Previous PageTable Of ContentsNext Page

Managing in the face of uncertainty: is there anything farmers can learn from finance?

Ben Jones

Mallee Focus, Box 2 Manangatang, Vic 3546. Email:
Birchip Cropping Group, Box 85 Birchip, Vic 3483.


Farmers have traditionally used rotations to manage risk, opportunity, and the biophysical requirements for a successful crop. Historically in the Victorian Mallee, a medic pasture and tilled fallow provided nitrogen, a disease and weed break, and stored water for a following cereal crop. In this rotation, financial losses and potential to capture opportunities were both constrained. Recent advances in varieties and agronomy have made Mallee farmers confident about sowing opportunity crops, but without the strong water-driven certainty of production that drives opportunity cropping in other regions. This review examined the management of uncertainty in finance, to see whether decision methods could be usefully applied to managing opportunity cropping in the Mallee. The capital asset pricing model (CAPM) benchmarks investment possibilities according to relative risk/opportunity, and over-/under-pricing. To suit farm management, the benchmarking could be made with respect to growing season rainfall or price. Real options valuation allows for explicit selection between present and future opportunities by working through future consequences and probabilities in a decision tree. Modularity allows parts of the tree to be summarized, and could allow aspects of opportunity crops (eg. future weed control) to be captured in a gross-margin style calculation. Both decision methods could be applied to choices between crop options in opportunity cropping on Mallee farms and the best approach may utilize strengths of both. The challenge for applying any new decision making methods is making them simple enough to be useful, while retaining some of the benefits of any added complexity.


flex cropping, decision support, simplicity


Advances in agronomy and varieties have freed farmers in the semi-arid Victorian Mallee from the constraints of disease and weeds that drove regular crop species rotation and/or use of fallows and pasture. Disease resistant varieties, monitoring and management options including fungicides mean that disease is not the driver it was. Provided herbicide rotation holds resistance at bay, incorporated-by-sowing herbicides, herbicide tolerant crops and integrated weed management practices remove the weed control driver. It is now conceivable to sow practically any sequence of crops, subject to herbicide residues, but the new freedom (a type of opportunity cropping) has also highlighted the risk-management role performed by traditional rotations, which enforced a conservative diversified enterprise strategy on those who followed them. This review aimed to canvas research in finance for new methods of decision that might provide Mallee farmers with ways of trading present and future opportunities, and the consequences that follow if seasons or, less frequently, prices, do not turn out as expected.

Opportunity cropping

Opportunity (flex in US) cropping as practiced in N NSW and Qld is essentially deciding to sow a crop only when a certain amount of moisture has been accumulated in the soil (Pollock et al. 2001). This makes sense where the soil is capable of storing enough moisture to grow a profitable crop with a minimum of follow-up rainfall, where the in-crop and sowing rain is unreliable and often low, and where a range of crops can be successfully sown depending on the time of year at which the amount of moisture has been reached. The process of accumulating water in the soil can be thought of as accumulating ‘certainty’.

In the Mallee, rainfall events are smaller and soil capacity to store water less (and variable). In-crop rainfall is more reliable but winter dominant, and there is not the range of crop types and timings in N NSW and Qld that allow many of the weed and disease issues to be overcome. Opportunity cropping in the Mallee must also be about accumulating a different type of certainty; that weeds, disease and herbicide residues allow a particular crop to be sown and harvested successfully. Given that the crop might not be profitable, the effect of the crop on a possible better opportunity the following year also needs to be considered.

Decision making in opportunity cropping

The major decision driver in opportunity cropping elsewhere is water accumulated, the soil x type of crop that can be sown, and whether it will emerge. Burt and Allison (1963) provided the first example of analysis to derive the decision rule, a major simplification being water at planting (for wheat) not being affected by moisture at planting the previous year (and whether there was a previous crop). Robertson et al. (2000) demonstrated some of the range of factors surrounding the crop choice decision in these systems and the potential for using crop simulation models, along with suitable consideration of factors not easily represented in the models. Exposure to this approach increases understanding, with the understanding being embedded in ‘policies’ which produce the optimal long-term return (Carberry et al. 2002). The short-term result of the policy depends on how well the problem was specified, and how well past probabilities reflect the distribution of future outcomes. Tactical departure from the long-term optimal policy can give greatly increased returns if the farmer gets it right (Kingwell et al. 1993); or the reverse (by implication).


Finance deals with the management of investments (eg. shares). There are some useful rough analogies between the decision challenges faced by investors and farmers (Table 1). The capital asset pricing model and real options valuation stand out as decision making aids from finance that may have application in Mallee opportunity cropping.

Table 1. Similarities between share investment/trading and the decision challenge faced by farmers.

The share investor

The farmer

Allocates a limited amount of capital among a portfolio of shares

Allocates a limited amount of capital, the choice is rather between paddock x crop combinations

Shares have known technical characteristics and history

Paddocks and crops have known technical characteristics and history

Future market performance is uncertain but there are forecasts of varying quality

Future crop performance in a paddock and price outcome is uncertain, but there are weather and price forecasts of varying quality

Can choose not to invest

Can choose not to crop/graze

There are better and worse times to buy and sell

Opportunities to sow/graze paddocks pass

Range of dividend-growth characteristics and sectors

Range of paddock/crop characteristics, immediate benefits (dividend) or to later crops (growth)

Can know more and specialize in better-known opportunities, or diversify to spread risk

Can specialize or diversify

Can outsource investment decisions, buy funds with particular investment policies

Can outsource decision-making, or use a rotation (policy) that fixes some decisions

The capital asset pricing model

The capital asset pricing model (CAPM) describes return on an investment as a function of the market return over time, and was developed as part of modern portfolio theory (for review, and some concerns, Fama and French 2004):

Investment Return = Alpha + Beta x (Market Return)

Beta describes the risk/return relationship between the asset and the market – less risky means a lower Beta, and Alpha describes any abnormal gain or loss due to mispricing of the asset, for example due to momentum selling/buying (depicted in Figure 1a). The Beta can be calculated separately for ‘upside’ (market gain) and ‘downside’ (market loss) risk (Ang et al. 2006) or allowed vary over time (Avramov and Chordia 2006). The optimal portfolio will contain investments with high Alpha, with Beta that suits risk/return preferences, and also (when diversifying to reduce risk) with Betas that are not correlated across the range of market return.

In the farm management application, each paddock/crop/management combination constitutes a potential investment. Each season an assessment of Alpha and Beta could be made to discriminate between better choices and relative risks, and to assemble appropriately risk-diversified ‘portfolios’ of choices given particular expectations of price and seasonal outcome. Rather than an abstract ‘market return’, it may make more sense to compare Alpha and Beta calculated with respect to rainfall (eg. growing season rainfall) or an indicator price. An example of this is given in Figure 1b, where linear regression has been used to estimate Alpha and Beta for wheat grown on two paddocks with different soil types, over a range of growing season rainfall. Both paddocks have similar Beta, but the Alpha is $120 higher for the sand paddock.

Figure 1. Diagrams of the capital asset pricing model in standard form (a), and as an example of enterprise choice, comparing the gross margin of wheat crops simulated on high nitrogen, clay soils with subsoil constraints, or sandy soils, with April-October rainfall in place of market return (b).

Past application of CAPM to the enterprise selection problem targeted a ‘best historical average’ portfolio of crops, rather than a forward-looking one a (Collins and Barry 1986). Historical average performance across crops was used as a ‘market return’, which seems less easy to interpret than rainfall or price, because historical agronomy, and production/rainfall and price relationships do not always apply to future crops.

Real options valuation: pricing future opportunities

Options are the right, but not the obligation, to make a choice at some future time. They are perhaps best known in the form of ‘put’ and ‘call’ contracts traded on exchanges, and much research has focused on methods of valuing them. Concepts from traded options have been applied to decision-making in the form of so-called ‘real options’. Real options research has been directed towards mining, where uncertainty in the resource, rate of extraction and metals prices and the possibility to start and stop production mirror some of the problems in agriculture. Mining industry uptake of real options has been slow, partly because of complex calculations involved (Davis 1998). The advent of simulation methods have provided realistic ways of solving options valuation problems (for example Longstaff and Schwartz 2001), and real options applications have begun to appear in agriculture (Isik et al. 2003; Odening et al. 2005).

Real options with farm management applications close to those aimed for in this review and directed at climate change, are the topic of a recent paper by Hertzler (2007). Hertzler presents the decision process in the form of decision diagrams (a development of decision trees). Decision diagrams summarise the cash flows resulting from each of the range of possible outcomes of a decision, represented as a series of ‘states’. The probability of each outcome path is used to weight cash flows to arrive at a single dollar ‘option value’ for each decision. Ideally different attitudes to risk could be incorporated into the weighting to reflect different risk appetites and attitudes to loss (as in Rasmussen 2003). Although the outcomes are fairly simple values for each possible decision, much of the value to be had may be from the process of constructing the decision tree/diagram and assigning probabilities and values.


This review presents two of the more practical concepts for farm management encountered in finance literature, but there are many more that offer useful insights into farm management issues. The advantage of the CAPM is its ability to summarise the inherent advantage (alpha) and risk/opportunity trade-off (beta) of a particular crop choice (for example) in a few key parameters. The few parameters can easily be related to season and farmer’s changing perceptions of likely seasonal outcome, but may disguise more subtle interactions and non-linearity. Expert knowledge is likely to be required to generate the parameters, which may be difficult to interpret without documentation of the generation process.

Real options also summarise the complexity of possible choices in a few key numbers, but handle non-linear outcomes well, and embed much of the generating process in the decision tree (there is some abstraction in the decision diagram). Some farmers may, however, find it difficult to get a sense of the relationships between decision outcomes and season and other variables (such as within-paddock spatial variation). Adding in some of the longer-term aspects of rotation decisions is also likely to lead to very big decision trees that are difficult to use. Hertzler (2007) recognise this and recommend their use at higher levels (for example policy), with modular ‘summary states’ to make decision diagrams tractable.

There is a clear opportunity to simplify and try to make the best of real options (and decision trees/diagrams) and the CAPM for use on-farm and amongst agronomists. Further research could flesh out some of the modular ‘summary states’ (eg. ‘paddock ready for cereal crop’, ‘grass problem requiring three year grass control’) so that they can easily be added to smaller paddock decision diagrams. Some of the relationships within decision diagrams (particularly with price, season) may benefit from CAPM-style presentation, so that inter-relationships are easier to conceptualise. In the future, a ‘gross margin’ or similar calculation for a particular crop on a paddock may include options values for items such as grass weed control, future foliar disease risk, nitrogen and water use by the crop. This would help to put some objectivity to the current ‘gut-feel’ approach to valuing future costs.


This work was part of the Grains Research and Development Corporation funded “Flexible Farming Systems” project (BWD0008). Thank you to Mallee Sustainable Farming’s “Reaping Rewards” project for APSIM-modeled yield on different soils used in Figure 1b.


Ang A, Chen J, Xing Y (2006) Downside Risk. Rev. Financ. Stud. 19, 1191-1239.

Avramov D, Chordia T (2006) Asset Pricing Models and Financial Market Anomalies. Rev. Financ. Stud. 19, 1001-1040.

Burt OR, Allison JR (1963) Farm Management Decisions with Dynamic Programming. Journal of Farm Economics 45, 121-136.

Carberry PS, Hochman Z, et al. (2002) The FARMSCAPE approach to decision support: farmers', advisers', researchers' monitoring, simulation, communication and performance evaluation. Agricultural Systems 74, 141-177.

Collins RA, Barry PJ (1986) Risk analysis with single-index portfolio models - an application to farm-planning. American Journal of Agricultural Economics 68, 152-161.

Davis GA (1998) Estimating volatility and dividend yield when valuing real options to invest or abandon. The Quarterly Review of Economics and Finance 38, 725-754.

Fama EF, French KR (2004) The capital asset pricing model: Theory and evidence. Journal of Economic Perspectives 18, 25-46.

Hertzler G (2007) Adapting to climate change and managing climate risks by using real options. Australian Journal of Agricultural Research 58, 985-992.

Isik M, Coble KH, Hudson D, House LO (2003) A model of entry-exit decisions and capacity choice under demand uncertainty. Agricultural Economics 28, 215-224.

Kingwell RS, Pannell DJ, Robinson SD (1993) Tactical responses to seasonal conditions in whole-farm planning in Western Australia. Agricultural Economics 8, 211-226.

Longstaff FA, Schwartz ES (2001) Valuing American options by simulation: A simple least-squares approach. Review of Financial Studies 14, 113-147.

Odening M, Musshoff O, Balmann A (2005) Investment decisions in hog finishing: an application of the real options approach. Agricultural Economics 32, 47-60.

Pollock KS, Meinke H, Stone RC (2001) The influence of climate variability on cropping systems in Central Queensland. In '10th Australian Agronomy Conference'. (Australian Society of Agronomy: Hobart, Tasmania).

Rasmussen S (2003) Criteria for optimal production under uncertainty. The state-contingent approach. Australian Journal of Agricultural and Resource Economics 47, 447-476.

Robertson MJ, Carberry PS, Lucy M (2000) Evaluation of a new cropping option using a participatory approach with on-farm monitoring and simulation: a case study of spring-sown mungbeans. Australian Journal of Agricultural Research 51, 1-12.

Previous PageTop Of PageNext Page