This piece is inspired by curiosities surrounding Aviary’s release 004: Lino Rodriguez as well as upcoming releases 005 and 006. AVIARY#004: Lino Rodriguez, featured in the April 2024 edition of Food & Wine Magazine, is available for pre-order through the Aviary website. To place this piece in context of the coffees that inspired it or to support this blog and the work I do, please consider ordering.

The way that climate change hit Huila, harvest expanded from a ten- or twelve-week symphony starting in August or September into staccato-pulses punctuated by silence running nearly continuously through the year. Since flowering and ripening became less predictable and often occurred simultaneously, it made the job of pickers and finding workers to pick difficult. The inconsistency made it less practical to hire migrant workers, who historically followed the harvest as it moved through the country—and even with an influx of migrants and refugees coming from neighboring Venezuela, finding farm labor remained a challenge. Why work on steep slopes picking coffee by hand for little money when the coca producers pay more and more often and the work is less physically demanding?

This made getting only red-ripe cherry challenging—but it’s what we’d planned to do.

Specialty coffee is as much about what you leave out as what you put in; the fundamental rule is “start with red, ripe cherry,” a lesson that is central to the quality improvement practices taught by CQI and Technoserve and one well-supported by both academic and industry research. Plus, picking 75% ripe cherry and 25% pods or green cherry instead of closer to just 50% fully-ripe means that there’s coffee left to ripen on the tree—fewer lower grades now, and a greater opportunity to pick and process that remaining coffee when it’s ripe later. This should, in practice, increase the percentage of coffees exportable through higher-paying specialty channels and should, in theory, improve a producer’s overall profitability. 

But: it’s easier said than done.

To get ripe, sorted cherry we’d have to pay pickers more—not-quite-twice as much more per arroba compared to prevailing local wages—and we’d have to add a second round of sorting with flotation at the wet mill before pulping to ensure as little as possible green cherry slipped through.

Sorting cherry at Lino Rodriguez’s wet mill in Palestina, Huila, Colombia in 2017

The pickers delivered Castillo cherry to the small washing station at Lino Rodriguez’s farm house in Palestina in the late afternoon one day in early November. After floating it, four of us took our time sorting underripes, leaves and sticks by hand into a plastic strainer sitting in the center. When darkness fell, we let the cherry rest there overnight.

The following morning, we pulped the coffee. Before we began, we‘d cleaned the pulper to clear it of any pulp or mucilage stuck to the screen and scrubbed the fermentation tanks. At other mills, where tanks were made of porous, uncleanable wood, I’d put down thick plastic sheeting as a cleanable membrane to avoid contamination or the expense of replacing infrastructure, but the tiled concrete here made cleaning contact surfaces quick and easy. During peak harvest, hygiene often gets overlooked or left behind in pursuit of capacity and efficiency, but it also can be a source of primary defects and lower both quality and yields. 

Capacity was less of an issue now; climate change saw to that.

This was a single-cultivar collection from one block on the farm, so the pulper was calibrated for that one type of cherry. By not feeding through coffee of different sizes and shapes and by keeping a band of ripeness from blood-red to burgundy, we could reduce the amount of skins, unpulped cherry, and pulper-damaged seeds that ended up in the fermentation tank. This too would promote quality and improve yields.

After pulping, we floated the coffee again in clean water to allow us to skim off any hollow, under-developed and low-density seeds as well as husk. This process takes time: to ensure you’re only leaving the highest-quality, densest seeds from the ripest cherry, we agitated the pile of coffee under the water every few minutes, then repeated our skimming and straining efforts. It’s slow, physical work; the drag of the paddle in the mass of pulped coffee swimming below a pool of cold water after a while gives the feeling of paddling a kayak.

By the time the coffee was sorted, floated, pulped, floated and skimmed, we’d discarded the insides of some 5-10% of the cherry that came in—coffee that would be processed separately and sold through other channels such as the strong domestic market or otherwise to instant coffee processors.

Since the Castillo on his mother’s farm, Betania, historically presented with a bit of vegetal astringency and an aromatic of “greenness”, we wanted to find a way to (1) improve cup structure, (2) enhance aromatics, and (3) integrate the citric acidity into the cup. We opted to inoculate the fermentation with LalCafe INTENSO, 2kg of which I’d smuggled in my backpack on my flight to Colombia. Through processing work I’d done previously in El Salvador, Mexico and Guatemala, I knew it would take at least 36 hours in the cool temperatures in Palestina to develop the aromatic profile I was looking for—so we set the minimum length of the fermentation to 36 hours and pitched the yeast after properly hydrating and attemperating it.

After 40 hours, we drained the tank and washed the coffee using the small, traditional washing channels, then carried the coffee to dry in the marquesina above Lino’s farm house, where it dried under partial shade for around 20 days.

The currency of the coffee trade is risk; the better we are managing and reducing risk, the more likely we’ll arrive at a positive outcome. For roasters, managing risk might mean buying only from importer spot offers, or only from certain places or processed in certain ways. For importers, it might mean limiting credit terms extended to roasters, or holding less coffee on spot and using SAS-Replace instead of SAS-NANS contracts. Risk for roasters and importers is fairly straightforward and knowable—and comes equipped with contractual escapes to diffuse risk up the value stream.

But for coffee growers and processors, risk is everywhere and often both unpredictable and unmanageable. The global effects of climate change can’t be overstated—from a change in flowering or ripening, to increased frequency and intensity of tropical storms, to frosts and droughts and heatwaves that damage productivity, quality and soil fertility, to changes in fermentation kinetics and drying and storage—and macroeconomic conditions triggered by events in the global north can wield devastating long-term consequences, such as was seen in the roya outbreaks following the global financial crisis in 2008-2010. A bridge collapse in Djibouti, attempted coups in Latin America, genocide in the Middle East, Russian invasion of Ukraine, and U.S. trade and foreign policy around the world have nothing to do with coffee farming—but the risks and effects of geopolitics harm smallholders in the form of rising costs, contract cancellation, delayed exports, delayed payments, and in the most dire circumstances, theft and violence. 

The economic fog of war settles far from battlefields across the world.

And because coffee—even specialty coffee—is commoditized, a buyer has the ability to obtain the coffee they require for their business, even if an individual farm should suffer collapse or a producer or exporter go bankrupt. If temperatures rise faster than shade can be planted and a farm loses fertility, or if roya strips leaves faster than resistant varieties can be planted and produce, a coffee grower might have no coffee to sell—but that doesn’t mean a roaster has no coffee to buy: buyers can go elsewhere. The fickle, brittle relationship between buyer and seller is at the heart of the burden of risk carried by coffee producers. If the coffee they produce doesn’t meet the expectations of a buyer, the contract might be canceled and they might be forced to sell through other, lower-paying channels, limiting or eliminating profitability and thus the producer’s ability to invest in inputs like fertilizer their farm, quality practices (like cherry selection), or cherry purchasing. If we don’t like or don’t want a coffee, we don’t have to buy it; but a coffee producer must sell what they produce—no matter if it’s an 84 or an 88. 

The scales are unbalanced, and while quality can and often does arise from luck or circumstance, it is often a casualty of this imbalance when risk comes to collect.

In Ethiopia, because washed coffee must legally be exported, cancelation of a contract can have cataclysmic consequences. Washing and drying station owners who collect and process cherry from smallholders typically finance purchase of the cherry they process throughout the harvest based on expected demand or provisional contracts. They carry this interest-bearing loan until export, at which time they’re paid. In the meantime, along with interest, the producer accumulates risk.

In December, I met a washing station owner in Ethiopia who for several years had an exclusive agreement with an importer. The year before we met, the importer had contracted two containers of coffee from the producer. After the producer moved the coffee to Addis for milling and to prepare for export, the importer canceled the contract, forcing the producer to sell their coffee through the horizontal market later in the harvest, after minimum registration prices fell—at a loss of 8 million birr. For the importer, exclusivity with a key, high-quality producer was a way to use their relationship and the promise of high prices to manage their own risk; by preventing this supplier from selling to other, competing importers, the extremely high-quality coffee he produced and his brand remained proprietary and inaccessible through other sources. But that exclusivity protecting the importer’s interests pushed risk onto the producer—and when they weren’t able to buy the coffee, the producer suffered.

This year, to manage their risk, that producer chose not to make washed coffee—instead opting for cherry-dried processes that could be sold through multiple channels should no export buyer appear. He wasn’t alone: all told, 20% fewer washed coffees were produced in Ethiopia during the 2023-2024 harvest.

Responsible buyers who wish to manage risk in a long-term sense—rather than focusing on their interests for a single season—are obligated to engage differently in their buying relationships, particularly when their interests oppose those of a producer’s. If a farmer or processor’s risks and interests are not accounted for in the balance of the business relationship beyond the promise of future payment, then that transaction may very well pose an existential threat to that producer.

The closer to the farm you go—and the further up the value stream you navigate—the effect of every dollar magnifies.

I’d contracted Lino’s coffee at the price his export partner paid for an 86.5 point coffee regardless of the outcome of our experiment, hoping we’d demonstrate that through processing practices we could improve the quality and thus the value of the coffee. I had confidence in the protocol and hoped the outcome would be positive but waived my rejection rights, anyway, as a way to share in the risk of our endeavor. No matter what happened, I’d buy the coffee at the price I’d promised. 

This is standard practice for the work I do—guaranteeing I’ll buy the result of any requests I have, paying a premium for the coffee to cover incurred costs, and never making requests without a sales contract to support it—and is the best way I can think of to create security in that relationship.

As a roaster or importer, I contract based on the price per pound as exportable green, in USD and pay an importer or exporter in USD—while the producer is paid in their local currency (in this case, Colombian pesos), typically before export (in this case immediately upon delivery to the dry mill). The exporter and importer each manage their risk for foreign exchange volatility in the interim—or profit from it—but FOB pricing only really matters insofar as we’re working to keep as much value as possible for that coffee where it was produced.

The price that the exporter pays to the farmer, or farmgate pricing—the in-the-moment, real price paid to a grower or processor for their coffee—is what determines a producer’s profitability.

For coffee growers, one of the biggest costs during harvest is the cost of picking coffee. To ensure that we would start with selectively-picked ripe cherry, instead of paying pickers 4,500 COP per arroba, they were paid 7,000. As a roaster, it was well worth the added cost. If I were to pay $4.65 per pound, landed, for that coffee, then sort out 10% of each roasted pound due to quakers or development issues, the costs would compound: I’d already paid for coffee that was milled, bagged, stored, and shipped—with all of those steps paid for through financing. And then I would pay 3% in carry to the importer to finance it, $0.12 per pound to ship from Continental Terminals to my roastery, and the utilities and wages to roast it then to remove 10% of the finished product. Very quickly, my “sorted” $4.65 per pound coffee actually cost me closer to $6 per pound—without any of that difference passing to the farmer. Increasing the pay to pickers to 7,000 COP per arroba, meanwhile, would add roughly $0.20 per pound to the cost of production—but if it were factored into the price, it would get me the quality I wanted while putting the additional money into the hands of the community and local economy in Colombia, rather than keeping additional value generated from sorting in the U.S., far from where the coffee was produced or the people who produced it.

But, of course: someone would have to be willing to pay.

Between 2016 and 2021, the number of active registered Colombian coffee exporters increased by 223%, and with this increase in exporters came increased market access. If a coffee producer didn’t like the price offered by the FNC or local association, they could try to sell to any of these other exporters. This isn’t true everywhere in the world: other countries with stricter regulations might prohibit the sale of coffee across zones, or might restrict eligibility to become an exporter, or might centralize coffee purchasing in some way so as to control pricing nationally; in other countries, lack of infrastructure, sector development or government support, or language and cultural barriers might limit the ability of smallholders to sell coffee outside of the local market or to anywhere other than to middlemen regardless of the number of exporters.

Colombia’s highly-developed and competitive export market meant that Lino, who had a longstanding relationship with a specialty exporter, had access to better pricing than that offered by the local market. Back then, for an 84-point coffee, the exporter would pay a price of 845.000 pesos per carga of parchment, or a premium of 137.000 pesos over the FNC pricing; for an 86.5 point coffee, they’d pay 986.000 (the FNC price was not tied to quality and unchanged, making this a premium of 278.000 COP over local prices).

The price paid to Lino by his exporter was based on a unit of measurement called a carga (125kg) of coffee in parchment: that is, dried, unsorted, unmilled coffee still in its husk. Before purchase, it’s standard practice in Colombia (see here) to determine a coffee’s “performance” or factor de rendimiento—the amount of parchment, in kilograms, is needed to mill a coffee into a standard exportable 70kg sack. The higher the factor, the lower the final price per kilo of parchment since it will require more parchment to create the exportable lot. In theory, this should incentivize quality practices that would result in higher percentages of higher grades by building an offer price connected to agricultural, agronomic and post-harvest practices—but weather patterns and climate change, monetary policies advancing the planting of certain cultivars (like Castillo) and luck, of course, all impact the factor. 

During milling and cleaning, we remove and sort the coffee: based on density, or color, or morphology or blemish. That is: we take coffee that was already picked (either purchased and financed from other growers, or through labor that was paid), pulped, fermented, dried and stored—financed all the while—and reduce the size of the exportable lot, all in pursuit of improving quality, reducing quakers, improving uniformity, and bringing the lot a little closer to perfection. The coffee that’s discarded—so-called defective beans, lower grades, and seconds—will be sold through other, lower-paying channels that may not even cover the short-term costs of production.

The FNC publishes a table of daily pricing with a base price for each producing region (based on carga, kilograms or arroba delivered) as well as the final price inclusive of a premium based on the lot’s factor. Notably missing in this calculator, however, is cup quality: the Invisible Hand of the specialty coffee industry and the characteristic that, since its inception, has determined whether a coffee qualifies as “specialty.” 

The exporters I work with in Colombia and elsewhere do offer premiums or higher prices based on quality—but this is not how the majority of coffee is traded around the world. If there’s no financial benefit to making the investments that may produce higher quality coffee and if they merely introduce or compound risk, why make those investments at all?

I scored the Castillo we processed together an 86.5—higher than Lino, a skilled cupper, did—but not as high as other cuppers scored it. I was thrilled with the outcome of the experiment, and thought it supported the theory that post-harvest practices could improve both quality and financial outcomes for coffee producers. After all, I was willing to pay more for quality—so I assumed that made it worthwhile.

But for Lino, even at the price I paid, it didn’t make sense.

Back then, other coffees were gaining popularity with coffee farmers in Palestina—Tabi and Pink Bourbon, for example, and Gesha—and those cultivars were drawing the attention of buyers in consuming countries. The prices Lino could receive and the potential profitability for exotic varieties—particularly if they cupped reliably higher than 87—far exceeded the price he could get for an “improved” Castillo, even accounting for the cultivar’s higher overall productivity. And these increasingly popular exotic coffees all offered at least modicum resistance to roya, so the risk mitigation offered by keeping the FNC- and Cenicafé-favored Castillos and producing potentially lower-quality coffee didn’t, for Lino, outweigh the potential upside of renovating the farm with other higher-cupping varieties. 

The FNC and Cenicafé had provided Castillo trees virtually for free to encourage their rapid adoption more than a decade earlier, and replacing those trees, even according to the FNC’s 7-year renovation schedule, would require a loan—loans that banks don’t tend to underwrite unless it will be used for the high-yielding, roya- and CBD-resistant Castillo cultivar. For banks, managing “risk” is about repayment: why loan money to plant trees that don’t come with government backing, which produce lower total yields, and which might be more susceptible to disease?

Lino’s education, training and business acumen—not to mention his market access—placed him in a position where he had the agency to make the investment in his family’s farm and their business. It’s a position of relative privilege: renovation is a risk, costly both in terms of the purchasing of thousands of seedlings as well as the loss of productivity of that sector during the 3-4 years before those new trees produce—if they produce at all. 

But it was a risk that Lino—who had training, education, competitive market access and long-term buyers who would pay for quality—was willing to take, and so in 2018 he renovated the Castillo on their farm with 3,200 Pink Bourbon seedlings. 

Three years later, he submitted a Pink Bourbon from his mother, Teresa de Jesús Ospina Rodriguez, to the Cup of Excellence. It placed fourteenth nationally, with an overall cup score of 87.82.

It would go on to sell for $24.10 per pound.

While the scores we assign to coffees are subjective and in some way result from the mutually agreed-upon delusion we call “quality,” there’s something behind it: quality as a way to describe essential character is something that seems both connected to and transcendent of culture, space and time. The fact that I am willing to pay more for a coffee because I prefer to drink it isn’t a moral judgment; it’s a preference. If consumers, too, are willing to pay more for the experience of one coffee over others means that, by nature of its very being, quality exists and, indeed, has value. That’s supposed to be the underlying truth and proposition of specialty coffee: by recognizing a coffee’s distinctive attributes and selecting it over others because of them, we aren’t treating it as a commodity. 

Coffee isn’t just coffee: it’s not a “mild Arabica,” it’s “a Colombian washed Pink Bourbon.”

It’s perfectly acceptable for us to want to buy coffees we enjoy; it’s our prerogative as buyers, roasters and consumers. But quality results from the intersection of skill, privilege and random chance—and there’s a moral hazard in the pursuit of quality above all else. The choices producers make for their own businesses and their own farms are a result of their agency and independence; what is in their best interest may not align with our own hopes or preferences.

If we, as buyers, don’t appropriately acknowledge and compensate for the measures producers must take and the risks they alone face in the pursuit of quality, then we’ve crossed the rubicon: the quality we’re demanding no longer has any value.

This piece is inspired by curiosities surrounding Aviary’s release 004: Lino Rodriguez as well as upcoming releases 005 and 006. AVIARY#004: Lino Rodriguez is available for pre-order through the Aviary website. To place this piece in context of the coffees that inspired it or to support this blog and the work I do, please consider ordering.