94. Cost of Production - the basis for understanding price

Feb. 12, 2020

Adele (L), a coffee farmer in Rwanda,
talks with researcher Ruth Ann (R).

Cost of production (CoP) is an accounting term. It refers to the "cost of production of any item or unit." Most definitions talk about three components of CoP:

1. Raw material
2. Direct labor costs
3. Overhead costs (e.g. management salaries, office supplies)

CoP gives the accountant a value for the costs incurred to produce a product or service. Once CoP is known, then profitability of the product or service can be calculated more easily.

A long lamented issue in the coffee industry is the difficulty to get a measurement, a value, for CoP for 1 kg of green coffee. More and more, the discussions about the "price crisis" and "fair trade" are focusing on how problematic and core this issue is. Say Roaster X just paid $2.75/lb FOB, about $6.05/kg FOB for green coffee. Is it a good price? It might be. The correct answer depends completely on the producers' costs.

At Artisan Coffee Imports, we have a perspective on this problem that goes back to our automotive roots. I grew up in southeastern Michigan and in an earlier part of my career I was a non-profit adviser to small manufacturers supplying the big auto companies. The companies that did well were the ones that had a fairly good understanding of their CoP, which enabled them to have productive discussions with their buyers about price.

After getting started in coffee around 2007 and seeing the parallels in coffee's long supply chain, I was eager to work with those who could help me understand costs and prices. Fortunately, I had the opportunity to be part of an amazing research project, the Feed the Future Africa Great Lakes Coffee support program in Rwanda and Burundi, from 2015 - 2018. My participation on this project enabled me to live in Rwanda for a year, and to utilize a large, new dataset to define and analyze cost of production of coffee in Rwanda.

In short, I wrote my masters thesis for an M.S. in Community Sustainability degree from Michigan State University on this topic. "Estimating Farmer Cost of Production: Implications for Sustainable Growth in Rwanda’s Coffee Sector" is the title of this 65 page labor-of-love, completed in August 2017. I've recently been able to format at least the executive summary to be more reader friendly, and therefore would like to share a link to the entire document HERE.

Summary of a Masters' Thesis
In case readers of this blog do not have time to read a 65 page thesis, I'd like to share a few highlights here. Tables and graphs are often the most interesting parts of any research paper, so let's focus on a few of those. In the literature review, I share Table 2: a summary of the results of 10 different coffee cost of production studies from around the world.

Table 2:  Ten CoP studies with country and CoP result

Sorted by country, then highest to lowest CoP
Study #	Study (Author – year)	Country	CoP Result US$ per lb. green coffee
1	USAID – AGLC – 2016	Rwanda	$0.64
2	Technoserve, 2011-2013	Rwanda	$0.26
3	NAEB – 1990s	Rwanda	$0.05
4	TMEA – Integrity – 2014	Burundi	$0.75
5b	COSA – 2015b	Costa Rica	$1.96
5c	COSA – 2015c	Guatemala	$1.32
5e	COSA – 2015e	Nicaragua	$0.92
5f	COSA – 2015f	India	$0.72
5g	COSA – 2015g	Vietnam	$0.52
5a	COSA – 2015a	Kenya	$1.52
6	Nyoro, Wanzla & Awuor 2001	Kenya	$0.68
8a	CIAT/CRS – 2015	Colombia – Narino – Off-farm	$2.38
9	IDH – Technoserve 2014	Colombia	$1.66
8b	CIAT/CRS – 2015	Colombia – Narino – Specialist	$1.58
8c	CIAT/CRS – 2015	Colombia – Narino Diversified	$1.40
5d	COSA – 2015d	Colombia	$1.12
8d	CIAT/CRS - 2015	Colombia nat’l avg.	$0.92
7a	Committee on Coffee Competitiveness, 2015	Colombia – state A	$0.83
10	CRECE-UTZ, 2014	Colombia	$0.78-$0.95
7b	Committee on Coffee Competitiveness, 2015	Colombia – state B	$0.72
7c	Committee on Coffee Competitiveness, 2015	Colombia – state C	$.68

Table 2 shows ten of the twenty or so papers reviewed for the study. The ten were selected because they give clear estimates of CoP. Using US$/lb green as a unit of measure to enable comparison, the range of values for Rwanda is $.05 to $.64. Integrity’s milestone study for Burundi estimated $.75/lb green for 2012. Estimates for Colombia range from $.68 - $2.38/lb green. This wide range of values can be attributed to both the underlying differences in costs across time and across regions and countries, as well as the variety of methodologies used in the estimation formulas.

After the literature review, we review the methodology of the research, especially the manner in which the survey was conducted in 2016. It was conducted with 1024 coffee farmer households randomly selected from farmer lists from 16 washing stations in 4 selected coffee-producing regions of Rwanda (Huye in the south, Rutsiro in the West, Kirehe in the East and Gakenke in the North.) So the calculation to understand the sample size of 1024 is 16 coffee washing stations x 64 farmers from each. This is a unique and extremely valuable database, providing the basis for establishing many new insights on Rwandan coffee, not just cost of production. Click here to see some of the publications that have already come out of this dataset!

CoP Punchline

To share the "punchline" first, we find the average CoP in Rwanda is 177/RWF/kg cherry (see Fig. 2 below). These costs for many farmer groups are higher than the average cherry prices being paid in Rwanda. This has serious implications for Rwanda’s long term production trend of specialty coffee. 177 RWF/KG cherry converts to $.64/lb. green, a more universally comparable metric, using Fx rate of 790 Rwf/$1 and cherry to green ratio of 6.4.

A key finding from the report is the confirmation that similar to other crops in many countries, cost of production in coffee in Rwanda tends to be lowest for the largest farms. The number of trees in on a coffee farm is significantly and inversely related to CoP (see Figure 1 below). It is recommended that this finding be used to guide both the design and the evaluation of farmer training programs. Programs should seek to target their resources to the size of farmer that will benefit most. 

Figure 1: CoP/Kg cherry is inversely related to number of trees in Rwanda.

CoP and Prices
More and more, the specialty coffee industry is realizing that programs and, importantly, pricing may need to be tailored by farm size to achieve sustainable margins in all size groups, (Trewick, 2015). For example, as shown in Figure 2 below, the floor price for cherry in Rwanda was set at 150 RWF at the beginning of the 2016 season (see line A in figure 2). On the left side of the graph, one can see how the smaller farmers will lose money when the cherry price is 150 RWF/kg cherry, whereas large farmers with 1001+ trees (20% of the sample in this research) make a slim margin, because their cost of production is below the average and below the floor price.

Figure 2: CoP/KG Cherry by Number of Trees with the 2016 cherry floor price

Based on these findings, stakeholders in Rwanda might look at other aspects of those with very few coffee trees, such as the inability to join cooperatives due to the fact that membership usually requires a minimum number of trees. A minimum tree requirement of 200 – 300 trees is common.  On average, cooperative membership has been shown to significantly correlate with lower cost of production in Rwanda (see section 6.2 of the report). With a holistic look at the situation of the “smallest of the small” farms, appropriate and specific interventions that would help create profitable scenarios for these farmers could be designed. 

Conclusion
Hopefully, these few highlights give a glimpse as to why understanding cost of production can be valuable. I know it has helped me as an importer to have confidence in my contracting methods and ways in which I interact with the producer organizations from which I buy. I tend to have a lot more emphasis on the cherry price paid to the farmer than other buying groups. Since I understand average cost of production is around 177 RWF/kg cherry (in some areas I have more numbers more specific for that exact geographic area), I push hard to get at least 300 Rwf/kg cherry to the farmer, roughly $1.50/lb green (using foreign exchange rate of 790, as used in the year the thesis was written). 300 Rwf/Kg cherry is the value the farmers told us in the survey was the value that they need to be profitable. As a buyer, I can see 300 Rwf/kg offers a reasonable gross margin to the farmer (30-40%), even those who are small farmers with CoP much higher than 177/Rwf/kg. We believe in the benefits that come to everyone in the supply chain when farmers are compensated for all the risk and effort required to grow great, specialty grade coffee, year after year.

Questions?
Have questions about cost of production? Please contact us at rachurch@artisancoffeeimports.com .

57. Cost of Production Presentation - New Research from Rwanda

June 25, 2016

The second presentation Artisan Coffee Group made at the Dublin World of Coffee Expo "Business" seminar room was an update on the new data from Rwanda on cost of production for farmers. The research comes about as a result of Ruth Ann Church's  dual role on the Africa Great Lakes Coffee (AGLC) support program. The project has implemented a baseline study, producing the data used in the presentation. Ruth Ann is the monitoring and evaluation coordinator for Michigan State University and she is eligible for that role due to her status as a masters student in the Community Sustainability department. Ruth Ann is writing her masters project this summer on "Understanding Cost of Production of Coffee in Rwanda." So the Dublin venue was a perfect opportunity to share what has been learned so far.

A key point about the research is that this is knowledge that can help us pay farmers more. At Artisan, we maintain that only when the costs in the entire supply chain are well understood, can management decisions at any point be made wisely and with true sustainability as a goal. Understanding the costs of the farmer are often the toughest for stakeholders to grasp for a number of reasons:

• Public data and credible research on costs to produce for smallholders in tiny countries is rare.
• Little research is even available on the methodologies for conducting cost to produce studies.
• Farmers, especially smallholder farmers, many of whom are illiterate, do not keep careful records of costs, making traditional cost accounting systems unhelpful.
• Coffee farmers are located in very remote places. Travel to reach them and identifying them is time-consuming.
• Coffee farmers speak local languages, like Kinyarwanda in Rwanda.
• There is great variety in the types of coffee farmers, the methods and practices they use, the plants and environmental conditions in which coffee is grown.
• In other words, conducting good research on farmers costs is expensive.

At Artisan, we feel these reasons explain why well-designed research on cost to produce of coffee should be heralded and given a great deal of attention. Thus, we were truly excited to bring the newest data in the industry on cost to produce to Dublin!

The sections of the presentation were as follows:
   1. Methodology of the baseline study

2. Components of cost of production (CoP)

3. Factors driving costs of production

4. Understanding declining farmer motivation to produce coffee.

1. Methodology

The AGLC baseline Survey was conducted early in 2016 on a sample of 1,024 households randomly selected from listings of 16 coffee washing stations (CWS) geographically dispersed across four major coffee-growing districts representing Rwanda’s four agricultural provinces (see map). 

Map: Rwanda - Sampled districts, washing stations (red dots), households (white dots). Source: Nathan Clay, AGLC staff.

The selected districts are Rutsiro (Western Province), Huye (Southern Province), Kirehe (Eastern Province), and Gakanke (Northern Province). The guiding objective of the Sector/CWS selection was to maximize geographic dispersion of the four CWSs in each district and also to ensure that the four would include two that are cooperatively owned and operated and two that are privately owned and operated. From the farmer listings at each of the CWSs, 64 farmers were randomly sampled for study, totaling 1,024 (16 CWS x 64 HH) coffee producing households in all.

The survey instruments were developed at the farm household and field levels. Sections of the questionnaire covered a diversity of topics, not only cost of production. The questionnaires were translated to Kinyarwanda and programmed for Samsung 7” tablets using CSPro Mobile software, and pretested in the field. Ten experienced enumerators were hired and were trained just prior to the pretest. Immediately following the pretest a series of debriefing sessions was organized and the survey instruments were revised based on the pretest results. The ten enumerators worked for 6 weeks in the field to complete the 1,024 farmer interviews, all at the farmers' homes and in their fields.

2. Components of Cost of Production (CoP)
Three major components make up our cost of production value:

1. Labor: Household (unpaid) labor and wage labor (by task).
2. Equipment, such as pruning shears, sprayers, masks, etc.
3. Purchased inputs (mulch, fertilizer, pesticide, etc.)

It's important to note that our data allows us to value the household's "own labor", which is unpaid labor, at a 'market value.' The market value was determined by taking the average of the daily rate the farmers in our sample told us they pay when they hire wage labor. The daily wage for coffee farm tasks was remarkably similar across tasks, except for sorting, which is paid at a lower rate than other tasks. The average rate is 700 RWF/day, which is about $.88/day, (less than a dollar per day). 

Splitting out the labor component into household labor vs. wage labor, a breakdown of the average CoP is as follows:

Cost of Production         Item	RWF	$US	% of total
Mean value of household labor:	35,868	45	33%
Mean value of wage labor:	44,313	56	41%
Mean value of equipment used:	7,506	10	7%
Mean value of purchased inputs:	19,838	25	18%
Total: for the 2015 season	107,527	136	100%

Dividing the total costs per household by the KGs each farmer produced (see table below), we calculate the average CoP per KG cherry. KG cherry is the production unit that is most common across all coffee producers globally.

A.                                        Cost of Production, RWF (cash value)	B.                                             Total harvest 2015 (KG) mean	C.                                  CoP RWF per KG cherry
104,479	1025	177

Converting RWF into $ we have $ .22/KG cherry.
Converting KG into lbs, we have $ .10/lb cherry.

These values are important in the Rwandan context, since the government board that sets the floor price for cherry at the beginning of each season uses a value for the farmers' cost to produce in its calculations. Today they are using 80 RWF/KG cherry, less than half of our estimate.

3. Factors Driving Cost of Production
The presentation continued with a discussion of drivers of cost of production, such as the number of trees, gender and percent of household income coming from coffee. As expected, the number of trees is found to be the most important driver. As the number of trees increases, CoP per KG cherry declines. (See chart below.) 
 

Female heads of household in our sample had significantly higher CoP than the average at 205 RWF/KG cherry. 

Percent of income coming from coffee is a significant driver when the percent drops below 25%. Then the CoP rises steeply to about 277 RWF/KG cherry. Once the percent of income from coffee is above 25%, however, CoP remains at about 150 RWF/KG cherry.

4. Understanding Declining Farmer Motivation
In this section of the presentation Ruth Ann showed analysis from other AGLC team members, namely from Dr. Dan Clay. The "discovery" in the data is that largeholder farmers (those with 1000+ trees) are not investing in coffee in Rwanda at the levels one would predict. Thus the gross margins (profits) per KG cherry for the largeholder farmers are below the margins mid-range farmers are receiving. (See chart below.)


Shown another way, in the bar chart below one can see that mid-range farmers produce more than their share of coffee given the percent share they have of trees. Largeholder farmers on the other hand, are producing significantly less KG of cherry than one would hope given their numbers of trees. 

 We explain the phenomenon in the above two charts with the explanation that largeholders are "divesting" from coffee because the returns are not there. They have other choices. There is a lack of financial motivation for these largeholders to spend their resources in coffee. We propose that a policy remedy is to raise the floor price of cherry to cover costs of production and provide a level of profit per KG cherry that is comparable to the competing rural products, such as banana and raising livestock.

[Thank you to Michigan State University for its generous support for my travel funds to get to Dublin.]

---