***********

This article, although it contains some empirical data, is essentially an attempt to give more precision and definition to possible arguments about, and approaches to, the question whether insecurity had a lot or a little to do with the development of the Colombian economy in the 19^th century. It also contains more general reflections on the relations between insecurity and development, or at least economic growth.

It is a polemical piece, and written to stimulate debate. It derives in part from my reading of certain articles and books, mostly by North American academics, which I think have been too influential in attributing the 19^th century backwardness of other parts of the hemisphere to political disorder, and which are coming to constitute a new orthodoxy, almost a new «black legend», one that I think mistaken. The arguments also derive from my reading — for me a frustrating experience — of the works of Douglass NorthFootnote 1.

I am neither an economic historian nor an economist. My reading of these works has been that of an ordinary historian, one with a propensity to insist that any argument should be sustained with the necessary evidence. Some of the methods of argument encountered, for their abandoning of empirical investigation, remind me of the excesses of the years of «dependency theory», when some — not all — thought that there was not much point in studying the contortions of the dependent countries as everything important was determined outside them. For example, on judicial reorganizations Alan Dye writes: «The economic consequences are difficult to measure but straightforward to describe.» (2006, p. 194). That strikes me as contradictory: something difficult to measure cannot be that straightforward to describe, at least not in a satisfactory manner, one suited to the demands of economic history, a branch of the discipline usually proud of its rigour. And North (Reference North1990, p. 107) in a poetic moment: «We cannot see, feel, touch or even measure institutions; they are constructs of the human mind.» It does not seem to me that all institutions are so impalpable, and if they are so, how is any verification of such theory possible? North himself, in the preface to this book, warns the reader that «The history I include is illustrative, designed to show the promise of the approach, but far from providing for the kind of hypothesis testing that must ultimately be done.» (1990, p. vii).

My preliminary conclusion is simple: disorder and insecurity did not have much to do with the degree of development achieved by Colombia in the 19^th century. It is probable that they were no help to the progress of the country, but they were not great hindrances. I am also sceptical about their importance elsewhere in 19^th century Latin America, and find uncertain inspiration in the writing on this theme concerning other parts of the world.

A contrary notion exists, that war can favour economic development. It is not new. Its first systematic defender in Europe was Sombart (Reference Sombart1913). The debate about the consequences of European wars was also the central theme of Nef (Reference Nef1950). Nef did not support Sombart, and concluded that one of the advantages of England in the two centuries before the industrial revolution was that she had suffered less from war than her European rivals, but his analysis is more impressionistic than precise.

The literature on the economic impact of international wars is surprisingly scant, as is that on the economic consequences of the best-known civil wars, such as the English Civil War of the 1640s and the War of Secession in the United States. There is an old debate, one with a nice exchange of insults but with little light shed and few conclusions reached, on the impact of the Hundred Years War on the English economy in the 14^th century, between the heavyweights McFarlane (Reference McFarlane1962), Stone (Reference Stone1962) and Postan (Reference Postan1964). None of them took much trouble to find any evidence, but it is worth quoting McFarlane, who questioned «…the dogmatic belief, inherited from the 19^th century, that wars can only be damaging to the societies engaged in them», and who criticized «the preference, inspired in part by the intractable nature of the financial evidence, for the oracular phrase, the more nebulous and commonplace the better.» «To say, for example,» he continues, «that England had been «bled white» by around 1314 is no substitute at all for a quantitative balance, no matter that it would have to be a crude one … Precision may be unattainable, but any attempt is better than such an empty phraseFootnote 2.»

Another rather more serious debate — an older and a longer one, beginning in 1861 — concerns the impact of the Thirty Years War on the economies of Germany and the low countries, and is summarized by Rabb (Reference Rabb1962) and by Kamen (Reference Kamen1968). The conclusions of both authors are tentative, pointing to much regional and local variation — stagnation here, rapid recuperation and even gain there –– and are supported by a wide range of precariously founded statisticsFootnote 3.

On the War of Secession in the United States, some historians reach more severe conclusions. McPherson (Reference McPherson2008) in his review of Neely (Reference Neely2007) writes that « … the Civil War wiped out two thirds of the assessed value of wealth in Confederate states, two fifths of the South's livestock, and more than half its farm machinery — not to mention one quarter of the Confederacy's white men of military age. While Northern wealth increased by 50 per cent from 1860 to 1870, Southern wealth decreased by 60 per cent.» The U.S. civil war destroyed the slave-based plantation economy of the South, the largest and most efficient example of such an economy in the Americas. However, no Colombian civil war had such a catastrophic economic denouementFootnote 4.

I emphasize that this essay is focussed on the 19^th century. Some of my arguments are pertinent to the 20^th century as well, and even the 21^st century, but it would be necessary to modify them in many ways to embrace the distinct economic conditions and conjunctures of more recent times. It is also concerned only with the impact of insecurity on the Colombian economy. If I conclude that that was marginal, it does not mean that I conclude that insecurity was unimportant in other respects.

Here are a series of arguments.

Were Colombian insecurity and disorder exceptional? I think not. Some historians — and among them economic historians who tend to pride themselves on being more precise, more rigorous, more faithful to the empirical evidence, apart from commonly more quantitative than their ordinary colleagues — abandon all rigour when they come to contemplate the so-called anarchy of the Latin American 19^th century, and begin to write like Thomas Carlyle wrote a century and a half ago. In a well-known essay, Carlyle wrote of «all South America raging and ravening like one huge dog-kennel gone rabid.» (Carlyle Reference Carlyle1843).

Do I exaggerate when I compare the febrile rhetoric of some of the «new institutionalists» among our economic historians with Thomas Carlyle? Present in embryo in the pioneer essay by John Coatsworth, it reaches surprising new heights in the apparently sober pages of the recent Cambridge Economic History of Latin America (Bulmer-Thomas et al. (eds) Reference Bértola and Williamson2006).

Leandro Prados de la Escosura seems to me there to exaggerate the prolonged nature of the negative effects of the wars of independence, as do a number of other authors. Even Venezuela, where the conflict was much more intense than in New Granada, recovered fairly rapidly. On the recuperation of an agricultural region of the country, one can consult the ingenious essay of the French precursor of ecology Boussingault (Reference Boussingault1849, pp. 4-7) who observed that during the wars of independence the waters of the Lake of Valencia had risen, owing to the recuperation of the neighbouring woods brought about by the cessation owing to the fighting of all agricultural activity. His translator, the Colombian General Joaquín Acosta, citing observations of the geographer Colonel Agustín Codazzi, noted that with the coming of peace the lake was receding again as the woods were cut down to expand the cultivation of coffeeFootnote 5.

Luis Bértola and Jeffrey Williamson assert that Latin America «experienced almost continuous war and civil strife between the 1820s and 1870s … practically all of Latin America experienced episodes of massive and prolonged civil strife. In six countries, internal civil wars raged more or less continuously for decades after independence.»Footnote 6

None of this has any serious basis in empirical research — the authors do not specify which were the six countries in continuous civil warFootnote 7.

This is not the place for an exhaustive critique, but their impressionistic view ignores long spells of good commercial order, for example that which prevailed in Argentina under Juan Manuel de Rosas, the idol of the foreign merchants and the estancieros of Buenos Aires. The early economic history of the independent River Plate is one of largely successful integration with the world economyFootnote 8.

Venezuela under José Antonio Páez, the head of what is termed the «conservative oligarchy», was also a country that up until the 1850s enjoyed a reputation for stability, something like that of Chile. On Venezuela, it is worth noting the opinion of Antonio José de Irisarri, an author who was well acquainted not only with his native Central America, but also with Mexico and most of South America. In the very pessimistic, and very eloquent, «Discurso Preliminar» to his Historia crítica del asesinato cometido en la persona del Gran Mariscal de Ayacucho — a good source, let it be said in passing, of Carlyle-like laments about the defects of formerly Spanish America when compared to the United States — he writes that «Venezuela and Chile are the only countries of this continent in which it has been seen that men have tried not to waste their time in useless questions, but to occupy themselves in what is most convenient; nonetheless it cannot be doubted that the progress of Chile has been greater than that of Venezuela, perhaps because the Venezuelans have been more impatient and have wanted to force nature to produce in a shorter time than is necessary results that inevitably take their time.» (Irisarri Reference Irisarri1846, pp. 19-20). I cite him here because along with Chile he singles out Venezuela, the region that suffered more than any other in the independence wars. Nor is it correct for New Granada to speak of «massive and prolonged civil strife … more or less continuous civil war.»Footnote 9

The intoxication with disorder assumes various forms in the new orthodoxy.

The first is that the adepts have an impressionistic way with numbers, uncritically adding together coups, revolts and irregular changes of government with no regard for their distinct natures — insecurity is, for example, not the same as irregularity — or for their real importance or lack of importance. This is a common vice in the cases of Mexico, Colombia and Bolivia. The confusion between insecurity and instability is particularly present in those who write about the first-half century of independent Mexico. In the Colombian case, we are all familiar with the figures of Gustavo Arboleda and others: nine national conflicts and sixty local revolts in the course of the century, but such counts give no indication about their seriousnessFootnote 10.

A second mistake — or omission — in the argument of lack of growth through insecurity is to consign to complete oblivion what passed in other parts of the world. Although the 19^th was not the most violent or conflictive century in the history of Europe — the 17^th and the 20^th surpass it in those respects — it was far from being an era of perpetual peace: it begins with some 15 years of Napoleonic warfare, with great destruction and loss of life, particularly of French lives. And consider France for the rest of the century: revolutions with varying degrees of violence in 1830, 1848, 1851 and 1870, the last combined with foreign invasion and the Paris Commune. France was a country that without doubt suffered more insecurity and disorder than Chile.

In the Americas, the longest civil war and the one most destructive in lives and property was the War of Secession in the United States. One is struck by how little has been written on the effects of these events on the economic history of these two countries, France and the United StatesFootnote 11.

The question then arises of the possibly distinct quality of these violences and insecurities. In what I have read of the «new institutionalists», I detect an implicit notion that there is something particularly pernicious, when it comes to economic development, in Latin American disorder, compared with disorder in other parts of the world. Clearly, the impact of a war, the possibilities of subsequent rapid recovery, may well be quite different in the cases of a rudimentary, «traditional», rural economy of primary producers and of an industrialized one. One thinks of the rapid recovery after the Second World War of Germany and Japan, although the advantage in rapid recovery does not seem to me to rest clearly with advanced industrial economies: rapid recovery can be seen in many more primitive economies, as Mill asserted and as can be seen in Colombia in the 19^th century. The question deserves more study.

Regarding the recent successes of many Asiatic economies, it might be argued that there the Second World War had positive results, bringing «creative destruction», and that the fault of the Latin American wars of the 19^th century was that they were not sufficiently destructive, that they did not go far enough in destroying the old structures that the neo-institutionalists deplore. If that was the case, then their arguments about the negative influence of disorder need to be modified. There are traces of such an argument in Coatsworth (Reference Coatsworth2006, p. 271): «… warfare that followed independence contributed to undermining the colonial institutional legacy that had inhibited economic growth and thus accelerated the creation of a more efficient economic organization throughout Spanish America.» Alas, apparently it did not accelerate it sufficiently. The reader is urged to conclude that the institutions of Latin America are perverse when the region is stable, as under the Empire, and perverse when it is unstable, as in the post-independence years. Such a conclusion is not logically impossible, but it does lead one to suspect these authors of bias.

Another trap is the indiscriminate surveying of «Latin America». As has been seen with Thomas Carlyle, it is all too easy to generalize condescendingly about a score of distant countries, where it is probable that things are going badly in at least one or two. However, anyone can gather from the reading of a manual of the economic history of the region, from Glade (Reference Glade1969) or Bulmer-Thomas (Reference Bulmer-Thomas2003), that some countries achieved much more growth than others, that Argentina and Uruguay, for example, ended the 19^th century with per capita income five or six times higher than Colombia. All the same, it would not be so easy to demonstrate that these comparatively successful republics had enjoyed greater order and security than the suffering tropical republic of ColombiaFootnote 12.

A further error, or possible source of error, is the tendency to assume that the direction of causation is from disorder to lack of development, rather than the other way round. It is clear that economic recessions, which frequently had their origin outside the country, increased the proclivity for civil war, weakening governments and increasing the numbers of those disposed to go to war. Recognizing this pattern ought to lead to a less severe judgement of native institutionsFootnote 13.

This problem of the direction of causation is also applicable to the institutionalist argument in a more fundamental way. It is by no means always clear as to what exactly institutionalists are referring to — a historian is likely to be puzzled by a reference to an era or country of «weak institutions», without more specification, as the term is used so widely and in such different senses, encompassing anything from systems of land tenure to the ethnic divisions of a population. But some particular institutions obviously cost money: poor countries cannot afford elaborate judicial systems, or high levels of policing, or clearly competent armed forces. Such things may well be as much the product as the cause of economic development.

For all these reasons we can conclude, in this preliminary phase of analysis, that it is not at all obvious that disorder and insecurity, in the most visible form of civil war, explain the backwardness, slow growth and relative lack of progress of the 19^th century Colombian economy.

It is of course easy to find contemporary Colombian writers deploring the wars and insecurity and attributing to them the lack of national progress, but political rhetoric has to be distinguished from more sober analysisFootnote 14.

Here follow some simple observations on the nature of «anarchy» in Colombia. The civil wars were relatively short. Even in the case of those wars usually considered national rather than regional, not all of them involved all the territory. The armies were small — neither the government nor the revolution had the wherewithal to organize large armies, and it was very difficult to sustain large concentrations of troops. Although there are exceptions — remember Ruanda — it is difficult to make waste on a grand scale or carry out large massacres with small and dispersed forces. The worst behaviour of the contenders generally took place in remote and sparsely populated parts (see Garavito Reference Garavito1897, p. 70).

For obvious reasons, it is not easy to establish reliable statistics for the losses in the various civil wars of the 19^th century. Even where they exist, contemporary estimates vary from side to side, and there is no way of knowing how many perished through disease, change of climate, etc. Garavito, calculated the civil war dead up until then at 40,000 (1897, p. 41). Faute de mieux, I repeat faute de mieux, Gaitán Daza (Reference Gaitán Daza1995, p. 201) gives the following approximations:

1830-1831: 1,800

1839-1841: 3,366

1851: 1,000

1854: 2,000

1860-1862: 6,000

1876-1877: 9,000

1884-1885: 3,000

1895: 2,000

1899-1902: 80,000

He rightly points out that these numbers are the more significant in proportion to the small population of the country in those years, notably in the cases of the wars of 1876-1877 and of 1899-1902 — «equivalents for 1992» of 27,000 end even 200,000 in the years of worst intensity, respectively. But the total is modest, especially if the last war, the War of a Thousand Days, which according to these figures killed three times as many men as all the others put together, is left out. Although it was the longest and probably the most intensely fought, the figure for the Thousand Days does not look very plausible to me. It is the sum of estimates of 10,000 for 1899, and 25,000 each for 1900 and 1901, and 10,000 is obviously excessive for 1899, as the war only began in the middle of October of that year. Contemporary estimates for the wars vary: Gaitán's figure for the Thousand Days (i.e. 80,000) is the same as the contemporary estimate of Holguín (Reference Holguín1908); there are others, higher and lower. There is little agreement, for example, on the number of deaths in the Battle of Palonegro, the largest encounter of the whole war: Arboleda Cortés (Reference Arboleda Cortés1901) gives as bajas de sangre — presumably dead and wounded — 1,600 on the government side and 3,200 on the side of the revolution. Florez Álvarez (Reference Florez Álvarez1939), in a careful later account, gives 2,000 Liberals dead and wounded … Note that the War of a Thousand Days, which began in October 1899, for elementary reasons of chronology cannot be held responsible for the backwardness of the preceding century.

In general, a degree of discipline was maintained in these wars: the social hierarchy was in great part preserved in the course of these conflicts. A temptation to be avoided is to emphasize the worst, or according to taste the more radical or extreme or lethal in the wars: the social struggles in the Valle del Cauca at mid-century, the burning of Colón by Pedro Prestán in 1885, the guerrilla warfare at the end of the War of a Thousand DaysFootnote 15.

Neither should one be lead astray by contemporary European and North American observers who with the racial and cultural prejudices of their time characterized South American conflicts as particularly savage. They appear to them necessarily, unquestionably more savage than their own regular wars; perhaps they unconsciously exclude colonial affairs. There is possibly an element of truth here, for the majority were civil wars, with the logic of civil wars: see the analysis of internal conflicts in Kalyvas (Reference Kalyvas2006). There is always some savagery in irregular war, and civil wars are never entirely regular — there was much irregularity in the U.S. civil war. Nonetheless, even Joseph Conrad, a European of broad sympathies, exaggerates the savagery of South American fighting in Nostromo and Gaspar Ruiz. The Carlist wars in Spain itself appear to me to have been more intense, more bloody, than the Colombian wars of the same era — and Spain appears less secure and stable in the 19^th century than many American republics, although I have yet to read a Spanish economic historian who attributes Spanish economic backwardness primarily to this insecurity and instability. In international relations, Colombia in the 19^th century and up until today has been a markedly pacific country: the exceptions have been a few skirmishes with Ecuador and Venezuela, which never amounted to wars, and one short war with Peru. The loss of Panama, the republic's worst trauma, was not the result of international war.

There are many contemporary memoirs of civil war, by rebels and defenders of governments, and the essentially civilian character of most of their authors prevents them from writing formal military history, so that they frequently contain first-hand accounts of their experiences free from the distortions of professional vanity. My impression is that the degree of violence in 19^th century Colombia was a low one, much lower than the violence of the second half of the 20^th century. In the 19^th century, there was much less of the unbridled, uncontrolled sectarian conflict that was to be the dominant note in Colombia in the 1940s and 1950s, with a prelude in the early 1930sFootnote 16.

I wish to emphasize, faced with the rhetoric of certain economic historians, that «anarchy» and «turmoil» are potentially analysable and quantifiable phenomena, and that they are notions that need definition. A scale of gravity could be designed for insecurity, something along the lines of the Beaufort scale for tempests or the Richter scale for earthquakes, which would go from a small riot or bochinche, the Colombian term, force 1, to the total war in the closing stages of the Second World War, force 10. On such an imagined scale it would be my opinion that Colombia would only rarely have reached force 2 or 3.

What of property rights in Colombia in the 19^th century? What is the evidence for their deficiencies and faults, and the consequences for economic development?

Those who paint a picture of general anarchy have forgotten that the Spanish Empire was one of the most formidable bureaucracies the world has known, and that not all of that was lost with independence. The notarial and registry system, notariado y registro, which was implanted from the very beginnings of the Conquest, persists in its essentials up to the present day. Doubtless it had its defects, but its coverage was always extensive. I do not share John Coatsworth's view, (Reference Coatsworth2006, p. 60), that land registry did not exist: «land registries were unknown …» That they had their defects is one thing, but to assert that they did not exist is undoubtedly an error, nor is it so clear that they were inadequate. Furthermore, in the developed world, the systems of land title and registry have varied a great deal — the English system shares few characteristics with the French, for example. The devotion of Colombians of all classes to official stamped paper, papel sellado, to solemnize transactions only died recently, and the system of notariado y registro remains in good health and continues to expandFootnote 17.

The existence of a legal infrastructure of this nature is not a proof of the security of property rights, but it is an indication that not everything is chaotic, and if it served no purpose people would not have had recourse to it, nor were its limitations always so clearly a significant obstacle to growth. One of the most common was the imprecision in land titles, a frequent source of conflict. However, those conflicts occur in times and in regions of economic expansion, for example the sustained expansion of coffee cultivation from the 1860s to the 1920sFootnote 18.

An argument has recently been made that these conflicts could have reduced the possible offer of Colombian coffee by as much as half. I confess that I find that conclusion, the product of a process of deduction without any hard empirical evidence, precarious and improbable. It is well known that various conflicts accompanied coffee expansion, but I see no evidence that it was significantly slowed by supposed legal deficiencies, in titles or property rights. The law that governed its rhythms and limits was the law of supply and demand in the world market, not Colombian law and its supposed imperfections, and its application in the tribunals of Santander, Cundinamarca, Tolima and Antioquia (see Sánchez Torres et al. Reference Sánchez Torres, Fazio Vargas and López Uribe2010)Footnote 19.

I have doubts about the use of «surrogates» in the formation of their models by some recent historians of the Colombian economy. They are certainly aroused in the case of the speculations about the effect on exports of conflicts in the coffee zone in Sánchez Torres et al. (Reference Sánchez Torres, Fazio Vargas and López Uribe2010). For the variable «conflict», they use the disputes over land, mostly from the archives relating to baldíos, registered in LeGrand (Reference LeGrand1986). But those conflicts could have been large or small, and may or may not have affected the production and productivity of the place concerned: Le Grand's book and other sources indicate that the degree of conflict varied a great deal. Without much more about the nature of each dispute, it is unsafe and unconvincing to treat them as a uniform variable in a speculative exercise on the possible quantity of exportable product. Neither can one reach the conclusion from a careful reading of Le Grand that all these conflicts were lost by the small colonos, as is posited by the model in question: the coffee zones, as is notorious, contain many small and medium-sized producers — see the observations on land policies in Melo (Reference Melo2007, pp. 156 et seq.) in Ocampo (Reference Ocampo2007). It is also not clear whether the multiplication of small properties would have increased exportable production: a preliminary hypothesis would be that in the era of expansion it was favoured by a mixture of large and small properties. It is not shown whether the number of conflicts was so large, and the statistics confirm what is known from other sources, that they became more numerous and widespread in the 1920s and 1930s, not in the principal years of expansion, still less in the 19^th century. The classic contemporary work of Rivas (Reference Rivas1899), which the authors do not cite, does not mention conflicts in the coffee zones, although it does mention them in the tobacco-growing area of Ambalema; that is of course no proof that there were none, but if they had been so important an obstacle to production they would certainly have left more contemporary evidence, more protest and commentary, as did those of the 1920s. The authors pass over problems as important as lack of capital and shortage of labour, Colombian costs of production and transport, fluctuations in the world markets — all are sacrificed on the altar of a neo-institutionalist analysis –– to reach the conclusion that «without conflicts the production of exportables would have been at least the double of that observed.»

I do not think that the expansion of coffee suffered from failings in property rights; rather, with the multiplication of small and medium holdings, it can be argued that more Colombians came to enjoy such rightsFootnote 20. Nor had such defects hindered the earlier rise of tobacco cultivation — litigation over land does not figure prominently in that history, although there were difficulties in the enforcement of share-cropper contracts. The export of quinine bark caused conflicts in certain parts, particularly in Santander — «quina wars» in the baldíos, state lands that were the subject of competing claims. All the same, legal deficiencies did not hinder the rise of either of these exports, nor do they account for their decline: compared with other causes, their effect was marginal.

Economic historians of the institutional school ought to have more respect for what the great English jurist William Blackstone (Reference Blackstone1765-1769) referred to as «the robust title of occupancy». It was the title recognized for a certain extension of land in the laws on the appropriation of baldíos, and it is a notion that must be current in all frontier societies. Many colonizers were interested in legalizing, registering and perfecting their titles at a certain stage, but they did not necessarily make this a fetish, a sine qua non of their entrepreneurial activity.

Let us look at other possible insecurities. Two of them derive from legislation that was at least in part designed to encourage economic progress, that concerning Indian reserves, the resguardos, and that concerning church property. In both cases, property rights can be said to have been violated, although they were not individual property rights. Neither example seems to me of great importance in 19^th century economic development in general. A proportion, far from the totality, of the resguardos disappeared, without giving the positive impulse to rural development that Liberal theory expected and with what are commonly regarded as negative consequences for the indigenas concerned. The church properties in Colombia did not amount to so much, and so far it has not been shown that the desamortización had any marked effect on the economic activity of the countryFootnote 21.

The basic security of the proprietor is not to suffer robbery or expropriation. Some commentators think that robbery and expropriation were common features of the civil wars, but this is too simple.

Without doubt, the civil wars were in many ways damaging. It is not the argument of this essay to deny the generally negative effects of disorder, but to be more precise about the importance or lack of importance of those negative effects, and to examine critically the argument that disorder was a principal cause of the poor record of the Colombian economy in the 19^th century.

It is necessary to examine the evidence, much more closely than it has been examined by those who refer too lightly to «anarchy» and «turmoil», and evidence does exist, from various sources. Concerning basic property rights, as in all wars there were expropriations and robberies. But in the case of expropriations there were limits. The typical cases were of cattle and horses, for rations and for mobility. Seizing cattle was commonplace: «Viva la guerra, muera el ganado!» This caused no surprise, and hacendados would routinely hide their animals as best they could, as well as attempting to protect their peons from being recruited. All the wars reduced the stock of cattle, sometimes in certain parts drastically, as in the Llanos during the Thousand DaysFootnote 22. The practice was regulated, and the regulations set out in the Código militar. Expropriators were meant to give out chits, which would in theory be paid after the government victory. Papers that were not worth much, but were better than nothingFootnote 23.

All the same, the detail is significant in relation to property rights. In the republican era, it was neither licit nor common to expropriate real property. The lands and real property of Spanish royalists was expropriated in the wars of Independence, but this was not the practice in the later civil wars. There are examples of explicit prohibition of the practice, as in the Constitution of the Estado Soberano de Santander of 1859. Exceptions did occur, and some can be found in the war of 1877 — the general repudiation was strong enough to produce prompt reversal in the examples I have encountered, and the restoration of the properties concerned to their original owners. There is a detailed account of one such episode in that war, the expropriation and auction of some rural and urban Conservative properties in Pasto, carried out by Liberals under the orders of the Radical Cesar Conto, which excited general opposition and which was reversed on the orders of another Liberal, the less Radical General Eliseo PayánFootnote 24.

This type of expropriation could not gain acceptance as part of the rules of the game. In a country of such political instability, it was not advisable to go too far in chastising the enemy. Everyone in the upper and even middle reaches of politics was aware that the winner of today might well be the loser of tomorrow, and this awareness produced at least a degree of prudence.

What of expropriation in the guise of forced loans? In the Memorias of the Ministers of Hacienda who served during and after civil wars, one can study in detail the measures taken to meet the extraordinary expenses they occasioned, and among these measures the voluntary and forced loans, with the lists of names of the lenders, willing — speculative — and unwilling, and the sums involved against each nameFootnote 25.

There was some order in these proceedings. The section on the last civil war in the Memoria of 1878 begins with the rejection by the federal government of a petition from the Radicals of the Cauca — the first signature is that of General David Peña, a Radical of particularly fierce reputation –– demanding that a fine of three million pesos, to be spent on the Ferrocarril del Pacífico, be exacted from the godos of Antioquia, according to them those principally responsible for starting the war. The response of the Secretary of Hacienda was that this would be neither legal nor advisable, and that he deplored any notion of confiscation or collective punishment: what was important was to restore peace, not to exacerbate the feelings of the defeated with that sort of economic vengeanceFootnote 26.

Forced loans were a measure of fiscal emergency, the end being to get resources rapidly. The victims were commonly the disaffected and the opponents of the government — from its supporters the government expected support too, supplies in kind, cattle and horses, or voluntary loans that after hostilities ended would receive preferential attention. With the forced loans, the government offered discounts for prompt payment, up to half the sum demanded. At times, it announced that it was forbidden to seek reductions, but it appears that deals were done, above and below board, as was logical given the urgency of raising funds. The recalcitrant were threatened with having their goods auctioned off, or with imprisonment, although these threats were rarely carried out. In the list of names and amounts for the Cundinamarca forced loan in the Memoria of 1878, there are some exactions above $1,000, but the majority are much lower: for 1876, the average for the 113 contributors named was $336. The total raised up to the end of the fiscal emergency in July 1877, according to the marvellously precise accounts of the Memoria, was $132,350 and seventy centavos from the forced loans, $86,616 and twenty centavos from voluntary loans and from voluntary contributions $85,099 and forty centavos. Thus, in this part of the country, the government raised more from its friends than from its enemies (Bernal 1978, pp. 47-61).

Even without formal expropriation, did politics and conflict influence the lot of proprietors in the 19^th century, to a degree that had a marked impact on the development of the economy? This is not a problem that has received careful study. The behaviour of most substantial property owners in times of conflict was prudent — there is plenty of evidence to indicate that the majority tried not to get much involved. It is well known from studies of 20^th century episodes of violence that in certain parts property in land, particularly medium-sized property, was affected, but up to now there is little evidence from the 19^th centuryFootnote 27.

How this might have affected development is a further question. We should again recall that Colombia became one of the world's principal coffee producers between 1870 and 1920, a half century that saw four of the so-called national wars: 1876-1877, 1885, 1895 and 1899-1902.

Another approach is to examine the vulnerability of the different sectors of the economy, as has already been sketched for cattle. The principal exports of the country in the century were gold, tobacco, quinine and coffee. None of these was particularly vulnerable to disorder, at least in the medium or long term.

Let us look at mining. For the purposes of this analysis, the sector can be divided into three categories: the large Antioqueño mines and a couple of Tolima enterprises, the Cauca and Chocó mines worked by slaves, and small placer mining. The large Antioquia mines and the placer mining there and elsewhere did not suffer much from conflict. Antioquia in general suffered less than other parts of the country, and was a region resistant to militarization: recruiting troops there was notoriously difficult, partly for the good prospect of gains and employment in small mining. Marmato and Supia, mining districts on the frontier between Antioquia and Cauca, were regarded as less secure, but this did not have a decisive effect on their developmentFootnote 28.

The silver mine of Frías in Tolima, worked by an English company that employed several hundred men at the end of the century, likewise suffered little: work went on throughout the Thousand Days War, in part thanks to the protection it enjoyed from the Liberal guerrilla General «El Negro» Marín, previously employed by the mine as a capataz. Cauca mining declined with the emancipation of the slaves, part of the conflicts of mid-century, but it is doubtful that many economic historians will wish to place slave-owning under the heading of insecure property rights. A counter-factual — neo-institutionalist? — argument that Colombia would have developed faster by maintaining slavery is for many reasons implausibleFootnote 29.

An obvious conclusion: the vulnerability of enterprises varied. Some large Antioqueño mines had good defences: they were situated away from the usual foci of fighting, and were not usually threatened with having their workers recruited or with excessive exactions. Others may have suffered more. Perhaps a closer look might discover evidence such as Stephen Haber has found for Mexican mines during the Revolution, that what there interested governments and revolutionaries was a flow of resources, and for that reason they had no interested in ruining enterprises that could provide such a flow (Haber et al. Reference Haber, Razo and Maurer2003)Footnote 30.

Neither tobacco, nor quinine nor coffee supports the thesis that insecurity was the primordial obstacle that explains the slow and inadequate growth of the 19^th century economy.

The causes of the decline of tobacco were other: the loss of quality through the exhaustion of the tobacco lands, the competition of Java, the changing habits and tastes of European smokers … not civil war or anarchy. Nor was the fortune of quinine or coffee determined by the ups and downs of public order. The price of coffee, which did have something to do with those ups and downs, did not depend on what passed in Colombia, but on what was already a world market. The direction of causality, in the case of the war of 1885, as has already been pointed out, ran from the fall in exports to disorder, not the other way round, and the same can be seen in the antecedents of other wars, as in 1859 and 1899Footnote 31.

Coffee had a certain robustness, a natural capacity to survive in troubled times. Coffee trees do not die, and after periods of neglect can be pruned back into good shape. Coffee can be stored for long periods without any great loss of quality. The armies never destroyed coffee plantations in the way the ancient Greeks cut down the olive trees of their enemies. A war would interrupt exports and cause a shortage of labour, but would not be a complete disaster. After the War of a Thousand Days, once recovering prices provided the incentive, the Colombian coffee industry recovered rapidlyFootnote 32.

Another sector to be looked at is transport.

There were some episodes of spectacular damage done to this sector, symbolic damage, such as the destruction of the steamer Union in the Magdalena River in 1840, during the Guerra de los Supremos. All the same, the detailed studies show that steam navigation was established on the river without much delay or difficulty when the available cargoes justified the investment. This coincided with the export of tobacco, which provided the downstream cargo and the earnings to pay for the increased imports that came up. The river steamers were certainly exposed in times of revolution, and there were many lost — the Battle of La Humareda, in the war of 1885, was such a holocaust. But the steamers were surprisingly cheap, and were soon replacedFootnote 33. The wars also meant difficulties and increased costs in horse and mule transport, and further interruptions in the efforts to build railways. Here too my conclusion is to recognize the negative influence, but to deny it any great importance.

Commerce too can be viewed as a sector. Frank Safford, in his thesis on Bogotá commerce in the century and in his account of the house of Vargas Hermanos in Carlos Dávila's collection of entrepreneurial histories, shows how that business confronted the risks and opportunities of civil war (Safford Reference Safford1965 and Reference Safford2002)Footnote 34. There was the option of restricting activity and holding the profits abroad until peace returned. There was the speculative option: look for unusual gains by importing more in the knowledge that competitors were importing less. Vargas Hermanos instructed their English correspondents how to send the merchandize with an appearance of being foreign property. The house did not always do badly out of civil war. In addition, there was insurance: at one time of disturbance they concluded that they were over-insuring, and that they had overestimated risk. They were not carrying on business in such a primitive or anarchic context as might have been supposed. In the phrase of Joseph Conrad's Nostromo, «the conditions of civilized business» were not entirely absent.

Each commercial centre of any importance had its circle of merchants and businessmen, large or small, representatives of that international bourgeoisie of the 19^th century. As is well known, Colombia's external trade in the early years of independence was largely with Jamaica, and that then gave way to direct trade with Europe. New Granadans made their contacts with Manchester, with Paris, with Bremen … . They travelled for this purpose, or sent family members to serve a commercial apprenticeship. They established their reputations, their credit, in all senses of the word. This was an informal freemasonry, sometimes reinforced by the formal version. Abroad, the principal suppliers of the Colombian market became accustomed to the slow commercial cycle and adjusted their terms to its distant realitiesFootnote 35.

Trade was rarely or never completely interrupted, as the wars did not affect all the territory, all the roads and rivers, all the ports. There are some spectacular examples of open communication in times of civil war, one of them the successful introduction in the middle of the War of a Thousand Days of the heavy machinery for the country's first sugar ingenio, La Manuelita, which required an enormous train of bullocks, mules and men to transport it from Buenaventura on the Pacific to Palmira in the Valle del Cauca. Just as Colombian soldiers elaborated a specific art of warfare for the context, merchants and entrepreneurs also adapted to the milieuFootnote 36.

One valuable source for evaluating specific economic losses caused by civil war is the evidence from the claims made through diplomatic channels. They cover not only the losses alleged by foreign nationals, but on occasion also what are really the losses of native businessmen who have managed to cover themselves with a foreign flag. Some of them are for sizeable amounts, and from time to time they had painful consequences for the Colombian government: events in Panama could be expensive, as was the «slice of melon riot» and the burning of Colón by Pedro Prestán in 1885; likewise the assassination of some German merchants in Bucaramanga by the criminal-anarchist gang La Culebra Pico de Oro, and the long-drawn-out Cerruti claim. Nonetheless, the majority of claims were for small run-of-the mill losses. The British diplomats complained of their compatriots’ tendency to exaggerate their losses, and made it clear that they would not support excessive claims, nor claims that disguised what was really native property. Their tone is often much more sympathetic towards the Colombian government than towards the claimantsFootnote 37.

To round off this tour d'horizon of sectors, what of finance and banks?

First, and it is not exactly a sector, one has to consider the relation of public finances to disorder. In a simplified version, the argument is that persistent strife impedes imposing any order on the budget of the national government, with dire consequences for national development: successive crises of public order have among their results the consequence that the needs of the government keep interest rates high in the local market to the prejudice of local enterprise, and that they prevent any recourse to relief by borrowing abroadFootnote 38. They also restrict the government's capacity for carrying out essential improvements in the national infrastructure, upset the rate of exchange and at times even lead to the sort of hyperinflation produced by the War of a Thousand Days. The picture is of a vicious circle embracing the entire century.

How can one put this argument into a more precise, and perhaps more realistic, perspective?

The history of the foreign debt of the republic is one of a series of defaults and renegotiations before it acquired a degree of stability in the first decade of the 20^th century. A prominent participant in the operations of the 1820s, Colombia did not figure in the subsequent waves of 19^th century lending to Spanish America, as did for example Argentina and Chile. Was this a great disadvantage? The day-to-day needs of the government were not large, and it was easier, and perhaps more economical, to meet them in part through recourse to internal borrowing: the advantages were that the lenders were close at hand and the negotiations rapid; London was a long way away and financial dealings there took their time, as well as implying a high political risk, as such negotiations were always likely to be the target of wild criticisms and scandalFootnote 39.

A simple conclusion is that for many years Colombian governments preferred not to have recourse to borrowing abroad, as a result of fiscal weakness which in part, but only in part, derived from internal disorder.

A counter-factual comes to mind: imagine a Colombia peopled by those peaceful angels that Victor Hugo according to legend thought suitable citizens for life under the libertarian 1863 Constitution of Rionegro, a Colombia without the civil strife that caused defaults, and therefore enjoying better credit abroad: what would have been the result?

With its weak fiscal base, the government's capacity for borrowing would have remained low. Its entrepreneurial capacity, its ability to put such resources to profitable use, would have remained low as well, and it is doubtful that the citizens of the country would have regarded grand government initiatives with much confidence. Those Colombians who preached the virtues of laissez faire and minimal government now appear to me much less utopian, much more realistic, than it has been the custom to judge them.

It is not at all clear as to what would have been the great works of national progress that a 19^th century government might have undertaken. There was no general accord about priorities — that is shown by the eternal debate about the merits of the Ferrocarril del Norte, a large undertaking, defended as a national necessity, but one of dubious economic viability. One result of larger resources in government hands might have been a multiplication of white elephantsFootnote 40.

At best, one can imagine some anticipation of the progress of the 1920s, when the country did derive some advantage from the millions that came with the first really prosperous decade in half a century. But even with the new phenomenon of U.S. investment, those millions would not have come without the previous advances in the economy, particularly the expansion of coffee, and petroleum and bananas were essentially new commodities, of the 20^th century, not the 19^th centuryFootnote 41.

*********************

Let us return to the thesis of Fernando Garavito, Influencia perniciosa de las guerras civiles en el progreso de Colombia. It is a remarkable work for a student of 1897, still worth reading, of a genuine concern for his country and rich in suggestions, avoiding the temptation to find a single cause for its instability. He makes observations on the climate of the tropics, and in Colombia's defence notes that the Guayanas, «populated by Frenchmen, Englishmen and Dutch», have not progressed much either. He cites Rafael Núñez on the deplorable Colombian mania for taking French ideologies too seriously. Among his criticisms of paper money, he lists the excessive facility it gives the government for beginning ill-conceived undertakings. His catalogue of the evil consequences of civil war, moral and material, contains much that is true. He certainly deserved his degree.

The economists who read him will take a particular interest in his quantitative exercises, where he calculates the losses of the country through deaths and disabilities from wounds, from the interruption of productive activity, the cost of armies, the destruction of capital, etc. They are quite sophisticated exercises, and their full analysis and criticism would be a lengthy matter. One pernicious influence runs into another, and the total cost rises and rises, to reach for example $1,200,000,000 for the suspension of work alone. The reader ends up suitably depressed, as the writer intended (Garavito Reference Garavito1897, p. 40).

One principal reservation about this sort of exercise is that in the end one does not know where to stop. Why limit oneself to wars? Why not add diseases and epidemics, the cholera, the Spanish influenza of 1918, the plagues of locusts and the ill effects of the ocean current El Niño?Footnote 42 The other is that already mentioned, that if what we are trying to clarify is the progress or lack of progress of the country in a comparative frame, then the same exercises have to be carried out on the other countries of the comparison.

Some final reservations about the neo-institutionalists. The explanatory power of the role of institutions in economic development seems to me still limited, and I am confused by the tendency of some authors to extend the boundaries of the notion of institution wider and wider, from the formal to the informal, from something more or less well defined, such as the judicial system, to the constitution and its workings, to the nature of the peasantry at the time of Independence … to, according to one distinguished economic historian of my acquaintance, the country's geography and climate. For any of this to be manageable, it has to be precise, and it has to make clear its frontiers with other categories of explanation, such as the older and familiar «factor endowments»Footnote 43.

In the case of the Colombian economy and its evolution after independence, I am inclined to share as a starting point José Antonio Ocampo's observations on the peasantry of the highlands and the lack of mobility of labour, which incites me to go back to read those who wrote on the economy around the time of the republic's birth: Pedro Fermín de Vargas (Reference Vargas1944), Antonio Nariño (Reference Nariño1797), Francisco Silvestre (Reference Silvestre1950), William Wills (Reference Wills1831). Taken together, they describe very well the disadvantages then attending.

I remain sceptical about the connections between economic development, at least in the crude dimensions such as gross domestic product and the growth rate, and institutions. I do not see the connections so clearly, nor the direction of causality in the changes, on one side and the other. Many institutions, such as justice and security, the last the particular concern of this article, are desirable in themselves, irrespective of their influence or lack of influence on the economy. This tends to be ignored when institutions are looked at by economists. The old economic history, the «pre-institutional», still has a lot more to say about the achievements and frustrations of Colombia's 19^th century.