A market-based mechanism to preserve Indonesia's forests

Wed Aug 19 2020

tags: exploration game theory mechanism design economics incentive public


Enormous swathes of virgin forest in Indonesia are being cut and burned down all the time and this contributes significantly to the climate crisis. How can we stop this? I propose a decentralised market-based mechanism that has the following desirable properties:

  • Scalable
  • Decentralised
  • Bounded cost for the philantropist
  • Owners of the land are better off


Norway has some sort of agreement with Indonesia where they pay Indonesia $1 billion to reduce their forest emissions: see links here and here.

The two countries signed the $1 billion pact in 2010, under the REDD+ (reducing emissions from deforestation and forest degradation) mechanism. In exchange for the funding, Indonesia would have to slow its emissions from deforestation, which accounts for the bulk of its CO2 emissions.

This is great and should be applauded. But this sort of enormous government-to-government approach has its problems. First of all, what can we do as private citizens, apart from lobbying our governments? Also, this approach is slow: the first payment was paid out almost ten years after the agreement was signed.

That it’s taken so long for the first payment to be announced is due to a combination of the structuring of the agreement and a change in the Indonesian government since the 2010 signing.

I present a market mechanism that might help safeguard the forests, but doesn't require long and protracted intergovernmental negotations and can work with only a few dollars of funding. The idea is probably not original and I would greatly appreciate if someone could point me to the relevant literature.

The mechanism in brief

In brief, the idea is as follows: philantropists can sell a ticket attached to some forested area that pays out if and only if that area remains pristine. People would bid on the price of this ticket and the highest bidder would stand to gain (Ticket Payout Price - Price Paid for Ticket). With satellite imagery, we can easily tell whether an area has been logged or not, and this would serve as a source of truth.

An explanation of the mechanism

Here's how it would work. Suppose we have a philantropist that wants to preserve some area of forest. And suppose his maximum willingness to pay for this area of forest is $50,000. And now suppose we have some satellite feed of this area (e.g. zoom.earth). Then what the philantropist can do is say the following:

"I will pay $50,000 to the person who holds this ticket if and only if $AREA remains forested by $DATE, as determined by satellite imagery. Otherwise, the $50,000 will be used to fund another similar ticket. [1]

And then he simply puts this ticket up for auction. Once someone wins the bid, the buyer pays the philantropist. The philantropist then puts his funds in escrow, and some trusted third-party determines whether or not the area has remained forested by the date specified in the ticket---whereupon the funds will either be released to the bearer of the ticket, or used to fund a similar future ticket.

How does this work? The buyer has an incentive to hunt down the farmer/owner of the forest and come to some deal to make sure that the farmer/owner doesn't cut down the land. This will probably take the form of a side payment. And if the farmer is savvy enough he can of course purchase the ticket himself and earn free money from doing nothing at all.

This mechanism has several attractive qualities.

Firstly, it's decentralised: there is no government intervention needed. This is also scalable: any person could conceivably sell such a ticket for as small or as large an area as they want, so if any citizen wanted to be a Keeper of the Grove it they could do it today, without having to lobby their governments to do something. All you'd need is a trusted third party to i) hold funds in escrow and ii) verify the forest cover of the area, both of which could be incentivised with some fees.


Why not pay the owners directly? Why the need for this complicated scheme?

How are you going to track down the owners of the land if they live in a forest in the middle of nowhere? Is there a unified land ownership and designation map? Is your Indonesian good enough to find out? How likely is it that that the land ownership map is constantly updated? What happens if ownership has changed hands without your knowledge? What happens if the land is co-owned by several smallholders in some sort of sharecropping arrangement? And so on.

An Indonesian friend of mine wrote to me:

A universal land designation map was initiated in 2014, but it hasn’t come to completion yet because the project has been delayed over and over again.

Rather than having to spend effort to track them down ourselves, the idea is to pay middlemen an "information arbitrage" fee: to provide the service of tracking down the owners of the land. The hope is that these middlemen who buy the ticket will have low-cost ways to contact and convince the farmers.

Why don't we pay NGOs to protect the forest instead?

(Thanks Filip)

Wouldn't this incentivise owners to just sit on the forests and not do anything to collect their payout?

That is the entire point of this mechanism, yes.

Are we worried that if we sell it to the highest bidder, the highest bidder might do unsavoury things?

We might be worried about unintended consequences. For instance, buyers of tickets might start coercing, intimidating, or assassinating farmers to prevent them from cutting down the forest. If so, it might be better to give the money to a good NGO---who will convince/cajole/massage the farmers, rather than to give money to someone unscrupulous---who will murder the owners of the plot---even if the NGO is more expensive.

To this, I have two semi-responses.

The first is that while this might be possible, another possibility is that the NGOs are horrendously inefficient and it might be better to give the money to an amoral corporation instead. Of course, there's no reason why NGOs can't bid on these tickets themselves if they are indeed the most efficient organisation. The other is that killing people is illegal (maybe this will give some people pause). Also, why do you need to kill someone if you can just give them money instead?

Nonetheless, I see the point. Of course when we optimise for price alone there are lots of things that we give up. This is a problem with any sort of market or competitive pricing mechanism, and isn't unique to this proposed mechanism.

What happens if there is collusion amongst the bidders?

The whole system falls apart, but I think any auction system falls apart once there is collusion amongst the buyers. However, if the market is thick enough then collusion would be difficult to sustain. Perhaps there might be some clever way for philantropists to demarcate areas to protect so as to minimise the likelihood of collusion.

Why can't the money go back to the philantropist if the forest burns down?

In a previous version of this mechanism I suggested that the $50,000 would either be released to the buyer, or returned back to the philantropist. I realised that this would incentivise "philantropists" to sell a lot of tickets, then commit arson so as not to pay out to make a profit. Any mechanism must ensure that the philantropist can never gain from this transaction. More strongly, it must ensure that nobody can possibly make any profit if the forest is cut down. (Does this always hold? I am unsure if this holds if people start coming up with complicated derivatives and financial instruments and what not)

A system like this doesn't benefit the poor farmers! Won't the surplus go to some finance bro?

I agree, and while the thought of this is indeed somewhat unsavory, I believe under a competitive market it should be the case that the finance bros will bid each other up until the surplus accrues to the farmers.

Questions I still have

I'd really be interested in knowing whether this mechanism works, and whether it has already been thought of by somebody else under a different name.


Of course, this works with any resource that we want to preserve, not just Indonesia's forests. You can even sell tickets on areas where the land has already been cut down, and pay out iff the land gets reforested. To handle the issues of escrow, we could do it on the blockchain.

Have I missed something? Interested in implementing this? Send me a message, please.


Thanks to Filip, Dylan, and Jonathan for reading versions of this article and giving very helpful comments/asking very illuminating questions.

Appendix: Things I cut, links, and text dumps

The key here is to realise that a) the forests are a positive externality in the sense that they sequester carbon and reduce global temperatures, and b) we are much richer than many of the Indonesians who own the forests, and they need to eat.


Bernardinus,* a village elder from the Basik-Basik clan, for instance, noted that “oil palm companies come here, obtain government permits and plant oil palm without asking permission from the traditional landowners. They think no one lives on this land because all they see is forest. Sure, Indonesia’s largest remaining forests are in Papua. However, people live in and depend on these forests. Indigenous people like Marind, Mandobo, Auyu, Jair, and many more. We own this land under customary law. And yet our voices are not heard”.

In a similar vein, Perpetua, a mother-of-four living in Khalaoyam village, stated that “Decisions about oil palm projects are made in Jakarta, not in Papuan villages. These decisions are made without our consent or participation. Of course, there are problems and conflict. It’s like planting a tree without making sure the soil is fertile and that there is enough water and shade for the tree to flourish. The tree will grow crooked and struggle to survive. It’s the same with oil palm projects in Merauke. The problem is that there is no consultation or consent”.

Consent on paper, however, does not guarantee consent in practice. Bernardinus and Perpetua are just two of dozens of community members who report not having been involved in or informed about oil palm developments now taking place in and around their home villages. Another is Kosmas, a Marind participant at a workshop on sustainable palm oil that I organized in July 2019, who told me, “I only found out about that oil palm was coming after the forest had already been cleared. Before that, I knew nothing about the project”.

In a small number of cases, government and corporate bodies have invited indigenous landowners to participate in sosialisasi before land developments. But sosialisasi, a term that can be translated loosely as ‘awareness raising’ or ‘public information dissemination’, is also problematic.

Firstly, sosialisasi often takes the form of a one-way transfer of information about the planned development from the government and companies to the communities, rather than a two-way conversation in which communities have a say in whether or not the project will go ahead, and under what conditions. Secondly, sosialisasi usually takes place in urban centres, such as Merauke City or Jayapura, rather than in the villages where indigenous communities are based. Those invited to attend these events are often handpicked by the government and companies and do not represent the broader interests or rights of the communities to which they belong. Thirdly, sosialisasi meetings are frequently supervised or attended by members of the police and military, who work with oil palm companies in the capacity of plantation security patrols and brokers. Many Marind are reluctant to speak out freely under such conditions for fear of consequent reprisal.

Meanwhile, Marind villagers who have already voiced their opposition to oil palm projects during consultations complained of being persistently approached, and sometimes intimidated, by company personnel and police officials to surrender their lands for agribusiness development. “They keep coming and asking us to change our minds,” reported village elder Geronimo. “We already said no, but they think that if they keep pushing us, we will agree to collaborate. That is not seeking consent. That is harassment”.

Why this harassment? Incentives not aligned. At the moment, oil companies want to skirt the law any way they can. If we can pay the oil palm companies or local governments more money to leave the land intact, this problem solves itself.

  1. The $50,000 needs to be "burned" (to use a crypto term) somehow. It cannot go back to the philantropist/to charity or to any third party, because then that person will have a perverse incentive to make sure the area of land is cut down in order to receive the payout. ↩︎