Volume 9

Patterns and Prospects of Haitian Primary Exports

Yves Bourdet and Mats Lundahl

Bourdet, Yves and Lundahl, Mats, Latin American Issues [On-line], 9.
Available: https://sites.allegheny.edu/latinamericanstudies/latin-american-issues/volume-9/


Yves Bourdet received his Ph.D. in Economics from the University of Lund, Sweden, and is a Research Fellow of Economics at the same university. He is the author of International Integration, Market Structure and Prices (1988) and Welfare (1991). His research interests focus on international trade, development economics and industrial economics.

Mats Lundahl also received his Ph.D. in Economics from the University of Lund, and is a Professor of Development Economics, and Chairman of the Department of International Economics and Geography at the Stockholm School of Economics. He is the author of Peasants and Poverty: A Study of Haiti (1979), The Haitian Economy: Man, Land and Markets (1983), Unequal Treatment: A Study in the Neo-classical Theory of Discrimination (1984), Incentives and Agriculture in East Africa (1990), and of numerous articles on economic development and discrimination. He is currently working on projects on Lesotho and the Southern African Customs Union, the role of the state in Haitian underdevelopment, and religion, economics and politics in the Dominican Republic.



Over the past few years, Haiti has displayed a negative growth of GDP and increasing deficits in its trade balance.1 These deficits have been financed by international aid (60 percent) and loans in the international financial markets. The foreign debt amounted to no less than US$ 600 million in 1986.2 During the same period, the Haitian population has continued to grow at a rate of approximately 2 percent per year.3 The result has been that per capita income dropped by 15 percent between 1980 and 1985.4 One of the reasons behind the negative growth and the worsening of the trade balance in Haiti is the decrease of primary exports.

A large part of the exports of developing countries originate from the primary sector of the economy. Agriculture, fishing, forestry and mining constitute the main components of this sector. The dominant place of primary exports in total exports from developing countries implies that their role may be of considerable importance for the development process. The failure of industrialization cum import substitution policies in promoting economic development to a large extent justifies this focus of attention. The purpose of the present work is to illustrate the role of primary exports for development and to analyze some of the possible factors behind their decline in the context of Haiti. We will also examine the future prospects of Haitian primary exports.

The analysis is divided into four parts. Part one examines briefly the contribution of primary exports to the process of economic development. Part two studies the composition of Haitian primary exports and its changes over time. It concentrates in particular on the last two and a half decades. These changes are partly the result of the government policies that have been in force since the Second World War. Part three analyzes the effects of these policies. This section also scrutinizes other constraints on primary exports in Haiti. It discusses the domestic factors that act on the supply side and prevent primary exports from expanding at a higher rate, but it also focuses on the non-domestic constraints that affect primary exports via the demand side. In the final section, we analyze the prospects for development of primary exports from Haiti and provide some policy recommendations.



The emphasis on industrialization and import substitution policies as the motor of development to a large extent is responsible for the lack of attention given to the primary sector in many developing countries. The fear of being dependent on a limited number of primary export products, the price of which is set in the world market, is another reason behind this lack of interest in primary products. The difficulties faced by developing countries when it came to selling their primary products during the Great Depression and obtaining manufactured goods during the Second World War also contributed to this attitude.5 Some of the recent economic discussion, on the other hand, has emphasized the role of the primary sector in economic development and one of the main conclusions drawn is that economic development is hardly feasible without the help of the primary sector.6

The contribution of the primary sector can take several forms. With the exception of minerals, the vast majority of products of the primary sector are labor-intensive. This means that developing countries with an abundant supply of labor have a comparative advantage in the production of such goods. If no obstacles to trade exist, it is thus likely that such products can easily be exported by developing countries for use by consumers or industries in other nations, especially industrial countries. Primary exports provide the economy with foreign exchange. This is particularly crucial at early stages of development when the industrial sector is still small. Foreign exchange is needed for the purchase of imported capital equipment necessary for the development of the manufacturing industry. Exports of primary products by countries which have a comparative advantage in their production thus permit the relaxation of foreign exchange constraints on the process of economic development.7

An increase of primary exports from an industry is very likely to induce an increase in the production of other industries delivering inputs to it. In the terminology of Albert Hirschman, such effects on the economy of a change in primary exports are called backward linkage effects.8 These are presumably more sizeable in agriculture than in mining (or other activities of an ‘enclave’ type) where a non-negligible part of inputs are imported from industrial countries.

Increased primary exports also affect the economy through their impact on incomes and taxes, i.e. through consumption and fiscal linkages.9 The former refer to the production of consumer goods for factors employed in the export sector and the latter to investments financed by export taxes. Consumption linkages could be substantial whenever the primary sector is large in terms of employment, and fiscal linkages are potentially sizeable due to the high rates of taxation on primary exports which exist in many developing countries.10 The concept of fiscal linkages takes us directly to the wider role of primary exports as a source of capital formation. Not only could an increase in primary exports result in larger tax revenues that may be used for investment purposes, but the increased incomes that a better export performance is likely to give rise to should contribute to capital formation through their impact on private savings as well, provided that suitable investments and a reasonably well functioning system of financial intermediaries exist.11

A sixth contribution that primary exports can make to the development process concerns employment. The large and rapidly growing labor force in most developing countries cannot be absorbed by their small industrial sectors in the foreseeable future. In Haiti, as we already know, the yearly rate of growth of the population is around 2 percent, meaning that with a time lag of some 12-15 years the labor force grows at about the same rate. This implies that the agricultural sector will remain the major source of employment at least until the end of the century. Due to the labor-intensive nature of production activities in the primary sector, an increase in primary exports can partly relax the employment constraints faced by developing countries.

Less clear-cut advantages emerge from activities, such as mining, which are of the enclave type. The main reason for this is the limited backward linkages which result from the fact that inputs, other than the mineral itself, are in most cases imported from industrial countries. A second reason is the capital-intensive production techniques which are usually employed in such activities and which reflect the factor price relations prevailing in industrial countries. This naturally suggests that the contribution to employment of the enclave type of activities is limited.

Finally, in the context of Haiti, there also exists a specific reason for developing primary exports. The AIDS disease and the alleged role of Haiti in spreading it has resulted in a sudden drop in foreign visitors and a drastic diminution of the receipts from the tourist industry since the early 1980s.12 Increased primary exports can provide the Haitian economy with foreign exchange and partly mitigate this new constraint on economic development.



The prosperity of Saint-Domingue during the eighteenth century was based on exports of primary products such as coffee, sugar, cotton and indigo. Later on, sisal and cocoa were added to the list of agricultural exports as well.13 Coffee, sugar and sisal have remained important export products during most of the present century. Indigo, on the other hand, practically disappeared as an important item of foreign trade during the wars of independence between 1791 and 1803,14 and cotton was virtually wiped out as an export product during the nineteenth century.15 However, cotton exports recovered between the two world wars. In 1935, they accounted for one-fifth of Haiti’s foreign exchange.16 A shift in relative prices in favor of food crops, such as corn and red beans, contributed a great deal to the rapid decline in cotton production after 1950.17 Exports decreased faster than output. By the mid-1960s, exports ceased altogether and Haiti became a net importer of cotton fiber. Cocoa, finally, became a leading export crop only at the very end of the nineteenth century.18

During the last three decades, the composition of Haitian exports has changed considerably. Table 1 presents the development since 1950. As can be seen, the share of primary exports in total exports has decreased over time. It dropped from some 60-70 percent during the 1950s to some 35-50 percent over the past decade. Two factors explain this decline: a decrease in primary exports in absolute terms and an increase in secondary exports.

Table 1

Historically, the main export product of the primary sector has been coffee. As indicated by Table 1, coffee accounted for around 60 percent of Haitian exports from 1950 up to the mid-1960s. This share dropped to about 45 percent in the 1960s and to 35 percent during the 1970-85 period. Sugar was second in importance during the 1950s. It was replaced thereafter by bauxite which made its appearance as an item in the list of Haiti’s primary exports in 1957 and which accounted for 9-16 percent of total Haitian exports during the 1960-82 period. Sugar and bauxite, however, virtually disappeared as export products in the early 1980s.

Three other primary export products are sisal, essential oils and cocoa. As shown in Table 1, the share of the former in total exports has decreased rapidly over time. It amounted to no less than some 20 percent in the early 1950s, but represents a negligible share of Haiti’s foreign trade nowadays. The share of essential oils in Haitian exports has increased somewhat over time and amounted to about 5 percent during the 1970-85 period. (It should be noted that essential oils are not primary exports in the strict sense of the word since they are subject to an industrial transformation process. However, in the present work, we will regard them as primary products because of the the very simple nature of this process.) The share of cocoa decreased during the 1960s but recovered thereafter. Cocoa accounts for around 3 percent of Haitian exports today. Finally, other primary products, such as molasses and cotton, represent a residual share of Haitian exports.19

A striking feature of Haitian primary exports is their changing composition over the past three decades. Whether this reflects changing patterns of comparative advantages is a moot question which, for lack of data, we will not deal with presently. In the next section, we will, however, discuss to what extent fluctuations and trends in individual products can be related to a certain number of factors that affect the supply and demand sides of primary product markets.



The main conclusion emerging from the previous section is that there has been a decline in the share of traditional export crops such as coffee, sugar, sisal and cocoa in Haiti’s foreign trade. It was suggested that this feature reflects an increase of manufactured exports and a decrease (or stagnation) of primary exports. The factors behind this decrease (or stagnation) are likely to vary both from product to product and over time. The purpose of the next section is to examine the main products individually and provide a tentative analysis of the factors which may explain the poor performance of the Haitian primary sector. (Unfortunately, the quality of the available data does not allow us to perform any econometric tests. Thus, the discussion should only be seen as an indication of probable explanations – not of the order of magnitudes.) The analysis deals with both domestic and international factors.

A. The Rybczynski Effect: Output at Constant Prices

When analyzing the development of Haitian primary exports over time, two types of distinctions are essential. The first one is between the development of the total output of the product in question and the share of exports therein. The second is between those changes of total output that take place at constant relative prices and those that are the result of changes in relative prices. We will come back both to the former distinction and to the development of relative prices below, in the context of the individual export products. Let us, however, first deal with the less intuitively obvious question of what will happen to the production of primary exports over time when the price of these items in relation to other products grown by the Haitian peasants remains constant.

One of the most powerful theorems in the theory of international trade is the Rybczynski theorem.20 This theorem deals with what happens with output when prices do not change but the factor endowment of the economy does. Assuming that peasant production can be divided into two components – export products and subsistence products – and that the production functions for both types of goods are linearly homogeneous, i.e. that for the sector as a whole no advantages or disadvantages of scale exist,21 assuming furthermore that only two production factors, in the present context land and labor, are employed and that competitive conditions prevail in the economy, the Rybczynski theorem states that if the endowment of one of the factors grows, the output of the product using this factor intensively will grow as well in absolute terms while the output of the other product decreases.

In the Haitian context, the interpretation of the Rybczynski theorem is the following.22 Over time, the population grows at approximately 2 percent per annum. With a certain time lag – say fifteen years – this translates into an increase in the labor force as well, and this increase, to the extent that it is not absorbed by the non-agricultural sectors of the economy or is lost through emigration, has to go into the agricultural sector. There, consequently, the output of relatively labor-intensive products will increase, while that ofland-intensive goods falls. The reason is that if commodities and factor prices are not allowed to change, we cannot change relative factor intensities either. Thus, the addition of the labor force must go into the relatively labor-intensive line of production (no factor intensity reversals are assumed to exist) but there it must be combined with some land too and this land can come only from the land-intensive line of production. Thus, the former expands, while the latter contracts. All that remains is to identify the ‘labor-intensive’ and the ‘land-intensive’ products, respectively. The available evidence indicates that the former are subsistence goods (corn, sorghum, beans and peas, rice, sweet potatoes, manioc, malanga, pigeon peas) while the latter are export crops (coffee, sugar cane, cotton, cocoa).23

Thus, as a result of the growth of the population and the labor force, over time there is a tendency for the output of agricultural export crops to fall in absolute terms even if the price of these crops does not change in terms of subsistence crops. This tendency is further reinforced by the fact that increased production of labor-intensive crops is prone to increase the rate of erosion, i.e. land is destroyed in the process, which in turn means that the Rybczynski theorem can be applied ‘backwards’: when the agricultural land area shrinks, the production of export crops contracts and that of subsistence crops increases. Thus, the cultivation of subsistence crops extends upwards on the steep mountain sides and since subsistence crops are not perennial, the land has to be laid bare at certain times of the year, for sowing and planting, which happens to be when the rains come – with the result that the topsoil is washed down.

In this way, the sequence acquires cumulative characteristics. Population growth leads to erosion, which in turn leads to further erosion (even without further population growth), and throughout, the output of export products falls. This sequence is on average valid for all the agricultural products discussed in the following. Thus, it will not be repeated below, but the reader should keep the mechanism in mind. This ‘Rybczynski effect’ may combine with or run counter to the effects caused by relative price changes, depending on whether the latter on average favor subsistence or export crops.

With this, let us turn to a discussion of the individual products.

B. Coffee

Historically, coffee has been the main export product of the Haitian primary sector, accounting for some 65 percent of total exports in the 1950s. However, the trend has been a downward one. During the first half of the 1980s, the share of coffee in total exports was a mere 30 percent, on average.24 Table 2 shows that the volume of coffee exports has declined both as a percentage of total output and in absolute terms. Which, then, are the factors behind this decline?

Table 2

The decrease of coffee exports can in principle be due to two reasons: a) a reduction of output, given the export share, b) a reduction of the export share, given the level of production. As indicated by Table 2, between the beginning of the 1950s and the late 1960s a contraction of output took place, followed by an increase during the next fifteen years, almost to the levels prevailing in the 1950s and early 1960s. The exported volume, on the other hand, shows a more or less decreasing trend throughout the period, whereas domestic sales have increased.

The thirty-seven years under consideration are not homogeneous. While both production and exports fell up to the beginning of the 1970s, the last two decades have been characterized by a relatively stationary output coupled with first an increase and then a decline of exports. This is the pattern we have to explain.

Beginning with exports, a potential reason for the declining trend is the commercial policy of Haiti’s main trading partners. Table 3 shows that coffee exports from Haiti are very concentrated and that the EEC countries and the USA are the main purchasers of Haitian coffee.25 Could it be that the trade policy of the United States and Western Europe caused the decline in Haitian coffee exports? This is hardly the case. The free market has by and large prevailed in international trade in coffee over the past two decades. All exports from Haiti enter the United States duty free.26 Access to the markets of other industrial countries was improved as a result of the Kennedy Round at the end of the 1960s and the beginning of the 1970s.27 Today, industrial countries impose a low tariff on coffee imports which amounts to no more than 6.8 percent for green and roasted coffee and 9.4 percent for processed coffee.28 Furthermore, only 11 and 17 percent, respectively, of imports are subject to non-tariff barriers in industrial countries.29

Table 3

Thus, as it appears, other factors than trade policy are responsible. These factors are the Rybczynski effect and the development of relative prices. When it comes to the development of the overall output volume, as we already know, the Rybczynski effect tends to hold coffee production back at constant prices. However, as indicated by Table 2, the difference between the first and last periods is a mere 4 percent. Thus, price factors must have counteracted the Rybczynski effect, at least since the late 1960s, since the population and the labor force have been growing throughout the entire period.

Table 4 summarizes the relative prices that bear on the production decisions of the coffee farmers. The decision to produce coffee and sell for export purposes rests with the growers. In standard economic theory, given the factor endowment of the sector, this decision is assumed to be governed, among other things, by the relative prices prevailing between coffee and other goods that the grower could produce. In the Haitian context, it is, however, sometimes argued that the price mechanism works only imperfectly, if at all, in the case of coffee.30 The peasants are tied to the intermediaries by means of debts and credits. Even if this should be true,31 it does not, however, mean that changes in relative prices do not affect the output mix. The main alternative to growing coffee is to grow subsistence crops that can also be marketed domestically, like corn or beans. Furthermore, coffee is sold, not only for export purposes, but in the domestic network of market places as well, for consumption within the country. This gives the coffee producers a twofold production option: coffee or subsistence crops, and, in the case of coffee, a twofold marketing option: for exports via the chain of spéculateurs and exporters, or domestically, via the internal market system.32

Table 4

By the same token, three sets of relative prices are of importance: those between exported coffee and subsistence crops, and domestically marketed coffee and subsistence crops, respectively, for the production decision and that prevailing between exported and domestically marketed coffee, for the sales decision. Looking at Table 4, we find that relative prices moved in favor of subsistence crops until the first half of the 1970s. By and large, the f.o.b. price of coffee showed a downward trend.33 At the same time, export taxes rose as a share of the f.o.b. price.34 Intermediary gross margins, on the other hand, did not rise until the mid-1960s.35 This, together with the Rybczynski effect, explains the downward trend in output. Furthermore, we observe that coffee prices moved in favor of exports until the end of the 1960s, which means that during the first part of the period under consideration, the decline in exports appears to have been due basically to the contraction of output, and not so much to a shift to domestic sales.

Figure 1

In the mid-1970s, the international coffee price increased drastically, as a result of frosts in Brazil (cf. Figure 1). As Table 4 shows, these increases were translated into an increase in terms of subsistence crops for the Haitian producers as well. Taxes were reduced, following a tax reform in 1973,36 while the gross margin of the intermediaries rose substantially.37 This, according to theory, should lead to an increased coffee production, ceteris paribus, and since at the same time the price of exports fell in relation to domestically marketed coffee, to decreased exports.

That, however, is not quite what we observe. Total coffee output fell during the latter half of the 1970s, meaning that the Rybczynski effect must have been stronger than the price effect. At the same time, exports increased and domestic sales decreased – contrary to what the theory predicts. Why this is so is not clear at all, but perhaps one should not make too much out of what after all must be considered as very shaky statistics, in particular as far as the price of the domestically marketed coffee is concerned. The prices of exported and domestically marketed coffee are not perfectly comparable, in that the latter contain a marketing cost component which the former do not, i.e. they are not net of costs:

A peasant may also sell coffee on the local market which generally brings a slightly higher price. This entails drying the coffee and pounding it, transporting it to the market and investing a certain amount in labor time to sell the coffee at the marketplace. Average daily sales may be quite low, particularly during the harvest when coffee is abundant, and thus it becomes uneconomical to invest days to sell coffee at a slightly higher price when the peasant could easily sell it in just a few minutes to the speculateur at a slightly lower price.38


It is worth noting that very often, coffee that is sold on the local market has been bought from a speculateur by a Madam Sara or other marchand to sell in the local marketplace. This is one explanation for the high price of local coffee – it has two middlemen extracting profits before final sale to a consumer.39

Thus, if marketing costs have increased, as they should, considering the strong overall increase in the price level in the latter half of the 1970s, with altogether 57 percent from 1975 to 1981,40 we also have an explanation of why exports did not fall and domestic sales did not increase: the figures in the last column of Table 4 are partly spurious.41 Also, the price quotations for the domestic market are difficult to interpret since they presumably refer to coffee of a quality which is both varying and not directly comparable with that of exports. For coffee of a given quality, disregarding transport costs etc., there should be no difference between the prices in the two markets.42

The conclusion, thus, is that a) to the extent that the price of both exported and domestically marketed coffee rose over that of subsistence crops the Rybczynski effect was overcome and b) to the extent that domestic marketing costs rose there was a tendency to switch from domestic sales to exports, during the latter half of the 1970s.

So far, it is possible to construct a coherent picture, with empirical data and economic theory matching each other reasonably well. When it comes to the first half of the 1980s, on the other hand, we run into problems. What the statistics tell us is that production and domestic sales increased while exports fell, both absolutely and in relation to domestic sales. The price of exported coffee fell in terms of subsistence goods. The world market price fell from the high levels of the second half of the 1970s. Unfortunately, no good data on domestic prices have been published so far, but the only movement compatible with the logic of producer choice based on relative prices is an upward one. If the domestic price went up, presumably mainly as a result of the increase of the population and income in the capital,43 this explains a) why production increased – the domestic price should have increased more than the price of subsistence goods and enough to overcome the Rybczynski effect, and b) why there was a shift from exports to domestic sales.44

This conjecture, however, leaves us with the problem of why the domestic price should diverge from what the producers receive when they sell for exports. It is indeed difficult to think of such reasons. Why should there be a sustained disequilibrium in the coffee market? There is a puzzle here, which for the time being must remain unsolved, unless we want to take the view that the production and domestic sales statistics upon which the reasoning is based are false, so that what happened, after all, was that total output, exports and domestic sales all declined. There are some foundations upon which such a conclusion could be based. When it comes to coffee volume statistics, only export quantities are reliable, whereas “the situation is very different when it comes to output because no systematic survey has ever been carried out, even less so on a consistent yearly basis.”45Thus, it could very well be the case that if we attempt to provide an explanation of the pattern in Table 2, we are analyzing a sequence of events that never took place. For the time being, for lack of information, we ought to remain agnostic. More research on both production volumes and the functioning of the domestic market for coffee is necessary.

To conclude, coffee exports from Haiti have declined over the period under consideration here. In the 1950s and 1960s there was a strong trend away from coffee production and exports, towards increased production of subsistence crops – as a result of the Rybczynski effect in combination with relative price movements. Thereafter, the picture is more complex, probably to a large extent due to the lack of reliable statistical information. The second half of the 1970s saw increased exports, quite probably mainly as a result of higher world market prices and lower taxes, while in the first half of the 1980s exports fell again, for reasons that are not perfectly transparent.

C. Cocoa

The interest in cocoa production was very limited in Haiti during the colonial period. However, cocoa became a leading export crop in Haiti around the end of the nineteenth century and the beginning of the twentieth century as a result of rising world demand. Cocoa exports thereafter declined up to the mid-1960s. Table 1 shows that their share of total exports decreased up to then and increased subsequently. This is further illustrated by Table 5 which shows that the volume of cocoa exported decreased during the 1950s and 1960s and increased substantially during the second half of the 1970s and the beginning of the 1980s. Unlike for coffee, it is impossible to calculate the export-production ratio for the whole period because of the paucity of production data and the lack of internal consumption data. The production data reported in Table 5 are mere guesstimates by the Interamerican Development Bank (1950-70), the International Cocoa Association (1971-79) and ECLA (1981-85) that do not seem to contain any trend. Not much attention should be paid either to production or domestic sales data.46 Presumably, the scope for domestic sales is still small. “The internal market at present [1984] appears weak”, reports Cristanna Cook in a survey of Haitian cocoa problems.47

Quite probably, the declining trend in exports from the early 1950s to the beginning of the 1970s was a result of the Rybczynski effect in combination with an unfavorable price trend vis-à-vis subsistence products. The world market price of cocoa was lower during the latter half of the 1960s than at the beginning of the 1950s, while simultaneously the price of subsistence crops rose.48 At the producer level, the situation appears to have been even more unfavorable. The Haitian government created a cocoa monopoly, the Haitian Manufacturing and Specialty Company, HAMASCOSA, in 1960, which paid low prices (1.75-2.25 gourdes per pound) to the producers. This situation was improved after the abolition of HAMASCOSA in 1978 and the world market for cocoa simultaneously improved, as shown by Figure 2. It may be conjectured that this turned relative producer prices in favor of cocoa (a producer price of 2.5 gourdes, 1977-78 and 3.75-4.5 gourdes, 1979),49 to an extent which was large enough to overcome the Rybczynski effect. Between 1984 and 1986, cocoa export prices dropped with 54 percent which made exports drop from their 1984 peak value.50

Table 5

Taxes, on the other hand, appear to have been of little significance. During the period under consideration, three taxes were imposed on cocoa exports: an export tax, a special or supplementary tax and an administrative tax.51 These taxes represented around 15 percent of the f.o.b. price of cocoa. Unlike coffee taxes, they were not increased during the 1960s.

Table 6 shows that cocoa exports from Haiti are very concentrated and that all exports go to industrial countries. As with coffee, one may therefore wonder whether these countries’ trade policies have affected exports of Haitian cocoa, preventing them from growing at a faster rate. This, however, does not seem to be the case. The United States, which is the main importer of Haitian cocoa (Table 6), does not impose any tariff on imports of cocoa beans. The EEC, on the other hand, reduced its rate of 9 percent to 5.4 percent as a result of the Kennedy Round of trade negotiations at the end of the 1960s. The tariff duties charged on processed beans by the United States and the EEC were reduced for the same reason.52 Today, industrial countries impose a very low tariff on cocoa beans. This amounted to an average of 2.6 percent at the very end of the 1970s. A somewhat higher tariff, 4.3 percent, is charged on processed beans. No non-tariff barriers, such as quotas or ‘voluntary’ export restraints, are imposed by industrial countries on imports of either beans or processed cocoa.53

Figure 2

One of the main objectives of the commodity agreements that have existed between cocoa producer countries since 1972 is to fix export quotas for participating countries.54 This may constitute a constraint for cocoa exports from Haiti. However, exporters and agricultural officials do not see the export quota accorded to Haiti as a major obstacle. This quota can easily be renegotiated since Haitian cocoa production accounts for a tiny share of world cocoa supply (between 0.1 and 0.2 percent).55

Table 6

The tentative conclusion which emerges from the analysis thus is that, while low prices combined with the Rybczynski effect contributed to reduce exports in the 1950s and 1960s, changes in the relative prices of subsistence crops and cocoa in favor of the latter product lie behind the increase in cocoa exports since the early 1970s.

D. Sugar

During the colonial period, sugar constituted the export product par excellence of the Saint-Domingue colony.56 Sugar was the crop that ensured the position of Saint-Domingue as the richest French colony at the end of the eighteenth century. Unlike coffee, however, sugar being a large-scale plantation product, sugar exports came to a complete halt after the national independence in 1804.57 They were only resumed at the end of the nineteenth century and on a large scale during the American occupation, after the establishment of the Haitian American Sugar Company, HASCO, which began producing in 1918.58 HASCO continued to be the largest sugar company, responsible for the lion’s share at the beginning of the 1980s, some 70 percent, of the cane crushed by the sugar mills in Haiti (some 40 percent of total output in 1982).59 Two other firms, Dessalines (Usine Sucrière des Cayes) and Citadelle (Usine Sucrière du Nord), accounted for the remaining 30 percent in the early 1980s. A state-owned factory, Usine Sucrière Nationale de Darbonne (USND), was constructed in 1983 and began producing in 1984.60

Figure 3

During the 1960s, sugar was the second most important export product in Haiti (see Table 1). An average of 40 percent of the total sugar output was exported during the 1957-69 period. The relative importance of sugar, however, declined in the following decade, partly as a result of the rise of bauxite exports, partly due to the decrease of sugar exports as the importance of the domestic market rose. (The development of exports from 1957 to 1985 is shown in Figure 3.) The export share of output amounted to a mere 20 percent 1970-82, i.e. it had been cut in half.61 Haiti even became a net importer of sugar in the mid-1970s and, at the beginning of the 1980s, sugar exports ceased for a couple of years. Figure 3 illustrates the rapid decrease of sugar exports, both in terms of volume and as a percentage of domestic output.

Why did sugar exports decline? Did total output contract or was it only the export share that decreased? Beginning with output, the available data are far from good. Estimates diverge between the various sources.62 Quite probably, all of them overstate the produced volume, a fact borne out by the results of aerial surveys of cane land combined with the best available yield estimates.63 Still, it can probably be safely conjectured that the long-run trend of output, from the mid-1950s to the early 1980s, was a downward one (at best one of stagnation). The area under cane in three of the four major cane producing plains, those of Cul-de-Sac, Léogâne and Cayes, shrank from 1956 to 1978, and adding the fourth large plain, the Plaine du Nord, where due to the opening of a new sugar mill in 1970, the area devoted to cane quadrupled, makes for a mere total increase of 6 percent. Also, the amount of cane handled by the sugar mills decreased markedly in the beginning of the 1980s.64 HASCO alone produced more sugar in 1950 than all four mills together in 1984.65

The stagnating or shrinking land area only provides part of the explanation of the stagnation of output. Simultaneously, the sugar content of the cane deteriorated, as indicated by Table 7.

Table 7

On top of the declining production, the domestic consumption of sugar appears to have risen slowly over the same period, among other things due to increased production of sugar-based alcohol (clairin or tafia).66 Thus, the explanation of the declining trend is to be found both on the production side and in a switch from exports to domestic sales.67

One of the most important factors behind the stagnation of cane production is a change in relative prices in favor of subsistence crops. The price of sugar cane and the price of sugar on the domestic market were fixed by a government agency, the Régie du Tabac, from 1961 to 1985. The former price, which was paid by sugar mills to cane growers, remained unchanged, at US$ 4.1 per ton, between 1948, when the government monopoly of wholesale sugar distribution began, and 1973.68 During the same period, the prices of subsistence crops increased significantly. For example, between 1953 and 1971, the prices of corn, red beans and sweet potatoes increased by 37 percent, 31 percent and 45 percent, respectively.69 The cane price, however, was increased beginning in 1973 and amounted to $ 13 per ton in the early 1980s,70 but this increase was not large enough to outweigh the increases in the prices of subsistence crops and to change the relative returns per unit of land in favor of cane production. According to a World Bank study from 1981, the yearly financial returns per hectare were $ 130 to $ 176 for sugar cane but $ 154 to $ 208 for rice, $ 300 for bananas and $ 380 to $ 540 for vegetables.71 As it seems, the low price of sugar cane to some extent could be blamed on the sugar mills, since these have constituted a vocal lobby whereas the diffused mass of cane growers have no representational voice.72 Between 1971 and 1980-84, the share of the growers in the ex-factory value of sugar fell from 51 to 29 percent, whereas the gross share of the mills increased from 49 to 71 percent.73 Thus, price factors appear to have reinforced the Rybczynski effect in the case of sugar.

In view of the changes in relative prices between sugar cane and other crops and the very different returns per hectare that they have given rise to, one may wonder why sugar production has not declined more rapidly. A hypothesis advanced by Leslie Delatour is that the existence of larger farms (in comparison to the average peasant holding) in cane growing regions impedes production from reacting strongly to shifts in relative prices.74 Larger landholders may invest in landownership basically for speculation purposes and are thus concerned with the level of the current income derived from agricultural activities only to a limited extent. This makes for a lower sensitivity to price changes.

The low cane price is not the only reason behind the bad export performance of the Haitian sugar industry. Another factor is tax policy. From 1948 to the early 1960s, the sugar tax was low. However, in the late 1960s and early 1970s, the marginal rate of taxation imposed on HASCO’s sugar exports was raised to very high levels (up to 80 percent when the f.o.b. price per bag of 100 pounds exceeded US$ 5). The two other companies were exempt from export duties up to 1975 as an industrial incentive, and the tax exemption for Citadelle was extended up to 1978 because of financial difficulties.75 In 1975, the maximum marginal tax rate was lowered to 30 percent but simultaneously a surtax applicable above a certain world market price was introduced.76

The discriminatory tax policy forced the formerly export-oriented HASCO company to concentrate on the growing domestic market, where the price was higher than in the world market. For example, in the spring of 1987, the ex-factory price was fixed at $24 per bag of 100 pounds, when selling in the domestic market, while the corresponding international price amounted to $7.50.77 From 1976 until 1985, when it was merged into the Direction Générale des Impôts, the Régie du Tabac handled all the sugar production of the three factories.78 Sugar was sold preferably on the domestic market where prices, which were fixed by the Régie, were higher than in the world market. On top of this, there was a tax differential between the production of sugar and the production of clairin, with a substantial tax on the consumption of centrifugal sugar (produced by the mills) and virtually nothing on clairin.79 There is thus no doubt that the Régie’s price and tax policy was a major source of the decline in sugar exports during the second half of the 1970s.

A third explanation of declining Haitian sugar exports may be the highly protectionist sugar policy of developed countries. A crucial characteristic of the sugar market is that there exist two competitive and separate source of supply: sugar cane and beets. Sugar cane is produced by tropical developing countries, like Haiti, while beets are a temperate product grown in Europe and North America. There is little doubt that the high protective barriers granted beet producers in developed countries, especially Western Europe, has inhibited the exports of the cane sugar industry in developing countries. Contrary to what happened with other tropical products, no improvement was made in access to developed countries under the Kennedy Round at the end of the 1960s.80 The replacement of tariffs and quantitative restrictions by a system of variable levies in the EEC in 1968 has increased the degree of protection for beet producers in Western Europe further yet.81 The EEC Sugar Protocol of the Lomé Convention allows 18 developing countries to export fixed amounts of sugar to EEC members.82 Haiti does not, however, belong to this group, which comprises other Caribbean countries such as Jamaica, Belize, Trinidad and Tobago. Another important characteristic of the EEC sugar policy is the subsidization of sugar exports, which explains why the share of the EEC in world sugar exports rose from less than 9 percent in the 1960s to more than 20 percent at present.83

Table 8

Exports of Haitian sugar to the United States, the main importer of Haitian sugar (Table 8), are also subject to quantitative restrictions. These have been strengthened substantially over the past two decades. In the late 1960s, an import quota of 30,000 tons was allowed for Haitian sugar exports to the US.84 In the early 1980s, it had been practically halved and amounted to 16,000 tons85 and, subsequently, it has been further reduced, first to some 12,500 tons86 and, thereafter, to 8,000 tons (1988).87 Note, however, that the Haitian sugar exports to the US have never attained the allowed quota over the past decade.

The sugar policy of the industrial countries has had two prejudicial effects for cane sugar exporters from developing countries. First, it has affected their export revenues. The loss in export revenue for developing countries was estimated by the World Bank at $7.4 billion in 1983.88 Secondly, the sugar policy of the industrial countries increases price fluctuations in the residual free world market for sugar and increases uncertainty for developing countries’ sugar companies. (The free market accounts for some 15 percent of the sugar traded in the world. The rest is channeled through preferential agreements.) Probably, all this affects developing countries’ production and export decisions and discourages them from exploiting their comparative advantages. According to one estimate, industrial countries’ sugar policy increases price instability in the residual world market for sugar by 25 percent.89 No doubt, part of the sharp fluctuations in the sugar world price observed during the last two decades can be ascribed to these policies.90

Total exports of Haitian sugar amounted to 25,000 tons per year on average during the 1960s and to 14,000 tons during the 1970s. The restrictions imposed on sugar imports by the United States can hardly be responsible for the decrease in sugar exports since the quota was never attained. Regarding the sugar policy of industrial countries, especially that of the EEC, and the preferences given by the EEC to those of Haiti’s competitors that belong to the ACP (African, Caribbean and Pacific Ocean) countries of the Lomé Convention,91 the tentative conclusion which emerges is that this may have acted as a deterrent to Haitian sugar exports, but not necessarily, since the US quota was not fulfilled.

A more important factor behind the bad export performance of the sugar industry is to be found on the supply side, in the declining sugar content of Haitian cane. All four mills have relied on outside producers of sugar cane instead of integrating backwards.92 At the beginning of the 1980s, HASCO had 7,000 cane suppliers, Dessalines 3,500 and Citadelle 1,300.93 This means that in practice it is impossible to control the quality (sugar content) of the cane. In addition, neglect, old equipment and poor management – i.e. X-inefficiency – are often advanced to explain the deterioration over time and the differences between mills (HASCO being far better than its competitors).94

The situation of the Haitian sugar industry has worsened drastically during the past few years. In the fall of 1986, the two state-owned sugar mills, the Usine Sucrière Nationale de Darbonne and Citadelle, closed their doors due to lack of efficiency.95 In the spring of 1987, the private-owned Dessalines and HASCO were also closing down their operations. The inflow of large amounts of smuggled sugar from the Dominican Republic was the main reason that was put forward by the two factories.96 Smuggling was encouraged by the high Haitian price and by the fact that the Dominican Republic could not sell as much sugar as before in the US because of the reduction of its sugar quota in 1986.97 If definitive, the closedown of HASCO means that Haiti is no longer a sugar exporting country.

Sugar is just one of the cane products. Molasses is a by-product which is produced by the same mills and which is also exported. As shown by Table 1, its share of total exports is rather small (around 0.8 percent for the 1950-84 period). As with sugar, a sharp decrease can be observed after 1976. The similarity of the decrease suggests that the same factors lie behind the bad performance of the molasses industry.

In conclusion, the Rybczynski effect, government price and tax policy, changes in relative prices in favor of subsistence crops, certain characteristics of the Haitian industry, and possibly also the sugar policy in some industrial countries, combined to explain the decline of sugar exports over the past two decades and the fall of the domestic sugar industry.

E. Sisal

Sisal was the most important export crop after coffee during the 1950s. This was illustrated in Table 1. One firm, Plantation (later Port) Dauphin, was responsible for the bulk of sisal production in Haiti (around 70-80 percent).98 Revenue from sisal export duties constituted a non-negligible source of revenue for Haiti during the 1950s and the early 1960s.99 Since the end of the 1950s, the volume of sisal exports has decreased continuously and sharply. This is portrayed in Figure 4. Today sisal has almost disappeared as an item in Haiti’s foreign trade since it only accounts for 0.1-0.2 percent of exports. Which are the reasons behind this fall?

Sisal prices decreased between the early 1960s and 1971. In 1971, Plantation Dauphin put a halt to its production. The rise in the price of oil in the mid-1970s brought about an increase in both the production costs of synthetic fibers and the price of sisal. This led Dauphin to reopen its doors in 1974. Figure 4 shows that this resulted in an increase in sisal exports from Haiti. However, the recovery was short-lived. The sisal price fell again after 1975 and Port Dauphin closed down its operations for the second time in 1976.100The company was purchased by the Haitian government at the beginning of the 1980s and kept functioning until 1983.

Figure 4

In the mid-1970s, 47 percent of the sisal land in Haiti was cultivated under the plantation system by large-scale producers like Port Dauphin, NABASA, Sajous and Gérard Théard.101 The remaining 53 percent was farmed by peasants whose production accounted for about 60 percent of total sisal output.102 The price received by peasants rose from $ 0.10 per kilo in 1971 to $ 0.59 in 1974.103 This contributed to the increase in sisal exports observed in the mid-1970s (see Figure 4). However, the short-lived upward change in the sisal price during the first half of the 1970s was not large enough to reverse the trend of relative prices in favor of subsistence crops. As illustrated when other export products were examined, the price of subsistence crops has increased rapidly and continually over the past two and half decades. It is this trend in favor of subsistence crops which lies behind the decrease in peasant sisal production and exports.

The main reason behind the falling prices appears to be that the world demand for sisal is stagnating. According to Jere Behrman, the world’s income elasticity for sisal is equal to zero.104 A shift in demand from sisal to synthetic fibers explains this result and is the main factor behind the rapid decrease in sisal exports. Another factor is the high degree of price instability that characterizes the world sisal market. For the 1965-81 period, only sugar exhibited a higher degree of price instability.105 This instability is all the more prejudicial to sisal production and exports since up to three years are required between the establishment of a new plantation and the sale of sisal on the market.

Today, most of the sisal output is absorbed by the domestic market. To some extent, exports of raw sisal fiber have been replaced by exports of finished fiber products, like twines, but not enough to compensate for the decline of the former.106 The role of sisal in the world economy has decreased tremendously over the past three decades. The emergence of substitute products explains this change. It constitutes the major reason behind the rapid decline of sisal exports from Haiti.

F. Essential Oils

A non-negligible share of export revenues comes from essential oil exports (vetiver, lime, amyris and bitter orange). These exports are of recent origin in Haiti since they began only 50 years ago when a French agronomist started the first distillery.107 Essential oil exports from Haiti expanded strongly during the Second World War when the main importer, the United States, was isolated from its traditional sources of supply in Asia.108 Their share of Haitian exports is today as large as that of cocoa (see Table 1). Figure 5 shows that the amount of essential oils exported increased rapidly

Essential oils are very concentrated. Table 9 indicates that vetiver oil accounts for the largest share and that exports of vetiver, lime oil and amyris together represent no less than 95 percent of essential oil exports. Roughly similar changes in the three main products lie behind the profile of essential oil exports over time. Vetiver oil exports increased rapidly up to the end of the 1960s, stabilized during the first half of the 1970s and have decreased since then.109 Lime oil exports increased rapidly up to 1974 but have decreased thereafter.110 Amyris oil exports followed an upward trend up to 1966 but this was turned into a slow downward movement after this year.111 Thus, the reduction of essential oil exports after 1974 (see Figure 5) is the result of a drop in the exports of the three main oils. Why did this fall take place?

Figure 5

The data on the price paid to the agricultural producers are not very good. Only scattered estimates are available. At any rate, as it seems, the price of vetiver root fell from 1960 to 1970-74 and remained constant thereafter.112 During the same period, the prices of subsistence crops increased significantly. For example, between 1960 and 1971, the price of corn, red beans and sweet potatoes increased by 98, 15 and 96 percent respectively.113 This implies that the rapid increase in vetiver oil exports up to the mid-1970s cannot be explained by a change in relative prices on the producer level. Vetiver root production appears to be inelastic with respect to relative prices because the land on which the vetiver roots are grown cannot be used for subsistence crops since it is rocky.114

For lime, the price to the producer increased with some 40 percent from 1970 to the early 1980s.115 During the same period, exports show a downward trend,116 which may indicate that the prise rise was not large enough to outweigh price changes in the competing markets for subsistence crops. In the case of amyris, hardly anything appears to be known regarding the price of the raw material.

Table 9

Turning to the factory side, Figure 6 provides evidence of slightly decreasing or stagnant prices of essential oils from the early 1950s up to the beginning of the 1970s. Thereafter, the trend has been an upward one, particularly for vetiver.

Figure 6

The competitive position of Haiti in the production of essential oils and an increase in the world demand can explain the growth of exports prior to the mid-1970s. The cosmetics industry in the United States and Western Europe is the main purchaser of amyris, vetiver and lime. Lime oil is also utilized by the soft drink industry. Almost all of the fourth important product, bitter orange, is bought by the French liqueur firm Cointreau. Haitian essential oils are internationally competitive. Haiti undeniably has a comparative (and even absolute) advantage in their production.117 Haitian vetiver oil production is the largest in the world, Haiti is the second largest producer of lime oil and amyris is produced only in Haiti. Note that this situation differs greatly from the one observed in the coffee, sugar or cocoa markets.

Income elasticities for essential oils have not yet been estimated. However, the world demand for essential oils seems to be growing in the long run. Roger Schwob has calculated that the yearly rate of growth of world demand for vetiver and lime oil are 2-3 percent and 7 percent respectively.118 He does not provide any figures for amyris but suggests that the world demand for this product is increasing moderately.119

The downward trend in essential oil exports after 1975 contrasts with the upward change in prices. The government pricing, marketing and taxation policies lay behind this apparent paradox. A public marketing organization, OCEAH (Office de Commercialisation des Essences Aromatiques d’Haïti), was granted the monopoly of essential oil exports in 1975.120 The main reason invoked by the authorities is that the bulk of the Haitian exports (80 percent) is purchased by one firm, the New York based Polarome, and that the creation of OCEAH can counterbalance the power of Polarome over prices.121 However, from 1981, the vetiver producers were allowed to negotiate their own export contracts as long as the sales were registered with OCEAH.122

No tariffs or quantitative restrictions are imposed on imports of essential oils in industrial countries. The two main export markets for Haitian essential oils are the United States and France (see Table 10). Demand shrank in both markets at the very end of the 1970s and in the early 1980s.123 In spite of this, the OCEAH maintained high export prices for vetiver and lime oils.124 This pricing policy resulted in a drop in the Haitian market shares in USA and France to the benefit of other exporting countries and a decline in the volume of essential oil exports from Haiti.125

Another potentially prejudicial factor may have been taxation, but as it seems, only to a minor extent. Prior to 1975, the export tax was moderate and amounted to 5 percent of the c.i.f. price.126 In October 1975, a levy for National Defense was introduced in addition to the 5 percent tax.127 In 1978, when the price of vetiver rose rapidly, this levy was increased and another levy per pound of oil was also added.128 Since then, however, the rate of taxation has decreased significantly. In 1982, it represented 4.8 percent of the vetiver export value, 10.1 percent of the lime export value and 9.8 percent of the amyris and bitter orange export value,129 low rates in comparison with coffee (26 percent on average, 1980-82).130 In 1985, taxes were abolished altogether.131

Table 10

Thus, since in the main taxation of the essential oil industry has been modest, taxes do not appear to have contributed very much to maintaining high prices for Haitian essential oils, thus favoring competitors on foreign markets – especially not in the amyris market where Haiti is the only producer and cannot be considered a price-taker in the world market. In this case, the main result of the OCEAH pricing policy has been a drop in demand. Since amyris exports increased up to 1970, the price of the product fell. In 1969, an association of essential oil producers was founded and this organization consciously restricted output in order to bring prices up again, a policy that was inherited by OCEAH in 1975.132 OCEAH’s performance was significantly worse for the other two products. The Haitian share of the US market was shrinking during the second half of the 1970s, and OCEAH proved incapable of remedying the situation.133 In the case of vetiver, OCEAH policy was highly detrimental. The organization apparently concentrated on maximizing tax revenues in the short-run and failed to adjust prices in the United States enough when the competitors (Indonesia) did. A loss of market share was inevitable.

Essential oils have become one of the leading export products of the Haitian primary sector over the past three decades. Growing exports prior to 1974 are mainly due to an increase in the world demand. The declining trend thereafter is mainly the result of the Haitian government’s pricing and marketing policies.

G. Bauxite

Bauxite has been the second most important export product of the Haitian primary sector over the past two and half decades. Its mining and export began in 1957 with the establishment of the American Reynolds company in the country.134 The bauxite sector was a completely export-oriented sector. Although it only represented 1 percent of the gross domestic product and 0.5 percent of the industrial workforce in the 1960s,135 it accounted for no less than some 11 percent of total Haitian exports.

Figure 7 depicts the development of bauxite exports over the past two and half decades. It shows an increase in the amount of bauxite exported by Haiti up to 1980 and a sudden stop in 1983. Why did exports first go up and then down?

The mining of bauxite is the first stage in a process leading to the production of semi-finished and finished aluminum products. The whole process in most cases is performed by highly integrated companies which mine bauxite, refine it into alumina and, in a third stage, smelt alumina into aluminum.136 Reynolds has extracted and exported bauxite from Haiti while the last two operations were performed in the United States, in spite of higher transportation costs for bauxite than for alumina.137 The presence of substantial economies of scale in the refinement of alumina explains this location in the United States.138 The amount of bauxite exported by Reynolds depended upon the demand for aluminium in the world and, also, upon the agreement between Reynolds and the Haitian government concerning the royalties and taxes that the former should pay to the latter in order to extract bauxite. This agreement decided the price of Haitian bauxite.

Figure 7

The demand for the final product aluminium and, consequently, for bauxite has increased rapidly over the past two decades. Jere Behrman has estimated the world income elasticity for bauxite to 2.3.139 The increase in Haitian exports of bauxite between 1957 and the early 1980s was probably to a large extent due to this factor. Political considerations may also have played a role. From 1969 on, Reynolds started getting a higher share of its bauxite supplies from Haiti, presumably as a result of the decision of the Forbes Burnham government in Guyana to tighten control over the bauxite industry there.140

Another characteristic of the demand for bauxite is its rather high price elasticity (-1.3 according to Jere Behrman).141 This means that an increase in the level of tax imposition on bauxite exports affects the volume exported significantly. The taxes paid by Reynolds increased from less than 5 percent of the export value of bauxite in 1957 to an average of 9 percent in the 1960s and 14 percent in 1971.142 Producer attempts to influence prices on the world market took place in 1974 when ten major bauxite exporters (Australia, Guinea, Guyana, Jamaica, Sierra Leone, Surinam, Yugoslavia, the Dominican Republic, Ghana and Haiti) formed the International Bauxite Association (IBA).143 In the Caribbean member countries, royalties and taxes on a specified level of bauxite production were increased by 700 percent during 1974 and 1975.144 This resulted in a drastic increase in the price of bauxite. It convinced the major American corporations to expand their investment outside the Caribbean and it led to a steep decline in the region’s world market share.145 This corporate policy has been supported by the fact that, unlike many other minerals, bauxite is one of the most abundant materials in the earth’s crust.146

It is likely that the royalty increase is the major factor behind the decision of Reynolds to pull out of Haiti. A complementary reason is that bauxite reserves in Haiti were limited. These reserves were estimated at 23 million tons in 1950 and 25 million in 1965147 and about 15 million tons had been extracted by Reynolds prior to 1983.148 Together, these two factors made bauxite exploitation increasingly uneconomical149 and in 1983, Reynolds stopped its production in Haiti.

It has been argued that the mining of bauxite has had a negative effect on coffee production and exports.150 The reason for this is that bauxite mining has led to the withdrawal of fertile land from cultivation of coffee (and subsistence crops) and has argument, coffee production in the south, where the bauxite mine is located, is compared with coffee production in other regions (see Table 11).

Table 11 provides only weak evidence of a decrease in coffee production in the southern region over time. The share of the south in Haitian production dropped from 51 to 46 percent between the 1950s and the mid-1970s. The bauxite mine is situated on the Plateau de Rochelois near the city of Miragoâne. More desegregated figures show that the share of the Petit-Goâve sub-region, where the bauxite mine is located, dropped from 18 percent of the coffee production in the south for the 1956-58 period to 5 percent for the 1972-75 period.151 Thus, factual analysis tends to confirm that the mining of bauxite is responsible for a contraction of the coffee production. However, it should be stressed that this contraction only accounts for a residual part of the slow decrease in Haitian coffee exports over the last two decades.

Table 11

Bauxite exports were a major source of foreign exchange earnings for Haiti between the very end of the 1950s and the early 1980s. A sharp increase in aluminum consumption in developed countries lay behind the upswing in bauxite exports from Haiti up to the early 1980s. Increasing export taxes and limited bauxite reserves explain the decision of Reynolds to pull out of Haiti and why Haitian bauxite exports came to a sudden end in 1983.



The share of primary exports in total exports from Haiti has decreased over the period examined. An increase in manufactured exports is one of the factors behind this trend,152but another is a decline of primary exports. We have shown that this decline has not affected all primary products equally. The situation, and thus the prospects for the future, varies from product to product. In this section, we will concentrate on these prospects and on possible policy measures for expanding Haitian primary exports.

Exports of minerals from Haiti disappeared in 1983 when Reynolds put a halt to the extraction and export of bauxite. In 1987, discussions were begun between the Haitian government and an American company, the LCP Corporation, which was interested in exporting marble and taking over the Reynolds installations and harbor facilities.153 These negotiations, however, had no concrete results.

Sisal is another primary product that virtually ceased to be exported in the early 1980s. No policy measures seem able to reverse this situation. A decrease in the price of sisal has no influence on the world demand for sisal and is unlikely to affect Haitian production and exports. According to Jere Behrman, the world demand price elasticity for sisal is equal to zero.154 Thus, lower prices will simply lead to lower incomes as well.

Sugar is a third product the export performance of which has deteriorated sharply over the past two decades. Our analysis suggests that foreign factors – i.e. trade and sugar policies in the United States and Western Europe – have played a limited role and that it is mostly domestic factors that explain the fall of the Haitian sugar industry. Public policy in Haiti can only act on the latter. A pricing policy that takes greater account of the prices prevailing in the world market should create a more competitive environment for the domestic industry. However, one may well wonder if this measure is sufficient for the recovery of the Haitian sugar industry which suffers from a high degree of X-inefficiency. Another major obstacle to recovery is the comparatively low sugar content of the Haitian cane and the failure of the endeavors to introduce varieties of cane with a higher sugar content in Haiti.155

At any rate, protection of the domestic sugar industry continues to be fairly high. In March 1987, a new tariff code established a 40 percent duty on imported sugar. At the same time, the excise duty on raw sugar was abolished, the one on domestically manufactured refined sugar cut in half and price control on both items removed.156

The conclusion which emerges for the future is thus that sugar exports will not recover because Haitian production is no longer competitive in the residual free world market for sugar. On the other hand, because of the permeability of the border between Haiti and the Dominican Republic, a public policy that maintains high prices in the domestic market is not sufficient to keep the domestic industry alive. If such a policy is pursued, smuggled sugar – and not domestic production – will satisfy the rather rapid rise in sugar consumption which results from increases in per capita income over time and which presumably characterizes Haiti as much as other developing countries (see the income elasticity for sugar in developing countries in Table 12).

Table 12

Sugar cane can also be used to make rum, the export of which can be increased because the Haitian rum, Barbancourt, benefits from a well-established image abroad. Rum exports from Haiti can be further encouraged by the inclusion of Haiti into the ACP countries of the Lomé Convention since the ACP rum quotas remain unfilled.157Negotiations are going on between Haiti and the European Community concerning this issue.158 However, rum exports can hardly make up for what has been lost through the decline of sugar exports.

The prospects for the three other export products, coffee, cocoa and essential oils, are a bit more optimistic. Their export performance can be improved provided that public policy relaxes the different constraints that impede their development. In the context of Haiti, an important constraint has been the existence of public marketing organizations. As argued above, these organizations have negatively affected the export performance of the cocoa and essential oil sectors. No doubt, the suppression of HAMASCOSA in 1978 and the abolition of the monopoly of OCEAH in 1981 have improved the working conditions of the cocoa and essential oil industries.159

Another major constraint on these primary exports has been the structure of the tax system. This was illustrated in some depth when the individual products were examined. The main objective of taxation policy in Haiti has been to maximize public revenue in the short run. This objective has led to high export duties that have represented an important source of government tax revenue. (Table 13 shows that they accounted for more than a quarter of government revenue between 1974 and 1978.)

Table 13

For coffee and cocoa, the higher the export tax, the lower the share of the f.o.b. price received by the producer will be. The reason for this is that Haiti is a price-taker in the coffee and cocoa export markets so that a tax increase cannot be passed on in the price. Without the increase in the export taxes on coffee during the 1960s, the changes in relative prices between subsistence crops and coffee would have been less favorable to the former. With lower taxes, presumably the production and exports of coffee and cocoa would have been larger. As regards the increase in the export tax on essential oils during the 1970s, its effect probably has been less detrimental for Haitian exports both since the export taxes were low in comparison with coffee and since Haiti is a price-maker in most essential oil export markets, so that tax rises to some extent can be passed on to the buyers.

A fiscal reform has been implemented in Haiti during the past four years.160 The long-term objective of the reform is to abolish export taxes and to introduce a new income tax and a value added tax instead. One aspect of the reform consists in the absorption of the Régie du Tabac into a new Direction Générale des Impôts. This was done in early 1985. Table 13 illustrates the changes in the Haitian tax structure for the 1980-84 and 1985-86 periods as compared with the 1974-78 period. It shows that export duties, which accounted for more than 25 percent of government revenue during the second half of the 1970s, have decreased rapidly thereafter and accounted for no more than 5 percent of government revenue in 1985-86.

In the present context, the diminution and, in a second stage, the removal, of export taxes is the most interesting aspect. From an initial level of 26 percent, the export tax on coffee was reduced to 23.4 percent in 1984, to 19.5 percent in 1985, to 10 percent in 1986 and it was definitively abolished in September 1987.161 Export taxes on essential oils were eliminated in March 1985.162 The large size of the coffee tax reductions – from 26 to 0 percent of the export value – suggests that tax reductions may have benefited coffee growers in the form of higher producer prices. Presumably, their effects on relative prices between coffee and subsistence crops should be substantial. The suppression of the tax imposed on essential oil exports in 1985, on the other hand, should only have a limited effect on producer prices because of the rather low level of the export tax prior to 1985. The increase in the relative price of coffee is likely to counteract the Rybczynski effect, at least to some extent, and to further coffee production and exports. However, it is yet too early to provide conclusive evidence on the effects on coffee production and exports since the trees take several years from planting time to mature and produce beans.163

Besides export taxes, producer prices are influenced by the share of the f.o.b. price that the intermediaries retain for services rendered and value added. The larger these intermediaries’ share is, the lower the producer prices will be and vice versa. The intermediaries’ share is rather large and varies somewhat from product to product. For example, it represents around 25 percent of the f.o.b. coffee price.164 Several authors have argued that the large share captured by the intermediaries, in particular the exporters, reflects their market power.165 According to them it is the non-competitive behavior of exporters which explains the low prices paid to farmers for their products and, thus, the bad export performance of the Haitian primary sector. If true, this means that the exporters’ market power constitutes a major constraint for primary exports and that one of the objectives of public policy should be to relax it by furthering competition.

Whether the market for export crops, in particular coffee, is competitive or not has been the subject of very heated discussion in the literature dealing with Haiti.166 Elsewhere, we have shown that the collusive argument by and large lacks empirical support although the state of competition cannot be regarded as perfect.167 The market power of the exporters is constrained by the world market and the fact that the prices paid to the peasants must be high enough to prevent peasants from dropping out of the coffee market. The very existence of a domestic market for coffee limits further the oligopsonistic power of exporters since peasants can sell their coffee on this market if they are not satisfied with the prices offered by intermediaries. The relatively high degree of entry, exit and mobility that prevails in coffee marketing also supports the competitive argument because such an instability can hardly exist in a collusive market characterized by a ‘quiet life’ and the sharing out of the market between the established exporters.168 From time to time, the leading exporters have attempted to limit competition but these efforts have tended to be short-lived.

The state of competition prevailing in the Haitian coffee marketing is not perfect. Nevertheless, the important role of the international environment, the presence of a domestic market and the rather high degree of entry and exit means that competition is far from being absent. The Haitian coffee market is much like an instable (contestable) oligopolistic market with short periods of collusion following long periods of keen competition.

At any rate, maintaining competitive conditions is important. A policy encouraging competition can expand exports since it results in higher prices for coffee growers. The same effects can be expected in the cocoa and essential oil sectors. Government support to the cooperative movement can be a part of such a policy since the presence of cooperatives presumably exerts competitive pressure on the private intermediaries. Other positive aspects of cooperatives are that they may offer higher prices to their members, since they do not have to earn the normal profit prevailing in the overall economy but will remain in operation as long as they do not run a loss,169 and that profits are reinvested in agriculture instead of being invested in other sectors of the economy. The low income elasticities that characterize coffee and cocoa (see Table 12) indicate, nevertheless, that the growth of exports can only be moderate.

Taxation and competition policies are not the only areas of public intervention that have been or may be used to affect primary exports positively. There also exist long-term measures whose enforcement lead to output and export expansion. For example, the improvement of the road infrastructure should lower transport costs and increase the prices received by the farmers. The improvement of the harvesting and processing methods by way of technical assistance could increase yields in the agricultural sector. Cooperatives can presumably play a role here. As regards the essential oil sector, which is extremely energy intensive, technical assistance can improve energy efficiency and besides reduce deforestation and erosion ensuing from the use of wood as a source of energy.170 Another possible area for public intervention in the essential oil and coffee sectors is that of examining the possibility of further downstream integration.171 The manufacture of value-added goods based on the processing of coffee and essential oils can form a strong base for the emerging patterns of exports. The case of cocoa appears to be different. Today, only cocoa beans are exported. Haiti ceased to be an exporter of processed cocoa in 1978 when the chocolate factory of HAMASCOSA closed it doors, probably as a result of the low quality of Haitian cocoa.172 This failure suggests that downstream integration is less likely to succeed in the cocoa sector.



As suggested in the introduction, one of the main reasons behind the negative growth and the worsening of the trade balance in Haiti is the decrease of primary exports. This fall concerns in particular sisal, bauxite and sugar. The decline exhibited by coffee, for the whole period examined, and essential oils, for the past decade, is much less pronounced. Cocoa is the only product the export of which has increased over the past two decades. A consequence of these changes is that the emerging pattern of Haiti’s primary exports exhibits more concentration.

We have argued that little can be done concerning traditional export products like sisal, bauxite and sugar. As regards the three remaining export products, coffee, cocoa and essential oils, their export performance can be improved somewhat provided that public policy pays more attention to the agricultural sector. The measures introduced since 1985 are steps in the right direction. Both the tax reform and the more active part taken in the formation of cooperatives are examples of a government policy acting to expand output of Haitian primary products. Several years are necessary in order to assess their impact on exports and to arrive at a more conclusive judgement. Whether these measures will be sufficient to counteract the Rybczynski effect in the long run is unfortunately more uncertain.



1. On the present economic situation in Haiti, see World Bank (1985a), International Monetary Fund (1986) and CEPAL (1986).return to text

2. CEPAL (1986), p.4.return to text

3. Lundahl (1979), p.193, Locher (1988).return to text

4. CEPAL (1986), p.2.return to text

5. Little, Scitovsky and Scott (1970), Chapter 2.return to text

6. See for example the articles contained in Lundahl (1985a) and Gillis et al. (1987), Chapter 15.return to text

7. Cf. Mamalakis (1965), Chapter 4.return to text

8. llirschman (1958), Chapter 6.return to text

9. See Hirschman (1977), pp.72-77.return to text

10. In the context of Haiti, however, where a predatory state has existed since the mid-nineteenth century, this has hardly been the case in practice. (Cf. Lundahl (1979), Chapters 7 and 8 and Lundahl (1985c).return to text

11. Cf. e.g. McKinnon (1973), Shaw (1973) and Fry (1988).return to text

12. See Latin America Bureau (1985), p.50, Francisque (1986), pp.139-41 and Cloutier (1987b).return to text

13. See Lundahl (1979), pp.40-45 and Francisque (1986), pp.117-36.return to text

14. Lundahl (1979), pp.260-62.return to text

15. Delatour (1984), p. 1.return to text

16. Ibid.return to text

17. Ibid, pp.2-3.return to text

18. For an overview of agricultural export products, with emphasis on the spatial factor, see CNRS (1985), planches 14-17.return to text

19. Of the minor export products, cotton is dealt with e.g. in JWK (1976) and Delatour (1984), meat in JWK (1976c), mangoes in JWK (1976d), copper in Caprio (1979), pp. 175-76 and fish in CNRS (1985), planche 18. Bananas were exported during the 1930s and 1940s. However, by the mid-1950s, the volume of banana exports had fallen almost to zero (Lundahl (1979), pp. (43-44).return to text

20. Rybczynski (1955).return to text

21. In agriculture, if constant returns to scale do not prevail, the most likely case is that decreasing returns do, for example as a result of the non-inclusion of factors like capital or management in a two-factor model containing labor and land only, as assumed in the main text. However, provided that the land-labor ratios differ enough between the two lines of production, the theorem carries over to this case as well (Hansson and Lundahl (1983).return to text

22. See Lundahl (1979), Chapter 5, for a detailed exposition of the argument.return to text

23. Ibid, p. 236.return to text

24. CF. Table 1.return to text

25. A striking aspect of Table 3 is the drop of the US share and the rise of the Italian share.return to text

26. UNCTAD (1969), p.21.return to text

27. Ibid., pp. 21-22.return to text

28. Tariff and non-tariff barrier figures are averages for Australia, Austria, the EEC (excluding Greece, Portugal and Spain), Finland, Japan, New Zealand, Norway, Sweden and Switzerland. (See World Bank (1986), p. 126.)return to text

29. Ibid. Since 1962, attempts have been made via four international coffee agreements to maintain high prices through export controls. These agreements are not responsible for the decline in Haitian coffee exports either since they never contained any reductions in these exports. (On the coffee agreements, see Gilbert (1987), pp.602-OS, and Gordon-Ashworth (1984), pp.210-18.)return to text

30. Notably by Girault (1981) and Seguino (1985).return to text

31. This issue is discussed in Bourdet and Lundahl (1988), where it is argued that the contention by and large lacks empirical foundations.return to text

32. Cf. Lundahi (1979), Chapter 4, and Bourdet and Lundahl (1988) for a description of the two marketing chains.return to text

33. Lundahl (1979), p. 144.return to text

34. Capital Consult (1983), pp. 10-14.return to text

35. Seguino (1985), p.48.return to text

36. Capital Consult (1983), pp. 10-14.return to text

37. Seguino (198), p.48.return to text

38. Ibid., pp. 14-15.return to text

39. Ibid., p. 15.return to text

40. Ibid., p. 12.return to text

41. Presumably, they are more reliable for the earlier part of the period, at least in that they don’t contain as many inflation-triggered raises in marketing costs. Between 1964 and 1970, the cost of living index rose with a mere 11 percent while between 1970 and 1975 the corresponding figure was 88 percent (ibid.). Thereafter the rate of price increases has remained high.return to text

42. Cf. Capital Consult (1983), p.7 and Bourdet and Lundahl (1988). Another reason for the decline in registered coffee exports over most of the period under consideration may be an increase in the amount of coffee smuggled across the border from Haiti to the Dominican Republic. Girault (1981), p.74, estimates that some 60,000 bags were smuggled per year in the mid 1970s. (Cf. also Helfenberger et al. (1987), p. 56, where a figure of 50,000 bags (no reference to year) is given.)return to text

43. Capital Consult (1983), p.98.return to text

44. A rise of the domestic price would also contribute to explaining why the direction of smuggling has been reversed. Seguino (1985), p.31, estimates that in the mid-1980s, some 40,000 bags of Dominican coffee entered Haiti illegally each year. (In the Dominican Republic, output exceeded both the export quota assigned to the country by the International Coffee Organization and local consumption plus exports (Capital Consult (1983), pp.5-6). To this came the unification of the Dominican exchange rate in 1985 which implied a defacto devaluation of the peso (Seguino (1985), p. 31).) This trend may have been reinforced by the gradual decrease of the Haitian export tax on coffee 1984-86 and its final abolition in 1987.return to text

45. Capital Consult (1983), pp. 46-47.return to text

46. Cook (1984). p.8.return to text

47. Ibid., p.7.return to text

48. Lundahl (1979), pp. 217, 219.return to text

49. Cook (1984), pp. 3-4. To what extent the marketing of cocoa is competitive so that price changes in the world market are passed on to the producers is difficult to know. In the mid-1980s, four firms, Novella (the largest), Wiener, Madsen and Bennett, dominated the market. All these firms exported coffee as well and the contact between exporters and producers goes via speculateurs, exactly as in the case of coffee. Frequently the same people are involved in both chains (ibid., pp. 14-15).return to text

50. International Monetary Fund (1986), p.45.return to text

51. Cook (1984), p.20.return to text

52. UNCTAD (1969), p. 27.return to text

53. See World Bank (1986), p.126.return to text

54. See Gordon-Ashworth (1984), pp.227-30 and 299.return to text

55. See Cook (1984), p.17.return to text

56. However, it should be pointed out that Saint-Domingue was far from being a monoculture economy. The relative importance of coffee increased over time, to the point where, at the outbreak of the French Revolution, coffee was almost as important as sugar from the export point of view (Lundahl (1979), p.259, cf. Trouillot (1982) for the importance of coffee production in a wider context). To this came exports of cocoa, cotton and indigo.return to text

57. Cf. Lundahl (1979), Chapter 6, (1984), (1985c), for a discussion of the reasons.return to text

58. Gaillard (1982), pp.27-29.return to text

59. Delatour (1983a), p.13. The rest goes to alcohol production and to direct consumption.return to text

60. Ibid., p.29.return to text

61. Ibid., p. 12. The production and export statistics have been calculated by the U.S. Department of Agriculture. Presumably, the production figures are overestimates. (Cf. ibid., pp.7-13.)return to text

62. Ibid., Chapter 1, provides a comparison.return to text

63. Ibid.return to text

64. Ibid.return to text

65. Banque Mondiale (1985), p.24.return to text

66. Delatour (1983a), p. 13 and CNRS (1985), planche 16.return to text

67. Domestic consumption of sugar on a per capita basis is very low in Haiti, a mere 11.6 kilos per year at the beginning of the 1980s (World Bank (1985b), pp. 1, 9).return to text

68. Lebeau (1974), p.24.return to text

69. Lundahl (1979). p.219.return to text

70. In most cases cane growers received less that US$ 13 per ton because sugar mills charged from $2 to $3 a ton for transportation. See Delatour (1983a), pp.55-56.return to text

71. Quoted by ibid., p.55.return to text

72. World Bank (1985b), p.2.return to text

73. Ibid., p.20.return to text

74. Delatour (1983a), pp.49-55.return to text

75. In 1982, the Citadelle was forced into receivership by the Banque Nationale de Credit (International Monetary Fund (1986), note, p. 16).return to text

76. Abdel-Rahman and Byrne (1979), p.95.return to text

77. Jean-Francois (1987).return to text

78. Delatour (1983), p.46.return to text

79. World Bank (1985b), p.2.return to text

80. UNCTAD (1969), pp. 26-27.return to text

81. Stevens and Webb (1983), pp. 321-47.return to text

82. World Bank (1986), pp. 142-43.return to text

83. Ibid., p. 114, Gilbert (1987), p.598 and Gordon-Ashworth (1984), p. 168.return to text

84. Lundahl (1979), p.283.return to text

85. Delatour (1983a), p.20.return to text

86. International Monetary Fund (1986), p.6.return to text

87. Personal communication from Giovanni Caprio, December 16, 1988.return to text

88. World Bank (1986), p.114.return to text

89. Ibid.return to text

90. Gordon-Ashworth (1984), p.182.return to text

91. The Lome convention grants free access to the EEC market to goods exported from the ACP countries.return to text

92. HASCO has constituted a partial exception here, although to a decreasing extent. In the 1940s, close to 50 percent of the cane processed came from the company’s own fields. In the early 1980, this figure had decreased to 25 percent (Delatour (1983a), pp.32, 30).return to text

93. Ibid., p.31.return to text

94. Ibid., pp. 26-30.return to text

95. Cloutier (1987a).return to text

96. US/AID estimates indicate that something like 20 percent of the domestic consumption of sugar is covered by smuggling from the Dominican Republic (Douzant-Rosenfeld (1988), p.52). Denise Douzant-Rosenfeld maintains that this is an underestimate and that the true figure is something like 30-50 percent (ibid.).return to text

97. Cloutier (1987a). An overview of smuggling is given in Douzant­Rosenfeld (1988). Cf. also Murphy (1988).return to text

98. On the history of sisal in Haiti, see Fatton (1975).return to text

99. Abdel-Rahman and Byrne (1979), p.84.return to text

100. Lundahl (1979), p.285.return to text

101. JWK (1976a), p. 10.return to text

102. Ibid., p.23.return to text

103. Ibid., pp.28-30.return to text

104. Adams and Behrman (1982), p.58.return to text

105. See Gordon-Ashworth (1984), pp.16-18.return to text

106. World Bank (1985b), p. 108.return to text

107. CNRS (1985), planche 15.return to text

108. Ibid.return to text

109. Delatour (1983b), p. 17.return to text

110. Ibid., p.24.return to text

111. Ibid., p.33.return to text

112. Ibid., pp.16 and 18.return to text

113. Based on Lundahl (1979), Table 5.13, p.219.return to text

114. Delatour (1983b), p. 19.return to text

115. Ibid., p.26.return to text

116. Ibid., pp.23-24.return to text

117. Kerinel-Torres (1983), pp.341-43.return to text

118. Schwob (1982), p. 5O.return to text

119. Ibid., p.51.return to text

120. Ibid., pp. 36-37 and Kermel-Torres (1983), p. 226.return to text

121. Delatour (1983b), pp. 43-44.return to text

122. Ibid., pp. 36-37 and CNRS (1985), planche 15.return to text

123. Delatour (1983b), p.5O and Kermel-Torres (1983), p. 343.return to text

124. Delatour (1983b), pp.50-52.return to text

125. On the drop market shares see Kermel-Torres (1983), p.343 and Delatour (1983b), p.50.return to text

126. Delatour (1983b), p.38.return to text

127. Ibid., p.39.return to text

128. Abdel-Rabman and Byrne (1979), pp. 95-96.return to text

129. Delatour (1983b), pp.38-43.return to text

130. Capital Consult (1983), p. 13.return to text

131. Banque Mondiale (1985), p.37.return to text

132. Delatour (1983b), p. 34.return to text

133. Ibid., pp.48-49.return to text

134. Garrity (1975), p.183. The agreement was signed in February 1944 (ibid., p.186).return to text

135. Ibid., pp. 183, 195.return to text

136. For a discussion of how the need for integration affects the negotiation strategies of the bauxite companies, see Girvan (1971).return to text

137. No tariffs are levied on bauxite imports by industrial countries (Giraud (1983), p.403). Regarding alumina, only the EEC countries are charging a tariff that amounts to 5,6 percent (ibid.).return to text

138. An alumina plant should produce at least 400,000 tons per year to reach the scale that corresponds to the lowest possible production costs (ibid., p. 409). The extraction volume in Haiti – about 700,000 tons during the 1969-82 period (see Figure 7) – made the location of such a plant in Haiti uneconomical because five tons of bauxite give two tons of alumina and an annual extraction of one million tons was, thus, required to reach the minimum efficient scale.return to text

139. Adams and Behrman (1982), p.60.return to text

140. Garrity (1975), p. 194.return to text

141. Adams and Behrman (1982), p.60.return to text

142. Garrity (1975), p.209.return to text

143. Bosson and Varon (1977), p.123 and Girvan (1987), pp.731-32.return to text

144. Gordon-Ashworth (1984), pp. 261-62. In the case of Haiti, Reynolds agreed to a new levy in December 1974 which made for a government revenue per ton which was about eleven times as high as the one prevailing at the time (Garrity (1975), p.215).return to text

145. Giraud (1983), p.434 and Girvan (1987), p.731.return to text

146. On the size and geographical distribution of world bauxite reserves, see Giraud (1983), p.430.return to text

147. Garrity (1975), p.230.return to text

148. Calculated from the sources in Figure 7.return to text

149. US/AID (1982), p.63.return to text

150. Garrity (1975), p.201.return to text

151. Girault (1981), Table IV-6, p.77.return to text

152. See Delatour and Voltaire (1980), Table 1.5, p. 13. Note, however, that manufactured exports have stagnated since the very end of the 1970s (International Monetary Fund (1986), pp. 10, 45). A critical view of the potential of manufacturing exports is provided by De Wind and Kinley (1988), Chapter 5.return to text

153. Haiti Progres (1987).return to text

154. Adams and Behrman (1982), p.58.return to text

155. The reason for this failure is that cane varieties which are rich in sugar are collected by people for direct consumption (Delatour (1983a), p.66).return to text

156. Personal corumunication from Giovanni Caprio, December 16, 1988.return to text

157. World Bank (1986), p.140.return to text

158. Hewitt (1987), p.629.return to text

159. That marketing and price public policy has been detrimental to primary exports from Haiti is further illustrated by the experience of the Regie du Tabac in the sugar sector (cf. above).return to text

160. On the fiscal reform, see World Bank (1985), p. 19 and Minister of Economy and Finance (1986).return to text

161. Seguino (1987), p.1 and Bulletin Trimestriel de Conjuncture (1987), p.44.return to text

162. Latin America Bureau (1985), p.46.return to text

163. In the short run, the main effect on output is through more careful harvesting of cherries.return to text

164. Capital Consult (1983), p. 13.return to text

165. See for example Girault (198 1), Seguino (1985), Kermel-Torres (1983) and Francisque (1986).return to text

166. Cf. Bourdet and Lundahl (1988) for an overview.return to text

167. Ibid.return to text

168. On the relation between market share instability and competition, see Caves and Porter (1978), pp.289-313 and McGuckin (1972), pp.363-70.return to text

169. Franzel and Martin (1985), p. 62.return to text

170. See Schwob (1982).return to text

171. Delatour (1983b), pp. 57-58 and UNIDO (1986).return to text

172. CNRS (1985), planche 16.return to text

Financial assistance from SAREC and SIDA is gratefully acknowledged. Thanks are due to Giovanni Caprio for constructive comments to a draft version.



Abdel-Rahman, A. M. and Byrne, J. (1979), Taxation in Haiti. Washington: International Monetary Fund

Adams, Gerard and Behrman, Jere (1982), Commodity Exports and Economic Development. Lexington, Mass.: Lexington Books

Banque Interaméricaine de Développement (1974), Possibilités d’investissement et développement du secteur rural en Haïti. Washington, D.C.

Banque Mondiale (1985), Haïti. Etude du secteur agricole. Volume I. Rapport Principal. Rapport No. 5375-HA, 14 June. Washington D.C.

Bosson, Rex and Varon, Bension (1977), The Mining Industry and Developing Countries. London: Oxford University Press

Bourdet, Yves and Lundahl, Mats (1988), Haitian Coffee Marketing Revisited. Mimeo, Department of International Economics and Geography, Stockholm School of Economics

Bulletin Trimestriel de Conjoncture (1987), No. 2, Port-au-Prince: Ministère de l’Economie et des Finances

Bulletin Trimestriel du Ministère du Commerce et de l’Industrie (1986), No. 22, October, Port-au-Prince

Capital Consult S.A. (1983), An Analysis of the Haitian Coffee Sector: Towards New Policy Directions. Mimeo, Port-au-Prince

Caprio, Giovanni (1979), Haiti – wirtschaftliche Entwicklung und periphere Gesellschaftsformation. Frankfurt am Main: Haag und Herchen

Caves, Richard and Porter, Michael (1978), Market Structure, Oligopoly and Stability of Market Shares, Journal of Industrial Economics, Vol. 26

CEPAL (1986), Rapport destiné à l’étude sur l’économie de l’Amérique Latine et les Caraïbes, Haïti, LC/MEX/R.46. Mexico

Cloutier, Jean-Pierre (1987a), ‘HASCO Closes Its Doors: Bitter Times for the Haitian Sugar Industry’, Haiti Times, Vol. 4, No. 4, May

Cloutier, Jean-Pierre (1987b), ‘Tourism Industry: Hope and Hard Work’, Haiti Times, Vol. 4, No. 4, May

CNRS (1985), Atlas d’Haïti. Talence: Centre d’Etude de Géographie Tropicale (CNRS) et Université de Bordeaux 3

Cook, Cristanna (1984), Issues of Importance to the Production and Marketing of Cocoa in Haiti, Mimeo, Department of Agricultural and Resource Economics. College of Life Sciences and Agriculture. University of Maine, Orono

Delatour, Leslie (1983a), The Centrifugal Sugar Sector in Haiti. Mimeo, Port-au-Prince: US/AID

Delatour, Leslie (1983b), The Essential Oil Export Sector of Haiti: Trends, Problems and Policy Options. Mimeo, Port-au-Prince: US/AID

Delatour, Leslie (1984), Cotton Production in Haiti. Mimeo. Port-au-Prince: US/AID

Delatour, Leslie and Voltaire, Karl (1980), International Sub-Contracting Activities in Haiti. Mimeo. UNCTAD-UNDP project # RLA-78-019; IDB project TC-78-06-17-0. Chicago

De Wind, Josh and Kinley, David H. III (1988), Aiding Migration. The Impact of International Development Assistance on Haiti. Boulder and London: Westview Press

Douzant-Rosenfeld, Denise (1988), ‘Haïti zone franche: les effets du commerce ouvert’, Problèmes d’Amérique Latine, No. 87

Fatton, Bernard (1975), Eléments d’information sur la production et la commercialisation du sisal en Haïti. Mimeo. Port-au-Prince: Institut Interaméricain des Sciences Agricoles de l’OEA, (IICA)

Franzel, Steven and Martin, Marc-Eddy (1985), Selected Issues in the Production and Marketing of Coffee in Haiti. Mimeo. Washington D.C.: Development Alternatives, Inc.

Francisque, Edouard (1986), La structure économique et sociale d’Haïti, Port-au-Prince: Imprimerie Henri Deschamps

Fry, Maxwell J. (1988), Money, Interest, and Banking in Economic Development. Baltimore and London: Johns Hopkins University Press

Gaillard, Roger (1982), Les blancs débarquent. Tome VI. Charlemagne Péralte le caco. Port-au-Prince: Presses Nationales, Imprimerie Le Natal

Garrity, Monique (1975), ‘The Multinational Corporation in Extractive Industries: A Case Study of Reynolds Haitian Mines, Inc.’, in: Sidney W. Mintz (ed.), Working Papers in Haitian Society and Culture. New Haven: Antilles Research Program, Yale University

Gilbert, Christopher (1987), ‘International Commodity Agreements: Design and Performance’, World Development, Vol. 15

Gillis, Malcolm; Perkins, Dwight H.; Roemer, Michael and Snodgrass, Donald R. (1987), Economics of Development. Second edition. New York and London: Norton

Giraud, Pierre-Noël (1983), Géopolitique des ressources minières. Paris: Economica

Girault, Christian (1981), Le commerce du café en Haïti. Paris: Editions du Centre National de la Recherche Scientifique

Girvan, Norman P. (1971), ‘Making the Rules of the Game: Company-Country Agreements in the Bauxite Industry’, Social and Economic Studies, Vol. 20

Girvan, Norman P. (1987), ‘Transnational Corporations and Non-Fuel Primary Commodities in Developing Countries’, World Development, Vol. 15

Gordon-Ashworth, Fiona (1984), International Commodity Control, A Contemporary History and Appraisal. London: Croom Helm

Guillaumont, Patrick (1982), Prix relatifs des produits primaires et développement. Paris: Editions du CNRS

Haïti Progrès (1987), ‘Contrat scandaleux! Le CNG brade notre patrimoine national’, 13 au 19 mai

Hansson, Göte and Lundahl, Mats. (1983), `The Rybczynski Theorem under Decreasing Returns to Scale’, Scandinavian Journal of Economics, Vol. 85

Helfenberger, André; Bailey, Martial; Azael, Ariel and Mathurin, Fequière. (1987), Assessment of the Situation of Coffee Production and Marketing in Haiti. Mimeo, Port-au-Prince: US/AID

Hewitt, Adrian (1987), ‘Stabex and Commodity Export Compensation Schemes: Prospects for Globalization’, World Development, Vol. 15

Hirschman, Albert O. (1958), The Strategy of Economic Development. New Haven and London: Yale University Press

Hirschman, Albert O. (1977), ‘A Generalized Linkage Approach to Development, with Special Reference to Staples’, in: Manning Nash (ed.) (1977), Essays in Economic Development and Cultural Change in Honor of Bert F. Hoselitz. Supplement to Economic Development and Cultural Change, Vol. 25

Institut Haïtien de Statistique (1977). Département de Finances et des Affaires Economiques. Guide économique de la République d’Haïti. Port-au-Prince

International Monetary Fund (1986), Haiti – Recent Economic Developments. SM/86/294. Washington D.C.

Jean-François, Philippe (1987), ‘Vérité sur la fermeture de la HASCO’, Journal du Commerce, No. 58, 4 May.

JWK International Corporation (1976a), Agricultural Policy Studies in Haiti: Sisal. Damien: Département de l’Agriculture, des Ressources Naturelles et du Développement Rural

JWK International Corporation (1976b), Agricultural Policy Studies in Haiti: Cotton. Damien: Département de l’Agriculture, des Ressources Naturelles et du Développement Rural

JWK International Corporation (1976c), Agricultural Policy Studies in Haiti: Meat. Damien: Département de l’Agriculture, des Ressources Naturelles et du Développement Rural

JWK International Corporation (1976d), Agricultural Policy Studies in Haiti: Mangoes. Damien: Département de l’Agriculture, des Ressources Naturelles et du Développement Rural

Kermel-Torres, Doryane (1983), L’économie des huiles essentielles en Haïti, Etude de géographie humaine sur la production et la commercialisation. Thèse de doctorat, Université de Bordeaux III

Latin American Bureau (1985), Haiti: Family Business. London

Lebeau, Francis (1974), Rural Sector Assessment of the Republic of Haiti. Mimeo, Port-au-Prince: US/AID

Little, Ian; Scitovsky, Tibor and Scott, Maurice (1970), Industry and Trade in Some Developing Countries, A Comparative Study. OECD Development Centre. London: Oxford University Press

Locher, Uli (1988), ‘Land Distribution, Land Tenure and Land Erosion in Haiti’, Paper presented at the Twelfth Annual Conference of the Society for Caribbean Studies, July 12 to 14, at the High Leigh Conference Centre, Hoddesdon, Hertfordshire, UK. Mimeo, Department of Sociology, McGill University, Montreal

Lundahl, Mats (1979), Peasants and Poverty: A Study of Haiti. London: Croom Helm

Lundahl, Mats (1983), The Haitian Economy, Man, Land and Markets. London and Canberra: Croom Helm

Lundahl, Mats (1984), ‘Defense and Distribution: Agricultural Policy in Haiti during the Reign of Jean-Jacques Dessalines, 1804-1806’, Scandinavian Economic History Review, Vol. 32

Lundahl, Mats (ed.) (1985a), The Primary Sector in Economic Development. London and Sydney: Croom Helm

Lundahl, Mats (1985b), ‘Toussaint L’Ouverture and the War Economy of Saint-Domingue, 1796-1802’, Slavery and Abolition, Vol. 6

Lundahl, Mats (1985c), ‘Government and Inefficiency in the Haitian Economy: The Nineteenth Century Legacy’, in: Michael B. Connolly and John McDermott (eds.), The Economics of the Caribbean Basin. New York: Praeger

McGuckin, Robert (1972), ‘Entry, Concentration Changes and Stability of Market Shares’, Southern Economic Journal, Vol. 38

McKinnon, Ronald I. (1973), Money and Capital in Economic Development. Washington, D.C.: Brookings Institution

Maizels, Alfred (1987), ‘Commodities in Crisis: An Overview of the Main Issues’, World Development, Vol. 15

Mamalakis, Markos. (1965), ‘Public Policy and Sectoral Development. A Case Study of Chile 1940-1958’, in: Markos Mamalakis and Clark Winton Reynolds, Essays on the Chilean Economy. Homewood: Richard D. Irwin

Minister of Economy and Finance (1986), Memorandum of Economic and Financial Policy. October 1986-September 1989. Mimeo, Port-au-Prince: Ministère de l’Economie et des Finances

Murphy, Martin J. (1988), ‘Dominican Responses to the Contemporary World Sugar Crisis’, Paper presented at the symposium Sugar Crisis in the Caribbean, International Congress of Americanists, 4 July 1988, Amsterdam.

Rybczynski, Tadeusz M. (1955), ‘Factor Endowment and Relative Commodity Prices’, Economica, Vol. 22

Schwob, Roger (1982), Situation et problèmes de l’agro-industrie des huiles essentielles en Haïti. ITC/DIP/65. Geneva: International Trade Center, UNCTAD/GATT

Seguino, Stephanie (1985), The Haitian Coffee Market: The Export Tax Reduction and other Issues of Market Competitiveness. Mimeo, Port-au-Prince: US/AID

Shaw, Edward S. (1973), Financial Deepening in Economic Development. New York: Oxford University Press

Stevens, Chris and Webb, Carole (1983), ‘The Political Economy of Sugar: A Window on the CAP’, in: Helen Wallace, William Wallace and Carole Webb (eds.), Policy Making in the European Community. Second Edition, Chichester: John Wiley & Sons

Trouillot, Michel-Rolf (1982), ‘Motion in the System: Coffee, Color and Slavery in Eighteenth-Century Saint-Domingue’, Review, Vol. 5

UNCTAD (1969), The Kennedy Round Estimated Effects on Tariff Barriers, Geneva

UNIDO (1986), Technical Report: Project Profiles for the Manufacture of Roasted-and-Ground Coffee and Soluble Coffee. DP/HAI/84/013. Vienna

US/AID (U.S. Agency for International Development) (1982), Food and Agriculture Sector Strategy for Haiti, Final Report. Mimeo, Port-au-Prince

World Bank (1985a), Haiti: Policy Proposals for Growth. Report No 5601-HA. Washington D.C.

World Bank (1985b), Haiti Agricultural Sector Study. Volume II. Annexes A-F. Report No. 5375-HA, 14 June. Washington D.C.

World Bank (1986), World Development Report 1986. New York: Oxford University Press

Tables and Figures

Table 1

Composition of Haitian Primary Exports, 1951-1985
(percent of total exports)
Sources: Sugar and essential oils, 1976-81: Bulletin Trimestriel du Ministère du Commerce et de l’Industrie,
No. 22, October 1986, pp. 81, 104-105. Rest of table: 1951-74: Institut Haïtien de Statistique (1977),
pp. 278-89; 1975-85: Unpublished data from the Institut Haïtien de Statistique et d’Informatique.
Essential Oils

Table 2

Total Coffee Production
Exports and Domestic Sales 1950-1986
(averages, thousands of 60-kilo bags)
Sources: 1950-81: Sequino (1985), p. 104; 1982-86: Bulletin Trimestriel de Conjoncture, No. 2, 1987, p.47
Domestic Sales
Export Share (percent)

Table 3

Geographical Distribution of Haitian Coffee Exports 1976-1981
Source: Bulletin Trimestriel du Ministère du Commerce et de l’Industrie, No. 22, October 1986, p. 97.
United States
West Germany

Table 4

Relative Prices of Coffee and Subsistence Products 1952-1983
Sources: Exported coffee/red beans, exported coffee/corn: 1952-80: Capital
Consult (1983), p. 20; 1981-83: Seguino (1985), p. 104; domestically marketed
coffee/red beans, domestically marketed coffee/corn: calculated from figures
in Capital Consult (1983), p. 20; exported coffee/domestically marketed coffee:
ibid., p. 20.
Exported Coffee
Domestically Marketed Coffee
Exported Coffee
Red Beans
Red Beans
Domestically Marketed Coffee

Figure 1

Table 5

Cocoa Production
Exports and Domestic Sales 1951-1985
(thousands of metric tons)
*1950 **1960 ***1970 ****1976-79
Sources: Exports: 1951-82: Cook (1984), pp. 2-3; 1983-85: Unpublished data
from Institut Haitien de Statistique et d’Informatique. Production: 1950-70:
Banque Interamericaine de Developpement (1974), p. 142; 1976-79: Cook
(1984), p. 5; 1981-85: CEPAL (1986), p. 2.
Domestic Sales

Figure 2

Table 6

Geographical Distribution of Haitian Cocoa Exports
Source: Bulletin Trimestriel du Ministère du Commerce et de l’Industrie, No. 22, October 1986, p. 98.
United States
West Germany

Figure 3

Table 7

Sugar Content of Cane
(tons of cane per ton of raw sugar)
Source: Based on Delatour (1983a), Tables VIII and IX, pp. 33-34.

Table 8

Geographical Distribution of Haitian Sugar Exports
Source: Bulletin Trimestriel du Ministère du Commerce et de l’Industrie, No. 22,
Octobre 1986, pp. 105-06.
United States

Figure 4

Figure 5

Table 9

Essential Oil Exports: Shares of the Three Main Products
Source: Based on Delatour (1983b), Table 2, p. 8.
Three Products

Figure 6

Table 10

Geographical Distribution of Exports of Essential Oils
Source: Bulletin Trimestriel du Ministère du Commerce et de l’Industrie, No. 22,
October 1986, p. 104.
United States
United Kingdom
West Germany

Figure 7

Table 11

Geographical Distribution of Coffee Production
The northern region comprises Cap-Haiten, Port-de-Paix, and Gonaives. The
central region comprises Saint-Marc, Croix-des-Bouquets and Port-au-Prince.
The southern region comprises Petit-Goave, Jeremie, Les Cayes and Jacmel. Source: Girault (1981), p. 77.
1956-1958 (average)
1972-1975 (average)

Table 12

Estimates of Long-Run Demand Elasticity of Three Haitian Export Products
Source: Adams and Behrman (1982), p. 58.
Coffee Price
Cocoa Price
Sugar Price

Table 13

The Composition of Government Revenue, 1974-1986
(percent of total revenue, averages)
* figures for 1986 are preliminary
** only 1985
Sources: 1974-78: Abdel-Rahman and Byrne (1979), p.16; 1980-84: World Bank
(1985), p. 162; 1985-1986: International Monetary Fund (1986), p. 73.
Taxes on Net Income and Profits
Taxes on Property
Taxes on Goods and Services
Value Added Tax
Taxes on International Trade
      Import Duties
      Coffee Export Duties
      Bauxite Export Taxes
Other Revenue and Taxes
Total Tax Revenue