 Hi everyone, thank you for coming. What I will present today is a paper that is about the trade-off between forest preservation and development. I will be very specific about what I want to talk about today. I will talk about deforestation that is driven by some agricultural land expansion. The relationship between deforestation and climate change is that deforestation is responsible for roughly 20% of the global greenhouse gas emissions. It constitutes the main source of emissions in some countries. Then in the international conferences about climate change, some countries have agreed that they could reduce their emissions only if they are compensated for the opportunity cost of doing so. This has created a lot of debate about the red mechanism, or the red plus. I will not really enter into all the debate about this mechanism and how to implement it, etc. What I want to focus on today is just about this notion of opportunity costs. What does it mean to reduce deforestation? The opportunity costs of reducing deforestation, are they just in the agricultural sector? Or is it more a global question about development? To be precise, what I will try to present is two possible versions of the opportunity costs of reducing deforestation. One is thinking of an economy that is purely agricultural and then if you reduce the amount of land that is available, it would imply that some of the rural labour force will not be used efficiently. But you can also think about a more diversified economy and then reducing deforestation and reducing the amount of land that is available would force some of the rural labour force to move from the agricultural sector towards different sectors that are potentially initially less profitable. And to present these two versions of the opportunity costs, I will use a model where deforestation is induced by trade and what I mean exactly is that the country has a comparative advantage in the agricultural sector and this is why it needs more land. And each farmer allocates their labour force efficiently between two activities. One is production and the other is land maintenance and land clearing. And then I build a very simple two sector growth model where there are two sectors, agriculture and industry and each sector has a specific factor. So land is a specific factor for agriculture capital is a specific factor for industry and both can be accumulated. And what I want to show you is that there are two ways to define the opportunity cost of reducing deforestation. One is what I will call the untied mechanism. So that's a transfer mechanism that would cover the opportunity cost in an economy that is mostly specialised in agriculture and then reducing the amount of land available which is reduced the amount of land per worker, per agricultural worker. Whereas the other possibility is to think about a tied mechanism and when I say tied it means tied to diversification. And in this case the objective is to shift some of the rural workers towards different sectors. So this is the presentation of the model briefly. So I have two stock of land. One is forested land and the other is agricultural land. And deforestation is just a conversion from one stock to the other. So NT is my agricultural land. And at the individual level each farmer decide how they allocate their labour between production and here you can see, well maybe you won't see very well but it's simply a cob-de-glace function where you use your land and some labour LA in order to produce and both inputs are necessary. And you can also use your labour to increase the amount of land that you are using. And this is the land dynamics at the individual level. So each farmer can increase its amount of land to a point where it is too big for him just to maintain this land. And we assume that they do not optimise their return intertemporary for the decision of allocation of labour. So each farmer can deforese basically until they reach this kind of optimal size for an agricultural lot and star. And this can be added so that at the aggregate level we have LT number of farmers and they use the amount of land that is at the aggregate level NT. And you can see that the aggregate level of land depends on the size of the rural labour force and the production function is relatively simple, it's linear in the size of the agricultural land. For the manufacturing sector it's a very standard way of representing it. So I just have capital and labour and capital can be accumulated through investment. So what I want to show you is that the context is deforestation that is driven by a comparative advantage in agriculture. So if you play a little bit with the model you would see that basically there are some specialisation patterns that would occur. And when you open to trade you will either specialise in the agricultural sector or in the industrial sector depending on the comparison between the world price P and what was the relative price that was revealed in the Otake case, PD. So if the world price is high, which is my first dot in the proposition, then the economy will move towards a purely agricultural economy. And I want to focus on this case because I think it's the most extreme case. So imagine that deforestation is not due to any market failure, it is really just following the dynamic of development of the country because the country has a comparative advantage in agriculture. And let's see how we can force or induce the country to limit this deforestation process without reducing the development potential of the economy. So this is how the transfer is important. So I focus on this case where the world price is high and I introduce a transfer which depends basically on two parameters SOG and SG. What I want to just tell you is that basically I provide a constant level of transfer to the economy, it's SOG. And then if the economy differs more than what was the agreement, there is a penalty. And this is my little SG. So that's basically an instrument with two components. And I will present two different ways to define this set of parameters SOG and SG. Either you consider that the economy is already in this process of specialization and you just consider at the steady state the extreme cases where all the population is already rural. So consider an economy where all the labor force is in the agricultural sector and think how can we save some forest in this context? So that's one way to define the set of parameters. And the other is, oh no, we should not let the economy derive in that direction. We should induce and maintain some diversification by defining the transfer mechanism in a different way, which is my tied mechanism. So this is how I try to define it in the most simple way. So the tied mechanism is willing to shift some of the labor force toward industry. And industry is in this case less profitable because the economy has a comparative advantage in agriculture. And this is a way to describe what is the incentive. So I define it endogenously. And the incentive to stop the foresting will depend on the difference between the actual world price and the world price that ensure diversification, which was my PD. The second mechanism accept the fact that the economy has been completely specialized in agriculture and then try to preserve some forest by making some of the labor force easel or less efficiently used. And this is what I call the congestion effect. And in this case, basically the farmers are not really allocating their labor between the two activity, which was land clearing versus production. They reduce their maintenance effort and this has also an impact on the output. So basically the opportunity cost is just decreasing the agricultural output here. And the incentive is endogenously defined and here you can see it's a little bit more complex. It depends both on the world price but also on the size. So N hat is the size of the agricultural land that is available for production, given the environmental target. And now what I'm interested in doing is basically comparing the cost of each type of transfer from the point of view of the donors, because there are two different ways of defining the parameters. Maybe one solution would be less costly or maybe more cost efficient than the other. And this depends on how you define the amount of constant transfer that is supposed to be given at each period, SOG. And the proposition says that the comparison between the two mechanisms depend on both on two different variables. It depends at the same time on the relative world price and also on the size of the forest that we want to preserve. So if the world price is high enough to give the country a comparative advantage in agriculture but not too high, so it's limited by the threshold value, then the mechanism that ensures diversification is more cost efficient. So that's the first point. However, if the relative world price is too high, then the mechanisms that induce diversification will be more cost efficient only if you want to preserve a very large amount of forest. If you just have a limited environmental objective, then it is better to let the economy be specialized in agriculture. And then you just have these congestion effects that comes from an inefficient allocation of labor in the agricultural sector. That's just an extension. So in the first part of the model, I consider that the forest has no value for the producers. And now I say, oh, but maybe in many of those developing countries, forest has some value. And one part of the value comes from those non-timber forest products. And if you consider that, then it means that there will be less deforestation in the laissez-faire situation. However, so that's what I say. However, it has some impact on the price at which you will switch to this situation where you specialize in the agricultural sector. And what is important for our comparison between the two preservation schemes is that the result that I've showed just before are basically preserved. It's just that the policies threshold are modified by introducing non-timber forest products. So just a little bit of discussion. Of course, it's a very rough presentation of the problem. I just want to, okay, I will go faster. There is a problem of enforcement here in when you want to decentralize the equilibrium. And maybe if you have a more decentralized economy, there will be less need for enforcement. So that's one potential benefit that is not taken into account in the model. And then there is a problem of durability. You need maybe some spillovers in the industrial sector so that people would remain in the industrial sector even when the transfer ends potentially. So that's a concluding remark. I think I've already said that. Thank you.