| Title |
Investigators | Department | Objectives | Approach Keywords | Progress Reports | Impact Statements | Publications | |
Project * COL00641 | |
| Title | *Market and Policy Analyses of Colordo Products Using Time Series and Panel Data Models |
| Investigator(s) | Davies, SP; |
| Department | Agricultural and Resource Economics |
| Objectives | The main goal of this project is to extend and apply several recent econometric methods to examine issues in international agricultural trade, marketing and policy applications of relevance to Colorado producers and agribusinesses. To reach this goal, there are several objectives: (1) to examine the US wheat, lamb and potato markets using cointegration models that have enhanced structural characteristics and can give evidence of strategic behavior within markets and among agribusiness firms; (2) to develop panel data models looking at, first, the impact of regulation on the state-level development in animal feeding operations for four different livestock industries and, second, changes in China's demand for feed grains in the face of market liberalization in the World Trade Organization (WTO); (3) to demonstrate important features in the econometrics literature that have not been used in empirical applications: with regard to cointegration models, effects of sample size on statistical performance will be evaluated as will the ability to use alternative normalization methods to better portray domestic demand and supply functions as well as export markets; with regard to the panel data methods used, the need to test for exogeneity of independent variables in random effects models will be shown, as will alternatives for estimating time-invariant variables in panel data models. |
| Approach | This project will use recent econometric and statistical developments in areas of time series models, especially cointegration analyses, and also recent panel data modeling approaches. The primary emphasis will be on a series of empirical applications, but several technical aspects will be addressed as well. With regard to time series models, this project will demonstrate the usefulness of cointegration methods in standard market applications. Moreover, it will use recent literature to demonstrate the increased flexibility in modeling market behavior associated with alternative normalizations of the system. Also, the importance of sample size will be examined via Monte Carlo methods because cointegration models often require the estimation of a large number of parameters. Neither of these aspects have been appreciated to their fullest in the agricultural economics literature. The second modeling approach, that of panel data models, has been used in econometrics for many years. While the number of applications has grown, there are several possibilities not appreciated, in particular the ability to use time-invariant variables in models. There also appears to be a continuing failure to test correctly for correlation between independent variables and the error term. These aspects will be demonstrated in several of the empirical applications. |
| Keywords | Panel data models, cointegration, wheat, Livestock, potatoes, lamb |
| Progress Reports | |
| 1996 | Research has proceeded in the area of dry bean exports using Johansen's cointegration techniques, with the main focus being a Phd dissertation by J. Rhodd. The main conclusions will relate to the nature of export shocks and the extent to whether they are stationary or non-stationary. Along with researchers from Nebraska, a session looking at various dimensions of the dry bean market will be submitted to the Western Agricultural Economics meetings. Also, cointegration techniques will be used to examine boom-bust cycles in the international market and whether the current changes in GATT and more general liberalization of world markets will lead to more stable or less stable international prices. Additionally, models using simultaneous panel data techniques that can be applied to world markets for selected agricultural commodities are being developed through the dissertation work of L. Chengula and S. Vickner. |
| 1997 | This project's emphasis is to assess the potential of several econometric methods in analyzing world markets that could be useful for Colorado commodities. The dissertations by L. Chengula and S. Vickner began this examination in assessments of simultaneous panel data models in the world cashew market and the domestic market for spaghetti sauce. Chengula's dissertation employed a non-linear 3SLS methodology using panel data to examine the behavior of India as a price leader in the world market for cashew kernels. He showed that the exercise of market power by India affected world price, but the dominant firm behavior varied across time. It was estimated that the price markups led to deadweight losses of about 6 percent of total sales. Vickner's dissertation spelled out a full-system methodology to determine how firms selling spaghetti sauce in ten US markets react to price changes by other firms. The market was modeled as a product differentiated oligopoly that appeared to have three different segments, premium, health conscious, and traditional. The price reaction elasticities and extent to which market power was exercised among five different brands was determined. This appears to be a methodology that could be transferred to international markets in commodities of interest to Colorado. Also, work proceeded using Johansen's Cointegration approach for the world wheat market, exports of pinto beans from the US and in several macroeconomic applications. Finally, a methodological extension of the subsector approach looking at strategic alliances was investigated in the dissertation of M. Luzius. In that research, he created a conceptual framework to evaluate strategic alliances and applied it to the marketing and distribution firms that sold organic products. |
| 1998 | The main objectives of this project are to analyze commodities of interest to Colorado producers and agribusinesses using recently developed econometric models and to refine the modeling techniques as necessary. The commodities included will be wheat, dry beans and selected differentiated products that are sold in retail outlets. These products will be examined using simultaneous equations panel data, traditional time series and cointegration models. The basic data set for a model of the US wheat market was collected and its fundamental properties were determined. Most of the variables appear to be non-stationary and thus are appropriate for use in the cointegration model being constructed. The model has three basic equations, including a domestic demand equation, a domestic supply equation and a demand for US exports. More cointegrating vectors are expected given that over ten variables are included in the model, and these may be long run relations between different prices and also perhaps a Apolicy@ vector. Additionally, substantial progress was made in estimating a cointegration model of the US dry bean export market, in which four main bean varieties were included. This should be finished during the coming year in the Ph.D. thesis by J. Rhodd. The main publications of relevance in 1998 included a book chapter on foreign direct investment by US agribusinesses, published with P.K. Bash, and a methodological article looking at methods of simultaneous panel data models, published with S. Vickner. In addition, however, several completed manuscripts have been circulated as working papers and are under review in journals. These are primarily focused in the area of strategic behavior in US agribusiness. One article examining the strategic behavior of firms selling a differentiated product at retail (spaghetti sauce) has been accepted in the Journal of Agricultural and Applied Economics, and two others looking at imperfect competition issues with differentiated products have been submitted to journals and circulated as working papers at the University of Kentucky. These articles all make use of special scanner data sets developed by IRI International, and are focused on examining the nature of pricing behavior when firms interact in an oligopolistic manner, which is summarized in the price reaction functions they have with other firms. The data sets used in these analyses are unique in their ability to identify short term strategic reactions, products that are competitively related, and the time dimensions of price leader and follower behavior. This should provide useful information to agribusinesses operating in the product lines analyzed. |
| 1999 | The main objectives of this project are to analyze markets of interest to Colorado producers and agribusinesses using recently developed econometric models. The products are examined using simultaneous equations panel data, traditional time series and cointegration models. Three research efforts analyzed the pricing and merchandising behavior at retail for selected differentiated products. These have resulted in published or submitted journal articles. One of the unique aspects of this work is that all articles make use of special scanner data sets developed by IRI International. The data sets used in these analyses are unique in their ability to identify short-term strategic reactions, products that are competitively related, and the time dimensions of price leader and follower behavior. One article published during this year in the Journal of Agricultural and Applied Economics examined the strategic behavior of firms selling a differentiated product at retail (spaghetti sauce). We examined five major brands of spaghetti sauce using a simultaneous panel data model that had ten equations (five supply and five demand equations). We found three market segments existed, where pricing strategies implemented by firms were followed closely by other brands within the segment. Using weekly, disaggregated data, we found high elasticities of supply and demand, evidence that firm sales are sensitive to pricing tactics. There was little overall market power found, but it tended to be higher among the major brands in the market, Ragu and Prego. About one third of the market power of most brands was estimated to come from tacit price collusion. An exception was Hunt's, that despite being a low priced brand, exhibited more market power from having its own market niche. A second article, due to come out in Agribusiness: An International Journal during this coming year, examined the pricing behavior in a classic duopoly, where two weekly price series taken from the canned pineapple sales of Dole and Del Monte were compared. Controlling for holiday and seasonal effects, we analyzed the behavior of these price series using Granger causality techniques and cointegration models. We found many perspectives on the price leadership role of Dole. For example, Dole maintained a mark-up of about 35% over the time period, and it was clear that Del Monte reacted to price changes of Dole while the reverse was not true. Also, the importance of using weekly data was seen as well, as most of the reaction to shocks to one price series occurred within less than twelve weeks. Finally, a third article under submission to the Journal of Food Distribution Research examined a brand-level demand model of the US frozen vegetable market to estimate market power and the returns to retailer coordination in this market. We found that more market power and higher price-cost margins were found in the private label brands, as opposed to national brands such as Bird's Eye and Green Giant, which gives retailers a competitive advantage with respect to pricing, promotions, merchandising and shelf-space allocations. |
| 2000 | The main objectives of this project are to develop and implement several panel data and times series econometrics methods and apply them to areas of interest to Colorado producers and agribusinesses. Johansen's cointegration methods were examined in two different articles (Cutler, Davies and Schmidt; Cutler et. al). In Cutler, Davies and Schmidt, we showed that the forecasting with Vector Error Correction (VEC) models is improved by imposing restrictions. In Cutler et. al., we used Monte Carlo techniques to examine Johansen's HJ test that is designed to jointly test for the existence of long run equilibrium vectors. We found that, in small samples, the number of cointegration vectors is severely understated by trace and maximum eignvalue tests, but that tends to disappear with sample size. The HJ test is biased in small samples, but more importantly, joint tests using this method are far too infrequently accepted . Even with a sample size of 600, HJ tests were only accepted 46.6% of the time, even though the null hypotheses that were used were in fact true. The remaining publications are empirical examples of the time series methods that should be of use to Colorado producers and agribusinesses. Using national level point-of-purchase data from scanner data sets, Vickner and Davies developed an empirical model of the canned pineapple market to examine price leadership behavior. Often, game theoretic models of this type have been complex and required simulation approaches to implement them. Our approach is much simpler, being based on vector error correction models, and demonstrated aspects of the strategic behavior of firms operating as product-differentiated oligopolies. We found that, as expected, Dole is a leader in the canned pineapple market and Del Monte follows. It was also clear that strategic behavior does not show up in the typical quarterly data often used, as most strategic interactions had run their course within thirteen weeks. A second paper (Vickner et. al.) compared market power and pricing conduct in the frozen vegetable market between national brands and store or private label brands. We used simultaneous equations methods but the same type of scanner point-of-purchase data. The model included brand-level demand equations with endogenous prices and showed that the price-cost margins for private labels are slightly greater than the national brands, indicating a small competitive advantage for those operations, and perhaps helping explain the growth of these products in retailing. The final empirical research done under this project was a time series model looking at the effects of lamb imports on prices and quantities of US lamb and wool markets (Marshall, Davies and Thilmany). This has been a contentious issue, with accusations of dumping and countervailing duties imposed on lamb from Australia. Using a Vector Error Correction model, this paper demonstrated that that indeed Australian and New Zealand lamb producers react to prices and quantities in the US, but they react in an equilibrating manner, rather than creating excessive supply and low prices from dumping. |
| 2001 | This project extends several recent econometric methods to examine issues of relevance to Colorado producers and agribusinesses. Broadly, these methods relate to time series approaches and issues in panel data estimation. A time series paper done with S. Vickner of the University of Kentucky was accepted in Agribusiness: An International Journal and will be published in 2002. This paper developed a vector error correction model using weekly point-of-purchase scanner data to investigate pricing relationships among brands in the domestic black and herbal tea industry. Johansen's likelihood ratio cointegration test established the existence of a long run relation between prices of Bigelow black tea, Twining black tea, and Celestial Seasonings herbal tea. Thus, these three companies look across narrow markets in making pricing decisions, which is consistent with an industrial organization concept called multi-market contact. The analysis suggests Bigelow leads and both Twining and Celestial Seasonings follow. Other papers focus on panel data approaches. One paper has been published this year, which used a panel of sixteen dairies over five years to evaluate medical and health costs in dairy farms. The main results showed that health care costs remained relatively constant from 1993 to 1998, but they increased with the use of Bovine Somatropin (BST) and decreased as the herd size grows (Mizuta, et. al.). Three papers with D. Park and A. Seidl of Colorado State University (CSU) highlight the use of time-invariant variables in panel data models. Data on key elements of the livestock industry have been collected over the years 1967 to 1997. Because policy effects cannot be collected across time, the recent policy setting is used to act as a proxy for the written stringency of regulations as well as actual enforcement. We used five variables to reflect the policy setting: the number of regulations on the livestock industry in a given state; the number of personnel in regulatory agencies related to the livestock sector in each state; the existence of right to farm legislation; and the existence of anti-corporate farming laws. These effects were used to show that policy influences are among the more important factors that determine the location of the livestock industry over the past several decades and that they affect pork more than beef. The final panel data analysis done during this past year is an examination of the effects of using municipal biosolids versus commercial nitrogen fertilizer in wheat production. The paper was based on a panel of data from two different sites in Eastern Colorado, which had distinctively different performance characteristics. This research has been reviewed in the Agronomy Journal and was written jointly with H. Lagae and K. Barberick of CSU. The paper used regression techniques for panel data to demonstrate the economics of commercial fertilizer use versus biosolids use. In particular, these results demonstrate the sensitivity of policy analysis to the choice of whether to pool the two sites or not. |
| 2002 | The research examined three non-stationary price series in the black and herbal tea markets, including Celestial Seasonings, a Colorado company. We asked whether multimarket contact, when firms compete in several different product lines, exists. If present, then defection from tacit price collusion by one player is punished by others in all related markets, so collusion will continue as costs of defection are high. This research should help agribusiness firms understand pricing tactics in their markets, and provide initial quantitative perspectives on the speed of reactions in prices and which firms have price leadership positions. Weekly, point-of-purchase scanner data for Twining, Bigelow and Celestial Seasonings prices were analyzed. Celestial Seasonings is an herbal brand, while the others are black tea products. Because the three series had unit roots, they potentially were linked in the long run, which can be assessed using Johansen's cointegration models. Two cointegrating vectors were found. The first vector linked Celestial Seasonings and Bigelow's prices, suggesting multimarket contact, and a second vector tied the Bigelow and Twining prices together, showing a long run relation in black tea. The coefficient signs inferred that prices move together positively. Speed of adjustment coefficients showed that Celestial Seasonings prices moved to counter disequilibria with Bigelow's price, and Twining's price moved in the black tea market. Thus, Bigelow appears to exhibit price leadership in both markets, and the other firms have to adjust (and do) to its pricing tactics. Finally, impulse response functions showed that, when Celestial Seasonings raised price, Bigelow followed quickly over the following thirteen weeks, implying the existence of multimarket contact. However, Celestial Seasonings and Twining were not able to react as strongly to the moves of Bigelow, further confirming its leadership position. A second emphasis was on effects of livestock operations on housing prices in Weld County, Colorado. A hedonic model of housing prices was developed using spatial econometrics techniques. The data were housing sales over three years, including key housing attributes, along with GIS calculations of the proximity of each sale to livestock firms, with the latter distinguished by species and size. The impact of livestock feeding on housing prices could be negative because of odor, or positive if the firm created an amenity for rural lifestyle, or if the firm added local job opportunities. These factors might vary according to species. Lower income housing had either little significance related to location, or the results were slightly positive. Thus, proximity to livestock operations, perhaps related to jobs, raised housing prices. For higher priced housing, the number of livestock operations, especially if beef or dairy operations, gave positive effects, presumably due to the amenity effect. Hog feeding, especially when in large operations, had a clear negative effect on higher priced housing. |
| 2003 | During this year, two pieces of research were published as papers or abstracts. First, a policy analysis of five regulatory effects on the location of livestock inventories across the US was published in the Review of Regional Studies. The data used a balanced state level panel from 1969 to 1997 for 48 states. The main statistical challenge was that the regulatory variables are time-invariant, as they had only been collected from a survey in 1998. Therefore , we used Hausman-Taylors method, so that this paper is one of the few applications of this powerful method for panel data in the literature. Two policy variables affected the size of animal inventories in positive ways: the number of regulations on the livestock industry in a state; and, a dummy variable for the existence of an anti-corporation farming law. We argued that these factors follow the location of livestock in a state. Thus, more livestock leads to more regulation and a greater likelihood that an anti-corporate farming law exists. Two other variables were negatively related to livestock inventories: the number of staff available to enforce regulations, and whether a given state levied fines or not. Thus enforcement of existing regulations had a logical negative effect on livestock industry location. A fifth variable, which captured whether states have agricultural zoning laws or not, was not significant. The overall conclusion is that actual enforcement of environmental regulations reduces the likelihood that an industry chooses to locate in that state. The second research output under this project revises a method proposed by Arnade and Pick (1998, Agricultural Economics, hereafter AP) to look at nonstationary determinants of seasonal unit roots. The contribution of this research is to identity problems with the AP approach, and then to propose a better method. Like AP, our approach distinguishes between nonstationary factors affecting amplitude versus phase shifts, that is, factors that influence permanent changes in the length of the season versus changes that cause the peak of a season to grow. The paper used grapes imported from Chile into the US as an example of the method. We used non-linear least squares, which appeared to be a stable, easily interpreted and easily implemented approach. The results showed that real interest rates were negatively related to the amplitude of the season, suggesting that the costs of holding inventories shortened the season permanently. Inflation decreased the amplitude, or peak, of the season during which Chile exports grapes to the US. This makes sense if inflation is not fully reflected in exchange rates, which was the case during the period examined. In further work, we will examine imports from Mexico and use more industry-related data. One intriguing possibility is the potential to examine the determinants of non-stationary behavior at the zero frequency level. This is something rarely touched in the large time series literature on cointegration because of the difficulty in extracting and understanding the common trends in those models. |
| 2004 | In a study of beef quality (Feuz et al., Sitz et al., Umberger et al.) consumers tasted steak samples in an experimental auction to determine willingness to pay for tenderness, marbling, aging and country of origin. Each consumer participated in multiple auctions over six time periods, so a unique experimental panel data set was obtained. Marbling and tenderness had statistically significant impacts on consumers palatability ratings for steaks. Tenderness significantly and positively impacted consumers willingness-to-pay values. There appear to be threshold levels of marbling and tenderness below which consumers discount steaks. Steaks from Australia were rated lower than U.S. steak samples. Dry-aging methods negatively impacted taste panel ratings and bids. Consumers were willing to pay more for the domestic steak, but market segments exist and some consumers prefer the taste and are willing to pay a premium for beef from Australia (19% of the consumers) and Canada (34% of the consumers). The Livestock Mandatory Reporting Act (LMRA) led to collection of retail scanner prices and volumes for meat products, and also feature (discount) prices and the percent of volume sold under feature. An analysis using this data was done by Ziehl (2004). Past and current feature and posted prices had significant coefficients across four primal cuts examined. However, there is evidence of varying relationships between these prices across primal cuts of beef. For example, the feature and syndicate prices for rib primal cuts interact, in that as one goes up, the other increases by about .5 percent. However, despite significance, in ground beef, the net cross price effect was zero. In rib primal cuts, as retailers increased the volume under featuring, they also decreased the feature price and increased the posted price. Thus, retailers may set the stage for later overall price increases when featuring expires. Another results is that a one percent increase in volumes sold under featuring raised total volumes sold by about one percent in some categories. Finally, there was also evidence that consumers stock up when steep discounts occur. Another research effort examined the economic contribution of subtherapeutic antibiotics via a sector model whose parameters are econometrically estimated. Removing subtherapeutic antibiotics (specifically Tylan and Rumensin) from feedlot cattle feed would increase the feed costs by $5.75 per head with losses of feed efficiency of $1.75 per head and losses attributed to increased liver abscesses of $4.00 per head. Placing these production changes into a beef sector model (with non-cattle prices held exogenous) a 5% increase in feed costs results in a 0.225% smaller cow herd, a 1.3% decrease in heifer additions to the herd, a nearly 0.1% decrease in overall beef slaughter and a 0.129% decrease in average carcass weights. Using previously published elasticity estimates, and a reduced beef output of roughly 16 million pounds, lost revenues to the industry are roughly $45 million when subtherapeutic antibiotics are banned |
| 2005 | Several papers in the project used panel data to evaluate consumer value for beef attributes. Results showed that WTP for Certified U.S. products is small, but slightly above costs of a labeling program. Other findings indicate consumers likely to purchase natural beef, and the factors motivating different segments. These studies illustrate market research useful for beef producers seeking value-added marketing opportunities, and portray the types of consumers who are fueling the growth in natural meats in the U.S. Experimental auctions were conducted in 2005 to determine the value of visual versus taste attributes in grass-fed beef. This panel will be analyzed to determine why certain consumer segments purchase and have a greater WTP for grass-finished beef. The importance of beef attributes, such as nutritional information or production practices (e.g. grass-fed, natural, organic), which can be labeled, will be compared to traditional beef quality attributes. One issue in this type of research is the extent of bias that exists when consumers are asked about new or hypothetical products . Methods are being developed based on the Orbit Procedure presented in a 1997 Journal of Econometrics article by Klein and Sherman to provide a more rigorous correction procedure for hypothetical bias in estimated WTP. One other focus of the project was the economics of animal disease. This involved first an extensive literature review and typology of animal disease research in economics. In general, animal disease studies posit outbreak scenarios defined by three elements: the location of the outbreak; timing between outbreak and recognition/response by health officials; and the strategy used to prevent and contain the outbreak. A time series of retail meat purchases was used to test whether consumer purchases were altered by the BSE event. Results suggested that hamburger purchases were affected, but pork purchases increased. The second research thrust compared a simple dummy variable for the BSE event to a media index capturing negative media coverage. In general, the dummy variable was at least as good at capturing the effect of the BSE outbreak on consumer purchases. The final area of the project considers strategic pricing in the retail meat case, where an econometric model analyzed the interaction between feature prices, syndicate (shelf) prices, and the percent of discount as pricing tools. Given seasonal demands and diverse meat cuts, managers may form strategic pricing groups and use these pricing mechanisms in different proportions. Feature pricing has a positive, significant relationship with volumes sold, a surprising result. This could result from an omitted variable or the complex interaction of pricing variables. Feature prices are associated with increasing syndicate prices, so that they might be used as a price discrimination tool. Thus, retail meat managers can use features to lower prices below consumers reference prices occasionally, but still can extract greater margins from a higher syndicate price when the meat is off feature. It should be noted, however, that the effect of feature prices on syndicate prices is quite small. |
| 2006 | Several time series applications were completed during last year. First, a study using weekly USDA-AMS fed cattle price data from 1992 to 2006 for five U.S. regional fed cattle markets, including Colorado, investigated effects of mandatory price reporting on spatial market integration. Results indicated that these markets have been, and remain, highly cointegrated after implementation of mandatory price reporting. Following introduction of mandatory price reporting, the five regional fed cattle markets have become more fully integrated. Secondly, the economic impact of food safety scares, specifically, bovine spongiform encephalopathy (BSE) in domestic cattle, was examined using time series techniques. Retail meat purchase data was used to test if consumer purchases were altered by the BSE event. Results suggested that hamburger purchases were adversely affected while pork purchases increased. The second research thrust compared a simple dummy variable representing the BSE event to a media index variable simulating the weight of negative media coverage. The simple dummy variable captured the effect of the BSE outbreak on consumer purchases better than the media data. Several research activities in this project provided perspectives on key Colorado products. Colorado cooperatives are important to producers of potatoes, wheat, corn, sugar beets, and millet. CSU researchers obtained consolidated financial statements for 505 cooperatives from 1995 - 2005 and calculated DuPont profitability ratios. Findings suggest that cooperatives compete against one another to capture equity and business from producer-members, to garner inputs and services from regional cooperatives, and to secure financing. The first quintile firms had, on average, a rate of return to local equity that was 7% higher than the next quintile and much higher than the remaining quintiles. Three other research activities examined consumer behavior related to beef products. First, choice experiments analyzed US consumers' preferences and willingness-to-pay for meat attributes in labeled ribeye beef steaks. Consumers value USDA food safety certification more than any of the other choice set attributes, including country-of-origin labeling, traceability and tenderness. As a result, indication of origin may only become a signal of enhanced quality if the source-of-origin is associated with higher food safety or quality. Secondly, survey data from Colorado consumers determined market segments for various natural beef products. Almost two-thirds of the differences among responses regarding the importance of meat attributes were related to production practices (e.g. use of antibiotics, hormones and environmentally friendly grazing) . A third analysis was done to determine sensory preference and value of fresh beef steak differing in aging technique. No significant differences for sensory traits of flavor, juiciness, tenderness, or overall acceptability were detected between wet-aged Choice samples and dry-aged Choice samples. Although more consumers preferred wet-aged samples, markets exist for dry-aged beef and consumers are willing to pay a premium for this product. |
| Impact | |
| 1999 | This research should help clarify and quantify pricing behavior in product-differentiated oligopolies in the US food system. The use of weekly, disaggregated brand level data affords better perspectives on firms behavior. The potential returns and extent of competition in these markets can be seen in this research. Agri-business firms will benefit from the research through enhanced understanding of merchandising and pricing impacts. |
| 2000 | The work on Johansen's cointegration approach will help empirical research, as the technique is appealing due to its ability to examine long run behavior. The empirical work should help Colorado agribusinesses understand pricing behavior in product-differentiated markets and avoid costly price wars. Finally, the lamb import research helps Colorado lamb producers understand their market. |
| 2001 | Understanding the driving forces behind changes in markets and the effects, unintended and intended, that changes in policy have, requires some type of modeling. These models can be of assistance to producers, agribusinesses and policymakers. The work done here provides examples of the range of Colorado participants in the food and fiber system that can be assisted through economic research. The results of the tea analysis will be presented personally to Celestial Seasonings, a Colorado company, to see if they find it of interest, and we will try to understand how they might use these results and if there other food manufacturers in Colorado could benefit from these analyses. Farmers can make use of the evaluation of health costs in dairy farms from the publication partially supported by this project and by presentations of extension specialists who use the research. Similarly, the analyses done for the biosolids paper will be summarized in extension bulletins and can be used as the basis for determining the regional potential for municipal biosolids use in Colorado. Policymakers in Colorado will be able to use the results from the livestock regulation studies to assess Colorado's relative position compared to other states in regard to regulation of livestock and some of the economic implications of that position. |
| 2002 | This research should help agribusiness firms understand pricing strategies in their markets, and provide quantitative perspectives on the speed of price reactions and firms that have price leadership positions. Processors such as Celestial Seasonings can promote related products to avoid costly price wars and strengthen their position in essential markets. Retailing firms can reflect this strategic price behavior correctly in their own promotion and pricing efforts. Research of this nature also can provide baseline outcomes for policy discussions regarding costs of promotion and oligopolistic behavior. Finally, the development of these methods permits a greater range of commodities to be covered and more issues to be analyzed. From a technical standpoint, this modeling approach has not been used to examine multimarket behavior and, thus, insights such as price leadership and reaction speeds have not been analyzed. With extensions and access to high quality data, these analyses can provide significant insights into pricing tactics by food processing and retailing firms and their welfare impacts. The second part of this research should help livestock producers determine some external costs and benefits of their location decisions. It should also help Weld County officials determine zoning costs related to livestock. Developers could benefit from this research, as they can determine the impacts of building variously-priced housing on the urban fringe. The spatial econometrics dimension should have numerous applications in the fast-evolving regional area of analysis. |
| 2003 | Environmental and agricultural policymakers at the state level should benefit from the livestock location paper because they can compute several heretofore unidentified economic costs associated with added environmental regulation and enforcement. Also, businesses will be able to assess effects on their output markets or input suppliers if these firms are subject to increased regulation or enforcement. Finally, from a technical vantage, it would not have been possible to do this analysis at all without the Hausman-Taylor technique. The second paper will provide an improved view of sources of change in the international competition to fruit and vegetable production here in the US. Businesses and policymakers can see the longer-term factors at work as a result of this work. Again, these insights are only available from the newer techniques that are being used. |
| 2004 | The red meat industry has seen many changes in the last decade, including new retailing strategies by supermarkets, new branding programs initiated by producers, packers and retailers, and changing consumer demand for various meat attributes. Like many food products, meat at the retailer can no longer be seen as a non-differentiated commodity. Also, the complexity of competitive and strategic interactions among different sectors of the meat industry has grown. Wal-Mart and other non-traditional retailers, from Whole Foods to club retailers, have changed the competitive landscape. The first and third research efforts help to shed light on the dynamics within the meat industry. We show some of the characteristics important to consumers through the research on the taste tests, and our pricing models have begun to illustrate the complexity of retailers strategies. Practically speaking, retail meat managers can use features to lower retail meat prices below consumers reference prices occasionally, but still must be able to extract an additional margin from the syndicate price when the meat is off feature. The second piece of research reported is thought to have two impacts. First, it provides economic analysis to supplement existing discussion regarding a proposed federal ban on sub-therapeutic antibiotics in livestock feed, a policy that McDonalds Corporation has pursued voluntarily since 2003. Second, the beef sector model is a foundation for additional sector impact models, especially those concerning animal diseases such as Foot and Mouth Disease. |
| 2005 | The panel data studies in this project illustrate the types of market research useful for beef producers seeking value-added marketing opportunities, and portray the types of consumers who are fueling the growth in natural meats in the U.S. The methods being developed based on the Orbit Procedure should provide a more rigorous correction procedure for hypothetical bias in estimated (WTP). The animal disease research provides stakeholders with insight into the consumer response to animal disease events. Quantifying consumer effects aids stakeholders in weighing benefits of animal identification systems against its costs. The study was published both within the international agriculture business literature, and as a book chapter from CABI publishing. Exposure of the literature review has informed industry stakeholders and economic modelers of existing approaches and future opportunities for modeling improvement. The research on retail pricing dynamics uncovers the relative responsiveness of consumers to pricing strategies assisting Colorado beef producer and retailers in determining optimal product and pricing strategies. |
| 2006 | The research provides stakeholders with insight into the consumer response to animal disease events. Quantifying consumer effects aids stakeholders in weighing the benefits of animal identification systems against its costs. Colorado cooperatives provide valuable input purchasing and marketing services to Colorado farmers. Moreover, these cooperatives are an integral part of the rural Colorado economic landscape. The research results from this area of the project have been directly applied to the training of board members for Colorado cooperatives through yearly Director Certification Programs. The educational emphasis is on financial decision making and improving the financial performance of Colorado cooperatives. The work on mandatory price reporting should help in the evaluation of this relatively new legislation, and provide useful results for other production-distribution systems that are becoming more vertically integrated. The various analyses of consumer responses to beef attributes will provide marketing information for processing and retailing forms as they try to design new products and understand their markets. As the researchers also have Cooperative Extension appointments, they will produce non-technical summaries of the research in a variety of fact sheets and presentations. |
| Publications | |
| 1997 |
CHENGULA, L. 1997. Estimation of Oligopolistic Behavior in the Cashew Kernel Export Market: A Simultaneous Equations and Panel Data Approach. Ph.D. Thesis. Colorado State Univ. Fort Collins, CO. 211p CUTLER, H., DAVIES S. and SCHMIDT, M. 1997. The Demand for Nominal and Real Money Balances in a Large Macroeconomic System: Evidence from Cointegration Analysis. Southern Economic Journal. 62(4):947-961. April CUTLER, H., DAVIES, S., RHODD, J. and SCHWARM, W. 1997. The Demand for M1 in a Large Macroeconomic System: Evidence from Cointegration Analysis. Journal of Macroeconomics. 19(1):53-78. Winter LUZIUS, M. 1997. Complementarity, Learning Effects, and Strategic Alliances in the Organic Food Marketing Channels. A selected paper presented at the American Agricultural Economics Association Meetings, Toronto, Canada. 21p. August LUZIUS, M. 1997. Strategic Alliances in the Organic Food Marketing Channels. Ph.D. Thesis. Colorado State Univ. Fort Collins, CO. 231p VICKNER, S. 1997. On Estimating Welfare Losses in Imperfectly Competitive Food Markets. Ph.D. Thesis. Colorado State Univ. Fort Collins, CO. 230p |
| 1998 |
Bash, P.K., and Davies, S.P 1998.Vertical integration in agribusiness foreign investment: transaction costs and contract choice. In The industrialization of agriculture: vertical coordination in the US food system, (Royer, J.S. and Rogers, R.T. eds.). Ashgate Publishing, Brookfield, VT. June Vickner, S.S., and Davies, S. P. 1998. A comment on Kinal and Lahiri's, 'On the estimation of simultaneous-equations error-components models with an application to a model of developing country foreign trade'. Journal of Applied Econometrics. 13(6):58. November/December Vickner, S.S., and Davies, S. P. 1998. A computational note on estimating simultaneous-equations error components models in `narrow panels'. Staff paper No. 378. Dept. of Agricultural Economics, University of Kentucky. September Vickner, S.S., Fulton, J.R. and Davies, S. P. 1998. Testing the Coase conjecture in the domestic frozen vegetables market. Staff paper No.379. Dept. of Agricultural Economics, University of Kentucky. September |
| 2000 |
Cutler, H., Davies, S.P. and Schmidt, M. 2000. Forecasting in a Large Macroeconomic System. Applied Economics. 32:1711-1718. Cutler, H., Davies, S.P., Eichhorn, R. and Schwarm, W. 2000. Reply to 'Testing Restrictions on Cointegrating Vectors, Journal of Macroeconomics, 22(4):707-711, Fall. Marshall, A., Davies, S.P. and Thilmany, D. 2000. Dynamic relationships between Imports and the US Lamb and Wool Markets. A paper presented at the national meetings of the American Agricultural Economics Association, Tampa, Florida, August. Vickner, S.S. and Davies, S.P. 2000. Estimating Strategic Price Response in a Product-Differentiated Oligopoly: The Case of a Domestic Canned Fruit Industry. Agribusiness: An International Journal. 16(2):125-140, Spring. Vickner, S.S., Davies, S.P., Fulton, J.R. and Vantreese, V.L. 2000. Estimating Market Power and Pricing Conduct for Private Label and National Brands in a Product-Differentiated Oligopoly: The Case of a Frozen Vegetable Market. Journal of Food Distribution Research, pp. 26-38, July. |
| 2001 |
Mizuta, Y., F. Garry, N. Dalsted, S. Davies, B. Wailes, P. Dinsmore, T. Field, and J. Green. 2001. Health Care and Medical Costs of Selected Colorado Dairy farms. Animal Sciences Research Report, pp. 19-25. Department of Animal Sciences, Colorado State University, Fort Collins CO. |
| 2002 |
Park, D. 2002. National, state and local economic influences of U.S. environmental policy and animal agriculture. Ph.D. Thesis. Department of Agricultural and Resource Economics. Colorado State University. Fort Collins, CO. Vickner, S.S., and Davies, S.P. 2002. Estimating strategic price response using cointegration analysis: the case of the domestic black and herbal tea industries. Agribusiness: An International Journal. 18(2): 131-144. |
| 2003 |
Davies, S.P., Breidt, J. and Nahuelhual, L. 2003. Identifying non-stationary components in seasonal unit roots. Selected paper of the 2003 Western Economics Association International Annual Meeting. July 14-16. Denver, Colorado. (Unpub). L. Nahuelhual, Breidt, J. and Davies, S.P. 2003. Changing seasonal patterns in domestic and import supply for us fresh grapes: an examination of seasonal unit root processes. Journal of Agricultural and Resource Economics. 28(3). (Abstract of a Selected Paper of WAEA Annual Meeting, July 14-16, Denver, Colorado.) Park, D., Seidl, A. and Davies, S.P. 2002. Environmental policy and industry location: the case of the US livestock production. Review of Regional Studies. 32 (2): 293-307. |
| 2004 |
Feuz, D.M., Umberger, W.J., Calkins, C.R. and Sitz, B. December 2004. U.S. Consumers Willingness to Pay for Flavor and Tenderness in Steaks as Determined with an Experimental Auction. Journal of Agricultural and Resource Economics. 29(3): 501-516. Jensen, G. 2004. Demand for Bell Peppers: An Application of the Barten Approach. MS Thesis. Department of Economics. Colorado State University. Johnson, K. 2004. Exogenous Shocks to the Beef industry: An Examination of Non Therapeutic Antibiotics. MS Thesis. Department of Agricultural and Resource Economics. Colorado State University. Johnson, K., Davies, S., Thilmany, D. and Umberger, W.J. August 2004. The Price Behavior of Retail Cuts of Beef, Pork and Poultry. Paper Presented in Organized Symposium on Advances in Panel Data Econometric Methods at 2004 American Agricultural Economics Association Annual Meetings. Denver, Colorado. Johnson, K., Pritchett, J. and Thilmany, D. 2004. Exogenous Shocks to the Beef Industry: An Examination of Non-Therapeutic Antibiotics. Selected paper abstract from 2004 Western Agricultural Economics Association Annual Meetings. Honolulu, Hawaii. Abstract published in the Journal of Agricultural and Resource Economics. 29(3): 593. Killinger, K.M., Calkins, C.R., Umberger, W.J., Feuz, D.M., and Eskridge, K.M. November 2004. A comparison of consumer sensory acceptance and value of domestic beefsteaks and steaks from a branded, Argentine beef program. Journal of Animal Science. 82: 3302-3307. Killinger, K.M., Calkins, C.R., Umberger, W.J., Feuz, D.M., and Eskridge, K.M. November 2004. Consumer visual preference and value for beefsteaks differing in marbling level and color. Journal of Animal Science. 82: 3288-3293. Killinger, K.M., Calkins, C.R., Umberger, W.J., Feuz, D.M., and Eskridge, K.M. November 2004. Consumer sensory acceptance and value for beefsteaks of similar tenderness, but differing in marbling level. Journal of Animal Science. 82:3294-3301. Pritchett, J. and Johnson, K. October 2004. Retail Meat Feature Pricing: Enhancing Meat Case Revenues? Food Distribution Research Society. Morro Bay, CA. Pritchett, J., Johnson, K. and Thilmany,D. August 2004. Consumer Trends and the New ERS Retail Meat Price Data Set. American Agricultural Economic Association Annual Meeting. Denver, CO. Sitz, B., Calkins, C., Umberger, W., and Feuz, D. January 2004. Consumer acceptance and value of beef from various countries of origin. Nebraska Beef Cattle Report MP-80A, p. 83-85. Sitz, B., Calkins, C., Umberger, W., and Feuz, D. January 2004. Consumer Preference and Value of Beef with Country-Of-Origin Labeling. 2004 Nebraska Beef Report. MP-80A, p. 81-82. Sitz, B., Calkins, C., Umberger, W., and Feuz, D. January 2004. Consumer acceptance and value of wet aged and dry aged beefsteaks. 2004 Nebraska Beef Report. MP-80A, p. 86-88. Umberger, W.J. and Feuz, D.M. Summer 2004. The Usefulness of Experimental Auctions in Determining Consumers Willingness to Pay for Quality Differentiated Products. Review of Agricultural Economics. 26:170-185. Umberger, W.J., Feuz, D.M., Perverse, S., Calkins, C.R. and Sitz, B. December 2004. Consumer Preference and Willingness to Pay for U.S. versus Australian and Canadian Beef. Selected paper abstract from 2004 Western Agricultural Economics Association Annual Meetings. Honolulu, Hawaii. Abstract published in the Journal of Agricultural and Resource Economics. 29(3): 589. Vickner, S. and Davies, S. August 2004. An EC3SLS Model of Corporate Governance in the US Restaurant Sector. Paper Presented in Organized Symposium on Advances in Panel Data Econometric Methods at the 2004 American Agricultural Economics Association Annual Meetings. Denver, Colorado. Ziehl, A. 2004. An Empirical Analysis of Consumer Retail Purchase Behavior for Beef Differentiated by Primal and Production Attributes. MS Thesis. Department of Agricultural and Resource Economics. Colorado State University. |
| 2005 |
Loureiro, M.L. and Umberger, W.J. 2005. Assessing Preferences for Country-of-Origin Labeled Products. Journal of Agricultural and Applied Economics. 37(1)(April): 49-63. Pritchett, J. and Johnson, K. 2005. Retail Meat Feature Pricing: Enhancing Meat Case Revenues? Journal of Food Distribution Research. 36 (1): 144-158. Pritchett, J., Thilmany, D., and Johnson, K. 2005. The Broader Economic Effects of Livestock Insurance and Health Management: Understanding Disease Outbreak Impacts on Allied Industries. In The Economics of Livestock Disease Insurance: Concepts, Issues and International Case Studies. eds. D. Hoag, D. Thilmany and S. Koontz. CABI Publishing, Cambridge, MA. ISBN: 0851990770. Pritchett, J., Thilmany, D., and Johnson, K. 2005. Animal Disease Economic Impacts: A Survey of Literature and Typology of Research Approaches. International Food and Agribusiness Management Review. 8 (1): 23-45. Schuck, E.C., Frasier, W.M., Webb, R.S., Ellingson, L.J., and Umberger, W.J. 2005. Adoption of More Technically Efficient Irrigation Systems as a Drought Response. International Journal of Water Resources Development. 21(4)(December):651-662. Sitz, B.M., Calkins, C.R., Feuz, D.M., Umberger, W.J., and Eskridge, K.M. 2005. Consumer Sensory Acceptance and Value of Domestic, Canadian, and Australian Grass-fed Beef Steaks. Journal of Animal Science. 83: 2863-2868. Umberger, W.J. 2004. Will Consumers Pay a Premium for Country-of-Origin Labeled Meat? Choices Magazine. 19(Winter 2004):15-19. Available online at: http://www.choicesmagazine .org/2004-4/cool/2004-4-04.htm Ziehl, A.R., Thilmany, D.D., and Umberger, W.J. 2005. A Cluster Analysis of Natural Beef Product Consumers by Shopping Behavior, Importance of Production Attributes and Demographics. Journal of Food Distribution Research. 36(1)(March): 209-217. |
| 2006 |
Pendell, D.L. and T.C. Schroeder. 2006. Impact of Mandatory Price Reporting on Fed Cattle Market Integration. Journal of Agricultural and Resource Economics 31(3):568-579. Pritchett, J. and D. Thilmany. 2006. The Cow That Stole Christmas? Exploring the Role of Media Coverage in Recent BSE Outbreaks. Western Economics Forum. Vol 4. No. 2. pp: 24-28. Sitz, B.M., Calkins, C.R., Feuz, D.M., Umberger, W.J., and Eskridge, K.M. 2006. Consumer Sensory Acceptance and Value of Wet Aged and Dry Aged Beef. Journal of Animal Science. 84:1221-1266. Thilmany, D.D., W.J. Umberger, and A.R. Ziehl. 2006. Strategic Market Planning for Value-added Natural Beef Products: A Cluster Analysis of Colorado Consumers. Journal of Renewable Agriculture and Food Systems. 21(3): 192-203. |