Infrastructures of Farmland Valuation in Australia

ABSTRACT

This article investigates how valuation practices change as land becomes financialised in Australia. Financial investors bring their own assumptions on how land value should be ‘measured’ to enhance its fungibility as a financial asset. Importantly, this includes a shift to a ‘future income’ approach to value, which produces frictions with established land valuation practices. Combining valuation and infrastructure studies, I examine how conceptualisations of value; techniques of standardisation, classification and comparison; devices for measuring and calculating; templates and databases; social networks and relationships; as well as multiple materialities facilitate, legitimise and stabilise land value. These infrastructures of value are crucial for making new conceptualisations of value productive while both enabling and limiting change in valuation practices. Value further emerges as an important infrastructure itself. Shared understandings of value, embedded and expressed in valuation infrastructures, work as a ‘conduit’, a mechanism necessary for people to come to agreements in exchange processes.

KEYWORDS:

• Valuation
• financialisation
• farmland
• infrastructure
• measurements
• Australia

Imagine for a moment you are a portfolio manager, working for a pension fund that is mandated to take care of the money people save for their retirement. In order to do this, you can either conservatively store their money in a safe place, so that it does not lose value; or you can make use of it in order to increase its value. Here, simply put, you have two options, you can store the money in a place that itself gains in value over time, or you can use the money to generate more money. What, however, if you were able to combine the two? As many of the traditional, ‘safe’ asset classes collapsed during the financial crisis in 2008, farmland investments were touted as such an ideal alternative, as they combined income streams with appreciation, or – more metaphorically speaking – ‘gold with yield’ (Fairbairn 2014). This promotion of farmland as an alternative financial asset class was underpinned by its long-term ‘fundamentals’ – a rising global population demanding an increasing amount of food that needs to be supplied from limited natural resources – and statistical data, indicating that farmland serves as a hedge against inflation and is uncorrelated to other asset classes. Moreover, it was argued that the value of land itself had steadily gone up over time. For example, the Savills Global Farmland Index, which collates data on the 15 ‘key farmland markets’,¹ recorded an average annualised value growth of 12% between 2002 and 2018 (Savills 2019).

These numbers on global farmland values had ‘a key function in the assemblage of farmland as an attractive financial asset’ in the post-2007 global land rush (Visser 2017: 192) and did ‘important work’ in rendering land investable (Li 2014: 598). A great deal of this work is due to the particular persuasiveness of numbers: people grant numbers authority, as they consider them as especially ‘credible’ representations of the world (Espeland & Stevens 2008: 416–417). Yet, indexes such as the Savills Global Farmland Index are not only highly selective in terms of the geography they represent (e.g. Africa, thus an entire continent, is not part of Savills’ ‘global’ index) and undifferentiated regarding different types of farmland or farming; there is also no further explanation as to how the data used for this index was actually produced. Rather, the index ‘naturalises’ value (Styhre 2013: 52) and presents farmland values as measurable ‘facts’, abstracted from the various practices and infrastructures that produce them.

This article looks behind these numbers and identifies the institutional settings, social norms, communities of practice, as well as technological and material components, which need to come together to constitute the economic value of land in Australia. I specifically examine how valuation practices are challenged as farmland becomes financialised. Constituting an asset’s appreciation, valuations are key factors within investment decision-making, such as the scenario described in the introductory vignette. Yet, despite the assertive number attached to it, value is not ‘measured’ but ‘infrastructured’ differently in different contexts, and this infrastructuring is subject to negotiation and change. Capitalising on an asset’s appreciation – a key goal of financial activity – thus hinges on potentially highly diverging practices of ascribing value to something. Financialisation, the proliferation of financial logics and rationales in various areas of society beyond the traditional ‘realm’ of finance (van der Zwan 2014), profoundly changes the way in which entities (i.e. things, ideas, beings, or activities) are valued.

This article investigates how infrastructures of value change as the valued entity – here land – is turned into a financial asset. While land has long been a ‘storage’ of value, turning land into an institutionalised financial asset class here refers to rendering it a rather liquid asset that can be more flexibly bought and sold, while also systematically and reliably producing future income streams for investors (Birch 2017; Birch & Muniesa 2020).² Financial investors, I argue, come with their own assumptions as to how land’s value should be ‘measured’, and what kind of indicators they need to make it fit within their portfolios and enhance its fungibility as a financial asset. Importantly, this includes a shift to a ‘future income’ based assessment of land value, which produces multiple frictions with the established land valuation practices.

Combining insights from valuation studies and research on infrastructure, I explore the changing land valuing practices in a twofold way, via the notion of infrastructures of value and through considering value as an infrastructure in itself. In line with the aim of this special issue, I seek to tease out the analytical usefulness of an infrastructure lens for tying together and making sense of a variety of empirical observations. I use the notion of ‘infrastructures of value’ (see Lammer and Thiemann, this issue) to examine how conceptualisations of value; techniques of standardisation, classification and comparison; devices for measuring and calculating; templates and databases; social networks and relationships; as well as multiple materialities facilitate, legitimise and stabilise land value. These infrastructures of value are crucial for making new conceptualisations of value productive while both enabling and limiting change in valuation practices. For the ethnographic material presented in this paper, the infrastructure lens further helps to bring often overlooked but crucial material components into the study of land’s financialisation (Sippel & Visser 2021). It lastly allows me to better understand how ‘unique’ land is turned into a ‘generic’ promise of a continuous cash flow by systematically inscribing it with stable and legitimate value.³ In other words, an infrastructure perspective helps investigating how something as incommensurable as land is made commensurable and thus accessible for global investment, while being mindful of the tensions and frictions inherent within this process. Within all this, I argue, the concept of value emerges as yet another form of infrastructure itself. Shared understandings of what value is and how it can be determined – as embedded and expressed in valuation infrastructures – work as a ‘conduit’, a mechanism that is necessary for people to come together in exchange processes, as they facilitate and stabilise agreement, and thereby allow for transactions to take place. If agreement on a shared understanding of value fails, there is no foundation – no infrastructure – to create this exchange.

This article draws on more than one year of fieldwork on the intersections between finance and farmland in Australia between 2016 and 2019. My research included participant observation at industry events and at the community and farm levels, as well as more than 70 interviews with various actors, including around 25 representatives of farmland investment companies and five professional farmland valuers. Reflecting the spectrum of actors from traditional valuers to more corporate actors, interviews took place in different settings, from rather moderately furbished rural offices to offices in the upper floors of urban high-rises, overlooking Sydney harbour or Brisbane River. Despite repeated requests, none of my interviewees was willing to take me along for a valuation or ‘due diligence’ (in the case of the investment companies) on the ground. This points to the sensitivity and confidentiality surrounding land valuations, and the difficulties involved in studying financial operations and ‘studying up’ more generally (Gusterson 1997; Mikecz 2012). Given this situation, I decided to ‘simulate’ the valuation procedure during the interviews and asked my informants to ‘walk me through’ the process of farmland valuation step by step.

The remainder of this paper is organised as follows. I first develop a conceptual framework linking valuation and infrastructure studies. I then depict the current infrastructure of farmland valuation in Australia. This is followed by an examination of how the emergence of financial actors on the farmland market has changed this infrastructure of value. I conclude by outlining the value of understanding value itself as an infrastructure.

Value, Valuations and Infrastructures

Value has been a contested issue in social research. Scholars, among others, have sought to identify what value is and where it comes from; addressed why people value certain things above others; examined the relationship between value and values; studied regimes of valuation; and inquired about the usefulness of value as a theoretical concept as such (e.g. Eiss & Pedersen 2002; Graeber 2001; Narotzky & Besnier 2014; Ortiz 2013; Pedersen 2008). Departing from the problem of ‘false representation’ – the idea of being able to identify whether some kind of value is ‘real’, whereas some other kind of value is not – this paper approaches value from a process-oriented, or ‘pragmatist’, perspective that holds that nothing is intrinsically valuable (Bigger & Robertson 2017; Muniesa et al. 2017). Instead, value is considered as grounded in social meaning, and therefore rooted in, and established through, social relations. Analytically and empirically, this means shifting from the ‘substantive’ – the analysis of what value ‘is’ – to the activity of ‘valuing’; it is the question of what makes something valuable to people, and the operations and social work of valuation, that becomes the focus of attention (Muniesa et al. 2017, 14–15).

Valuations – as an institutionalised way of determining (mostly) economic value – consist of a number of operations that enable comparisons, such as categorisations, classifications, standardisations and hierarchisations of things, ideas, beings, or activities (Lamont 2012). Valuing, moreover, requires inter-subjective agreements on the specific criteria or matrixes based on which these operations should take place, and on the actors who are legitimised to undertake them (see also Boltanski and Thévenot 2006 [1991]). These operations involve, and rely on, various technologies, such as instruments, tools and devices for measuring, gauging, assessing and scaling, and are embedded in efforts of legitimation, stabilisation, routinisation and institutionalisation of value. Moreover, as Scott (1998) shows, certain measurements not only measure an existing reality in terms of one (or a few) variables (thereby reducing the complexity of what is measured); rather, reality is forcefully (at times violently) reshaped in order to make measurements and calculations about constant revenue possible.

Bundling together diverse sets of practices and operations with the material devices, these perspectives on valuations can be fruitfully expanded through an ‘infrastructure’ lens. Scholarship on infrastructure focuses on the material and immaterial systems or networks that enable the flow of goods, people or ideas, and make their exchange over space possible (Larkin 2013). Infrastructure scholars have studied a broad variety of subjects, from roads, bridges, pipes, sewers, wires and data centres (e.g. Harvey & Knox 2015; Özden-Schilling 2016; Vonderau 2019), to accounting practices, statistical models, automated algorithms, knowledge, social norms and customs, and people (e.g. Elyachar 2011; Lampland & Star 2009; Simone 2004). As Larkin points out, infrastructures are not simply ‘out there’. Rather, identifying and defining an infrastructure, and what belongs to it or not, is ‘a categorical act’ (Larkin 2013: 330; see also Star & Lampland 2009; for shifts in emic understandings of what constituted the infrastructure producing value in raspberry farming, see Thiemann, this issue). Studying something through an infrastructure lens thus means looking at things, components, or practices of the social world as they fulfil certain key – infrastructuring – functions. One such key function, which is of particular interest in this article, is the capacity of infrastructures to make other things possible; they act as important facilitators in society (Bernards & Campbell-Verduyn 2019: 777). Further elements of infrastructures include (Bernards & Campbell-Verduyn 2019; Star 1999): their embeddedness or implication in other material or immaterial structures; their openness to the use of various actors, beyond those who created them; their centrality as they crucially shape the way in which activities (can) take place; and the durability, stability and persistence they provide over time. Infrastructures are linked with the particular conventions and standards of a community of practice, and are learned as being part of that community of practice. Finally, infrastructures are not void of power; rather, an infrastructural lens can help emphasise questions of distribution, power and governance (Bernards & Campbell-Verduyn 2019: 780–783). Examining the practices of farmland valuation in Australia through an infrastructure lens thus emphasises how various conceptual, methodological, technological, practical, social and political components combine with the material world in facilitating, institutionalising, legitimating and stabilising value judgements – thereby allowing for the concept of value itself to work as an infrastructure within these exchange processes.

To further understand the durability and persistence of this infrastructure of value, and the emerging frictions, incompatibilities and destabilisations – specifically as farmland is becoming financialised – two additional perspectives are useful. First, there is an important distinction between processes of valuation as the inscription of economic value, which often takes the form of a metric (e.g. market price), and processes of commensuration as ‘the valuation or measuring of different objects with a common metric’ (Espeland & Stevens 2008: 408; see also Styhre 2013). Achieving commensurability of objects is a crucial component of financialised valuations, as will be demonstrated below. Second, based on understanding financialisation as ‘the ingraining of financialised metrics and reasonings in spaces and situations where they were previously non-existent or less common’, Chiapello considers financialised valuations as ‘valuation processes equipped by models, instruments and representations belonging to the explicit knowledge underpinning the approach and practices of finance professionals’ (Chiapello 2015: 17). Following Chiapello, it is critical to examine (1) the various and potentially changing viewpoints and ‘sets of assumptions’ based on which farmland is valued, which are bound up with the people performing the valuation and those requesting it; and (2) the different calculation methods and metrics used within the valuation process.

In what follows, I first delineate how farmland is currently valued in Australia. Based on that, the subsequent section examines how valuation infrastructures are destabilised and change as farmland becomes financialised.

Valuing Farmland in Australia: Determining Value, based on Market Price

Land in Australia is mostly valued by professional, independent valuers who undertake valuations for various purposes, from land transactions to assessing mortgage security to the annual valuation for taxation. Valuations are requested from different parties, including those who are interested in selling or buying (e.g. family farmers, companies, investment groups) as well as banks and the state. To date, the ‘direct comparison’ and ‘summation’ approaches have been the most commonly used methods to value farmland in Australia (Lane 2016). In a nutshell, both conceptualise the value of farmland as economic value that is derived from the price paid for a similar plot of land on the farmland market. Below, I describe the different steps and of this established infrastructure of value.

Measuring, Classifying and Comparing Properties

When asked to value a property, Paul,⁴ a professional valuer in his late 40s who has been working for a valuation company in rural New South Wales (NSW) for some 20 years, starts by collecting information about the property. This usually includes a direct inspection of the property, during which he takes photographs and notes on the four key elements: land, water supply, structural improvements and housing, which according to the direct comparison and summation approach constitute the value of the property. He thus identifies: different soil types (on soil mapping as infrastructuring of terroir see also Ana, this issue); existing rivers, creeks and irrigation infrastructure; sheds, fences, cattle yards, or paddocks; and collects information on the living areas (kitchens, bathrooms, furniture, etc.). He also notes existing livestock and permanent crops, as these are part of the valuation. To facilitate this procedure, Paul uses a tablet that runs a program developed for valuations. This program includes the valued elements, such as ‘land’, along with their pre-defined subcategories, for example the soil category ‘red clay’, and allows him to enter this information directly during the inspection. Nevertheless, Paul describes himself as a bit ‘old-fashioned’ as he never leaves for an inspection without pen and paper: ‘What I find is that my batteries generally run out before I finish, or you want to write something down that doesn’t quite fit in [the template of the program]’ (Paul, NSW, 2016).

Subsequently, Paul tries to find a ‘comparable sale’, that is to say a transaction that is as recent, close and comparable in terms of the size and characteristics of the property to the valued property as possible. Finding the ‘right’ comparable sale is not always easy, as there are not that many sales of properties in rural areas, especially when it comes to larger transactions. Ideally, a comparable sale is not older than 12 months; but depending on the activity in the land market, sales of up to two years might still be included. If the valuer has a good relationship with a real estate agent, s/he can also use a sale that has been contracted but not yet settled, which Paul considers as the ‘most recent market evidence’ a valuer can get (Paul, NSW, 2016). Moreover, regional proximity is key. A neighbouring property for example always qualifies as a ‘comparable sale’ (Paul, NSW, 2016). However, the valuation of the rarer large properties – which due to their economies of scale are considered more valuable than smaller properties – often requires including the national market to find an adequate comparison. That is why Paul also keeps a record of larger transactions outside of his core area of activity.

If a comparable sale is found, the last step is the ‘sale analysis’. This means that the valuer identifies the same elements – land, water, structural improvements and housing – for the comparable sale and, based on the price the property sold for, breaks them down into comparable price units. These are then compiled in the final report for the client, which is automatically generated based on a template and the information entered: ‘We put all that together, and the system spits out a report … magic number down the bottom!’ (Paul, NSW, 2016). In the report, this ‘magic number’ – the final monetary value that is ascribed to the property – is accompanied by ‘contextualising information’, which explains – or legitimises – how the number came about and how it should be interpreted in the light of the developments in the land market and other relevant markets.

As demonstrated so far, when valuing land, valuers undertake various operations for classifying, measuring and comparing farming properties, for which they use different devices, programs and databases. As access to information is key within the valuation, various forms of databases are an important component of the infrastructure of value. As Lammer and Thiemann argue (see introduction to this special issue), infrastructures shape value by relating particular actions or beings to larger wholes. Valuers use public and private databases that provide prices paid for properties, as well as the latest information on water and livestock markets, farm equipment prices and rural real estate. Additionally, valuers build their own databases where they store information from their previous valuations along with further information they collect both online and offline. As Paul explained, some information disappears from the internet after a while, so he prefers to save it for himself. He also collects material from local newspapers as some of the agents in his area were ‘really old’ and only printed their ads in the local newspaper (Paul, NSW, 2016). Thus, although valuers increasingly rely on digital technologies, ‘analogue’ elements such as pen and paper or local newspaper are still used in case the digital technology breaks down or to allow for the greater flexibility and independence these analogue components entail.

While all these databases and devices are important components of the infrastructure of value, they are not sufficient to develop a final value judgement. Value judgements also need to be grounded in an ‘accurate’ understanding of the social embeddings of prices and markets to become legitimate and justifiable, as the next section argues.

Controlling for the Social Components Within the Infrastructure of Value

The databases and devices the valuers use provide numbers, but within the valuer profession these numbers cannot be taken at ‘face value’. As the valuers are working towards providing their clients with a ‘waterproof’ value number to facilitate transactions, tracing the social context of these numbers and ‘controlling for the social’ becomes an essential part of their valuation work. Social relationships and people – the valuers’ networks of buyers and sellers, real estate agents and transaction managers, as well as the valuers themselves – are thus a critical component within the infrastructure of value (on people as infrastructure see also Lammer, this issue).

First, valuers need to make sure that the denoted price found in the database is indeed a ‘relevant’ price for determining the property’s market value as it is understood in economic theory, that is to say a market price where two unrelated parties came to an agreement on a price as part of an ‘arm’s length’ transaction. With the diffusion of the ‘efficient market hypothesis’ in economic theory, market prices have come to be seen as ‘the best estimation of the true worth of goods’ (Chiapello 2015: 19). Conceptualised as reflecting the largest number of expectations regarding the future, market prices are considered ‘superior’ to other forms of value calculation. Hence, valuers reach out to their networks and local contacts to find out whether the involved parties had a pre-existing relationship with each other, and if so, what this entailed. For example, land sales between neighbours are considered carefully as they are often directly arranged between the two parties without going on the market. Sales between family members are generally considered as ‘non-relevant sales’ for determining value:

‘We just won’t use those because we don’t know what’s behind it. Sometimes, they give it to them cheap, or sometimes it’ll say a figure, but there might be something in behind what they gave them, like animals – so we just leave those alone.’ (Steve, NSW, 2019)

This quote shows that – in line with their training in orthodox economic theory – valuers see only ‘pure’ market transactions as suitable for their value judgements, while personal relatedness and kinship relationships appear as an incomprehensible black box disguising the ‘fair market value’. Further to identifying who was involved in a land transaction, valuers talk to the different parties involved in the deal. For example, valuers try to find out if the land was sold under financial pressure, because of retirement, or to buy property elsewhere. Also, real estate agents and transaction managers provide background information on the sales process, as Steve explains:

‘[W]e know a lot of agents around, and they know us. So we ring them up. [A]nd then they’ll talk about what people that were trying to buy it were thinking. We’ll say, “Was there a lot of interest in it?” And they’ll say, “Yes, there was,” or, “no, there wasn’t. People were thinking this. We had locals looking to buy it, or we had investors looking or big companies,” and they’ll give us an overview of it. And that helps us to work out what we think it’s worth.’ (Steve, NSW, 2019)

Similar to the prices, markets are also closely monitored by the valuers. For example, valuers keep track of the ‘set prices’, that is to say the prices at which farmland is advertised, and compare them with the final prices. The valuers’ market research can further include their own ‘fieldwork’. Auctions are the ideal occasion for this kind of market research, as they are considered the ‘best’ indicator of the market value on the day. However, also here understanding the social dynamics involved is crucial, so as not to ‘misjudge’ a price, as Stuart, an experienced valuer in his late 50s who mostly values large properties in NSW, pointed out to me. To illustrate his point, he told me the story of a farmer who failed to buy his neighbour’s property. When the property eventually came back on the market many years later, the farmer went to the auction determined to ‘pay any price’ so as not to miss out on this opportunity again. Thus, Stuart concluded, even an auction price – as the otherwise best indicator of the ‘pure’ market price of the day – can be ‘distorted’.

These examples illustrate the importance of the social insights and individual assessments of the valuers within the infrastructure of value. In an interesting twist, my interviewees often pointed to these social dimensions within the process of valuing, such as Shane, a valuer in his late 40s based in Brisbane, who told me that valuations to a certain extent always remained ‘biased’: ‘It’s all at the end of the day still very much a subjective exercise because it’s our opinion, which is based on our experiences and our own personal insights’ (Shane, QLD, 2016). At the same time, all this research on the social components is done with the intention of ‘peeling off’ the social to identify the ‘true value’ defined in their economic textbooks, as Stuart’s auction story illuminated. Valuers, it seems, live an inherent contradiction: they need to dig deep into the social side of value productions to formulate justifiable value judgements, which they recognise as having some degree of subjectivity – but they also need to strip off the social again in order to endow their judgement with legitimacy. Gaining this legitimacy through ‘objectivity’ is crucial for value to fulfil its infrastructuring function of enabling people’s participation in exchange processes. As valuer Steve told me: ‘If we’re out by a certain degree, it can be bad, whether it’s low or high’ (Steve, NSW, 2019). If the valuer’s judgement placed the value too high, and the buyer received a loan from the bank based on that value, they might struggle to service the debt, or receive their money back if they had to sell. If the value was too low, it equally affected how much they could borrow and whether they could buy the property in the first place. Given this critical facilitating function of value as infrastructure – and valuers as the actors performing this infrastructuring – value judgments can be contested: ‘We get some irate people on the phone or talking to us about where we’ve ended up with a figure […] so yeah, it’s very contentious. We get some very upset people’ (Steve, NSW, 2019). The established methodology and valuation procedures and the databases, measuring devices, social networks and relationships not only enable valuations technically and practically, they also help to justify, legitimise and institutionalise the potentially inconvenient or contested value judgements of valuers. In this way, the infrastructure of value provides value as infrastructure with the stability, durability and legitimacy it needs in order to work as a facilitator in a context of multiple actors, all of whom have strong interests at stake. This stability, however, cannot be taken for granted, as infrastructures of value are not insusceptible to change. If new powerful actors emerge, who bring new interests and rationales to the valued object, established value judgements and procedures can become contested.

Making Farmland Commensurable: Changing Infrastructures of Value in the Context of Financialisation

Asset appreciation is a key goal of financial activity – a goal, which also applies to the financial investors who have purchased Australian farmland over the past 15 years. As demonstrated in the previous section, the appreciation of Australian farmland currently depends on the intricate relationship between the market prices people paid for farmland and the value ascribed to it by valuers, based on their assessment of these prices. How does the entrance of financial actors – who rely on farmland’s appreciation to fulfil their investment mandate – and the financial mind-set they bring with them, influence and potentially alter, the infrastructure of value? This section gives three examples that illustrate how the infrastructure of value is destabilised and renegotiated in the context of financialisation.

Paying ‘Above Market Value’

Land markets in Australia have historically been influenced by climatic conditions (especially droughts), the establishment of new transportation infrastructures (e.g. roads and airports), the development of global commodity markets, the introduction of new crops, as well as movements in the water market (since water and land can be traded independently). Since the late 2000s, these dynamics have been combined with the (re)emergence of financial actors as a particularly cash-rich group of actors purchasing farmland (Larder et al. 2018). The effects of these actors on the farmland market are widely felt throughout the sector, as they are reshaping the ‘relevant whole’ of farmland transactions by adding an additional layer of global comparison to the previously mostly local or regional assessments of land value.

One important perception of financial actors that was shared by many interviewees throughout my research is that they are more likely to pay ‘over market value’. Paul’s account of a financial investor who ‘knocked on the door [of a farmer] and said, “We want to buy it [the land],” and made the guy a stupid offer, like crazy offer. It was an AUD 10 million offer’ is a typical, often heard story in this regard (Paul, NSW, 2016). Compared to other sales in his district, Paul finds that financial actors are ‘paying way above market for everything. Normal sales, people aren’t paying that money’ (Paul, NSW, 2016). Paul’s assessment of the prices that financial actors (allegedly) pay shows that there is a friction within the infrastructure of value, namely the way he learned to formulate his value judgement by methodologically grounding it in the market prices, and his perception of the prices paid by ‘external actors’ as being ‘over market value’ given their mismatch with the prices paid by ‘internal’ actors. While according to the valuation methodology these higher prices should simply constitute a higher value, his value judgement is grounded in ‘local prices’ and has not (yet) been adapted to the new international situation.

While valuers such as Paul scale their comparison of prices at the local and regional, or at most national levels, ‘external’ financial investors work with an international or global scale. From their perspective, Australia is a highly profitable, low-risk investment context, where farmland is still ‘undervalued’. Especially for the large financial players investing on behalf of pension funds, the ‘investable universe’ in agriculture is quite small, as one informant told me:

‘If you’re a Dutch pension fund you can’t own land in Canada. You can’t own land in China. It’s probably too risky to own land in the Ukraine and those places. Africa is 54 different countries, but a lot of political instability. So, they don’t like Africa as a rule, risk being their first filter. [S]o then they start to look at where in the world can you invest in these things? It really comes down to North America, Australia, and New Zealand as the sort of investable universe.’ (Farmland investor, NSW, 2016)

Thus, within increasingly globalised and financialised farmland markets – which are also shaped by different legal infrastructures that determine access to land and what thus enters into valuation practices as comparable land – actors, competing for properties, operate within entirely different investment rationales and scales, which inform their value assessments and willingness to pay. As a result, there is an increasing ‘value gap’ between ‘external’ and ‘internal’ assessments of land value, leading to frictions and incompatibilities with the existing infrastructure of farmland valuation. Within the newly emerging infrastructure of value, it is not the local (or even national) community anymore that sets the price for land, but land value assessments of actors taking on a ‘global’ gaze.

Changing the Valuation Methodology

Financial investors, moreover, look at agriculture by using the same standard ‘return on capital’ investment indicators they apply to other sectors (Lane 2016: 1). In other words, land’s value needs to be (made) commensurable with other financial assets, which beyond a data-based, numeric representation of the farm also concerns the way in which the value of farmland is conceptualised. As several of my interviewees reported, there is a strong push from the corporate sector to change the farmland valuation methodology from the comparison and summation approach to the income-based approach, which is a much more common valuation method in the realm of financial investment. In its most basic sense, the income approach ‘is based on the broadly interpreted principle of anticipation: a property’s value is a today’s value of income it is anticipated to generate in the future’ (Kucharska-Stasiak 2019: 68). Thus, the temporality of value changes. Rather than grounded in past market prices, within the income approach value lies in the asset’s capacity to generate a future income for an investor, with the challenge of calculating the present worth of this future income. Often this is done by discounting the future because the future is considered as distant and thus involves opportunity costs and uncertain outcomes (Doganova 2018; Muniesa & Doganova 2020; see also Fairbairn 2020: 55–57). Below, I argue that changing the valuation methodology for farmland is both a conceptual issue and calculatory challenge, as well as reflecting financial interests and contestations over who holds power within the valuing profession.

Historically, the income approach was rarely, if ever, used to value Australian farmland (Lane 2016: 3), and there are a number of reasons why. First, instead of relying on market prices and their interpretation, the income approach requires the valuer to make assumptions on the likely profitability of the farm, which includes numerous decisions on the farm management the valuers are not necessarily in the position to make. A further challenge is the lack of historical farm data to sustain these assumptions, as this data is either not kept in the first place, or not made available to the valuer (Stuart, NSW, 2019). As Lane writes, in the best case valuers ‘will get copies of stock books to understand the stocking densities relative to seasons, maybe a property plan, a summary of capital expenditure in recent periods and some rainfall data’ (Lane 2016: 3).

Second, farm incomes depend on highly variable factors grounded in the specific materiality of farmland and farming, such as climatic conditions, pests, or diseases, which further combine with global commodity prices and the substantial gaps between farm incomes depending on farm size and type of management (Lane 2016). Here, the challenge is not only how to include these highly variable factors in an income-based approach to valuing farmland, but also how much they are seen as affecting the value of land. In other words, is land’s value seen in relation to the income it produced historically, or is anticipated to produce in the future – or is the materiality of land itself seen as having (a certain) value independent of these factors? Lane (2016: 5), for example, points to this materiality and suggests that land’s value does not disappear, even if its average return over a certain period of time is a loss due to climatic circumstances, as there is still productive land when it rains.

Despite these compatibility issues, the methodology for the valuation of farmland in Australia is currently being renegotiated as financial investors come with their own assumptions as to how value should be ‘measured’, and what kind of indicators they need to make land fit within their portfolios. However, the methodology is not just a conceptual and ‘calculatory’ issue, it also involves concrete financial and professional interests. Given their investment mandate, financial actors have a sustained interest in raising land values and, as Stuart indicated, might consider the income approach as more conducive to this goal than the comparison and summation approach (Stuart, NSW, 2019). While being sceptical as to whether the income approach will actually lead to higher land values, he and others see the push for the methodology change as motivated by the finance industry’s reliance on, and pressure for, continuously increasing land values. Supporting this point, one interviewee told me about a conversation he had with a representative of a farmland investment company who he sees is ‘fighting a battle’ to change the way farms are valued: ‘[I] said why are you fighting this battle? You guys are trying to change the whole industry. He said: “It’s only because my board keeps going on about it because they haven’t had revaluations that look good”. So, their portfolio is not going up in value so they look to try and value it a different way’ (NSW, 2016).

The change of valuation methodology, moreover, raises professional issues in terms of who is considered a ‘legitimate’ valuer. Usually, valuations were made by companies that exclusively specialised on valuations, and therefore considered themselves ‘independent’. In recent years, global real estate companies, which have taken over a significant portion of large-scale land transactions in Australia, have started to offer valuations as well. Given their mostly corporate clientele, and their US American background in terms of valuation techniques, these companies are more inclined to use the income approach. The ‘traditional’ valuers observe this development critically, as they see a conflict of interest, if land transactions and valuations are made within the same company.

In sum, this section has revealed how financial actors bring with them new ways of ‘measuring’ value, which, however, produce some friction with the understandings of value within the existing infrastructure of value and the materialities of farming. At the same time, the different elements within the infrastructures of value also mutually influence one another (see also Thiemann, this issue), and farmland’s materiality itself is adapted to fit the new valuation methodology, as the next section argues.

Adapting Land’s Materiality to the new Valuation Methodology

Taking inspiration from the work of William Cronon and James Scott, valuation practices not only ‘measure’ what is ‘there’, they are also productive as they (make actors) both ‘re-view’ and alter the materiality of the land itself (see also Lammer, this issue).⁵ Cronon (1991: 36), for example, argues that perceptions of woodlands through the lens of notions such as ‘timber’ already point to its prospective usage as resources for urban growth, while Scott (1998: 11–22) shows that tools of simplification of scientific forestry not only measured, but actually transformed forests. In a similar vein, financialised infrastructures of value are not independent of the entity they evaluate. In order to function, the entity itself needs to be made compatible with the (new) measurements conducted. The example below illustrates this point.

The ‘investable universe’ for financial investment in agriculture is currently not only limited on a global scale (as shown above), it is also limited within Australia, as many properties do not fulfil the criteria that financial actors are looking for. Hugh, who has a background in agricultural science, agronomy and financial investment and work experience with family offices and institutional fund management, identified this ‘gap’ in the early 2010s and turned it into his ‘business niche’. At the time, he saw that financial investors had raised capital and were ready to ‘spend a bucket of money’, but needed specific material conditions to deploy this capital:

‘A lot of those funds come with a very specific mandate. So, they can buy orchards, but they cannot go and build orchards. They can’t do greenfield because that’s got a different risk profile. And we’ve had them say to us, “Well, the minute you put a tree in the ground and run a drip tube past it, we can buy that.” It’s an immature orchard. There’s no development risk. Now they can. But they can’t go back 12 months or 18 months earlier or two and buy a bare paddock and put the infrastructure in. So, well, we’re good at that bit. We know our way around it. We’ve got quite a fair bit of practice now. Let’s build them.’ (Hugh, Victoria, 2019)

By addressing the gap of ‘investment-grade assets’ in Australia, Hugh capitalises on the combined benefit from – as he puts it – converting ‘low value’ properties into ‘high value’ properties and the ‘huge scarcity premium’ that financial investors are willing to pay for these rare ‘investment ready’ properties (Hugh, Victoria, 2019). From an infrastructure of value lens, however, it can be argued that rather than converting ‘low value’ into ‘high value’ land, the land simply had no value for financial investors before, because it was not compatible with their valuation methodology. In this sense, Hugh’s infrastructuring transforms the materiality of the land to create an ‘investment-grade asset’ that has a number of material features, such as quality natural resources, commodity specialisation, appreciable scale and infrastructural equipment. Importantly, this infrastructuring also includes putting in a digital infrastructure of data capture systems, which produce the kind of data that make land’s value legible for financial investors.

In sum, Hugh’s strategy stands for a further dimension in which infrastructures of value are changing in the context of financialisation. Land’s value is not only assessed from a ‘global’ gaze at investment contexts and in regard to its commensurability with other financial assets; the financial valuation methodology also transforms land’s materiality – from the kind of produce grown on the land to its infrastructural equipment – to make it accessible within the new infrastructure of value. Together with valuation methodologies, databases, devices and so on, the land’s material layout and related infrastructures are essential for turning farmland into a financial asset. Without the notion of infrastructures of value, these material aspects could easily be overlooked, not only in public understandings of the financial sphere as supposedly divorced from ‘the real economy’, but also in the research on the financialisation of farming, which only recently started to address the material components and hurdles involved in this process (see Böhme 2021; Fairbairn 2020; Visser 2021).

Conclusion

This article examined the practices of valuing land in Australia, and specifically addressed how these valuation practices change as land is financialised. I demonstrated how financial investors assess land’s value based on their own assumptions about how value should be ‘measured’, with the aim of rendering land a financial asset commensurable within their portfolios. This notably involved a shift to a ‘future income’ based assessment of land’s value, which resulted in a number of frictions and compatibility issues with established land valuation practices. In particular, financial actors have challenged the existing infrastructure of value in three ways: (1) in terms of spatial and temporal scales, as they have added an additional layer of global comparison to the assessment of land value and shifted the temporality of land’s value from being grounded in the past to lying in an anticipated future; (2) in terms of scope, as they have added land’s commensurability with other financial assets to land’s value assessments and (3) in terms of land’s materiality, as the financialised infrastructure of value requires not only a different kind of data, but also a material transformation of the land.

Approaching this topic through the combined lens of valuation and infrastructure studies, the notion of ‘infrastructures of value’ has proven valuable to analyse the various conceptual, methodological, technological, practical, social and material components, which endow land with stable, institutionalised and legitimate value. What is more, this lens has been critical to detect infrastructures of value as crucial for both enabling as well as limiting change in valuation practices, thereby pointing to the facilitating, stabilising and legitimising function of the infrastructure of value. This facilitating, stabilising and legitimising function, I argued, is required for the concept of value itself to work as yet another form of infrastructure, namely as a prerequisite and indispensable component of exchange processes – albeit embedded within institutions and power relations.

Emerging out of my ethnographic case,⁶ I suggest that concepts and understandings of what value is, and how it should be measured, are in themselves an important form of infrastructure as they work like a ‘conduit’ between the different parties involved in exchange processes – and who without this agreement on value would have no foundation to come together. Understood in this way, value is intricately linked with the practices of valuation, and their embedding in infrastructures of valuation. Value’s infrastructuring significantly depends on the stable and institutionalised working of the infrastructures of valuation, and can fail if these infrastructures become destabilised, challenged and contested.

I revealed this intricate relationship between value as infrastructure and the infrastructures of value by demonstrating how methodologies and valuation procedures, databases and measuring devices, as well as networks and relationships not only enable valuations technically and practically, but also help to justify, legitimise and institutionalise the potentially inconvenient or contested value judgements of valuers. In this way, the value infrastructure centralises valuations and provides these value judgements with the durability and legitimacy they need to fulfil their infrastructuring function. However, I also revealed how this infrastructuring function of value is impeded, if infrastructures of value become destabilised. In the case presented here, this destabilising occurred as financial actors emerged, who brought new financial interests and rationales to land, which has led to frictions and incompatibilities with the previously established infrastructure of value. ‘Thinking with infrastructures’ (see Lammer and Thiemann, this issue) thus provides us with a helpful analytical lens to better grasp the intricate relationship between various understandings and concepts of value, and how these are being affected and change valuation practices within the context of financialisation.

Acknowledgements

Many thanks to the guest editors of this special issue for their very thorough reading of this paper and helpful suggestions on sharpening its argument. Thank you also to Atakan Büke, Cynthia Gharios, Moritz Dolinga, Nicolette Larder, Juan Ignacio Staricco, and Tim Weldon, who provided valuable comments and support at different stages of the paper, as well as Esther Riley for her editorial work.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The research presented in this paper has been supported by three research grants at the universities of Leipzig, Queensland, and New England (DFG-funded research project C04 of the SFB 1199; ARC-funded Discovery Grant DP 160101318; and DAAD funded Australia-Germany Joint Research Cooperation Scheme).

Notes

1 These markets are Argentina, Australia, Brazil, Canada, Denmark, France, Germany, Hungary, Ireland, New Zealand, Poland, Romania, United Kingdom, United States and Uruguay (Savills 2019).

2 This ‘assetisation’ of farmland (Ducastel & Anseeuw 2017; Ouma 2020; Visser 2017) is part of a larger ‘financialisation’ of agriculture and food (cf. Gertel & Sippel 2016; Bjørkhaug et al. 2018; Fairbairn 2020).

3 On the dialectic between uniqueness and genericness in evidencing terroir see Ana, this issue.

4 All names used for interviewees in this paper are pseudonyms.

5 Like quantitative measurements, qualitative evaluations of ‘alternative’ products (e.g. ‘ecological rice’ in China) also require material changes such as shifting to certain crops and transforming landscapes for eco-tourism (see Lammer, this issue).

6 This argument of ‘value as infrastructure’ emerges from my specific ethnographic material. By suggesting this understanding of value, I do not mean to say that there cannot be situations in which infrastructures are arranged around something else than value accumulation, or that nothing else than agreements on economic value can bring people together (e.g. to redistribute or reciprocate land or other things).