The impact of coffee quality attributes and ratings on specialty coffee bean prices

With a growing interest worldwide, coffee is one of the most significant agribusiness commodities, coming only after oil as a traded economic good in the global market. Indeed, it is widely acknowledged as a crucial driver of the global economy. In 2021, coffee exports were estimated to be worth a substantial USD$ 36.3 billion.^[1] Brazil and Vietnam are the leading coffee-producing and exporting countries globally. Brazil contributes approximately 33% of the world's coffee production, while Vietnam's share is approximately 14%^[2]. Although more than 124 coffee species have been identified around the world^[3], Coffea arabica and Coffea canephora are the most commercially and economically utilized^[4], accounting for 70% and 30% of the overall commercial production of coffee, respectively. Arabica and robusta coffees are diverse in several aspects, including optimal growing climate, physical properties, chemical identities, and the characteristics of the beverage resulting from the roasting of their seeds^[5]. Coffee beans with unique flavor profiles that are produced in special geographical microclimates are known as specialty coffees. Specialty coffees have high niche markets and fetch premium prices due to the great care taken over the growth and preparation of the beans; they also differ from commercial coffee in terms of taste, quality, and cultivation methods. Currently, specialty coffee represents one of the fastest-growing sectors. Specialty coffee beans are graded based on a 100-point scale to assess their quality, with beans scoring 80−100 points being recognized as specialty grades and those scoring below 80 points are categorized as commercial varieties^[6].

This study aims to explore the factors affecting the prices of specialty coffee beans. Furthermore, this study investigates how the relationship between coffee bean quality indicators and roast degree influences coffee bean prices. The results of this study will be important to both customers and producers alike because they will guide consumers in their purchasing decisions while assisting sellers in establishing data-driven pricing decisions. The following section presents an overview of the previous studies that have investigated coffee prices with regard to price volatility, international trade, agricultural cooperatives, and coffee quality.

Coffee price volatility

Coffee is a popular beverage and is consumed in various forms. A limited number of countries in Africa, Asia, and Latin America produce most of the coffee traded internationally. A study conducted in Ethiopia measured and compared the price risks facing producers, wholesalers, and exporters to identify the groups most affected by price volatility over the past three decades, along with the Ethiopian coffee value chain. It was revealed that the producers' prices were found to be the most volatile, followed by wholesale prices and export prices, respectively^[7]. However, the welfare effects of coffee price volatility for Ethiopian coffee producers were estimated, following the elimination of coffee price volatility, with the results revealing the welfare gain from eliminating coffee price volatility was limited and that the gain per producer came to a meager USD${\$} $ 0.76 per year^[8]. An experimental approach was used to examine the patterns and underlying causes of the excessive price volatility of coffee and cocoa—two soft commodities of critical importance to many of the poorest commodity-dependent developing countries—in the Netherlands. In terms of historical coffee and cocoa prices, coffee price volatility has a different or heterogeneous impact, depending on the nature of the market shock. The indirect effects of oil prices on the coffee and cocoa markets were also assessed, using co-integration and error correction models. The findings demonstrated that there is a long-run causal relationship between oil prices and coffee and cocoa prices^[9]. Also, a study in Uganda indicated that there is a long-term causal relationship between oil prices and the price of coffee^[10]. A descriptive analytical approach was used to create a data set involving two series of future coffee prices and crop reports, which were the daily closing prices for arabica and robusta coffee from Close Futures reported by Intercontinental Exchange (ICE) between January 2004 and December 2014 in the USA and Brazil. The study indicated that crop reports generally affect price fluctuations. This effect is particularly strong when the reports provide information following boom periods in Colombia, Brazil, and Vietnam, which are the world's main producers^[11]. Additionally, the effects of climate change and commodity price volatility in Nigeria on the coffee output were analyzed. The findings confirmed that the coffee output in Nigeria is influenced by both climate change and the international commodity price of coffee^[12].

International trade and coffee prices

The price setting mechanism is largely determined by availability of buyers and the existing market forces in the importing and exporting markets, respectively^[13]. The coffee sector was analyzed as a potentially important source of export diversification in Zambia^[14]. The results show that Zambian coffee exhibits asymmetric short-run supply adjustments related to a long-run equilibrium, whereby production rises significantly after prices rise, while changing little after prices fall. Production, exchange rates, and prices have significant positive impacts on the export volume of coffee. Production variables, exchange rates and prices positively and significantly affect coffee exports in South Sulawesi in Indonesia^[15]. The clear gap in the transmission of international coffee prices in the Mexican coffee market was studied, using a descriptive-analytical approach. The Conditional Error Correction Model (ECM) was used to analyze the relationship between the Mexican and international coffee markets by modeling the price of coffee in the Mexican market (the dependent variable) related to Mexican coffee production and the international price of coffee (independent variables). Time series data were analyzed for a period of 16 years using monthly observations, reaching a total of 192 observations. The following findings were the most important of this study: a 1% increase in the international price of coffee was 0.9% higher than the Mexican price; Mexican production has no impact on its local price; and the collapse started in 2015 and caused a loss of statistical significance in the long-run relationship among prices until 2017^[16]. Using a descriptive-analytical approach, an analysis of the world's largest coffee producers was conducted in Varna, Bulgaria^[17], to identify the challenges in accessing and producing coffee, as well as international coffee prices and the dynamics of their changes. The most notable findings of this study were that the coffee trade of four multinational companies—Nestlé, Kraft Foods, Procter & Gamble, and Sarah Lee—has led to a reduction in their economic growth due to a crisis related to the negative effects of climate change, which affects the production of coffee. The current adverse conditions in the global coffee industry is associated with the availability of cheap labor and high labor costs in coffee-producing countries. From an economic point of view, the crisis also reveals the failure of export-oriented models, wherein significant financial resources are allocated to countries processing coffee but not to countries producing this specific type of crop.

Agricultural cooperatives and coffee prices

The effect of the pricing mechanisms of coffee on the sustainability of Agricultural Marketing Cooperatives (AMCOS) in the coffee-growing districts of Rombo in Kilimanjaro and Mbozi in Tanzania were assessed. The results showed that unfavorable coffee prices, access to extension services, pests, coffee diseases, unreliable coffee markets, shortages, and untimely accessibility of farm inputs were the main challenges related to changes in the price of coffee as well as the sustainability of AMCOS^[18]. Cooperative direct export is one of the institutions involved in the coffee value chain in Indonesia. Compared to exporting through exporters, this route is more beneficial since an acceptable price is charged and higher value-added coffee is sold^[19].

Coffee quality and prices

The factors affecting the sales price of arabica coffee include the quality of the coffee and the technology used in post-harvest processing^[20]. The main quality indicator used as a criterion to determine the price of coffee is the quality grade of the coffee. Based on the scores of green bean physical and cup qualities (0–100) and according to the Specialty Coffee Association (SCA) of America, coffee that scores 80 points or more out of 100 is classified as specialty coffee^[21]. Consumers in Taiwan, China, evaluated various attributes of local specialty coffees to help coffee retailers develop strategies for increasing sales in the secondary market^[22]. In terms of intrinsic quality, atypically, as shown by the empirical results of the analysis, consumers in Taiwan, China, do not prefer coffee attributes such as extra aroma and a strong sour taste. A study undertaken in Mexico involved the survey method to extend the framework used in previous research to understand specialty coffee purchasing behavior. Simple and multilevel mixed-effects logistic models were tested. Most notably, the study found that the probability of buying specialty coffee increased when consumers inhaled the coffee aroma when the coffee shop had a value-added business model and had diversified into sales of ground coffee and coffee drinks, and when the coffee shop was located in a city with a high education index. In contrast, the probability of buying specialty coffee decreased when a coffee professional tasted coffee before purchasing, when coffee shops did not provide the option to roast the coffee, and when coffee shops were in large cities^[23].

The data for this research were obtained from the Coffee Review^[24,25], a website that specializes in reporting specialty coffee bean prices and evaluating specialty coffee bean quality to guide coffee consumers worldwide in their purchasing decisions. A 100-point scale evaluation system is used according to the following interpretation, shown in Table 1^[26].

After removing any missing or incomplete observations, the total number of observations reached 1,601. Table 2 shows a summary of the statistics of the collected data, indicating that the average score for acidity and aftertaste is fair while the aroma and body aroma were good. Surprisingly, the average score for flavor is within the very good range.

Furthermore, the results show that the average retail price of specialty coffee beans is USD${\$} $ 24, while the average size is 10.8 ounces. The internal consistency for the coffee attribute ratings was examined using Cronbach's Alpha. The obtained score was within the acceptable range (0.76) with a 95% bootstrap confidence interval ranging from 0.73 to 0.79.

The reported consumer ratings for coffee attributes reached as low as 60 for acidity, body, and aftertaste. The attributes that achieved the maximum ratings were acidity, aroma, and body. The roasting levels obtained from the original data consisted of light, medium, dark, medium-light, and medium-dark. However, we merged medium-dark with dark and medium-light with light for the sake of organization and brevity. Also, the coffee beans' origins were classified according to their continent rather than their country of origin to reduce the data dimensions and reduce the degrees of freedom in the subsequent analysis. It was noticed that the majority of coffee beans originated from Africa, followed by Central and South America, respectively. Coffee beans produced in Hawaii, Papua New Guinea, Haiti, the Dominican Republic, and elsewhere were labeled as Other.

To examine how the factors influence coffee bean prices, the following model was used (also known as the hedonic pricing model):

where, P denotes the specialty coffee bean retail prices, size is the weight of a sold bag in ounces, Quality reflects the variables of the quality attributes and consumer ratings, and R and O are the dummy variables, representing roasting level and country of origin.

The results of the estimating model in Eqn (1) are reported in Table 3, along with the goodness of fit measures (r-square, adjusted r-square, and regression standard error).

The results for goodness of fit, as measured by regression standard error, r-squared, and adjusted r-squared, show that the log-log model is a better fit than the first model with no logs. Consequently, the results of model 2 will be the main focus of the analysis. The results of the Breusch-Pagan heteroskedasticity test shows that the null hypothesis of constant error variance (homoskedasticity) is rejected. Thus, the reported standard errors in Table 3 are heteroskedasticity robust standard errors.

The sign of the size coefficient in both models confirms the economic theory of inverse demand, which states that there is a negative relationship between quantity and the retail price of specialty beans, in this case. In other words, an increase in the quantity of specialty coffee beans offered for sale reduces the retail price that the customer pays per gram. This aligns with the findings of others^[27−30] who found that quantity (the size of a bag of coffee) negatively influences the selling price of specialty coffee. Thus, buying specialty coffee beans in larger bags, whether in pounds or kilograms, will result in greater savings for the consumer. Specifically, the results showed that a 1% increase in the size of the coffee bag offered for sale reduces the retail price by 0.13% on average, when all of the other factors remain constant.

All of the control variables representing coffee quality attributes (aroma, acidity, body, flavor, and aftertaste rating) were found to have a positive effect on the retail price of coffee beans. This result confirms the findings of prior research^[31], which found that the quality score positively affects the price of specialty coffee. Aroma and coffee flavors (such as sweet, floral, and fruity) as well as acidity were found to have a positive impact on specialty coffee prices^[28]. Conversely, the present results differ from other findings^[28], which reported that the aftertaste score does not affect the price of specialty coffee. Nonetheless, the present findings show that flavor rating and aftertaste rating have the largest impact. Thus, decision-makers in the specialty bean industry should seek advice from expert sensory evaluators and collect consumer feedback since highly rated coffee beans will enable them to charge a higher price for their beans. On the other hand, the results of the first model show that light, medium, medium-dark, and medium-light roasted coffee beans are, on average, priced lower than dark roast coffee beans. This result contrasts with the findings of previous studies^[28], which found that roasted coffee flavor does not affect the coffee price. Furthermore, coffee beans originating from Central America, South America, and countries such as Hawaii, Papua New Guinea, Haiti, and the Dominican Republic are priced higher, on average, than those from Africa. This result is consistent with prior studies^[28,29] which found that the country of origin is a significant determinant of specialty coffee pricing.

In the third model, the log-log model was extended (model 2) by adding interaction terms between roast level and quality attributes, as shown below:

The interaction between buyer location and quality score has been previously investigated^[30]. In this paper, we are interested in exploring the existence of potential interaction between the specialty coffee bean quality attributes and roast level, and the effects of such interaction on pricing. The impact of a change in quality attributes can be found by taking the partial derivatives of $ \mathrm{ln}{P}_{i} $ with respect to $ \mathrm{ln}{Quality}_{i} $:

Equation (3) shows that the impact of a change in quality variables on price varies, based on the roast level. Thus, when the roast level is (R = 1), as in the case of light and medium roasts, Eqn (3) is used. However, when the roast level is dark (R = 0), the effect of a change in the quality variables on the price is simply β₂.

The results in Table 3 reveal a statistically significant interaction between aroma and roast levels, which confirms that the roasting level is a critical factor in coffee aroma formation^[32]. The sign of the interaction terms between the acidity and roast level was negative, confirming the negative association between overall coffee acidity and roasted coffee quality^[33]. Furthermore, all other quality attributes interact significantly with roast level. A significant interaction was found between acidity and light-roasted coffee beans. Also, all roasting levels significantly interact with the coffee body, resulting in a strong positive influence on price, which is consistent with previous findings^[34] that showed that the influence of roasting degree is more pronounced on body and acidity, respectively.

Furthermore, when the coffee roast is light, a 1% increase in aroma, acidity, flavor, and aftertaste rating increase the selling price by more than one percent, with other factors remaining constant. Moreover, the results did not show a significant interaction between acidity and medium-roast coffee beans that may influence the selling price.

Thus, we recommend that decision-makers in the specialty coffee bean industry study coffee flavor rating scores and consumers' aftertaste ratings since they have the strongest influence on the retail price of specialty coffee. Moreover, sellers are recommended to sell the finest specialty coffee beans in small-sized bags only to maximize their revenue.

The main limitation of the present model is that it does not include variables representing coffee processing (such as natural, washed, and honey processed) or altitude, which can influence the retail price, because data on these variables were not available. Thus, it is recommended that future researchers study these and other variables such as store type (online or physical location) to gain a further understanding of the factors that affect the retail price of specialty coffee.

Coffee is the favorite beverage of many people worldwide, and its popularity has led consumers to search for the finest coffee beans. Specialty coffee beans are sought by consumers due to their unique flavor and pleasant smell. Specialty coffee retail prices, whether sold online or in stores, vary significantly. Therefore, this study attempted to identify the factors that affect specialty coffee bean prices to help consumers save money on their purchases and to provide producers with recommendations for identifying the most important attributes that influence the selling price of their beans. This study used 1601 observations regarding price, size, origin, and the sensory evaluation of specialty coffee bean attributes (aroma, acidity, body, flavor, and aftertaste rating). The results for Cronbach's Alpha showed an acceptable internal consistency for the coffee attribute ratings. The results of the linear regression models show that as the size of the coffee bean bags (in ounces) increased, the further the retail selling price is expected to be reduced. This study recommends that consumers interested in cost savings should buy their favorite coffee beans in larger quantities and only purchase smaller quantities when they want to taste new brands, flavors, roasteries and so on. Conversely, the sellers of specialty coffee beans are recommended to sell the finest high-quality beans in small bag lots to obtain higher revenue. Also, the results show that specialty coffee beans originating from Asia, Central, and South America are, on average, priced higher than those from Africa. Aroma, acidity, body, flavor, and aftertaste ratings were all found to have a positive influence on specialty coffee bean prices. Also, this paper found a significant interaction between roasting level and coffee quality score, represented by acidity, body, flavor, and aftertaste ratings. Thus, the impacts of quality attributes on price were demonstrated as varying, based on the roasting level. This study recommends that producers and sellers of specialty coffee beans should regularly seek feedback from consumers on flavor and aftertaste ratings since this information will help them to charge higher prices for their beans. Furthermore, it is recommended that future researchers should collect data directly from coffee consumers using surveys to investigate the impact of consumers' socio-demographics characteristics (such as race, gender, household size, education level, etc.) on the quantity of specialty coffee beans consumed. This would help to estimate the price elasticity of demand for specialty coffee beans, which will, in turn, help to determine the responsiveness of quantity demands to changes in specialty coffee bean prices.

The authors confirm the contribution to the paper as follows: study conception and design: Al-Mahish MA, Alfayadh RA; data collection: Al-Mahish MA; analysis and interpretation of results: Al-Mahish MA, Alfayadh RA; draft manuscript preparation: Al-Mahish MA, Alfayadh RA. Both of the authors reviewed the results and approved the final version of the manuscript.

The data that support the findings of this study are available on the kaggle.com website repository^[24]. These data were derived from the following resources, available in the public domain: www.kaggle.com/datasets/hanifalirsyad/coffee-scrap-coffeereview, www.coffeereview.com/review/

We extend our sincere appreciation to Dr. Shamseldein H. Ahmed for his valuable assistance with the English language translation.