With a book of this scope, size and complexity, it is worthwhile outlining a brief plan for reviewing it. So I propose to consider first, its novelty; second, its overall argument; and third, its relative strengths and weaknesses. There is inevitable overlap between the three areas, but I hope to give prospective readers a clear accounting of each.

First, as to novelty, I am unaware of any comprehensive economic history of the Caribbean in any other language. One that integrates the performance of the Spanish, French, English and Dutch territories is nearly, by definition, inconceivable in the basic sense of that word. I have always admired the works of Franklin Knight, John Parry and Philip Sherlock, and Eric Williams, but while these treat economic matters, they are in no sense modern economic histories per se. Bulmer-Thomas himself made extensive reference to the Caribbean in The Economic History of Latin America since Independence (Cambridge University Press, 2003), but here, his account of the Cuban economy since the revolution alone (which, in reality, takes key data series back to the 1820s) amounts to far more. Haiti, largely terra incognita, adds to the richness of the account. Bulmer-Thomas' startling conclusion – that Haiti did not fall behind economically until the turn of the twentieth century – must obviously lead to a thorough reassessment of the Haitian Revolution (1791–1804) and its impact on the dominant plantation complex in sugar. Haiti and indeed all of the Caribbean grew more slowly due to the abolition of black slavery; ironically, then, the relative costs of the Haitian Revolution were reduced. The construction of the time series on which such work is based is arduous, involving both thorough synthesis and index number construction to whose assumptions the results may be quite sensitive. Running the numbers for one country over a brief period is hard enough; I cannot imagine doing so for the entire Caribbean since 1815. Fortunately, Cambridge University Press has afforded Bulmer-Thomas almost half as much space to explain his sources and methods (as well as a website, multimediafluids.uam.elogim.com/9780521145602, for the Statistical Appendix, which is accessed under ‘Resources’) as it has to discuss his findings. Do not take Bulmer-Thomas at his word: trust, but verify.

Second, as to argument, in The Economic History of Latin America since Independence, Bulmer-Thomas essentially argued that highly productive but small tradables sectors were swamped by the effects of large non-tradables sectors characterised by low productivity. Here the approach must be somewhat different: small open economies in the Caribbean had relatively large tradables sectors. If the overall export/GDP ratio in 1960 was 33 per cent (according to Bulmer-Thomas' data), it is scarcely believable that, in 1860, the ratio of tradables to non-tradables was any less than 50 per cent, and surely it was higher. As a first approximation, then, the tradables sector was the economy. An effective mechanism to transmit its productivity gains to the rest of the economy was less important than the relative prices of exports (the net barter terms of trade) and their purchasing power over imports (the income terms of trade), because these largely determined expenditure per head, which Bulmer-Thomas reasonably chooses as a proxy for output per head (or national income) before 1900. In the nineteenth century, Cuba, French Guyana, Guyana and Trinidad tended to do relatively well. While there was certainly considerable persistence in ranking into the twentieth century, Puerto Rico and the Dutch Antilles emerged with greater prominence. Cuba's performance was most strikingly inconsistent. Income per head there was exactly the same in 1958 as it was in 1903, after wild gyrations. Bulmer-Thomas accurately labels this ‘a shocking indictment of [US] development policy in and towards the island’. In fact, the burden and benefit of comparative colonialism in the Caribbean is never far from the surface of Bulmer-Thomas' narrative. It is a complaint going back to the Macmillan Committee (1931), if not before: the British understood how to run an empire, if not the world, but the Americans, alas, never quite learned.

Third, strengths and weaknesses. This book is a remarkable achievement, but it would be a pity if Bulmer-Thomas' conclusions were the last word, rather than more of a starting point. For example, I disagree with his handling of the question of the Jamaican economy, which fell out of the top quartile ‘income’ rankings in 1820, never to return. As Seymour Drescher has argued, the British pinned their hopes on a ‘mighty experiment’, the axiomatic efficiency of free labour after the abolition of slavery. In Jamaica the experiment was a failure, and its nemesis was Cuba, where slavery remained firmly in place. You do not have to be an apologist for chattel slavery to understand, after 40 years of argument, its economic efficiency. The Parliamentary Papers are simply full of this discussion, and the impact of the admission of Cuban sugar into Britain had remarkable effects on both economies. This was no secret. And, yes, Cuba prospered in the nineteenth century even though its net barter terms of trade deteriorated. But, as Bulmer-Thomas and British consular officials on the island understood, the productivity-corrected terms of trade improved. In this sense, Cuba was a lot like the US South, where cotton was in a similar position. The ingenuity of planter capitalists in a tight corner, however distasteful, is considerably more impressive than the acumen of socialist planners, however well intentioned. On the other hand, I could not agree more with Bulmer-Thomas in his evident dislike of the current fashion of using hurricanes as the deus ex machina for economic change in the Caribbean. Climate certainly has something to do with comparative advantage, but I sincerely doubt that the weather does. There is a difference.

I trust this provocative, erudite and well-nigh exhaustive history will be translated into the relevant languages of the Caribbean as soon as possible. And I hope that every student of Latin American history – not just regional specialists – will read it as well. It is a stunning achievement.