Agriculturalizing finance? Data assemblages and derivatives markets in small-town New Zealand

Abstract

The financialization of agriculture appears to be proceeding apace. In New Zealand, the creation of a futures market for dairy lends weight to this story. What is less well understood about the process of financialization in agriculture, however, is how exactly it is proceeding. This paper focuses on NZXAgri, an offshoot of the New Zealand sharemarket operator NZX, which is tasked with the creation of the dairy derivatives market, and on the data infrastructure that is being assembled to underpin this trading space. The making of NZXAgri has been a complex process, resulting from the dissipation of a previous agriculture data assemblage during neoliberalization, and now with multiple political and economic projects partially aligned under its banner. Meanwhile, the emerging data assemblage relies on all manner of material and immaterial relational work to produce the necessary dairy production information for consumption by international financial actors. It is this kind of assembling work that is shaping the financialization of agriculture, and it requires constant negotiation with the diverse agencies of farmers and their rural contexts. This suggests that we are seeing the agriculturalization of finance alongside the financialization of agriculture.

Keywords

Financialization agriculture data assemblage financial markets derivatives

Introduction

In 2010, NZX, the New Zealand stock exchange operator, set up a futures market for whole milk powder, followed by markets for skim milk powder, butter, and milk. For the first time, financiers, corporations, and private citizens could participate in New Zealand’s dairy commodity market without having to handle the commodity itself, while farmers would have the opportunity to manage their risk, or if they so desired, vastly increase it in the hope of larger returns. From one point of view, the making of a futures market for dairy in New Zealand represents a pivotal moment in the financialization of the country’s agricultural sector, and given the role of the large New Zealand dairy farmer collective Fonterra in the global dairy trade, the ongoing financialization of agriculture globally (Bruno et al., 2017; Clapp et al., 2017; Gertel and Sippel, 2016). But one of the surprising things about the making of this market is not so much that it is happening, but where it is happening. The small town of Feilding (pop. 16,500) in the Manawatu district of New Zealand’s North Island is about as far from the glass towers and fast lifestyles that we associate with finance as we might imagine, and yet it has a vital role in the making of this market. Feilding is not an international financial center, or a national financial center, or even a regional financial center (nearby Palmerston North would claim that title). Such scalar hierarchies are of little use for understanding Feilding’s role in finance. But understanding the role of places like Feilding in the making of financial markets, and the contingent historical circumstances that led to it, can reveal much about the kind of work required to make the financialization of agriculture happen.

The narrative in which financialization represents a late stage in capital accumulation (Lapavitsas, 2013) and for whom agriculture is its latest target is increasingly questioned (Ouma, 2014, 2016; Williams, 2014). Not least because of the long history of entanglements between finance and agriculture going back decades, if not centuries (Gertel and Sippel, 2016; Martin and Clapp, 2015). A key concern with the conventional financialization narrative is that it has little to say about how finance works in practice or how financial markets are actually achieved (MacKenzie and Millo, 2003; MacKenzie et al., 2007; Ouma, 2016), and so its grasp of the consequences of financialization for particular geographical settings is relatively limited (French et al., 2011). The geography of finance is necessarily diverse as it seeks to enrol historically and geographically specific agencies and situations, requiring various combinations of systematic and improvised strategies of alignment and visibility (Christophers, 2011; French et al., 2011; Williams, 2014). Understanding what this looks like requires shifting our focus from ‘financialisation at work’ to ‘financialisation as work’ (Williams, 2014: 410) and considering how different financializing sites, such as New Zealand dairying, are precarious socio-technical assemblages stitched together to give life to neoliberal, technopolitical projects that focus on remaking relationships between capital, information, and agriculture. Our approach in thinking about financialization as work can be characterized as a deliberate strategy of using ‘weak theory’ that focuses on unpacking the work of making financial markets rather than developing ontological claims as to what financial markets are (Carolan, 2016; Gibson-Graham, 2014). Consequently, the paper eschews claims that the relationships developing between agriculture and finance have inherent characteristics. Rather what we argue is that while these relationships have powerful effects, they are contingent and contextual rather than universal.

Recent interventions by Ouma (2016) and Williams (2014) on financializing work in agriculture point to two aspects of this process that demand more attention, and reflect the need to treat a priori claims about the necessary relationship between finance and agriculture with care. One: farmers. This group is often represented as little more than victims of the ever-expanding reach of finance capital. They also tend to be treated as a singular group, rather than a category cut across with social and geographical divisions and stratifications (Williams, 2014). Understanding how and why farmers are enrolled, and the agency they have in constructing financial assemblages, is necessary for properly understanding financializing work in agriculture. Two: rural places. Most stories of financialization are stories of ever more powerful international financial centers, or of ‘excluded’ spaces where access to finance is difficult or only available through ‘subprime’ avenues. Ouma (2016) makes the point that grasping the incorporation of agriculture into finance means getting beyond the office towers of international finance centers and into the places where the necessary connections with agriculture can be made. Williams (2014: 423) suggests that ‘ag space’, a term used by financiers when they talk about investing in agriculture, usefully ‘captures the extent to which ‘agriculture’ actually represents a distinct social and institutional space, a particular field of engagement, visualization, and intervention that exists somewhere between finance and farming, sharing features of both while being irreducible to either one’. Smaller places many of us have never heard of, such as Feilding, are a part of this space, and they play vital roles. These two points also ask us to see the relationship between finance and agriculture in mutually constitutive terms. While we acknowledge the ways in which financial interests in agriculture have changed the latter, we also argue that the reverse relationship exists, insofar as financial actors are being reshaped through their relationships with agriculture.

We add to this another aspect that deserves greater attention: the role of data. The rise of ‘big data’ is already a much storied phenomenon (Mayer-Schönberger and Cukier, 2014), with critics pointing to its implications for industry, automation, surveillance, privacy, knowledge production, and capitalism more generally (Boyd and Crawford, 2012; Dalton and Thatcher, 2014; Graham and Shelton, 2013; Kitchin, 2014). ‘Big data’ has been promoted as the next big thing in agriculture, partially because of the increasing amount of information being collected about production and environmental conditions under the auspices of precision agriculture (Carolan, 2016). Curiously, however, despite the simultaneous growth in critical data studies, and the growing importance of agricultural ‘big data’, the relationships between the two have not been systematically unpacked by agrifood scholars (Bronson and Knezevic, 2016). Yet, as Bronson and Knezevic (2016) argue, now is the time to critically interrogate agricultural data relationships because the socio-technical assemblages that create and circulate data are still in flux, as are the projects that seek to use emergent data assemblages to further particular political goals. It is this fusing of data assemblages, technical expertise, and knowledge practices to ‘constitute, embody, or enact political goals’ that makes these developing agrifood relationships technopolitical (Hecht, 2009: 15). Agrifood data assemblages simultaneously represent a means of generating new insights into opaque processes and flows, and a structuring power that increasingly redefines what knowledge is and how it can be used to embody and enact the conditions of political possibility (Folkers, 2017; Freidberg, 2014; Hecht, 2011; Mitchell, 2002). The technopolitics of data is not new. The existing literature argues that the ability to measure and collect data about agriculture has been central to processes of state-formation and colonization, and in turn, this has contributed to the rendering of agriculture visible in ways that are useful to finance capital (Gertel and Sippel, 2016). Moreover, the availability of information is central to the making of financial markets, and in particular futures markets, where a lack of information generally precludes the participation of actors in the market. Consequently, if we are to critically grasp the evolving nature of an ‘ag space’, we need to examine the constitutive, mutually dependent relationships between financial actors and data, and the contextual circumstances in which they take shape.

This paper examines the New Zealand dairy futures market as a market in the making. The evidence for this paper comes from a combination of semi-structured interviews with informants working in the agri-financial sector and the wider agricultural industry, archival research, the examination of government policy, and extensive examination of the reportage of agriculture and finance in the New Zealand media. The interpretation of this material, especially the interviews, was workshopped by the authors and the wider Making Markets Collective based at Massey University. The paper argues that central to the making of a dairy future market is the construction of an appropriate data assemblage that links farmers, analysts, companies, traders, and investors. Data, like money, is neither weightless nor frictionless. Both stutter from place to place, organization to organization, sticking and congealing in some relationships, while sliding off others. As money and data flow they bear down on relationships, shaping how they are imagined and practiced, creating new forms of agency, redistributing existing ones, and transforming subjectivities. This paper analyzes how the data assemblage that has been stitched together has been partially framed by the pursuit of different, but intersecting, technopolitical projects by commercial and governmental actors. It examines how the weight and friction of financialized agricultural data is quietly transforming relations in New Zealand agriculture, and in the process producing novel forms of agency and subjectivity.

The next section provides a critical review of existing literature on the financialization of agriculture. It then makes a case for linking this to an emerging literature on data assemblages as a way of grasping financializing work. The remainder of the paper follows the NZX owned financial intelligence company NZXAgri and its involvement in the development of a data assemblage intended to support the operation of the dairy futures market. We argue that this emerging agri-financial data assemblage represents a particular form of financializing work that brings together of an array of data devices and practices as well as constructing new farmer subjectivities around risk and future expectations.

Financializing agriculture

Financialization, risk, and subjectivity

Agriculture, and by extension food, has long been shaped by financial concerns (Gertel and Sippel, 2016; Martin and Clapp, 2015). As Cronon (1991) shows, the development of the corn fields of the American Midwest went hand in hand with the building of the elevators that made corn markets synonymous with both financial innovation and chicanery in the 19th century. However, recently scholars have argued that the relationship between agriculture and finance has begun to shift in distinctive ways (Burch and Lawrence, 2009). This includes the growth of instruments such as derivatives, direct investment by financial institutions in commodities, the development of finance divisions in agrifood companies, the expansion of farmland investment funds, and investment in agricultural equipment, technology, input, and information businesses. In significant part these new relationships have been designed to allow large institutional investors to gain exposure to different commodity sectors. A key theme of this financial work has been an emphasis on the security and stability of agricultural investment returns vis-à-vis other investment categories following the 2008 global financial crisis (Ouma, 2016; Williams, 2014). The result has been the remaking of agriculture into a financial investment space.

Social science scholars have developed a number of lines of critique in response to the evolving relationship between agriculture and finance (Baines, 2017; Brooks, 2016; Bruno et al., 2017; Burch and Lawrence, 2009, 2013; Clapp et al., 2017; Field, 2016; Gertel and Sippel, 2016; Isakson, 2014). This work has exposed the increasing scale of financial investment in agriculture and the evermore central role of financial actors in the food system. In turn, specific attention to agricultural financialization has been connected to wider debates about the role of finance in the global economic system, and the sense of an epochal shift in the trajectory of capitalism (Christophers, 2015; French et al., 2011; Lapavitsas, 2013; Leyshon and Thrift, 2007).

One conventional line of critique is to argue that finance and agriculture represent ontologically and spatially separate spheres brought together in an ‘unnatural coupling’ that disrupts the normal dynamics of agricultural commodity markets, leading to speculative bubbles and food price volatility (Ghosh, 2010; Russi, 2013). A second line of critique is rooted in the political economy approach to ‘food regimes’ (Friedmann and McMichael, 1989). Here the emergence of a ‘third food regime’ characterized by the deep entanglement of finance and agriculture is seen as representing the outcome of wider changes to the spatio-structural dynamics of global capitalism (Burch and Lawrence, 2009). Collectively, these critiques have given scholars an analytical grasp of the dynamics of finance/agriculture relationships and a strong sense of their global spatio-structural dimensions that has brought important questions of food security and food sovereignty to the fore. However, they largely fail to provide adequately nuanced accounts of the complex spatio-historical relationships between finance and agriculture, the contested forms of imagination and practice involved, and the evolving agencies of different interrelated actors (Ouma, 2014, 2016; Williams, 2014). A point demonstrated by Gill (2016) in his examination of the failure of a particular agricultural investment in Gambella, Ethiopia because of an inability to take account of the particular socio-ecological dynamics of the Baro river ecosystem. Too often the presentation of financial capital as a monolithic steamroller passes over the complex, contested and contingent work of actually assembling financial projects and creating agricultural investment spaces.

The approach of this paper is rooted in the dialogue between the geography and cultural economies literatures on markets, technicality, data and assemblage (Bennett et al., 2008; Berndt and Boeckler, 2009; du Gay and Pryke, 2002; Hall, 2011). In particular, it emphasizes the importance of approaching financialization as a form of work (Williams, 2014), and of understanding the constitutive role of particular geographies in the creation and maintenance of financial markets and projects (Christophers, 2014a, 2014b; Hall, 2011). Borrowing from Çalışkan and Callon (2009, 2010) we argue that financialization as work involves the conceptualization of financial agency, the arrangement of technologies and institutions, and the materiality of those things that are going to be translated into financial instruments. The performative dimensions of economic theories, and specific financial equations, has been well documented (Callon, 2007; MacKenzie, 2006; MacKenzie and Millo, 2003; MacKenzie et al., 2007). However, within this framework there is significant scope to pay attention to the multiple scales and sites of theorization enacted ‘in the wild’. So rather than understanding financialization as involving the diffusion of a core logic, we see it as being built from dispersed, contingent, and reflexive responses to particular circumstances undertaken by a range of actors. The results are unstable, ad hoc conceptualizations that set about attributing and distributing agency in uncertain ways. This uncertainty is amplified by the reality that in order to do work in the world, financial conceptualizations require the assemblage of infrastructures, devices, and expertise in the context of already existing assemblages with a weight and momentum of their own.

The result is that we need to analyze ‘the technicality of financial markets; the role played in these markets by technologies and by systematic forms of knowledge; the concrete, material practices of trading, risk management and regulation’ (Beunza et al., 2006: 721). Understanding financialization as technical and material work leads to a concomitant realization of the unstable, independent materialities of those objects being placed in financial relationships. The organization of the production, processing, and consumption of agricultural commodities has always been shaped by the materialities of those commodities, such as variations in quality, perishability and bulkiness, the rhythms of seasonality, breeding and technological change, and the restlessness of consumer preferences (Henry, 2017; Henry and Roche, 2013). Material recalcitrance builds into financialization an uncertainty that must be constantly conceptualized and rendered into the technicality of financial work. This uncertainty, suggests Sassen (2012), means that ‘incomplete knowledge’ is a systemic limitation that presents an ongoing challenge to the aspirations of financial actors and which leads them to engage in ‘various forms of translation and enrolment and to create new ways of rendering agriculture visible to finance and novel strategies for calculating and performing value’ (Williams, 2014: 409).

The assemblage of contingent, uncertain financial assemblages requiring constant maintenance work does not occur on the head of a pin (Christophers, 2014a; French et al., 2011). Rather the performance of legibility and calculative practices identified by Williams (2014) as central to financialization rest upon the entangled making and remaking of financial spaces and subjects. This involves the transformation of individual places, such as farms, to make them legible. Such work is not new. Both Hacking (1990) and Porter (1986) in their seminal works on the co-development of state power and statistics argue that agriculture was an early focus for the collection of statistical data. In this vein, Murdoch and Ward (1997) show how agriculture in the United Kingdom was gradually brought into existence as an economic sector, able to be subject to governmental calculation, through the development of systems to collect agricultural statistics from farms. Statistical representation enabled governments to create systems of classification for farms and farmers which provided the analytical framework for future interventions around grants and subsidies, technology transfer, and ideas of best farming practice. Significantly, argues Murdoch and Ward (1997), an effect of statistical representation was the ontological constitution of farms as businesses, and the stabilization of ideas of who, or who was not, a farmer. The constitution of a new space – the farm as business – went hand in hand with the creation of a new subject – the professional farmer – both of which could become objects of governmental concern. The techniques by which ‘ag space’ is constituted have long histories and diverse geographies, and they highlight the uneven but nonetheless intricate entanglement of agricultural subjects with financial networks (Hall, 2012).

Data assemblages

The story that Murdoch tells us points to the role of statistics in shaping how the work of governing has been conceived and achieved for over two centuries, and brings us to the second theme of this paper. Statistics are not something simply produced by the state to make their job easier: they define the very grounds, objects, and tools of governmental power (Desrosieres, 1998; Hacking, 1990; Porter, 1995). While the state produces statistics, the opposite is also the case. Recent decades, however, have seen a proliferation of data, produced by actors we would not normally think of as part of the state. This proliferation, and the emblematic notion of ‘big data’, has been made possible by advances in computing power and digital storage infrastructures (Graham and Shelton, 2013; Kitchin, 2014; Mayer-Schönberger and Cukier, 2014). It is enabling new ways of seeing, calculating, and governing by diverse entities within and beyond the state (Amoore and Poitukh, 2016; Wyly, 2014).

The conventional story of ‘big data’ is that there is now so much data being constantly produced from smartphones, appliances, traffic cameras, social media, weather sensors, credit cards, farm equipment, and so on, all stored in cloud-based infrastructures, that it is changing everything about how we live. As more of the world is ‘datafied’ into a form that makes it analyzable with the right algorithm, so we can learn and act with greater certainty (Mayer-Schönberger and Cukier, 2014). More critical accounts of ‘big data’ point out that the hype ignores the history of previous data-driven revolutions that have not produced the results promised (Barnes and Wilson, 2014). Meanwhile, the growth and centrality of big data to decision-making processes mean that their powers of prediction are really just a function of the fact they increasingly shape how the world works (Shearmur, 2015) while those that sit outside of data collection systems are uncounted and ghettoized (Graham and Shelton, 2013). Nevertheless, in this world where data is becoming more valuable as it becomes more ubiquitous, companies that once thought of themselves as simply marketing, accounting, or equipment companies are transforming themselves into data companies. Moreover, even for those companies that do not produce or work with data that qualifies as ‘big’, this new orientation to where value can be produced and captured is changing business practices.

Kitchin (2014) grounds the world of data by pointing out that data exists in ‘data assemblages’ made up of various apparatuses and elements. These include the computers, servers, fiber-optic networks, and scanners that make up the computing infrastructure, the places where these are located and pass through, the people that act on and through it in various capacities, the organizations in which all of this is housed and the various standards, protocols and regulatory systems they have in place, the funding streams, business models, and data markets that finance it, and the various philosophies, modes of thinking, theories, and models that shape and condition the assemblage and the many texts these are inscribed upon (see Kitchin, 2014: 25, for a more comprehensive list). Apart from the complexity and extensiveness of data assemblages, what we can take from this is that data assemblages have to be actively made out of material and immaterial resources and practices (Pickren, 2018). This means that they have a geography and a history (Barnes and Wilson, 2014; Graham and Shelton, 2013). As these assemblages have been assembled and reassembled over time they have gone through moments of transformation that have produced new agencies and effects, both expected and unexpected. After all, it is these assemblages that datafy the world. In so doing, they link the people, things, and places being converted into data in various ways with one another and with the various agencies who would act on, with, or through them. This new social topology can have real consequences for those that are a part of it.

The notion of assemblage also brings us back to the encounter between finance and data, or more precisely, the encounter between the assemblage of financial work and data assemblages. Seeing finance as the fragile effect of an assemblage involving the knitting together of theories, devices, and materials and the making of space and subjects requires us to consider its complex topological relationships. Conventional narratives describe a process of financial network expansion, unfolding from a core and gradually enrolling and transforming actors as they are reached on the network’s margins. The core-periphery spatiality of this imagination of financial networks can be contested by looking carefully at the variegated performances of space found in assemblages. Here topological relations are shaped by the dynamic between stability and fluidity, and between presence and absence (Allen, 2016; Law and Mol, 2001). In prosaic terms this means carefully following the intricate weave of connections and order that do not neatly correspond to our existing maps and their assumptions about the relatedness and ordering of financial relationships. In what follows we do this by tracing the history and geography behind the emergence of a futures market for dairy in New Zealand. It draws on an analysis of historical documents, including NZX’s annual reports, and interviews with workers at NZXAgri, located in Feilding in the Manawatu.

Assembling NZXAgri

In early 1987, North Canterbury farmers Rod McKenzie and his cousin David Meares faced the problem of comparing schedule prices for lamb from different processing companies (Cronshaw, 2008). Under the brisk wind of neoliberal restructuring that was profoundly reshaping state – agricultural relationships in New Zealand, the existing, national schedule of meat prices had been stopped to allow variations in market prices between processors to be, theoretically, more clearly signaled to farmers (Calder and Tyson, 1999). However, with the demise of the commensuration infrastructure provided by a single, national schedule for lamb pricing (Henry, 2017), the competing schedules offered by processing companies treated variables such as killing charges, meat levies and transport costs in different ways, which meant that it became very difficult for individual farmers to accurately compare them. McKenzie and Meares devised a spreadsheet that accounted for these variables and enabled livestock farmers to compare prices between processing companies. By April 1987, a weekly spreadsheet that drew on information from Fortex, PPCS, and Challenge Meats was being faxed to subscribers, and The Press in Christchurch had begun buying the information (de Lacy, 1997). From this starting point the market intelligence service, which become known as Agrifax, rapidly expanded its clients to include newspapers, magazines, radio networks, regional television, and corporates across New Zealand. It also began incorporating market data and analysis of trends for a range of agricultural commodities beyond lamb (de Lacy, 1997). Over the next 20 years Agrifax became the de facto provider of agricultural market data, and increasingly analysis, in New Zealand with a business model that was based on the transformation of freely available data into commensurable forms for users. However, by the mid-2000s, Agrifax was increasingly confronting the unwillingness of some companies to supply data as they began to recognize the commercial value of the production and market data they generated (Barber, 2006; de Lacy, 1997).

In April 2006, Agrifax was sold to the New Zealand share market operator NZX, becoming NZXAgri in the process (NZX, 2006). The purchase of Agrifax by NZX should be seen in the context of the NZX’s own transformation throughout the 2000s. In different ways both Agrifax and NZX were products of New Zealand’s neoliberal transformation during the 1980s and 1990s. Since its establishment in 1870, the New Zealand Stock Exchange (NZSE) had been a mutual association, described as ‘parochial, inward-looking, guild-like, insecure, prone to procrastination, and only superficially democratic’ (Grant, 1997: 409). However, by the 1980s, provoked and enabled by economic restructuring and commercial deregulation, the NZSE had begun to become a more aggressive financial actor, working within a new corporate structure, and developing a much wider range of investment options. Demutualization in 2002, followed by the transformation of NZSE into the publically listed company NZX, saw the threads that had been evolving in the 1990s being woven together into an explicit diversification strategy. Operating with a new CEO, Mark Weldon, this strategy involved expanding the commercial operations of NZX to include information and data services, with its CEO arguing that part of the ‘muscle’ of the economic value generated by the company would come from stimulating demand for market information (NZX, 2005). The 2006 acquisition of Agrifax was the initial part of this project (NZX, 2008, 2009), and was spruiked as the first building block of an agricultural version of the New York based financial and data analytics firm Bloomberg. By 2010, NZX was being described by its CEO as an ‘integrated information, markets and infrastructure company, not merely a ‘stock exchange’’ (NZX, 2010: 6). Whereas international exchanges were tending to focus on equities, Mark Weldon was arguing that NZX had begun concentrating on the tools (data analytics) to dig into the goldmine, and that while ‘we cannot control the supply of gold,… as long as there is enough of it…, the tools are worth selling’ (NZX, 2010: 8). In this context, NZX’s focus on agricultural data through NZXAgri was framed as unique in terms of its focus on agriculture and its level of connection and embeddedness, distinguishing the analysis provided by NZXAgri from purely ‘quant’ based international exchanges. The development of an agricultural data analytics business within NZX went hand in hand with the company’s concomitant strategy of developing new financial products. The emerging interlinkage of data analytics, financial products, and agriculture, particularly the dairy sector, within NZX was clearly articulated by Weldon in 2010 when he argued that as markets grew the demand for information grew as well, identifying that ‘NZX’s long-term position in information is strong, with information sales poised to benefit from dairy futures’ success’ (NZX, 2010: 8).

The genesis of Agrifax in the breaking up the pricing infrastructure that had evolved since World War One was repeated more broadly across the agricultural sectors with the restructuring of the Ministry of Agriculture in 1987 and the shift to ‘user pays’ for previously publically available data (Dunn, 1987), the concomitant restructuring of the Department of Scientific and Industrial Research (DSIR) into commercially focused, competing companies (Galbreath, 1998), the shifting of priorities for statistical collection with the demise of agricultural production surveys in the mid-1990s (Edlin, 1997), and the recognition of the commercial value of agricultural data by agribusinesses (Bishop, 2006). The net result was that by the 2000s, publically available agricultural data was fragmented and had largely disappeared into private agribusiness companies. The resulting lack of public agricultural data led one of our interviewees to observe that some people look at the dairy sector New Zealand ‘… as a bit of a black hole for data’ (Jan, interview). Unlike the United States Department of Agriculture’s extensive data collection and publishing, or similar work in Australia, in New Zealand the dominance of Fonterra meant that, ‘… data just doesn’t exist for New Zealand because it is considered commercially sensitive’ (Jan,¹ interview). This lack of information was framed by the interviewee as an opportunity for NZXAgri because in this context, ‘Any information that we can provide to provide some transparency is certainly valued’ (Jan, interview). Another interviewee developed this point by arguing that the relationship went beyond simply providing transparency into a form of symbiosis that meant ‘the more turnover we can get in a market the more information is generated and the more we can repackage it and sell it back to the audience’ (Paul, interview).

By late 2007, helped by the Agrifax purchase, the majority of NZX’s revenue was coming from selling data rather than more traditional sources such as listing fees (Allen, 2007). Alongside the shift to selling data, NZX developed instruments that would enable a greater scope for investment in New Zealand’s agricultural sector. Historically, in New Zealand the agricultural sector has been characterized by a dearth of publically listed companies that can be directly invested in. Instead, the sector has been dominated by a mix of privately owned companies, many owned by international actors, and farmer-owned cooperatives such as Fonterra. Only to a limited extent have managed funds become interested in New Zealand farmland and processing assets. In this context, NZX identified that the launch of its first set of derivative products in 2010 would represent a fundamental turning point in the trajectory of the company as it sought to create ‘a multi-product derivatives franchise that develops global distinctiveness’ and would enable more investment opportunities in agriculture (NZX, 2009: 17). Dairy was specifically identified as the key sector for the development of this ‘derivatives franchise’.

The focus on the dairy sector reflected the strong relationships that been developing between NZX and Fonterra throughout the period leading up to the introduction of a dairy derivatives market in late 2010. Fonterra was established in 2001 from the merger of the two largest key remaining New Zealand dairy cooperatives (Kiwi and NZ Dairy Group) and the New Zealand Dairy Board (Lind, 2013). At board level a number of NZX directors had strong connections to the dairy industry. Henry van der Hayden (NZX board member 2005-8) was a founding director of Fonterra, and its Chair between 2002 and 2012, while another director, Chris Moller, had been the New Zealand Dairy Board’s CFO, and Fonterra’s Deputy Chief Executive. The shift in NZX’s strategy towards developing instruments for agricultural investment coincided with a major review by Fonterra of its capital structure in 2007 intended to solve problems around the cooperative’s access to capital, and the redemption risk posed by a requirement to buy back shares from farmers under certain circumstances. The outcome of this review was improved incentives for farmers to hold onto Fonterra shares beyond their normal limit, and changed mechanisms for valuing those shares. The review also led to the establishment of a private farmer-owned share market within which Fonterra shares could be traded, and the associated creation of a fund selling investment units to ‘friendly’ external investors to help farmers purchase Fonterra shares. The ‘Trading Amongst Farmers’ market was established in 2012 and run by NZX.

Underpinning much of this change was sustained investment by the NZX in the infrastructure required to run complex derivative markets. This was not a simple case of the infrastructure following the development of new financial instruments, rather a more co-constitutive relationship emerged. According to Mark Weldon: ‘in order to fund increasingly expensive IT platforms, and operate in an increasingly costly regulatory environment, we needed to… both grow our growth cash flow, and grow it in areas that were not correlated at the revenue line, with our ‘traditional’ cash equities, stock exchange business’ (NZX, 2011: 6). Markets require infrastructures to work, but infrastructures require markets to fund them. In this context, NZX announced that it would significantly upgrade its technology platform as part of its diversification strategy (NZX, 2003). This began in 2004 with the development of the Direct Market Access trading engine that allowed for automated ordering, and the launch of the Australian Electronic Communications Network intended to connect the Australian and New Zealand capital markets as part of a consortium of broking houses in Australia in 2006. Key to these developments was work on improving the settlement and clearing infrastructure in New Zealand’s financial sector. This was a major recommendation of the Capital Markets Development Taskforce established by the Labour-led government following the Global Financial Crisis in 2008 with the goal of developing a blueprint to transform New Zealand’s capital markets (Capital Markets Taskforce, 2009).

The Capital Markets Taskforce represented a moment when the particular technopolitics being practiced by the NZX through its transformation intersected with a wider set of aspirations about the role of finance in national development. The Capital Markets Taskforce observed that agriculture was highly underrepresented in existing capital markets (Capital Markets Taskforce, 2009) and recommended the development of a specialized agricultural capital market center. This center would incorporate activities ranging from the commercialization of innovation, the creation of public markets that reflected the particular characteristics of agricultural cooperatives, and the development of derivatives markets – an idea taken up by NZX. Mark Weldon argued that historically markets such as the Chicago Board of Trade had developed around geographically embedded industries, and that New Zealand should be looking to develop a global hub for international dairy markets, and in particular futures markets (New Zealand Herald, 2010), that would leverage its advantage as a ‘price setter’ in international dairy markets. These aspirations reflected the particular macro dynamics of the dairy trade which is characterized by the vast majority of dairy products being traded domestically, but with New Zealand’s dairy companies, particularly Fonterra, being very large players in the much smaller international dairy trade.

In 2008, Fonterra launched its Global Dairy Trade (GDT) auction platform to sell some of its commodities.² Part of the purpose of the development GDT was to provide an instrument to discover, and make transparent, dairy commodity prices. In mid-2010 NZX announced that it would be launching a futures market for Whole Milk Power (WMP). The goal in developing this market for WMP futures was to have the global dairy industry look to New Zealand for a reference price and give farmers and processers a clearer sense of market expectations for future prices (Rural News, 2010). NZX was not alone in seeking to develop a dairy futures market. Existing futures markets were being run by EEX’s Agricultural Commodity Exchange, London’s International Financial Futures Exchange, and the Chicago Mercantile Exchange. However, Kathryn Jaggard, NZX’s derivatives consultant, maintained that what was being developed in New Zealand reflected the preference of the dairy industry for cash settlements that avoided issues with logistics, perishability, and quality (Rural News, 2010).

The new NZX futures market began trading in early October 2010, with prices based on those paid at the fortnightly GDT auctions. NZX argued that the GDT auction price was transparent to all market participants, even though basing it on information derived from NZAgri would have driven business to NZX as market actors looked for pricing information (Fox, 2010). The new market started slowly with perception concerns, regulatory issues in the United States keeping a lot of potential actors out of the market, and the difficult temporality of trading globally from New Zealand. Despite this, NZX’s new CEO, Tim Bennett, reaffirmed the exchange’s belief in the market’s potential and its own ongoing investment in the market, even extending the market’s trading hours in mid-2013 and again in 2017 to enable European based brokers to participate before the end of their trading day, and to position the market within the hours of the main United States trading session.

The WMP futures market, which had been expanded to include a range of other commodities including skim milk powder and butter, was joined in 2016 by a milk futures market. The new milk futures market replaced Fonterra’s short-lived Guaranteed Milk Price scheme and represented another way of trying to provide farmers with new tools to insulate themselves against milk price volatility. In distinction to the commodity futures market which had been dominated by large, international ingredients companies, the milk futures market was aimed specifically at New Zealand farmers facing particularly volatile milk prices at the time. In April 2016, the new milk futures market was approved by the New Zealand’s Financial Markets Authority and began trading in May 2016. Andrew Hoggart, President of Federated Farmers, welcomed the new market, arguing that it provided farmers with risk management tools that put them on a level playing field with their international competitors. He did suggest, however, that farmers intending to use the futures market needed to get professional advice; a warning widely repeated. One source of advice and information would be NZXAgri’s data business AgriHQ.

NZXAgri as data providers

AgriHQ, the brand for the data business of NZXAgri, produces a considerably more comprehensive array of data for its customers from its base in Feilding than it did in its early days. Not just beef and lamb prices but dairy futures, forestry plantings, and other agricultural data. This information is made available in a series of reports, samples of which are available for free on the website or in the back pages of New Zealand Farmers Weekly, but in their complete form only through subscription. Data assemblages are comprised of organizations as well as data, and it is often through these that particular business practices, strategies, and systems of thought come to shape the assemblage and the nature and uses of the data (Kitchin, 2014). Since the purchase of the company by NZX in 2006, this work of collecting, collating, and distributing data in this way is understood as part of a larger strategy of providing the information that would support markets – specifically futures markets in New Zealand agriculture – operating on existing NZX infrastructure. The tidy-looking tables of the reports reinforce the notion that AgriHQ provides credible, timely, and useful information services for their customers, putting together objective knowledge about an objective world that may be changing and dynamic but in ways that can be measured, understood, and responded to (Beer, 2015; Desrosieres, 1998; Hacking, 1990; Porter, 1995).

But AgriHQ provides more than just information support for their customers: they try and make the latter into knowledgeable actors with specific agencies relating to market participation (Çalışkan and Callon, 2009, 2010; Callon, 2007). Indeed, workers at AgriHQ understand this, describing their work as providing information in order to get people to participate in the market. Here they are performing an economic theory around the problems of information asymmetries and their effects on the confidence of market participants (see Akerlof, 1970). At one point in our interviews, one AgriHQ worker is quite explicit:

this centres around an economic theory… that the more somebody knows around something then the more confidence they have to trade and the more on average they will actually pay for it. It’s someone’s theory of lemons or something. It’s like if you have two lemons at the supermarket³ and one you know absolutely nothing about, and one you know some information about even if it might not be the greatest history or story behind it, it still creates some certainty around it. You will then actually pay, in theory, more for that one you know at least something about. (Glenn, interview)

This performance is making, or at least trying to make, a market (Callon, 2007; MacKenzie, 2006; MacKenzie and Millo, 2003; MacKenzie et al., 2007). The effect is to provide their subscribers not simply with information support that they use in decisions they would have made anyway, but with new capacities and knowledges that transform them into subjects who may be able and willing to participate in a futures market. For farmers, these data networks make new risks and opportunities visible and calculable. They change the topology of farmers’ worlds, so that particular relations to other actors and places become more or less important (Allen, 2016), and the relation to the derivatives market becomes more conceivable and sensible. This makes agricultural futures markets possible. The point here is that the data assemblage does not simply support the market but constitutes it because it constitutes the necessary agencies that might participate. These subjectivities are a part of the data assemblage and they and their agencies cannot be separated from it (Kitchin, 2014). So what does this assemblage look like?

The data assemblage that AgriHQ positions itself in has both material and immaterial resources and practices (Pickren, 2018). For one, the materiality of the data assemblage has changed dramatically since the days of Agrifax, which comprised a national grid of phone lines through which spoken words and images, in the form of facsimiles, could be sent. Now, AgriHQ is in an office with fiber-optic cables connecting them across international space via the internet to data sources and customers both inside and outside of New Zealand. Words, images, numbers, videos, all can travel almost instantaneously across space for very low costs. The volatility that futures trading is meant to help manage is put down to such shifts in materiality by one ArgiHQ worker:

I think that is largely because people are a lot more aware of what is going on in a market. The information flows around the market are much better. Whereas, (in the past)… you had to send a telegram to find out that somebody has just shipped a thousand tons of lamb to the UK and it took three months for the ship to arrive. Now we fly product around the world and we can influence supply. (Paul, Interview)

AgriHQ not only distribute data more widely on these networks, they bring in more data as well, scavenging data from both paid and free sources – such as US agricultural statistics – and configuring them in line with their own format for redistribution. They also conduct their own datafication exercises (Mayer-Schönberger and Cukier, 2014), producing exclusive data through surveys conducted on the phone and through email with various farmers and stocking agents, and they collect data on livestock weights through teams of animal weighers and scales kept on call at saleyards around New Zealand. It is no use imagining AgriHQ’s data as a spreadsheet in a file folder on a computer in their office in Feilding from which analysts extract material for their reports. They are better understood as mustering and shepherding databases and datapoints into their report format from the various networks that extend out of their building. Made commensurable, these are distributed along many of those same networks and others. The material resources and practices of the assemblage and the digital flows it enables produce the market agencies required for the futures market.

But the assemblage is more than just technology-driven connections. A key set of both material and immaterial practices that AgriHQ uses to construct and direct the data assemblage are relational in the sense that they are about human relationships. This relational work is central to maintaining the data assemblage. For example, generally people are willing to participate in the price surveys even though they are giving up time and information because in the survey they feel as though they get something back in the discussion about what is happening in the markets they are working in, particularly given not everything AgriHQ picks up is published in their reports. According to Glenn, ‘a big part of the job is actually managing the relationships with those contacts.’ The maintenance of trust is the maintenance of the assemblage. Additionally, particularly given a significant proportion of their content is not exclusive, AgriHQ provide analysis of the data in their reports, explaining the consequences for their customers of what the data is communicating. This is relational work that maintains AgriHQ’s project of futures market-making at the center of the data assemblage. The history of AgriHQ and its placement in a rural center are useful for them here. Most of the analysts have rural backgrounds and have attended one of New Zealand’s agricultural universities: Massey or Lincoln. This commonality is important for these relationships:

If you are talking to famers and writing for farmers then it is quite hard for them without a rural background. It’s so easy to write something in a report that you think is a safe enough comment when actually you are coming across as a complete numpty. (Glenn, interview)

While for Jan:

here in Feilding the bonus is, even if they don’t come off a farm, they are in some ways connected to agriculture whether that is through friends or partners, there is always some connection there. It just means that… everybody on the phone is kind of saying the right things and having the right conversations and you’re just that much more connected to the day to day and to what’s happening.

This demonstrates the point that data assemblages have histories and geographies that feed into how they work (Barnes and Wilson, 2014; Graham and Shelton, 2013), and the extent to which they must be actively made and maintained to be able to produce the desired effects – in this case the production of market agencies.

This is not to say that the conversion of these agencies into market actors is inevitable. What the very requirement for relational work reveals is the extent to which there are competing projects, desires, ideas, opinions, attitudes, emotions, and comportments across the various components that are brought together to constitute this data assemblage. The origin story of Agrifax underlines the point that farmers and others in agri-business are not just datapoints on the one hand or passive receptacles for the information AgriHQ provides on the other: they are actors in the making of the assemblage through their own projects and ideas (Williams, 2014). For example, some exporters do not provide information to AgriHQ because they believe that this gives their buyers an advantage. Meat processors often protest that AgriHQ might say in their reports that prices are going up when they have not lifted theirs, producing tension in their day-to-day sales relationships. Sometimes people ask to not have their animals weighed at the saleyards. There is variation across sectors in this regard as well, as Glenn explains in relation to sheep and beef:

Probably the biggest barrier in the sheep and beef industry is the people. They have a reputation for being pretty stuck in their ways and resistant to change. Whether that is through livestock agents and companies, or some of the meat processors themselves who have been in the industry for forty odd years and doing the same thing.

AgriHQ’s workers tend to explain this as a different attitude or set of values in relation to data transparency. It illustrates the variation across the assemblage that must be constantly managed and pulled into line. For the NZXAgri futures market project to be successful, it needs to overcome, or come to terms with, the many other projects that actors across the assemblage are participating in. Any of these alternative projects could work against the integrity of the assemblage. This is to not even mention the way that the materials being datafied could undermine the data itself by resisting the standardization required to render it measurable (Henry, 2017; Henry and Roche, 2013). When it appears in a spreadsheet in a report, data looks like something static. But really it is a flow, not just of numbers, but of ideas and emotions and words and attitudes and material agency, all of which must be harnessed and managed to produce the spreadsheet in a credible way. This is the work of the data assemblage.

Conclusion

In this paper, we have deliberately focused on financialization as work (Williams, 2014) and the centrality of the assemblage of data relationships to this. In tracing the emergence of New Zealand’s dairy futures markets we have uncovered the genealogies and topologies of the financial data assemblages required to make those markets work. These data assemblages now represent increasingly invisible infrastructures that shape the possibilities of financial markets and of the multiple actors that constitute them. The critical work of making finance visible requires us to look at these assemblages, their genealogies, embedded technopolitics, materialities, and emergent topologies.

Rather than the conventional view of financialization that takes the logic, organization and trajectory of capital as given, we have highlighted the specificity of finance, the reliance upon particular material arrangements, and the creation of situated ways of knowing. The data infrastructure we have examined has a genealogy produced by the neoliberal restructuring of agriculture in the 1980s and 1990s that dissolved existing sources of commodity pricing and market information in New Zealand. Seen from this angle, the formation and development of Agrifax represented less the imposition of a set of financial logics on farming, and more the situated effort to try and create tools to enable farmers, and other actors, to exercise marketplace agency more effectively. The data assemblage, and the agency created by Agrifax, represents the exercise of a technopolitics that has deep roots in those farmer subjectivities of independence and self-reliance through which the lived experiences of neoliberal restructuring have been refracted (Ouma, 2016; Williams, 2014). Simultaneously, however, this technopolitics should be seen as operating at scales beyond the individual. Here we can see the fusing of financial subjectivites with investment instruments and capital markets infrastructures as the technical means through which the political goals of company diversification, dairy dominance, and national economic development might be realized. None of these projects completely displace anything else, meaning that within the data infrastructure supporting New Zealand’s dairy futures markets are heterogeneous logics that need to be constantly worked on to maintain degrees of functional alignment. It means that the assemblage has fissiparous qualities that threaten its durability, but which also provide nooks and crannies for a multiplicity of financial logics and practices to survive.

In this ‘ag space’ intricate topological relationships fold and connect financializing relationships in unexpected ways. The development of the futures market by NZX, its own data analytics businesses, and the connected GDT platform, are part of aspirations to create an innovation and financial hub for the international dairy trade in New Zealand. So rather than being at the margins of the dairy commodity trade, the various infrastructural platforms being created represent an active attempt to constitute and embed the very rules of the game. Place has a key role here (French et al., 2011; Hall, 2011). The place of Feilding at the center of a data assemblage that works with, and produces, complex, variegated flows of data across New Zealand and globally represents another fold in the financializing topologies. Likewise, the personal performances of financializing by analysts at NZXAgri that relies upon their simultaneous performance of rurality suggests that the authority and logic of finance is possibly quite weak and needs careful shepherding to order to do any work.

This raises the question of who changes when we start to approach finance as work in a situated fashion. Our tendency is to frame those actors driving financializing work as unchanging, always complete, and ultimately colonizing in their ambitions and practices, and to also assume the obverse for those subjected to financializing work (French et al., 2011; Hall, 2012). In this view, financial work is done to subjects, and they are transformed as a consequence. Yet what we have shown in this paper is that while the creation of new data assemblages, new infrastructures, and new markets are potentially generative of different farmer subjectivities and relationships, we can also argue that it is generative of other effects as well. NZX, its analysts and their infrastructures, and the other actors in financializing agricultural relationships in New Zealand, have not stayed the same. In finding ways to insert themselves into agricultural relationships they have had to build new infrastructures, create new ways of knowing the uncertain, variable materialities of agricultural production, and to perform particular versions of rurality in order to be trusted. In this context, a question to ponder is the extent to which the changes wrought by the relationship between financializing work and agriculture should encompass a sense of how financializing work is itself transformed by the specific problems of translating agriculture into financial worlds and logics. We offer this question as an example of the critical work that needs to be done to expose actually existing financializing work, its technopolitical forms and aspirations, its subjectivities, and the worlds that these make. We would argue that finance has been transformed by its relationships to agricultural logics, subjectivities, and practices, and in some sense it might be fair to talk about the agriculturalizing of finance as much as the financializing of agriculture.

Footnotes

Acknowledgements

This paper was based on research and workshops conducted by the Making Markets Research Collective at Massey University. We would like to thank the other members of the collective for their input: Carolyn Morris, Aisling Gallagher, Stephen FitzHerbert, Nick Lewis, and, in particular, Cassandra McTavish who conducted and transcribed the interviews. The paper was presented at the 2017 Agrifood Conference in Bandung, Indonesia, and we would like to thank attendees for their generous feedback. Finally, our thanks to the three reviewers and Brett Christophers for their incisive reviews that helped us to sharpen the argument. All errors remain our own.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Akerlof

(1970) The market for “lemons”: Quality uncertainty and the market mechanism. The Quarterly Journal of Economics 84: 488–500.

Allen

(2016) Topologies of Power: Beyond Territory and Networks. London: Routledge.

Allen

(2007) Weldon believes Kiwis can fly. The Press, 14 February, p. 4.

Amoore

Poitukh

(2016) Algorithmic Life: Calculative Devices in the Age of Big Data. Abingdon: Routledge.

Baines

(2017) Accumulating through food crisis? Farmers, commodity traders and the distributional politics of financialization. Review of International Political Economy 24: 497–537.

Barber

(2006) Price setting and arcane business. New Zealand Herald, 3 July. Available at: http://www.nzherald.co.nz/the-country/news/article.cfm?c_id=16&objectid=10389396http://www.nzherald.co.nz/the-country/news/article.cfm?c_id=16&objectid=10389396 (accessed 24 April 2018).

Barnes

Wilson

(2014) Big data, social physics, and spatial analysis: The early years. Big Data & Society 1: 1–14.

Beer

(2015) Metric Power. London: Palgrave Macmillan.

Bennett

McFall

Pryke

(2008) Culture/economy/society. Journal of Cultural Economy 1: 1–7.

10.

Berndt

Boeckler

(2009) Geographies of circulation and exchange: Constructions of markets. Progress in Human Geography 33: 535–551.

11.

Beunza

Hardie

MacKenzie

(2006) A price is a social thing: Towards a material sociology of arbitrage. Organization Studies 27: 721–745.

12.

Bishop

(2006) Alliance says: ‘We’re no sacrificial lamb’. Straight Furrow, 31 January, p. 1, 7.

13.

Boyd

Crawford

(2012) Critical questions for big data: Provocations for a cultural, technological, and scholarly phenomenon. Information Communication and Society 15: 662–679.

14.

Bronson

Knezevic

(2016) Big data in food and agriculture. Big Data & Society 3: 1–5.

15.

Brooks

(2016) Inducing food insecurity: Financialisation and development in the post-2015 era. Third World Quarterly 37: 768–780.

16.

Bruno

Buyuksahin

Robe

(2017) The financialization of food? American Journal of Agricultural Economics 99: 243–264.

17.

Burch

Lawrence

(2009) Towards a third food regime: Behind the transformation. Agriculture and Human Values 26: 267–279.

18.

Burch

Lawrence

(2013) Financialization in agri-food supply chains: Private equity and the transformation of the retail sector. Agriculture and Human Values 30: 247–258.

19.

Calder

Tyson

(1999) Meat Acts: The New Zealand Meat Industry 1972–1997. Wellington: Meat New Zealand.

20.

Çalışkan

Callon

(2009) Economization, part 1: Shifting attention from the economy towards processes of economization. Economy and Society 38: 369–398.

21.

Çalışkan

Callon

(2010) Economization, part 2: A research programme for the study of markets. Economy and Society 39: 1–32.

22.

Callon

(2007) What does it mean to say economics is performative? In: MacKensie

Muniesa

Siu

(eds) Do Economists Make Markets? On the Performativity of Economics. Princeton: Princeton University Press, pp. 311–357.

23.

Capital Markets Taskforce (2009) Capital Markets Matter: Summary Report of the Capital Market Development Taskforce. Wellington: Capital Markets Taskforce.

24.

Carolan

(2016) Publicising food: Big data, precision agriculture, and co-experimental techniques of addition. Sociologia Ruralis 57: 135–154.

25.

Christophers

(2011) Follow the thing: Money. Environment and Planning D: Society and Space 29: 1068–1084.

26.

Christophers

(2014a) Geographies of finance I: Historical geographies of the crisis-ridden present. Progress in Human Geography 38: 285–293.

27.

Christophers

(2014b) The territorial fix: Price, power and profit in the geographies of markets. Progress in Human Geography 38: 754–770.

28.

Christophers

(2015) Geographies of finance II: Crisis, space and political-economic transformation. Progress in Human Geography 39: 205–213.

29.

Clapp

Isakson

Visser

(2017) The complex dynamics of agriculture as a financial asset: Introduction to symposium. Agriculture and Human Values 34: 179–183.

30.

Cronon

(1991) Nature’s Metropolis: Chicago and the Great West. New York: W. W. Norton.

31.

Cronshaw

(2008) Leaving the farm. The Press, 3 October, p. C8.

32.

de Lacy

(1997) Happy 10th birthday, Agrifax. The Press, 11 September, p. 35.

33.

Dalton

Thatcher

(2014) What does a critical data studies look like, and why do we care? Society and Space Open Site. Available at: www.societyandspace.org (accessed 17 April 2018).

34.

Desrosieres

(1998) The Politics of Large Numbers: A History of Statistical Reasoning. Cambridge, MA: Harvard University Press.

35.

Du Gay

Pryke

(2002) Cultural Economy. London: SAGE.

36.

Dunn

(1987) MAFTech hardline on user-pay: Farmer for former-free data. Dairy Exporter, May, p. 6.

37.

Edlin

(1997) Williamson torpedoes years of agricultural research. The Independent, 1 August, p. 26.

38.

Field

(2016) The financialization of food and the 2008–2011 food price spikes. Environment and Planning A: Economy and Space 48: 2272–2290.

39.

Folkers

(2017) Existential provisions: The technopolitics of public infrastructure. Environment and Planning D: Society and Space 35: 855–874.

40.

Fox

(2010) Dairy futures role queried. The Press, 9 January, p. 19.

41.

Freidberg

(2014) Footprint technopolitics. Geoforum 55: 178–189.

42.

French

Leyshon

Wainwright

(2011) Financializing space, spacing financialization. Progress in Human Geography 35: 798–819.

43.

Friedmann

McMichael

(1989) Agriculture and the state system: The rise and decline of national agricultures, 1870 to the present. Sociologia Ruralis 29: 93–117.

44.

Galbreath

(1998) DSIR: Making Science Work for New Zealand: Themes from the Department of Scientific and Industrial Research, 1926–1992. Wellington: Victoria University Press in Association with the Historical Branch.

45.

Gertel

Sippel

(2016) The financialisation of agriculture and food. In: Shucksmith

Brown

(eds) Routledge International Handbook of Rural Studies. London: Routledge, pp. 274–287.

46.

Ghosh

(2010) The unnatural coupling: Food and global finance. Journal of Agrarian Change 10: 72–86.

47.

Gibson-Graham

(2014) Rethinking the economy with thick description and weak theory. Current Anthropology 55: S147–S153.

48.

Gill

(2016) Can the river speak? Epistemological confrontation in the rise and fall of the land grab in Gambella, Ethiopia. Environment and Planning A 48: 699–717.

49.

Graham

Shelton

(2013) Geography and the future of big data, big data and the future of geography. Dialogues in Human Geography 3: 255–261.

50.

Grant

(1997) Bulls, Bears & Elephants: A History of the New Zealand Stock Exchange. Wellington: Victoria University Press.

51.

Hacking

(1990) The Taming of Chance. Cambridge: University of Cambridge Press.

52.

Hall

(2011) Geographies of money and finance I: Cultural economy, politics and place. Progress in Human Geography 35: 234–245.

53.

Hall

(2012) Geographies of money and finance II: Financialization and financial subjects. Progress in Human Geography 36: 403–411.

54.

Hecht

(2009) The Radiance of France: Nuclear Power and National Identity after World War II. Cambridge: MIT Press.

55.

Hecht

(2011) Entangled Geographies: Empire and Technopolitics in the Global Cold War. Cambridge: MIT Press.

56.

Henry

(2017) Meat, metrics and market devices: Commensuration infrastructures and the assemblage of ‘the schedule’ in New Zealand’s red meat sector. Journal of Rural Studies 52: 100–109.

57.

Henry

Roche

(2013) Valuing lively materialities: Bio-economic assembling in the making of new meat futures. New Zealand Geographer 69: 197–207.

58.

Isakson

(2014) Food and finance: The financial transformation of agro-food supply chains. Journal of Peasant Studies 41: 749–775.

59.

Kitchin

(2014) The Data Revolution: Big Data, Open Data, Data Infrastructures and Their Consequences. London: SAGE.

60.

Lapavitsas

(2013) Profiting without Producing: How Finance Exploits Us All. London and New York: Verso.

61.

Law

Mol

(2001) Situating technoscience: An inquiry into spatialities. Environment and Planning D: Society and Space 19: 609–621.

62.

Leyshon

Thrift

(2007) The capitalization of almost everything: The future of finance and capitalism. Theory Culture Society 24: 97–115.

63.

Lind

(2013) Till the Cows Came Home: Inside the Battles That Built Fonterra. Wellington: Overland Publishing.

64.

MacKenzie

(2006) An Engine, Not a Camera: How Financial Models Shape Markets. Cambridge: The MIT Press.

65.

MacKenzie

Millo

(2003) Constructing a market, performing theory: The historical sociology of a financial derivatives exchange. American Journal of Sociology 109: 107–145.

66.

MacKenzie

Muniesa

Siu

(2007) Do Economists Make Markets? On the Performativity of Economics. Princeton: Princeton University Press.

67.

Martin

Clapp

(2015) Finance for agriculture or agriculture for finance? Journal of Agrarian Change 15: 549–559.

68.

Mayer-Schönberger

Cukier

(2014) Big Data: A Revolution That will Transform How we Live, Work and Think. New York: Eamon Dolan/Mariner Books.

69.

Mitchell

(2002) Rule of Experts: Egypt, Techno-Politics, Modernity. Berkeley: University of California Press.

70.

Murdoch

Ward

(1997) Governmentality and territoriality: The statistical manufacture of Britain’s ‘national farm’ Political Geography 16: 307–324.

71.

New Zealand Herald (2010) Why not have a big agriculture hub? Dairy is already here. New Zealand Herald, 27 May. Available at: http://www.nzherald.co.nz/business/news/article.cfm?c_id=3&objectid=10647760 (accessed 24 April 2018).

72.

NZX (2003) Annual Report 2003. Wellington: NZX.

73.

NZX (2005) Annual Report 2005. Wellington: NZX.

74.

NZX (2006) Annual Report 2006. Wellington: NZX.

75.

NZX (2008) Annual Report 2008. Wellington: NZX.

76.

NZX (2009) Annual Report 2009. Wellington: NZX.

77.

NZX (2010) Annual Report 2010. Wellington: NZX.

78.

NZX (2011) Annual Report 2011. Wellington: NZX.

79.

Ouma

(2014) Critical review: Situating global finance in the Land Rush Debate: A critical review. Geoforum 57: 162–166.

80.

Ouma

(2016) From financialization to operations of capital: Historicizing and disentangling the finance-farmland-nexus. Geoforum 72: 82–93.

81.

Pickren

(2018) The global assemblage of digital flow. Progress in Human Geography 42: 225–243.

82.

Porter TM (1986) The Rise of Statistical Thinking, 1820-1900. Princeton: Princeton University Press.

83.

Porter

(1995) Trust in Numbers: The Pursuit of Objectivity in Science and Public Life. Princeton: Princeton University Press.

84.

Rural News (2010) WMP futures to lead the world. Rural News, 31 August.

85.

Russi

(2013) Hungry Capital: The Financialisation of Food. Hants: Zero Books.

86.

Sassen

(2012) Global finance and its institutional spaces. In: Knorr Cetina

K and

Preda

(eds) The Oxford Handbook of the Sociology of Finance. Oxford: Oxford University Press, pp. 13–32.

87.

Shearmur

(2015) Dazzled by data: Big Data, the census and urban geography. Urban Geography 36: 965–968.

88.

Williams

(2014) Feeding finance: A critical account of the shifting relationships between finance, food and farming. Economy and Society 43: 401–431.

89.

Wyly

(2014) The new quantitative revolution. Dialogues in Human Geography 4: 26–38.