Copyright Sociological Research Online, 2000

Anne Margrethe Brigham (2000) 'Land Reform: A Foundation For Industrial Growth In Developing Countries?'
Sociological Research Online, vol. 5, no. 3, <http://www.socresonline.org.uk/5/3/brigham.html>

Abstract

In this paper I examine whether or not land reform can contribute to growth in the agricultural surplus of developing countries. There are two possible channels where we might expect an effect: the first is agricultural productivity and the second is "on farm consumption". I start with an introduction to the theories on the relationship between agrarian structure and farm productivity. The theoretical literature on the effects of land reform on productivity is inconclusive. Therefore, I perform an empirical test on the relationship. I find that land reform is at least not detrimental to productivity. Next I provide a short summary of the literature on the relationship between land reform and "on farm consumption". This literature is more conclusive. My conclusion suggests that land reform can have a positive effect on agricultural surplus (and thereby on industrialization) through both growth and redistribution.

Keywords:

Agricultural Surplus; Chile; Developing Countries; El Salvador; Labour Productivity; Land Productivity; Land Reform; Nicaragua; On Farm Consumption"; Peru.; Philippines; The Domoinican Republic; Zimbabwe

Introduction

1.1

One of the main differences between the developed and the developing countries besides the gap in per capita income is the relative size of the industrial and the agricultural sectors of the economy (Bhaduri 1993:150). While less than five percent of the population in the former group of countries are employed in the agricultural sector, the percentage can be up to 80 in many of the latter (Jazairy et al. 1992). As labor productivity is generally higher in the industrial sector than in the agricultural, this characteristic can to a large extent explain the relatively low income per capita in the developing countries. According to Lewis (1954), Fei and Ranis (1964; 1966) and Kaldor (1989) "disguised unemployment" exisist in developing country agriculture. By this they mean that the same output could be produced with less people, but otherwise unemployed people are offered work as a form of "social security". A transfer of people (with low labor productivity) from the agricultural sector, to jobs in the industrial sector (where they can be more productive) will thus reduce rural poverty, as the same agricultural output can be divided among less people. For this transfer to be possible, growth in the industrial sector is needed.

1.2

The majority of developing countries need to import production capital (machinery) for the industrial sector, but most of them operate under serious constraints on their balance of payments. Thus agriculture has an important role to play in the industrialization process as suppliers of foreign exchange, as well as of raw materials and food. Industrial growth is thereby dependent on growth in the agricultural surplus (which is the share of the agricultural production that is not consumed within the agricultural sector) (Bhaduri 1993, Skarstein 1997:58-60; Lewis 1958:23). Since most developing countries have a heavily skewed distribution of ownership to agricultural land, a crucial question is whether it is possible to increase the agricultural surplus and improve (or at least not deteriorate) the conditions of the rural poor at the same time.

1.3

In the wake of the Green Revolution there was widespread belief that agricultural growth would have a trickle down effect on the poor, -even where land was concentrated in the hands of a small elite. The logic behind this belief was that the appliance of new technology would boost production and reduce food prices, and thus give the poor access to more food. This enthusiasm was, however, based solely on availability and not "entitlements" (Osmani 1993: 1) ^[1]. When there in the last decade has been renewed interest in land reform as a poverty alleviating policy, it is a result of growing acceptance of the fact that poverty cannot be alleviated unless there is some kind of redistribution of wealth and income (Tyler et al. 1993:3; Alexandratos 1995: 318).

1.4

But, it is far from certain that land reform is a panacea for alleviating poverty. Disruptions in credit and marketing relations, management practices etc. can lead to reduced agricultural production, which may leave beneficiaries worse off than they were before, - and in addition reduce the agricultural surplus available for industrialization. The poverty alleviating effect -in both the agricultural and the industrial sector- is questionable if the agricultural surplus available for industrialization decline as a result of a drop in production. Moreover, if production levels are kept up, and the living standard of the beneficiaries improve, it will most likely lead to an increase in the farmers own consumption of agricultural products, and the marketed surplus may decline even if the production is not disrupted. Thus, there may be a conflict of interest between the beneficiaries of the reform and the urban poor, as well as the rural poor that did not benefit from the reform (and may want to seek employment in the industrial sector).

1.5

In this paper I will examine whether or not land reform is beneficial to industrial growth. There are two possible channels where we might expect an effect, both of which act via agricultural surplus available for industrialization. The first is agricultural production, and the second is farmers' own consumption. In the following section I will start with an introduction to the theories on the relationship between agrarian structure and agricultural production. The theoretical literature on the effects of land reform on production is inconclusive. Therefore, I will perform an empirical test on the relationship in the latter part of the section. In this part I find that land reform is at least not detrimental to production. In the third section I will provide a short summary of the literature on the relationship between land reform and "on farm consumption". This literature is more conclusive. In the conclusion (in section four) I carefully suggest that land reform can have a positive effect on industrial growth through both growth and redistribution.

Land Reform and Agricultural Production

2.1

This section, which will discuss how the agricultural surplus can be affected by land reform through changes in production levels, has two parts. In the first part I will have a theoretical discussion on how production may be affected when large estates are divided into many small farms or transformed into cooperatives. In the second part I will perform a preliminary test on these theories by studying trends in agricultural production and productivity in seven countries that have been through a land reform. The results of this test indicate that land reform does not (always) have a negative affect on agricultural production.

Theoretical Perspectives

2.2

The level of agricultural production is determined by four main factors: land area, land productivity, number of workers, and labor productivity. Thus, agricultural production can increase as a result of colonization, population growth and improvements in land and labor productivity. I expect that only land and labor productivity will be influenced by land reform. In fact, one of the main arguments for land reform (the most important argument is that it promotes equity) has been that there is an inverse relationship between farm size and land productivity. According to this "inverse relationship theory", land reforms where large operating units are divided into smaller holdings will not only reduce poverty because of the redistribution of land from the rich to the poor, but also because it would lead to an increase in land productivity and thereby in agricultural production. The theory that small farms have higher land productivity than the larger ones can be traced at least as far back as to Adam Smith's legendary book "The Wealth of Nations" (1776). Based on a comparison of the conditions of large estates with smallholdings in their neighbourhood, he argued that small farms and were more productive mainly because they were more adaptive to new technology (Smith (1776) cited in El-Ghonemy 1990: 133). He claimed that "great proprietors were seldom great improvers, because their style of life (dresses staff, house, etc.) constituted a state of mind and habit that led to the neglect of their vast estates" (ibid.: 137).

2.3

Much has been written on the subject since (see Dyer 1991: 87-88 for a list of major contributors). There is a whole range of studies that support the theory of the inverse relationship, but there are also quite a few that come to the opposite conclusion (Ellis 1989; El-Ghonemy 1990:128). Among the studies that have found that small farms have higher land productivity, different causal explanations may be found. On the one side are those that suggest that land reform may be an effective development strategy. The most common of these explanations are that (Bharadwaj 1974: 11-31; Ellis 1989: 198-199; El-Ghonemy 1990: 128-132; Tyler et al 1993; Thiesenhusen 1995:10):

1. a larger part of the holdings are in productive use on small farms than on large farms;
2. small farms plant more crops per year than the large farms;
3. small farms plant higher value crops while large farms are often engaged in land extensive production (such as livestock);
4. a larger portion of the farms are irrigated on small farms than on the large ones;
5. labor intensity is higher on small farms;
6. large land owners often invest in land for other reasons than to use it as a productive resource, such as for social prestige and political power, or as a portfolio investment in periods with high inflation;
7. because small farms are often family run they have higher flexibility with regard to seasonal employment.

2.4

In addition we might expect increased labor productivity when large estates are divided into smaller, privately owned farms because the peasants will work for themselves and not for a landlord.

2.5

The issue of better motivation to work hard on family farms brings us to the most important argument against land reform as a development strategy. This argument is that the inverse relationship is a result of desperate families fighting for survival from too small pieces of land. By trying to make a living from inadequate holdings of land, family farms use the land much more intensively than their larger counterparts. Over time, such intensive farming may degrade the quality of the soil and give rise to even worse hardships. According to Dyer (1991:87) the inverse relationship "...arises from the desperate struggle of poor peasants for survival on below subsistence plots of land...(and the) ...redistribution of land on the basis of the inverse relation argument therefore, far from alleviating poverty and creating employment opportunities, will only deepen and perpetuate extreme levels of exploitation and poverty".

2.6

There are also, as I mentioned above studies that show that land productivity increases with farm size (Patnaik 1972; Dyer 1991). The causality behind this positive relation between farms size and farm productivity is most often explained as (Ellis 1989: 194-195; El-Ghonemy 1990: 123):

1. better educated managers on the larger farms;
2. division of labor and specialization on large farms, and also that;
3. larger farms can generate more revenues and credit and can therefore invest in more machinery and irrigation as well as fertilizers

.

2.7

Dyer (1991) argues that while there might have been an inverse relationship between farm size and land productivity before the Green Revolution, this relationship has broken down because large farms can generate the capital needed to keep up with technological developments while the small farms can not.

2.8

The effects of land reform on agricultural production do not only have to do with the relationship between the size of the farms and productivity. As often as not, land reforms consist of turning large privately owned farms into state- or communally owned cooperatives. In reforms where large estates are turned into cooperatives the operational size of the farm is usually kept more or less intact. For this type of reforms the productivity question is whether communally owned farms are more or less productive than privately owned estates.

2.9

The most apparent issue is whether labor productivity increases when the workers are (part) owners instead of wage laborers. But cooperatives may also want to put a larger part of the farmland into productive use and produce more crops per year, and thereby increase the productivity of land (Grimm 1998: 214).

2.10

While the cooperative idea has quite a few followers in developing countries, it has not been able to break through the belief in private property in the West. The most common perception in the western world is that the cooperatives are less productive because of free riders, incentives problems and difficulties of (co-)management. Management problems have been found to be pervasive in communally owned cooperatives that are managed by all owners communally. Examples from South Africa have shown that such cooperatives may easily become paralysed by disagreement, because shareowners feel threatened by sharing decision making, or because investment procedures are difficult to understand for people with only basic schooling. Productivity gains have also been forgone because the owners have avoided mechanization in order to keep up employment levels (Grimm 1998:214).

2.11

From the review above, we have seen that there is no clear answer to the question about the effects of land reform on the level of agricultural production in the theoretical literature. We will now study this debated relationship in a few historical cases of land reform.

Empirical Indications

2.12

In this section I will analyse trends in agricultural production as well as land and labor productivity in agriculture in seven developing countries where land reform has been implemented.^[2] By studying these trends before, during and after the reforms, we will get an indication as to whether the land reforms influenced agricultural production in the specific countries. We will also be able to study the effect of land reform on land and labor productivity, which we have seen constitute major forces behind changes in the level of agricultural production.

2.13

Since we are discussing the effects of land reform primarily from the point of view of its impact on agricultural surplus available for industrialization, the analyses will be performed on nationally aggregated data on production. Thus, we will not be able to consider whether the "beneficiaries" of the reforms have become better or worse off.

2.14

In line with the theoretical debate in the first part of this section, I have divided the countries into two groups according to the main characteristic of the land reform.^[3] The first type of reform is where the operational size of the affected farms has been reduced because large estates have been divided into small, privately owned farms. The countries studied in this group are Zimbabwe, The Philippines, and Chile after 1974. The second type is where the operational size is kept (more or less) intact but the ownership has been transferred from private to state or communal. The selected countries in this group are Nicaragua, El Salvador, the Dominican Republic, Peru and Chile between 1967 and 1974. We will look for common features resulting from the land reforms irrespectable of type, and we will study whether the trends have different patterns from one type to the other^[4].

2.15

Before we proceed with the analysis, it is in its place to give a caveat. Since land reform is a controversial policy because it challenges the vested interests of a (generally) politically powerful elite, a great deal of socio-political unrest is associated with its implementation. We may consequently be measuring effects of strikes, revolutions or civil wars, without being able to separate these from the effects of the land reforms. And since we will not control for population growth or colonization-or for climate, prices, or changes in (agricultural) policies that may also affect productivity of land and labor-the results of this empirical exercise can only be takes as indications that will have to be investigated further^[5].

Figure 1: Agricultural production trends in selected countries where the operational size has been reduced (Indices, 1979-1981=100)

a) Phillippines

b) Zimbabwe

c) Chile

Figure 2: Agricultural production trends in selected countries where estates were turned into cooperatives (Indices, 1979-81)

a) Nicaragua

b) El Salvador

c) The Dominican Republic

d) Peru

Source: FAO (1995)

Note: Where the "per 1000 ha." trend line is not visible, it lies under the total trend line.^[6]

2.16

Attending to the examination of the reforms, we will start with figure 1a) which displays the trends in agricultural production and the productivity of land and labor before, during and after the 1988-reform in the Philippines. We can see that both the production and the productivities decreased in the years immediately before the reform, but by the time the reform was initiated, they had climbed to pre-unrest levels and have shown a relatively steady growth until 1995. There was a decline in all three trends in 1991, which may possibly be linked to the eruption of Mount Pinatubo the same year.

2.17

Looking at the Zimbabwean trends in figure 1b) we see a somewhat different pattern: After a sharp increase in production in the year of the reform, there has been an upward trend in the agricultural production disrupted by rather large drops for the whole period. After the land reform in 1980 the scope of these variations increased, but production in good years have been higher after 1980 than before the reform. Since Zimbabwean agriculture is largely rain-fed, draughts that occurred in the same years may explain the relatively large drops in production in 1983 and 1992 (FAO 1995; World Bank 1997:241). Labor productivity has however declined since the reform, while the productivity of land has varied in pace with total production. This means that increases in the total agricultural production after the reform in Zimbabwe is mostly a result of better productivity of the land. This finding is in line with the (inverse relationship) theory described above, since the Zimbabwean reform is of the type where large estates have been broken up into smaller, privately owned farms.

2.18

The Chilean reform (figure 1c) is hard to classify as it started out as a cooperative reform under Allende in 1967^[7], but was soon countered by privatization after the military coup in 1973. When the military regime took power in 1973, they annulled the land reform laws and returned 28 % of the reformed land to the previous owners, 5% was auctioned, 10% went to the public sector for later to be sold to individuals, while 57% remained in the reform sector (Thiesenhusen 1995:110). It may therefore be fruitful to study trends between the reform in 1967 and the contra reform in 1973, and from 1973 to 1994 separately.

2.19

The trends in agricultural production between 1967 (when Allende came to power and strengthened the commitment to the ongoing land reform) and 1974 show an almost unchanged total production, while land productivity fell slightly and labor productivity increased correspondingly. There was however a relatively large drop in the production in 1974 which can probably be attributed to the political turmoil of the military coup. In the period after the counter reform, the total production and the productivity of both labor and land have practically doubled compared to the levels during the "Allende reform period".

2.20

Although these three cases represent the same type of reform, they reveal quite different trends, but neither of the three experienced a long-term reduction in the aggregate production.

2.21

Moving our attention to figure 2, and the collectivization type of reforms, we see that there was a large drop in the agricultural production in Nicaragua (figure 2a) in the two years before the land reform was initiated in 1981. We can also see that the production per unit of land under permanent crops and pasture declined, and more so than the decline in labor productivity.

2.22

This sharp drop in agricultural production can be explained by the violent revolution that took place in 1979. During the insurrection, 35 000 people died and 10 000 were seriously injured. GDP per capita fell by 35 per cent between 1977 and 1979, and the country's infrastructure was severely damaged. Output of traditional peasant grown crops such as beans and maize plummeted, and the managers on many of Samosa's vast holdings abandoned the farms. Sugar production dropped by 14 percent, milk by 50 per cent and cotton by 82 per cent (Thiesenhusen 1995:127). In the nearly twenty years since the land reform, the production has not restored it's volume from the pre-reform era, but it has gained some momentum during the last couple of years. The lasting low productivities compared to pre-revolutionary levels have been attributed to a combination of continued conflict (between the Sandinistas and the US backed contra revolutionary Contras), insecurity of property relations, lack of investment, shortage of labor and distrust of Sandinista policies (Ibid.:131).

2.23

The picture is a rather different one in the El Salvadorian case (Figure 2b). In this country there was a drop in agricultural production as well as the productivity of land and labor in the immediate years after the reform in 1980. This decrease coincides with the year of the killing of 10.000 campesinos by right wing death squads, and the on start of overt civil war. But the production level was restored within one decade, and climbed above the pre-reform levels in 1992 and 1993. We can also see that labor productivity increased slightly more than land productivity until 1992, then the relationship was reversed. There was however a decline in all of the three trends in 1994, and we cannot from these data tell whether it is temporary or more permanent.

2.24

The trends in The Dominican Republic (figure 2c), and in Peru (figure 2d) during the period after the reforms in 1972 and 1968, respectively, are somewhat similar: unsteady growth in total production and productivity, and lower growth in labor productivity than in land productivity. Also similar is the absence of a large drop in production in the immediate years after the land reforms in these countries. This may be explained by the fact that these land reforms did not result from civil wars, which in other cases-like in Nicaragua and El Salvador-have destroyed economic and social structures, as well as human resources.

2.25

While keeping in mind that there are too few cases to generalize to the whole population of land reforms, there are some aspects that are common for all or most of these cases. The first, surprising common denominator is that five out of the seven (all except for Nicaragua and El Salvador) did not experience sharp declines in agricultural production immediately after the reforms. The second common denominator is that after the reforms the agricultural production has climbed to levels higher than those experienced before the reforms, in all countries except Nicaragua. The third general feature is increased fluctuation in production between years. This is most striking in the Zimbabwean case, where the reform was of the type where large estates were divided into smaller individually held units, but it is also evident in El-Salvador, the Dominican Republic and Peru. The reason for this may be that smaller farms and cooperatives are unable to set aside capital needed to invest in technology to mitigate the effects of changes in climatic conditions^[8].

2.26

We were also looking for two recognizably different patterns in the three trends according to the type of reform, but this was not found.

2.27

If we were to draw any conclusion from this comparative exercise it would be that agricultural production will not necessarily fall (in the long run) after land reforms of these two types. And, the two types of reform do not have a clear cut different effect on land and labor productivity, or on agricultural production. But, as we do not know what the counterfactual outcome might have been, we do not know if the production and the productivies would have been even higher or lower if these land reforms had not taken place. We can, however, say that the trends in these seven countries have shown that agricultural surplus will not necessarily decline as a consequence of lower production after land reforms. In the next section we will turn our attention to how the agricultural surplus can be affected by changes in farmers own consumption after land reform.

Land Reform and "On Farm Consumption"

3.1

The analysis above has only taken us one step along the way in our discussion of how land reforms may affect the agricultural surplus available for industrialization. In this section we will discuss how the consumption of agricultural products in the agricultural sector can be influenced by changes in land property rights. We are mainly interested in how much of a farm's produce, whether its food, wood or the like, the owners of the farm retain for themselves. Of additional interest are increases in the demand for agricultural products from other producers within the agricultural sector. The level of consumption of all types agricultural products within the agricultural sector is of importance for how much food and raw materials that will be available for the industrial sector.^[9]

3.2

According to Lippit (1978:58) and Barraclough (1991: 129) income elasticity for food is high where consumption is very low, and relatively low where consumption is high. Consequently, it is likely that beneficiaries of land reform who (in many cases) were living near subsistence levels before the reform will increase their consumption of agricultural products.

3.3

It has proved difficult to find data on the type of consumption we are looking at, but an example from the Chinese land reform in the 1950's supports the above theory. From Table 1 (below) we can see that while grain production increased in China during the height of the collectivization process, the surplus of grain available for the urban sector declined.^[10]

Table 1: Surplus of grain available for rural areas in China, 1954/55-1956/57. (Millions of tons)

		1954/55	1955/56	1956/57
Total grain output		169,5	183,9	192,7
Total collections and purchases		53,9	52,0	49,9
Re-sales to rural areas		24,7	20,2	24,5
Available for urban consumption, exports and government stockpiles		29,2	31,8	25,2

Source: Eckstein,1966:312

3.4

In the above section we saw that the land reform in Nicaragua led to a drop in agricultural production for a period of at least twenty years. In this and similar cases, the beneficiaries of reform may not have been enabled to increase food consumption. But, in the other cases, where production levels were quickly restored, it is probable that the beneficiaries retained more of the farm output for themselves. Whether this reduction will be in absolute terms, depends on the relation between the effects of the reform on productivity relative to the effects on "on farm consumption".

3.5

If the overall economic development strategy for a country is to move people with low marginal productivity in the agricultural sector to more productive jobs in industry, there may be a conflict of interest between an improved standard of living in the agricultural sector and industrial growth. First, an increase of "on farm consumption" may reduce the surplus available for the industrial sector. Second, in countries with limited ability to import food and raw materials because of balance of payments constraints, a reduced surplus may lead to higher prices on food and raw materials. Higher prices on food can induce wage claims, which in connection with more expensive raw materials may reduce industrial profits and investments. Third, since the bulk of many developing countries' exports are agricultural products, increased "on farm consumption" may curtail the export capabilities and thereby reduce the capability to import capital goods for industrialization.

3.6

A common argument against the more equitable distribution of income that is sought through land reform is that capital accumulation provided by the large landowners will disappear, and consequently reduce investments in industry. This argument is based on the assumption that the rich do not consume much more than the poor, but save (and invest) most of their income. Studies from Chile have, however, shown that large landowners spent about 84 percent of their income on consumption of foreign travels and imported articles, while only 14 percent was invested. Of these investments, considerable portions were in land, foreign stocks and bonds, or in the construction of luxury hotels and apartments (Barraclough 1973: 30-31).

3.7

There is an additional effect of land reform on industrialization that we have not discussed yet, because we have only focused on the supply side of the relationship between the two sectors. But, insofar as the reforms serve to redistribute income from the rural rich to the rural poor, the demand for industrial products from the agricultural sector will most likely rise (Barraclough 1973:30). An "internal market" strategy for industrial growth, which relies on increasing demand from the country's poor, will give more immediate poverty alleviation than an export led strategy, where the industrial sector's need for food and raw materials are favoured above increased consumption in the agricultural sector.

3.8

We have seen that there is wide agreement in the literature that land reform will, if production is not too severely disrupted, lead to increased "on farm consumption". This will reduce the agricultural surplus compared to if consumption levels had not risen. Whether or not this reduction will be in absolute terms, depends on the post reform productivity.

Conclusion

4.1

This essay has discussed some of the ways in which land reforms in developing countries may affect the agricultural surplus available for industrialization. In the first section we looked into the relationship between property rights in land and the productivity of land and labor. The questions at issue were if small farms have higher productivity than large farms, and if privately owned farms are more productive than communally owned farms. The comparative study of production trends in seven countries that have implemented land reform since the 1960's could not provide any firm conclusions regarding these relations. We were, however, able to make the partial conclusion that the agricultural production in a country does not necessarily decrease after land reforms, neither when the reforms are of the type of parcelization into privately owned smallholdings, nor of the collectivization type.

4.2

The way in which land reforms may affect the agricultural surplus through increased "on farm consumption" was discussed in the second section. The dominating theory about this relationship is that the lower the initial levels of food consumption, the higher the expected increase in "on farm consumption".

4.3

In conclusion we can see two mechanisms that may induce a conflict between land reform and industrial growth via the former's impact on agricultural surplus. The first mechanism is that the agricultural surplus available for industrialization may decline if productivity suffers from the reform. The second is that the agricultural surplus may decline even when productivity is kept up or improved, if the consumption of agricultural products in the agricultural section increases. Of these two effects, the former is the most problematic because if productivity falls (enough), neither the industrial sector nor the rural poor will benefit from the reform. If agricultural output is sustained or increased by land reform, but the agricultural surplus declines as a result of increased "on farm consumption", the effect is less serious because at least the rural poor will have increased their standard of living.

Notes

¹Entitlements are the ways by which a person's rights to food (and other commodities) are determined (Sen, 1981).

²The sample in my analysis is constrained by the availability of time series data. Data on aggregate agricultural production that are comparable over time is only available from 1961. This is why important reforms such as the ones implemented in Mexico, Bolivia, Taiwan, Japan, South Korea, and Kenya are left out of the analysis (as well as the Ethiopian case where there is almost a total lack of data).

³A typology of reforms into these two distinct groups is somewhat problematic, because most reforms are a mix of the two, and because the nature of the reforms often change over time. I have classified the countries according to the most common ownership structure in the reformed sector in the first decade after the reform. The sources for my classification are Bruce (1998) for Zimbabwe, El-Salvador, Peru, and Chile, Adams (1995) for the Philippines, Dorner (1992: 40- 41) and Stanfield (1989: 334) for the Dominican Republic, and Thiesenhusen (1995: 87-114 and 128-129) for Chile and Nicaragua.

⁴To be defined as a land reform, the ownership to a substantial part of the agricultural land in a country must have been redistributed in either of these two ways as a result of political intervention (as opposed to by the "invisible hand" or by manipulation of taxes or factor prices. The source as to whether reforms have been substantial is Thiesenhusen (1989: 10), Adams (1995) and Bruce (1998: 41). The year of implementation of the reforms refers to the year where the reform law was passed. For the analytical purpose of studying the effects of the reforms, it may, artificial to use this year as break-off point, as the implementation of the reforms may stretch out in time. We will keep this in mind when studying the trends.

⁵In order to control for these variables I will need to compare the productivity trends from the part of the country's agricultural land that has been "reformed", with the trends in the rest of the agricultural sector. I am now in the process of collecting such data, and the results of my analysis will be presented in my PhD thesis on the effect of land reform on food security in developing countries.

⁶Source: FAO 1995 (Time series data for "The State of Food and Agriculture". 1995. FAO, Rome)

⁷The reform process started earlier, but did not redistribute enough land to be considered significant until 1967.

⁸The question of optimum farm size is an important issue, not only in relation to weather, but also with regard to the type of farming (extensive vs. intensive, dry land vs. irrigation, etc) as well as input and market price fluctuations.

⁹For simplicity, I will refer to these two types of consumption combined when I use the terms "on farm consumption" and "consumption of agricultural products in the agricultural sector".

¹⁰The urban sector does not necessarily coincide with the industrial sector, because there may be industrial activity in rural areas too. The example may non the less be illuminating to the discussion.

Acknowledgements

I would like to thank Rune Skarstein, Amit Bhaduri, Jonathon Moses and one anonymous reviewer for the useful comments on earlier drafts of this article. Obviously, any remaining shortcomings are my responsibility.

References

ADAMS. Martin. 1995. "Land Reform: New Seeds on Old Ground?" in Natural Resource Perspectives, Overseas Development Institute, London.

ALEXANDRATOS, N. 1995. "World Agriculture: Towards 2010". FAO, Rome.

BARRACLOUGH, S. 1991. "An End to Hunger?", Zed Books Ltd, London.

BARRACLOUGH, S. 1973. "Agrarian Structure in Latin America", Lexington Books, Lexington.

BHADURI, Ahmit. 1993. "Alternative Development Strategies and the Rural Sector", in Economic Crisis and Third World Agriculture, edited by A. Singh and H. Tabatai, pp149-178. Cambridge University Press, Cambridge.

BHARADWAJ, Krishna. 1974. "Production Conditions in Indian Agriculture. A Study Based on Farm Management Surveys", University of Cambridge, Department of Applied Economics, Occasional Paper 33, Cambridge University.

BRUCE, J. 1998. "Learning from the Comparative Experience of Agrarian Reform" Paper presented at The International Conference on Land Tenure in the Developing World, with a Focus on Southern Africa, University of Cape Town, January 1998.

DORNER, Peter. 1992. "Latin American Land Reforms in Theory and Practice: a Retrospective Analysis", University of Wisconsin Press, Madison.

DYER, Graham. 1991. "Farm Size- Farm Productivity Re-examined: Evidence from Rural Egypt", in The journal of Peasant Studies, Vol. 19, No. 1, pp. 59-92, Frank Cass, London.

EL-GHONEMY, M. Riad. 1990. "The Political Economy of Rural Poverty," Routledge, London.

ELLIS, Frank. 1988. "Peasant Economics. Farm Households and Agrarian Development", Cambridge University Press, Cambridge.

FAO. 1995. "SOFA 95". Time-series data for The State of Food and Agriculture. FAO, Rome.

FEI, John C. H. and RANIS, G. 1964. "Development of the Labor Surplus Economy". Richard D. Irwin, Homewood, Illinois.

FEI, J. C. H. and RANIS, G. 1966. "Agrarianism, Dualism and Economic Development, in: Adelman, Irma and Eric Thorbecke (eds), The Theory and Design of Economic Development, John Hopkins Press, Baltimore.

GRIMM, Josef. 1998. "Opportunities for Formal and Informal Tenure Systems for Farm Workers", Paper presented at The International Conference on Land Tenure in the Developing World, with a Focus on Southern Africa, University of Cape Town, January 1998.

JAZAIRY, Idriss, Mohiuddin Alamgis and Theresa Paniccio. 1992. " The State of Rural Poverty" The International Fund for Agricultural Development, New York University Press, New York.

KALDOR, Nicholas. 1989. "Equilibrium Theory and Growth Theory", in F. Targetti and A. P. Thirwall (eds): The Essential Kaldor, Duckworth, London, pp. 411-33.

LEWIS, W. Arthur. 1954. "Economic Development with Unlimited Supply of Labour", in The Manchester School, Vol. 22, No. 2, pp. 139-91.

LEWIS, W. Arthur. 1958. "Unlimited Labour: Further Notes", The Manchester School, vol. 26, no.1, pp. 1-32. Manchester.

LIPPIT, Victor D. 1978. " Economic Development in Meiji Japan and Contemporary China: a Comparative Study", in Cambridge Journal of Economics, Vol. 2, pp. 55-81, Academic Press Inc, London.

OSMANI, S. R. 1993. "Growth and Entitlements: The Analytics of the Green Revolution". UNU/WIDER, Helsinki.

SEN, Amartya. 1981. "Poverty and Famines. An Essay on Entitlement and Deprivation". Oxford University Press, Oxford.

STANFIELD, D. 1989. "Agrarian Reform In the Dominican Republic", in Thiesenhusen, W. Ed. Searching for Agrarian Reform in Latin America: the Dominican Case, Praeger, New York.

SKARSTEIN, Rune. 1997. "Development Theory. A Guide to Some Unpopular Perspectives". Oxford University Press, Oxford.

THIESENHUSEN, W. 1995. "Broken Promises", Westwiew Press Inc, Boulder.

THIESENHUSEN, W. Ed. 1995. Searching for Reform in Latin America, Praeger, New York.

TYLER, G. J, El-GHONEMY, R and COUVREUR, Y 1993. "Alleviating Rural Poverty Through Agricultural Growth", The Journal of Development Studies, Vo. 29, No.2, January, pp. 358-364, Frank Cass, London.

WORLD Bank. 1997. "African Development Indicators". World Bank, Washington DC.

Copyright Sociological Research Online, 2000

20