Tuesday, April 14, 2015

Economics of Quality and Price: Analysis of CoE Data

Dr. Norbert Wilson and Adam Wilson deliver an engaging presentation at SCAA Seattle
April 11, 2015
Two researchers with a great "how we met" story discussed their interesting project to dissect Cup of Excellence auction data from 7 countries and 8 years. Dr. Norbert Wilson, an Agribusiness professor at Auburn University and Adam Wilson, a coffee professional who recently moved to Washington state from Alabama, shared the findings from their paper published in the June 2014 issue of Agricultural Economics. This published paper caught my attention so much because of its economics approach to price, without looking to evaluate certification in any way. So I proposed and organized the panel, and then had the great pleasure to lead it!

This duo met at an Auburn cafe named Mama Mocha's while Adam was a barista there. Norbert, being naturally curious, welcomed the conversations he'd have with Adam while ordering his daily cup of  'joe. One thing lead to another until Adam was no longer going to be a math teacher. Instead, he was pursuing a masters degree in agribusiness at Auburn University and delving deep into the statistical complexities of quality and price in coffee.
You can learn a lot from your barista.

Talk to your customers, they may open new opportunities.

Coffee shops are fantastic classrooms.

Wilson and Wilson created a hedonic model to predict coffee price -- a practice well established in the wine industry, and already applied to analyze coffee using CoE auction data three prior papers. The model's structure is basically:

Price = X0 + X1Quality + X2Reputation+ X2Buyer
To me, the most interesting question this model answers is "how much more will a buyer pay for an additional point of the CoE quality score?"  Wilson and Wilson were able to re-confirm the earlier research that showed a strongly statistically significant relationship between quality and price. Their work even corrected for a technical error in previous work. Wilson and Wilson used a statistical technique that accounts for the fact that CoE scores are actually truncated at 87 - the CoE competition eliminates all coffees with scores < 87.

We also learned that 
  • North American buyers, more than Asians or Nordic buyers, on average pay higher prices for high quality coffees.  
  • Reputation variables like the region of origin and the rank in the CoE competition also significantly impact price.
  • Reputation of variety, however, is not found to be significant in this data set. A shocking "non-result" in the minds of many coffee connoisseurs. There were re-assuring explanations, however, such as the fact that these coffees are already considered an elite group of coffees when the coffees are scored.
The sensory quality of the coffee determines price. 

Increases in the quality increase price.

The effect of quality on price is buyer-region specific.

The single greatest impact on price comes from winning first place in the competition.

The price of a winning coffee varies by the buyer’s country.

Recommendations for further research – more quantitative work on price; non CoE datasets.

Ruth Ann Church (blog author and panel moderator), Adam Wilson, Norbert Wilson
There was a good sized and actively interested audience -- just not in the front seats.