REGROUPEMENT DES TERRES AU JAPON ET DANS D'AUTRES ÉCONOMIES RIZICOLES DE L'ASIE INFLUENCÉES PAR LE RÉGIME DES MOUSSONS

Dans les économies rizicoles de l'Asie influencées par le régime des moussons, la petite taille et la fragmentation des exploitations agricoles sont devenues des obstacles au développement rural durable. Ces obstacles se sont révélés importants dès lors que les exploitations se sont intégrées dans une économie à salaires élevés et ont été renforcées par le fait qu'elles ne sont pas passées sous le régime de la propriété privée. Cet article démontre que, sous un régime de propriété privée, les parcelles fragmentées pourraient être consolidées en unités foncières compactes favorisant le leasing ou la production en coopérative, grâce à l'échange de droit de propriété ou d'utilisation. La terre pourrait être aussi passer à un régime de propriété publique en vue d'être exploitée par des «agriculteurs experts». Ces deux changements de régime permettraient d'atteindre des économies d'échelle.

CONSOLIDACIÓN DE LA TIERRA EN EL JAPÓN Y EN OTRAS ECONOMÍAS BASADAS EN EL ARROZ EN RÉGIMEN DE PROPIEDAD PRIVADA EN EL ASIA MONSÓNICA

Al entrar en las economías de altos ingresos, la pequeña parcela agrícola fragmentada característica de las economías basadas en el arroz del Asia monsónica se ha convertido en un obstáculo al desarrollo rural sostenible, y el problema no ha podido ser resuelto bajo el sistema de la propiedad privada de la tierra. En este artículo se propone que las parcelas fragmentadas de propiedad privada puedan consolidarse mediante un sistema de intercambio. Se podrán formar unidades compactas de tierras destinadas a la generación de ingresos y a la producción en cooperativas. Las tierras también podrán ser de propiedad pública y ser alquiladas por cultivadores expertos. Con ambas modalidades se podrian lograr economías de escala.

Land consolidation in Japan and other rice-based economies under private landownership in monsoon Asia

Jian-Ming Zhou1
European University Institute

Now that they have entered the high-wage economy, the small and fragmented agricultural holdings characteristic of the rice-based economies of monsoon Asia are an obstacle to sustainable rural development and the problem has not yet been resolved under the private landownership system. Through the exchange of ownership and location this article proposes that, enabled by private landownership, fragmented plots could be consolidated into compact land units for leasing or cooperative production. Land could also be turned over to public ownership and contracted out to expert farmers. Both solutions would achieve economies of scale.

In general, the Asian monsoon climate brings rains from May to October and a dry period from November to April. Only rice suits this climate, and it has been the major crop for centuries2. Up to the end of the Second World War, a feudal farm landownership had been dominant and most peasants owned little or no land and were either tenants or wage labourers, although there were some peasant land- owners. Farm work had to be done by hand with simple tools. Reclamation of new land had reached its limit. In the rainy half of the year, rice cultivation required highly labour-intensive, sophisticated and coordinated work, resulting in labour shortages. The consequent demand for more labour led to a high population growth rate, low cultivated land per caput and small-sized and fragmented individual (family) farming units3. In contrast, during the dry half of the year, the fewer work opportunities caused serious unemployment, underemployment or disguised unemployment4. Poverty was widespread and persistent. The economies were predominantly agrarian, with the presence of some industry in big cities creating a dual economy5. With natural conditions remaining the same, this economic situation changed after the Second World War, first in Japan and then also in Taiwan Province of China and the Republic of Korea.

The Japanese model of rural development starts in 1946 and consists of nine major components.

  1. Institutional changes for a mixed individual-cooperative economy.

    1. The land reform in the late 1940s, which provided for the compulsory purchase of the land of resident landlords with more than 2 ha and all the land of absentee landlords, allowed land to be sold to peasants for individual ownership, protected tenants from eviction (Rothacher, 1989), set land rent at a very low level and imposed a 3 ha ceiling on landholdings (Hayami, 1988). This was a huge incentive for peasants to increase output, while there continued to be numerous small and fragmented farms (measuring on average 1 ha [Hayami and Yamada, 1991] and composed of as many as 30 plots [Kristof, 1996]).

    2. The establishment of rural cooperatives - mainly service cooperatives - which provided forward and backward services and financing to the individual farming units. The direct agricultural production process, however, was under the independent control of the individual farming units (Kojima, 1988).

  2. Government policies supporting rice production and rural development. With the principal aim of achieving rice self-sufficiency, these rice price support policies included: farm credit and subsidies, technological research and extension services, import protection and activities in support of components I and III-VIII. There was also technological progress, embodied in components III-VIII). Five further steps (components III-VII) were implemented with the aim of reaching full employment:

  3. Construction of rural infrastructure. This mainly included irrigation, land improvement, transportation, communications, electrification and education, which established the technical basis for further rural development.

  4. Higher yields and multiple cropping of rice and other grains. The use of high-yielding varieties and more fertilizers made a considerable contribution to raising both land and labour productivity and also made the release of labour from grain cultivation possible.

  5. Diversified cropping and non-crop agriculture.6 This development increased peasants' income, changed agricultural structures and necessitated the establishment of rural enterprises for processing, transporting and marketing of crop, livestock, fishery and forestry products.

  6. Off-farm employment.7 The availability of jobs in both urban and rural enterprises has further increased peasants' income, changed rural structures and promoted urbanization.

  7. Peasant migration to cities and towns. Those who migrated were mainly the male, able-bodied labour force, leaving the aged and women in agriculture. As peasant farmers could get jobs also in the dry half of the year, full employment was achieved and wages rose. Hence, the post-full employment step in the Japanese model:

  8. Agricultural mechanization with small machinery. Mechanization sharply reduced the agricultural labour force without affecting output. The first transition (from agriculture to industry) was thus completed, a shortage of labour arose and the second transition (industry to services) started8 in Japan around 1960 (FAO, 1972). Rice self-sufficiency was achieved in 1955 (Hayami, 1988), per caput productivity was raised, equity in income distribution was reached and poverty eradicated (Oshima, 1987; 1993). Except for the import protection in component II above, these eight stages are significant for other economies. At this advanced stage of rural development, all the major obstacles imposed by the monsoon had been overcome except for:

  9. The small and fragmented farms.9 The major causes of fragmentation were as follows. As people became richer, rice consumption, although still necessary, declined. In the high-wage economy, the income from rice production turned out to be much lower than that from diversified cropping, non-crop agriculture and off-farm employment. If rice farms could not be viable,10 farmers would abandon rice production so that self-sufficiency was no longer attainable. In order to make rice farms viable, the income from rice production needs to be raised by enlarging farm size (more than 3 ha) so that large machinery may be used, labour saved, costs reduced and increasing returns to scale gained, as evidence later showed (Nishimura and Sasaki, 1993; Hayami, 1988).

Therefore, from 1961 on, farmers' purchase of land was subsidized by the government. In 1962, the landholding ceiling was relaxed. However, there were not enough land sales. On the supply side, part-time farming became dominant, and many able-bodied males commuted to off-farm employment while their wives and elderly parents farmed. They had no incentive to sell land: off-farm income was high, distances between towns and villages were short, transportation was convenient and there was no need to pay high rent for city dwelling, hence the added benefit of being able to enjoy less pollution and a rural home for retirement. Moreover, as industrialization proceeded, land prices soared and sales in the future were therefore destined to be more profitable. On the demand side, because land prices went well over the income surplus gained from rice production, it became unprofitable for full-time farmers to enlarge farm size through land purchase. Thus, much land remained under inefficient use by part-time farmers. In effect, it is the shortcomings of private landownership that have hampered land sale.
Hence the resort to land leasing. In 1970, rent control was removed, and land could be returned to landlords on the termination of contracts of more than ten years. In 1975 and again in 1980, leases for shorter periods were legalized. However, although land leases exceeded sales, progress was very limited. On the supply side, landlords still feared that, once let, land would be lost (Hayami, 1988). On the demand side, because the small farm was composed of many fragmented plots located in different parts the village, the leaseholder could not join them up into large units or change their shape into roads, canals, ponds, etc. (since the ownership belonged to the leaser). Therefore, large machinery could not be used. In this way too, it is the shortcomings of private landownership that have constrained both land leasing and the efficient use of the leased land.
Since economies of scale could not be achieved, the number of viable farms was reduced. Thus, farmers and cooperatives organized political lobbying for protection and the ruling party was obliged to yield, fearing a loss of votes (Hayami, 1988). In 1960, a "cost of production and income compensation scheme" was designed. As the monopsonist buyer, the government bought rice at a predetermined price and sold it at a lower price, thereby subsidizing rice farmers, while an accompanying law prohibited rice imports. Rice prices increased to ten times the level of world prices in the 1980s (Schaede, Lowe and Tokunaga, 1996).
As a result, in this period the domestic budget deficit for rice rose to more than US$7 000 million, leading to international protests, especially from the United States. The GATT Uruguay Round of multilateral trade negotiations in 1993 stipulated a "phase-in" of rice imports of 10 percent of the total market until 2005. Therefore, when Japan experienced a disastrous harvest in 1993, rice had to be imported for the first time in 1994 from Thailand, China, the United States and Australia (Schaede, Lowe and Tokunaga, 1996). Rice self-sufficiency was over. In 1996, two-thirds of Japan's food consumption consisted of imported, cheaper food. Further liberalization is expected (Kristof, 1996). However, under these circumstances, it is difficult for rice farmers with small and fragmented farms to survive and for the government to restore rice self-sufficiency. Subsidies have to continue. A grant of 6 000 billion yen was included in the 1995/96 budget for farmers to adjust to the new regime (Schaede, Lowe and Tokunaga, 1996). The critical issue is therefore how to consolidate and enlarge the small and fragmented farms.
Small and fragmented farms were efficient in a low-wage economy since they were conducive to development and diffusion of land-saving and scale-neutral technology. But in the high-wage economy, they hamper the achievement of economies of scale. This problem is common to all rapidly industrializing economies with limited land resources and a reduced working population in agriculture (Hayami and Yamada, 1991). Of other rice-based economies under private landownership in monsoon Asia, Taiwan Province of China and the Republic of Korea have replicated the Japanese model. Indonesia, Malaysia, Thailand, Bangladesh, India, Pakistan, the Philippines, Sri Lanka, Bhutan and Nepal may be regarded as at lower stages of the model. Once their industrialization has led them into the high-wage economy, this small and fragmented farm structure would also prove to be non-beneficial.
Therefore, the small and fragmented farms have become the remaining obstacle in the sustainable agricultural and rural development of monsoon Asia11. This problem has been taken as an argument against land reform from the feudal landownership system in other rice-based economies of monsoon Asia, on the grounds that the Japanese experience demonstrates that the benefits of scale economies will be lost if estate farming is dissolved (Koppel, 1993; Takahashi, 1993). Although this problem has been the subject of substantial analysis carried out by many leading economists in this field for many years, solutions have not yet been found (Bray, 1986; Francks, 1995; Hayami, 1988; Hayami and Yamada, 1991; Oshima, 1993).
The author's proposals for overcoming this remaining obstacle unsolved by the Japanese model are as follows.

Proposal I
Establishing a mixed economy of consolidated private farmland and public infrastructure land.

Consolidated private farmland. Fragmented plots could be reorganized into large land units through the exchange of location and ownership. Following is an illustration of such a system (the numbers are imaginary; the sizes of farms in figures are not proportionate to the grades):
Suppose a village has m (e.g. five) household farms - Fm: m = 1, 2, ..., 5; each farm has up to n (e.g. ten) plots located in different places - Pn: n = 1, 2, ... , 10.
Also suppose F1 has six plots, F2 seven plots, ..., F5 ten plots; (see Table 1). Each plot can be assessed on i factors (e.g. five: area, fertility, access to water, sunshine, convenience for transport) - Qi: i = 1, 2, ..., 5. Each factor can be given a different weight - W: W = [0, 1] (area and fertility may receive higher weights and, in general, a smaller area of better land could be exchanged with a larger area of poor land; some plots may be given zero as weight for a factor, e.g. if a plot on a northern hillside does not receive any sunshine, it may receive zero for this factor).
Following the assessment, grades could be given to each farm, e.g. F1 = 10, F2 = 20, F3 = 30, F4 = 40, F5 = 50. Then, after the main (autumn) harvest, assigning joined larger land units to each farm according to its grade through exchange, the land belonging to each farm should be given the most practical shape possible (e.g. the length of the plot should not be more than three or four times its breadth) (Skovgaard, 1950, p. 44). After this reorganization, each household would privately own one or a few (preferably no more than three) joined, but therefore relatively larger, farmland units (see Tables 2 and 3).

TABLE 1
Before reorganization - fragmented farms

F1P1

F4P1

F3P1

F2P1

F5P1

F1P2

F2P2

F5P2

F2P3

F5P3

F2P4

F4P2

F3P2

F4P3

F5P4

F1P3

F3P3

F1P4

F4P4

F5P5

F4P5

F5P6

F2P5

F3P4

F4P6

F3P5

F5P7

F1P5

F2P6

F3P6

F4P7

F5P8

F5P9

F4P8

F3P7

F2P7

F3P8

F5P10

F1P6

F4P9

TABLE 2
After reorganization - two-plot farms

F1P1

F3P1

F5P1

F2P2

F4P2

F2P1

F4P1

F1P2

F3P2

F5P2

TABLE 3
After reorganization - one-plot farms

F1

F2

F3

F4

F5

Some discreteness of plots may be rational. For example, a farmer may need both summer and winter pasture in certain hill areas, or land suitable for seed nurseries or for growing rice, or varieties of soil and situation in certain types of mixed farming. There is also a local custom of working both an upland plot and a plot on riverbanks and islands where work is done in entirely different seasons (Binns, 1950). Many farms in mountain regions consist of three separate estates - in the plains, in the middle levels and on the high levels. The solution may be to lighten the task and the expense of the peasants by regrouping as much as possible the lands which they possess at the various levels, and by reducing to a minimum the capital invested in construction (Government of Switzerland, 1950). Thus the numerous plots of a farm may be consolidated into two or three, rather than one, joined land units. Needless to say, different localities have specific peculiarities which are beyond the scope of this general discussion but should be taken into consideration in decision-making.

Public land for major infrastructure items. Among the large land units, major infrastructure items (main roads, canals, dams, irrigation and drainage network, electricity facilities, ponds, warehouses, other buildings, etc.) could be built as either

  1. jointly owned assets - the infrastructure items could belong to the government or village and those households and not invested in their construction according to their respective investment shares; the land beneath them should belong to the government or village; or
  2. exclusively public property with both the investment and land beneath them belonging to the government or village. The main reason why such land should belong to the public is that private owners have the right to withdraw their land if they wish, which would exert harmful externalities on the whole community.

Expenses are necessarily incurred in the above process. For private landholders, some permanent crops, buildings and other infrastructure in the old plots would have to be removed and compensated and new buildings and other infrastructure in the new farms be built and subsidized. Some peasant farmers might be asked to migrate to other areas and be subsidized. Public infrastructure implies public finance; fees for organizational purposes, e.g. setting up ad hoc committees, inviting external experts (surveyors, appraisers, etc.) who are supposed to be fairer than internal personnel. For land purchase, if consolidation is combined with land reform, i.e. extra fields belonging to landholders are sold to peasants in consolidate form directly or indirectly (bought by public authorities then sold to peasants). These expenses should be borne by the central and local governments, village committees and landholders in the form of government grants and loans, bank credits and personal payments. The village committees and landholders should be involved in decision-making and allocation.
The process of exchange of private plots for consolidation would not be easy. Individual considerations would constitute obstinate difficulties to consolidation. The farms resulting from consolidation could differ considerably in size, type, topography and value. Some farmers will almost certainly get better bargains than others and probably a still larger number will fear that others may do so. Satisfactory consolidation may result in leaving some farmers with poorer-quality land than they had before and it may not be possible to accommodate all the farmers. This would be compounded by the inertia of peasant traditions. For example, one family could claim that its plots are the heritage of its ancestors and could not be given away. Another may feel unfamiliar with the new plots. There will also be concern about financial matters. For example some farmers may worry that permanent crops, buildings, etc., in the old plots would not be sufficiently compensated (Binns, 1950).
Such realities imply that, to fulfil its objectives, consolidation may be voluntary or partly voluntary or may need to be compulsory.

Voluntary consolidation. Combined private action may take the form of the voluntary cooperation of a group of farmers, either through a formal cooperative society or through some looser form of ad hoc association. Such spontaneous efforts should be assisted and encouraged by governments. However, owing to the above-mentioned individual considerations, complete success is unlikely to result from purely private enterprise (Binns, 1950).
An example of voluntary consolidation was the attempt by the freehold farmers (peasant owners) themselves to consolidate the village fields of Oster Hjermitslev, Denmark, in 1820. Having been unable to agree on a rational consolidation scheme, the farmers' land remained split up in 12 different places throughout the village. In 1917, nearly 100 years later, although some amalgamation of the parcels of land had taken place, the situation remained unchanged, demonstrating the limits to consolidation by voluntary private action. Denmark's experience is that, where consolidation has been influenced by the peasant farmers themselves, the process has not been carried out well (Skovgaard, 1950).
In all voluntary and partly voluntary schemes, individual interests are a determinant factor. There is the obvious difficulty of getting a large number of people, however well-intentioned, to remain in substantial agreement throughout the proceedings. Even a very few recalcitrants may wreck or seriously impair the best of schemes. Thus, government intervention is required. Government action may be divided roughly into action in which compulsion is a reserve to facilitate voluntary efforts (partly voluntary consolidation) and action based primarily on compulsion (compulsory consolidation) (Binns, 1950).

Partly voluntary consolidation. Here, government intervention towards consolidation has aimed at encouraging voluntary action and supporting it by financial and other inducements, legal reserve power and technical assistance. Such activities need to educate public opinion, with very careful and intensive preparation. For carrying out consolidation, agreement can be by a substantial majority or simple majority of those landholders who control more than half of the land concerned. However, owing to the difficulty of securing the consent of a large number of smallholders, even the requirement of the consent of a simple majority has been found to prevent consolidation where it is most needed. This has led some countries to permit the competent authority to order consolidation where a substantial minority of the landholders, or even a single landholder, desire it, if the authority certifies that the consolidation is in the public interest. The general pattern is for the consolidation process to be initiated by the government, usually in response to a local request, and for it to be carried out if a sufficient proportion of the landholders concerned give their consent (Binns, 1950). Below are two successful cases of consolidation in Ireland and Switzerland.
Ireland. Before 1870, the ownership of the land in Ireland was concentrated among a small number of landowners with about half a million tenant families who enjoyed no security of tenure or protection against arbitrary increases in their rents and who, on eviction, were not entitled to compensation for improvements they had made. In the areas suitable for grazing, smallholdings were consolidated into large pasture and tenants there were forced to emigrate abroad or to other areas. Rural congestion was caused by overopulation in the poorer and lighter lands as well as in some other areas of the country. Hence the "congested area", which refers to any district where a considerable number of holdings were of such size or quality as not to afford a reasonable means of livelihood for their occupiers. While the concentration of ownership was made secure through the system of undivided succession, the tenants' holdings were excessively subdivided and the number of tenants progressively increased. As a result, a tenant's farm could have as many as 20 or 30 scattered plots (Government of Ireland, 1950).
During the period 1870-1946, a series of legislative acts were passed to allow the authority (Land Commission) to purchase (initially voluntarily but later compulsorily) from the landlords all tenanted land of the country, untenanted land in congested counties and the untenanted land outside such counties as was required for the relief of rural congestion. The authority could purchase other untenanted land by voluntary agreement with the landlords who could also keep demesne (land used by themselves). Afterwards, rather than selling tenants the ownership of the fragmented farms already held by them (as in the case of Japan), the authority consolidated, enlarged, improved and rearranged them by constructing self-contained land units, including roads, fences, drains, dwellings, outoffices, peat fuel supplies, etc., then sold them to tenants, former evicted tenants, former employees, migrants, cottiers and so on. The lords were paid by a special government stock and the government then charged the peasant purchasers annuities over a period of 66.5 years (Government of Ireland, 1950).
The rearrangement of holdings was an intricate and tedious process, for the tenants were often very exacting, and much tact and discretion as well as technical skill were required by the authority inspectors engaged on this work. Where possible, the rearranged holdings were brought up to an economic standard. But even where this was not possible, the rearranged and enlarged holdings were much more valuable than the old ones and greatly improved the living conditions of the tenants. The redistribution was partly voluntary because the authority induced the tenants to agree to the exchange by enlarging the holdings with additional land (either from migrants' surrendered holdings or from adjoining untenanted lands), and raising their standard with better housing and other improvements (Government of Ireland, 1950). It was also partly compulsory since the authority otherwise could refuse to sell land to them.
Any substantial economic improvement of conditions in the congested rural areas depended primarily on the migration of landholders from these areas to holdings provided for them on untenanted lands elsewhere as well as on the utilization of migrants' surrendered holdings by the remaining tenants. The migration was also partly voluntary since the migrants were induced by the consolidated, enlarged and improved farms in new areas (Government of Ireland, 1950), and partly compulsory, since the authority held the power not to sell them land in their original areas.
Switzerland. Until the nineteenth century, land reform against feudalism in Switzerland distributed farm ownership to tenants who already tilled the land thus maintaining fragmentation (just as in Japan). Various cantons (prefectures) tried voluntary consolidation schemes but failed. In 1912 the Swiss Civil Code stipulated partly compulsory consolidation: "When improvements to the soil (control of watercourses, drainage schemes, reforestation, roads, consolidation of fragmented forests and of rural lands, etc.) cannot be carried out except by the owners as a body, and when measures necessary to this end have been decided upon by two-thirds of the persons interested who possess more than one-half of the land concerned, the other owners shall be compelled to adhere to this decision." Appeals could be made to the consolidation authority (syndicate); then to the administrations of the commune, district, or canton, or to an arbitration tribunal consisting of specialists; and then to the central appellate commissions whose decision was final, so that landowners had sufficient opportunity to defend their interests but could not block the consolidation process (Government of Switzerland, 1950).
The procedure was based on the principle of compensation in kind, with due regard to the value of the properties surveyed (in terms of land, permanent crops, buildings, etc.). Next, private ownership and the location of lands were exchanged, and this process was supplemented by a road plan that would permit the rational use of the redistributed lands, the construction of outlets for the drainage of surface water, etc. Permanent crops and buildings in the old plots which had to be removed were compensated in cash. As a general rule, communal, cantonal and federal grants represented a total of 60 to 70 percent of the expenses connected with consolidation (Government of Switzerland, 1950).

Compulsory consolidation. The intervention of the central government or local authorities could also be mandatory, as the Danish example shows.
Until the 1750s, Danish agriculture was dominated by the manorial system, together with a few freeholders. Manors were owned either by the Crown or, more usually, by lords. Of this land, about 15 percent was held directly by the manorial owners as demesne, and the remainder by copyholders for life, who were dependent on the lords and received military protection. The land was fragmented. Thus, the land reform of the eighteenth century was aimed at not only giving land ownership to copyholders but also at consolidating land (Skovgaard, 1950).
The first government measures in the 1750s relied on the agreement of peasant farmers to consolidate the manorial fields and did not produce significant results. The consolidation operations carried out on the Bernstorff Manor were of greater importance. On this manor, comprising about 200 ha of demesne and about 1 600 ha of copyhold, land was worked by 41 copyhold farmers in three villages. All the land was mapped, surveyed and classified and was then divided into blocks of square or rectangular form as far as the configuration of the land permitted. The blocks were not of equal size, as the principle was that the copyholders belonging to each village should be on the same economic footing after consolidation as they were before, and copyholders who received poorer land were compensated by obtaining larger areas. Thus in one village the area of the smallest farm as compared with that of the largest farm was in the ratio 1:2; in the two other villages the ratio was 2:3. After consolidation, the distribution of the farms among the copyholders was made by lot, and the peasants who had to move from the village to their fields received a subsidy for new buildings. When the consolidation had been carried out, the lords sold all the farms to the copyholders, who thus became owner-peasants (Skovgaard, 1950).
The Counts of Bernstorff, as the local authority, carried out the consolidation scheme from high patriotic and social ideas. The copyholders were not enthusiastic about the consolidation and indeed the peasants who were to move their farmsteads fiercely resisted the measure. But in a few years the arrangement proved to be of value. In 1783 the former copyholders erected an imposing memorial in token of their gratitude to the Counts of Bernstorff (Skovgaard, 1950).
These operations thus became the model for much subsequent work in the following years, as a number of lords carried out consolidation activities according to the same principles. The consolidation process was most successful in places where the local authorities ignored the views of the peasants. The government then combined all existing rules of law on consolidation in a new act, the Consolidation Act of 1781, which became the fundamental law for all consolidation in Denmark. It was a comprehensive enactment which prescribed in detail all rules governing consolidation, the technical details being in accordance with the procedure adopted at the Bernstorff manor and with experience gained there. It stipulated that every owner of land in a village was entitled to demand that his or her land be consolidated and combined into a single lot or, where this was not possible, two or at most three lots. This provision applied regardless of whether other owners of land were willing to take part in the consolidation. The authorities, after consultation with all landholders of the village, proposed consolidation of the whole area, despite owners' wishes to to the contrary, in order that the consolidation of land belonging to one particular owner might not interfere with a subsequent complete consolidation. All the holders of land were bound by law to contribute to the aggregate consolidation expenses pro rata, according to the size of their land. A later act of 1792 further demanded recalcitrants to bear the cost of any delay in the consolidation. As a result, the land reform, together with consolidation, was essentially complete before 1835, in which year only about 1 percent of the copyhold land was not yet consolidated (Skovgaard, 1950).
For Japan and other rice-based economies under private landownership in monsoon Asia, partly voluntary consolidation may be the best solution, but compulsory consolidation should not be excluded whenever and wherever necessary.
After consolidation, there could be two ways to operate the land.

Proposal II
An alternative reform proposal consists in compulsorily buying rural land at appropriate prices for public ownership and setting up a dual land system under the management of villages and regulation by central and local governments. Land for living could be distributed equally to rural residents for their use for housing and a certain amount of agricultural production for self-consumption. If a person has formally migrated to the urban sector, a proportionate amount of land should be withdrawn from their household. Land for production should be contracted for a long term to expert farmers who aim at a higher output of rice and other products so that large land units could be formed and large machinery used. Contracts could be transferred and renewed according to market principles. Within the contract period, if the output target is not reached for reasons other than natural disaster, land quality is destroyed, or production abandoned etc., the contract could be stopped and sanctions applied. If land improvements have been carried out, awards could be granted. Production cooperatives or companies (as presented above but without private land shares) could also be set up in which expert farmers could work together. Wage labour could also be hired. Hence a mixed economy of public landownership, individual management and capitalist wage labour employment. If some production becomes surplus, fields could be used for other (even non-agricultural) productive purposes. The above-mentioned shortcomings of individual leases and production cooperatives or companies could be avoided. Details such as whether the public ownership be at the state or local government or village level or land prices and the fees and lengths of land contracts should be determined through a gradual process of experiments, public debates and expert consultations.
These first two proposals might be useful for other rice-based economies in monsoon Asia under private landownership when small and fragmented farms become an obstacle to sustainable rural development.

Proposal III
In those rice based-economies in monsoon Asia that have not yet implemented or completed land reform from the feudal system, reform could be achieved without maintaining the fragmented plots as Japan did. Peasant farmers could be given one or a few, but relatively larger, joined land units as in Ireland and Denmark.

Proposal IV
Population control should be strengthened, otherwise, dividing the parents' farms among many children for inheritance purposes would refragmentize land units.

Proposal V
Raising economies of scale in land should be gradual and follow the progress of diversified cropping, non-crop agriculture and off-farm activities. If, before the absorption of surplus labour by the development of these sectors, many land units were already transferred to large-scale farmers using large machinery, small peasant farmers would find it hard to survive. Thus, land reform providing limits on landholdings and rent as well as protection of tenants from eviction would be necessary. As these sectors developed, such controls could gradually be relaxed.

BIBLIOGRAPHY

Binns, B.O. 1950. The consolidation of fragmented agricultural holdings. Washington, DC, FAO.

Bray, F. 1986. The rice economies - technology and development in Asian societies. London, Blackwell.

FAO. 1972. FAO Production Yearbook 1972. Rome, FAO.

FAO. 1993. FAO Production Yearbook 1993. Rome, FAO.

FAO. 1995. Current thinking and activities. Paper prepared by the Sustainable Development Department for the Oversight Panel on Sustainable Development, 22-24 November. Rome, FAO.

Francks, P. 1995. Agricultural adjustment and
the East Asian development model: the
Japanese example. Paper presented at a
seminar at the School of Oriental and African Studies, London.

Government of France. 1950. Consolidation of agricultural holdings in France. In B.O. Binns, ed. The consolidation of fragmented agricultural holdings. Washington, DC, FAO.

Government of Ireland. 1950. Relief of rural congestion in Ireland. In B.O. Binns, ed. The consolidation of fragmented agricultural holdings. Washington, DC, FAO.

Government of Switzerland. 1950. Consolidation of land in Switzerland. In B.O. Binns, ed. The consolidation of fragmented agricultural holdings. Washington, DC, FAO.

Hayami, Y. 1988. Japanese agriculture under siege. New York, St Martin's Press.

Hayami, Y. & Yamada, S. 1991. The agricultural development of Japan - a century's perspective. Tokyo, University of Tokyo Press.

Kojima, R. Dec. 1988. Agricultural organization: new forms, new contradictions. China Qtly.

Koppel, B. 1993. Land policy problems in East Asia: understanding the past and moving toward new choices. In B. Koppel & D.Y. Kim, eds. Land policy problems in East Asia - toward new choices - a comparative study of Japan, Korea and Taiwan. Honolulu, Hawaii, USA, East-West Center and Kyounggi-Do, Republic of Korea, Korea Research Institute for Human Settlements.

Kristof, N.D. 1996. Japanese family farm: a way of life in crisis. International Herald Tribune, 3 January.

Nishimura, K.G. & Sasaki, S. 1993. Agricultural land reform and the Japanese farm land market. In B. Koppel & D.Y. Kim, eds. Land policy problems in East Asia - toward new choices - a comparative study of Japan, Korea and Taiwan. Honolulu, Hawaii, USA, East-West Center and Kyounggi-Do, Republic of Korea, Korea Research Institute for Human Settlements.

Nurkse, R. 1953. Problems of capital formation in underdeveloped countries. London, Blackwell.

Oshima, H.T. 1987. Economic growth in monsoon Asia. Tokyo, University of Tokyo Press.

Oshima, H.T. 1993. Strategic processes in monsoon Asia's economic development. Baltimore, MD, USA, The Johns Hopkins University Press.

Rothacher, A. 1989. Japan's agro-food sector. London, Macmillan.

Schaede, U., Lowe, D. & Tokunaga, Y. 1996. Economy - Japan. The Far East and Australasia 1996. Europa Publications.

Skovgaard, K. 1950. Consolidation of agricultural land in Denmark. In B.O. Binns, ed. The consolidation of fragmented agricultural holdings. Washington, DC, FAO.

Takahashi, A. 1993. The evolution of Japan's land policies in the East Asian context. In B. Koppel & D.Y. Kim, eds. Land policy problems in East Asia - toward new choices - a comparative study of Japan, Korea and Taiwan. Honolulu, Hawaii, USA, East-West Center and Kyounggi-Do, Republic of Korea, Korea Research Institute for Human Settlements.

1 The author sincerely thanks Professors Stuart Holland, Mario Nuti, Christopher Howe and Dr Jim Riddell as well as many others who have given instruction and help.

2 Monsoon Asia contains 19 rice-based economies: China (mainland), Japan, the Democratic Republic of Korea, the Republic of Korea and Taiwan Province of China in East Asia; Cambodia, Indonesia, Laos, Malaysia, Myanmar, the Philippines, Thailand and Viet Nam in Southeast Asia; and Bangladesh, Bhutan, India, Nepal, Pakistan and Sri Lanka in South Asia.

3 "Farm" (or farming unit) as used in this paper means "agricultural holding". It refers to all land that is used wholly or partly for agricultural production and is operated by one person - the holder - alone or with the assistance of others, without regard to title, size or location (FAO, 1972). Fragmentation of an agricultural holding is defined as the state of division of the holding into several discrete plots (Government of France, 1950; Binns, 1950).

4 (i) Those who are willing and able to work but cannot find work are unemployed; and (ii) among those employed, those who are working less than full time and want more hours of work are underemployed (Oshima, 1993); (iii) those of the population engaged in agriculture who could be removed without reducing agricultural output, even though the technical methods in use remain unchanged, are disguisedly unemployed (Nurkse, 1953).

5 Although Japan was developed, its industrialization was based on its import of foods from and export of industrial goods to colonies. Its agriculture was relatively stagnant (Oshima, 1987).

6 Diversified cropping implies a shift from a monoculture or a few crops (mainly grains) to a larger assortment of crops (roots and tubers, pulses, oil crops, vegetables, fruits, berries, tree nuts, etc.). (Oshima, 1993; FAO, 1993). Agriculture - depending on the context - in a broad sense includes cropping (farming), animal husbandry, fishery, forestry and hunting (Oshima, 1993) (the importance of hunting has been declining owing to environmental protection); in a narrow sense, it may refer only to cropping (farming).

7 Off-farm employment of farm families denotes their employment in non-agricultural sectors, i.e. industry and services. Industry contains mining, manufacturing, construction, public utilities, transportation and communication. Services comprise banking, real estate, business, public services that require the highest level of education and retail trade, restaurants, domestic and other personal services that only need minimal education (Oshima, 1993).

8 The first transition is completed when the share of the agricultural labour force in the total labour force (about three-quarters) has fallen, while the share of the industrial labour force has risen, to roughly 1/4-1/3. The second transition is concluded when the service sector overtakes the industrial sector in size of labour force (Oshima, 1987).

9 "Farm size" may refer to the area of land, or number of households, of the farm. The large farm size advocated in this paper for monsoon Asian rice-based economies denotes the large size in land area of farm whose basic operation unit is one household which may receive help from governments, collectives and cooperatives and hire non-family labourers. But experiments on managing a collective/cooperative farm as an agricultural enterprise including a number of households should be carried out.

10 Farms that earn income per farm household member equal to, or above, that of non-farm employees who are living in rural areas are "viable units" (Hayami, 1988).

11 In 1991, FAO/Netherlands Conference on Agriculture and the Environment defined the essential and interdependent goals of Sustainable Agricultural and Rural Development as "Food security, to be obtained by ensuring an appropriate and sustainable balance between self-sufficiency and self-reliance; employment and income generation in rural areas, particularly to eradicate poverty; and natural resource conservation and environmental protection."(FAO, 1995).