Scholarly article on topic 'Land Inequality or Productivity: What Mattered in Southern Vietnam after 1975?'

Land Inequality or Productivity: What Mattered in Southern Vietnam after 1975? Academic research paper on "Law"

CC BY-NC-ND
0
0
Share paper
Academic journal
Asia & the Pacific Policy Studies
OECD Field of science
Keywords
{""}

Academic research paper on topic "Land Inequality or Productivity: What Mattered in Southern Vietnam after 1975?"

ASIA & THE PACIFIC POLICY STUDIES

Australian National 1 University

Australian Government Department of Foreign Affair* and Ihde

Asia & the Pacific Policy Studies, vol. 3, no. 2, pp. 300-319 doi: 10.1002/app5.127

Original Article

Land Inequality or Productivity: What Mattered in Southern Vietnam after 1975?

Minh-Tam T. Bui* and Arayah Preechametta

Abstract

1. Introduction

Land redistribution and agricultural collective production were the key components of agrarian reforms implemented by the Vietnamese Communist Party in the south of the country after 1975. Land inequality was serious in the region under the Republic of Vietnam's regime. The new government struggled with agricultural collectivisation contributing to the decline in rice productivity. This study explains the persistence of a market-based agricultural production in the southern economy under the new political regime. Beside the economic reasons and arguments of local peasants' everyday politics cited in the literature, we argue that the de facto political power of the middle-class landowners was an important factor impeding the performance of agricultural cooperatives. It also implies that agricultural productivity was more vital than land inequality during the study period. We apply the model of Acemoglu and Robinson explaining how de facto political power helps elites to maintain their economic institutions in spite of a political change.

Key words: land inequality, agrarian reform, collectivisation, de facto political power, Vietnam

JEL Classification: P21, P26, P32, Q15

* Bui: School of Economics and Public Policy, Srinakharinwirot University, 114 Soi Sukhumvit 23, Wattana, Bangkok 10110, Thailand; Preechametta: Faculty of Economics, Thammasat University, 2 Prachan Rd., Phranakorn, Bangkok 10200, Thailand. Corresponding author: Bui, email: <buithiminh@swu.ac.th>

Very soon after the victory in the American war in April 1975, the Communist Party of Vietnam (CPV) embarked on the process of establishing socialism in the south of Vietnam, following the model applied in the north of the country. In the agriculture sector, two key policies were implemented with regards to land reform and agricultural collective production as prerequisites for a socialist economy. The ultimate objective was to make the agriculture sector generate surplus and foreign exchange for the country's industrialisation and at the same time strengthen the power of the new government all over the unified country.

Land has been a vital economic issue as well as politically sensitive topic in the agricultural economy of Vietnam where the majority of its workforce is engaged in the agriculture sector (over 80 per cent in the late 1980s) and a large share of GDP has been contributed by agricultural products (again over 80 per cent in the late 1980s). Land inequality was a serious economic problem confronting the Saigon government of the Republic of Vietnam, which was often referred to as South Vietnam prior to 1975. A large share of cultivated land belonged to a small number of rich landlords. Over 80 per cent of the cultivated area in the region was tenanted (Callison 1983). After several reform efforts, which redistributed land to farmers, the tenancy rate was reduced to around 60 per cent in 1970 (Salter 1970). Tenants in the south paid absentee landlords a rent of 25 per cent or more of harvested output. Land distribution in the Mekong Delta showed

© 2016 The Authors. Asia and the Pacific Policy Studies

published by Crawford School of Public Policy of the Australian National University and John Wiley & Sons Australia, Ltd This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.

a high level of inequality with a Gini coefficient of 0.8 in 1966 (Tuan 2001).

Land inequality and tenancy issues in rural southern Vietnam were accompanied by the exploitation of tenants by landlords, which was often cited as the cause of unrest and insurgency or rebellion in the region. It was proposed by western media at the time that the defeat of the American army and the Saigon government by the Vietnamese communists was because of their ignorance of the severity of land inequality in the rural south (Sansom 1970). Meanwhile, the revolutionary force led by Viet Minh gained considerable political support from poor tenant farmers and the landless in the region in exchange for the promise of landownership if victory was achieved. It was the farmers' hope that land inequality would improve under the new communist regime.

Land reforms were intended to serve as prerequisites for collective farming where farmers were forced to produce collectively. It turned out to be a long-term source of tension between the CPV and southern rural groups from 1975 to the late 1980s. In some southern provinces, it remained a source of dispute until the early 1990s. Meanwhile, collectivisation, an important feature of Vietnamese agriculture in the north from the mid-1950s to the 1980s, was not successful in the south of the country. Paddy rice productivity dropped by 13.2 per cent and 19.7 per cent, respectively, in 1977 and 1978 compared with the 1976 level. Southern rice output dropped from 7 million tons of paddy per year to around 6 million tons in the years following the country' s unification (Pingali & Xuan 1992). It was, indeed, claimed as a failure of the CPV's initial objectives. The reform perpetuated inequality, causing disputes among peasants and creating unrest in those areas (Dang 2010).

Why was there such a failure in the implementation of the economic policies of the communist government in the south of Vietnam? Did the farmers in the south look forward to a more equal land distribution under the communists' rule? What were the key factors hampering the outcome of agrarian reform in the region? Why did market production in the

agriculture sector persist and the collectivised production fail in spite of the change in the political regime? Further, why did collectivisation operate in the north of the country for a long period from the late 1950s until early 1980s, but have a short life in the south? Ultimately, what was the aspiration among farmers, land equality or agricultural productivity?

There is an extensive literature that attempts to find the underlying causes. Tri (1990) claims that collectivisation violated the voluntary principle, forcing farmers to join cooperatives. Another reason cited by Pingali and Xuan (1992) is that land tenure insecurity led to peasant disinvestment in their land leading to low productivity. Other economic reasons put forward were the well-established market-based economy in the south before 1975 (Akram-Lodhi 2004) as well as the co-existence of a free market for agriculture inputs and outputs in parallel with the cooperative system. On the other hand, the power of everyday resistance of farmers against the state was believed to be an important obstacle to the implementation of the collectivisation system (Kerkvliet 1995). Similarly, the political power of Vietnamese peasants in transforming national policy was also emphasised by Kerkvliet (2006). Another study showed that the key players responsible for the failure of post-1975 land reform and collectivisation were ordinary villagers and local cadres (Dang 2010).

Our study, however, applies the model of de facto political power and institutional persistence developed by Acemoglu and Robinson in 2006 to explain the situation in southern Vietnam after the political upheavals in 1975. The model highlights some important mechanisms for understanding simultaneous change and persistence in institutions. It claims that it is possible for the same specific economic institutions to exist under different political institutions. A simple static model provides a very useful framework for understanding how the persistence of certain types of economic institutions could have lasting effects on economic outcomes.

The current research aims to explain the failure of collectivisation following land reform in

southern Vietnam after the country's unification in 1975 and the persistence of the market economy in agricultural production under the new communist regime. We argue that, in addition to several reasons cited in the literature, the failure of collective farming was also because of the de facto political power of southern middle landlords, middle-class peasants (tang lop trung nong) who impeded the land reform and collectivisation conducted by the communist government. Such power is obtained from local government officials in exchange for economic rents, in combination with the ineffective de jure political power of local authorities and cadres, which led to the persistence of private commodity agricultural production in spite of the dramatic change in the political regime. These political reasons, indeed, were as important as the economic reasons for maintaining agricultural productivity and sufficient economic rents for farmers.

The remainder of this paper is structured in five sections. Section 2 briefly describes the major agrarian reforms and collectivisation policy implemented in the south of Vietnam after 1975. Section 3 presents the theoretical model of Acemoglu and Robinson (2006) on de facto political power on which the analytical framework of the study is built. The supporting evidence on southern Vietnam is in Section 4. Section 5 provides some concluding remarks.

2. Land Reforms and Agricultural Collectivisation in the South of Vietnam

The southern region of the country, which was often called South Vietnam, had undergone several significant land reforms since 1954 when the country was divided into two independent states demarcated by the 17th parallel of latitude following the Geneva Accords. The southern president Ngo Dinh Diem implemented a number of land ordinances during 1955-1957 aimed at rent reduction and tenancy contracts in combination with land redistribution and credit programs. This program and those of Diem's successors were, however, largely unworkable or ineffective (Tyson 1973). Land inequality remained high following the reform. The Gini coefficient only

reduced slightly from 0.84 in 1955 to 0.80 in 1966 (Tuan 2001).

In 1970, Nguyen Van Thieu continued land reform policy under the 'Land to the Tillers' law. The program was backed by the United States, which hoped for peasants' political support against the expanding National Liberalisation Front (NLF) from the north. The new land laws sought to eliminate tenancy and establish a nation of small farm owners (Salter 1970). Land was granted free of charge to tenants with compensation to former owners paid by the government. Legislation limited landownership to 15 hectares in the south and

5 hectares in the central areas of the country for those who directly cultivated their land. Farmers could also buy more land from the government with no-interest loans paid over

6 years. The acquisition was limited to 3 hectares in the south and 1 hectare in the central regions. By March 1973, nearly 2.7 million acres of land had been turned over to 800,000 million tenant farmers in the south (Tyson 1973) and farm tenancy had been almost eliminated. As a result, by early 1975 in the Mekong Delta, 70 per cent of the rural population was middle-class peasants who owned 80 per cent of cultivated land and 60 per cent of total farming equipment. The dominant economic institution in agriculture in the south was very commercialised with advanced mechanical production on the basis of secured property rights and private ownership (Dang 2010). It was, indeed, the objective of Thieu's land reform policy to bring middle-class living standards to farmers in a society that advocated capitalism as the main economic policy (Tyson 1973).

Immediately after the unification, the new revolutionary government decided to keep the economy and market structure intact, and private landownership was respected. CPV documents expressed clearly that a multi-component economy with various forms of ownership should be maintained within a certain period of time (CPV 2004a). In the very first years after 1975, rural areas in the south were thriving because of favourable weather conditions and abundant production inputs left over from the former administration (Liem

1995). Starting from late 1976, however, the situation deteriorated because of the shortage of fertilisers and pesticides and production machinery that had broken down or become obsolete. The state, then a monopolist in the trade and distribution of all agricultural inputs and equipment, was also faced with a shortage of foreign exchange for imports. In this context, the government called for the creation of collective agricultural production in the south. The model required a completion of land reform.

2.1. Land Reform

The main purpose of land reform, as documented by the CPV, was to eliminate the vestiges of feudal and colonialist exploitation and to provide the landless and the land-poor a means of production as well as to facilitate peasant solidarity in production (CPV 2004b). On the 23 September 1976, the Politburo of the CPV passed Resolution 235-CT/TW on a land redistribution campaign to authorise land appropriation and the banning of private land sales. This process followed the stereotype in socialist countries and a similar model conducted in North Vietnam in the 1950s (Box 1). The 4th National Party Congress at the end of the same year unveiled the second 5-year plan (1976-1980) for the unified country moving towards large-scale socialist production (CPV 2004b).

Minh [Viet Nam Doc Lap Dong Minh Hoi or League for the Independence of Vietnam, established in the north and later opposed the South Vietnam), in the meantime, were seeking to maintain production on lands they controlled and to obtain support from the poor farmer. They undertook a land reform program in which farmers' rents were materially reduced and lands were confiscated and assigned to poor farmers. However, as they do now in Vietnam [South Vietnam], the Communists have misrepresented their objectives to the peasants. They failed to indicate that their ultimate objective is to communize the land... (Salter 1970: p. 726).

In 1977, a central committee was established under the name Ban cai tao nong nghiep mien Nam (the Agricultural Transformation Committee in the South—ATCS) to reinforce the agricultural reform and to conduct a land survey (CPV 2005b). The survey revealed that 20-30 per cent of households who were agricultural labourers in the region held no cultivated land as shown in Table 1.

The ATCS completed its task in 1978, reclassifying rural society into four main classes as shown in Table 2, based on land holdings of households, quantity of livestock and the production equipment they possessed. The fact whether landowners directly cultivated their own land or hired labour was also taken into account. The richest class, named 'rural capitalist', owned over 7-15 hectares

Table 1 Estimation of Landownership in the Mekong Delta After 1975

Landowners Land surface

Category (households) (%%) (%)

Landless 20-30 0

Poor peasants 50-60 50-60

Middle-class 12-17 20-27

peasants

Rich peasants 2 10-12

Sources: ATCS, Agricultural Transformation Committee in the South (1984) and Callison (1983).

Box 1: Land reform under communism

...As exemplified by Soviet Russia and Communist China, agrarian reform is simple enough: it is a means to political power, based on a promise to the peasant: land-in exchange for his political support. Once the Communists are in power, all the land is confiscated; peasants become farm hands on collectives, communes and state farms. And harsh production and delivery quotas complete the rude awakening from an exhilarating but all-too-brief experience of free holding. (Ladejinsky 1964: p. 446)..From 1945 to 1954,.insurgency was rampant in the rural areas.the Viet

Table 2 Land Holding Classification Under the Two Tenure Systems

Land area owned household (ha)

Agricultural Transformation Committee in the South in 1977

'Land-to-the-Tiller' Program in 1970

Less than 0.5

Rural capitalist (with farming equipment)

Rich landlord

Medium-high landlord

Medium-low landlord

Small landlord

Poor peasants

Rich landowner Middle-class landowner

Poor peasants

Sources: ATCS (1984), Liem (1995) and (Dang 2010).

and production equipment and were regarded as exploitative capitalists who were inappropriate under the socialist economy. Private ownership of farming equipment was then outlawed and existing equipment was confiscated. Raising livestock was prohibited (Raymond 2001).

Table 2 describes the differences in identifying rural classes based on their land holdings under the two land policies of Thieu's administration and the new communist regime. With the new classification, poor peasants accounted for 20 per cent of the southern population with a land holding share of only 10 per cent. The largest class was medium-low landlords constituting 50-60 per cent of the total number of southern landlords while holding a similar share of the land area (Liem, 1995). Such a classification was important to the new government in order to implement the so-called 'eliminating land-based exploitation' policy, aimed at landlords, rich and middle peasants. These groups, in the administration' s view, would pose an obstacle to the introduction of socialism in rural areas. The second aim of the reform was to redistribute those confiscated lands to landless and land-poor households— a process referred to as 'land readjustment'. According to the new classification, the middle-class peasants were the major force (50-60 per cent) in southern agricultural production.

The land redistribution in the Mekong delta during the period from 1975 to mid-1978 was deemed not to have achieved its targets. Reports from the ACTS showed that only 191,931 hectares were expropriated from 'exploitative' rural capitalists, rich and mediumhigh landlords (Dang 2010). Another survey by Dinh (1994) estimated the landless and

land-poor households who accounted for 1831 per cent of the rural population, owned only 10 per cent of cultivated land in 1978. In some provinces in the Mekong delta like An Giang, a big rice growing centre, 41 per cent of the rural population, including non-farm households, were considered landless or land-poor Lam (1985). Confronted with this lack of progress in land redistribution, the Central Committee of the CPV decided to expand collective farming along with the continuation of 'land readjustment' in 1978—the key year of the 5-year plan (CPV 2004a).

2.2. Collectivisation

The collectivisation process commenced in the north of Vietnam after vigorous land reform between 1953 and 1955. It was further reinforced during the period 1976-1980. Almost two decades later, the model was replicated in the south. The process of collectivisation in the north is described in Box 2.

Box 2: Agricultural collectivization in the north of Vietnam

According to Pingali and Xuan (1992) and Liem (1995) the process in the north included three phases: the formation of work-exchange teams and mutual aid teams (MATs—to doi cong, to doan ket) (1956-58), the establishment of low-rank cooperatives (hop tac xa bac thap) (195860), and the advancement of cooperatives to high-rank or large-scale cooperatives (hop tac xa bac cao) (1976-80) following the Soviet model of collective farms. A large number of high-rank agricultural

cooperatives and state farms were established. By 1979, there were 232 state farms, responsible for 11.6 per cent of agricultural production (Fforde & Vylder 1996). Together with thousands of smaller cooperatives, they were the key economic unit, responsible for a major part of the agricultural output of the country.The initial form of collective production consisted of work-exchange teams or MATs where farmers retained ownership of land and control of crops but were encouraged to assist each other during periods of peak labour demand. Farmers in each team, normally within a hamlet (thon, were compensated as a group for their pooled labour (Raymond 2008). Following the MAT model, subsequent collective production involved agricultural production cooperatives. Farmers were forced to join cooperatives and work collectively to fulfil the crop procurement quota imposed by central authorities (Raymond 2001). Labour was divided into brigades which specialised in each stage of production, under the brigade leaders' supervision and cooperative management. Farmers' efforts were recorded in work points by brigade leaders. They also decided on the agricultural input distribution, resource allocation, production and output purchases, according to the state plan. Remunerations to farmers were only realised after the harvest. Cooperatives determined the portion to be sold to the state, taxes, production costs and other fees, before sharing profits among farming members.

In the southern provinces, large-scale cooperatives had never been established. The first wave of collectivisation took place from 1978 to 1979, when the 'exploitative group' was forced to join cooperatives or production units. Their land and production assets, livestock, equipment were also expropriated to be shared assets under the management of cooperatives. Medium-low landlords were persuaded,

convinced or even threatened to join production units (Liem 1995). The process also recruited the landless and land-poor who cultivated fields far from their residence to build state farms. Resistance to local authorities, therefore, did not only come from landlords but also from land-poor households.

Besides the forced collectivisation, farmers had to comply with food obligations by selling their crops to the state following procurement quotas fixed by the central government. The procurement price was irrationally lower than the free or 'black' market price. This regulation fueled the anger of southern peasants who refused to sell their produce to state trading agencies. To force farmers to comply with regulations, domestic trade across provinces was prohibited. Farmers' access to the free market to buy consumer goods and agricultural inputs was restricted. Trade flows from rural to urban areas were strictly controlled, leaving the state as the only buyer for farmers' rice and livestock. In urban-rural trade, the difference between the state-administered price for agricultural goods and the official retail price of industrial inputs and consumer goods was a source of dispute. The state had a tendency to buy agriculture goods at low prices and sell high-priced manufactured goods, a process known as 'price scissors' (Tri 1990). This fact further dampened economic relations between peasants and the state.

3. The Theoretical Model for an Analytical Framework

The persistence of private household-based farming in the south of Vietnam after 1975 can be a good case for applying the model from Acemoglu and Robinson (2008) in an economy experiencing political change. The original model draws some implications of changes in political institutions for economic institutions. A change in political institutions alters the distribution of de jure political power, but creates incentives for investments in de facto political power of the elites. Therefore, the new political regime may choose economic institutions favouring the elites. The model was used to explain the economy as a consequence

of the de facto power of elites in the southern United States, even after the civil war had ended.

A more specific model of Acemoglu and Robinson (2006), which is a static version of the previously mentioned model in Acemoglu and Robinson (2008) raised the question, why did the southern economic institution persist? Despite losing the Civil War, political elites in the south managed to sustain their political control in the region. They also successfully blocked economic reforms that might have undermined their power and derailed any political reforms they opposed. Although the economic system using slaves was abolished, southern elites still possessed considerable de facto power through their control over economic resources, their greater education and their ability to engage in collective action.

The model provides a very useful framework for understanding some important mechanisms for simultaneous change and persistence in institutions. Institutional persistence, in this context, refers to the persistence of a cluster of economic institutions. These different specific economic institutions may exist under different political institutions. The simple static model can, however, show how the appearance of change in certain dimensions of specific institutions does not necessarily mean a change in economic institutions that are essential for the allocation of resources in society. Details of the Acemoglu and Robinson (2006) model with mathematical expressions are presented in Appendix 1.

3.1. Analytical Framework

Relying on the settings of Acemoglu and Robinson' s (2006) theoretical framework, this study specifies two different choices of economic institutions, which are collective agricultural production, and market-based household production based on private land tenure and property rights. Meanwhile, the two political regimes in this case can be assigned to the former Republic of Vietnam and the later communist/socialist government of Vietnam. We borrow the aforementioned theoretical framework to explain that given

the crucial change in political regime to the communist system, one might have anticipated a significant change in economic institutions. However, the market-based agricultural production persisted, and the communist government's efforts to impose collective farming were not, after all, a success.

We modify the original models of de facto political power and institutional persistence (Acemoglu and Robinson 2006) for the case of southern Vietnam after the country's unification in 1975 by explicitly specifying the role of the state in the process of making choices between economic institutions. The state or government was not specified or implicitly included within the elite group in the original model. With modifications, we can explain the difficulties in the implementation of the policy of land redistribution by the communist government and the imposition of collective agricultural production in the southern region. The model can also explain why there was resistance and even confrontation of peasants against local authorities and why the market-oriented agricultural renovation policy was finally initiated a decade later in 1986. Our framework is summarised in Figure 1.

The system includes a central government, which determines the two major command policies on land distribution and the collectivisation of agricultural production all over rural areas. The later policy is supposed to be carried out when the former is complete. The implementation of those policies is conducted at district and village levels by local officials or local cadres. There are two main groups of population in the rural society, namely, elite landlords and land-poor or even landless peasants. The possession of land for cultivation of the two groups is significantly different. The interactions of political power and economic rents determine the dominance of economic institution, which can be either collective farming or market household-based agricultural production.

Under the non-communist administration, agricultural production follows the competitive market system. Property rights of land and private ownership of farming equipment are

Bui: Land Inequality or Productivity Figure 1 Analytical Framework for Rural South Society After 1975

CENTRAL COMMUNIST GOVERNMENT

Land reform with redistribution policy toward equality

Collectivized agricultural production

guaranteed by the government. The landless peasants work for landlords and receive wages. Wages are determined in the competitive market and are equal to the marginal product of labour for agricultural production. The peasants' source of income is only wages, whereas the elite' s income is derived from land rentals on the area under their ownership. If rural elites are over-represented in rural society and the political system, they will make decisions on agricultural policy.

Under the policy of collective farming, however, elite landlords have to transfer part of their land to the landless. The poor farmers do not actually own the land but make contributions from their redistributed land to the collective

production as they are compelled to join the cooperatives. The landlords, therefore, are in fact, only middle landowners under the classification of the former regime. Remunerations from the cooperative system to farmers include two components. The first component is from wages determined administratively by cooperatives, usually at a lower level than the marginal product of labour. The second income component is output sales from farmers' directly cultivated land under their ownership if they can somehow illicitly seize it from local cadres. Farmers make decisions on the allocation of time spent working for cooperatives and working on their own plots. Output must be sold to the state at a low price pre-determined by the state.

As for the land-rich elites, under the new communist administration, they are also forced to join cooperatives. However, there is the alternative option of compromising with local officials. If they are content to provide some economic rents or 'bribery' to local officials, they are allowed (or more precisely, are 'ignored' by local cadres) to do business on their own land, and they can sell their output on the free 'black' market. Their income, as a result, must deduct the cost of bribery or 'transaction costs' from total sales. If the elites decide not to opt for bribery, they have to join cooperatives as farmers.

The central government, under the cooperative system, gains from the premium between the market price and the actual price of agricultural output. The total gain of the state is the sum of the agricultural surplus of the central government and economic rents obtained by local officials. If the cooperative system is eliminated and the market system is in effect, the total gain of the state will only consist of tax payments from both elites and farmers. The state' s decision on the de jure economic institution to be adopted depends on which brings the state the greater gains. If the sum of the local cadres' economic rents and price-margin gains is larger than the tax, the reform toward market production will not occur.

In summary, we argue that it was the de facto political power of the southern middle landowners and capitalists who impeded collectivised production, in combination with the ineffective de jure political power of local authorities and cadres that led to the persistence of the commodity agricultural production institutions. The collectivisation efforts of the government failed in spite of the dramatic change in the political regime to socialism. In rural society, the de jure political power belonged to local cadres at a grassroots level who actually conduct state policy. However, de facto political power was obtained by a considerable proportion of middle landowners. This power, which was further supported by the strong resistance of poor farmers to the local authorities, secured the prevalence of commercial farming in the region.

In that sense, our framework has evolved from the original model in two directions. First,

an additional agent is included in our model— the state or central government, which imposes property rights and corresponding economic policies on different rural classes. Meanwhile, the local government implements and manipulates those policies in daily interactions with the rural society. Second, the economic institution that prevails is determined by the de facto political power obtained by rural landlords at the cost of economic rents paid to local cadres.

4. Evidence on de facto Political Power in Southern Vietnam

Given the scarcity of quantitative information on the period of this study, we adopt qualitative instruments in testing this aforementioned model in the case of southern Vietnam. By reviewing literature and bringing supporting evidence from other studies to shed some lights on the key argument, this research is able to show that middle-class peasants (tang lop trung nong) in rural economy in the south significantly contributed to the persistence of private economic mechanism of the region. Literature looking at the failure of the collectivisation campaign in southern Vietnam after 1975 provides different explanations, which are both economic and political. The economic arguments have often been drawn from the violations of incentive structure and the dominance of the free market. Tri (1990) pointed out that the key principles of collective organisation of willingness and voluntarism were not valid in the south of Vietnam where farmers were forced to join collective production and to contribute their farming assets. Second, peasants in the south boycotted collective farming and organised strikes because there was no incentive structure within cooperatives. Within the brigades, labour efforts were recorded as one point per day regardless of the intensity and quality of their work, whereas income of cooperative members was fairly uniform. Liem (1995) noted that cooperative members hurried to finish their 1-day work point and spent most of their time on their own residential land. Similarly, evidence from the north indicated that around 75 per cent of household income came from private plots

through private sales on the free market (Kerkvliet 1995). Later, in the 'reform' period, it was evidenced that incentives that induced farmers' efforts mattered greatly for the total factor productivity in Vietnam' s agriculture during the period 1976-1994 (Che et al. 2001). Third, the failure of collectivisation after 1975 proved the dominance of free markets. Whenever there is price control or rationing of markets, free or black markets emerge. During the collectivisation period, markets for factory inputs, credit for agriculture production, consumer goods and agricultural products were all rationed by the state, and there were corresponding black markets for all of these. The CPV leaders admitted that farmers' free market sales made it impossible for them to be collectivised (CPV 2005a). Share-cropping, wage labour and money lending were widespread in the south (Raymond 2001).

In the following section, an emphasis will be placed on the evidence regarding political aspects in the aforementioned framework. The evidence supports four key points of the argument: (i) farmers in the south had experienced democratic political power under the previous Saigon regime; (ii) property rights for land-ownership were the core of agricultural production in the region; (iii) local authorities could exercise their political and economic power simultaneously in the rural society; and (iv) there was de facto political power gained by middle-class landowners through compromise and trading off economic rents with local authorities.

First, it has been demonstrated in various studies by western analysts that the use of everyday resistance by farmers to the state's imposition of collectivised production played a critical role (Raymond 2001; Kerkvliet 1995) in the failure of collectivisation. This failure is, in our opinion, partly due to the fact that political power was embedded in southern peasants and rural middle class. It should be noted that under the Saigon administration, tenants and farmers in the south had greater knowledge and experience of the process of acquiring democracy as well as political power. As landowners, peasants had status and dignity in their community and sought participation in

and control of its institutions. They had, in fact, a stake in their society and tended to support a government that recognised and supported that stake (Salter 1970).

Wurfel (1957) described the participation of farmers and tenants in provincial joint committees made up of provincial leaders, officials of the agricultural service, landlords and tenants. These committees were established each year in each province to solve landlord-tenant disputes, and they also fixed average yields for rent basis in lease agreements or tenancy contracts. Tenants were able to negotiate the rate of land leasing at 15 per cent of crop output with landlords. Ladejinsky (1964) noted that in contrast to experiences in other Asian countries, tenants in southern Vietnam were more hostile to land reform than the landlords. There was also the Tenants' Union, which was part of Vietnam' s largest labour federation in the south at that time, and their lawyers could represent tenants in the court in any disputes with landlords (Wurfel 1967). It can be said that rural society in Vietnam had experience in exercising political power at the village level, which made them different from those in the north of the country.

Second, private ownership of land was a key factor for the success or failure of agricultural production. Security of tenure should provide incentives for families to invest in the land and apply agricultural technology in their cultivation. As Ladejinsky (1964) put it:

It is generally believed among rural society that it is relatively easy to increase production, but only if the cultivator's relationship to the land and the state's treatment of him and of agriculture create incentives to invest, to improve the land and to raise productivity.

The land redistribution process conducted by the revolutionary government was supposed to be supported by the poor as they were the main beneficiaries, but this was not the case. Farmers in the Mekong Delta recalled the land tenure system before 1975 under the Saigon regime as 'acceptable' even though land holding was not equally distributed among households. The system was based mainly on private

ownership and commercialised agricultural production, allowing even the landless and land-poor to live comfortably with wages or crop shares paid by the land-rich. Local people neither appreciated the land redistribution policy of the CPV nor thought it justified (Dang 2010).

Peasants in the south used to have private ownership of land for their cultivation. Even in the cases of tenancy, they were able to maintain full control of their production. Salter (1970) noted that in the south and central lowlands, farmers, whether they owned or rented land, were managers of production. The landlord did not usually participate in the production process and provided neither seeds, credit, farm equipment nor marketing outlets. Once private ownership is ensured, the profit maximisation principle of individual households will follow. Results from US-supported cadastral surveys in the south of Vietnam carried out by the Stanford Research Institute in 1968-1969 showed that tenant farmers were fully willing and desirous of purchasing the land provided it was available at reasonable prices (Salter 1970). In the case of the Mekong Delta after 1975, farmers destroyed their equipment when they were forced into cooperatives (Liem 1995), simply because their individual ownership over their plots was not guaranteed.

Third, on the opposite side of the power balance, who were the elites in the rural society? Before unification in 1975, Wurfel (1967) pointed out that the economic elites had been dominated by the French and the Chinese, while the political elites had been confined to the ethnic Vietnamese. Formerly, the great landlords were those who held significant amounts of both political and economic power. Following the defeat of the Saigon government by Vietnam's communist forces, the political and economic elites of the old administration were either eliminated, fled the country or moved to urban areas to avoid the rural political turmoil (Tri 1990). Who were then the new elites under the new communist regime? During the implementation of the land redistribution policy, it was apparent that there were new conflicts between middle landowners and the new ruling authorities from the CPV, who were also involved in the management

of agricultural cooperatives. During the period of land redistribution, local cadres could arbitrarily allocate land plots to farmers, which indicated a certain economic power, originating from their political power. In large-scale or high-level cooperatives, the cultivation of crops, the division of labour and distribution of the harvest were bureaucratically rather than privately managed, and land was owned by the state.

Later, it became clearer that political cadres also controlled rural economic life. The cadres themselves were, in fact, the elites in the village (Raymond 2001). The distinction between political and economic elites would be blurred in some cases because of the involvement of party cadres in the economic management of cooperatives. The cooperative management teams included managers, accountants and inspectors who decided on the allocation of consumer goods, prices for agricultural produce or the distribution of agricultural supplies. Their economic power was therefore trivial.

The last point of this political argument focuses on how the middle landowners achieved de facto political power. As noted earlier, under the new classification of the communist government in Table 2, the middle-class landowners were actually middle landlords and rich peasants of the previous Saigon regime. Their cultivated land was less than 15 hectares and not appropriated by the Land-for-the-Tiller program of president Thieu. They accounted for 50-60 per cent of the rural population. As noted earlier, after the war, at the beginning of the unification period, the revolutionary government kept the existing southern production structure unchanged as they realised the importance of this middle class. Canh (1988) pointed out that the individual farm business (ho lam an ca the) in the rural south played an important role in the rural economy, and their business operation would continue for a considerable period of time as long as the advantages of collective farming had not been verified.

In addition, there was also a certain extent of compromise by the new government in conducting their land appropriation policy. According to the CPV's policy, land and farms of

foreign capitalists, of landlords who fled out of the country and of commercial capitalists (tu san mai ban) must be totally confiscated while those of other landlords were taken with considerations. Landlords were distinguished between ordinary landlords (dia chu thuong) and revolutionary landlords (dia chu khang chien) who were treated better if they refused to donate their land and whose assets were not appropriated (CPV 2005b). These middle landlords were reported to have strong connections with the southern national liberalisation front (NLF) (Mat tran giai phong dan toc mien Nam) during the wartime, providing financial as well as logistical support to revolutionary soldiers. Following the end of the war, their families were rewarded as veterans by the communist government and some even became local cadres (Liem 1995). Some of those even made considerable donations to the new communist government (Tri 1990), which could be considered as a form of investment to obtain 'soft' political power.

In addition to certain favourable treatment given by the central government policy toward middle landowners, this group also found the way to negotiate and compromise with local authorities who directly implemented government policies. On the power balance between middle landowners and local cadres, the former possessed both human and physical capital, which were desperately needed by the later for the operation of cooperatives. For example, Raymond (2001) noted a survey on cooperative committee cadres revealing that most of them did not complete primary education and half of them completed only Grades 1 or 2 or were even illiterate. Meanwhile, the upper-class and middle-class farmers, who were often the most knowledgeable and productive, were excluded from decision-making positions within cooperatives. As a result, there was conflict between cooperative committees and local farmers on economic benefits, operations and bureaucratic management.

In this power balance, local farmers have economic resources to trade for political power with local officials. Middle peasants in the south possessed their own economic resources apart from land. Those with machinery and livestock did not face collectivisation. Instead,

they were forced to make these assets available to cooperatives at administratively determined prices (Akham Lodhi 2004). In other words, the collective production had to rely on the physical capital of this group. In many cases, there was also armed resistance of farmers against local authorities and police to protect their own assets, and they refused to join cooperatives (Liem 1995). During the land readjustment, landowners were able to use kinship, local informal social networks and local norms to minimise the impact of land redistribution (evade and minimise the impact of state policy). At a later stage, they also employed various tactics to retrieve land that had been redistributed to someone else (Dang 2010).

In addition, de facto power of rural middle landlords was supported by corrupted local cadres. A member of the VCP Politburo, Le Thanh Nghi made an assessment in 1981 that cadres had been corrupted and 'misused their authority to exploit peasants' (CPV 2005a). Government agencies and army units had seized lands and hired poor villagers to work as sharecroppers (Raymond 2001). On the other hand, Kerkvliet (2006) pointed out that the local cadres in rural areas often played a role between central authorities and peasants. They had strong ties with and an affinity with the villagers. Local cadres often interpreted central directives according to their own priorities and modified state policies to match local reality.

Liem (1995) specified cases where landlords were able to negotiate with local authorities to allow them to keep a small part of land for their own cultivation as the price of joining cooperatives and donating production tools and equipment. In this way, the central reform policy was bent to benefit rent-seeking local officers. Similarly, Dang (2010) also pointed out that local cadres as key actors at the local level who, through malpractice, had a strong influence on the process and results of land redistribution. This group of middlemen between the central government and peasants in many cases modified land adjustment state policies for easier implementation and to accommodate peasant concerns. There were a number of forms of bribery including

the transfer of surplus non-confiscated lands of rich landlords to relatives of local cadres. In this way, the de jure power of the government was undermined while the de facto power of landlords was implicitly increased. As a matter of fact, the central government ordered the provision of more cadres to local areas and intensified scrutiny and inspection of cadres regarding their moral ethics and implementation of central policies. The party determined to punish corrupted, collaborative cadres who purposely manipulated state policies (CPV 2005a).

The land-rich peasants in the south were obviously very dissatisfied with the land readjustment policies of the revolutionary government. In addition, the land-poor peasants also, in many cases, refused to accept the redistributed land. The policy of 'offering rice and sharing clothes' (nhuong com se ao) initiated by the government was not suitably applied in the south where both peasants and landlords deeply acknowledged the hard struggle they had experienced to actually own a piece of land. Land to them also meant 'sweat and tears' . Taking land from others for free was considered a weird practice (Dang 2010). In addition, some poor households were reluctant to receive distributed land also because they were obliged to cultivate high-yield rice rather than the traditional rice that they were familiar with. Others took those lands to avoid being relocated to the 'new economic zones' in remote areas (Liem 1995). The land-rich households, in contrast, tried to circumvent land appropriation by transferring parts of their lands to their children, relatives and even friends. This practice was even encouraged by local officials. As a result, many landless and land-poor households actually received no land during the period 1975-1978 (Dang 2010).

4.1. Results of Collectivization

Resistance against the new government and local cadres included both a refusal to join cooperatives and direct confrontation. By the end of 1979, collectivisation in the south had covered only 33.5 per cent of peasant households and 27 per cent of the cultivated area. The VCP

also failed to force a substantial proportion of southern farmers to produce grain collectively. By the end of 1982, only 1,639 cooperatives had been created in southern Vietnam, involving 24.4 per cent of rural households. However, in the entire Mekong Delta, cooperatives covered only 0.6 per cent of the area according to the General Statistical Office (GSO) in 1991. Only six new cooperatives had been formed in the delta provinces of Ben Tre and Long An by 1982 and only 10 had been created in Ho Chi Minh city. The VCP Politburo member, Le Thanh Nghi, evaluated the collectivisation situation in the south as a failure (CPV 2005a). Only 9 per cent of peasant families and 7 per cent of cultivated land had been collectivised by 1981 (Raymond 2001).

Southern farmers' attempts against collective production were different in many ways. Although the sale of land was prohibited, farmers continued to sell redistributed land granted to them by the new government. In many cases, appropriated land was returned to its previous owners by farmers. If farming equipment was about to be confiscated, it was damaged beforehand. Raymond (2008) pointed out that the number of functioning tractors in the Mekong Delta declined by 76 per cent between 1975 and 1983. Agricultural products were sold to the free market by farmers, rather than to the state. In 1978, during a food shortage, a record number of peasants used rice to distill alcohol or to feed their livestock. Huu-Tho (1982) estimated 1.5 million tons of rice was used to feed pigs each year throughout the country, but the practice was particularly prevalent in the Mekong Delta. In some extreme cases, violent unrest was reported in provinces between farmers and local cadres (Dang 2010).

After collective production in agriculture was abandoned and private farming was legalised, Vietnam' s rice output rocketed, elevating the country to the third highest rice exporter. It turned out that farmers who produced a little for the state during collectivisation were quite able to produce huge surpluses for the private market (Raymond 2001).

The aforementioned arguments also further clarify the reasons for the long existence of collective agricultural production in the north,

compared with their short-life in the south. In the north, almost all villages had been effectively collectivised by the 1960s and remained so for a generation (Wurfel 1994). The process of collectivisation there did not seem to be as much a struggle as in the south because of landownership. One of the main reasons was the existence of large areas of communal land in the north since the colonial period, whereas southern land was mostly privately owned (Table A2). This fact made it easier for the collective production team to pool labour. It was a fact that productivity increased during the early years of collectivisation from 1.3 tons per hectare in 1955 to 1.9 tons per hectare in 1960 (Table A3). In addition, when the north started collectivisation in the mid-1950s, their rice productivity was still very low because of their obsolete methods of production. Meanwhile, collectivisation in the south was after 1975, when agricultural production utilised more machinery, water-pump irrigation and high-yield rice varieties (Wurfel 1994; Akram-Lodhi 2004).

With two different starting points, farmers in the south were aware of the great reduction in their benefits if they joined cooperatives. Comparing the benefits that peasants received from cooperatives or production units, Liem (1995) noted that there was a big difference in the returns that each class of peasants used to obtain from their own land under the Saigon administration (2 tons of paddy per hectare on average). On the other hand, the returns from joining cooperatives were not enough even to feed their families. Poor peasants could survive for 15-30days at most. Small landlord's survival was longer at 1.5-3 months, while middle landlords could have enough food to eat for 4.5 months. For the first time in the Mekong Delta, food shortages were witnessed and famine was threatening many localities.

The negative effect of collectivisation on output was more apparent in the south. The total paddy rice production of 6.37 million tons in 1976 plummeted by 21 per cent to 5.03 million tons in 1978 before it recovered in 1979 and 1980 following policy changes (Tri 1990). Collective production caused both southern and northern Vietnamese farmers to

produce less. Vietnam' s national rice production fell from 11.83 million tons in 1976 to 10.63 million tons in 1977 and 9.79 million tons in 1978 (GSO 1991). Rice productivity also dropped by 13.2 per cent and then 19.7 per cent, respectively, from the 1976 level of 2.2 tons per hectare (table A3). In addition to the resistance to collectivization, excessive rains from typhoons struck both northern and southern Vietnam in 1979, destroying crops in several provinces.

As a result, the communist government was unable to meet its targets for grain procurement in southern Vietnam and food shortages became a serious issue several years after unification (CPV 2005a). In 1976, state grain procurement reached only 50 per cent of its targets and even less in 1977. In the Mekong Delta, state food procurement decreased from 950,000 tons in 1976 to only 398,000 tons in 1979 (Long 1988). There was a serious food shortage throughout the whole country during the period 1976-1980 resulting in the imports of 1.7 million tons of food per year (Liem 1995).

5. Conclusion

The failure of land reforms and collectivisation in the south of Vietnam after the country's unification has been studied extensively in the literature. The arguments fall into two main categories: lack of a competitive market incentive structure and private landownership that influence farmers' efforts (Liem 1995; Akram-Lodhi 2004; Che et al. 2006) and the daily resistance of local farmers against the communist state (Dang 2010; Kerkvliet, 2006). Our study documents these explanations. It has been clear that in addition to the economic reasons, the political economy of rural society in the south and its inheritance from the previous regime contributed substantially to the unsuccessful imposition of collective agricultural production in the region.

We further argue that it was not only the resistance of the local farmers but the de facto political power of the southern middle-class peasants or middle landowners (tang lop trung nong), accounting for a large share of the rural

population, who impeded the land reform and collectivisation efforts of the communist government. Such power was traded off with local government officials for economic rents. The de facto political power helped farmers in the south maintain their agricultural production in a competitive market mechanism. In combination with the ineffective de jure political power of local authorities and cadres, this power led to the prevalence of commodity agricultural production and the failure of the collectivisation policy of the ruling government. This persistence occurred in spite of the dramatic change in the political regime in the region. Our argument is supported by a theoretical model explaining the persistence of economic institutions given simultaneous changes in political regimes in Acemoglu and Robinson (2006).

This failure in the southern part of the country also requires an explanation for the success of the same model in the north. There was a different agrarian structure before collectivisation in the northern and the southern regions of Vietnam. In the north, collectivisation transformed colonial agriculture with the aim of turning small-scale commodity production into larger-scaled and more mechanised production. In the south, before the collectivisation campaign, the market-based structure operated efficiently in the agriculture sector with more physical capital and higher technology. It is claimed that collectivisation in the south sought to transform a production system that had already been changed through two agrarian reforms in 1956 and 1970 (Akram-Lodhi 2004). Under a different economic structure and political setting, the prototype application simply resulted in failure as admitted by the administration. In 1986, a senior official of the State Planning Committee of Vietnam indirectly stated that during the 10-year period (1976-1986), the party had made lots of errors in defining the targets as well as the steps to take in socialist transformation and economic management in southern Vietnam (CPV 2006). They also failed to take into consideration southern characteristics and arbitrarily applied the northern model to the southern economy, which was not appropriate (Tri 1990). Economists in the country at that time

acknowledged that after the first 2 years following liberalisation in 1975, agricultural production slowed down because of the incorrect policy of forced collectivisation and inadequate fixed prices for farm produce according to Nguyen (1982) cited in Tri (1990).

The experience of the collectivisation process four decades ago in the southern part of the country still serves as a good lesson for continuing economic reform by the ruling party in Vietnam these days. Land tenure has never been a calming issue in Vietnam, even nowadays in the context of the country' s industrialisation reducing the agricultural use of land. The power of farmers in this new context and the security of their most important asset—land—still deserve the careful consideration of the authorities, especially when the agriculture sector is still regarded as one of the key drivers for economic growth in Vietnam.

Within the scope of this article, we have not attempted to model the stylised facts and typical features of the rural society in southern Vietnam during the period of the study. An integration of mathematical modelling for those facts into the original static model of Acemoglu and Robinson (2006) will be the topic of future research.

ACKNOWLEDGEMENTS

The authors are grateful for the financial support from the Thailand Research Fund. This study is part of the research project entitled "Spatial Dynamics and Persistence of Inequality". We thank the anonymous reviewer for providing valuable comments to improve our paper. Special thanks are dedicated to participants at the 9th International Winter School on Inequality and Social Welfare Theory: Public Policy and Inter/Intra-generational Distribution (Canazei, January 2014), the 8th Vietnam Economists Annual Meeting (Thai Nguyen, June 2015) for helpful suggestions and to Anthony Reardon for proof reading the article.

September 2015.

References

Acemoglu D, Robinson JA (2006) De Facto Political Power and Institutional Persistence. The American Economic Review 96(2), 325-330.

Acemoglu D, Robinson JA (2008) Persistence of Power, Elites, and Institutions. The American Economic Review 98(1), 267-293. DOI: 10.2307/29729971.

Akram-Lodhi AH (2004) Are 'Landlords Taking Back the Land'? An Essay on the Agrarian Transition in Vietnam. The European Journal of Development Research 16 (4), 757-789.

ATCS (1984) Bao cao tinh hinh ruong dat. Ban cai tao nong nghiep mien Nam, Ho Chi Minh city.

Callison CS (1983) Land-to-the-Tiller in the Mekong Delta: Economic, Social and Political Effects of Land Reform in Four Villages of South Vietnam. Center for South and Southeast Asia Studies. University Press of America, Berkeley, Lanham.

Canh H (1988) Cach mang xa hoi chu nghia va su bien doi co cau xa hoi nong thon Viet Nam [Socialist Revolution and the Social Structural Change in Rural Vietnam]. Xa hoi hoc, 1-2(1988).

Che TN, Kompas T, Vousden N (2001) Incentives and Static and Dynamic Gains from Market Reform: Rice Production in Vietnam. The Australian Journal of Agricultural andRe-source Economics 45(4), 547-572.

Che TN, Kompas T, Vousden N (2006) Market Reform, Incentives and Economic Development in Vietnamese Rice Production. Comparative economic studies 48(2), 277-301.

CPV (2004a) Van kien Dang toan tap. (Tap 36 -1975). Nha xuat ban Chinh tri Quoc gia, Hanoi.

CPV (2004b) Van kien Dang toan tap. (Tap

37-1976). Nha xuat ban Chinh tri Quoc gia, Hanoi.

CPV (2005a) Van kien Dang toan tap. (Tap 42 - 1981). Nha xuat ban Chinh tri Quoc gia, Hanoi.

CPV (2005b) Van kien Dang toan tap. (Tap

38-1977). Nha xuat ban Chinh tri Quoc gia, Hanoi.

CPV (2006) Van kien Dang toan tap. (Tap 471986). Nha xuat ban Chinh tri Quoc gia, Hanoi.

Dang TD (2010) Post-1975 Land Reform in Southern Vietnam: How Local Actions and Responses Affected National Land Policy. Journal of Vietnamese Studies 5(3), 72-105. DOI: 10.1525/vs.2010.5.3.72.

Dinh TH (1994) Qua trinh bien doi ve che do so huu ruong dat va co cau giai cap o nong thon dong bang song Cuu Long (19691975) [The Process of Change in Land Tenure and Class Structure in the Rural Areas ofthe Mekong Delta (1969-1975)]. NXB Khoa hoc Xa hoi, Hanoi.

Fforde A, Vylder S (1996) From Plan to Market: The Economic Transition in Vietnam. Westview Press, Boulder.

GSO (1991) Statistical Yearbook [Nien giam Thong ke]. Tong cuc Thong Ke, Hanoi.

Huu-Tho N (1982) Reflection in the Food Problem. Vietnam Courrier 3, 21.

Kerkvliet BJT (1995) Village-State Relations in Vietnam: The Effect of Everyday Politics on Decollectivization. The Journal of Asian Studies 54(2), 396-418. DOI: 10.2307/2058744.

Kerkvliet BJT (2006) The Power of Everyday Politics: How Vietnamese Peasants Transformed National Policy. Journal of Vietnamese Studies, 1(1-2), 499-501. DOI: 10.1525/vs.2006.1.1 -2.499.

Ladejinsky W (1964) Agrarian Reform in Asia. Foreign Affairs 42(3), 445-460.

Lam QH (1985) Cach mang ruong dat a mien Nam Viet Nam [Land revolution in Southern Vietnam]. Khoa h c xä h i, Hanoi.

Liem LT (1995) Chinh sach cai cach ruong dat Vietnam (1954-1994) [Land Reform Policies in Vietnam (1954-1994)]. (Vol. 100). Nam A, Paris.

Long NV (1988) Some Aspects of Cooperativization in the Mekong Delta. In DMaC White (ed.), Postwar Vietnam: Dilemmas in Socialist Development: Doi Moi in Comparative Perspective (pp. 165208). Southeast Asia Program, Cornell University, Ithaca.

Nguyen XL (1982) The Family Economy of Co-opertive Farmers. Vietnamese Studies 13, 108-128.

Pingali PL, Xuan VT (1992) Vietnam: Decollectivization and Rice Productivity Growth. Economic Development and Cultural Change 40(4), 697-718. DOI: 10.2307/1154630. Raymond C (2001) The Insoluble Internal Conflicts of Agricultural Collectivization in Vietnam.

Raymond C (2008) "No Responsibility and No Rice": The Rise and Fall of Agricultural Collectivization in Vietnam. Agricultural History 82(1), 43-61. DOI: 10.2307/20454780. Salter M (1970) The Broadening Base of Land Reform in South Vietnam. Asian Survey 10 (8), 724-737. Sansom RL (1970) The Economics of Insurgency in the Mekong Delta of Vietnam. The MIT Press, Cambridge. Tri VN (1990) Vietnam's Economic Policy Since 1975. Institute of Southeast Asian Studies, Singapore. Tuan DT (2001) Workshop of the World Social Forum. IRAM-APM-CONTAG, Porto Alegre.

Tyson JL (1973) Land Reform in Vietnam: A Progress Report. Asian Affairs: An American Review 1(1), 32-41. Wurfel D (1957) Agrarian Reform in the Republic of Vietnam. Far Eastern Survey 26, 81-92. Wurfel D (1967) The Saigon Political Elite: Focus on Four Cabinets. Asian Survey 7 (8), 527-539. DOI: 10.2307/2642594. Wurfel D. (1994). Change and Continuity in Vietnamese Villages: The Consequences of Decollectivation. Paper presented at the 13th Conference of the International Association of Historians of Asia, Tokyo.

Appendix A 1.

A 1. The Sketch of the Theoretical Model

The model (Acemoglu and Robinson 2006) assumes a small finite number of land-owning elites (E) and a large finite number of citizens (C) or workers. The choice of economic institution in the model is the form of labour market institutions, which can either be competitive with wages equal to marginal products or monopsonic with a lower wage level. The

key economic decision in this model is the form of economic institutions (in this case is the labour market), which affect the rents of the elite. Labour markets can either be competitive (t =1) with wages equal to marginal product or monopsony (t = 0) where wages are kept at lower level and the elite can make profits equal to R > 0.

Meanwhile, the political system (S) can either be a democracy (D) or non-democracy (N). One of the important features of this model is the distinction between de jure and de facto political power. De jure power is a type of power allocated by political institutions (such as constitutions or electoral systems), whereas de facto power emerges from the ability to engage in collective action, or the use of brute force or other channels such as lobbying or bribery. When S= D, de jure political power is more evenly distributed and workers have greater power in the political process because they are in the majority. With a non-democratic system S=N, the elite has greater de jure political power.

In either system, collective decisions are made by the group with greater overall power, which is a combination of de jure and de facto. The elites can invest to gather further de facto political power. The cost of this effort is expressed in terms of the final goods. If elite agent i spends an amount of ff as a contribution to activities increasing their group's de facto power, then the elite's de facto political power of the whole group will be

pe = (0 > 0)

On the other hand, the political power of the citizens is given by

PC = m + nI (S = D)

where I (S = D) is an indicator function for democracy n > 0 and m is a continuous random distribution and strictly decreasing density function.

If PE > PC, the elite decides t = 0, otherwise the citizens decide t =1, then m is realized depending on whether PE > PC. The equilibrium of the game is therefore straightforward under democracy or non-democracy.

The choice of ff for elite agents is conditional on the political regime S when all other elite agents choose ff#(S). Total elite power in regime S will be

pE = 0(2 ff(s) + ff'j

In a non-democracy, the agent will choose ff to maximise profits net of investment cost ff while R is the return from a monopsonic labour market.

-ff + ^ff(S = N) + ff'j j R

The second term of the equation represents the probability of the elite having greater power than the citizens and thus each elite agent receives the return from monopsony, R.

In a democracy, the elite i faces a maximising problem with the only difference is the presence of n representing the greater de jure power of the citizens in a democracy.

01 xe(s = d) + e; ) - n \R

The first-order conditions show the main result if there are positive contributions by the elite in their de facto power in both political regimes, that is, ^8'(S=N) > 0 and ^(S=D) > 0; this implies that

F(0CE0i(S = D)) - n)

= F (0(^0i(S = N)))

In other words, the elites' investment in their de facto power in democracy is more than suf-

ficient to maintain overall political power. Therefore, they can decide on their favoured economic institution of a monopsonic labour market (t = 0). It means that the probability of economic institutions favouring the elite is independent of whether the political regime is democratic or non-democratic. Economic equilibrium institutions are a result of the effects of these two sources of power. There can be equilibrium changes in political institutions and thus de jure political power; but these induce offsetting changes in the distribution of de facto political power. In a democracy, the elites invest more in their de facto power to ensure the same probability of maintaining overall political power. The probability of economic institutions favouring the elite is independent of whether the political regime is democratic or non-democratic; in other words, when elites who monopolise de jure political power lose this privilege, they may still exert disproportionate influence in politics by increasing the intensity of their collective action (for example, in the form of greater lobbying, bribery, or brute force) and thus ensure the continuation of the previous set of economic institutions.

It is also straightforward with this model to verify that d PrgR~01 > 0, suggesting when rents are higher because of monopsony in the labour market R, the greater is the investment of the elite in their de facto power and the greater is the likelihood that they will succeed in imposing their favoured economic institutions. In addition, the probability of imposing a monopsonic labour market depends on the technology that generates de facto political power for the elite, that is, higher p, also increases Pr(T = 0) so thatd ^g^011 > 0.

Table A1 Land Distribution Before and After the Agrarian Reform in the North, 1953

Before the reform After the reform

Households (%) Land (%) Households (%) Land (%)

Landowners 1.1 18 2.3 1.1

Rich peasants 1.8 4.7 1.6 1.8

Middle-class peasants 41.7 39 36.5 11.7

Poor peasants 40 25.4 43 40

Landless 12 6.3 13 12

Others 2.6 1 3.6 2.6

Monasteries 0.1 1.3 — 0.1

Communal land 0.7 4.3 — 0.7

Source: Tuan (2001) from Statistics General Office, 1930-1980, Hanoi.

Table A2 Land Distribution in the South Before and After Land Reforms

Landowner (%) Land surface (%)

Category 1955 1966 1955 1966

Landless 46.7 42 0 0

0.1-4.9 ha 38.6 45.3 16.4 27.4

5.0-19.9 ha 7.8 10.5 13 33.3

20.0-49.9 ha 5.6 1.6 24 15.6

50.0-99.9 ha 0.7 0.5 12.5 12.1

more than 100 ha 0.5 0.2 34.1 11.6

Source: Tuan (2001) from Callison (1983).

Table A3 Rice Productivity and Rice Availability, 1942-1986

North South Total

Yield per Per capita Yield per Per capita Yield per Per capita

hectare rice output hectare rice output hectare rice output

Year (tons/ha) (kg/year) (tons/ha) (kg/year) (tons/ha) (kg/year)

1942 1.2 190 1.3 420 1.3 280

1955 1.3 150 1.3 230 1.3 190

1960 1.5 200 2.2 390 1.9 280

1965 1.5 180 2.0 300 1.8 230

1976 2.1 210 2.2 270 2.2 240

1980 1.9 160 2.2 290 2.1 220

1984 2.5 204 2.9 331 2.8 266

1986 2.6 200 3.0 330 2.8 260

Source: Pingali and Xuan (1992).

Table A4 Rice Production Performance, North and South Vietnam, 1950-1987

Time period

Growth in cultivated area per annum

Growth in yield per hectare per annum

Growth in total rice production Population per annum growth rate

Vietnam (North and South)

1950-1955 2.79

1956-1965 0.33

1966-1975 1.59

1976-1981 1.14

1982-1987 0.08 North Vietnam

1950-1955 0.22

1956-1965 0.85

1966-1975 -0.24 South Vietnam

1950-1955 5.63

1956-1965 -0.13

1966-1975 3.18

0.74 0.16 1.82

2.06 4.81 2.31

2.63 3.8 1.91 2.81

4.64 5.48

2.05 2.72 3.1

2.6 2.6

2.4 2.09

3.3 3.8

Source: Pingali and Xuan (1992).