Olena Sambucci and Julian Alston
One of the challenges to be tackled by the economics team on the EV project is to devise an approach to quantify the value of changes in the quality of fruit as a result of variable rate management practices implemented in the experimental vineyards. As a part of our preliminary analysis, we use market-level data for each major category of grapes (table, juice, wine) to analyze the distribution of prices, payoffs to changes in the timing of harvest, and potential payoffs to improvements in quality.
We routinely use data from the California Crush Report (published by USDA/NASS) for our analysis of the market for winegrapes. The Report is the main source of detailed regional data on average returns per ton and tons crushed by variety. These data are used to calculate the total value of wine grape production, as reported in the Annual Statistical reports published by USDA, as well as major industry publications, and academic studies. Table and raisin grape varieties are included in the Report, but only a very small share of these grapes are crushed (usually less than one percent, when it is not possible to sell them for their intended purpose). Therefore, we normally only concern ourselves with data in the Crush Report as they relate to winegrape varieties.
While analyzing these data, we discovered a persistent difference between our calculations of average returns per ton and those included in the Crush Report. The source of the error turned out to be variation among regions in the structure of winegrape industry: in regions with lower-priced grapes, the calculations of average returns per ton in the Crush Report included almost 100 percent of all grapes crushed, since they were sold to others and not crushed to growers’ accounts; in the regions with higher priced grapes, a much greater share of production was crushed to growers’ accounts and excluded from calculations of average returns per ton in the Report. At the level of district and variety the prices in the Crush Report were correct, but when aggregated to state averages the calculations weighted lower-priced grapes more heavily because they excluded tonnage crushed to growers’ accounts.
Given what we know about the structure of the winegrape industry, we developed an alternative method for computing average crush prices, which takes into account the varying shares of production crushed to growers’ accounts and gives more accurate estimates of average prices per ton. We were able to show that using average prices published in the Crush Report to calculate the total value would understate the true total value of the crush by 14–20 percent, depending on the year, compared with our proposed method. This is a fairly significant difference to any user of these data!
We will be using the adjusted numbers from now on when we work with data from the Crush Report. In addition, we sent our calculations to the California office of the National Agricultural Statistical Service and published a paper in the Journal of Wine Economics to share our findings with other researchers working with the data.
The lesson here is that even when using information from well-established and familiar sources it is important to make an effort to understand how the data were created (what Bruce Gardner called “factology”). We stumbled upon this problem with our data while working on something completely different, and were able to suggest a better approach and share it with others. In economics we call this a positive externality, and it is an example of how research in one area can sometimes provide unexpected benefits elsewhere.
Alston, J., Anderson, K., and Sambucci, O. (2015). Drifting Towards Bordeaux? The Evolving Varietal Emphasis of U.S. Wine Regions. Journal of Wine Economics, 10 (3), 349–78.
California Department of Food and Agriculture (CDFA). (2017a). Grape Crush Report, Final 2016. March 10, 2017. https://www.nass.usda.gov/Statistics_by_State/California/Publications/Grape_Crush/Final/index.php.
California Department of Food and Agriculture (CDFA). (2017b). Historical Crush Reports. http://www.nass.usda.gov/Statistics_by_State/California/Publications/Grape_Crush/index.asp.
Gardner, Bruce L. “How the Data We Make Can Unmake Us: Annals of Factology.” American Journal of Agricultural Economics 74(5)(1992): 1066–75.
Sambucci and Alston (2017). Estimating the Value of California Wine Grapes. Journal of Wine Economics, forthcoming.