Many books have been written about the incorporation of the Caribbean region, South Asia, Africa and Latin America into the global economy. Remarkably, few have dealt with Island Southeast Asia or Maritime Southeast Asia as a macro-region. For the Caribbean nations, it has been amply discussed how the legacies of the plantation economies consisted of meagre economic growth and massive unemployment. Conversely, scant attention has been given to the question how societies in Island Southeast Asia were turned into providers of cheap commodities and how this impacted their long-term development prospects. This silence is even more remarkable considering some striking parallels with Caribbean socio-economic trajectories. Today, emigration of millions is the fate of Island Southeast Asia, as it is for the Caribbean region. To break the silence and to invite further discussion I wrote The Making of a Periphery: How Island Southeast Asia Became a Mass Exporter. After reading the review by Dr AguilarFootnote ¹ on this book in a previous issue of this journal, I felt that it could be worthwhile to highlight some of the main points of my argument about the peripheral integration of Island Southeast Asia in the global economy. I am grateful to the editors of the International Journal of Asian Studies for granting me the opportunity to do so.

No doubt, nations whose subordinate role it was in the global economy to produce cheap commodities were severely limited in their development potential and saw extensive migrations culminating in today's mass international emigration of people trying to escape from poverty. This happened in Island Southeast Asia as well as in the Caribbean region, amongst others. As Nobel Prize winner and Arthur W. Lewis, who was born in St. Lucia, has pointed out, the reason that even highly productive sectors within these peripheral economies pay low wages can be explained by the existence of “unlimited supplies of labour.”Footnote ² Interestingly, his diagnosis did not originate in the Caribbean, but as he explained himself, while he strolled the streets of Bangkok in 1952. Likely, he saw a bustling streetlife that barely concealed massive underemployment. It is appropriate to quote this great economist himself remembering his eureka moment:

Lewis' publication on the “unlimited supplies” met with both “applause and with cries of outrage” in 1954, which is over 65 years ago. Perhaps, today he would have elicited even more criticism as from a neo-classical point of view his thesis on the “unlimited supplies of labour” is heterodox, because it rejects the assumption that labour markets will move towards an equilibrium. However, his theory of dual economies and peripheralisation is part of the economic field. Lewis has left a lasting legacy, has inspired prominent economic historians such a Jeffrey Williamson and – relevant to this discussion – has inspired my book, as indicated by the word “periphery” in its title.Footnote ⁴

One crucial point made by Lewis is about demography and another about the lasting negative legacies of plantation economies. The concept of labour abundance is both a derivative from demography and a relative one as in rural societies it is invariably related to the availability of land. With little technological development, labour and land are the two factors of production that really matter. I used these very classical premises from development economics as a starting point to revisit theories that divide the world in macro regions according to their place and role in the world economy. Wallerstein's world-systems analysis is the most famous, but Daron Acemoglŭ applies the same broad, and one might say, homogenising approach. In their bestseller Why Nations Fail Acemoglŭ & Robinson explain global divergences from a historical perspective using a single theory pertaining to institutions regulating economic life.Footnote ⁵ I recognise and appreciate the crucial role historicity plays in their economic analysis. However, an unavoidable drawback of unifying theories is that these homogenise our understanding of complicated and diverse processes of long-term historical change.

Island Southeast Asia exhibits a variety of demographic patterns, and, in contrast to the Caribbean region, exhibits historically highly diverse labour regimes for commodity production ranging from slavery to Chinese kongsis, to haciendas, to extensive foreign-owned plantation conglomerates but also important sectors of independent smallholder production. Although no one will deny that colonial powers aimed to develop Indonesia, the Philippines and Malaysia as commodity producers with cheap labour, it still allowed for major subregional differences regarding colonial regimes and the concrete constellations of production, labour recruitment and migration. In this respect Luzon, parts of the Visayas, the western parts of the Malay Peninsula and Java with their large agricultural estates and mines experienced very different trajectories compared to, for instance, West Kaliamantan or the eastern provinces of Luzon where smallholder cash crop production of rubber, copra and abaca dominated.

I argue that plantation economies in Island Southeast Asia performed worse in terms of benefits for local populations than smallholder export production. It is a point that deserves further research, but it is a plausible one, for which it is not difficult to find evidence. Plantations are known for relying on cheap labour. This comes as no surprise as they usually must compete in global markets largely on the basis of price. For the Caribbean we know all too well where this labour came from: millions of Africans were kidnapped, enslaved and transported across the Atlantic Ocean to produce sugar, tobacco and other crops for Europe and America. In Malaysia, plantations and mines imported Chinese and Indian labour on a massive scale. But, for the Philippines and Indonesia it was natural demographic growth that guaranteed abundant labour supplies, as my book amply demonstrates.

I concluded that the relatively successful smallpox vaccination in Java, Luzon and parts of the Visayas in the early years of the nineteenth century was responsible for a population growth that was unique for the region. De Bevoise suggests that smallpox vaccination in the Philippines compared badly to Java, but my contention is that where Spanish control was relatively tight in Luzon and the Visayas, the results were comparable with those of Java.Footnote ⁶ The vaccine resulted in a precocious demographic growth of over 1.5 percent per annum. Together with a stagnant manufacturing sector and declining agricultural productivity, this created the abundant labour supplies for the developing plantation economies. I was somewhat surprised to see that Dr Aguilar did not point out for the reader that my point is not entirely uncontroversial.

However, high-population densities alone cannot explain the rise of plantation systems. Invariably, I conclude in my book, locally existing dependency relations facilitated the recruitment for plantation estates. We see in Java that the coerced Cultivation System, which was introduced in 1830 and was gradually phased out from the 1860s onwards played a crucial role in creating the powerful Java sugar industry. This system was crucially based upon the collaboration of the local aristocracies. After the Cultivation System had been phased out in the 1860s village elites supported the plantation economy in their role as labour recruiters and by forcing villagers to rent their land to plantations. They shared in the profits for each worker and for each piece of land they managed to deliver to colonial plantation economy.

Many of the plantation economies in Southeast Asia were based upon local systems of labour recruitment, often containing elements of dependency on patrons. These could, for instance, be the leader of the labour gang of cane cutters, or hacenderos in Negros, or wealthy farmers in Java. Such dependencies were much more efficient for the producers than coercive systems such as slavery or indentureship. These dependencies could also sustain substantial migration systems. In Java, Luzon and the Visayas we see huge migration flows of impoverished rural labourers towards sugar-producing regions in the nineteenth century. I have reconstructed the size of these flows and the data are available online (Bosma Reference Bosma2017).

Within Island Southeast Asia, Java, Luzon and parts of the Visayas stood out in terms of population density and increasing availability of labour. For most of the immense space of Island Southeast Asia, labour recruitment required extensive investments. In this thinly populated part of the world, slavery was the dominant mode of labour recruitment, probably until the mid-nineteenth century. Here, I agree with James Francis Warren who has argued that this persistence of slavery, and slave raiding, was not the result of economic decline but very much the result of incorporation into the global economy.Footnote ⁷ Meanwhile, Chinese self-governing units of labourers, kongsis, who extracted tin, gold or diamonds, cultivated gambier, pepper, pineapples and milled cane had developed their own systems of labour recruitment. Over the course of the nineteenth century, the Chinese migration systems developed into massive trafficking of workers from China. Through indentured labour contracts British and Dutch colonial authorities both tried to regulate and encourage these flows on behalf of plantations and mines driven with European and American capital.

Although systems of dependency in labour-abundant rural societies in Island Southeast Asia included neither slavery nor indentured labour, we see a continuation of slave-based commodity production and the introduction of indentured labour contracts by the colonial governments of the Netherlands Indies and British Malaya, in the less densely populated parts of those regions. At this point the Nieboer–Domar thesis comes in view, postulating that land-abundant and thus labour scarce economies tend towards extra-economic coercion, be it slavery, serfdom or indenture. I also submit, however, that indentured labour contracts were far less effective at bringing labour to plantations and mines than the more informal networks held together by patron–client relationships, which were often undergirded by debt bondage. Throughout Island Southeast Asia the most effective and cheapest way to recruit labour was via existing patron–client networks. One might disagree with this or not, but this is the point I am making.

Although much of Island Southeast Asia became involved in export commodity production at the turn of the twentieth century, the foreign-owned plantation or mine was far from the dominant mode of production. In fact, the region exhibits an immense variety in population densities, labour regimes and penetration by colonial powers. One of the explanations I cited in my book is that smallholder production usually outcompetes plantation production and hence that plantations could only exist with state-backing. Rubber was a case in point, but even in the case of sugar this argument applies. Furthermore, not all plantations are in the hands of foreigners and in that respect the sugar conglomerates of the Philippines and Java, for example, differ fundamentally. As I argued, both Javanese and Philippine workers in these sectors were badly paid and malnourished, but in terms of remittances abroad it made a difference, as most of the sugar factories that had been refurbished with American capital had come into Philippine hands by the 1930s. We can outline a similar contrast in the New World. Both in Pernambuco and the British Guiana cane workers suffered in the 1920s and 1930s, but Pernambuco was part of country that had already been independent for about a century, whereas most of the profits made in Guiana ended up with Booker McConnell in London. I think my book convincingly shows that the concepts of peripheralisation and colonialism are distinct and as such the words should not be used interchangeably. I think this point is somewhat missed in Dr Aguilar's review.

In Luzon and Java plantation economies were successfully embedded in existing agrarian systems. Local elites played a crucial facilitating role in this transformation of existing agrarian and taxation systems for colonial export production. Dr Aguilar suggests that I am utterly exaggerating the role of patron–client relationships in the peripheralisation of Island Southeast Asia. I understand very well why we should be careful using patron–client relationships as an explanation, since so much, and perhaps too much, has been written about clientelism in the Philippines. But that is not a good reason to dispose of this notion, it should instead oblige us to unpack it and try explaining what it is, why it exists and what it does. It exists because of the desire for social security in agricultural societies that are ecologically vulnerable. In these societies it makes sense that the economically more vulnerable look for protection with the stronger elements in their societies. Dependency is exacerbated by the existence of high interest rates in rural societies, which engender indebtedness. In two ways patron–client relations were key to the functioning of sugar plantations in the Visayan Island of Negros, for example. This was first, to keep labour tied to the hacienda and second, to hold labour gangs together.

My explanation for how Island Southeast Asia became a mass exporter first of commodities and subsequently of labour would be rather different from what Acemoglŭ and Robinson suggest in their Why Nations Fail for the simple reason that colonial relationships and commodity production allow for widely divergent trajectories of labour relations and economic development. Here again, I want to stress the distinction between plantation and smallholder production. In the latter case far more money stays within the local economies, as for example Lynn Hollen Lees has argued for British Malaya and Bambang Purwanto for Sumatra.Footnote ⁸ The point I made is that Java suffered from economic stagnation and its population from malnutrition at the peak of the colonial plantation economy. Per capita income lagged most other parts of Island Southeast Asia, where independent peasants produced rubber, copra or coffee for the global markets. Once Indonesia and Malaysia had become independent nations in 1949 and 1965 respectively, their governments branded plantations colonial institutions and encouraged smallholder cultivation. They did so for a perfectly sound reason: to ensure the revenues would benefit local development. Unfortunately, this decolonisation was never completed. Palm oil, one of the world's most important tropical commodities, has been a driving force in the establishment of new plantation regimes in Indonesia and Malaysia, which are the world's first- and second-largest producers of this commodity.

Over the past few decades, we have seen the return of appalling coerced labour conditions that were supposed to have been buried alongside colonialism. Palm oil plantations cause not only grave ecological damage, but also serious human rights violations. The routes and methods of labour recruitment have their roots in colonial times and sometimes even in old practices of enslavement. Patron–client relationships play an important role, as they provide the resources to enable migration, but also keep the migrants in a condition of dependency. Although in the past the commodities were a crucial source of colonial revenue for which labour had to be mobilised, today the export of labour has become the source of revenue, and all the attending practices of coercion, deceit and exploitation – all the human rights abuses – have been extensively documented. It involves several millions of people in Island Southeast Asia, although we will never know precisely how many, because much of this migration has been undocumented or based upon overstaying their visas. Dr Aguilar suggests that I overstate the seriousness and scale of the abuse of vulnerable labour migrants. I don't. The sources I quote in my book tell both the stories of widespread abuse and how migrants have tried to circumvent and resist abusive systems.

The peripheral position of Southeast Asia in the world of today is the result of a long-term development, as many scholars from Immanuel Wallerstein to Daron Acemoğlu have pointed out. But high demographic growth and local systems of labour bondage are much neglected elements in the making of a periphery. My intention was to write a book that invites us to rethink the geography of colonialism and the role plantations have played in economic marginalisation.