A new vision for agriculture
momagri, movement for a world agricultural organization, is a think tank chaired by Christian Pèes.
It brings together, managers from the agricultural world and important people from external perspectives,
such as health, development, strategy and defense. Its objective is to promote regulation
of agricultural markets by creating new evaluation tools, such as economic models and indicators,
and by drawing up proposals for an agricultural and international food policy.
Document sans titre

Testing international agricultural models (1)

The foundations of a new international model

(1) This part is an excerpt translated from a memorandum written by Bertrand Munier, chief economist of momagri and director of the GRID, Nicolas Drouhin, researcher at the GRID and Michel Trometter, head of research at the INRA and associate researcher at the CECO.

Summary :

A. The seven criteria
> Criteria 1 : Reliance on foreign countries
> Criteria 2 : Taking into account climatic and market risks
> Criteria 3 : The effects on poverty
> Criteria 4 : Growth and effects on future generations
> Criteria 5 : Taking innovation into account
> Criteria 6 : The link between the environment and market stability
> Criteria 7 : Sustainable growth and the future of the planet

B. Conclusion
Document sans titre

Calls for the total or partial abolition of agricultural subsidies in Europe, and in industrialized countries in general, are legion. They are often put forward by prominent public figures who use their media clout to support arguments they think they find in simulations produced using highly technical economic models.

The most common theory asserts that the total liberalization of agricultural markets would cause agricultural prices to drop and the standard of living to rise leading to, according to the World Bank, earnings of 340 billion dollars for the world community out of which 240 billion, or 2/3, would go to LDC...

Reactions to these statements are ideological in nature (they condemn “ultra-liberalization” or “globalization”, for example). The risk is that these arguments do not have much weight, or at least not more than any other political stance. These reactions can also be irrational, accusing the models on which the simulations are based of being “false”.

The result in each case is despair on the part of those who are not heard with all the demonstrations that can be seen on the roads or in the streets. There is also despair that is not seen but which produces tangible results like the attitude on the part of the agricultural world towards the French referendum on the 29th May 2005.

This report shows that it is possible to function within a free market economy, to accept globalization, to recognize the need for economic models for international agricultural exchanges without necessarily accepting the results of simulations based on these models at face value.

Indeed, these results are obtained using simplistic hypotheses and an approach that is totally divorced from reality!

For example :


> They consider that fluctuations in demand are totally elastic in relation to prices…that is to say; we would eat twice as much food if agricultural prices were cut by half.

> They do not take into account the preferential agreements that are vital for the LDC.

> They consider that the agricultural sector is totally independent from its environment and exists in a sort of "bubble", uninfluenced by energy and transport costs, health and safety regulations or environmental factors.

They are based on the premise that the economy is experiencing full employment.

> Finally, these models are based on the principle that, according to pure 19th century economic theory, supply adjusts automatically to demand, without taking into account the impact of climatic factors, financial speculation and available stocks.

This means that the models that are used are logically incomplete and that the authors of the assertions mentioned earlier


> either manipulate the models and the resulting simulations to show,
> or lead us to believe that these models show,
> or naively believe themselves that these models show.

what well-informed scientists would not hazard to conclude.

In a nutshell, these models, which nevertheless are the basis for all the technical and political decisions taken within the framework of the WTO, and by extension within the CAP, are crude toys that are a danger for the world because they totally skew our vision of the future.

Examples in other sectors of the economy unfortunately confirm this statement. Indeed, the total liberalization of the textile sector was supposed to be beneficial for Europe. However, this did not take into account a key player, China, a country that was totally ignored by the models that served as the basis of the 1994 Multifiber Agreement.

In the same way the liberalization of the American energy sector, without any regulatory mechanisms, was a real catastrophe for energy security, especially in California.

It is possible to do much better and it is extremely urgent to create a model for international agricultural exchanges. This model, besides promoting economic efficiency (we cannot justify producing less well-being with a given quantity of resources or wasting resources to achieve a certain level of well-being) will also take into account the following essential objectives:


> the war on poverty
> environmental protection and biodiversity
> innovation …

hat is why momagri, in response to this false, flawed and dangerous vision of agricultural reality, has asked its team of economists to “test” the principal international agricultural models against the seven criteria that any international agricultural economic model must take into account.

These seven criteria will govern the construction of a “fair international agricultural model”, the NAR model, which is currently under construction and which is based on a totally new architecture according to the principles of game theory.

Page Header
Document sans titre

A. The seven criteria

These criteria are:

Criteria 1 : Reliance on foreign countries.
Does the model allow us to evaluate the potentially negative effects –including from the point of view of efficiency- of excessive dependency on foreign countries ?

Criteria 2 : Taking into account climatic and market risks.
Does the model take into account the fact that agricultural production and exchanges take place in a world full of potential risks and that this can have an impact on efficiency?

Criteria 3 : The effects on poverty.
Does the model take into account the objective, proclaimed by all, to overcome poverty in the world and measure the effects on poverty of a given organization of international agricultural exchanges ?

Criteria 4 : Growth and effects on future generations.
Does the model take into account the impact on future generations of the organization of agricultural exchanges ?

Criteria 5 : Taking innovation into account.
Does the model evaluate the impact of international agricultural exchanges on the capacity of the food industry to innovate and equal access to innovation throughout the world ?

Criteria 6 : The link between the environment and market stability.
Does the model allow for analysis of the link between the environment and market stability within the economic modeling of international exchanges ?

Criteria 7 : Sustainable growth and the future of the planet.
Does the model establish the link between agricultural exchanges and global changes i.e. environmental protection and biodiversity, each being considered as a shared economic resource ?

Once tested against these seven criteria the principal models used for international decision-making appear to be incomplete in their analysis of the effects of the organizational rules we can propose and therefore do not allow us to reach conclusions about the decisions that need to be made.

As it has been pointed out by Bureau and Gohin (researchers at Dublin’s Trinity College and the INRA respectively) in a recent article “the people who run the (World) Bank do not always have the scientific scruples and the necessary caution when interpreting their results in the political arena”.

Which models were tested?

The WOAgri economists “tested” the models most often used as a reference, i.e. the FAPRI model (Fapri), the OECD model (Aglink), and the World Bank model (Linkage). Two other alternative models have appeared on the scene recently and are starting to be well-known: the Cirad model (ID3) and the CEPII model (Mirage).

However, the FAPRI and OECD models cannot really be used in international negotiations.

The OECD model is in fact centered on the OECD countries and integrates developing countries in only a very limited way. This is normal in that its purpose is to promote a common future scenario for the OECD countries based on agricultural policy scenarios.

FAPRI is a highly detailed partial equilibrium model. The other side of the coin is that this model integrates the developing countries in a marginal manner and many of its hypotheses are exogenous to the model and are totally fabricated. With such limitations its use to analyze the impact of liberalization on different regions of the world is so questionable that the outcome of our test was predictable.

Page Header
Document sans titre



The consequences of the evolution of trade on consumers remain unclear in most models as well as the interactions between productions in different sectors, for example the way downstream production can depend on access to upstream goods. Thus, in the World Bank model uneven competition is not taken Into account. However, there is always a risk that supplies can be blocked by an embargo or a cartel.

The World Bank model supposes that there is a seamless transfer of factors between sectors inside each country which minimizes the impact of excessive dependence on a cartel for food consumption. The example of petroleum in the 70s is significant in relation to this last point, as well as the cartel of “majors” in the energy sector in the United States.

Along the same lines, severe environmental crises that could modify the supply for one or more countries, e.g. the risks linked to bird flu, cannot be taken into account. However, such a disaster would prevent consumption worldwide.

Therefore, if we look at the way Economics and Society are associated, the World Bank model integrates in a way that is unclear a minimum level of subsistence without explaining how this works or how it is financed. Moreover, the integration of the right to food at a reasonable price goes way beyond the concept of a minimum level of subsistence and can concern certain types of goods.

This means that the World Bank model implies that there will always be food available for a given country’s consumers thanks to imports, on a large scale if necessary. But how this is achieved is left to the discretion of the economic modeler.

The World Bank model, when it comes to the questions of dependence and sovereignty in relation to food supply, cannot be used to make recommendations for the organization of agricultural exchanges.

Page Header
Document sans titre



If we look at the different risks that affect production and agricultural exchanges, the information on prices and the nature of goods is imperfect and this is not taken into account by any of the models except the one used by the Cirad.

Moreover, in agriculture the elasticity of demand is limited and this is very important here because any error in anticipation on the part of a large number of producers has massive repercussions on market risks. However, this fact which has been known for two centuries is generally not taken into account by any of the models except the Cirad model. It is indeed necessary to integrate delays between production choices and selling the produce on the market. These market risks can be limited by taking out insurance (the impact of insurance is taken into account by the FAPRI and Cirad models), unlike other market risks.

In the context of risk another tool should be modeled: commodity futures. This could limit speculation bubbles, at least in certain conditions. Nothing of the sort can be found in the models of international institutions like the World Bank.

Climatic risks are presented as negligible in all the models with the exception of major world catastrophes. However, certain local climatic crises can have short-term effects on farmers if there are no insurance or state funds (locusts, drought…) and long-term effects on depopulation, reduced water supply, biodiversity…)

None of the important international models – and especially the World Bank model – take explicitly and completely into account the risks, the way economic agents deal with them, the consequences this behavior can have on the results of the market and the level of well-being that is the result. The Cirad model is a prototype worth pursuing in this area.

Even though the Fapri model integrates certain aspects of the question (but only partially and only related to American behavior) it has the major drawback of being a partial equilibrium model where everything that is not directly related to the agricultural sector is ignored. This makes it a highly unreliable model for carrying out simulations.

Page Header
Document sans titre



The World Bank model considers that there is a “representative consumer” (i.e. all the inhabitants of the same country have the same tastes and the same initial provision) with aggregate revenues. This prevents measuring the effects on poverty of any agricultural policy or organizational measure. It does not integrate the effects of the liberalization of trade on poverty. Even though the authors of the model do stipulate that is possible to reconsider these simplistic hypotheses (and obviously it needs to be done) no one has tried so far.

Therefore, this type of model does not take into account the distribution of global earnings from international exchanges which can, in the end, be carried out to the detriment of the poor. The sharing of earnings depends on the elasticity of agricultural supply in the LDC which in turn depends on the elasticity of substitution of different factors (capital, land, labor) between the two sectors of agriculture (modern and traditional) and between these sectors and industry in the LDC. However, this elasticity, in the World Bank model, is defined in a totally discretionary manner by the economic modeler.

Consequently, the World Bank model cannot be used to assert that there will not be simultaneously an increase in prices worldwide and a decline in the wages of unqualified workers and therefore increased poverty among the poor in underdeveloped countries. If certain simulations produce the opposite result this can only be attributed to the numerical values subjectively assigned to elasticity factors. This point is absolutely decisive. Indeed, if the poor become even poorer this will in turn have  repercussions on rural depopulation, ghettos, crime and emigration towards the developed countries ( not to mention the temptation and despair which can lead to terrorism) etc.

In the same way, to continue with consequences linked directly to the economy, the cycles of poverty that the analysis of endogenous growth has highlighted are becoming more violent (see the following point). In the end, we will once again find ourselves questioning efficiency, an objective which the model is, nevertheless, supposed to help us pursue. Another model, like the one put forward by the Cirad, shows that the fight against poverty can justify (or plead in favor of) a certain form of protection for agriculture. “The stability of basic commodities can be considered a public good given its importance for social peace”.

 Finally, the preferential agreements that allow developing countries to sell their products at prices higher than the international market price are not taken into account by the World Bank model. Taking into account preferential agreements seems only logical considering that out of 148 WTO member states only 9 trade with the EU without a preferential agreement. These agreements, which are more and more numerous, are taken into account by the Cepii model. That is why models like the Cepii and Cirad models really contrast with the World Bank model when it comes to the effects of the liberalization of trade on poverty. And it is interesting to note that these models propose complementary hypotheses that provide a picture that is closer to the reality of economic exchanges.

Page Header
Document sans titre



All the models that, directly or indirectly, like the World Bank model, formulate hypotheses based on a constant rate of savings, are only variations of the first growth models of the 50’s applied to several goods.

However, it is well-known that these models, whatever the values of the parameters they use, and independently of the empirical data used to estimate them, lead to, through theoretical construction, the conclusion that all economies converge towards a unique growth rate. The world dynamic is seen though rose-colored glasses, but maybe the color does not correspond to the real context it is supposed to represent.

It is important to underline here that it is not because a model uses large and serious databases (which is the case for the World Bank) that the conclusions are not the result of choices made by the economic modeler upstream from the empirical study.

On balance, accounting for the construction of human capital and the incentives for this in terms of education, for example, are lacking as well as, consequently, identification of the pitfalls of underdevelopment :


> the “tax bite” due to the fact that poverty creates social costs (fighting crime, etc.)
> the cycle: poverty poor sanitation low investment
> the cycle: poverty fertility insufficient investment underdevelopment

A study based on the models of endogenous growth, i.e. based on the factors that the economic agents bring out or, at any rate, influence by their behavior, seems therefore necessary.

It is thus vital that the model used to predict the effects of liberalization of international agricultural trade be designed in such a way as to identify these effects when they exist. In general this is not the case. In particular all the models that use, like the World Bank, an artifice like the representative consumer theory are incapable, through theoretical construction, to correctly measure this sort of impact.

In this respect, if macroeconomics has progressed dramatically with the first models that include overlapping generations, the macroeconomics of international exchanges is slow in adopting this approach and has not really evolved beyond the general macroeconomics of the 40’s.

So-called “endogenous growth” theories have considerably enriched the problems and tools used in economic growth analysis for the last 20 years.

It is a shame that the integration of these new methods in the study of applied economics, in this case the study of the effects of the liberalization of international trade, is so slow.

Page Header
Document sans titre



Technical change and its generalization have as an objective either to reduce production costs, to create new products, or to adapt practices to new constraints or new standards (environmental, as seen in the next two sections, sanitary…).

However, innovation and change automatically imply Research and Development and R&D in turn imply protection of the innovation. And finally protection of innovation means giving the “owner” a temporary monopoly on the use and dissemination of the innovation.

None of the models studied takes into account the way the profits generated by innovation are shared.

Thus, the World Bank model introduces technical change in its dynamic version by way of production factors. It is, nevertheless, interesting to note this excerpt: Equations G2 to G4 are indexed by ik, these are the sectors with endogenous productivity increases. Normally it excludes agriculture and possibly the natural resources sectors. Nothing could be clearer!

Moreover, as in the model, we have no idea where technical change comes from and who benefits from it (royalties, for example). Therefore the sharing of surpluses created by growth due to technical change is difficult to analyze. And since Brazil is considered an LDC, we can only hazard a guess as to the fair distribution of surpluses created by the liberalization of agricultural exchanges.

In addition, by ignoring the way intellectual property rights are applied the models for international agricultural exchange fail to evaluate the way (fairly or unfairly) that the opportunities for growth are distributed between the countries participating in the exchange. By doing this they contradict the affirmative research of economic efficiency. Because if protectionism is not justifiable in general, because it becomes sustained, it is the same for intellectual property rights. If these lead to exclusion of new competitors in a sector then what is the purpose of these rights other than creating a situation where there is no competition?

Thus, by promoting the liberalization of international agricultural exchanges, the current economic models can lead to a new kind of anticompetitive situation with devastating effects.

It should be noted that none of the models demonstrate this aspect: the sharing of potential growth. None of them can therefore assert that free trade limited to the aspects that they use for economic modeling can, at a given time in the future, permanently lift a certain number of people above the poverty line.

And the variables that could be useful for a better analysis of fair international trade from the point of view of intellectual property (direct investments in foreign countries and preferential agreements) are not taken into account at all by the current models.

The effects of short-term and long-term economic dynamics are systematically underestimated by all the known models because they ignore the legal aspects of innovation in agriculture and the food industry.

Page Header
Document sans titre



When considering the stakes between the Economy and Society, integrating the notion of sustainable growth means integrating a third axis, the environment.

The notion of sustainable growth means specifying the function of cost (in order to internalize the externality that the environment represents) as is the case with general equilibrium models that calculate the greenhouse effect with one of the objectives being to identify the function of the marginal cost of reducing CO2 emissions.

Currently it is interesting to note that it is impossible to take retaliatory action against a country that does not respect its environment if the externalities do not cross over its borders. This means that in terms of environmental policy a country’s choices (or available tools) are limited in that making a polluter pay so that he will pollute less would introduce economic distortions that would penalize his industry in relation to others.

But sustainable growth can also be linked to activities carried out by farmers that are not necessarily related to agricultural production. These can involve maintenance of the local environment and are subject to “payments” that should not be counted as farm production subsidies.

Preferential agreements developed in the Cepii model are rightfully taken into account, whether in terms of price or transfer of technologies. In the Cirad model, on the other hand, the objective is clearly sustainable growth, but without the environmental aspect

None of the models takes into account the impact of the environmental dimension on international exchanges despite the fact that environmental issues are more and more important in international negotiations as can be witnessed in the reform of the CAP in 2003 in which rural development and multifunctionality are key elements.

Page Header
Document sans titre



Within sustainable development there are forms of impact that are directly identified as caused by a source of pollution or deforestation (the impact of soil erosion on productivity). This is local impact but there is also global impact that can be characterized by climatic changes, biodiversity and desertification (each phenomenon having an impact on the others).

The introduction of these global changes will require that every country be able to adapt in such a way that has never even been imagined before for local risks.

This means that, once again, to limit irreversible damage, flexible decisions have to be made based on the evaluation of known or probable risks. Indeed, with global changes we are entering the age of collective public good and its funding. Can we accept that the countries that choose to ignore these risks to the detriment of others receive preferential treatment?

The introduction of global changes in international trade only reinforces the conclusions put forward in the previous section.

The introduction of global changes in international trade only reinforces the conclusions put forward in the previous section. None of the models previously mentioned simulate correctly the impact of direct investments in foreign countries or the preferential agreements concerning the transfer of technology.

Capacity building models should be made in order to try to identify the distribution of gains in growth prospects between the different partners. The World Bank notes that “the use of commercial penalties to enforce the respect of standards (work and environment) threatens the poorest countries on the world market”

Page Header
Document sans titre

B. Conclusion

The World Bank model is the most widely used model. Its simulations are used to support the theory that the immediate and complete suppression of all support for agriculture will increase well-being worldwide and in particular will benefit the poorest populations.

As we have seen, the simulations that rely on this model are far from being sufficiently founded and researched for their conclusions to be accepted without considerable reservations. The assertions published recently in the press are the result of a naive or incorrect use of the model as it exists today.

The seven criteria that we defend in order to help build sustainable agriculture do not allow us, for the time being, to accept the new alternative models without some reservations. Partial equilibrium models like the FAPRI or the OECD models are even less reliable than the World Bank model when it comes to evaluating agricultural policies and in particular the immediate and complete suppression of agricultural subsidies.

The Cepii and Cirad models are a step in the right direction in economic modeling on many points. However, on other points the modeling choices are questionable.

However, none of the models fully pass the test of the seven criteria we have defined based on the observation of a fairly wide consensus. When it comes to intellectual property and innovation the stakes are completely neglected by all the models. In the same way the questions linked to sustainable development are the ones that are the most frequently ignored by all economic modelers.

The introduction of the notion of imperfect information in the Cirad model underlines, on the other hand, the interest of agricultural policies that could reduce the uncertainties that farmers are confronted with. The question of uncertainty and imperfect information seems fundamental to us. Any new model for international exchanges cannot ignore this issue and consequently must be founded on a more general principle of economic equilibrium.

This does not mean we cannot hope for an improvement in economic modeling.  On the contrary, we need urgently to build a new model that allows us to accept the results of simulations as reliable forecasts in which we can have sufficient confidence. 

This is the objective of momagri which has started building the NAR (New Agricultural Regulations) model around the seven criteria and will be able to produce its first results in the second semester of 2006.

Page Header
momagri model
Paris, 20 June 2019