Risk management in the momagri model: A decisive step in agricultural modelling ____________________________________________________________________
Among the seven modules in the NAR model, and that no current international agricultural model takes into account, the way in which farmers form opinions about the future and manage risk is one of the modules that holds important consequences for the understanding of agricultural markets.
In most of the models referred to, farmers do not have to solve the problems caused by the existence of risk on agricultural markets: they are supposed to know in advance the average price of a product for the following year (even for the following years if the product is perennial) and they adjust their productions almost perfectly, taking this figure into account, to competitive market balance. Here we catch the main models used in agriculture “red-handed”: their perception of the behaviour of farmers is not only simplistic but above all implausible. And this perception leads to consequences that are manifestly foreign to current observations.
First, the world in which farmers decide on the areas they will use for each product and the levels of production is a world with an uncertain future. Farmers must first estimate future prices, and we all know from experience that whatever precautions we take and the techniques we use, our anticipations are not always perfect. The anticipations of farmers however are not naïve, and the producers of field crops know how to use past observations to make a significant estimate of what will happen the following year. The way they manage the situation varies according to the period:
> when the situation has been stable, farmers remember the observations they have made of the market over a longer period;
> when the period has been marked by major changes, farmers remember the observations they have made over shorter periods.
It is important that these facts are integrated into a model of the behaviour of agricultural production.
We could of course ask ourselves if these anticipations, once they have emerged from the farmers’ thoughts and reflexes, lead to decisions concerning their production the following year – or the following years for productions where maturity is slow – where they would be considered by farmers to be simple fixed prices. After all, this is what the formalization of several existing models in the field implies, even though their writers are not new to the profession.
In fact, writers of these models who have frequent and focused contacts with farmers know that this is not true.
Farmers are, on the contrary, acutely aware of the fact that these anticipations are only averages, and that many other incidences will probably occur to affect these averages. It is important to take this into account: how, otherwise, can the psychological climate in which producers are submerged and which conditions their behaviour, be reflected?
Economic models must not be based on pure mechanisms, as if people were machines. They must also reflect the “morale”, the psychological attitudes of those who make decisions about successive productions. This is what the equations in the NAR model do. They show very precisely the farmers’ attitude to risk.
This aspect of things is even more important because the way farmers manage risk leads in a way to market instability.
As long as a market economy prevails (which of course is desirable) – whatever the structure of these markets – prices will fluctuate. The best we can and must maintain for farmers is limited but effective volatility, based on a model that accurately reflects reality. Promises of imminent price “stability” – stability that is almost complete, but never effective – constantly repeated for the last thirty years by model creators, who have ended up believing that tomorrow the world will slip into the exaggerated simplicity of their models, have proved to be a fantasy. They will no longer convince anyone. And globalization will be rejected until we give farmers good reasons to believe what we tell them.
But we have to go one step further. Do these attitudes to risk convey the same psychological reflexes from period to period?
If, as is usually the case in economic models, the attitude to risk in the NAR model was limited to a simple introduction of a correction for the risk, repeated more or less in the same way from period to period for a given producer, we would have made very limited progress in the deplored “mechanisms” mentioned above. And, all things considered, is this characteristic of the behaviour of farmers?
Observation of the markets and the discussions that the NAR team has had over recent months with various farmers suggest that it is nothing of the sort. The reality is more subtle. The NAR model remains a model – therefore a simplification of reality – and it does not claim to convey all the complexity of the agricultural world, but its aim is at least to reflect the specificities of agriculture compared to other economic activities.
The equations of the NAR model will therefore introduce a new dimension, that of the optimism or pessimism of the agricultural economic players. The NAR model tries to convey the fact that optimism increases after prosperous periods, and decreases after a succession of “bad years” – which we know are more numerous than good years in agriculture – through equations elaborated by our team. To do this, they therefore do not include standard economic risk modelling.
Finally, farmers’ decisions are far from being individual isolated decisions, as representatives of “pure” competition maintain. Contact between farmers exists in all countries in the world, whatever the technology, whatever the standard of living. The opinions and psychological attitudes of decision makers in the agricultural world are at least in part influenced by external opinions, expressed within various organizations, cooperatives, unions or those giving technical advice etc. Recent discussions outside the general assembly of a federation of cooperatives, reinforces our conviction on this point. Here again, we will try to convey this state of affairs in our equations.
No, the NAR model will not be just another model, but a real decision-making tool, which will enable international decision-makers and negotiators to make decisions on a more satisfying, credible basis that reflects reality.
Chief Economist of WOAgri