Progress 08/01/17 to 07/31/22
Outputs
Target Audience:From July 2021 to July 2022: Scholarly audiences reached in this period We targeted a scholarly audience by preparing one book manuscript and submitting it to a university press. This manuscript is currently being reviewed by outside experts. Co-PI Neil Reid delivered the following conference presentation over this period: N.Reid, S. Goetz, P. Gottlieb, and A. Hira. Craft brewing, neolocalism, and the emergence of local hop production. Presented to the Southern Regional Science Association Annual Conference, Austin, TX, April 9th, 2022. Industry audiences reached in this period We reached multiple stakeholder audiences in the cranberry industry. We define a local agricultural cluster to include all elements of a commodity's supply chain, as well as outside experts. For cranberries, our list of stakeholders included growers, suppliers, specialized distributors (e.g., Ocean Spray), trade associations, extension personnel, and government agencies. We communicated with all of these audiences over the reporting period, and we do not distinguish among them here: 40 in-person interviews were conducted in Wisconsin and Quebec An online survey was administered to stakeholders in the cranberry industry throughout North America A report on the cranberry industry was completed, reviewed by industry stakeholders, and posted on the research website of co-investigator Anil Hira for free download. Members of the general public reached in this period The web posting by Anil Hira gave us the potential to reach members of the general public. At present, only members of the public with a special interest in the cranberry industry are likely to seek out this online resource. Before July 2021: Our interactions with cranberry industry stakeholders began before 2022 and included an additional 60 interviews, site visits, surveys, one executive briefing at Ocean Spray headquarters, and consulting. We interviewed industry stakeholders in Massachusetts, New Jersey, Washington State, Oregon, Quebec, and Wisconsin, as well as USDA officials in Washington, DC. Dr. Anil Hira gave a presentation of the clusters research to the NACREW (cranberry researcher and extension conference) and to the Washington/Oregon cranberry growers' meeting in 2019. Members of the cranberry industry participated in development of our research design from the very beginning of the project. This is also true for our case study research project on racehorse breeding. The Jockey Club and the Standardbred Breeders' Association provided us with proprietary data on registered stallions, with the understanding that the resulting publication (Sustainability, January 2020) would clarify geographic aspects of national decline in the racehorse breeding sector. These equine trade associations were briefed on the results by co-author Dr. Karyn Malinowski. Between 2017 and 2022, scholarly audiences were reached by means of fifteen conference presentations, one published book chapter, one accepted book chapter that was reviewed by editors, and three published articles. This list does not include the manuscripts completed in 2022 but not yet submitted for publication. Changes/Problems:The data provided to us in 2018 under a special tabulation contract with NASS were incomplete. This problem was rectified by the contractor: a new set of tabulation results was delivered to us in 2019. Because we received the 2012 census data in two separate blocks, our empirical work in support of grant goals 1 and 2 was effectively duplicated. Using a partial dataset, our first set of results was disseminated at conferences throughout the year 2019. Using the complete dataset of 208 commodities, our second set of results exists today only in manuscript form. In retrospect, this duplication of effort had a silver lining. Had we not moved ahead to analyze the partial dataset in 2018-2019, we would have been unable to present any results at scholarly conferences before Covid-19 made travel impossible. We were fortunate to get conference feedback in 2019, because it helped to inform our subsequent work on the complete list of commodities, which we conducted during the "lockdown." In addition to the special tabulation, we applied to NASS for the right to access suppressed agricultural census data in a secure location. The purpose of that request was to see if our study findings would change if we measured the geographic concentration of commodity acreage instead of commodity operations. By the time NASS was ready to consider our application for direct access to suppressed data, its secure data sites were shut down due to Covid. NASS took some time to transition to a new system that would allow access to suppressed data remotely from user's offices, under strict security protocols. We paid a fee for this remote access, but our access was not fully operational at the time of this final report. Our research on the geography of commodity acreage will therefore take place after the end date of this grant period. Covid restrictions posed a special challenge for the cranberry case study because of the need for air travel and face-to-face interviews. The investigator in charge of that subproject, Dr. Anil Hira, was forced to make up for lost time in 2021 and 2022. A few planned meetings were cancelled, but this had no noticeable effect on the ultimate work product. What opportunities for training and professional development has the project provided?The project has provided significant training of graduate and post-doctoral students. We expect future training to occur as research outputs are translated into publication formats for lay audiences (e.g., extension or USDA rural development center factsheets). How have the results been disseminated to communities of interest?Our methods of dissemination over the 2021-2022 reporting period are fully detailed in the section above on "target audience." What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
GOAL #1 1) Major activities completed / experiments conducted To test this hypothesis, regression analysis of a 208-commodity dataset was completed. Results of the analysis were written up in a completed article manuscript: Gottlieb, P., E. Dobis, N. Reid, S. Goetz, A. Hira, and Z. Tian, 2022, "Empirical investigation of the relevance of localization economies to production agriculture." This manuscript is important because localization economies--the economist's phrase for "industry clusters"--are virtually unexplored within production agriculture. This is in spite of the fact that the concept has become central to economic development practice in all other industries. To emphasize its importance, note that development of the theoretical framework we test in this manuscript led to Paul Krugman winning the Nobel prize in economics in 2008. 2) Key outcomes or other accomplishments realized Change in knowledge: We learned that across commodities, a high level of environmental conservation activities and high diversity of products and services were positively correlated with geographic clustering, controlling for other known causes of clustering. This constitutes evidence for the idea that clustering may be caused by the need to share knowledge on particular agricultural practices, rather than being explained entirely by the existence of common geographic "anchors." New methods for rescaling Theil indices to account for the bias caused by having fewer distributed "objects" than geographic units were developed during this reporting period. A similar method of rescaling Herfindahl-Hirschmann indices for geographic data was published in the journal Sustainability in 2020. These new methods will have wide applicability within mathematical geography. GOAL #2 1) Major activities completed / experiments conducted A dataset of 129 commodities was used for a statistical cluster analysis that sorted each commodity into one of four disjoint groups. In 2022, the results of this statistical analysis were written up in an article manuscript: Gottlieb, P., E. Dobis, A. Hira, S. Goetz, and N. Reid. 2022. "A new typology of U.S. agricultural commodities. 2) Key outcomes or other accomplishments realized For reasons of data availability, the typology created in the manuscript above is informed by geographic behavior, but it is not a typology of geographic clusters. The advantage of this approach is that the typology is more generally useful for agricultural policy making than would be the case if it distinguished only among different categories of geographic clusters. Early in this project, a typology of different kinds of geographic clusters was created a priori, using our understanding of the case study literature. That cluster typology was reported in a 2017 presentation to the North American Regional Science Association. It also appears in the 2019 conference paper prepared for the Agricultural and Applied Economics Association Meetings. Change in knowledge: A small group of specialized commodities that we have called "exotic," including alpacas, bison, maple syrup, wool, and Christmas trees, was robustly identified across all statistical models (i.e., very stable group membership). The most prominent characteristic of this exotics group is diverse sources of revenue. A second contribution of the typology involved dividing fruit and vegetable commodities into a direct marketing/value added group and a group that lacks those two attributes. These two groups differ systematically on other dimensions as well, a fact that has not generally been recognized. Commodities like wine grapes, hops, and racehorses have the potential to earn significant economic rents. We hypothesized that the higher these economic rents, the more likely the commodity would be to cluster geographically. This hypothesis was confirmed in 2020 by comparing top-earning stallions to less successful stallions. This finding sheds additional light on one category within our a priori typology of agricultural clusters: the so-called "craft" model. GOAL #3 A three-page report was prepared over the 2021-2022 period that contains recommendations for federal level policy making arising from this project. Starting in 2018, co-PI Neil Reid began a similar exercise with respect to recommendations at the local level. Local recommendations on agricultural cluster "best practices" were presented at the North American Regional Science Association meetings in Fall of 2019. Case study recommendations arising from this project are difficult to generalize to all commodities. For example, the study we published on racehorse clusters found that state policies on gambling played a significant role in determining which states would capture a dominant share of a "shrinking pie." This represents an unusual policy lever, to say the least. Our case study research on hops discovered that this commodity has been de-clustering nationally so that it can be closer to craft beer manufacturing establishments. The downstream portion of the hops supply chain (beer) has been moving in the direction of craft production with local branding and local sourcing of ingredients: this is an example of local supply chain development. A more dispersed hops landscape has started to replace a single exporting cluster--the Pacific Northwest, with a larger set of hop-growing clusters now oriented to several metropolitan areas. The geographic transformation of hops production is a natural response to a revitalized beer industry that is characterized by significant product differentiation, customer-driven localism, and connoisseurship. Some, but not all, commodities may be able to emulate this pattern. GOAL #4 By identifying causes for geographic concentration other than climate, soil, and transportation, the manuscript described under goal #1 showed that geographic behavior in agriculture (including observed clustering) is not strictly deterministic. This finding should be kept in mind when tracking geographic changes attributed to global warming, and when planning potential remedies. The finding that links environmental conservation activities to geographic clustering is especially important because it reflects the possibility that neighbors growing the same commodity are sharing best environmental practices. Previous studies had found that among growers of a particular commodity, cluster-based learning can contribute to greater climate change resilience. Our multi-year cranberry case study found that climate change is an important factor in the relative performance of the various local clusters. Climate change has benefitted the Wisconsin and Quebec clusters especially. This book-length case study highlights the value of taking a "portfolio approach" to multiple growing regions, to help prepare for an uncertain climate future. An open question raised by this research is how national-level interests within North America ought to be managed. GOAL #5 Our experience with the datasets collected as part of this project leads us to believe that data on the location of multiple commodities in the U.S. are adequate (unless limited for confidentiality reasons). Data on the frequently combined practices of direct marketing and value-added production are also adequate. Data on innovation and technology, however, lag behind what is available for North American Industrial Classification System (NAICS) industries. Input-output matrices could also be improved in a way that would facilitate the analysis of local supply chains that is necessary for effective analysis of industry clusters. The IMPLAN input-output system, for example, contains a maximum of 14 agricultural commodity groups. Some of these are highly aggregated (vegetables and melons) while others are separated (cotton, tobacco, sugar).
Publications
- Type:
Books
Status:
Under Review
Year Published:
2022
Citation:
Hira, A., P. Gottlieb, N. Reid, S. Goetz, and E. Dobis. 2022. Cooperatives in clusters: The experiment to overcome agricultural commodity cycles in the cranberry industry.
- Type:
Journal Articles
Status:
Other
Year Published:
2022
Citation:
Gottlieb, P., E. Dobis, N. Reid, S. Goetz, A. Hira, and Z. Tian. 2022. Empirical investigation of the relevance of localization economies to production agriculture.
- Type:
Journal Articles
Status:
Other
Year Published:
2022
Citation:
Gottlieb, P., E. Dobis, A. Hira, S. Goetz, and N. Reid. 2022. A new typology of U.S. agricultural commodities.
|
Progress 08/01/20 to 07/31/21
Outputs
Target Audience:
Nothing Reported
Changes/Problems:Covid continued to make it impossible to move ahead on the cranberry case study and on the analysis of confidential Census data at a USDA-NASS facility. Thankfully, both of these issues have now been resolved. What opportunities for training and professional development has the project provided?
Nothing Reported
How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?Goal 1: Data collection and analysis will be completed. Goal 2: A manuscript will be completed. Goal 3: Non-data research material, such as interview notes, will be organized and manuscripts will be written. Goal 4: The cranberry experience on this will be written in a separate memo, and additional research will be conducted for future work. Goal 5: A White Paper on this subject will be completed.
Impacts
What was accomplished under these goals?
GOAL #1: To test the hypothesis that certain known drivers of transactional clustering behavior are active in agriculture: that they are correlated with observed geographic clustering, controlling for factors like climate requirements and industry size. During the reporting period August 2020 to July 2021, progress was made on two tasks: (1) Creation of the same kind of dataset that was used for the regression analysis presented at the November 2019 meetings of the North American Regional Science Association, but for a larger sample of commodities; (2) Creation of an alternative dependent variable built using data on acreage by commodity by county, rather than operation counts. Creating this new variable requires access to confidential Census of Agriculture data. Collection of the raw data was stalled by the Covid pandemic: the standard procedure had been to visit the NASS data repository in Washington, D.C., which was closed in 2020. In the summer of 2021, we received word that we would be able to use a license already held by the New Jersey Agricultural Experiment Station to access this Census data remotely. We have learned about the software available on the NASS Census server that would allow us to create continental scale indices using raw county scale data. We are currently writing the code that will allow us to emerge from online sessions with cluster indices for each commodity that will not compromise the confidentiality of the Census micro-data. GOAL #2: To create a typology of geographic clusters within production agriculture. A cluster-related typology of agricultural commodities was presented at the meetings of the Agricultural and Applied Economics Association on July 23rd, 2019. No additional work was done on the typology during this reporting period. The next task in this subproject is to conduct a literature review on pre-existing typologies of agricultural operations in the U.S., including typologies that have little to do with geographic behavior. GOAL #3: Using both case studies and the empirical results described in goals 1 and 2, to develop systematic recommendations on regional economic development practices for agriculture that are based on cluster theory. The proposed field work on the cranberry industry, conducted by co-PI Anil Hira, has been restarted. Dr. Hira's university and the governments of Canada and the U.S. have relaxed certain travel restrictions. Co-PI Neil Reid continues to collect materials for a more formal and comprehensive manuscript on "best practices" for organizing a local agricultural cluster. This is a continuation of the work he presented at the November 2019 meetings of the North American Regional Science Association. GOAL #4: To better inform climate resilience policy by recognizing that not all agricultural location behavior is strictly deterministic. The pending cranberry case study, utilizing an industry that has significant self-reported climate effects, will provide most of the research information for eventual achievement of this objective. GOAL #5: To advise USDA, NASS, and others on survey questionnaires and economic development approaches that take a fuller account of the roles played by innovation and knowledge-sharing in space. This activity is still pending.
Publications
|
Progress 08/01/19 to 07/31/20
Outputs
Target Audience:Land grant scholars, land grant extension personnel, USDA researchers and program leaders, industry leadership, individual growers. Changes/Problems:The proposed cranberry industry case study, to be managed by co-PI Anil Hira, cannot be completed before July 2021 grant deadline due to Covid-19 restrictions. This component of the grant is important enough that we plan to apply for a no-cost extension into summer of 2022. What opportunities for training and professional development has the project provided?Two post-doctoral scholars working at Penn State University have received funding under this grant. Both are co-authors on grant-related articles that are published and in the pipeline. A Rutgers doctoral student in Animal Sciences, Jennifer Weinert, provided valuable assistance on the equine study that was published in January 2020. Dr. Zheng Tian's main area of research on agricultural clusters is described immediately above. He has also been helpful in thinking through the regression and multivariate analyses that constitute deliverables under this grant. Dr. Elizabeth Dobis has been a significant asset to the research team, leveraging her expertise in NASS commodity datasets and GIS. In March 2020, Dr. Dobis left Penn State University to join the USDA Economic Research Service in Kansas City. Her work at ERS will focus primarily on rural public health. As time permits, she will continue to work on her grape/winery and hops case studies (see previous years' grant reports) and move them toward publication. How have the results been disseminated to communities of interest?A large number of presentations were given at scholarly conferences. Here is the complete list for the reporting period: Gottlieb, P.D., S. Goetz, N. Reid, E. Dobis, A. Hira, and Z. Tian. Innovative business practices and geographic clustering in a large sample of US agricultural commodities. Presentation at the meetings of the North American Regional Science Council, Pittsburgh, PA, November 14, 2019 (repeated on November 16). Reid, N., P.D. Gottlieb, E. Dobis, A. Hira, and S. Goetz, Identifying best practices for fostering collaboration in agricultural clusters. Presented at the meetings of the North American Regional Science Council, Pittsburgh, Pa, November 13-19, 2019. Gottlieb, P.D., J. Weinert, E. Dobis, and K. Malinowski. Thoroughbred and Standardbred stallions as key assets in racing industry clusters: Geographic dynamics 1995-2017. Presentation at the meetings of the North American Regional Science Council, Pittsburgh, PA, November 15, 2019. Tian, Z., P.D. Gottlieb, E. Dobis, A. Hira, N. Reid, and S. Goetz. Detecting Agricultural Clusters with the Spatial-IO Location Quotient. Presentation at the meetings of the North American Regional Science Council, Pittsburgh, PA, November 15, 2019. Dobis, E., N. Reid, A. Hira, P.D. Gottlieb, and S. Goetz. Trends in the co-location of grapes and wineries vs hops and breweries: 2002 TO 2017. Presentation at the meetings of the North American Regional Science Council, Pittsburgh, PA, November 15, 2019. Hira, A., and P.D. Gottlieb, N. Reid, S.J. Goetz, and E. Dobis. Understanding clusters as a source of industrial competitiveness. Presentation at "Industry Dynamics" Conference associated with Oxford University Press volume on this subject. Kyoto, Japan, September 25, 2019. What do you plan to do during the next reporting period to accomplish the goals? Expand the commodity dataset being used for the grant's regression analysis. As part of a special tabulation, NASS sent us agricultural census data on business practices by commodity in two installments, separated by approximately one year. We will add the second installment of commodity data to the pre-existing dataset. Visit a secure NASS facility to collect data on commodity-specific acreage. For our geographic analysis, we have been using data on the count of operations by county by commodity. Acreage by commodity is also of interest, but it is frequently suppressed at the county level. We do not currently know how NASS will handle this request in light of Covid social distancing requirements. Complete the grant's proposed regression analysis and write it up in manuscript form. Prepare the previously-presented commodity typology analysis for publication. Draft a White Paper that connects our theoretical geographic research to local agricultural development activities and to relevant policy programs of the USDA, both inside and outside of NIFA A1661. (Case study research on the cranberry industry is currently halted because of COVID-19 restrictions, including those related to international travel.)
Impacts
What was accomplished under these goals?
GOAL #1: To test the hypothesis that certain known drivers of transactional clustering behavior are active in agriculture: that they are correlated with observed geographic clustering, controlling for factors like climate requirements and industry size. Regression results addressing this hypothesis were presented at two sessions of the November 2019 North American Regional Science Meetings, held in Pittsburgh, PA. GOAL #2: To create a typology of geographic clusters within production agriculture. A cluster-related typology of agricultural commodities was presented at the meetings of the Agricultural and Applied Economics Association on July 23rd, 2019. A manuscript describing that work is being prepared for publication. GOAL #3: Using both case studies and the empirical results described in goals 1 and 2, to develop systematic recommendations on regional economic development practices for agriculture that are based on cluster theory. CASE STUDY RESULTS FOR THIS REPORTING PERIOD: The team's case study on craft breweries and the growing of hops has been published, with the following citation: N. Reid, S. Goetz, E. Dobis, P. Gottlieb, and A. Hira. 2020. "The rise of craft beer and its impact on demand for local hops," in A. Kratzer and J. Kister, eds., Rural-urban linkages for sustainable development. London: Routledge Press. This study found that the rise of craft breweries is increasing opportunities for growing hops outside of the crop's traditional concentration in the Pacific Northwest. The rise of the craft brewery industry has increased the interest in growing hops closer to multiple metropolitan areas. Some of the new hop growing is taking place at "farm breweries," which are the beer equivalent of winery-vineyards. Craft brewers have also increased their demand for new and different hop varieties. Because growing hops in Oregon benefits from both internal and external economies of scale, there are clear trade-offs between the benefits of maintaining this historical commodity cluster in the Northwest, on the one hand, and the creation of complete supply chains that are localized, closer to the brewer and the beer consumer, on the other hand. Two members of the study team collaborated with researchers at the Rutgers Equine Science Center to conduct a geographic analysis of registered Thoroughbred and Standardbred stallions standing at stud throughout the U.S. The citation for this published research is: Gottlieb, P.D., J.R. Weinert, E. Dobis, and K. Malinowski, 2020, The evolution of racehorse clusters in the United States: Geographic analysis and implications for sustainable agricultural development, Sustainability 494, January, doi:10.3390/su12020494. This case study found that the relative growth and decline of racehorse clusters in individual U.S. states is characterized by vicious and virtuous cycles--as cluster theory would predict. Against a backdrop of systemic decline in registered stallions since the mid-1990s, what has emerged is a "winner take all" map, with Kentucky dominating the Thoroughbred industry and Ohio dominating the Standardbred industry. Similar winner-take-all effects are evident at a substate scale. In addition, it appears that policies related to on-track gambling and sire stakes programs can explain which states became winners and which became losers, in terms of the long-term change in stallion counts.We currently have no evidence that the fate of registered racing stock in a particular state was shared by the much larger number of horses with no connection to racing. GOAL #4: To better inform climate resilience policy by recognizing that not all agricultural location behavior is strictly deterministic As stated in our July 31, 2019 progress report, our cranberry case study is the best way for this project to provide new empirical information on this critically important issue. Our map research shows that the North American cranbrerry industry is characterized by clearly-defined and separated geographic clusters, each of which is almost entirely within a single U.S. state or Canadian province. Growers are especially well organized at the cluster level: they identify closely with their own state, its land grant cranberry program, and its department of agriculture. The largest single purchaser of cranberries, the Ocean Spray coop, also recognizes this fact of multiple "grower communities." Ocean Spray seeks to maintain a diversified portfolio of growing areas, in part to guard against the effects of climate change. The challenge posed by climate change is evident to all industry stakeholders, because cranberries do best in cooler climates. The qualitative research being conducted by co-PI Anil Hira of Simon Fraser University, who is currently waiting for Covid-19 clearance, cannot fail to provide generalizable insights into agricultural cluster responses to climate change. GOAL #5: To advise USDA, NASS, and others on survey questionnaires and economic development approaches that take a fuller account of the roles played by innovation and knowledge-sharing in space. Recommendations to USDA and NASS based on our statistical analysis are pending. In order to advance a discussion of actionable approaches to agricultural development that are informed by cluster thinking, co-PI Neil Reid was given the task of presenting a conference paper on "Identifying Best Practices for Fostering Collaboration in Agricultural Clusters." This paper drew on Dr. Reid's extensive knowledge of the industry cluster literature, as well as his previous USDA-funded work helping to organize the greenhouse and nursery industries in northwestern Ohio. The citation to this presentation is: Reid, N., P. Gottlieb, E. Dobis, A. Hira, and S. Goetz, Identifying best practices for fostering collaboration in agricultural clusters, Presented at the 66th Meetings of the North American Regional Science Council, Pittsburgh, Pa, November 13-19, 2019. GOALS NOT ENUMERATED IN THE ORIGINAL GRANT PROPOSAL Penn State University post-doctoral research associate Zheng Tian of the Northeast Regional Center for Rural Development now receives some funding under this grant (see "opportunities for training" below). Dr. Tian has pioneered a new approach to identifying industry clusters on a national map. He has also applied this approach to agricultural industry clusters. In this second body of work, an important goal was to expand the supply chain beyond the production-based definition that is used, for example, by the National Agricultural Statistics Service. This requires the use of an input-output matrix. Dr. Tian, co-authoring with other investigators on this grant, has published one article on this topic and has presented preliminary results from a second paper. The citations are as follows: Tian, Z., P.D. Gottlieb, and S. Goetz. 2019. Measuring industry co-location across county borders. Spatial Economic Analysis, October, doi: 10.1080/17421772.2020.1673898 Tian, Z., P.D. Gottlieb, E. Dobis, A. Hira, N. Reid, S. Goetz. Detecting agricultural clusters with the spatial-IO location quotient, Presentation at the meetings of the North American Regional Science Council, Pittsburgh, PA, November 15, 2019
Publications
- Type:
Book Chapters
Status:
Published
Year Published:
2020
Citation:
Reid, N., S. Goetz, E. Dobis, P. Gottlieb, and A. Hira. 2020. The rise of craft beer and its impact on demand for local hops. In A. Kratzer and J. Kister, eds., Rural-urban linkages for sustainable development. London: Routledge Press.
- Type:
Journal Articles
Status:
Published
Year Published:
2019
Citation:
Tian, Z., P.D. Gottlieb, and S. Goetz. 2019. Measuring industry co-location across county borders. Spatial Economic Analysis, October, doi: 10.1080/17421772.2020.1673898
- Type:
Journal Articles
Status:
Published
Year Published:
2020
Citation:
Gottlieb, P.D., J.R. Weinert, E. Dobis, and K. Malinowski. 2020. The evolution of racehorse clusters in the United States: Geographic analysis and implications for sustainable agricultural development. Sustainability 494, January, doi:10.3390/su12020494.
- Type:
Book Chapters
Status:
Awaiting Publication
Year Published:
2021
Citation:
Hira, A., P.D. Gottlieb, N. Reid, S.J. Goetz, E. Dobis. Understanding clusters as a source of industrial competitiveness. Prepared for Oxford Handbook of Industry Dynamics, ed. M. Kipping, T. Kurosawa, and E. Westney. London: Oxford University Press.
|
Progress 08/01/18 to 07/31/19
Outputs
Target Audience:Land grant scholars, land grant extension personnel, USDA researchers and program leaders, industry leadership, individual growers. Our interaction with the cranberry industry, in particular, is helping us gauge industry receptiveness to some of our ideas. At the same time, each commodity (and geographic cluster within each commodity) has its own peculiar culture, so such findings will not automatically transfer. Changes/Problems:A special tabulation received from NASS in November of 2018 was missing several of the commodities originally requested, including horses and mushrooms. This problem was fixed in July 2019. Because we have two datasets, the first one partial and the second one complete, we will wind up doing a number of analyses over again on the larger dataset. A request to NASS for access to a secure facility to look at suppressed acreage data was not acted on by the NASS review board until July 2019, although it was originally submitted in February 2018. This lag did not affect our ability to meet any deadlines, because we were able to test all of our concepts using data on commodity operation counts, which are NOT suppressed. NASS has responded quickly and cooperatively when certain oversights on their part were brought to their attention. We are very grateful for this agency's willingness to work with us on this project, and to do so at a reasonable price. What opportunities for training and professional development has the project provided?We focus here on the activities of the two post-doctoral scholars working on this project, Elizabeth Dobis and Tian Zheng of Penn State University. In addition to on-site mentorship, professional development opportunities were provided in workshops hosted by Indiana University that included staff as well as representatives of the Economic Development Administration (Zheng). Results were presented at various professional conferences, including the American Association of Wine Economists (Dobis), as well as the various regional science conferences SRSA, WRSA and NARSC (both). Dr. Dobis attended the July meetings of the Agricultural and Applied Economics Association, where she was able to network and discuss her work on the geographic cluster indices required for this project. How have the results been disseminated to communities of interest?Book chapters, journal articles, and conference presentations in various stages of the publication pipeline. See products. What do you plan to do during the next reporting period to accomplish the goals?Planned presentations at the November 2019 meetings of the National Regional Science Association in Pittsburgh, PA: Regression analysis of observed clustering behavior as a function of commodity-specific business characteristics Locational behavior of grapes versus hops The rise and decline of historical equine clusters, measured using data on standing stallions Economic and commodity development practices informed by industry cluster theory Application of the Spatial-IO-Location Quotient (SILQ) to agricultural supply chains We will receive and clean a larger dataset on commodity business characteristics from NASS, which was received in raw form in July 2019. We will finish gathering data for the cranberry case study. We will refine our regression and typology papers by doing a visual scan of dot density maps for all commodities in the 2012 Census. With preliminary approval already in hand, we will apply for access to a secure NASS site in order to collect data on acreage by commodity that is suppressed in the public use file. We will begin building a theoretical model of agricultural clustering that takes both fixed factors and co-location into account. Ellison-Glaeser (1997) is our current template, but it has significant drawbacks for the case of agriculture. Research on grapes, hops, and horses will continue through toward publication.
Impacts
What was accomplished under these goals?
GOAL 1 On November 20, 2018, we received from NASS the census data that we requested on farm business practices organized by detailed commodity (these are the potential "drivers of clustering behavior" in our model). Although we did not receive data for all of the commodities we requested, we received enough observations to help us select a case study commodity and proceed with the promised empirical analysis. The first step was to clean the data and match it with our pre-existing data on operation counts by commodity by county. We received data on dozens of individual census questions, so we needed to reduce these business practice variables to a manageable number. The topical organization of the census survey, among other factors, suggested that it would be useful to create numeric indices to represent the following eight concepts: capital intensive, labor intensive, direct sales, value-added production, diverse sources of income, conservation practices, other innovative practices, and professionalized management. These indices for each commodity were created in early 2019. They serve as key inputs into both the regression analysis described under this goal and the multivariate analysis described under goal #2. A continuing challenge under this goal has been to develop metrics that reliably capture the kind of continental-scale geographic distributions we would expect to see as a result of so-called "social" factors, like proximity to urban populations or to other growers in the same industry. For example, a simple metropolitan/non-metropolitan county distinction appears to be collinear with the specialized growing conditions found in California and the Northeast. In late spring of 2019 we developed a new metric that distinguishes between highly-concentrated commodities that have multiple clusters, and those having only one or two. This new measure shows promise for identifying urban orientation. A second important body of measurement work is new to this grant. Penn State post-doc Tian Zheng has developed a technique that identifies particular industry clusters in particular places. The technique uses both input-output tables and measures of county proximity to ensure that identified sectoral clusters are both large enough and "connected enough" to matter. Note that the problem of placing on a map particular clusters that surpass some threshold of size or statistical significance is not a promised deliverable under this grant. Dr. Zheng's technique, however, highlights the critical issue of supply chain relationships in cluster thinking. More important, he will soon apply his technique to production agriculture, as an official contribution to the present project. During this reporting period, we finished collecting data on the weight/density of various types of produce. These data can be used to proxy the need for a particular commodity to be near urban markets---to the extent transportation costs are still relevant. GOAL 2 We now have our first set of empirical results related to this goal, which we presented at the meetings of the Agricultural and Applied Economics Association on July 23rd, 2019. We used our data on geographic distributions and business practices as inputs into a statistical clustering algorithm that classified more than 100 commodities into five groups (conference paper) and four groups (conference presentation). The results made sense, although they did not clearly identify knowledge-driven clustering behaviors that were distinguishable from known phenomena, such as urban orientation. The most robust groupings, differing form each other on both geographic behavior and business practices, were an innovative group of mostly specialty crops that engage in direct sales and value-added production and have a heavy orientation to metropolitan areas; this contrasted with a group of large field crops having none of those urban-oriented practices, large capital to labor ratios, non-metropolitan location (but with strong contiguity), and strong environmental practices. The commodities that fit into neither of these two categories have interesting attributes that warrant further study. GOAL 3 1) Case studies. Using maps and the business practice variables described under Goal #1, we have settled on cranberries as our primary case study commodity for this project. This commodity clearly clusters in five to six discrete locations in North America. It is well organized in its various states and provinces, and significant national leadership is provided (at least informally) by the Ocean Spray cooperative, a very large buyer of the crop. Co-PI Anil Hira of Simon Fraser University has written an interview guide and a mail questionnaire, which we are now reviewing with the help of industry insiders. Several interviews of trade association representatives have already been conducted. The interview phase of this primary research exercise is expected to be completed by September 2019. Dr. Hira's survey research is always aimed at measuring factors like cluster governance, information flow, levels of trust, and innovative capacity. His cranberry case study will therefore be extremely applied and actionable in economic development terms. 2) BONUS CASE STUDIES: Although not formally promised in the grant application, we have collectively been researching, presenting, and writing scholarly articles on the following agricultural commodities: wine grapes, hops, and racehorses. 3) Recommendations for economic development practice. Co-PI Neil Reid is writing a paper on the implications of geographic cluster theory for economic development in production agriculture. His own experience, as well as the economic development literature, suggests some prosaic but proven steps that help to build trust. These include such things as "hiring a Cluster Manager who is respected by cluster members, choosing a neutral meeting place for cluster meetings, and identifying a first collaborative project that has a high chance of success and clearly demonstrates the benefits of collaboration." GOAL 4 Many of the findings in this grant (not to mention a large pre-existing literature) will show that commodity location is not strictly deterministic. We have begun networking with climate/land use specialists so that we can address this issue more completely in a subsequent funded project. GOAL 5 A very preliminary conclusion we have drawn from the work described under Goal #2 is that the Census questionnaire does a better job of capturing certain correlates of urban orientation--direct sales, diverse sources of income like agritourism, and value-added production -- than it does capturing innovation, technology, and knowledge in a more general sense. A second conclusion is that our current statistical approach to the creation of a commodity typology dovetails nicely with earlier efforts by USDA to create a farm typology with even wider applicabilty than our own (e.g., the work of USDA analysts Hoppe, et al, 2000). For example, we see systematic relationships in our dataset that link fruit and vegetable production, exotic "hobby" commodities (bison for example), high off-farm income, smaller farm sizes, and less professional management.
Publications
- Type:
Book Chapters
Status:
Awaiting Publication
Year Published:
2019
Citation:
Reid, Neil, Stephan J. Goetz, Elizabeth A. Dobis, Paul D. Gottlieb, and Anil Hira. The Impact of the Craft Beer Revolution on the American Hop Industry, in Armin Kratzer, Jutta Kister, Frank Zirkl (eds). Rural Urban Linkages for Sustainable Development, Taylor & Francis Publishers.
- Type:
Book Chapters
Status:
Submitted
Year Published:
2020
Citation:
Hira, Andy, Paul Gottlieb, Neil Reid, Stephan Goetz and Elizabeth Dobis (in progress) Understanding Clusters as a Source of Industrial Competitiveness, in Kipping and Kurosawa (eds.) Handbook on Industry Dynamics (Oxford Press).
- Type:
Book Chapters
Status:
Other
Year Published:
2020
Citation:
Paul Gottlieb and Andy Hira (in progress) Recent industry dynamics in production agriculture, in Kipping and Kurosawa (eds.) Handbook on Industry Dynamics (Oxford Press).
- Type:
Journal Articles
Status:
Under Review
Year Published:
2020
Citation:
Tian, Zheng, Paul D. Gottlieb, and Stephan J. Goetz. Measuring Industry Co-Location across County-Borders , revised and resubmitted to Spatial Economic Analysis. Also presented at the SRSA Conference and the RSA Global Conference.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2019
Citation:
Dobis, Elizabeth A., Anil Hira, Paul D. Gottlieb, Stephan J. Goetz, and Neil Reid. 2018. "Grapes and Wineries in the United States: Geographic Location and Industrial Co-location." Working Paper. Also presented at AAWE conference.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2019
Citation:
Gottlieb, Paul, Elizabeth Dobis, Anil Hira, Stephan Goetz, Neil Reid, The new location theory applied to production agriculture: A multivariate approach. Presentation at the Meetings of the Agricultural and Applied Economics Association, Atlanta, GA, July 23, 2019.
|
Progress 08/01/17 to 07/31/18
Outputs
Target Audience:Regional scientists, Geographers, Agricultural economists, Community development extension professionals, Scholarly experts on wine and beer Changes/Problems:It was hoped that data on dollar sales by individual commodity could be used for statistical analysis, in addition to acreage data. Another possible use of sales data would be to calculate value-to-weight ratios for individual commodities. In the Census of Agriculture, however, sales data are available only at a higher level of commodity aggregation than that at which teh study is being conducted. Therefore, county sales data will not be used to weight any geographic cluster indices. Other data sources, such as retail prices, may be used to create value-to-weight ratios. What opportunities for training and professional development has the project provided?Three post-doctoral associates are actively involved in this research. One undergraduate research assistant was involved early in the project's first year. How have the results been disseminated to communities of interest?Five research presentations have been given at scholarly conferences throughout the world. Two draft research papers have been prepared and one of them has been submitted to a scholarly journal. Agreement has been reached on the writing of two book chapters. What do you plan to do during the next reporting period to accomplish the goals?We are waiting for a special tabulation to be completed by the National Agricultural Statistics Service (NASS). We are also waiting for clearance from NASS to visit a secure site to examine suppressed data on commodity acreage. Once we have these data, we can complete the proposed regression analysis and select new case study commodities. The bulk of the regression analysis will be done in the coming year. It is expected that new case study research will begin in spring of 2019.
Impacts
What was accomplished under these goals?
GOAL 1: Literature review. We have collected and analyzed a large number of articles and books on measurements of disproportionality or inequality that would be suitable for data grouped by U.S. counties. Equally important, we discovered that our plan to identify a statistical correlation between geographic clustering and its root causes in particular industry practices has been done before. Rosenthal and Strange (2001) did something similar for a set of U.S. manufacturing industries. Kim, Barkley, and Henry (2000) explored "industry characteristics associated with the clustering of establishments in three-digit SIC manufacturing industries in nonmetropolitan areas." Our statistical analysis will extend this work to the agricultural sector both inside and outside of metropolitan areas. Dependent variable. We collected data on operations by detailed commodity by county before submitting the grant proposal. These data are sufficient for implementing the proposed regression analysis. However, following works like Ellison and Glaeser (1997) and Daniel (2005), we would like to incorporate not only establishment counts, but also measures of operation scale into our geographic disproportionality measures. On February 26, 2018, we filed a formal request with NASS to access suppressed data on acreage by detailed commodity by county. These data will allow us create more sophisticated measures of continental-scale geographic clustering. This request to NASS is pending. Independent variables. On December 29, 2017, we submitted a formal request to NASS for summary data on 60 Census questions for each of 179 crop and livestock commodities. As of the date of this report this special tabulation is being performed by NASS personnel, although it has been delayed somewhat by the resource demands of the 2017 Census of Agriculture. It is difficult to find data sources outside the Census of Agriculture that can be used to compile explanatory variables at the level of commodity detail we require. We have, however, acquired a table from the University of Georgia that lists the shipping density of various plant and livestock commodities.We will supplement these data with a supermarket survey and use the numbers to proxy the cost of transportation and thus potential urban orientation of commodities. GOAL 2: A preliminary typology of clustering commodities was presented at the Meetings of the North American Regional Science Association conference in Vancouver, BC, on November 9, 2017. The typology was based on cluster metrics calculated across different kinds of areal units, such as climate zones or metropolitan versus non-metropolitan counties. In addition, to improve our cluster typology and geographic metrics, we have prepared national maps of operations for about 20 detailed commodities. GOAL 3: Case studies. This project will incorporate work on the geography of hops, wine grapes, and horses that was started by members of our team before the grant was awarded. Co-PI Anil Hira has been assigned the task of developing case studies on commodities in addition to these. Before selecting case study commodities, he will look at our regression results to identify those commodities that follow the norm and those that are outliers. The kinds of maps shown above will also be useful. Because case study research is time-consuming, it is important to select cases with care. We have therefore postponed our newer case study work until the regression results are in hand. Recommendations for economic development practice. Co-PI Neil Reid has been assigned the task of writing a White Paper on the implications of geographic cluster theory for economic development in production agriculture. Dr. Reid will address this question in the context of current federal programs for regional agricultural development and commodity marketing. The paper will benefit from Dr. Reid's significant experience in the field, as he worked to develop cluster-conscious collaborations among nursery operators situated in Northwest Ohio. Nothing to report on GOALS 4 and 5 for the current period. REFERENCES Daniel, Karine. 2005. "Elements Sur La Geographie de L'agriculture Aux Etats-Unis et Dans l'Union Europeenne: Les Productions Agricoles Se Concentretn-Elles (Stylised Facts on the Location of Agricultural Activities within the United States and the European Union: Do Agricultural Productions Agglomerate With English Summary)." Revue d'Economie Regionale et Urbaine, no. 4 (October): 533-56. Ellison, G., and E. L. Glaeser. 1997. "Geographic Concentration in U.S. Manufacturing Industries: A Dartboard Approach." Journal of Political Economy 105(5): 889-927. Kim, Y., D. Barkley, and M. Henry. 2000. Industry characteristics linked to establishment concentrations in nonmetropolitan areas. Journal of Regional Science 40(2): 231-259. Rosenthal, S., and W. Strange. 2001. The determinants of agglomeration. Journal of Urban Economics 50: 191-229.
Publications
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2017
Citation:
"Toward a Typology of Agricultural Industry Clusters in the United States." Presented at the meetings of the North American Regional Science Association, Vancouver, British Columbia, November 9, 2017.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2018
Citation:
"Craft Beer and the Geographic Expansion of Hop Production." Presented at the annual Conference of the Mid-Continent Regional Science Association, Kansas City, Missouri, June 6-8, 2018.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2018
Citation:
"The Rise of Craft Beer and its Impact on the Demand for Local Crops." Presented at the International Geographical Union Commission on the Dynamics of Economic Spaces Conference on Rural-Urban Linkages for Sustainable Development: An Economic Geography Perspective, Innsbruck, Austria, July 19-21, 2018.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2018
Citation:
"Craft Breweries and Urban Tourism." Presented at the meetings of the National Association of Community Development Extension Specialists, Cleveland, Ohio, June 11, 2018.
- Type:
Conference Papers and Presentations
Status:
Other
Year Published:
2018
Citation:
"Grapes and Wineries in the United States: Geographic Location and Industrial Colocation." Presented at American Association of Wine Economists, 12th Annual Conference, June 10-14, 2018 (presented on June 12, 2018), Ithaca, NY, USA
|
|