Guido Tabellini (IGIER,Bocconi University) April 2007

Slides:



Advertisements
Similar presentations
The Well-being of Nations
Advertisements

Empirical Analysis Doing and interpreting empirical work.
Chapter 12 Simple Regression
Econ 140 Lecture 131 Multiple Regression Models Lecture 13.
Multiple Regression Models
ECON 6012 Cost Benefit Analysis Memorial University of Newfoundland
Inductive Generalizations Induction is the basis for our commonsense beliefs about the world. In the most general sense, inductive reasoning, is that in.
The inference and accuracy We learned how to estimate the probability that the percentage of some subjects in the sample would be in a given interval by.
Chapter 10 Confidence Intervals for Proportions © 2010 Pearson Education 1.
Stats Methods at IC Lecture 3: Regression.
Chapter 9 Fundamentals of Hypothesis Testing: One-Sample Tests
Chapter 13 Simple Linear Regression
Chapter 9: Testing a Claim
Confidence Intervals for Proportions
Statistics 200 Lecture #9 Tuesday, September 20, 2016
Chapter 10: Comparing Two Populations or Groups
Chapter 14 Inference on the Least-Squares Regression Model and Multiple Regression.
Selecting the Best Measure for Your Study
Unit 5: Hypothesis Testing
Chapter 9: Testing a Claim
Chapter 9: Inferences Involving One Population
Chapter 9: Testing a Claim
Confidence Intervals for Proportions
Chapter 25 Comparing Counts.
Significance Tests: A Four-Step Process
Chapter 10: Comparing Two Populations or Groups
Findings of examined articles
12 Inferential Analysis.
Elementary Statistics
More on Specification and Data Issues
Hypothesis Tests for a Population Mean,
CHAPTER 26: Inference for Regression
Cross Sectional Designs
Chapter 11: Inference for Distributions of Categorical Data
Significance Tests: The Basics
Significance Tests: The Basics
15.1 Goodness-of-Fit Tests
12 Inferential Analysis.
Paired Samples and Blocks
Chapter 9: Testing a Claim
Chapter 26 Comparing Counts.
Chapter 7: The Normality Assumption and Inference with OLS
Chapter 13: Inference for Distributions of Categorical Data
Chapter 10: Comparing Two Populations or Groups
Chapter 9: Testing a Claim
Chapter 10: Comparing Two Populations or Groups
Confidence Intervals for Proportions
Chapter 9: Testing a Claim
Chapter 10: Comparing Two Populations or Groups
Chapter 26 Comparing Counts Copyright © 2009 Pearson Education, Inc.
Chapter 10: Comparing Two Populations or Groups
Chapter 9: Testing a Claim
Chapter 10: Comparing Two Populations or Groups
Chapter 10: Comparing Two Populations or Groups
Chapter 10: Comparing Two Populations or Groups
Chapter 10: Comparing Two Populations or Groups
Chapter 9: Testing a Claim
Chapter 9: Testing a Claim
Chapter 9: Testing a Claim
Chapter 26 Comparing Counts.
Chapter 10: Comparing Two Populations or Groups
Confidence Intervals for Proportions
Chapter 9: Testing a Claim
Chapter 10: Comparing Two Populations or Groups
Unit 5: Hypothesis Testing
Chapter 9: Testing a Claim
Chapter 10: Comparing Two Populations or Groups
Significance Tests: The Basics
F test for Lack of Fit The lack of fit test..
Presentation transcript:

Guido Tabellini (IGIER,Bocconi University) April 2007 Culture and institutions: economic development in the regions of Europe Guido Tabellini (IGIER,Bocconi University) April 2007

Abstract Does culture have a causal effect on economic development? Culture is measured by indictors of individual values and beliefs, such as trust and respect with others and confidence in individual self determination Two historical measures: literacy rate at the end of 19th century and political institutions in place over past several centuries

Introduction Since the seminal work of economic historians like North(1981), it has become almost commonplace to view history as main determinant of current economic development A widespread interpretation is that history shapes current economic performance through “institutions” The fact that same institutions function so differently in different environments suggests that informal institutions play an important role

Culture Like institutions, it is a vague concept Specific indicators of culture, that can be interpreted either as social norms or as individual values, are correlated both with historical patterns and with current economic development

Introduction Previous research schematically shows: Historical institutions => Contemporary institutions => Economic Development This paper instead uses within country variation at the regional level to explore: Historical Institutions => Culture => Economic Development

How does Tabellini measure culture? By aggregating at the regional level individual responses collected in opinion polls of World Value Surveys in the 1990s Specific indicators of values and beliefs: trust, respect for others, confidence between individual effort and success After controlling for country fixed effects, regional education and urbanization rates, it’s related with two historical variables: regional literacy rates and indicators of political institutions in the period from 1600 to 1850

Related Literature Greif (1994) – emphasizes the interaction between culture and institutions, he points out how the different cultures of Maghribi and Genoese traders in the late medieval period led them to develop different institutions Glaeser et al. (2004) – effect of history on current economic development reflects the accumulation of human capital which influences institutional outcomes

Related Literature More related to this study are: Barro and McCleary(2003) provide evidence that religious beliefs are correlated with economic growth in a sample of countries Guiso, Sapienza and Zingales(2004) study the effect of social capital on individual financial habits Spolaore and Wacziarg(2005) find that income differences between countries are positively correlated with genetic and geographic distance, and interpret this evidence as suggestive of cultural barriers to the diffusion of innovations across countries

Data on Output, Education, Urbanization and Culture Sample consists of 69 regions in 8 European countries: France, Germany, UK, Italy, Netherlands, Belgium, Spain and Portugal Per Capita Output: Current economic development is measured by per capita gross value added (GVA) in international prices, focused on after 1990 period where culture measure is available. Variable, yp9500 is dependent variable in analysis

Education The goal is the study the direct link between culture and economic development, so Tabellini wants to avoid using culture just as a proxy for human capital in the region. Thus, always controls for regional differences in education of adult population measured by enrolment in primary and secondary schools in percent of the population of relevant age group This variable is called school

Urbanization in 1850 As a proxy for regional economic development in previous centuries, past urbanization rates are used To measure it variable urb_rate1850 is constructed, defined as fraction of regional population that lived in cities with more than 30 000 individuals around 1850

Culture World Value Surveys are designed to measure a variety of cultural traits. Four cultural traits: 1) mutual trust and 2) respect for others. These encourage welfare enhancing social interactions. 3 and 4) confidence in virtues of individualism. Symptomatic of an entrepreneurial environment where individuals seek to take advantage of economic opportunities

Trust Considered the following question in the survey: “ Generally speaking, would you say that most people can be trusted or that you can’t be too careful in dealing with people?” The level of trust in each region is measured by the percentage of respondents who answer that “Most people can be trusted” (the other possible answers are “Can’t be too careful” and “Don’t know”). This variable is called trust

Respect and Obedience “Here is a list of qualities that children can be encouraged to learn at home. Which, if any, do you consider to be especially important? Please choose up to five”. The variable respect is defined as the percentage of respondents in each region that has mentioned the quality “tolerance and respect for other people” as being important Lack of trust and lack of respect for others are typical of hierarchical societies, where the individual is regarded as responding to instinct rather than reason, and where instinct often leads to a myopic or harmful course of action The variable obedience is defined as the percentage of respondents that mention “obedience” as an important quality that children should be encouraged to learn

Control To measure this cultural trait we construct a variable, called control, from the following question in the survey: “ Some people feel they have completely free choice and control over their lives, while other people feel that what we do has no real effect on what happens to them. Please use this scale (from 1 to 10) where 1 means “none at all” and 10 means “a great deal” to indicate how much freedom of choice and control in life you have over the way your life turns out”. The variable control is defined as the unconditional average response in each region (multiplied by 10)

pc_culture Thus have four related but distinct measures of culture: three indicators expected to promote economic development (trust, control, respect), and one that might hurt it (obedience) pc_culture, is a summary measure of these cultural beliefs Since this principal component is negatively correlated with obedience, while it is positively correlated with trust, control and respect, we take it to be a net measure of the aspects of regional culture that favour economic development

Other summary measures of culture To facilitate the interpretation, we have also extracted the first principal component from the positive beliefs only (trust, control and respect), called pc_culture_pos, as well as the first principal component from the two questions on the desirable qualities of children (obedience and respect), called pc_children. Since this variable is positively correlated with respect and negatively correlated with obedience, it is once more a net measure of the aspects of norms that favour economic development Finally, since the principal component only captures the variation that is common to all beliefs, while these norms could have more than one relevant dimension of variation, I have also computed an alternative summary measure called sum_culture, defined as the sum of the three positive beliefs (trust, control, respect) minus the negative belief (obedience)

Output and Culture Some of the correlation between per capita output and culture apparent from Figures 1 and 2 can simply reflect the influence of other common determinants, such as education, historical levels of economic development or national institutions To remove the effect of these other variables, we have regressed per capita output (yp9500) on a set of dummy variables (one per country), school enrolment in 1960 (school), urbanization rates in 1850 (urb_rate1850) and the various measures of culture

Output and Culture Naturally, we cannot safely assume that culture is independent of current levels of economic development. On the contrary, all our variables measuring culture are likely to be influenced by the current economic situation. Controlling for current education in each region (the variable school) and for past economic development as measured by past urbanization rates (the variable urb_rate1850) removes some of this correlation Nevertheless, reverse causation remains a fundamental concern. Hence, the estimated coefficients reported in Table 3 could be biased and cannot be interpreted as reflecting a causal effect of culture on output. To cope with this problem, in the remainder of the paper we rely on instrumental variable estimation, using other historical variables as instruments for culture

Indicators of Culture How do these cultural indicators relate to the measures of institutions widely used in existing cross-country analysis? And do they explain cross country differences in per- capita income? A recent wave of the World Value surveys, conducted in 1999-2000 and covering a larger sample 15 of countries, is made available. From this third wave, Tabellini constructed the same indicators of culture at the country (rather than regional) level, for almost 50 countries

Indicators of Culture Even without suggesting a causal interpretation, these regressions are nevertheless remarkable. They show that these cultural indicators are meaningful, and highly correlated with variables that have attracted so much interest in the recent analysis of cross country differences in economic development But separating the effect of culture from that of institutions is more credibly done in the sample of European regions, where one can control for common political and economic institutions at the national level, and where unobserved heterogeneity is less problematic. This will be the focus of remainder

Identification of the model Y = regional per capita output C = culture Yo = urbanization in 1850 X = school enrollment in 1960 and institutions e = error term Problem: C and e are likely to be correlated!

Possible solution: instruments Cov(z,x) ≠ 0 Cov(z,e) = 0 Cov(z,y) = 0

Culture can be viewed as shaped also by cultural features of earlier generations, denoted by Co Co is reasonably excluded from (1), but unfortunately we don’t observe it Yet, we can apply the same logic to Co , which refers to past social interactions and hence is influenced by historical features

λi = parameters Xo = literacy rate around 1880 and constraints on the executives in 1600-1850 v = error term The instruments we use isolate the variation in culture that is exogenous, from the possibly endogenous variation in culture due to v. Now δ only exploits this exogenous variation in culture

The issue has now been shifted away from whether culture is endogenous to whether our historical variables are valid instruments Our estimation strategy rests on two premises History shapes culture Literacy and institutions are assumed as valid (rather strong assumption)

IDENTIFICATION of ESTIMATION and HISTORICAL DATA Instruments are very important for understanding the economic models.They help us to learn the current economic situation of a country. Literacy and institution are most importants according to Tabellini

Literacy In 1880 To capture regional differences in educational histories, Tabellini collected data on the literacy rate around 1880 by region Literacy changes from region to region.Except Netherlands and Portugal because Tabellini only found national data of these two countries

Early Political Institutions If we look to institutions of countries, we can see differences because of “Constraints of Execuive”. Constraints of Executive means democracy. Without it a country will be dictatorship In table 6 we see data about years 1600s, 1700s, 1750, 1800 and 1850

The Effect of Culture on Output

HOW CREDIBLE IS THE IDENTIFICATION? Past literacy rates could be linked with current per capita output, if for instance regions mostly illiterate specialised in agricolture The over-identified model with the two historical variables and the orthogonality test could not be sufficient Instruments can have a direct effect on current development through omitted variables (like capital stock)

Controlling for the sectoral composition of employment To rule out the suspition that literacy rates affect current output we add another variable, the employment share in agricolture in 1977, as an additional control. agr_share is treated as exogenous and is negatively correlated with pc_culture As shown in the table, the estimated coefficient of agr_share is significantly different from zero in the output regression, but not in the equation for culture This give strenght to the validity of past literacy rates as an instrument for culture

Table 13 or link to it

Just-identified models As a further check of the orthogonality test on the validity of the two historical variables, we add them to the second stage regressions one at time, then in a just-identified model If the instruments are valid, the estimated coefficients ought to be close to zero The results are displayed in the table and confirm Sargan over-identified test Yet, we notice also that the coefficients of pc_culture changes across the two specifications

Table 13 again, circles on the results

Controlling for the capital stock Capital stock could be an omitted variable with a direct effect on current development We use the capital stock data of some regions of Italy in 1979 As a result, culture retains a positive and significant estimated coefficient on the second stage, while initial capital, even if positive, is not statistically significant in the output regression We can conclude from the Italian sample that capital stock probably doesn’t affect current output

Orthogonality test With the Sargan statistic we failed to reject the null hypothesis and demonstrated with good approximation that our instruments are valid The next step is a further check to examine three of the possible errors the Sargan statistic could not find: Bootstrapped statistic Montecarlo simulation Different model

#1 The failure to reject the null hypotesis of exogenous instruments could be not so robust if due to specific features of the sample In order to test this, we bootstrap the Sargan statistic randomly replacing one observation from the sample with a random draw from a similar sample. Then we replicate the instrumental variables estimates 1000 times From the figure we can see that only in about 70% of the replications the statistic does not reach the threshold of 3.84 Therefore, the rejection of the over-identifying restrictions may not be very robust

Figure 7

#2 To check the power of the Sargan statistic and control if both the instruments are actually valid we run a Montecarlo simulation setting the coefficient of culture on output first at 0.21 and then at 0.86 We consider four different cases: both historical variables are valid instruments, only one of them a time, or none The next step is to implement the Sargan statistic to detect if one of the two instruments is not valid The test turns to be quite reliable more if we consider literacy and not pc_institutions as valid and also when the bias in the instrumental variable estimates is large

Table 14

#3 Sargan test is not able to detect if the assumptions at the base of the model are wrong, or more precisely Hypotethical model: Historical institutions => Culture => Economic development True model: Historical institutions => Economic development => Culture In addition, income and culture are measured at the same point in time. Because of there are no available measures of culture for earlier time periods at a European level and much of the relevant variation come from Italian regions, we take into consideration a particular example occured in Italy

In 1946, Italy held a popular referendum in favour or against the monarchy. It’s easy to say that a vote in favour of the monarchy was likely to reflect backward cultural values and mistrust for democratic institutions The percentage of regional votes in favour of the monarchy is taken as a measure of cultural backwardness in the immediate post-war period

Data in the table show that votes are strongly correlated both with literacy and institutions and with contemporaneous measures of culture When entered separately, both historical variables have a negative and significant estimated coefficient and the same negative relationship stands between pro_monarchy and pc_culture Despite the small number of observations, we can affirm the persistence in cultural traits, due to the worse nowadays cultural attitudes of the regions that voted in favour of the monarchy. The evidence supports the view that history has an effect on culture and this is possibly independet from development

Table 15

Conclusion If past political institutions and past literacy rates are valid instruments for culture in the output regression, this historical variables are correlated with some specific measures of regional culture: trust and respect for others and confidence in individual self-determination Distant history appears to be an important determinant of current economic performance. Thus, early historical institutions have shaped current institutions On the other side, formal institutions seem in same cases less important than informal. In fact we still find an economic legacy of early institutions in regions that have been ruled by the same formal institutions for centuries

MAIN FINDINGS: Early political institutions emerge as a significant source of current economic performance, also in regional comparisons The component of culture explained by the historical variables is an important determinant of regional economic performance

CHALLENGES: It’s not possible to reject that culture entirely explains the economic legacy of history Both culture and formal institutions are likely to shape the functioning of institutions and the behavior of agents, but culture is still largely a “black box” The low labor productivity of economically backward regions won’t go away soon, indeed it is connected with the persistence of culture and institutions. This means income transfers and public investment are not a solution, while governments should take into consideration long-term investments like education and lower real wages in regions with lower productivity

Thank you for your attention 