Shady business: Why do Puerto Rican coffee farmers use conservation agriculture practices?

What is in conservation agriculture for coffee farmers in Puerto Rico? As the title indicates, the answer to this question is not clear. In fact, I would dare to say that the answers evidence provides so far are not answers at all, but rather a series of new questions.

With this commentary I want to share some recent research I have conducted in the area. Of conservation agriculture. I will tell you why this is important for those who care about economic development among agricultural communities and related environmental issues; I will also provide some background of what’s going on in Puerto Rico; give you a brief description of the methods I used to approach the problem quantitatively; tell you about the data I had to work with; and show you the results from my analysis. I will give you my interpretation of the results and conclude with some recommendations for improving current agricultural and environmental policy in PR.

If your thing is solving world poverty, coffee should be in your radar. It is an important commodity traded internationally—second only to oil (!). Also, it employs over 25 million families, mostly small scale farmers living in poverty.

Alternatively, if your cause is for the environment, this research is important because over 70\% of the world’s coffee is produced in the most biologically diverse regions of the planet, and sometimes, coffee farming can cause significant environmental damage.

There are roughly two methods for growing coffee. Farmers can have shade plantations, where the coffee trees grow surrounded by a dense canopy of other trees that provide shade along other benefits. Alternatively, farmers can have a sun plantation, where the coffee trees grow without the shade cover. Respectively, these farming methods are normally referred to as shade and sun plantations. Sun plantations are what you probably imagine when you think of coffee farming: clear rows of coffee tress over a mountain top kissed by the tropical sun.

When done correctly, a shade plantation can function as a natural forest and provide a wide range of ecosystem services that are good for sustaining biodiversity within the farm. In addition, the shade cover can benefit farmers economically: the ecology of a shade plantation is such, that the shade trees nourish the soil and allow it to sustain productivity for longer. Also, coffee grown under shade is said to taste better, allowing farmers to charge higher prices for a better quality bean.

Of course, there is no such a thing as a free lunch. Growing shade coffee comes at a cost: forgone short term productivity. In fact, sun plantations are significantly more productive than shade plantations in the short run and cheaper to manage since there is no need to prune and maintain the shade trees.

In Puerto Rico, the department of natural resources and the environment (DRNA) is very interested in sun plantations (or at least it was before hurricane Maria wreaked havoc in the island). Since 2013, the DRNA has been buying and leasing land near the major coffee producing region of the island in an effort to expand the protected areas covered by forest.

The DRNA recognizes that it would be much cheaper to nudge farmers to grow shade coffee than acquiring their lands and rehabilitating their plantations. To nudge farmers, the DRNA is considering improving existing assistance programs that encourage farmers to grow shade coffee. (Although, I speculate the priorities of the agency have changed since the last hurricane season.)

Before the DRNA is ready to propose new policies, it must understand the microeconomics of coffee farming in PR. As you now know, deciding whether or not to grow shade coffee is not a simple decision: farmers need to consider different time horizons (which will in turn trigger considerations over how much risk they are willing to take), possibly different production functions and technologies (it is difficult to use tractors in shade plantations), dynamics in different markets (those pertaining to the market for specialty coffee which is sold to high end consumers and the dynamics corresponding to the market for average quality coffee which will likely go in beverages and instant coffees), and in PR, the availability of multiple incentive programs, some favoring sun plantations and some favoring shade plantations.

With my research, I intend to understand the economics behind farmers’ behavior in Puerto Rico. I set up the problem analytically as a system of simultaneous equations where growing shade coffee is impacted by whether or not a farmer participates in programs favoring, shade or sun coffee. In turn, recognizing that the decision to participate in these programs is no independent from current practices (or as we say in economics lingo, endogenous), participating in any of these programs is impacted by whether or not the farmer already grows shade coffee.

Both decisions are affected by a wide range of other factors that are farmer–related and farm–related such as age, gender, educational attainment, years of experience, farm size, farm income, price of coffee received by farmers, access to specialty markets, current management practices that include inter-cropping, multi-cropping, leaving unused land in the farm, etc.

In particular, I am interested in learning whether receiving assistance for growing sun or shade coffee was a significant factor determining behavior, and whether the price of coffee and access to specialty markets are important factors. In other words, I want to know if assistance programs work and if markets are working for the farmers.

The data I use for the empirical analysis comes from a survey of 89 randomly selected farmers in the regions targeted by the DNRA (and area with easily 1,000 private properties, in an island with approximately 4,500 coffee farms according to the latest USDA census).

The survey was conducted by ecologists at North Carolina State University and it was designed to answer ecology questions. The survey provided a much less than ideal dataset for answering questions of economics and it became clear from the beginning that the ultimate goal of this research was exploratory in nature. Instead of putting much faith in finding precise quantitative measures, this study would serve to shed light on the areas that need more research and to identify what kind of questions could be asked in a future survey.

Given that disclaimer, I will tell you a about the sample of 89 farmers. Of the 89 farmers, 73 receive some kind of assistance for growing coffee: shade or sun. Almost all of those (72) receive assistance for growing sun coffee. In total, 30 receive benefits for growing shade coffee. Of those, almost all (29) were receiving both kinds of assistance.

The average farmer is old, male, and has been working on the farm over 20 years. Surprisingly, this is a very well educated sample of farmers: 39% of them have college degrees or higher. I suspect this education variable to be related with access to non-farm income. Unfortunately, I have no way to verify this as this question was not in the survey.

Average annual farm income is low. The average farm generates just over $16,000 in revenues per year. The majority of the farmers are just growers, although some do part of the processing. Almost none roast their coffee.

Prices per pound show variance within the sample, and 12% of farmers have access to specialty markets. The observed variation in prices came as a surprise. I was expecting prices to be constant because in PR prices are heavily regulated by the Department of Consumer Affairs and because 85% of the coffee is bought by a single company: Coca–Cola.

Lastly, 55\% of farmers reported leaving part of their farm uncultivated and unused. In these areas farmers left naturally occurring forests to grow unmanaged. I highlight this characteristic because I think it is an important part of the story as I am about to tell it.

A simple look at the data shows something very funky was going on with coffee farmers: When examining subsamples of farmers by type of program they benefit from, I find that 23\% of farmers receiving assistance for growing shade coffee ONLY plant sun coffee, while 14\% of farmers receiving assistance for growing sun ONLY plant shade coffee.

This contradiction is an indication that if assistance programs are doing anything at all, they may be having unintended effects. And that is exactly what I found with the econometrics analysis.

In the econometrics analysis, I consider three 3 cases: one looking at participation in sun programs, another at participation in shade programs, and one looking a participation in any program.

The most interesting finding of this study is that receiving support for growing sun coffee is the most important determinant of whether farmers decide to grow shade coffee or not. Receiving sun assistance increases the likelihood of growing shade by up to 88%!

The second most important determinant is having a graduate degree.

A second group of interesting findings are the determinants of participation in shade programs. The likelihood of participation in these programs increases for farmers already growing shade (unsurprisingly, a selection bias problem), for farmers with access to specialty markets (which is relieving for us economists advocating for markets), and for farmers who leave unmanaged wild forested areas in their farms.

Having natural forests is also a substitute for having shade in the farm, and I think that this is the key for answering why farmers are using conservation agriculture in PR.

Let me now interpret the results: Why are farmers growing shade coffee in PR?

I think that they do so to keep the definitions of their management practices flexible. Let me explain. Having an ambiguous definition is advantageous because if an agent from the Fish & Wildlife Service asks farmers about their management practices, they can point to the dense canopy in their farms and say their plantation is a shade plantation. Having an ambiguous definition is also convenient for when Department of Agriculture agent asks farmers about their management practice, they can say their plantation is a sun plantation. In response, the DA agent will presumably ask how are these sun plantations when there are shade trees in the land. To which farmers could answer that these are not shade trees but natural unmanaged forests they have not taken care of yet.

Although the results do not necessarily imply this answer, it is supported by the inconsistency between the programs farmers benefit from and the practices they choose, and with the popularity of leaving wild forest areas in their farms.

So, mystery solved?

Not at all. The fundamental reasons for why farmers want to leave the definition of their management practices flexible remains unanswered. At this point, all I can do is speculate and offer three answers that are consistent with the data:

  • Rent-seeking: This is a well–educated sample of farmers. Maybe they are savvy farmers taking advantage from the lack of coordination between agencies offering support.
  • Preventive reasons: The average farm does not generate a lot of revenue. Maybe these are cash–strapped farmers that sustain both types of management practices to guarantee some level of support.
  • Other: Maybe there are other forces driving farmer behavior that are unrelated to program participation, such as access to crop insurance. Or maybe, there is so much confusion over the requirements and benefits of the multiple available programs (farmers have access to over 30 programs administered by at least 4 agencies), that farmers just behave erratically and do anything they can.

There are some clear implications for policy design from this study.

It is evident that there are loopholes and inefficiencies with the current system. Instead of tweaking it, the DNRA may want to consider other mechanisms to nudge farmers to grow shade.

If the story is of cash–strapped farmers, maybe they want to implement some income–support program.

If the story is of difficulties accessing specialty markets, the DNRA may want improve farmers’ access to these markets. For example by having conversations with the department of consumers affairs and allowing shops to charge higher for better quality coffee, or by supporting the formation of cooperatives so that farmers can guarantee a minimum volume of good quality beans to the retailers, or maybe supporting the establishment of a shade–coffee certification program, which does not exist in PR.

Ultimately, the findings of my research are suggestive, but not necessarily indicative, that the case of conservation incentives programs in Puerto Rico is one of good intentions accompanied by poor oversight and a lack of institutional coordination: all factors that enable a form of rent–seeking among coffee farmers in PR. Based on this study, federal environmental agencies interested in improving the targeting of existing programs should be wary of displacing or even duplicating the efforts from antagonistic state programs that favor mono-crop type of coffee cultivation as they seem to be, ironically, the most important diver of the decision to adopt conservation practices among Puerto Rican coffee farmers.

For the Department of Natural and Environmental Resources of Puerto Rico to meet its expansion of protected habitat area in the island by encouraging farmers to adopt shade coffee management practices, it is possible that simple income-transfer programs that allow farmers to afford switching from sun to shade coffee are perhaps a more efficient way to promote biodiversity conservation practices than the current schemes in place. Alternatively, conservation agencies may consider devoting their resources to new routes that also induce conversion to shade coffee cultivation such as encouraging the creation of shade coffee certification programs and farmer cooperatives, or supporting other means

for improving farmer access to gourmet markets, high–end retailers and consumers. After all, farmers who had access to specialty markets were up to 50% more likely to participate in shade programs than farmers who lacked the opportunity of charging higher prices and reaching different consumer groups.

If you like this story and want to learn more, you can take a look at the paper I presented this summer at the Annual meetings of the American Agricultural Economics Association in Chicago.