1600 AD    1, 1, 0.75 (rule was
                divided by domestic
                religious wars)
1601-1650 AD    1, 1, 1 (mostly under
                unified control)
1651-1700 AD
1701-1750 AD
1751-1800 AD
1801-1850 AD                              1,0.75, 1 (British
                                          settlement)
1851-1900 AD                              1,1,1 (gained
                                          independence from
                                          Great Britain)
1901-1950 AD

Notes: The values in each entry reflect [ST.sup.PRESENCE],
[ST.sup.AUT0N0MY], and [ST.sup.COVERAGE]. For instance,
[ST.sup.PRESENCE] is 1, [ST.sup.AUTONOMY] is 0.5, and
[ST.sup.COVERACE] is 1 for Egypt in every 50-year period from 1 to
850 AD.

Source: Putterman (2004).

TABLE B3
Examples of State History Construction for Inexperienced
States by Continent

Region          Africa          Asia               America
Country         Kenya           Philippines        Uruguay

1-50 AD         0, 0, 0         0, 0, 0            0, 0, 0
                (nonexistence   (nonexistence      (nonexistence
                of states)      of states)         of states)
51-100 AD
101-150 AD
151-200 AD
201-250 AD
251-300 AD
301-350 AD
351-400 AD
401-450 AD
451-500 AD
501-550 AD
551-600 AD
601-650 AD
651-700 AD
701-750 AD
751-800 AD
801-850 AD
851-900 AD
901-950 AD
951-1000 AD
1001-1050 AD
1051-1100 AD
1101-1150 AD
1151-1200 AD
1201-1250 AD
1251-1300 AD
1301-1350 AD
1351-1400 AD
1401-1450 AD
1451-1500 AD
1501-1550 AD                    1,1, 0.75 (ruled
                                by two local
                                Islamic
                                sultanates)
1551-1600 AD                    1,0.75, 0.75
                                (partially
                                occupied by the
                                Spanish)
1601-1650 AD                    1, 0.5, 1 (rule
                                was unified by
                                the Spanish)
1651-1700 AD
1701-1750 AD                                       1, 0.5. 1
                                                   (occupied by
                                                   foreign forces)
1751-1800 AD
1801-1850 AD                                       1, 0.72. 1
                                                   (independent
                                                   since 1828)
1851-1900 AD                                       1,1.1 (under
                                                   independent rule)
1901-1950 AD    1, 0.5, 1       1, 0.75, 1
                (became a       (local autonomy
                colony of       coexisted with
                the British)    control by the
                                United States)

Region          Europe                 Oceania
Country         Moldova                Papua
                                       New Guinea

1-50 AD         0, 0, 0                0, 0, 0
                (nonexistence          (nonexistence
                of states)             of states)
51-100 AD
101-150 AD
151-200 AD
201-250 AD
251-300 AD
301-350 AD
351-400 AD
401-450 AD
451-500 AD
501-550 AD
551-600 AD
601-650 AD
651-700 AD
701-750 AD
751-800 AD
801-850 AD
851-900 AD
901-950 AD      1. 0.5, 1
                (ruled by
                Kievan Rus)
951-1000 AD
1001-1050 AD
1051-1100 AD
1101-1150 AD
1151-1200 AD
1201-1250 AD    1,0.5, 1 (pledged
                allegiance to the
                Tartars)
1251-1300 AD
1301-1350 AD
1351-1400 AD
1401-1450 AD    1, 0.5, 1 (became
                part of the
                principality of
                Moldavia until 1513)
1451-1500 AD
1501-1550 AD

1551-1600 AD    1, 0.5, 1
                (belonged to the
                Ottoman Empire)
1601-1650 AD
1651-1700 AD
1701-1750 AD
1751-1800 AD
1801-1850 AD    1, 0.5, 1 (largely
                occupied
                by Soviet Union)
1851-1900 AD
1901-1950 AD                           1, 0.5, 0.75
                                       (occupied by the
                                       British, Germans,
                                       and Dutch)

Note: The values in each entry reflect [ST.sup.PRESENCE],
[ST.sup.AUTONOMY], and [ST.sup.COVERAGE].

Source: Putterman (2004).

ABBREVIATIONS

C3-I: Climate Change Cooperation Index

EPLEX: Employment Protection Legislation

FAR: Fractionally Resampled Anderson-Rubin

GDP: Gross Domestic Product

LOT: Legal Origins Theory

SD: Standard Deviation

UNFCCC: United Nations Framework Convention on Climate Change

doi: 10.1111/ecin.12441

REFERENCES

Acemoglu, D., S. Johnson, and J. A. Robinson. "The Colonial Origins of Comparative Development: An Empirical Investigation." American Economic Review, 91, 2001, 1369-401.

Acemoglu, D., C. Garcia-Jimeno, and J. A. Robinson. "State Capacity and Economic Development: A Network Approach." American Economic Review. 105, 2015, 2364-409.

Albouy, D. Y. "The Colonial Origins of Comparative Development: An Empirical Investigation: Comment." American Economic Review. 102, 2012, 3059-76.

Alesina, A., A. Devleeschauwer, W. Easterly. S. Kurlat. and R. Wacziarg. "Fractionalization." Journal of Economic Growth. 8. 2003, 155-94.

Andersen, R. Modern Methods for Robust Regression. London, UK: Sage Publications, 2008.

Anderson, T.W., and H. Rubin. "Estimation of the Parameters of a Single Equation in a Complete System of Stochastic Equations," Annals of Mathematical Statistics, 20, 1949, 46-63.

Ang, J. B. "Are Modern Financial Systems Shaped by State Antiquity?" Journal of Banking & Finance, 37, 2013a, 4038-58.

--. "Institutions and the Long-run Impact of Early Development." Journal of Development Economics, 105, 2013b, 1-18.

Barbier, E. B., R. Damania, and D. Leonard. "Corruption, Trade and Resource Conversion." Journal of Environmental Economics and Management. 50.2005.276-99.

Becerra. O., E. A. Cavallo, and C. Scartascini. "The Politics of Financial Development: The Role of Interest Groups and Government Capabilities." Journal of Banking & Finance, 36, 2012, 626-43.

Berkowitz, D., K. Pistor, and J.-F. Richard. "Economic Development, Legality, and the Transplant Effect." European Economic Review, 47, 2003, 165-95.

Berkowitz, D., M. Caner, and Y. Fang. "The Validity of Instruments Revisited." Journal of Econometrics, 166, 2012, 255-66.

Bernauer, T., and T. Bohmelt. "National Climate Policies in International Comparison: The Climate Change Cooperation Index." Environmental Science and Policy, 25, 2013, 196-206.

Besley, T., and T. Persson. "The Origins of State Capacity: Property Rights. Taxation, and Politics." American Economic Review, 99, 2009, 1218-44.

--. Pillars of Prosperity: The Political Economy of Development Clusters. Princeton, NJ: Princeton University Press, 201 I.

Besley, T., E. llzetzki, and T. Persson. "Weak States and Steady States: The Dynamics of Fiscal Capacity." American Economic Journal: Macroeconomics, 5, 2013. 205-35.

Botero, J. C., S. Djankov. R. L. Porta, F. Lopez-de-Silanes, and A. Shleifer. "The Regulation of Labor." Quarterly Journal of Economics. 119, 2004. 1339-82.

Burkett. J. P. C. Humblet, and L. Putterman. "Preindustrial and Postwar Economic Development: Is There a Link?" Economic Development and Cultural Change, Al, 1999, 471-95.

Caballero, R. J., K. N. Cowan, E. M. R. A. Engel, and A. Micco. "Effective Labor Regulation and Microeconomic Flexibility." Journal of Development Economics, 101.2013, 92-104.

CIA. The World Factbook. Washington, DC: Central Intelligence Agency, 2015.

Comin. D., W. Easterly, and E. Gong. "Was the Wealth of Nations Determined in 1000 BCT' American Economic Journal: Macroeconomics, 2, 2010. 65-97.

Congleton, R. "Political Institutions and Pollution Control." Review of Economics and Statistics, 74. 1992. 412-21.

Damania. R., P. G. Fredriksson, and J. A. List. "Trade Liberalization, Corruption, and Environmental Policy Formation: Theory and Evidence." Journal of Environmental Economics and Management, 46. 2003, 490-512.

David. P. A. "Clio and the Economics of QWERTY." American Economic Review, 75, 1985, 332-37.

Diamond, J. Guns, Germs and Steel: The Fates of Human Societies. New York: Norton. 1997.

Djankov, S., R. L. Porta. F. Lopez-de-Silanes, and A. Shleifer. "The Regulation of Entry." Quarterly Journal of Economics, 117, 2002, 1-37.

--. "Courts." Quarterly Journal of Economics, 118, 2003, 453-517.

Eliste. P., and P. G. Fredriksson. "Environmental Regulations, Transfers, and Trade: Theory and Evidence." Journal of Environmental Economics and Management, 43, 2002, 234-50.

Farzin. H. Y., and C. A. Bond. "Democracy and Environmental Quality." Journal of Development Economics, 81, 2006. 213-35.

Fredriksson, P. G., and E. Neumayer. "Democracy and Climate Change Policies: Is History Important?" Ecological Economics, 95, 2013. 11-19.

Fredriksson. P. G., and J. Svensson. "Political Instability, Corruption and Policy Formation: The Case of Environmental Policy." Journal of Public Economics, 87, 2003, 1383-405.

Fredriksson. P. G., and J. R. Wollscheid. "Legal Origins and Climate Change Policies in Former Colonies." Environmental and Resource Economics, 62, 2015, 309-27.

Gallup. J. L., A. D. Meilinger, and J. D. Sachs. "Geography Datasets." Center for International Development, Harvard University, 2010.

Galor, O., and D. N. Weil. "Population, Technology, and Growth: From Malthusian Stagnation to the Demographic Transition and Beyond." American Economic Review, 90, 2000, 806-28.

G-Econ. "Geographically-based Economic Data." Yale University. 2008.

Gennaioli, N., and I. Rainer. "The Modern Impact of Precolonial Centralization in Africa." Journal of Economic Growth, 12, 2007, 185-234.

Glaeser. E. L., and A. Shleifer. "Legal Origins." Quarterly Journal of Economics, 117, 2002, 1193-229.

Greif. A. "Historical and Comparative Institutional Analysis." American Economic Review. 88, 1998, 80-84.

Haltiwanger, J., S. Scarpetta. and H. Schweiger. "Assessing Job Flows across Countries: The Role of Industry, Firm Size, and Regulations." World Bank Policy Research Working Paper No. 4070, 2006.

Hodgson, G. M. "The Approach of Institutional Economics." Journal of Economic Literature, 36. 1998, 166-92.

Hotte, L., and S. Winer. "Environmental Regulation and Trade Openness in the Presence of Private Mitigation." Journal of Development Economics, 97, 2012, 46-57.

Hsu, A., J. Emerson. M. Levy, A.d. Sherbinin. L. Johnson, O. Malik, J. Schwartz, and M. Jaiteh. "The Environmental Performance Index." New Haven. CT: Yale Center for Environmental Law and Policy, 2014.

ILO. "Employment Protection Legislation: Summary Indicators in the Area of Terminating Regular Contracts (Individual Dismissals)." Geneva, Switzerland: International Labour Office, 2015.

IPCC. "Climate Change 2014: Synthesis Report. Contribution of Working Groups 1. II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change." Geneva, Switzerland: IPCC. 2014.

Klerman. D., and P. G. Mahoney. "Legal Origin?" Journal of Comparative Economics, 35, 2007, 278-93.

Klerman, D. M., P. G. Mahoney, H. Spamann, and M. I. Weinstein. "Legal Origin or Colonial History?" Journal of Legal Analysis, 3, 2011, 379-409.

Kremer. M. "Population Growth and Technological Change: One Million B.C. to 1990." Quarterly Journal of Economics, 108, 1993, 681-716.

La Porta, R., L.-D.-S. Florencio, A. Shleifer, and R. W. Vishny. "Legal Determinants of External Finance." Journal of Finance, 52, 1997, 1131-50.

--. "Law and Finance." Journal of Political Economy, 106. 1998, 1113-55.

La Porta, R., F. Lopez-de-Silanes, and A. Shleifer. "The Economic Consequences of Legal Origins." Journal of Economic Literature, 46, 2008, 285-332.

Lafontaine, F., and J. Sivadasan. "The Microeconomic Implications of Input Market Regulations: Cross-Country Evidence from Within the Firm." Ross School of Business Paper No. 1069, 2007.

Levine, R. "Law, Endowments and Property Rights." Journal of Economic Perspectives. 19, 2005, 61-88.

List, J. A., and D. M. Sturm. "How Elections Matter: Theory and Evidence from Environmental Policy." Quarterly Journal of Economics, 121.2006. 1249-81.

Lovei, M. "Phasing Out Lead from Gasoline in Central and Eastern Europe, Health Issues, Feasibility, and Policies." World Bank Technical Paper No. 397, 1998.

Mamatzakis, E., M. G. Tsionas, S. C. Kumbhakar. and A. Koutsomanoli-Filippaki. "Does Labour Regulation Affect Technical and Allocative Efficiency? Evidence from the Banking Industry." Journal of Banking and Finance, 61, 2015, S84-98.

Marchand, S. "The Colonial Origins of Deforestation: An Institutional Analysis." Environment and Development Economics, 21, 2016, 318-49.

McNeill, J. R., and W. H. McNeill. The Human Web: A Bird's-Eve View of World History. New York: Norton, 2003.

Michalopoulos, S., and E. Papaioannou. "Pre-Colonial Ethnic Institutions and Contemporary African Development." Econometrica, 81, 2013, 113-52.

--. "National Institutions and Subnational Development in Africa." Quarterly Journal of Economics, 129, 2014, 151-213.

Mulligan, C., and A. Shleifer. "The Extent of the Market and the Supply of Regulation." Quarterly Journal of Economics. 120, 2005, 1445-73.

North. D. C. Understanding the Process of Economic Change. Princeton, NJ: Princeton University Press, 2006.

Nunn. N., and D. Puga. "Ruggedness: The Blessing of Bad Geography in Africa." Review of Economics and Statistics, 94, 2012, 20-36.

Oliva, P. "Environmental Regulations and Corruption: Automobile Emissions in Mexico City." Journal of Political Economy, 123, 2015, 686-724.

Oto-Peralias, D., and D. Romero-Avila. "The Distribution of Legal Traditions around the World: A Contribution to the Legal-Origins Theory." Journal of Law and Economics, 57, 2014, 561-628.

Pierre, G., and S. Scarpetta. "Do Firms Make Greater Use of Training and Temporary Employment When Labor Adjustment Costs Are High?" IZA Journal of Labor Policy, 2, 2013, 1-17.

Ponzetto, G. A. M., and P. A. Fernandez. "Case Law vs. Statute Law: An Evolutionary Comparison." Journal of Legal Studies, 37, 2008, 379-430.

Putterman, L. "Can an Evolutionary Approach to Development Predict Post-War Economic Growth?" Journal of Development Studies, 36, 2000, I -30.

--. "State Antiquity Index (Statehist) Dataset. Version 3.1." Brown University, 2004.

--. "Agricultural Transition Year: Country Data Set." Brown University, 2006.

--. "Agriculture, Diffusion and Development: Ripple Effects of the Neolithic Revolution." Economica, 75, 2008, 729-48.

Putterman, L., and D. N. Weil. "Post-1500 Population Flows and the Long-Run Determinants of Economic Growth and Inequality." Quarterly Journal of Economics, 125. 2010, 1627-82.

Reyes, J. W. "Lead Exposure and Behavior: Effects on Antisocial and Risky Behavior among Children and Adolescents." Economic Inquiry, 53, 2015, 1580-605.

Roberts, R., and K. Mann. "Law in Colonial Africa," in Law in Colonial Africa, edited by R. Roberts and K. Mann. Portsmouth. NH: Heinemann. 1991, 3-58.

Roe. M. J. "Juries and the Political Economy of Legal Origin." Journal of Comparative Economics, 35. 2007. 294-308.

Schapera, 1. Tribal Innovators: Tswana Chiefs and Social Change. 1795-1940. New York: Humanities Press, 1970.

Schindler, M., and M. Aleksynska. "Labor Market Regulations in Low-, Middle- and High-Income Countries: A New Panel Database." IMF Working Paper No. 11/154, 2011.

Shea, J. "Instrument Relevance in Multivariate Linear Models: A Simple Measure." Review of Economics and Statistics, 49, 1997, 348-52.

Sigman, H. "Decentralization and Environmental Quality: An International Analysis of Water Pollution Levels and Variation." Land Economics, 90, 2014, 114-30.

Sokoloff, K. L., and S. L. Engerman. "Institutions, Factor Endowments, and Paths of Development in the New World." Journal of Economic Perspectives. 14. 2000, 217-32.

Tosh, J. Clean Leaders and Colonial Chiefs in Lemgo: The Political History of an East African Stateless Society c. 1800-1939. Oxford, UK: Oxford University Press, 1978.

World Bank. "World Development Indicators 2015 Database." Philadelphia, PA: World Bank, 2015a.

--. "World Governance Indicators 2015 Database." Philadelphia, PA: World Bank, 2015b.

JAMES B. ANG and PER G. FREDRIKSSON *

* The authors are grateful for helpful comments on this article from the editor Chuck Mason; three knowledgeable referees, Roger von Haefen, Kathryn Perry, and Melissa Lane; the seminar participants at Florida Gulf Coast University; and the participants at the SSRCdemocracy Working Group on Climate Change Workshop. Princeton University. The usual disclaimers apply. James Ang acknowledges financial support from Singapore Ministry of Education Academic Research Fund Tier 1, and Per Fredriksson acknowledges research support from the College of Business, University of Louisville.

Ang: Associate Professor, Department of Economics, Nanyang Technological University, Singapore 637332, Singapore. Phone +65 65927534, Fax +65 67955797, E-mail james.ang@ntu.edu.sg

Fredriksson: Professor, Department of Economics, University of Louisville, Louisville, KY 40292. Phone 502-8524858, Fax 502-8527672, E-mail per.fredriksson@louisville.edu

(1.) See, for example, Congleton (1992), Damania, Fredriksson, and List (2003), Fredriksson and Svensson (2003), Barbier, Damania, and Leonard (2005), Farzin and Bond (2006), List and Sturm (2006), Hotte and Winer (2012), and Oliva (2015).

(2.) The literature on the very long run determinants of economic development include, for example, Kremer (1993), Galor and Weil (2000). Comin. Easterly, and Gong (2010), and Putterman and Weil (2010).

(3.) See Klerman and Mahoney (2007) and Roe (2007) for critical views of the LOT by La Porta, Lopez-de-Silanes, and Shleifer (2008).

(4.) Botero et al. (2004) show that British common law legal origin is associated with a substantial decrease in labor regulations.

(5.) See Michalopoulos and Papaioannou (2013), Supplementary Appendix Table 16.

(6.) One may suspect that statehood experience may be highly correlated with democratic capital and economic development. Their correlations are surprisingly low: -0.121 between State and GDP per capita and -0.138 between State and democratic capital.

(7.) Territorial integrity appears important for building bureaucratic capacity, possibly due to the associated stability.

(8.) This identification strategy is based on the well-known proposition of Diamond (1997) that agricultural transition significantly increased food supply, which enabled settled agricultural villages with small-scale political entities governed by supra-tribal authorities to compound into larger polities, leading to the emergence of fully Hedged states. Dramatic improvements in agricultural productivity also enabled polities to enhance their fiscal capacity through raising more tax revenues. Hence, the onset of the institutionalization of power relations triggered by the Neolithic transition was a key catalyst for state formation and the development of state capacity. This hypothesis suggests that an earlier transition to agriculture is expected to have a positive influence on the length of statehood experience.

Caption: FIGURE 1 Distribution of State (1-1800 AD) across the World

Caption: FIGURE 2 The Correlation between Climate Change Policy Index (C[C.sup.policy]) and Statehood Experience (State)

Caption: FIGURE 3 The Relationships between Climate Change Policy Index and Statehood Experience across Different Legal Traditions

Caption: FIGURE 4 Partial Effect of State x Common Law LO

TABLE 1
Descriptive Statistics of Key Variables

Variable                            Observations    Mean

C[C.sup.policy]                          78        35.35
  C[C.sup.policy]--common Law LO         24        33.65
  C[C.sup.policy]--Mixed Law LO          12        34.23
  C[C.sup.policy]--0ther LO              42        36.64
State                                    78         0.32
  State--Common Law LO                   24         0.24
  State--Mixed Law LO                    12         0.26
  State--Other LO                        42         0.38
Common Law LO                            78         0.31
Mixed LO                                 78         0.15
Other LO                                 78         0.54
State X Common Law LO                    78         0.07
State X Mixed Law LO                     78         0.04
State X Other LO                         78         0.20
Manufacturing (% GDP)                    76        13.34
Income per capita (logged)               76         7.14
Democratic capital                       68         0.42
Control of corruption                    78        40.98
Political stability                      78        37.72
Ethnolinguistic frac.                    72         0.53
Latitude                                 78         0.22
Landlocked                               75         0.31
Elevation                                78       560.83
Distance to coast                        78       429.94
Precipitation                            78        95.10

Variable                              SD     Min.      Max.

C[C.sup.policy]                     10.64   14.49      62.74
  C[C.sup.policy]--common Law LO     9.13   18.75      56.77
  C[C.sup.policy]--Mixed Law LO     10.12   14.49      52.16
  C[C.sup.policy]--0ther LO         11.61   16.46      62.74
State                                0.26    0.00       0.93
  State--Common Law LO               0.27    0.00       0.81
  State--Mixed Law LO                0.27    0.00       0.79
  State--Other LO                    0.25    0.00       0.93
Common Law LO                        0.46    0.00       1.00
Mixed LO                             0.36    0.00       1.00
Other LO                             0.50    0.00       1.00
State X Common Law LO                0.18    0.00       0.81
State X Mixed Law LO                 0.14    0.00       0.79
State X Other LO                     0.26    0.00       0.93
Manufacturing (% GDP)                7.69    2.89      38.67
Income per capita (logged)           1.46    4.82      10.62
Democratic capital                   0.35    0.00       1.00
Control of corruption               26.39    1.46      97.57
Political stability                 26.17    0.48      90.38
Ethnolinguistic frac.                0.25    0.00       0.93
Latitude                             0.15    0.00       0.67
Landlocked                           0.46    0.00       1.00
Elevation                          510.29   31.64    2672.87
Distance to coast                  454.08    3.68    1875.13
Precipitation                       65.63    2.91     259.95

Notes: The descriptive statistics provided in the table include
up to 78 countries used in the baseline regressions. Sources and
definition of data are described in the text and in the Appendices.

TABLE 2
Main Results

Dep. Var. =          (1)       (2)          (3)
C[C.sup.policy]

State               5.346                17.194 **
                   (1.027)                (2.188)

Common Law LO                 -2.996       6.186
                             (-1.155)     (1.595)

Mixed Law LO                  -2.409       5.710
                             (-0.714)     (1.355)

State X Common                          -28.069 ***
Law LO                                   (-2.914)

State X Mixed                            -23.365 *
Law LO                                   (-1.743)

R-squared           0.018     0.018        0.130

Observations         78         78          78

Notes: Robust Standard errors are used and t statistics
are reported in the parentheses. The intercept estimate is not
shown. The omitted legal origin group is "other legal origins,"
which includes 39 French and 3 German civil law countries.

* Significant at 10%; ** significant at 5%; *** significant
at 1%.

TABLE 3
Robustness Analysis: Including Control Variables

Dep. Var. =             (1)          (2)         (3)         (4)
C[C.sup.policy]

State                  12.99       14.03 *      13.09      15.71 *
                       (1.64)      (1.81)      (1.56)      (1.99)

Common Law LO          5.93 *       2.18        3.53        1.98
                       (1.68)      (0.54)      (0.66)      (0.45)

Mixed Law LO            2.23        3.97        3.58        3.26
                       (0.54)      (0.85)      (0.71)      (0.79)

State X Common       -27.13 ***   -20.88 **   -22.30 **   -21.99 **
Law LO                (-3.00)      (-2.11)     (-2.01)     (-2.15)

State X Mixed          -17.43      -25.56*     -19.56      -26.22*
Law LO                (-1.30)      1-1.75)     (-1.24)     (-1.76)

Manufacturing         0.54 ***
(% GDP)                (3.03)

Income per capita                  2.50 **
(logged)                           (2.39)

Democratic capital                              1.43
                                               (0.26)

Control of                                                 3.79 *
corruption                                                 (1.76)

Political                                                   -0.55
stability                                                  (-0.28)

Ethnolinguistic
frac.

Latitude

Landlocked

Elevation

Distance to coast

Precipitation

R-squared              0.243        0.195       0.089       0.184

Observations             77          77          68          78

Dep. Var. =             (5)         (61         (7)
C[C.sup.policy]

State                 13.66 *     15.94 *      6.30
                      (1.69)      (1.73)      (0.641

Common Law LO          5.47        7.03        5.91
                      (1.35)      (1.61)      (1.05)

Mixed Law LO           3.86        5.54        0.97
                      (0.80)      (1.07)      (0.15)

State X Common       -24.61 **   -28.43 **   -25.75 **
Law LO                (-2.39)     (-2.60)     (-2.16)

State X Mixed         -20.68      -21.42      -20.08
Law LO                (-1.41)     (-1.52)     (-1.U)

Manufacturing                                0.55 ***
(% GDP)                                       (2.80)

Income per capita                              2.07
(logged)                                      (1.24)

Democratic capital                             2.64
                                              (0.55)

Control of                                     4.29
corruption                                     (114)

Political                                      -0.71
stability                                     (-0.27)

Ethnolinguistic        -6.98                   3.34
frac.                 (-1.30)                 (0.48)

Latitude                           0.03        0.09
                                  (0.40)      (0.86)

Landlocked                         -0.64       -0.22
                                  (-0.19)     (-0.05)

Elevation                          0.00        0.00
                                  (0.99)      (0.75)

Distance to coast                  0.00        0.01
                                  (0.13)      (1.07)

Precipitation                      -0.00       -0.00
                                  (-0.01)     (-0.14)

R-squared              0.150       0.155       0.339

Observations            72          75          65

Notes: Robust standard errors are used and t statistics are
reported in the parentheses. The intercept estimate is not shown.
The omitted legal origin group is "other legal origins." which
includes 39 French and 3 German civil law countries. The following
countries are excluded due to unavailability of data: Israel
(column 1); Zimbabwe (column 2): Barbados, Grenada, New Zealand,
Sao Tome and Principe. Canada, United States, South Korea,
Australia, Israel, and Seychelles (column 3); Democratic Republic
of the Congo, Kyrgyzstan, Sao Tome and Principe, Barbados,
Seychelles, and Grenada (column 5); and Barbados, Grenada, and
Seychelles (column 6). The last column includes all control
variables and hence has only 65 observations.

* significant at 10%; ** significant at 5%; *** significant at 1%.

TABLE 4
Alternative Statehood Periods

                      (1)               (2)              (3)
Dep. Var. =      State History     State History    State History
C[C.sup.         up to 1500 AD     up to 1500 AD    up to 1650 Al)
policy]            (State 1)        (State 1) &       (State 1)
                                   1501-1800 AD
                                     (State 2)

Slate 1          12.960 (1.596)    9.893 (1.135)    14.299(1.655)

Common Law LO    3.873 (1.091)    6.404 (1.097)     5.415 (1.395)

Mixed Law LO     2.601 (0.591)     9.099(1.233)     3.929 (0.843)

State 1 X         -26.194 ***        -24.362 *       -27.307 ***
Common Law LO       (-2.737)         (-1.849)          (-2.680)

State 1 X           -15.928       -7.234 (-0.588)      -18.612
Mixed Law LO        (-1.351)                           (-1.470)

State 2                            6.660(0.810)

State 2 X                         -5.317 (-0.397)
Common Law LO

Stale 2 X                             -21.138
Mixed Law LO                         (-1.209)

R-squared            0.148             0.166            0.152

Observations           75               75                75

Geographic            Yes               Yes              Yes
controls

                       (4)               (5)
Dep. Var. =       State History     State History
C[C.sup.          up to 1650 AD     1751-1800 AD
policy]            (State 1) &        (State 2)
                   1651-1800 AD
                    (State 2)

Slate 1           11.603 (1.217)

Common Law LO     6.763 (1.045)     7.031 (0.994)

Mixed Law LO      8.757 (1.022)     2.917 (0.339)

State 1 X           -28.707 *
Common Law LO        (-1.919)

State 1 X        -13.139 (-0.986)
Mixed Law LO

State 2           5.888 (0.667)     0.186 (0.990)

State 2 X        -1.330 (-0.086)       -0.328
Common Law LO                         (-1.311)

Stale 2 X            -14.029           -0.146
Mixed Law LO         (-0.720)         (-0.426)

R-squared             0.164             0.094

Observations            75               75

Geographic             Yes               Yes
controls

                       (6)                (7)
Dep. Var. =       State History      State History
C[C.sup.          up to 1750 AI)     up to 1750 AI)
policy]             (State 1)         (State 1) &
                                      1751-1800 AD
                                       (State 2)

Slate 1          15.503 * (1.726)    15.409 (1.528)

Common Law LO     6.539 (1.565)      5.596 (0.808)

Mixed Law LO      5.046 (1.022)      3.321 (0.389)

State 1 X           -28.166 **         -34.780 **
Common Law LO        (-2.643)           (-2.026)

State 1 X        -20.792 (-1.539)   -21.804 (-1.561)
Mixed Law LO

State 2                              0.025 (0.136)

State 2 X                            0.124 (0.349)
Common Law LO

Stale 2 X                            0.098 (0.294)
Mixed Law LO

R-squared             0.156              0.163

Observations            75                 75

Geographic             Yes                Yes
controls

Notes: Robust standard errors are used and t statistics are
reported in the parentheses. The intercept estimate is not shown.
The omitted legal origin group is "other legal origins." which
include 39 French and 3 German civil law countries. The geographic
controls included are latitude, landlockedness. mean elevation,
distance to the nearest coast or river, and precipitation. The
estimations exclude Barbados, Grenada, and Seychelles.
*
Significant at 10%; ** significant at 5%; *** significant at 1%.

TABLE 5
Further Robustness Checks

                                      (2)             (3) Robust
                      (1)           Include           Regression
                   La Porta        Continent        (For Presence
                  et al. Data       Dummies          of Outliers)

State               16.93 *      9.74 ** (3.83)     15.64 * (1.84)
                    (1.79)

Common Law LO        6.54         2.48 (0.66)        6.31 (1.27)
                    (1.56)
Mixed Law LO                      1.80 (0.43)        4.93 (0.76)

State X Common    -29.18 ***       -22.47 ***         -28.80 **
Law LO              (-2.72)         (-6.36)            (-2.34)

State X Mixed                    -16.56 (-1.75)     -20.31 (-1.30)
Law LO

Settler
mortality

Colonial
origins
(F - stat)

R-squared            0.176           0.323              0.144

Observations          75               75                 75

Geographic            Yes             Yes                Yes
controls

                        (4)               (5)              (6)
                      Include          Include            Using
                      Settler          Colonial          Presence
                     Mortality          Origin           of State
                                        Dummies

State               14.82 (1.46)      7.34 (1.05)     8.95 ** (3.29)

Common Law LO       6.07 (1.31)      -4.57 (-1.05)     6.19 (1.09)

Mixed Law LO        4.11 (0.68)      -5.93 (-1.27)     4.84 (1.25)

State X Common        -22.88 *         -16.33 *      -17.61 * (-2.56)
Law LO                (-1.85)           (-1.76)

State X Mixed         -22.26 *          -10.27        -12.30 (-1.47)
Law LO                (-1.95)           (-0.83)

Settler               -0.01 **
mortality             (-2.29)

Colonial                              8.03 *** =
origins                                 0.000)
(F - stat)

R-squared              0.264             0.426            0.122

Observations             51               75                75

Geographic              Yes               Yes              Yes
controls

                        (7)                (8)
                       Using              Using
                      Autonomy         Coverage of
                      of State            State

State              8.65 ** (3.19)    14.75 ** (4.39)

Common Law LO       5.70 (1.15)        7.27 (1.28)

Mixed Law LO        3.77 (0.99)         5.74(1.66)

State X Common       -20.08 **          -23.24 **
Law LO                (-2.89)            (-3.26)

State X Mixed      -13.99 (-1.69)     -16.13 (-1.68)
Law LO

Settler
mortality

Colonial
origins
(F - stat)

R-squared              0.123              0.149

Observations             75                 75

Geographic              Yes                Yes
controls

Notes: Robust standard errors are used and t statistics are
reported in the parentheses. The intercept estimate is not shown.
The omitted legal origin group is "other legal origins," which
includes 39 French and 3 German civil law countries. The geographic
controls included are latitude, landlockedness, mean elevation,
distance to the nearest coast or river, and precipitation. The
continent dummies are Asia, Africa, America, and Europe (Oceania is
the excluded group). Due to the unavailability of data, Barbados,
Grenada, and Seychelles are excluded from all columns. Column 4
also excludes another 24 countries, including Armenia, Azerbaijan,
Botswana, Cambodia, Cape Verde, Cyprus. Georgia, Israel, Jordan,
Kazakhstan, Kyrgyzstan, Lesotho. Malawi, Mongolia, Mozambique.
Philippines, Sao Tome and Principe, Swaziland, Syria, Tajikistan,
Turkmenistan, Uzbekistan, Zambia, and Zimbabwe. "Significant at
10%; "significant at 5%; ""significant at 1%.

TABLE 6
Addressing Endogeneity Issues

                                       (1)                  (2)
                                   First-Stage          First-Stage
                                   Regression            Regression
                               (Dep. Var. = State)      (Dep. Var. =
                                                          State X
                                                           Common
                                                          Law LO)

Common Law LO                  -0.215 ** (-2.577)     -0.056 (-1.142)
Mixed Law LO                     -0.016 (-0.156)        0.014(0.733)
Agr. transition                 0.045 ** (2.238)       0.002 (0.425)
Agr. transition X Common        0.041 ** (2.314)     0.095 *** (12.936)
  Law LO
Agr. transition X Mixed          -0.021 (-1.137)      -0.002 (-0.786)
  Law LO
State
State X Common Law LO
Slate X Mixed Law LO
First-stage F statistic for           19.16                55.88
  excluded instrument
Shea's partial ft-squared             0.114                0.263
Anderson-Rubin Wald test               --                    --
FAR test
R-squared                             0.500                0.777
No. obs.                               73                    73
Geographic controls                    Yes                  Yes

                                      (3)                  (4)
                                  First-Stage         Second-Stage
                                  Regression           Regression
                                 (Dep. Var. =         (Dep. Var. =
                                    State X         C[C.sup.policy])
                                 Mixed Law LO)

Common Law LO                   -0.000 (-0.021)       13.874(1.513)
Mixed Law LO                     0.059 (0.759)        9.694(1.233)
Agr. transition                  0.001 (0.075)
Agr. transition X Common         0.000(0.019)
  Law LO
Agr. transition X Mixed        0.050 *** (6.046)
  Law LO
State                                                 39.300(1.380)
State X Common Law LO                              -50.394 ** (-2.107)
Slate X Mixed Law LO                                -30.112 (-1.261)
First-stage F statistic for          12.98                 --
  excluded instrument
Shea's partial ft-squared            0.455                 --
Anderson-Rubin Wald test              --            10.65 (p = 0.013)
FAR test                                            6.861 (p = 0.147)
R-squared                            0.749                0.017
No. obs.                              73                   73
Geographic controls                   Yes                  Yes

Notes: Robust Standard errors are used and t statistics are
reported in the parentheses. The intercept estimate is not shown.
The omitted legal origin group is "other legal origins," which
includes 39 French and 3 German civil law countries. The geographic
controls included are latitude, landlockedness, mean elevation,
distance to the nearest coast or river, and precipitation. The
estimations exclude Barbados, Grenada, Fiji, Sao Tome and Principe,
and Seychelles.

* Significant at 10%; ** significant at 5%;
*** significant at 1%.

TABLE 7
Additional Environmental Policy Outcomes

                                (1)                 (2)
                               C3-I              Max. Lead
                                                  Content
Dep. Var.                                       (Inversed)

Stale                      1.514 (1.344)        0.22 (0.81)
Common Law LO              0.537 (1.046)      0.39 ** (2.57)
Mixed Law LO               0.249 (0.483)      0.39 *** (2.82)
State X Common Law LO   -3.346 *** (-2.665)   -0.56 * (-1.71)
State X Mixed Law LO      -1.174 (-0.840)       0.01 (0.02)
R-squared                      0.124               0.469
No. obs.                        74                  43
Geographic controls             Yes                 Yes

                               (3)                  (4)
                          Environmental        Environmental
                           Performance         Regulation in
Dep. Var.                     Index             Agriculture

Stale                      1.77 (0.41)          10.98 (0.21)
Common Law LO            8.46 *** (2.78)      69.73 *** (3.04)
Mixed Law LO               0.23 (0.03)        60.62 *** (3.86)
State X Common Law LO   -21.67 *** (-2.95)    -89.50 * (-1.75)
State X Mixed Law LO        7.16(0.57)       -151.59 ** (-3.15)
R-squared                     0.478                0.548
No. obs.                        52                   26
Geographic controls            Yes                  Yes

Notes: The alternative dependent variables are measured on
different scales and hence sizes of the estimates are not directly
comparable with earlier results or across table columns. Robust
standard errors are used and t statistics are reported in the
parentheses. The intercept estimate is not shown. The omitted legal
origin group is "Other legal origins," which includes 39 French and
3 German civil law countries. The geographic controls included are
latitude, landlockedness. mean elevation, distance to the nearest
coast or river, and precipitation.

* Significant at 10%; ** significant at 5%; *** significant at 1%.

TABLE 8
Labor Market Regulation Outcomes

                               (1)               (2)
                             Social             EPLEX
                            Security
Dep. Var.                     Laws

State                      0.18 (0.93)      0.18 ** (2.42)
Common Law LO              0.10(0.80)        0.01 (0.30)
Mixed Law LO               0.13 (1.12)       0.13 (1.59)
State X Common Law LO    -0.51 * (-1.95)   -0.41 ** (-2.71)
State X Mixed Law LO      -0.49 (-1.69)    -0.27 * (-1.80)
R-squared                     0.483             0.421
No. obs.                       37                 42
Geographic controls            Yes               Yes

                               (3)                (4)
                          Unemployment       Unemployment
                            Insurance      Benefits Coverage
Dep. Var.                Benefits Ratio

State                      0.10(1.44)         0.18 (1.13)
Common Law LO            0.12 ** (2.09)     0.42 ** (2.44)
Mixed Law LO               0.12 (1.64)        0.12 (1.48)
State X Common Law LO    -0.26 * (-1.94)   -0.86 ** (-2.41)
State X Mixed Law LO     -0.24 * (-1.79)     -0.12 (-0.60)
R-squared                     0.411              0.473
No. obs.                       38                 35
Geographic controls            Yes                Yes

Notes: The alternative dependent variables are measured on
different scales and hence sizes of the estimates are not directly
comparable with earlier results or across table columns. Robust
standard errors are used and t statistics are reported in the
parentheses. The intercept estimate is not shown. The omitted legal
origin group is "Other legal origins," which includes 39 French and
3 German civil law countries. The geographic controls included are
latitude, landlockedness, mean elevation, distance to the nearest
coast or river, and precipitation.

* Significant at 10%; ** significant at 5%; *** significant at 1%.