ANDREA ESTRELLA, STEVE BOUCHER, and CHRISTOPH SAENGER share results of a recent research collaboration between the University of Muenster’s TRANSSUSTAIN research project, University of California Davis, and the International Coffee Organization that could help build an ongoing benchmark for farmer profitability.
One of the reasons why a benchmark for farmer profitability doesn’t yet exist is the complexity of each question we have to ask along the way. To start, we have to first establish what economic sustainability looks like; one way to do this is to ask whether or not a producer is “breaking even.” But what exactly does that mean? First, we need to identify a time dimension. In the case of coffee, it makes sense to consider the period beginning immediately after last year’s harvest through the end of this year’s harvest. Next, we need to identify an indicator that will tell us if the producer “broke even” or not this year. We use the conventional indicator of gross margin, which is simply the difference between the revenues generated from coffee production over the past year and the costs of producing that coffee. Revenues, in turn, is a relatively straightforward concept, found by multiplying the volume produced by the price received by the farmer.
So far so good. But what about costs? This is where things get a bit tricky, for two reasons. First, coffee is a perennial crop, implying that the level and structure of costs change over the life of the farm from installation, to maintenance of young, non-producing plants, through peak years of production. Second, coffee production is dominated by small farmers who tend to complement hired labor with unpaid family labor.
These features have important implications when we define and measure this year’s costs, for example, for a producer with a five-year-old plantation. On one hand, we should include a portion of the fixed costs the producer incurred to install the coffee plants as well as the implicit costs of using their own land, machinery, and equipment. On the other hand, we should also include the opportunity cost of unpaid family labor they used to manage and harvest the coffee farm throughout the year, i.e., how much is their family giving up by employing their spouse and children on the farm instead of working for a wage elsewhere?
Once we’ve determined a way to calculate costs (see Calculating Cost: A Methodology, on page 18), we then apply our methodology to a dataset. The data used in this study were collected as part of TRANSSUSTAIN, a project run by researchers at the University of Muenster who seek to evaluate the effectiveness of voluntary sustainable standards. We use data from a random sample of approximately 1,900 coffee farmers in the main coffee growing regions of Colombia, Costa Rica, and Honduras. In Colombia and Costa Rica, farmers were selected from the membership rolls of coffee cooperatives. In the case of Colombia, 745 farmers were selected from three cooperatives in the traditional Coffee Belt (Eje Cafetero). In Costa Rica, 503 farmers were selected from five cooperatives in the Valle Occidental and Los Santos coffee growing regions. Finally, in Honduras, 659 farmers who work with a large foundation were selected from the North, South, and West coffee growing regions.
Sample farmers were administered a survey which collected detailed information on a range of topics including production, marketing, and costs; data were collected between March 2016 and December 2017. The recall period for Costa Rica and Honduras was the 2015–16 season, but given that coffee is harvested throughout the year in Colombia’s Coffee Belt, the recall period in Colombia was the 2015 calendar year.
Building an “Average”
Once we applied the methodology to our data, we get an initial picture: Fig. 1 presents the average costs per hectare as well as a breakdown of average costs into two categories: annual operating costs (those required to deliver a year’s harvest) and capital recovery costs (land, machinery, equipment, etc.). Per-hectare costs range from US$1,558 in Honduras to US$3,316 in Colombia to US$4,044 in Costa Rica.
Closer inspection of Fig. 1 reveals a number of interesting patterns: First, in each country, annual operating costs dominate long-run capital recovery costs. Operating costs range from 75% of total costs in Costa Rica to 92% of total costs in Colombia. The lower fraction in Costa Rica corresponds to both the higher land values and larger holdings of tractors and other machinery among Costa Rican coffee farmers compared with the other two countries.
Second, labor is far and away the largest cost category. Paid and unpaid labor account for approximately 56% of total costs in both Honduras and Costa Rica and up to 75% of total costs in Colombia. Harvest labor represents over 70% of total labor costs in each country, not surprising given that mechanical harvesting is rare or nonexistent in each of these three countries.
Third, unpaid labor is significant in both Colombia and Honduras, where it represents about 18% of total costs. Neglecting to account for unpaid labor would lead to serious over-estimation of the profitability of coffee farming in these two countries. In Costa Rica, in contrast, coffee farmers tend to pay family laborers so that unpaid labor represents less than 4% of total costs.
Given the labor intensiveness of coffee production in these three countries, the economic sustainability of the sector is highly sensitive to trends in local wages. Indeed, an important explanation for the lower total costs in Honduras is significantly lower wages; the daily harvest wage in Honduras averaged US$8.4, compared to US$16.3 and US$22.2 in Colombia and Costa Rica respectively.
Finally, even though capital recovery costs are small relative to annual operating costs, neglecting them would lead to over-estimates of sustainability, especially in Costa Rica where land prices are high and sample farmers have significant investments in machinery.
Fig. 2 shifts the focus from cost per hectare to cost per pound of green coffee. Average per-pound costs of production were similar in Colombia and Costa Rica, at US$1.4 and US$1.3 per pound respectively At US$0.8, per-pound costs were approximately 40% lower in Honduras. This is not surprising given the significantly lower labor costs in that country.
Of the two graphs, Fig. 2 moves us closer to a measure of sustainability as it identifies the price per pound the “average” farmer in each country would have needed to receive in order to just break even. To get a feel for how well the “average” farmer did, consider that the average farmgate prices per pound of green coffee reported by the International Coffee Organization (ICO) for the 2015–16 production year were US$1.19 in Colombia, US$1.25 in Costa Rica, and US$0.88 in Honduras. Based on these prices, the “average” farmer in Honduras would be slightly above the break-even point. In both Costa Rica and Colombia, however, the situation is more dire as the ICO price falls short of the average cost per pound.
Moving Beyond “Average”
So far, we have looked at average costs. This is useful because it gives a rough sense of what price farmers would need to receive in order to break even – but there is no such thing as an “average” farmer. Instead, each farmer makes different decisions regarding how to manage their coffee farm. These choices may also affect quality and thus the price they receive for their coffee. Farmers may also face different unit prices for their inputs if, for example, they purchase in bulk through a cooperative versus purchasing small quantities as an individual buyer. Finally, farmers may be affected by different unexpected weather or market-related shocks that, in turn, affect yields, costs, and quality. In short, we know there exists significant variation in both costs incurred and prices received by coffee producers. As such, it’s likely that there is a significant variation in profitability for farmers across and even within regions.
Understanding this heterogeneity (lack of “sameness”) is important in order to evaluate the severity of the “coffee price crisis.” For example, we saw above that the average price received by producers in Honduras was US$0.88 per pound, while the average cost per pound was US$0.80. Should we conclude then that all is well in Honduras’ coffee sector? Certainly not, at least not without additional information. If the vast majority of farmers in Honduras receive the US$0.88 price and have a cost structure similar to the “average” farmer at US$0.80 so that most farmers have a positive gross margin, then we might conclude that production seems economically sustainable (at least in this specific year). If instead, we see a lot of variation in gross margins around the “average” farmer, so that a significant fraction of farmers are far from breaking even while others are doing quite well, then our conclusion would be quite different.
So what do we see for sample producers in the three countries? Fig. 3 gives us a clear sense of the variation in gross margins by plotting the cumulative distribution functions of gross margin for each country.
Perhaps contrary to expectations, profitability was highest in Honduras. Consider the point (0, 0.25) on the light gray curve, which represents the cumulative distribution function for Honduras. This implies that 25% of the sample in Honduras had a gross margin of zero or less. In other words, 25% of the farmers in the Honduras sample failed to break even, while 75% were breaking even or better. The distribution of gross margins in Costa Rica (middle curve) is quite similar to that of Honduras, with approximately 28% of farmers at or below the “break even” point. How do we reconcile the similarity of the distribution of gross margins with the significant differences in cost per pound across these two countries? Readers familiar with coffee production in these two Central American countries will likely know the answer. Honduran farmers receive significantly lower prices for their coffee (compared to Costa Rican farmers), but they remain competitive because they have significantly lower costs. Costa Rican farmers have higher yields per hectare, which lowers their per-pound production costs, and also receive higher prices than their Central American peers.
The situation in Colombia is more concerning. In the year of the study, 53% of sample farmers operated at a loss, with many operating at significant losses. For example, just over 25% had gross margins less than (−)0.5 (i.e., their costs per pound exceeded the price they received by at least US$0.5). A combination of rising labor costs plus a set of adverse weather shocks which lowered coffee quality (and price) help explain these results.
In order to monitor and evaluate the economic sustainability of coffee producers, we must fully account for costs. But as we’ve seen here, measuring those costs is difficult given the complexity of coffee production, including the perennial nature of coffee and the common use of unpaid family labor by producers.
The findings of this study suggest reasons for concern: Over a quarter of sample producers in Honduras and Costa Rica failed to break even in the 2015–16 coffee year, with the situation significantly worse in Colombia where just over half of sample producers failed to break even. And, importantly, although much emphasis has been placed on the need for smallholders to raise quality in order to receive price premiums, these numbers indicate that this is only a viable “solution” if the additional costs required to meet quality standards do not overwhelm the price premium. This is one interpretation of the relatively higher profitability of sample farmers in Honduras, who produce lower quality coffee with significantly lower costs compared to those in Costa Rica and, especially, Colombia.
This study highlights the importance of research initiatives such as TRANSSUSTAIN that generate high-quality data on production costs for large, representative samples of producers. These data allow us to understand the degree to which current trends in farmgate prices for coffee are jeopardizing the economic viability of producers. Collecting reliable, high-quality cost data that captures the diversity of farming systems within countries is challenging and resource-intensive needed in the coffee sector to develop the capacity to track costs and monitor the economic health of its producers in different origins. To this end, initial efforts by the ICO to explore potential methodologies to systematically collect cost data are encouraging. The ICO-led sector-wide dialogue brings together stakeholders from both the private and public sectors to address the impact of coffee price levels on livelihoods of farmers and provides an opportunity to increase market transparency. The collection of data on cost of production has been identified as one of the priorities for joint action with a commitment to work towards a mechanism for a global benchmarking of production costs.
Finally, as Jeffrey Sachs’ Coffee Sustainability report states, continuing with a business-as-usual approach will lead to a widespread coffee crisis, further harming the livelihoods of producers and heightening supply risks. Strong concerted action among industry players, governments, NGOs and individuals is needed to ensure the long-term economic viability of the coffee sector.
STEVE BOUCHER is an Associate Professor in the Department of Agricultural and Resource Economics at the University of California Davis. ANDREA ESTRELLA is a Research Economist at IMPAQ International, LLC. She completed her PhD at the University of Muenster and was a Visiting Scholar at UC Davis. CHRISTOPH SAENGER is Senior Economist at the International Coffee Organization.
Calculating Cost: A Methodology
Many methodologies exist to account for costs. We broadly follow the methodology used by the Cost and Return Studies maintained by the California Agricultural Issues Center at the University of California, Davis. We separate costs into two general categories: annual operating costs and long-term capital recovery costs. Annual operating costs include all costs associated with maintenance and harvest throughout the year. We further separate operating costs into the following three categories: i) Hired labor; ii) Unpaid family labor; and iii) Materials and inputs. A range of options exist for the valuation of unpaid family labor, but here we choose a simple approach and value unpaid family labor at 60% of the average local wage paid by sample producers for each activity.
Capital recovery costs spread the cost of major investments over the life of the coffee plantation. We separate these into the following three investments: i) Installation of the plantation; ii) Land; and iii) Machinery and equipment. Again, a range of methods exists for calculating capital costs, but for the purpose of this study we proceed as follows. Installation costs include the cost of land preparation and planting of seedlings. In each of the three countries in our study, we first calculate the average installation cost per hectare for those sample farmers who installed new trees in the 12 months prior to the survey. We then compute the capital recovery cost associated with installation as the average per-hectare cost in each country divided by 20 to spread the cost evenly over the productive life of a coffee plantation. The capital recovery cost of land is calculated as the annual interest payment on a 20-year loan for purchasing the land. Finally, the capital recovery cost associated with machinery is calculated as 50% of the current value of the producer’s assets divided by 10, which is an approximation of their productive life.
 https: //www.uni-muenster.de/Transsustain/
 Specifically, in Honduras the data collection took place from March 2016 until December 2016; in Colombia from July 2016 until October 2016; and in Costa Rica from February 2017 until December 2017.
 Cost and return studies for a wide range of perennial and annual crops grown in California as well as a detailed description of the cost accounting methodology are available at https://coststudies.ucdavis.edu.
 For simplicity, we include interest on loans for operating costs in the inputs and materials category. Given that a relatively small fraction of farmers borrowed for operating costs, these interest costs are negligible for sample farmers, representing less than 10% of input costs.
 See “Calculating Cost: A Methodology” for a summary of methods to value unpaid labor.
Special Thanks to Our Issue 11 Advertisers
Are you an SCA Member? Claim your complimentary subscription to the print version of 25 Magazine in English at sca.coffee/signmeup.
Not a member? Join now and support the mission of our non-profit trade association at sca.coffee/join.