339. Assuming that farmers maximise profit

Agricultural economists building models of agricultural production often use simplistic assumptions about what motivates farmers. The simplest possible assumption — that farmers maximise profits — is still quite commonly made in agricultural economic models. Can this ever be defended? 

One of the papers in the recent AEPP Special Issue of papers on adoption of agricultural innovations is on this topic. The paper is by Alfons Weersink from the University of Guelph and Murray Fulton from the University of Saskatchewan. You can hear my interview with Alfons about the paper here. My conversation with Alfons is available as an episode of the AEPP podcast series.

They point out that many motivations other than profit maximisation have been identified in research about the uptake of new practices by farmers. Other factors that influence them can include risk, leisure, family, social norms, peer pressure, environmental concerns, and altruism.

Weersink and Fulton also picked up a table from an old Pannell Discussion (PD103) in which I suggested that the factors that influence adoption of new practices are likely to vary in different stages of the adoption process. Early on in the process, when farmers are learning from others about the practice, social, cultural and personal factors are likely to play a relatively large role. Later in the process, when farmers have personal experience in using the practice on their own farm, the influence of outsiders will be reduced to some extent. Influences like social norms could still play a role, but the farmer is no longer primarily relying on outsiders for their information.

The finding that farmers are more complex than we might like to assume is reinforced by research in the booming  field of behavioural economics. This has identified a large number of biases that affect people’s decision making. Some relate to the way our brains process information, and some are more about our responses to social situations. This work tells us that farmers will often diverge from profit-maximising behaviour even if they think that’s what they are going for.

Having read the Weersink and Fulton paper a few times (in the course of editing the special issue), it got me thinking about the role of profit in adoption of innovations. Here are some further thoughts.

I think that the main thing driving long-term adoption of a practice is its relative advantage (how well it enhances the achievement of the farmer’s goals). An important element (but, as I said, not the only element) of relative advantage for most farmers is the effect of the practice on profit. There is plenty of evidence that, although various factors influence farmers’ decisions, profit remains one of the main ones. It doesn’t mean that farmers are motivated by wealth for its own sake, but they seek to make money for what it allows them to do.

At the same time, the cognitive biases identified by behavioural economics don’t seem to me to be so large that profit-motivated farmers would make decisions that are really poorly correlated with their objectives.

If I was putting a number on it, I’d say that for Australian farmers, profit explains something like 70% of the variation in adoption between different practices. We can potentially do a more sophisticated analysis to get a more accurate understanding of adoption, but even if we don’t, a 70% understanding is likely to be very useful.

Economists should think carefully about whether some other factor is likely to overwhelm considerations of profit, but in most cases profit at least gives a useful indication of the broad trend in behaviour. If a scientist, an extension agent or a policymaker wants to know whether a new practice or technology is likely to be adopted by farmers, an analysis of its effect on farm profits is bound to be a useful first step.

Why not just include those other motivations in the economic models, to pick up some of the remaining 30%? Because it is much more difficult. The big advantage of modelling profit is that it can relatively easily be quantified, whereas most of the other factors can’t.

So, in summary, my view is that basing agricultural economic models on profit maximisation is often not a crazy thing to do, because it is relatively easy and reasonably useful.

Like all of the papers in the special issue, Alfons and Murray’s paper is free to access.

Further reading

Streletskaya, N.A., S.D. Bell, M. Kecinski, T. Li, S. Banerjee, L.H. Palm-Forster, and D.J. Pannell. 2020. Agriculture Adoption and Behavioral Economics: Bridging the Gap. Applied Economic Perspectives and Policy 42(1), 54-66.

Weersink, A. and M. Fulton. 2020. Limits to Profit Maximization as a Guide to Behavior Change. Applied Economic Perspectives and Policy 42(1), 67-79.

7 Comments

  • 19 May, 2020 - 8:10 am | link

    Hi David

    Mostly true in my view, but the extent to which profit will be important will also depend on whether it is a major strategic change or a relatively minor change. The effect of elements of uncertainty, ambiguity, and risk overlap with that, with a major change being more uncertain & will be more likely to affect the viability of the farm. My article from a while ago is relevant here:
    Murray-Prior R & Wright V 2001, ‘Influence of strategies and heuristics on farmers’ response to change under uncertainty’, Australian Journal of Agricultural and Resource Economics, vol. 45, no. 4, pp. 573-598.
    Sometimes people will have very biased views of profitability also, which are hard to overcome.
    However for some decisions (e.g. annual wool sales) it is perception of price and profitability that can sometimes be – I can’t think of another word but wrong – that will affect their decisions. Reminds me of some Murdoch press commentators beliefs about global warming. See Fig 3 in:
    Murray-Prior R & Wright V 2004, ‘Use of strategies and decision rules by Australian wool producers to manage uncertainty’, Australian Farm Business Management Journal, vol. 1, no. 1, pp. 56-71.
    Therefore it is not what we think is profitable, but what each farmer perceives as profitable that is the key issue.

    • 19 May, 2020 - 5:31 pm | link

      Thanks Roy. I agree with all that. In some work that I never published we got some price predictions for wool from a sample of growers. It was striking how variable they were. Some were obviously not sensible, but most of the variation just reflected high uncertainty about markets. I guess that when we are predicting adoption, we should remember that there is an opportunity for farmers to learn over time, so the wrong ones won’t necessarily stay wrong forever. (But they might!)

      Yes, of course, farmers’ decisions are based on their own perceptions about the relevant variables, not what is included in some economic model somewhere else. This is true for all relevant factors that would influence adoption, not just those related to profit.

      But I still hold that the sort of economic models I’m talking about here are still pretty useful for predicting broad adoption trends and identifying some of the factors that might inhibit adoption.

  • Thilak Mallawaarachchi
    19 May, 2020 - 9:05 am | link

    Hi Dave,
    We are all crazy. But, thinking more deeply, our models are not just concerning profit maximisation. In our models, in particular optimisation models, we have constraints that account for those other motives. And our cost functions include the cost of adhering to those other concerns that you mentioned. Then wouldn’t our profit in the objective function be the residual? For me, the insight I try to gain from those models is about the opportunity cost of meeting those other important constraints that addsto our utility in different ways.

    As you always inspire me, it is all about getting the message across.
    Cheers
    Thilak

    • 19 May, 2020 - 5:24 pm | link

      Thanks Thilak.

      I’m glad you made this point. Yes, it’s true that it is possible to build contraints into an optimisation model as a simple way of representing an objective that is not represented explicitly in the objective function of the model. And, yes, many models that have been set up to measure profit have also been used to measure the financial cost of meeting other objectives. This is another reason by building an economic model can be more useful for exploring broader issues than non-economists may realise.

      I was thinking about a different use of such models: to predict adoption behaviour. Putting a constraint on the model (e.g. to limit environmental damage) is not such an obvious thing to do in that situation. I guess if you were confident about the consequences of a non-profit objective, you could do it that way, but then that is not a prediction coming from the model, it’s a prediction you’ve made outside the model and built into the model as a constraint.

  • 19 May, 2020 - 10:58 am | link

    This nicely supports Vic Squire’s line;
    Livestock keepers will not undertake change unless;
    1. It is profitable, and
    2. It conforms to a significant extent with their experience and values

    In Rangeland stewardship in Central Asia: Balancing Improved livelihoods, Biodiversity Conservation and Land Protection. Science+Business Dordrecht. Squires V.R 2012

    • 20 May, 2020 - 1:40 pm | link

      Hi David
      As a farmer i am often amused by what people who sit in rooms and talk about farmer behaviour say, present company excluded!
      This morning i was speaking with an esteemed academic about how we judge success in our farming business, (3,500 cattle, team of 10, southern NSW) and made the point that profit (dollars) is important but we also count our social impact highly (having meaningful and interesting employment with a young team, + work experience students from Charles Sturt Uni (local), having the ability to judge risk and adapt in our decision making) and being confident about our environmental hoofprint (and the metrics of how we measure that)
      i think (with no data) that there are changes in farming systems based on gender equality in the decision making. This is after 40 years of working in Ag and having a 50:50 mix in the business.!
      Always enjoy your discussions, thank you.

  • Leave a Reply

    Your email address will not be published. Required fields are marked *

    Please solve this to show you\'re a real person *