Author Archives: sebastianfleitas


Gibrat’s Law and the British Industrial Revolution

Alexander Klein (University of Kent) and Tim Leunig (LSE)


Abstract This paper examines Gibrat’s law in England and Wales between 1801 and 1911 using a unique data set covering the entire settlement size distribution. We find that Gibrat’s law broadly holds even in the face of population doubling every fifty years, an industrial and transport revolution, and the absence of zoning laws to constrain growth. The result is strongest for the later period, and in counties most affected by the industrial revolution. The exception were villages in areas bypassed by the industrial revolution. We argue that agglomeration externalities balanced urban disamenities such as commuting costs and poor living conditions to ensure steady growth of many places, rather than exceptional growth of few.

Reviewed by Sebastian Fleitas

On August 24th a couple of philosophers of science asked themselves: ¨What is Economics good for?¨ and they provided a provocative answer in the Opinionator blog of the New York Times. The main argument of the philosophers was that:

¨..the fact that the discipline of Economics hasn’t helped us improve our predictive abilities suggests it is still far from being a science, and may never be.¨

Although tempted, I will not enter the debate here but will discuss something related. Suppose you are asked to predict the population of English cities 100 years from now. Do you think London, Birmingham, Manchester and Liverpool will still be the most populated ones? Will the medium size cities of today grow as much as the smaller cities? I suppose that you will be inclined to think that the size that the city already has will be correlated with its future growth and then it is useful for prediction. Why? I imagine that you could think that the causes that have generated the growth of the cities for the last 50 or 100 years will be in one way or the other pushing their growth in the next 100 years. Alexander Klein and Tim Leunig show that this intuition is not correct, at least for England and Wales during the nineteenth and the first decade of the twentieth century.

Prediction: The image in the NYT Blog

Prediction: The image in the NYT Blog

Klein and Leunig’s work is part of a growing scholarship who has given greater attention to Gilbrat’s Law: an empirical regularity that postulates that the growth rate of places is identical for places of all initial sizes. This paper, circulated in NEP-HIS 2013-08-16, tests Gibrat’s law for the period of the most important event in the recent economic history: the British Industrial Revolution.

The Industrial Revolution seems to be such an important ¨shock¨ as to change completely the process behind the growth of the cities in the past. As the authors point out, the British Industrial Revolution implies four major interrelated changes: population grew at an unprecedented rate, both total and per capita income rose, England ceased to be a largely agrarian nation in which people’s locations were tied to the land, and the nineteenth century witnessed a transport revolution covering almost every form of transport and the energy used for transportation. Even more, this period is even more relevant given that it encompasses stage without planning nor zoning rules, so that urban growth was the result of many interacting forces.

The authors find that Gibrat’s law broadly holds, although small villages in areas bypassed by the industrial revolution tended to violate it. Moreover, although the results are similar during the entire nineteenth and the beginning of the twentieth century, it is possible to find some qualitative differences. In the nineteenth century they find a more rapid growth of large towns and cities than small ones. In contrast, at the beginning of the twentieth century in the areas that had been affected by the industrial revolution there are no differences in the growth rates of places of different sizes, because counties were at that time mature. Based on the models of Eeckhout (2004) and Córdoba (2008), the paper also offers an explanation for why Gibrat’s law largely holds even during the rapidly changing environment of the British Industrial Revolution because of the balance between positive agglomeration externalities and negative externalities of high commuting costs and large disamenities of urban life.

Ashton, T. S.. The industrial revolution, 1760-1830. London: Oxford University Press, 1948.

Ashton, T. S.. The industrial revolution, 1760-1830. London: Oxford University Press, 1948.

Two methodological factors in this paper are worth noting. First, the authors have set a unique, authoritative and comprehensive data set of all cities, towns and villages in England and Wales in the periods 1801-11, 1841-51, and 1901-11. They take advantage of the fact that Great Britain undertook the first census in 1801, and has had decennial censuses ever since. A very important factor for the test is that the descriptive statistics show that the growth in population was led primarily by the growth in the size of existing places, rather than the emergence of new places. The fact that England and Wales were populated in 1801 implied that there was no frontier to open up and that, even when some new settlements were developed (for instance along railway lines), population growth took place overwhelmingly in places that already existed. Therefore, the key urban issue in the period then is to understand what caused the growth of some cities instead of what originated new cities.

The second methodological issue is that they use a non-parametric regression analysis to test Gilbrat’s Law. In a first approach, they approximate Gilbrat’s Law using the unconditional relationship between the standardized growth rate (defined as the difference between the growth rate of the place and the sample mean divided by the sample standard deviation) and the log of population size of the place at the start of the period. The non-parametric approach (generally used in this literature) is a key factor here. Since it allows for more flexibility in the functional form of the relationship between growth and initial size, the authors are capable of showing patters of the unconditional relationship between these two variables and can test whether Gilbrat’s Law holds for the whole initial size distribution. In a second approach, the authors test another characterization of Gibrat´s Law that makes the relationship between the variance of the growth rate and the initial size. This second approach is very useful since even when the initial size could not be useful to predict the growth of the cities it could be a predictor of the variance of the growth rate.


Birmingham 1732 (taken from Wikipedia)

At this point, I would like to say that I think this research line would advance significantly if the authors considered the inclusion of other factors to explore the relation between growth and size conditional on these factors. As they establish, the Industrial Revolution completely changed many factors that can be influencing the growth rates of the cities. Just as an example, imagine what could happen with two cities that are identical in all aspects except that one is close to a coal mine. It is likely that the city close to a coal mine will grow more, just because the transport costs of this new energy source are lower. Obviously these factors included in the agglomeration externalities that balance with the commuting cost or disamenities of urban life (negative externalities). But including those factors in a regression between growth and size would provide a more reliably consistent estimation of the (partial) effects of size and the other factors on city growth. In this sense, a multivariate regression of the growth rates over size and other factors would have a more clear the interpretation (and measure) of the economies (and diseconomies) of agglomeration that are suggested and discussed in this paper and that drove the growth of the cities during the period.

To sum up, I think the paper made a very important contribution robustly showing that there is no unconditional relationship between growth and initial size (besides some small villages bypassed for the Industrial Revolution). The authors provide useful explanations of why this could have happen along of the lines of the models in Eeckhourt (2004) and Cordoba (2008). In short, the authors show that the British Industrial revolution was such an important ¨shock¨ that changed completely the process of growth of the cities, changing the balances between economies and diseconomies of agglomeration. With this in hand and with the incredible database they have constructed, they have also opened the possibility of formally testing the conditional (partial) effects of size and other factors on city growth. This would help to understand the dynamics behind city growth during this key period of time. Furthermore, while the (lack of) unconditional relation between city size and growth rate leaves us ill-equipped for predicting future city growth, the identification of the conditional partial effect of size on growth could improve our “predictive abilities”.

Fiscal Policy during high unemployment periods: still a bad idea?

Are Government Spending Multipliers Greater During Periods of Slack? Evidence from 20th Century Historical Data

Michael T. Owyang, Valerie A. Ramey, Sarah Zubairy


A key question that has arisen during recent debates is whether government spending multipliers are larger during times when resources are idle. This paper seeks to shed light on this question by analyzing new quarterly historical data covering multiple large wars and depressions in the U.S. and Canada. Using an extension of Ramey’s (2011) military news series and Jordà’s (2005) method for estimating impulse responses, we find no evidence that multipliers are greater during periods of high unemployment in the U.S. In every case, the estimated multipliers are below unity. We do find some evidence of higher multipliers during periods of slack in Canada, with some multipliers above unity.


Review by Sebastian Fleitas

For a very long time the size of the expenditure multipliers has been one of the most vivid economic debates. For instance as recently as 2009, when the Obama administration proposed a fiscal stimulus package, there was a heated discussion regarding the relative size of the expenditure and tax multipliers. The reason fuelling this narrative is perhaps clear: ascertaining the potential impact of a particular proposed measure is key when designing the fiscal policy.

The paper by Owyang, Ramey and Zubairy, which was distributed by NEP-HIS on 2013-02-08 tries to answer this question: Are government spending multipliers greater during periods of slack for the US and Canada when we look at the historical data? The argument behind it is to consider that the expenditure multipliers will be greater in times of crisis, that is, during periods without full employment of labor and capital in the economy. This argument follows the idea that to wake up animal instincts, you need to have something in the forest when guys go out to hunt.

Image in Barro's comment on expenditure multipliers debate on 2009 in Stanford blog.

Image in Barro’s comment on expenditure multipliers debate in 2009 in Hoover Institution Stanford University’s blog (

The answer that the authors offer is counterintuitive, which makes the paper very interesting. They find that the expenditure multipliers were higher in periods with high unemployment in Canada but they were the same for both periods in the US. To arrive to this conclusion the authors first construct high frequency (quarterly) historical data for the US and Canada. The procedure they follow to build the database is documented in an online available annex of the paper (here). After this process they have data on GPD, GDP deflator, government spending and the unemployment rate for the period 1890q1 to 2010q4 for the US and from 1921q1 to 2011q4 for Canada. The other key variable is the “news” variable, which reflects the changes in expected present value of government spending in response to military events as in Ramey (2011), which in turns directs to Ramey (2009).


Regarding the econometric approach, the authors use Jorda’s (2005) local projection technique to calculate impulse responses. The idea in Jorda (2005) is that, in contrast to VAR approaches which  linearly approximate the data generating process to produce optimal one period forecasts, when we are looking at impulse response analysis we should care about the estimation of longer horizons. In this context, it is a better approach to estimate the impulse responses consistently by a sequence of projections of the endogenous variables shifted forward in time onto their lags using ordinary least squares (OLS) with standard errors addressing heterogeneity and serial correlation. The authors estimate a set of OLS regressions of different number of leads of the log of per capita government expenditure and GDP, over their lags and the variable news for periods with high and low unemployment and a quadratic trend. The coefficient for the variable “news” is the impulse response at that certain number of lags.

Finally, the paper made me think of three comments. First of all, the paper shows a very interesting and creative way to proceed when the data needed for the study is actually not available for that historical period. Besides combining sources of information, the authors constructed quarterly series of the variables. Since the paper was prepared for the American Economic Review Paper and Proceedings, it is a very short paper and the procedure to construct the variables is explained not in the paper but in the Annex. Given the lack of data, assumptions about the data generating process must be made. However, and besides the obvious limitation of space, the reader could miss an explanation about the assumptions that are made in the methods used and, also, what implications these assumptions have for the results, in particular about what is the source of variation that allows the identification of the coefficients. Maybe a section in the paper or in the appendix discussing these issues can shed light about what are the potential problems of different assumptions.

The last two comments are related to what is exactly the interpretation of the results. The first one directly follows from the last sentence of the paper. The authors state that they do not adjust for the fact that taxes often rise at the same time as government spending, which turns these multipliers not equal to pure deficit financed multipliers. However, it seems plausible that the effect of the multiplier on the GPD depends on whether this increase in the government was financed by taxes or by debt. If that is the case, and if the episodes when the former and the latter happen are mixed in a non-random way between the periods of high and low unemployment, then it is possible that the value of the coefficients can reflect not only the effect of the exogenous shock but also the effect of different ways to finance it.

A joke?

A joke?

The last comment relates to the consistent estimation of the parameters of the model. In the paper the “news” about military expenditure is taken as the only source of exogenous shock in this economy during the period of two years, four years and the time of the peak of each response. This “news” variable reflects exogenous innovations to the expenditure from a military source. However, it would be relevant for the paper to discuss the existence of other (non-military) sources of exogenous shocks to the expenditure. The relevance of this issue is because, given that the estimation of the parameters of interest is done by OLS, the consistency of the estimates requires zero covariance between the ¨news” and the error term of the equation, and this assumption can be violated if there exist this kind of non-military shocks and they are correlated to military “news”.

Overall I think this is a very interesting paper because of the results they find and also because of the construction of historical data. I found the results very puzzling and therefore a big motivation to continue trying to understand the relationship between GDP and public expenditure.

Money for Nothing? Banking Failure and Public Funds in Michigan in the early 1930s

The Effects of Reconstruction Finance Corporation Assistance of Michigan’s Bank’s Survival in the 1930s

Charles W. Calomiris (, Joseph R. Mason ( ), Marc Weidenmier (, Katherine Bobroff (



This paper examines the effects of the Reconstruction Finance Corporation’s (RFC) loan and preferred stock programs on bank failure rates in Michigan during the period 1932-1934, which includes the important Michigan banking crisis of early 1933 and its aftermath. Using a new database on Michigan banks, we employ probit and survival duration analysis to examine the effectiveness of the RFC’s loan program (the policy tool employed before March 1933) and the RFC’s preferred stock purchases (the policy tool employed after March 1933) on bank failure rates. Our estimates treat the receipt of RFC assistance as an endogenous variable. We are able to identify apparently valid and powerful instruments (predictors of RFC assistance that are not directly related to failure risk) for analyzing the effects of RFC assistance on bank survival. We find that the loan program had no statistically significant effect on the failure rates of banks during the crisis; point estimates are sometimes positive, sometimes negative, and never estimated precisely. This finding is consistent with the view that the effectiveness of debt assistance was undermined by some combination of increasing the indebtedness of financial institutions and subordinating bank depositors. We find that RFC’s purchases of preferred stock – which did not increase indebtedness or subordinate depositors – increased the chances that a bank would survive the financial crisis. We also perform a parallel analysis of the effects of RFC preferred stock assistance on the loan supply of surviving banks. We find that RFC assistance not only contributed to loan supply by reducing failure risk; conditional on bank survival, RFC assistance is associated with significantly higher lending by recipient banks from 1931 to 1935.

Review by Sebastian Fleitas

The systemic risk of bank failures, and its macroeconomic consequences, led the Fed to take action when some banks started to fail in 2008. How much money did the Fed give to the banks in 2008? And even more important, was this money helpful to avoid banking failures? The latter question seems to be a key question every time that the government is implementing a program to try to stem bank failures and to reduce the economic cost of financial disintermediation.

Detroit skyline, circa 1930

The paper by Calomiris, Mason, Weidenmier and Bobroff, distributed by NEP-HIS on October 6th,2012, assess the success of a public support program aimed at banks in financial distress. This through assistance provided by the Reconstruction Finance Corporation (RFC), ,a government-sponsored enterprise, to Michigan’s banks in the  1930’s.

Calomiris and friends offer a very interesting description of the timing of the crisis and a regression analysis of the impact of the RFC assistance. The period of analysis, from January 1932 through December 1934, covers two sub-periods: the first in which bank failures occurred sporadically; and a second sub-period in which the failures were concentrated and coincided with regional and national panics.

The banking crisis of 1933 in Michigan is situated in the middle of the period of analysis. This is a very important episode as it can be seen as a prelude to the national banking disaster as well as the Michigan hosting the automobile industry, an industry on the raise and of future importance for the national economy.

The role of the RFC changed between the two sub-periods. During the first period, the RFC main action was to help banks advance money on loan. The risk involved in these loans was mitigated through their short duration, strict collateralization rules and high interest rates. Although these rules protected the RFC from losses, they also limited the effectiveness of the RFC lending policy. However, on March 9, 1933 the Congress passed an act altering the original mandate, allowing the RFC to purchase preferred stock in some financial institutions that were considered as likely to survive. This opened the possibility for the RFC assistance to be more effective in the second sub-period than in the first one.

1940 Reconstruction Finance Corporation RFC Cartoon

Econometric estimates then try to identify the effect of RFC assistance. Specifically whether in light of an increasing rate of bank failures, the federal government had decided offer support to banks with greater risk of failure. In this sense, the dummy variable of RFC assistance is an endogenous variable, and this problem has to be addressed in order to consistently estimate the effect. To deal with this problem, the authors use two different estimation techniques and they use two sets of instruments. First, they use a set of instruments that indicate the correspondent relationships of each bank, that indicate the extent to which the bank was important within the national network of banking and also the correspondent relationships with Chicago and New York. Second, they included county specific characteristics that might have affected RFC assistance without affecting bank failure risk.

The authors conclude that the loans from the RFC did mitigate the risk of bank failure  but rather, that recapitalization (in the form of the purchase of preferred stocks) increased the likelihood of bank survival. Reasons why preferred stocks assistance was more effective included: a) because unlike loans, it neither increases the debt of the bank nor the liquidity risk or collateral requirements, b) the RFC was selective when choosing who was included in the program, and c) the RFC was able to prevent abuse from assisted banks. In general, they conclude that these results suggest that during a banking crisis, effective assistance requires that the government takes a significant part of the risk of the bank failure.

January 1931, Chester Garde

Emprical estimates in this paper concur with previous results in the literature. But by incorporating Michigan this papers offered added granularity and also improves in the use of econometric techniques used to address the issue of the effect of the RFC in banks failure. However, I think the paper could be improved by a more thorough discussion of the instruments used, in terms of why they can be assumed to be related to the RFC assistance and not directly related to bank failure. This is especially important because the results of the first stage estimations cast some doubt about the suitability of some of the instruments selected. Regarding the first set of instruments, one variable indicates the connections of a bank within the national network of banking and another one the relationships with Chicago and New York. However, in the first stage the effect of these two variables over the RFC assistant have different signs and their statistical significance depend on the period and specification of the model. A second concern is that they use the variable “Net due to banks over total assets” but this instrument is not significant in any first stage estimate. Banks with more creditors or debtors could be more important to save, but it could also be the case that these banks are more indebted with other banks because they are facing problems and thus they have more risk of failure. Regarding the second set of instruments, these variables generally fail to be consistently significant and the mechanisms through which they affect the decisions of the RFC without affecting the hazard of failure are not completely clear. Was the main proportion of the business of the banks concentrated at the county level at those times? Does the political importance of the county matter to allocate the assistance, even when the authors say that the manipulation of the RFC by Congress or the Administration was mitigated? Is the unemployment rate in the county in 1930 unrelated with the risk of failure of the banks during the crisis? A more deep consideration of these issues could help to understand why these variables are good instruments and why the results of the first stage estimations look like they do.

To sum up, this paper provides new evidence about the role of the RFC during the important period of 1932-1934. Furthermore, this paper addresses an issue that is relevant today: the efficiency of public funds to avoid bank failures. The general conclusion the authors achieve is that an effective assistance involves that the government assumes a significant share of the risk of bank failure. As in the thirties, in the present the government has spent lots of money trying to avoid the systemic risks related with the failures of some banks. This and other related papers in the literature can help us to understand the effects of a banking crisis in the real sector and the efficiency of public policies that try to reduce its negative impacts. This particular historical experience can not only shed light about what happened in that opportunity but also give us insights to approach these situations when they appear again, in particular to design better economic policies.

Breaking News: The old rapid peripheral industrial growth

The spread of manufacturing to the periphery 1870-2007: eight stylized facts.

by Agustín Bénétrix (, Kevin O’Rourke ( and Jeffrey Williamson (


Abstract: This paper documents industrial output growth around the poor periphery (Latin America, the European periphery, the Middle East and North Africa, Asia, and sub-Saharan Africa] between 1870 and 2007. We provide answers to the following questions: When and where did rapid industrial growth begin in the periphery? When and where did peripheral growth rates exceed those in the industrial core? When was the high-point of peripheral industrial growth? When and where did it become widespread? When was the high-point of peripheral convergence on the core? How variable was the growth experience between countries? And how persistent was peripheral industrial growth?

Review by Sebastian Fleitas

This paper was distributed by NEP-HIS on 2012-07-29, and starts with a fabulous quote namely:

“To a large extent, world economic history since 1800 has been the history of how the international economic system adjusted to the dramatic asymmetric shock that was the Industrial Revolution”.

Besides catching the reader’s attention, this powerful statement points to the globalisation of manufacturing as one of the most important economic events of the 19th and 20th centuries.

Authors document industrial output growth between 1870 and 2007 for core (“developed”) countries and for the periphery (Latin America, the European Periphery, The Middle East and North Africa, Asia and the Sub-Saharan Africa). Different phases of trade policy are highlighted: the globalization of the late 19th Century, the interwar period; the period after WWII, and the 1980s onwards.

Men working their own coal mines. Early 1900s, USA. Taken of Wikipedia.

Stylised facts emerging from their results provide insights into the process of globalization of manufacturing (from the core to poor countries in the periphery), these include,

  • Rapid peripheral industrial growth began relatively early in all peripheral regions, and
  • Peripheral industrial growth rates were uniformly higher than those in the three original industrial leaders between 1920 and 1989.
  • Rapid industrial growth was not a consequence of a small number of countries in each region but a response to the spreading of this growth to more countries.
  • The less industrialized countries saw statistically higher industrial growth rates between 1920 and 1989, with this convergence being higher during the ISI period.
  • Although some countries show a very persistent industrial growth, rapid long run industrial growth was not particularly persistent in the 20th century.
  • They find that industrial growth rates were more volatile in the periphery than in the core for every year except for the interwar period. In both regions, core and periphery, the volatility of industrial growth has reduced over time.
  • This paper makes several contributions first and foremost is the database itself. The possibility to reconstruct industrial output series for 134 countries for the period 1990-2007 and 23 for the period 1870-1889 is a remarkable effort. As they went further back in time, the authors had to rely increasingly on individual country sources and they acknowledge “many generous colleagues” for sharing the data with them. In this sense, setting up this kind of databases is possible thanks to many colleagues that are working to reconstruct different databases all over the world. A second contribution is that this paper documents and clearly shows us that the rapid peripheral industrialization had different paces and starting points for regions and countries. Moreover, this allows us to see that this is not a phenomenon unique to the recent past. These conclusions are important to understand, for example, that the process of the BRIC´s countries (Brazil, Russia, India and China) is more a historical trend than a completely novelty. Finally, this paper contributes to show that the relationship between openness and industrialization is not straightforward. In this sense the authors state that it may be contingent on other factors, just as appears to be true of the relationship between openness and growth, more generally.

    BRIC countries: Brazil, Russia, India and China

    Besides these contributions, there are also some comments to make at this point that I think are good to better understand the scope of the statements in the paper. To begin with, the paper discusses the evolution of the industrial output but not the industrial output per capita. Although the paper makes a very good job describing the spread of the industrial production over the world, the implications that this spreading has on the development process remain unclear. The relationship between economic growth and industrialization is given by several channels like structural change (labor shares in production) and productivity in industry (output per capita), to mention two of the most important. In this paper, the authors do not discuss labor shares in industry and the output per capita. Moreover, the output per capita is only used as the variable of convergence: they show regressions of manufacturing output growth rates against initial levels of per capita output. Then, what the regressions show is that less industrialized countries saw statistically higher industrial growth rates. However, this fact does not imply, for example, that these countries have been closed the gap of industrial product per capita since, for example, the population has grown much more in the developing countries than in the developed ones. In this sense, it seems that more research is needed to understand what are the process and determinants underlying the convergence and divergence of industrial output per capita and how they influence the process of economic growth.

    A second comment is that the paper does not make a difference among different process of industrialization in terms of the goods they produce for domestic consumption and their patterns of international trade. On the one hand, the specialization in different sectors matters because it has been stated that different sectors have uneven capabilities to boost economic growth. On the other hand, the big changes in international trade policies make the research about the sectorial composition of the industrial output even more important. During this period, international trade has changed from being mostly inter-sectorial trade based on comparative advantage to intra-sectorial trade based on other kinds of dynamic technological advantages. In this sense, the processes that can boost industrialization during some context are not the same processes that generate it in other contexts. Moreover, the composition of the industrial output is very significant during the period of the ISI when some peripheral economies were almost closed to international commerce. There is a possibility that the rapid economic growth of the industry during this period in some peripheral countries took place in industries heavily protected and where (even after the learning period) international commerce would turn impossible. In that sense, it is not straightforward what this rapid industrial growth represents in terms of economic development and more research would determine if it was good news for those economies or merely inefficient replacement of trade.

    To sum up, the process of industrial development over the world in this almost 140 years has yielded very different results. For example, some Asian countries have achieved a remarkable industrialization while Latin American countries have been diverging in productivity terms against United States in the long run. The common characteristics and the differences between all these paths established a very good scenario to understand, for example, the determinants of structural change or the relationship between trade and industrialization. Among others, these questions contribute to understand the process of economic development in the long run. This paper is a contribution and an invitation to carry on this task.

    Political Power and Regional Economic Performance

    Who is the boss here? Regional power and participation of the Caribbean coast in the ministerial cabinets, 1900-2000.
    (Original title: ¿Quién manda aquí? Poder regional y participación de la costa Caribe en los gabinetes ministeriales, 1900-2000)

    by Adolfo Meisel Roca (


    Abstract (Translated from Spanish by the reviewer)

    It is well known that regional identities are very strong in Colombia and that they have had an influence in the politics of the country. One of the dimensions that Presidents of the Republic take into account for the composition of their cabinet is regional origin. In this paper we analyse the regional origins of Colombian ministers at cabinet level during the twentieth century. This to identify whether particular regions dominate the number of ministers in the cabinet and whether individuals from a particular region were overwhelmingly appointed to a specific ministry. To analyze the regional participation in the cabinet, we constructed a database with the names of the 702 persons that served as ministers during the twentieth century. The paper focuses on the Caribeann region, because we wanted to study its participation in the national political scene, during a century in which its economy lagged behind in comparison to the central region of the country. However, to articulate a comparative regional perspective, we have also widely discussed the cases of Antioquia and Bogotá and, to a lesser extent, of other regions of the country. The results show the strong influence of Antioquia and the departments of the coffee-growing region on the cabinets of the first part of the twentieth century, as well as the rise of Bogotá in the last decades of the century.

    Review by Sebastián Fleitas

    This paper was distributed by NEP-HIS on 2012-04-17. It addresses the role of the Caribbean region in the executive branch of the Colombian government, given that it was a region that fell behind in terms of economic development, human capital and political power throughout the twentieth century. The regional inequality in Colombia is well-know. It is an issue of economic and political policies to the extent that each president of Colombia has had to take into account while aiming for a balanced regional composition of his cabinet. To describe and interpret the links between political power and economic development at the regional level, Adolfo Meisel constructs a database with the information about regional origin for all the 702 persons that were ministers during the twentieth century in Colombia.

    Colombian Regions (number two is Caribbean Region)

    Colombian Regions (number two is Caribbean Region)

    I think that this paper makes two important contributions. The first one is that the article describes in a very precise way that the Caribbean region had a low participation in the executive branch during the twentieth century and that in fact this low participation contrasts with the increased influence of the capital city (Bogotá) and the coffee-growing region. This trends can be seen in the fact that no president was born in the Caribbean region; in the participation of the Caribbean region in the total number of ministers during the century (about 12%); and also in the fact that the participation of ministers born in the Caribbean region in key areas (like the Treasury, Trade, Defense, Education or Economic Development) was lower than than the average participation of individuals from other regions. Meisel also advances some possible explanations regading the change in regional political power in Colombia: the backwardness of the Caribbean region, the rise of the coffee-growing region and the consolidation of Bogotá as the most important economic center.

    Second, this paper addresses the important issue of political power and economic development, making focus in the regional dimension in a country with strong regional inequalities. The literature on the linkages between political power and economic development has a long tradition. More recently it has had a big impulse with the work of Acemoglu and some of his co-authors (see Acemoglu, et. al., 2005, the recent book by Acemoglu and Robinson´s Why Nations Fail as well as Anna Missiaia’s entry to NEP-HIS entitled Acemoglu on Past, Present, Future and Beyond). However, similar efforts to understand the institutional factors and the role of political power in economic development in Latin America are few and far apart (such as the work of Roderic A. Camp for Mexico). This issue is of great importance in Latin America due to the persistence of institutions that have been blamed of deterring economic development. Moreover, Latin America exhibits significant levels of inequality that could lead to worse economic performance if the elites use their political power to perpetuate themselves using these extracting institutions.

    Cartagena de Indias (The most famous Caribbean region city)

    There are, however, two major concerns that limit the scope of the contributions in this paper and in particular, regarding the linkages proposed between regional political power and economic development. The first one relates to the fact that the author uses the place where the minister was born in order to assess the regional origin, which in turn is used to assess the political power of the regions during the twentieth century.

    It is not straightforward that the origin of the minister is a good indicator of the political power of a region and even more that this type of political power could be transformed into a better economic performance of that region. At least two reasons can be argued. First, the region where the minister was born might not be a good indicator of his commitment with that region, either because this commitment is not strong before being elected or because after being elected his preferences can change. For instance, whether President Obama was born in Kenya or Hawaii is somewhat besides the point as he took office as Senator from Illinois.

    As a cabinet minister rather than an elected official, a politician might wish to maximize, among other things, their continuity in the Executive branch, which depends more on the President’s decisions than on the regional support for a particular individual. On the other hand, there is not a direct relationship between this kind of political power and the economic performance of a region, because even when if a minister wants to favor a region, policies enter in a political process in which the other ministers take part, and also a complex process of negotiation in the legislative branch (were possible lobbyists can take action). For these reasons, I think that the paper would benefit if the author discussed a little bit more the relationship between the executive and legislative branch, the role of regional lobbyists and the existence of regional development policies, in order to try to isolate the effect of the regional composition of the cabinet on the performance of the Caribbean region.

    Casa de Nariño (Colombian Government House)

    A second important issue that Meisel could address is the existence of endogeneity between political power and economic development. It is key to address this endogeneity in order to make conclusions and to determine the causality in the relationship. An example may help to make this point more clear. Assume that a region that has better performance over time due to different commodity endowments and invests some of the proceeds to improve its educational system. Assume that eventually the more educated people serve in the cabinet. In this example, the composition of the cabinet will show that people from more developed regions have a greater participation in cabinet, but note that the causality goes from a better performance to increased participation.

    But the things can be the other way around. Assume now that exogenously a president born in the central region decides to appoint only ministers from his region for the Cabinet. Assume also that these people decide to allocate the resources for education in a very unequal way, favoring this region. Assume that eventually by this process the people from this region get more educated than the people from other regions and that this increase in human capital boosts economic growth. Under the assumption that the more educated people have better opportunities to be in the Cabinet, then the data will again show that people from more developed regions have a greater participation in cabinet, but note that this time the causality goes from the political power to economic development.

    These two very simplified examples predict the result that the author finds for the composition of the Colombian Cabinet during the twentieth century. However, the reasons for this are very different in each example. In the first example economic development generates political power while in the second political power causes a better economic performance. Overall, I find that the discussion of the linkages between political power and economic development are really important to economic development. This is particularly true for Latin American countries, where the existence of a huge inequality could be seen as a hint of the presence of an economic elite that reproduces its privileges over time. However, in order to advance in the understanding of the linkage between political power and economic development, the big challenge is to find exogenous shocks that allow to draw conclusions about the direction of the causality.