James Cooley Fletcher

At the beginning of the 19th century there was almost no vestige of Protestantism in Brazil. From the 16th century the country was colonized basically only by Portuguese, who resisted the advance of Protestantism during the same period. Huguenots and Dutch Reformers tried to colonize parts of Brazil in the 16th and 17th centuries, but with little or no lasting effects. Only after the arrival of the Portuguese royal family in 1808 did this picture begin to change.

First came the English Anglicans. England rendered a great help to Portugal in the context of the Napoleonic Wars, and thus the subjects of the English crown gained religious freedom on Brazilian soil. This freedom soon extended to German Lutheran immigrants who settled mainly in the south of the country from the 1820s. However, it was only with the American missionary work, from the 1840s and 1850s, that Protestantism really began to settle in Brazil.

James Cooley Fletcher was one of the people who contributed most to the establishment of Protestantism in Brazil. Quoted frequently by historians, he is, however, little understood by most of them and little known by the general public. Born April 15, 1823 in Indianapolis, Indiana, he studied at the Princeton, Paris, and Geneva Seminary between 1847 and 1850 and first came to Brazil in 1852. In 1857 he published the first edition of The Brazil and the Brazilians, a book which for many decades would be the main reference regarding Brazil in the English language.

Fletcher first came to Brazil as chaplain of the American Seamen’s Friend Society and a missionary of the American and Foreign Christian Union. However, shortly after his arrival in the country, he made it his mission to bring Protestantism to the Brazilians. His performance, however, would be indirect: instead of preaching himself to the Brazilians, Fletcher chose to prepare the ground for other missionaries. For this he became friends with several members of the Brazilian elite, including Emperor Dom Pedro II. Through these friendships, he managed to influence legislation favorable to the acceptance of Protestantism in Brazil.

Although Fletcher anticipated and aided missionaries who would work directly with the conversion of Brazilians to Protestantism, his relationship with these same missionaries was not always peaceful. Some of the missionaries who succeeded Fletcher were suspicious of him because of his contacts with Brazilian politicians. It is true, Fletcher had an agenda not always identical with that of other missionaries: while others wished to focus only on the conversion of Brazilians, he understood that Protestantism and liberalism were closely linked, and that the implementation of the first in Brazil would lead to the progress propelled by the second. For this very reason, Fletcher had no problem engaging in activities that at first glance would seem oblivious to purely evangelistic work. He promoted, for example, the immigration of Americans to Brazil, the establishment of ship lines linking the two countries, the end of slavery in Brazil and commercial freedom.

James Cooley Fletcher is generally little remembered by Brazilian Protestants, although he has contributed decisively to the end of the Roman Catholic monopoly in the country. He is also little remembered by historians, but this should not be so. Fletcher was one of the people who contributed most to the strengthening of religious freedom in Brazil, and also to a combination of religious, political, and economic beliefs. It was precisely because of his religious beliefs that he believed in the political and economic strength of liberalism to transform any country, including Brazil.

Can we trust US interwar inequality figures?

This question is the one that me and Phil Magness have been asking for some time and we have now assembled our thoughts and measures in the first of a series of papers. In this paper, we take issue with the quality of the measurements that will be extracted from tax records during the interwar years (1918 to 1941).

More precisely, we point out that tax rates at the federal level fluctuated wildly and were at relatively high levels. Since most of our inequality measures are drawn from the federal tax data contained in the Statistics of Income, this is problematic. Indeed, high tax rates might deter honest reporting while rapidly changing rates will affect reporting behavior (causing artificial variations in the measure of market income). As such, both the level and the trend of inequality might be off.  That is our concern in very simple words.

To assess whether or not we are worrying for nothing, we went around to find different sources to assess the robustness of the inequality estimates based on the federal tax data. We found what we were looking for in Wisconsin whose tax rates were much lower (never above 7%) and less variable than those at the federal levels. As such, we found the perfect dataset to see if there are measurement problems in the data itself (through a varying selection bias).

From the Wisconsin data, we find that there are good reasons to be skeptical of the existing inequality measured based on federal tax data. The comparison of the IRS data for Wisconsin with the data from the state income tax shows a different pattern of evolution and a different level (especially when deductions are accounted for). First of all, the level is always inferior with the WTC data (Wisconsin Tax Commission). Secondly, the trend differs for the 1930s.

Table1 for Blog

I am not sure what it means in terms of the true level of inequality for the period. However, it suggests that we ought to be careful towards the estimations advanced if two data sources of a similar nature (tax data) with arguably minor conceptual differences (low and stable tax rates) tell dramatically different stories.  Maybe its time to try to further improve the pre-1945 series on inequality.

Empire effects : the case of shipping

I have been trying, for some time now, to circle an issue that we can consider to be a cousin of the emerging “state capacity” literature (see Mark Koyama’s amazing summary here). This cousin is the literature on “empire effects” (here and here for examples).

The core of the “empire effect” claim is that empires provide global order which we can consider as a public good. A colorful image would be the British Navy roaming the seas in the 19th century which meant increased protection for trade. This is why it is a parent of the state capacity argument in the sense that the latter concept refers (broadly) to the ability of a state to administer the realm within its boundaries. The empire effect is merely the extension of these boundaries.

I still have reservations about the nuances/limitations of state capacity as an argument to explain economic growth. After all, the true question is not how states consolidate, but how they create constraints on rulers to not abuse the consolidated powers (which in turn generates room for growth). But, it is easy to heavily question its parent: the empire effect.

This is what I am trying to do in a recent paper on the effects of empire on shipping productivity between 1760 and 1860.

Shipping is one of the industry that is most likely to be affected by large empires – positively or negatively. Indeed, the argument for empire effects is that they protect trade. As such, the British navy in the 19th century protected trade and probably helped the shipping industry become more productive. But, achieving empire comes at a cost. For example, the British navy needed to grow very large in size and it had to employ inputs from the private sector thus crowding-it out. In a way, if a security effect from empire emerged as a benefit, there must have been a cost. The cost we wish to highlight is the crowding-out one.

In the paper (written with Jari Eloranta of Appalachian State University and Vadim Kufenko of University of Hohenheim), I argue that, using the productivity of the Canadian shipping industry which was protected by the British Navy, the security effect from a large navy was smaller than the crowding-out from high-levels of expenditures on the navy.

While it is still a working paper which we are trying to expand and improve, our point is that what allowed the productivity of the Canadian shipping industry (which was protected by Britain) to soar was that the British Navy grew smaller in absolute terms. While the growth of the relative strength of the British Navy did bolster productivity in some of our tests, the fact that the navy was much smaller was the “thing in the mix that did the trick”.  In other words, the empire effect is just the effect of a ramping-down in military being presented as something else than it truly is (at least partly).

That’s our core point. We are still trying to improve it and (as such) comments are welcomed.

Japanese adoption

A recent article in the Economist describes the results from a study of an interesting Japanese custom. Traditionally (and according to the civil code until 1945), company ownership and leadership were passed on through primogeniture. This custom has continued to be practiced, and the study found solid evidence that family-managed companies outperform professionally managed companies. It could be argued that family unofficial training, continuity, and trust are at the root of this, but the authors find a different reason: adoption.

Japan and the US lead the world in adoption rate, but Japanese adoption is not necessarily what you would think. Over 90% of Japanese adoptions are of adults, who are usually men adopted into childless (or more specifically son-less) families for business purposes. This is also quite traditional, and since Japanese birth rates are extremely low, Japanese businessmen are likely to continue adopting talented, ambitious single men. This is often also accompanied by marriage to the daughter of the businessman (while women participate in business in Japan, they are more rarely the executives), which is referred to as mukoyoshi and combines familial with business ties.

This practice has the advantage of allowing for the trust, mutual investment, and long-term planning and teaching based on family relationship while avoiding the risk of having an unfit successor. In fact, in reading this, I am reminded of another culture’s custom of adoption: ancient Rome.

Ancient Rome also had a custom of adoption among its upper classes because of the prestige associated with old patrician family names and the incentives of inheritance laws (for a time it also served as a political means of a patrician gaining the tribuneship, as the Gracchi did). Similarly to the Japanese businessmen, adopted sons tended to excel (the earliest and best example is the Republican general Scipio Aemilianus), and it served as a method of political alliance and developing shared interest. It also became a typical method of succession for Roman emperors in the case that no son was available, a deal needed to be made between factions, or an emperor had a favored successor to whom he was not related. Adopted emperors included Trajan (adopted by Nerva as deal with the army, successful and beloved military leader), Hadrian (great builder and patron of the arts), and Marcus Aurelius (famed general and philosopher), and were generally well-reputed and effective leaders. In contrast, the sons of emperors (such as Domitian and Commodus, who were both infamously insane and harmful, or Maxentius, whose usurpation followed a blood-based claim) or blood-based successors (such as Caligula, who was blood-thirsty and childish, Nero, who was brutal and wasteful, and Elegabalus, who had was more interested in orgies than leadership).

It seems that the same potential reasons underlying Japanese business success through adoption were also at work in antiquity: emperors who were chosen and bred based on their abilities provided the continuity associated with heredity but with the advantage of meritocratic selection. This raises the question of whether this advantage has somehow been passed over by the Western world in those cases where we have avoided family-based succession in business (based on worries about nepotism and/or to gain the advantages of meritocratic selection). Even more interesting, does the adoption practice maintain the freedom of choice and opportunity found in Western careers while also conferring the ability to maintain trust and continuity that family-based succession offers?

The Return of Cyclical Theories of History

“People and states oscillate between peace and war, freedom and slavery, order and disorder. They tire easily. Even happiness soon grows wearisome. No sooner do they begin to enjoy the benefits of wise and just government than they demand more wisdom and a different kind of justice. Factions spring up. Everyone is on the lookout for new privileges. The equilibrium that was so hard to strike crumbles. Wild hopes are embraced. The system collapses. Everything has to be built up anew on the ruins of the past”.

Jean, D’Ormesson, The Glory of the Empire

This, from D’Ormesson’s excellent 1971 fictional history The Glory of the Empire, could stand in for many such statements from thinkers who have held to a cyclical view of political development: Polybius, Machiavelli, Vico, Spengler, and Arthur Schlesinger. Now, after a period of eclipse, cyclical histories are back in fashion. Tyler Cowen in his excellent new book The Complacent Class heralds their return:

The biggest story of the last fifteen years, both nationally and globally, is the growing likelihood that a cyclical model of history will be a better predictor than a model of ongoing progress. (Cowen, 2017, p. 200)

The leading modern day cyclical theorist is undoubtedly Peter Turchin. For my money Turchin’s best book is Secular Cycles (co-authored with  Sergey A. Nefedov). Their innovation (building on an argument made by my GMU colleague Jack Goldstone in his 1991 book Revolution and Rebellion in the Early Modern World) is to take the Malthusian model of economic cycles and add to it a model of elite competition.

Tuchin and Nefedov show that periods of demographic expansion are often associated with the growth of elite incomes and inequality (as population growth causes rents to rise and wages to fall). More elites competing over the surplus, however, puts fiscal pressure on the surplus-extraction machine that we call the state. Elite overproduction thus brings about a political crisis. Secular Cycles applied this model to medieval and early modern England and France, Russia and ancient Rome. Turchin’s most recent book applies it to the United States.

Another recent cyclical account that has caught my attention is that of Bas van Bavel. His recent book The Invisible Hand? (OUP 2016) has been favorably reviewed by Branko Milanovic (strangely Bavel doesn’t cite Turchin). I quote in full from Branko’s review:

Van Bavel’s key idea is as follows. In societies where non-market constraints are dominant (say, in feudal societies), liberating factor markets is a truly revolutionary change. Ability of peasants to own some land or to lease it, of workers to work for wages rather than to be subjected to various types of corvées, or of the merchants to borrow at a more or less competitive market rather than to depend on usurious rates, is liberating at an individual level (gives person much greater freedom), secures property, and unleashes the forces of economic growth. The pace of activity quickens, growth accelerates (true, historically, from close to zero to some small number like 1% per year) and even inequality, economic and above all social, decreases . . .

But the process, Bavel argues, contains the seeds of its destruction. Gradually factor markets cover more and more of the population: Bavel is excellent in providing numerical estimates on, for example, the percentage of wage-earners in Lombardy in the 14th century or showing that in Low Countries wage labor was, because of guilds, less prevalent in urban than in rural areas. One factor market, though, that of capital and finance, gradually begins to dominate. Private and public debt become most attractive investments, big fortunes are made in finance, and those who originally asked for the level playing field and removal of feudal-like constraints, now use their wealth to conquer the political power and impose a serrata, thus making the rules destined to keep them forever on the top. What started as an exercise in political and economic freedom begins to look like an exercise in cementing the acquired power, politically and economically. The economic essor is gone, the economy begins to stagnate and, as happened to Iraq, Northern Italy and Low Countries, is overtaken by the competitors.

This is a great summary of the main idea of the book.  And the idea of endogenous economic cycles is an intriguing one.

But my impression of Bavel’s argument is less favorable. I am more inclined to the views expressed in this more critical review of the book by Peer Vries who lauds the ambition of the project but wishes that the execution was better. Like Vries I think there are issues with defining and measuring the growth of factor markets. I think that purely internalist stories of rise and decline might make sense for some preindustrial societies but are much less compelling for the more interconnected early modern world let alone for the post-1800 period. All in all, the book lacks a clearly laid out theoretical framework and suffers for it.

This said, cyclical patterns in history and in particular what one might call political cycles should get more attention. We should not be looking to date Kondratiev waves or other such pseudo-scientific phenomenon, but we should seek to build models and explanations that explain the pattern of ups and downs, growth efflorescence followed by crises and collapses, that characterized preindustrial history.

On British Public Debt, the American Revolution and the Acadian Expulsion of 1755

I have a new working paper out there on the role of the Acadian expulsion of 1755 in fostering the American revolution.  Most Americans will not know about the expulsion of a large share of the French-speaking population (known as the Acadians) of the Maritimes provinces of Canada during the French and Indian Wars.

Basically, I argue that the policy of deportation was pushed by New England and Nova Scotia settlers who wanted the well-irrigated (thanks to an incredibly sophisticated – given the context of a capital-scarce frontier economy – dyking system) farms of the Acadians. Arguing that the French population under nominal British rule had only sworn an oath of neutrality, they represented a threat to British security, the settlers pushed hard for the expulsion. However, the deportation was not approved by London and was largely the result of colonial decisions rather than Imperial decisions. The problem was that the financial burden of the operation (equal to between 32% of 38% of the expenditures on North America – and that’s a conservative estimate) were borne by England, not the colonies.

This fits well, I argue, into a public choice framework. Rent-seeking settlers pushed for the adoption of a policy whose costs were spread over a large population (that of Britain) but whose benefits they were the sole reapers.

The problem is that this, as I have argued elsewhere, was a key moment in British Imperial history as it contributed to the idea that London had to end the era of “salutary neglect” in favor of a more active management of its colonies.  The attempt to centralize management of the British Empire, in order to best prioritize resources in a time of rising public debt and high expenditures level in the wars against the French, was a key factor in the initiation of the American Revolution.

Moreover, the response from Britain was itself a rent-seeking solution. As David Stasavage has documented, government creditors in England became well-embedded inside the British governmental structure in order to minimize default risks and better control expenses. These creditors were a crucial part of the coalition structure that led to the long Whig Supremacy over British politics (more than half a century). In that coalition, they lobbied for policies that advantaged them as creditors. The response to the Acadian expulsion debacle (for which London paid even though it did not approve it and considered the Acadian theatre of operation to be minor and inconsequential) should thus be seen also as a rent-seeking process.

As such, it means that there is a series of factors, well embedded inside broader public choice theory, that can contribute to an explanation of the initiation of the American Revolution. It is not by any means a complete explanation, but it offers a strong partial contribution that considers the incentives behind the ideas.

Again, the paper can be consulted here or here.

Turkey’s Referendum: Authoritarianism and Electoral Fixes

As previously indicated I will be posting an appendix to my posts on Coup and Counter-Coup in Turkey, referring to Ottomanism, Kemalist republicanism and related issues. The sixteenth of April referendum does require a response of a more immediate kind. The referendum was on amendments to the Constitution largely concerned with transforming Turkey from a parliamentary republic, which it has been at least in principle since the formation of the Republic of Turkey in 1923. Recep Tayıp Erdoğan had already been breaking the Constitution since 2014 when he was elected President of Turkey after more than a decade as Prime Minister. On becoming President he transferred the chief executive power to the presidential palace (which has more than 1000 rooms and was built on Erdoğan’s orders using executive privilege to override court bans on building on the land concerned). This in itself tells you everything you need to know about the decay of Turkish democracy, from a starting point which was itself not a shining beacon to the world of purist constitutional democracy.

Erdoğan’s ambitions for an executive presidency in Turkey precede his elevation to that role. The shift from a President elected by the National Assembly to a President elected by popular vote came from a referendum of 2007, though Erdoğan is the first President of Turkey to take the office in this way. Kemal Atatürk (1923-1938) and then İsmet İnönü (1938-1950) were powerful presidents in a parliamentary system. This paradox arose because Turkey was a de facto one-party state in which only the Republican People’s Party had seats in the National Assembly from 1923 to 1943, though the Free Republican Party won seats in local elections. The elections of 1946 resulted in an opposition party and some independent deputies entering the National Assembly. In 1950 İnönü became the first Turkish (or Ottoman) leader to give up power peacefully as the result of elections and the Republican People’s Party became the first political party to give up power in this manner. During the one party period, the President was the dominant figure in the Republican People’s Party and therefore acted as the head of government, though with a Prime Minister and some genuine divisions of responsibilities.

Not only did Erdoğan start using powers he had not been given in the constitution in 2014, the whole evolution towards a presidential system has been done in a way to benefit him personally. He was Prime Minister over three terms from 2002 to 2014, an office to be abolished in 2019 as a result of the recent referendum, and was the head of government. He has acted as effective head of government since 2014, regardless of the Prime Minister having this role. The office of Prime Minister will still exist until 2019 and most of the constitutional changes as a result of the referendum will not come into force until then. Erdoğan can then have the two terms of executive presidency in addition to the term he is currently serving which will still constitutionally limit his powers in ways that mean the Prime Minister should be head of government. In this sense, the system is working as in the days of the one party system. So Erdoğan can serve as head of government in Turkey from 2003 until 2029. Furthermore he may be able to add another term if the National Assembly goes to an early election during his second term.

The President will have the power to dissolve the National Assembly, and this could easily happen simply because the President wishes to have a third term. The amended constitution at least restricts the President to two five-year terms plus most of a third term in special conditions. So the possibility exists of Erdoğan running the Turkish government from 2003 to 2034, a highly unusual situation in any democracy and one likely to undermine the democracy in question, particularly as the powers of the President now include control of appointment of senior judges, senior civil servants, senior bureaucrats, the right to issue decrees as laws, the right to appoint all cabinet ministers without National Assembly approval, and the right to appoint two vice-presidents without National Assembly approval. Theses figures will not be required to answer questions in the National Assembly and, like the President, will benefit from lifetime immunity with regard to alleged crimes committed in office. That is to say, the President and his associates will have immunity for life unless the National Assembly votes to suspend the immunity, with a high enough majority required to make this unlikely unless there is a massive collapse in the number of AKP deputies, or of Erdoğan’s control of the AKP.

We cannot even say that these changes designed to produce a President above normal democratic constitutional checks and balances, dominating the whole governmental process and state machinery in a way unprecedented in Turkey’s multi-party history, have been agreed to by a genuine majority vote. The referendum was held in state of emergency conditions, which still prevails. A state of emergency in which opposition journalists have been detained in large numbers on flimsy charges as ‘terrorists’, opposition deputies (from Kurdish rights-leftist HDP) have been detained on a similar basis. State media and most private media groups operate as media organs of the AKP. The state of emergency has been applied in a particularly harsh way in the southeast (Kurdish majority) part of the country where elected local government has been replaced by central government appointees. There are 500,000 displaced persons in the southeast resulting from PKK terror and the state security reaction which led to military bombardment of whole towns and urban districts until they were reduced to rubble. It was clearly not easy for them to vote and it looks like a large number did not vote. Extreme intimidation of No campaigners was the norm in the southeast where large numbers of HDP election observers were denied access to polling booths. Intimidation of the No campaign took place elsewhere if in a less extreme way and public spaces were dominated by Yes publicity.

The count was itself full of flagrant irregularities. The number of polling stations recording a 100% vote for Yes was far greater than the number recording 100% AKP votes in recent elections. There was another party campaigning for the Yes, the hardline nationalistic MHP, but the party split over the leader’s support for Yes and all the evidence is that overwhelmingly most MHP voters did not vote Yes. Given that some AKP voters defected from Yes, not many but no less than 5%, there was no reason to expect an increased number of polling stations with a 100% vote and of course suspicion is in order about polling stations which recorded such results in the past.

An AKP politician who is a member of the national electoral board (itself packed with AKP appointees) requested that all ballot papers used to vote, but not carrying a stamp to show they have been authorised for that polling station by an official, should be counted. This is illegal but the request was granted. AKP apologists were quick to say that opposition requests in the past for counting such ballots (because maybe they were stuck together when the official was stamping papers) were granted in the past. Small illegality does not excuse large illegality and the scale of counting of such ballot papers was much larger than previously.

During the count the state news agency was announcing results before they were released by the electoral board. A strange and suspicious situation. The international media, following the state news agency, failed to see the discrepancy also confusing the percentage of ballot boxes opened with votes cast, giving a very misleading impression of a big lead amongst most votes cast early in the evening. The election board results then went off line and were not shared with the opposition. When results came back online they showed a very different pattern than before the break in service, more in line with the state news agency announcements.

Given the close (51.4% for Yes) nature of the official result, there are a number of reasons to think that No won in votes cast, and even if we ignore the voting irregularities, there is reason to think that in a less intimidating atmosphere, particularly in the southeast, more Yes voters would have cast a vote. In these circumstances I suggest that even on a very cautious reckoning, the number of votes cast for No was at least 51% and that with a less intimidating and disruptive atmosphere, another 2% would have gone to No. To say on this basis that 53% voted No is I believe a very cautious estimate. There is very probably a clear majority of Turkish voters against the new presidential system, at least 55%.

The narrow result clearly caused embarrassment to Erdoğan and the AKP who had predicted a very big victory for Yes. They have gone quieter since then, but with no let up in authoritarian measures. Just two days ago Wikipedia was blocked in Turkey, several thousand state employees were dismissed using emergency powers and a number of NGOs were closed in a similar way. The context of the supposed Yes victory gives hope that there is opposition to the Erdoğan/AKP destruction of liberal democracy, but recent measures suggest they are as determined as ever to use the tools of state to obliterate opposition, or just any sense of independence from the party-state machine.