Soft budgets and elastic debt: farm liabilities in the agrarian political economy of post-Soviet Tajikistan

ABSTRACT

Since Tajikistan’s independence, market reforms and pressure from international donors have brought changes to the state’s role in the economy. The official narrative holds that the post-socialist state reduced its control over agriculture, but there are still various mechanisms through which it exerts control over farming. In this paper, I examine Tajikistan’s post-socialist agrarian change through the prism of farm debt. Farm debt used to be an accounting nuisance in Soviet agriculture as a result of so-called soft-budget constraints. In the political economy of post-socialist transformation, farm cotton debt has been transformed into indebted land. I classify this debt ‘elastic’ for its ambiguous nature. It ties farmers to land and makes farmers’ independence illusory. With an in-depth analysis based on original ethnographic insights, I aim to provide a theoretical contribution to the way in which debt is conceptualised and politicised in post-Soviet Tajikistan.

KEYWORDS:

• Agrarian change
• political economy
• post-socialism
• debt
• soft budget constraint
• cotton
• Tajikistan

View correction statement:

Corrigendum

1. Introduction

‘Before we would start working in cotton there, I would need to know who owns those farms’, the local employee of an international donor told me in 2013. I wondered: ‘Who owns those farms? Those are individualised farms holding a proper title. Doesn’t that say enough?’

Wandering through the central market of Dushanbe, on any given morning, one can only be impressed by the range of fresh fruit and vegetables on offer at every stall. The socialist shortage economy seems to be firmly in the past. Meanwhile, foreign and domestic agribusinesses are slowly emerging. One tends to conclude that agrarian modernisation in Tajikistan has finally begun in earnest. However, behind this façade, farm practices and rural livelihoods are restricted by rent-seeking elites in areas of the state’s benign neglect.

Tajikistan’s agrarian economy was set for reform in 1991, in accordance with post-Soviet transformation across the former Soviet Union. Although the state officially withdrew from the agricultural sector to abide by conditions set by international donors, strongmen¹ personally close to the presidential family have found ways to safeguard control over particular areas and agricultural revenue streams. The production of cotton – condemned as much as praised as the ‘white gold’ (cf. ICG 2005; Spoor 2007; Van Atta 2009) – has played a prominent role in shaping local pathways of agrarian transformation in post-Soviet Tajikistan. One way in which this takes place is through sustained farm indebtedness.

In this paper, based on longitudinal fieldwork, including interviews with farmers, rural dwellers, state officials (at local, district and national level) and foreign consultants employed by international donors, I will analyse the role of cotton debt in the process of post-socialist agrarian change in Tajikistan.² In so doing, I have a threefold objective. First, I aim to discuss the particularities of post-socialist agrarian transformation in Tajikistan, through the prism of debt. By describing how the role of debt has changed in this setting, from a collectively held farm liability to an ‘elastic’ individually held debt, I aim to show that post-Soviet agrarian change in Tajikistan has not resulted in far-reaching farm privatisation. Instead, a primarily nominal farm individualisation has taken place. A second and related objective is to contribute to the anthropology of debt. Building on concepts of accounting in socialist economies (Kornai 1980; Bockman, Fischer, and Woodruff 2016), I will demonstrate how farm debt emerged due to the continuation of ‘soft budget constraints’ in the cotton economy and how the function of the debt, and its perception, was transformed. Debt became a mechanism that tied people to land and controlled farm labour. In so doing, I aim to provide a more in-depth ethnographic understanding of this debt problem than recent analyses (see for instance, Van Atta 2009; Kassam 2011; Nekbakhtshoev 2016). My final objective is to contribute to understanding land governance in this little-explored context of post-Soviet Central Asia, where neo-patrimonialism characterises the political economy of agrarian change (see also Boboyorov 2012; Sehring 2009). In this way this paper engages literature concerning post-socialist agrarian change (cf. Verdery 2003; Trevisani 2007; Allina-Pisano 2008) and the scholarly debates on debt, agrarian debt and debt peonage (cf. Graeber 2011; Gerber 2014; Guérin 2013).

My paper is structured as follows. The following, second, section provides the theoretical underpinning of the paper. It builds on economic anthropology to address issues of land, property and debt. I describe socialist farm accounting rationales as inherent features of a (agrarian) political economy. The third section proceeds with an analysis of the emergence of cotton debt in Tajikistan. In the fourth section, I analyse the transformation of cotton debt that has occurred since 2009 and the implications of debt for rural livelihoods. The fifth section concludes the paper.

2. Defining debt and property in different economic regimes

2.1. Accounting rationales and shortage economies

The ways in which production factors such as land, labour and capital are valued and appreciated depend on the particular economic and political regime (Appadurai 1986). Land has a very peculiar characteristic in the sense that it is fixed: it has ‘imperfect substitutability, it is illiquid, and has limited divisibility’ (Fairbairn 2014, 782). Land, and also labour and capital, differ from commodities or assets, as Polanyi (1944/2001, 76) noted they are fictitious commodities: ‘it is with the help of this fiction that the actual markets for labour, land, and money are organised’.

In the Soviet Union, land and labour were not commodified. Farmland that belonged to a kolkhoz or sovkhoz was not included on the farms’ balance sheets. Moreover, prices were not used to allocate resources,³ but were ‘mere extras in bureaucratic coordination’ (Kornai 1992, 148). Exceeding the budget did not always have repercussions for the farms’ functioning. Debt was regarded as a purely technical matter, to be solved by internal cross-subsidisation or by heightening crop procurement prices (Kornai 1992; Kitching 2001). This accounting rationale underpins Kornai’s (1980) concept of ‘soft budget constraints’. A firm with soft budget constraints decides on production levels based on the available input, not on production costs and consumer demand. In this way, a rational price evaluation is absent, and

accordingly,  …  demand is hardly constrained by solvency considerations. The firm, as buyer, tries to acquire as much input as possible in order that shortage could not hinder production. At the other side, the firm, as seller, faces an almost insatiable demand. (Kornai 1980, 29)

This is why the socialist economy can be seen as an ‘economy of shortage’ (Kornai 1980). Prioritising continued production over profit and attributing importance to overcoming unemployment were clear political decisions (Kornai 1980; Bockman, Fischer, and Woodruff 2016).⁴ As such, a debt in Soviet agriculture was inherent to the socialist agrarian political economy with implications for the way in which it was perceived.

That being said, although the concept of soft budget constraints has often been used to explain the functioning of (in economic terms) underperforming firms in socialist economies, capitalist economies also feature soft budget constraints, in cases where the state or private creditors consider a firm’s survival important and bail it out (cf. Visser and Kalb 2010). As a result, the ‘hardness’ of budget constraints depends on the (local) political economy, rather than on the overall economic regime, and this hardness can vary within an economy across localities and over time (Amelina 2002; Woodruff 1999).

Amelina (2002) consequently qualified such frequently oscillating budget constraints as ‘elastic’, instead of ‘soft’ or ‘hard’. As Amelina describes in the context of post-Soviet Russia, farms’ accounting rationales and budget constraints became more locally defined with post-socialist transformation, shaped by bargaining and manoeuvring between agents of the state and farm leadership. A debt could be tolerated in one season, but authorities could harden budget constraints in the next, so making the possibility of bankruptcy real. As a result, interests and actions of the local state could conflict with nationwide economic reforms (Woodruff 1999).

2.2. The anthropology of debt

2.2.1. Debt (as property) in the post-socialist realm

If we see the transition as a project of cultural engineering in which fundamental social ideas are resignified – including not only ‘markets’, ‘democracy’, and ‘private property’ but also ideas about entitlement, accountability, and responsibility – then the (re)creation of private property is evidently a critical locus for this cultural project. (Verdery 1999, 54)

Ethnographies on post-socialist land reform (cf. Allina-Pisano 2008; Verdery 2003; Trevisani 2007) show how the division of assets and liabilities unfolded along with the breakdown of the Soviet bloc and the attempts to create market economies (see also Stark 1996). Collective debt, private gains and value appropriation were part and parcel of this process. Liabilities and debt tended to be individualised, as were farmland and farm assets.

The process of redefining property in the post-socialist setting has not necessarily proceeded along formal registers (cf. Verdery 2003; Allina-Pisano 2008). The specific allocation of property was often obscure. People held a variety of overlapping rights. Those with administrative or allocative power held the authority to determine who would get what; power constellations were highly important for the final redefinition and redistribution of socialist property.

2.2.2. Debt in agriculture

In an agrarian context, debt has often been associated with the commodification of agriculture and contract farming schemes (McMichael 2013; Watts 1994; Gerber 2014). From the perspective of institutional economics, contract farming is regarded as a way to minimise transaction costs and the possibility of market failure (Djanikbekov et al. 2013). In international development aid discourse, the inclusion of smallholders into commodity chains and global food markets tends to be seen as a way to relieve rural poverty and improve global food security. Critical analyses point, however, at the often-asymmetric power relations and socioeconomic differentiation inherent in contract farming and outgrower schemes (McMichael 2013; Watts 1994).

Notably, the characteristics of contract farming, like vertically coordinated production chains and standardised production patterns, are not typical of agrarian capitalism, and can be found across time and place. For instance, production relations in colonial plantations, feudal estates and socialist agriculture also featured an industrial labour process with a highly specified division of tasks. These entailed labour bondage to varying extents (cf. Veldwisch and Spoor 2008; Ransom and Sutch 1972; Byres 2016; Guérin 2013).

Particularly in the production of cash crops that cannot be (directly) consumed (e.g. sugar, cotton and coffee), farmers are inclined to engage in more complex interdependencies. And yet, single-crop contracts or loans restrict farmers’ options to diversify and increases farms’ risks (McMichael 2013). With production risks left to the farmer, a fluctuating production season may result in perpetually increasing debt. These, in turn, may force farmers into treadmills of production, with the risk of dispossession (Gerber 2014). ‘The contract … defines the social space of autonomy and subordination that the grower occupies in relation to the labour process. It maps … the topography of direct and indirect control’ (Watts 1994, 27).

In so doing, debt can become an exercise of power in contract farming arrangements, as being instrumental in fulfilling the terms of the contract (cf. Guérin 2013; McMichael 2013). Along these lines, Byres (2016, 444) described the thirlage imposed upon farm tenants to landlord services in feudalist Scotland. Ransom and Sutch (1972) analysed spiralling debts in the cotton sector in post-bellum America. Here, the local merchants’ monopoly in supplying credit to farmers allowed the merchants to enforce the growing of cotton over food crops. These analyses highlighted that debt, through its often-coercive nature, ‘transforms or reinforces patterns of capital ownership and therefore the power structure’ (Gerber 2014, 743). In more nuanced terms, debt shapes people’s behaviour and it structures society, just like barter and gift (cf. Mauss 1954; Humphrey and Hugh-Jones 1992; Graeber 2011), and debt exists in every society and regime of accumulation. Complexities and entanglements and the ways in which debt is perceived, however, differ from one context to another.

2.3. Landed property, authority and autonomy

Understanding the emergence of debt in agriculture, as well as its relation to land, brings one back to the juncture of property, authority and autonomy (Verdery 2003; von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006). Property rights over land have been at the core of debates in development studies and ‘transition’ literature, in which advocates of neoliberal reform have explained failures in agrarian development through a lack of private property rights over land (cf. De Soto 2000; criticised by von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006). I follow critiques in thinking that this reflects a naïve understanding. For instance, by only juxtaposing the strands of ‘private’, ‘public’ and ‘communal’, one may overlook salient aspects of the users’ authority and autonomy over production (Ribot and Peluso 2003; von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006). Relations of production, and duties and liabilities that come with property, mediate access to resources. Farmers may hold private ownership over farmland, but external factors like the state, the market, technologies and knowledge (Ribot and Peluso 2003, 173) may condition property utilisation. As a result, land use rarely depends on the landowner alone. ‘Behind the mask of private ownership often lurk complicated arrangements, and management rights often become more important than the residual ownership aspect of property’ (von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006, 20). Analysing the ‘bundles of power’ over resource access and use – rather than focusing only on the ‘bundle of rights’ – is indispensable for understanding the way in which resource revenues are distributed in society. As I will demonstrate below, and as the introductory sentence of this paper also pointed to, the crucial issue in Tajikistan has not been whether the title deed has been transferred – that is, the property rights over land – but whether production relations have altered. This is because production relations ultimately determine the distribution of production revenues.

3. Cotton debt in Tajikistan

Building on the concepts discussed above, in this section I explore the genealogy of cotton debt in Tajikistan. I will describe its early emergence, and in the section that follows I will analyse how farm debt was transformed after 2009 in a specific locality, and how debt has impacted farm livelihoods since then.

Tajikistan’s initial farm debt situation echoes developments in other post-socialist republics in the 1990s, especially the problem of perpetual farm debt. In contrast to Central and Eastern Europe, where debt settlement and farm reorganisation were accomplished early on, several countries belonging to the Commonwealth of Independent States (CIS) were confronted with persisting and accumulating farm debt (cf. Csaki, Lerman, and Sotnikov 2001).⁵ The complex nature of the farms’ organisation meant that debt resettlement schemes seen in capitalist economies (following ‘standard financial ratios’; Csaki, Lerman, and Sotnikov 2001, 44) could not fit the post-socialist farm enterprises. One aspect they could not cover, for instance, was the farms’ important social role in the countryside. Nevertheless, the problem of farm debt in most of the former Soviet Union was resolved in the 2000s. This was not the case in Tajikistan, where the regime held an interest in upholding debt, as I will describe in more detail below.

3.1. A prelude to Tajikistan’s post-Soviet farm debt: the futures system

The narrative of Tajikistan’s persistent farm debt phenomenon started in 1996, when the Tajik government abolished the crop procurement system and state financing of agriculture, to meet the requirements of ‘transition’ schemes set by international donors. At that time, the government feared that this measure would mean the demise of cotton production, since farms would be left without inputs. As it was under the Soviet regime, cotton revenues have been critical for the post-Soviet economy. Cotton has been the principal agricultural export commodity. In 2014, cotton exports comprised USD 132.1 million, equalling 13.5 per cent of total exports (TajStat 2016). The state budget has also benefited from high taxes levied on various segments of the domestic cotton production complex, which cumulatively have added up to over 30 per cent of total national tax revenues (World Bank 2012).

In order to compensate for state withdrawal, the government set up a triadic system through which the cotton-producing farm enterprises (the former kolkhozes and sovkhozes and newly established dehqon farms – as the commercial farms that were to succeed former Soviet farms)⁶ could receive advances (working capital in cash and in kind, such as seeds and fertilisers), in which cotton served as collateral. This became known as the futures system, and was in fact a reintroduction of the former Soviet credit supply system, similar to the tovarnii kredit in Russia (as described by Kitching 1998, 7). Through the futures system a few regime insiders – strongmen – seized control over the cotton sector, which meant that the privatisation of the cotton sector resulted in crony capitalism, typifying the neo-patrimonial power configurations of the post-Soviet state.

The futures system involved state, domestic and international actors. Different state agencies played a regulating role in quality certification and exports. The National Bank of Tajikistan (NBT), which also provided guarantees for credits from international agents, issued export licenses. Licenses were issued to only a small number of domestic ‘futures companies’ and the functioning of investment banks and futures companies was under the control of a few strongmen as cronies of the state. Their responsibilities overlapped (Nakaya 2009). With their extensive reach futures companies almost entirely replaced the former authority of the Soviet regime in the cotton economy; the critical differences were that the companies were privately held, and driven by a desire to accumulate private capital (ICG 2005; Van Atta 2009). Farms’ inclusion in these schemes was involuntary and movement of cottonseed outside prescribed localities was prohibited (Van Atta 2009; Porteous 2003). The futures companies’ monopsony empowered them to manipulate the farm gate prices of cotton, without adjustment to the (usually higher) world market prices (Kassam 2011, 7). Farm inputs were also regularly overpriced and delivered late. In this way, the futures system facilitated extraction of capital from the countryside. Export earnings remained in a few hands within the regime (ICG 2005; Van Atta 2009; Boboyorov 2012).

The actors in the cotton sector knew the critical role that cotton production had for the state. The companies as well as farms assumed there was no risk of bankruptcy as they anticipated state support in this regard. This situation encouraged irrational input supply and demand by cotton-producing farms. As was the case in Soviet agriculture, input supply was based on expected output. If the expected production was inflated, it was possible to request greater inputs than what was needed (cf. Kassam 2011). Hence, the soft budget constraints that had featured in the socialist economy continued in post-socialism (as also noted in other post-Soviet and post-socialist contexts by Amelina 2002; Woodruff 1999; Csaki, Lerman, and Sotnikov 2001; Shtaltovna, Van Assche, and Hornidge 2012). However, the former in-kind budgets that characterised Soviet agriculture, as well as farm debt, had become monetised in this post-Soviet setting (as also observed by Burawoy 2001 in the Russian context).

In the late 1990s and early 2000s, the reorganised Soviet farms (by that time renamed as farm associations and dehqon farms) had no choice but to work through futures contracts, as very few banks offered agrarian financing. Resorting to barter outside the futures system would not solve the need for specific inputs, since the private market in agricultural supplies was weakly developed. This also meant that these reorganised farms had difficulties in obtaining inputs for crops other than cotton, which in turn led them to increase input demands for cotton and divert these to the production of other crops.

Kassam (2011, 7) therefore maintained that cotton debt was ‘fictitious’, stating that it was merely ‘a paper loss on the account of the [collective] farm’ – little more than a form of private taxation on the diversion of inputs. Following this line of argument, one can conclude that the perception of post-Soviet farm debt was similar to that of Soviet farm debt and that it had few repercussions for farms. However, I contend that debt was much more than that, consistent with Van Atta (2009) who characterised the debt as fictitious because of the fact that futures companies accumulated capital through which strongmen enriched themselves. Debt had a significant role, firstly in securing cotton production. In recent years, in specific localities, debt has tied people to their land. There was no proper documentation held by farmers on their indebtedness, but it became increasingly difficult to pay off their debt, which meant that a treadmill of production emerged. ‘Basically the futurist [sic] is saying: “look, you have no choice. You have to keep producing and if you want to keep producing, take our terms”’ (interview, foreign consultant, 17 February 2015).

3.2. State efforts towards liberalising the cotton economy

Driven by international donor pressure to spur farm reorganisation and by decreasing cotton production figures, the Tajik government, since the early 2000s, has made several attempts to solve the debt issue, mainly through legislative changes. In 2002, the futures companies' local monopsonies were formally eliminated and all farmers were then free to select their outlet for cotton regardless of their location (Van Atta 2009). Nevertheless, farmers’ options were limited: apart from persistent pressure by local elites on farms to take their terms, the farms also struggled with cash shortages and transport problems.

An important decree was promulgated in 2003, which entailed that farms’ cotton debt was to be redefined into a debt per hectare, with repayments to start in 2005. This decree transformed cotton debt into a land-tied debt – the so-called indebted fields (qarzi zamin). The redefinition gave the farm debt problem in Tajikistan a very peculiar characteristic, since – as in most cases of farm debt restructuring in the former Soviet Union – land remained outside the debt settlement procedure (as in the cases described by Csaki, Lerman, and Sotnikov 2001). The consequence of this decree in Tajikistan was that farm shareholders would be held liable for a part of the collective dehqon farm’s debt once they withdrew their share to start their own farm. Even though this procedure was not in accordance with the Tajik Civil Code, which was proven by donor-initiated pilot court cases, the decree was introduced.⁷ Notably, court cases were not likely to be initiated in localities where farmers lacked the backing of an international donor.

3.3. Donor interference

Several international donors had their own vision about debt resolution, but their projects all had little impact. Strongmen thwarted any meaningful changes, since debt served as a lever of control over farms. Tajik authorities saw technical solutions for the problem, and expected international donors to step in with financial aid. Structural change in production relations did not take place. As a foreign consultant commented:

If you read all the cotton sector project reports, from about 2002 to about 2007, they all say, ‘well, the problem is that our cotton yield is not high enough. We produce about two tonnes of cotton per hectare. As soon as we produce three, then we can solve our problems’. So they brought in new grades and standards, they intended to bring in new, better seed, but nothing worked, because the farmers couldn’t afford it, and … you know, the good farms can get funding anyway, since they [well-run farms] are not running up debts. (interview, 17 February 2015)

Pressured by international donors, the Tajik government adopted Cabinet Resolution (111) on ‘Freedom to Farm’ in March 2007, which would guarantee farmers freedom selection of crops and crop outlets. The futures system would also be abolished at the end of 2007. However, these changes remained only on paper. On the ground, the situation did not change and at the end of 2007, the total nationwide cotton debt had risen to USD 548 million⁸ (Governmental Decree 663), but debt was unevenly spread throughout the country, with higher debt concentrations in the southwest Khatlon region (Government of Tajikistan 2007), where political domination over the cotton sector was more pronounced (cf. Hofman and Visser forthcoming).⁹ ‘Seventy-four per cent of the cotton processing was controlled by five firms, of which the two largest controlled 20 per cent of the national cotton market’ (Nakaya 2009, 266; see also KasWag AgriConsulting Worldwide 2008; Government of Tajikistan 2007). One of the primary actors was the head of the National Bank, who held private interests in the cotton sector. After the revelation of hidden sovereign guarantees for cotton finance, discovered by the International Monetary Fund (IMF), he had to step down.¹⁰

The revelation of this scandal coincided with an extremely cold winter in which international donor assistance was critical to the population in the form of food and emergency aid. These crises provided donors the opportunity to more strongly pressure the Tajik state to reform the cotton sector. It resulted in the adoption of an extensive agrarian reform programme (the ‘Cotton Debt Road Map’) with land reform and a debt resettlement scheme at its core. The government would write off all cotton-related debt accumulated by farms by 1 January 2008 (Decree 663 and Resolution 312), and the ‘Road Map’ programme declaration pointed out that

Investors who obtained [this] finance through Kredit Invest will be obliged to take this loss in full, while independent investors and commercial banks will have their debt write-off offset by government securities. Non-cotton loans made by investors will be repaid to the [National Bank of Tajikistan] at concessional rates. (World Bank 2010, 2)

The ‘Road Map’ also implied an end to the role of the National Bank in issuing export licenses for cotton. This had previously given the state a powerful mechanism to control the market. Scrapping it was meant to create a more market-based environment for cotton ginneries, with more transparent pricing. The World Bank set up projects to raise the rural population’s awareness of farmers’ rights and the debt resolution programme, and would start a small number of local pilot projects on debt restructuring.

To summarise this section, farm debt emerged as a result of the socialist legacy of soft budget constraints which characterised the agrarian political economy. This debt was related to factors both internal and external to the organisation of the farm. As I will demonstrate in the next section, the debt logic changed locally after 2009, as it became subject to local power plays and social inequality. As alluded to above, debt sheds light on social relations and societal transformation – which is also the aim of the present paper.

4. Debt continued: the local political economy of cotton

With the ‘Road Map’ in place, the cotton debt problem gradually seemed to disappear from the international donor agenda, even though the political will to resolve the debt problem was questioned. The government had earlier proposed to leave debt restructuring with futures companies, with which donors disagreed, since this would create transparency issues (KasWag AgriConsulting Worldwide 2008). But in fact this is what eventually happened. The political will to change production relations in cotton farming, through which the debt had emerged and was upheld, was absent. Whereas in particular districts cotton growing dehqon farms were freed from debt, with more freedom to farm as a result, this has not been the case everywhere. Hence, the state and donors’ benign neglect of particular areas has enabled elites to persevere in claiming debt. One of these areas is a locality in southwest Tajikistan, where cotton debt ‘survived’. I describe this case below, based on ethnographic fieldwork undertaken in three periods: April–August 2012; May–September 2013; and September 2014–February 2015. While temporarily residing in the village, I had numerous, daily conversations with villagers and over 50 farmers as well as local and district authorities, including the land committee and the district Agroprom (the agricultural department of the district government). These conversations did not directly address cotton debt, as this topic was heavily politicised. Instead, my longitudinal research, in the form of frequent revisits over the years, with a Tajik field assistant, generated trust. This has allowed me to gain insight into the history of the village, its social structure, people’s livelihoods and, gradually, the on-the-ground reality of cotton debt.¹¹

4.1. Land and liabilities in Jaloliddini Balkhi

The district of Kolkhozobod – literally: ‘developed through the kolkhoz’ – was a prominent district in Soviet times.¹² Flourishing research stations were located in the southern villages, where Soviet experts were employed to conduct research on subtropical crops and trees. The inhabitants had been resettled there by the Soviet regime from Tajikistan’s northern mountainous districts in the 1950s in order to work on the newly established cotton kolkhozes.

After the Soviet breakup and the Tajik Civil War (1992–1997), district and local power shifts gradually took place, including a change in kolkhoz leadership. Through this change, the government aimed to safeguard political representation and fulfil its political goals on the ground; one of these goals was securing cotton production (Boboyorov 2012). Moreover, the power reshuffling meant a loss of villagers’ representation at several levels of government.

When state procurement of crops was abolished in 1996, as explained above, futures companies in the hands of a small number of well-connected strongmen took control of the cotton value chain. This also happened in the district of Jaloliddini Balkhi. With privatisation of the cotton sector, the district became subject to one strongman’s power. Particularly important in his political power and ability to influence authorities were his ties to the ruling presidential family, as he had married into this family.¹³

Local authorities – such as Agroprom – then followed his (and therewith the regime’s) interests, controlling and prioritising cotton farming over other kinds of production. Production plans set by the central regime were pushed through by district authorities (particularly as the district governor holding office during my period of fieldwork held a ginnery), who, in turn, pressed lower authorities (jamoats) and collective dehqon farm chairmen to safeguard compliance (on similar observations see Boboyorov 2012). For local and district authorities, fulfilling mandates was important in order to secure succession. Local authorities yearly gathered the farmers to urge them to transport their cotton harvest to specific ginneries. In this way, the regime’s interests in the cotton economy were endorsed from the top to the bottom.

The entanglements of state and private interests made it very difficult for villagers to distinguish actors and interests, even though the paternal state had vanished. As a result, the already low level of trust in local authorities decreased, as villagers articulated the loss of representation: ‘what to do, the old rais [collective farm chairman] is dead, and there is no one listening to us’ (interview with a landless villager, 17 July 2012). People perceived protests to be worthless, since legal aid centres, lawyers and prosecutors, in locals’ eyes, were all related to a small circle of strongmen (as a dehqon farmer stated, 31 July 2013).

The lack of public investment resulted in dilapidated infrastructure, with consequences such as waterlogging, severe soil salinisation and desertification. Authorities ignored the villagers’ pleas for support, including complaint letters to local authorities and to ministries in the capital Dushanbe. Even though the Ministry of Agriculture had analysed the area and had declared it to be severely affected, land taxes had not been lowered. As a farmer complained:

Every time when the tax payment comes, I show the letters, but … the only thing they demand … is the payment. If they are not impressed by a letter stamped by the ministry, why should they listen to a farmer in the first place? (Interview 20 August 2013)

4.2. Qarzi zamin: the debt puzzle. Why did debt persist?

During my first weeks of field research in 2012, I attempted to study the process of farm restructuring in the locality. The issue of farm debt (as indebted fields, ‘qarzi zamin’) repeatedly came up in conversations with farmers. The villagers held contradictory opinions on the existence of cotton debt.

I was puzzled. Why, if there had been a nationwide debt write off in 2008–2009, did debt persist here? Or had debt been newly incurred? Who had debt and how high was it, and who had no debt? What was on paper? The international donors who had previously been involved in the debt resettlement programme had completed evaluation reports, and state officials working both in the capital Dushanbe and in the district centre refuted my observations. The local authorities also claimed that they had no part in the debt issue, as I observed in an interview with the secretary of the local government (1 December 2014):

‘Dehqon farmers had five years’, the secretary says. ‘And then?’ I ask. ‘Yagon kas namedonad’ (no one knows), the fields are salinised. Not on the other side of the road, the situation is better there. The situation may improve. Smaller farms have no debt anymore, only the bigger ones. It is their fault if dehqon farmers are still in debt’.

From the top to the bottom of the hierarchy, the authorities repeatedly argued that cotton debt belonged to the past.

According to the collective farm chairman, after five years debt would increase again (interview 29 July 2013), as if debt was ‘frozen’, as Boboyorov (2012, 429) observed in a nearby locality where farmers would only have to start paying it off once their farm production began to return a profit. I understood only over time that debt was written off on paper but not in practice. The strongman had not accepted a loss. Many questions called for the need to analyse the division and implications of farm debt on the ground.

4.3. The allocative right to divide assets and liabilities

Local farm reorganisation had been erratic since the 2000s. As production conditions worsened, the large collective dehqon farm – the successor of the kolkhoz – as well as smaller (collective) dehqon farms accumulated cotton debt.¹⁴ As several villagers recalled, a top-down push for farm reform took place in 2010, when officials from the capital came to issue land certificates ‘on the spot’, particularly with the goal to restructure the large successor of the kolkhoz.

The national authorities, under pressure from international donors, had thought it legitimate to intervene, since the process of farm reorganisation had halted over the years and the collective farm shareholders had not expressed an eagerness to take land out. There was no competition for land, in contrast to some other localities in Tajikistan.

The land committee went from house to house to inform and urge the shareholders: ‘Please take the land, it is empty, start a farm, take your share’ (interview with a farmer 14 September 2013). ‘Please take your share, there is a huge garage and you can also take things from there’ (recalled in an interview with a former brigade head, 16 July 2013) – seemingly ignoring the fact that many assets were long sold or appropriated. It should be noted that the shares existed only on paper. They were attached to the shareholders’ former working unit (brigade), but the plots had not been physically demarcated.

The land committee gave the shareholders three options: first, they could take their share and assemble other shareholders to start a dehqon farm; second, they could keep their share and stay with the collective dehqon farm; or third, they could refuse their share and officially resign. The ceiling of a farm size was set at four hectares, as the authorities thought that people might not have the power to farm more land (interview with the collective dehqon farm's former accountant, 16 July 2013).

With the top-down push of farm reorganisation in 2010, female-headed households had not been inclined to start a farm. Countrywide, though, a reasonable number of women de facto run dehqon farms.¹⁵ Here, the lack of de jure female-owned farms was the result of women’s subordinated position in this conservative setting and the perceived lack of labour power within the household, in the absence of men. When migrating male household members brought in more money than they thought to earn from farming, households were also less inclined to take on individual farming and the private risks that would come along with it.

If people resigned, their land share was added to the governmental ‘reserve fund’, which meant this land became uncultivated. That part of land was no longer under the responsibility of the collective dehqon farm. That being said, while this meant that the total size of the collective dehqon farm shrank, and while most farm assets were long gone, the debt of the large collective dehqon farm was still part of the farm restructuring process. Like land, the single, collective debt was converted into several individual ones: each individual debt corresponded to the size of the new, individual dehqon farm (between USD 750 and 1500 per hectare). In this way, farm reorganisation transformed the collective liability into a private liability.

The speeding-up of the decollectivisation process after 2010 could be regarded as a reflection of the authorities’ successful attempt to individualise farmland, and the chairman’s role in this. It echoes Allina-Pisano’s (2008, 172) observations in the Ukrainian/Russian Black Earth region, where ‘some apparent privatisation success stories depended on directors’ successful manipulation not only of language but also of farm liability’. Through a strategy of ‘recombining property’ (cf. Stark 1996) the chairman was able to appropriate former collectivised farm assets for private use (like the former kolkhoz garage and the good fields near the mountain), while transferring the liabilities to others.

The chairman’s authority was highly disputed among villagers, as he controlled who would get what. Drawing on Verdery’s (2003) concept of allocative rights, we may understand how the chairman gained privileged access to farm assets prior to and shortly after the restructuring. This is similar to the state of affairs during the Soviet regime – when the state held ownership over the means of production – in which a head of a lower level of authority was meant to allocate, control and regulate use-rights, as a social and government agent (as described by Abashin 2017).

The chairman held a privileged position vis-à-vis the strongman through which he could instrumentalise his role. He had become the intermediary between the strongman’s futures company and local farms, as he facilitated contracts between them. His role also involved the responsibility to collect farmers’ tax payments on behalf of the company (interview with a farmer, 1 December 2014). For some small farmers, continuing contracts with the strongman’s company seemed the easiest way to obtain necessary inputs and machinery and advice, which they lacked. At the same time, this meant that relations of dependency were reproduced. The chairman kept his status as ‘rais’ (collective farm chair) in locals’ eyes and discourse, which led to the striking fact that some dehqon farmers appeared not even to know who the actor claiming the debt was, as it all was mediated by the chairman. They termed the company the ‘sponsor’ or ‘f’jucher’.

4.4. From an asset to a means: debt to discipline and bond labour

One might wonder why former shareholders started a farm under these conditions. Debt could be seen as a source of patronage. Having a dehqon farm meant access to irrigated land in the village, and access to land – with or without the pressure to grow cotton – meant access to some part of land for food crops, and perhaps future prosperity. In other settings, a debt tied to land could also be regarded as an implicit price for land. However, the severe constraints on production and the lack of fertile land in this setting refute such thoughts. Nonetheless, several villagers decided to start their own dehqon farm. This was because they wished to have a landed base in their village, as a fallback option, also since off-farm employment opportunities were limited due to the lack of industries nearby.

None of the farm households, however, could rely on revenues from farming alone. They had highly diversified sources of livelihood. Among these, migrant remittances were critical in providing for family and farm needs.¹⁶ Capital accumulation from farming (i.e. from below) in this setting, even by dehqon farmers without debt, was exceptional. In this regard this locality is not unique in Tajikistan, since individual farming in many other parts of the country cannot provide a meaningful income.

The debt was not corrected for inflation or adjusted with interest, and the payments had not been collected after 2010, but land use by the smaller dehqon farms had remained strictly monitored by the local authorities and the collective dehqon farm chairman. Farmers were required to report their cultivation details and were investigated thereupon by authorities, and once in a while the ‘revizor’ (the state’s inspector) passed by.

They [the futures company] have all our personal information. … There is a penja [interest-based penalty] [for the debt]. … People have five years to pay it off, but they [the futures company] did not come for two years in a row now. We’re all in debt. (interview with a farmer, 1 December 2014)

These controls suggest that the actual value of the debt was unimportant. The officially nullified debt had become an extra-economic means to bond farmers’ labour and discipline farm practices; indebted fields tied people to the land, and through the interlocking pressure and control from local authorities, the debt secured cotton production. The futures contract scheme through which debt had emerged had long been abolished, but the ‘politics of contract farming’ (cf. Clapp’s 1994; Burawoy 1985) had survived.

The formal requirement was that 20 per cent of debt should be paid off per year, in cash or in cotton. Pay off in kind was more realistic for cash-strapped farm households, which the strongman was sure to know. It created another lever to push farmers to sell cotton to the strongman’s company. At the same time, the in-kind payment allowed the strongman to manipulate the severity and duration of indebtedness, namely through suppressing farm gate prices of cotton.

Yet one may question whether the debt was the single factor to coerce dehqon farmers to grow cotton. The Tajik saying that ‘cotton is the country’s wealth’ (‘Pakhta boygarii davlat ast’) could be understood as a justification for the state’s pressure to grow the crop. With a lowered tax for cotton cultivation (compared to other crops), the government has long continued to encourage cotton production. Beyond these incentives there are other factors that may ultimately lead farmers to grow cotton, and for rural households to be involved in cotton production. First, transport facilities and marketing options for large quantities of perishable crops have been limited. Second, the legacy of Soviet agriculture plays a role, where farmers sometimes lack the knowledge needed to produce other crops. Cotton cultivation and concomitant labour mobilisation have been shaping the structures of daily village life for several decades. Third, their access to credit and inputs has, for a long time, been limited to advances for cotton cultivation (with which credit was highly politicised); and, finally, cotton stalks, as a source of fuel for cooking, continue to be essential in the countryside in the absence of gas and reliable electricity supplies. There are few alternatives to these by-products.

However, diversification of cropping patterns has clearly taken place in areas where debt peonage has disappeared: on fields of the few debtless dehqon farmers in this locality, and also in neighbouring settlements.¹⁷ Profits from crops other than cotton are significantly higher but restrictions in market outlets and governmental pressure, like here, severely limit farmers in exploiting the potential of farming. Especially such local divergence typifies the highly localised nature of the agrarian political economy in Tajikistan and its implications for farming patterns and rural livelihoods.

4.5. The elasticity of debt

As noted, villagers and dehqon farmers held contradictory perceptions of the implications of being in debt. The ones who started in 2010 had never received an official document stating the size of their individual debt, but had been required to sign documents with which they had accepted debt.

The farmers who were aware of the earlier debt write off were hoping for a second (country-wide) bailout, or for the debt to wither away one day. In this way, even if the debt was real, farmers anticipated support, as had been the case in earlier years. The individualisation of farm assets and debt had been fraught with ambiguity, having been shaped by local power plays between elites, authorities and farmers. In contrast to what (neo-) classical economists (De Soto 2000; Csaki, Lerman, and Sotnikov 2001) might have expected from a transformation away from socialism, it was non-economic rather than financial capital that has become decisive in building a farm livelihood. According to the former kolkhoz accountant (interview 16 July 2013), some powerful, well-connected people had been able to bargain debt down to zero when they started their individual farms; the debt attached to the fields they acquired would stay with the former collective dehqon farm (whose indebtedness, as a consequence, increased). As a result, it was only a few farmers who fared well: the ones who had separated from the large collective in the 1990s thanks to their political and social capital, as was reflected in earlier years in their privileged position on the kolkhoz in Soviet years, like one prospering family with three of five sons individually farming. For outsider, well-connected actors who were willing to farm, debt could be waived – such as for the district governor, but also for the collective dehqon farm chairman. As local strongmen, these two farmed former collective dehqon farms’ fields in the area with relatively good conditions. ‘If you cultivate gold there, you will harvest gold. If you cultivate gold here, you will harvest sand’, (interview with the collective farm chairman, 29 July 2013).

To this end, and building on Verdery’s (1996) notion of land elasticity and Amelina’s (2002) elastic soft budget constraints, I term this debt ‘elastic’. Where collective farmland shrank and stretched in the decollectivisation process in Verdery’s Romanian context, debt was pliable and perceived in different ways, and some indebted farmers appeared more troubled than others, regardless of their amount of debt.

By bargaining over payments, debt and yield levels, some farmers could accrue benefits to lessen hardship and could offset the labour spent on cotton (somewhat). Debt then appeared less hard or harmful (i.e. more advantageous) for the farmers with political connections and social capital, since they could slightly ignore pressure by authorities, like the farming sons mentioned before. The farmers with less political and social capital had limited options, as the elastic nature of the debt made appealing to courts complicated, even beyond the high political stakes, social inequality and skewed knowledge on legal issues that made taking legal steps unrealistic in the first place. One farm household managing a collective dehqon farm with fields at the edge of the village was severely indebted. The family members often expressed their worries about ‘spontaneous’ visits by authorities. Local authorities, in turn, made cynical comments about this farm household and harassed the family by demanding informal payments once in a while (observed during a visit to local authorities 3 August 2013). Elasticity resulted in uneven control over land, and the insecurity of land tenure rendered the farmers highly dependent. In all cases debt tied local farmers to land, which meant it prevented a rural exodus or a diversification of cropping patterns through which the local base could become a meaningful source of livelihood. The debt seemed nearly unresolvable, which in fact challenges the label of ‘debt’ (in the way defined by Graeber 2011).

4.6. Debt and everyday resistance

Debt was a burden in that it implied that declared profits would be siphoned off. Paying off might seem appealing, as it would ‘untie’ the farmer from the land, with more autonomy as a result. Nevertheless, farmers perceived this as almost impossible and preferred to hide small profits and underreport harvests; debt encouraged them to practise creative accounting. Some farmers sought to secretly sell cotton at a (greater) distance with the aim to hide earnings, but the majority of farmers sold their cotton at one outlet.¹⁸ So, while farming did display capitalist features – such as profit-oriented calculus and a reliance on wage labour – these features were not related to the cotton economy. This refutes Gerber’s (2014, 736) statement that ‘indebtedness shapes capitalist rationality’. In fact, such a corollary was prevented here by the elasticity of debt.

Alternative forms of land tenure also appeared: by leasing land from other village farmers, farmers could avoid production obligations and claims over revenues, while liabilities remained with the formal landowner – like a chairman of a small collective dehqon farm who registered an individual dehqon farm in his private name, in order to obtain a bank loan.¹⁹ Others tried to convince authorities that cotton seeds had failed to germinate in order to obtain permission for alternative crop cultivation.²⁰ Farmers thus looked for strategies to benefit from farming, and through specific ways one was able to diversify cropping patterns more than another.

The farmers’ covert strategies had minimal impact on the relations of power and hegemony, as they did not directly challenge structural conditions. Farmers’ strategies, like their arguments over failed germination of cotton seeds, and their secretive sales of harvested cotton, were rather ‘everyday forms of peasant resistance’ (cf. Scott 1986; Kerkvliet 2009) than actions that would bring meaningful change.²¹

4.7. The wider consequences: debt, farm reorganisation and land abandonment

Regardless of the pervasive nature of debt, having indebted farmland was not a source of shame, as the entire farm community was in debt. As one farmer put it, ‘it is not my debt, but my fields that are indebted’ (interview, 6 December 2014).

Even so, debt had severe implications for farm investment and farm development. For instance, some farmers had considered expanding their farm holdings, but withdrew their applications once they realised the significant debt that would come along with the land-use certificates. This clearly exemplifies the consequences of debt for farm reorganisation (as also noted by Ruben and Lerman 2005 in Nicaragua).²²

In particular, the remaining collective dehqon farms struggled with increasingly uncultivated lands. Collective dehqon farm shareholders were unmotivated to undertake barely remunerated work. The district authorities actively sought to find new uses for the fields. They allocated fields for rice cultivation to governmental institutions, and in 2011 they granted fields to a Chinese investor, legitimising the deal with the argument that the abandoned fields exemplified people’s lack of willingness and ability to farm. ‘When applying for land at the hukumat (district authorities) they told us “davai, davai [come on, come on], take the debt; if you do not take them, then don't farm the fields”’ (interview with a farmer, 20 August 2013). ‘The jamoat [local authorities] said, “if no one takes the lands, we will give them to the Afghan or the Chinese”’ (interview with a former kolkhoz member, 20 July 2012). The villagers were particularly anxious about the Chinese land investment since the Chinese investor had been given land without debt, which once again shows the elastic nature of the debt. Yet it has not been because of lack of entrepreneurship or farming skills that land has been abandoned, as the quoted villager added: ‘If all the lands were without debt, we could also take it!’ Rather, post-Soviet hierarchies played an important role in land use and land abandonment. Power configurations defined who would have effective control over land and production revenues, with clear implications for land use.

5. Conclusion

With Tajikistan’s independence and the ensuing civil war, kolkhoz land and assets became conflicted property. By investigating the transformation and appropriation of debt in this setting, I aimed to contribute to post-socialist studies, as well as to studies on agrarian debt in general (cf. Gerber 2014; McMichael 2013).

First, this paper sheds light on the way in which the specifics of farm accounting depend on the political economy, rather than on the economic regime (Amelina 2002; Visser and Kalb 2010). I showed that Tajik cotton debt originated from the Soviet legacy in the cotton sector, which partially retained its soft budget constraints – as was the case in other post-socialist settings (cf. Amelina 2002; Shtaltovna, Van Assche, and Hornidge 2012; Burawoy 2001). In Tajikistan, this enabled loss-making cotton farms to survive, while also safeguarding cotton revenues. Both farmers and futures companies were aware of the impossibility of going bankrupt.

As this paper demonstrated, debt has been redefined and transformed from its Soviet understanding. This case exemplifies that even though debt may feature in different societies across time and space, implications and perceptions differ. Particularly after the implementation of the donor-driven debt resettlement scheme, debt became a more concealed ‘mechanism of capital, land and labour control’ (Gerber 2014, 732). The strongman used liabilities to generate private wealth, in order to seize the generated surplus (in fact a mode of primitive accumulation as described by Marx 1894/2010). There was never a political willingness to write off debt.

Second, the pliability of debt bears a resemblance to the varying nature of budget constraints as described by Amelina (2002), and Verdery’s (1999) ‘fuzzy rights’ and responsibilities over land. If we see debt as negative property, ‘fuzziness, then, will lie precisely in the lack of routinized rules and crystallised practices around private property in the context of postsocialism, as well as in the constraints on exercising bundles of power’ (Verdery 1999, 55). The debt appeared ‘fictitious’ (cf. Van Atta 2009; Kassam 2011), but I contend that it can best be qualified as ‘elastic’. This elasticity typifies the local agrarian political economy in which local power configurations determine the hardness of the debt. Whereas a debt imposed on one farmer was ‘hard’ and restrictive, for another a debt seemed less constraining. The fact that the Chinese investor acquired fields without debt in 2011, as if debt had suddenly disappeared, points once more at this elasticity.

Whereas analogies of debt exist in other post-socialist localities (cf. Allina-Pisano 2008; Shtaltovna, Van Assche, and Hornidge 2012) the political economy of debt described in this paper is unique, embedded in post-Soviet and local post-conflict power configurations. At the same time, beyond its Soviet and civil war legacies, cotton debt in Tajikistan shares characteristics with agrarian debt in other regions. The implications of farm debt on reorganisation have also been observed in other settings (cf. Ruben and Lerman 2005 in Nicaragua, and Allina-Pisano 2008 in the Black Earth region), while the emergence of farm debt in Tajikistan, as a consequence of asymmetric power relations in contract farming, reflects more general characteristics of debt in contract farming as observed in other contexts by McMichael (2013), Byres (2016), and Guérin (2013).

Finally, by examining the local trajectory of post-socialist agrarian change through the prism of debt, this paper sheds light on the consequences of the political economy of cotton for the fate of the former Soviet farm worker. Examining the context only from a narrow property rights point of view could be misleading. These Tajik farmers may hold only one component of the typical bundle of rights, and even that might have a peculiar form. Specifically, the right to (individually) cultivate land brings with it the obligation to do so with cotton. By unpacking the bundle of rights, one can grasp the variation within it (cf. von Benda-Beckmann, von Benda-Beckmann, and Wiber 2006), and thus account for local variations even though legal labels are the same. As von Benda-Beckmann, von Benda-Beckmann, and Wiber (2006, 11) stated, ‘much theory [on property rights] reifies the individual as an actor, granting owners far more agency than they have in real life’. The de jure ‘individual’ Tajik farmer may de facto be less autonomous than envisioned by international donors.

It is therefore debatable whether post-socialism has meant a social and political change, or if it rather is a continuation of institutionalised neo-patrimonial power configurations, in which regime interests filter down to the local level through a chain of command and control. The ‘Road Map’ implemented after 2008 has not created a road to prosperity. Despite donor pressure to resettle and solve the debt, the Tajik state was unable to remove the debt’s structural causes. Freedom to farm has continued to be frustrated. If there has been a road to prosperity, it has been a narrow one for the happy few. What used to be a state revenue stream from the export of cotton has been captured by a small clique of rentiers. In this way, the localities outside the attention of the international donors have come under a debt-based mode of governance. This political economy of debt entails that a rural exodus is prevented, where people otherwise might have left. Debt restricts villagers from exploiting the potential of farming. Forgotten by the state, unobserved by international donors, and left to the vagaries of strongmen, rural dwellers express a strong wish to leave. The power asymmetries over land and farm revenues in these areas of benign neglect inhibit agrarian development, render people governable and depress rural livelihoods.

Acknowledgements

I am grateful for the support provided by the Catharine van Tussenbroekfonds, the Land Deal Politics Initiative (LDPI) and the Leiden University-based Leiden Asia Centre (LAC) for enabling field research in Tajikistan. I especially thank my supervisors, Oane Visser and Frank Pieke, as well as Don Van Atta, Hafiz Boboyorov and Max Spoor, for sharing insights and providing valuable comments on earlier versions of this paper.

Disclosure statement

No potential conflict of interest was reported by the author.

ORCID

Irna Hofman http://orcid.org/0000-0003-3141-0638

Additional information

Notes on contributors

Irna Hofman

Irna Hofman is a PhD candidate and part-time lecturer at Leiden University Institute for Area Studies. Her doctoral research focuses on post-Soviet agrarian transformation in Tajikistan, and also includes the study of Chinese investments in Central Asian agriculture. She is a rural sociologist by training, and her work and interests are focused on rural livelihoods, agrarian and social change, and transition economies. She was a temporary research fellow (2014) at the Leibniz Institute of Agricultural Development in Transition Economies (IAMO, Halle/S in Germany).

Notes

1 In this paper, I use the term ‘strongmen’ rather than ‘cotton elites’ (cf. Boboyorov 2012, 413) to refer to elites tied to the ruling regime by kinship, regional identity or marriage. In my view the term ‘strongmen’ better denotes the significant unequal power relationship between strongmen and farmers. It is also used to indicate that the economic base of such strongmen goes beyond the cotton sector as having high stakes in other sectors of the economy. 

2 Legislative information has been cross-checked with the legal database managed by the Tajik government Adlia (www.adlia.tj).

3 This was different outside the state-regulated economy, where people could accumulate capital through their labour.

4 For a discussion and interpretation of Polanyi's view on socialist accounting, in which maximum productivity and social justice prevail, see Bockman, Fischer, and Woodruff (2016).

5 The high inflation of the early 1990s meant that debt inherited from the Soviet period quickly decreased (Csaki, Lerman, and Sotnikov 2001). For this reason, ‘farm debt’ in the remainder of this paper refers to debt incurred in the post-Soviet period.

6 The first law on dehqon farming was promulgated in 1992, after which former Soviet farm shareholders could start a dehqon farm on state or collective farmland. Since the mid 2000s all reorganised farms became known as dehqon farms, the term used for the commercial post-Soviet farm. Three types of dehqon farms exist: collective, family, or private dehqon farms; the difference depends on the management and ownership structure. Most large-scale Soviet farms were initially transformed into collective dehqon farms, which has been followed by more gradual individualisation in later years. Nowadays, large farms with large numbers of shareholders are frequently collective dehqon farms. Otherwise, differences between the types of dehqon farms are often not directly observable on the ground. Where I may incidentally use only the term ‘farmer’, this refers to the dehqon farmer; and this may refer to both women and men. The main way to start a farm is to apply for a land-use certificate with the district authorities. Land-use rights are generally granted for inheritable use, while the state has retained ownership over land. 

7 A collective (dehqon) farm (a legal person) cannot transfer a debt to an individual or family dehqon farm (which at that time was an entity without the status of legal person and not the legal successor of the collective farm).

8 Farm debt for government-supplied utilities has always been less severe, and has been written off by the government more frequently.

9 The quote at the beginning of the paper, from an employee of an international donor organisation working in the northern region of the country, sheds light on the fact that landed property is more dominated by strongmen in the southwestern region (see also Hofman and Visser forthcoming).

10 The same person was later appointed the deputy prime minister in charge of the national process of agrarian reform (see also Van Atta 2009), and in 2015 appointed director of Agroinvestbank, even as he maintained his high stakes in the agrarian economy. In his different positions, he has been able to pursue his private interests in the agrarian economy.

11 Observations and interviews were noted in a field diary, later analysed using computer software (Atlas.ti).

12 The district's name was changed to Jaloliddini Rumi in June 2007, and changed again in February 2016, to Jaloliddini Balkhi.

13 Besides the cotton economy, this strongman has vested interests in the financial and industrial sectors.

14 The main collective dehqon farm, as successor of the kolkhoz, gradually decreased in size over the years (TajStat 2013, 40). Between 2009 and 2010 it decreased by more than 40 per cent, reflecting the top-down push described, after which smaller farm units were established. Notably, only farmers with smaller farms, who had started in the 1990s when farm debt was lower or non-existent, had been able to pay off debt over time.

15 I thank the reviewers for raising the issue of gender. Farming has feminised in the past decades as a result of high male labour migration, and beyond that farm work in Tajikistan is highly gendered, where women primarily undertake all the manual farm work and men carry out the mechanised practices. In the absence of men, women have to undertake both farm work and reproductive work.

16 Labour migration is an essential source of livelihood throughout the country. In 2013, remittances equalled over 50 per cent of the country's gross domestic product (GDP), with which Tajikistan ranked highest worldwide (World Bank 2016).

17 A comparison based on (district) statistics is not possible, given the fact that there are clear intra-district differences. My argument is based here on personal observations of crop diversification; Nekbakhtshoev (2016) also noted a relatively more diversified cropping pattern in the neighbouring district. 

18 Whereas the number of cotton ginneries has increased in recent years, it is questioned whether these really compete. A large number of ginneries in the districts in southwestern Tajikistan are controlled by the strongman, and sales of cotton seed is also restricted (see also Boboyorov 2012, 424).

19 Whereas farmers have faced difficulties to obtain short-term credits, banks in general have refused to finance indebted farmers.

20 If farmers can convince authorities that resowing has failed, they may cultivate other (food) crops which can be consumed or sold freely, for instance rice; cultivation of this crop is regarded as beneficial to decrease soil salinity. In this regard, there is a competition between crops, resembling the situation Veldwisch and Spoor (2008) observed in Uzbekistan. 

21 As Scott (1986, 30) explained, ‘peasants’ everyday resistance’ may result in structural change, but not at all times. Peasants ‘may marginally alleviate exploitation … they may change the course of subsequent development, and they may more rarely help bring the system down. These, however, are possible consequences. Their intention, by contrast, is nearly always survival and persistence’.

22 Nekbakhtshoev (2016) observed elsewhere that collective dehqon farm chairmen instrumentalised debt to discourage farm fragmentation, by demanding repayment from shareholders once they wanted to separate. This was explicitly not at stake here.