<%BANNER%>

Effectiveness of Marketing and Promotion Expenditures in the Florida Tomato Market considering the Presence of Food Safe...

Permanent Link: http://ufdc.ufl.edu/UFE0044329/00001

Material Information

Title: Effectiveness of Marketing and Promotion Expenditures in the Florida Tomato Market considering the Presence of Food Safety Incidents
Physical Description: 1 online resource (64 p.)
Language: english
Creator: Einsohn, Samuel Joseph
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: advertising -- marketing -- promotion -- tomatoes
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Much research has been performed on the topic of commodity promotion over the past 30 years but few of them have considered food safety as a factor in the creation of relevant price and demand models. Despite Florida tomato growers’ efforts to control growing practices and prevent salmonella outbreaks, the Florida Tomato Committee attributes national food safety concerns to recent decreases in the consumption of Florida tomatoes. This research study addresses this concern while simultaneously evaluating the effectiveness of Committee expenditures on education and promotion activities by utilizing regression analysis to quantify the impacts of food safety and promotion on the price of and demand for Florida tomatoes. Additionally, an interaction term for promotion and food safety was included in an augmented model to determine whether or not promotion efforts have helped to mitigate the effects of food safety incidents. A benefit/cost analysis was also performed to quantify the returns on investment for producers. Results of the analyses yielded statistically significant results for most variables. Promotional returns on investment for producers were positive and the effects of food safety were negative, however, promotion efforts have not mitigated these results.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Samuel Joseph Einsohn.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Vansickle, John J.
Local: Co-adviser: Weldon, Richard N.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044329:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044329/00001

Material Information

Title: Effectiveness of Marketing and Promotion Expenditures in the Florida Tomato Market considering the Presence of Food Safety Incidents
Physical Description: 1 online resource (64 p.)
Language: english
Creator: Einsohn, Samuel Joseph
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: advertising -- marketing -- promotion -- tomatoes
Food and Resource Economics -- Dissertations, Academic -- UF
Genre: Food and Resource Economics thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Much research has been performed on the topic of commodity promotion over the past 30 years but few of them have considered food safety as a factor in the creation of relevant price and demand models. Despite Florida tomato growers’ efforts to control growing practices and prevent salmonella outbreaks, the Florida Tomato Committee attributes national food safety concerns to recent decreases in the consumption of Florida tomatoes. This research study addresses this concern while simultaneously evaluating the effectiveness of Committee expenditures on education and promotion activities by utilizing regression analysis to quantify the impacts of food safety and promotion on the price of and demand for Florida tomatoes. Additionally, an interaction term for promotion and food safety was included in an augmented model to determine whether or not promotion efforts have helped to mitigate the effects of food safety incidents. A benefit/cost analysis was also performed to quantify the returns on investment for producers. Results of the analyses yielded statistically significant results for most variables. Promotional returns on investment for producers were positive and the effects of food safety were negative, however, promotion efforts have not mitigated these results.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Samuel Joseph Einsohn.
Thesis: Thesis (M.S.)--University of Florida, 2012.
Local: Adviser: Vansickle, John J.
Local: Co-adviser: Weldon, Richard N.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044329:00001


This item has the following downloads:


Full Text

PAGE 1

1 EFFECTIVENESS OF MAR KETING AND PROMOTION EXPENDITURES IN THE FLORIDA TOMATO MARKET CONSID ERING THE PRESENCE O F FOOD SAFETY INCIDENTS By SAMUEL JOSEPH EINSOHN A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FL ORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Samuel Joseph Einsohn

PAGE 3

3 To my darling wife Judith, my love for you grows deeper with every passing moment. If it were not for y our unwavering faith and trust I would not be the person I am. Your wisdom and support are the sources of my strength and perseverance

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my committ ee, Dr. John VanSickle and Dr. Rick Weldon Their flexibility and su pport throughout my course s and thesis work has been invaluable. Without the funding and supervision of Dr. VanSickle my graduate school experience would likely not have been possible. He is truly a brilliant economist and wonderful teacher who always made himself available to me when I needed his help I would also like to acknowledge the UF/IFAS Food and Resource Economics Department, their professors, and especially Jessica Herman for all of her guidance. Finally, I would like to thank my family w ho neve r stopped encouraging me throughout all of my pursuits.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 ABSTRACT ................................ ................................ ................................ ..................... 9 CHAP T ER 1 INTRODUCTION ................................ ................................ ................................ .... 11 A Brief History of the Market Order ................................ ................................ ......... 12 The Florida Tomato Industry ................................ ................................ ................... 13 Harvest and Processing ................................ ................................ .......................... 13 Trends in Market Share, Imports, and Exports ................................ ....................... 14 Trends in Yield and Prices ................................ ................................ ...................... 16 Promo tion Activities ................................ ................................ ................................ 17 Developments in Food Safety ................................ ................................ ................. 17 Researchable Problem ................................ ................................ ........................... 19 Relevant Hypotheses ................................ ................................ .............................. 19 2 LITERATURE REVIEW ................................ ................................ .......................... 21 General Approaches to Promotion Evaluations ................................ ...................... 21 Quantity Dependent vs. Price Dependent Models ................................ .................. 22 Recent Price Dependent Evaluations ................................ ................................ ..... 23 Re cent Quantity Dependent Evaluations ................................ ................................ 24 Introduction of the Food Safety Variable ................................ ................................ 27 Benefits and Costs of Promotion ................................ ................................ ............ 28 3 METHODOLOGY ................................ ................................ ................................ ... 29 Data Collection ................................ ................................ ................................ ....... 29 Regressions Analysis ................................ ................................ .............................. 29 New Quantity Dependent Models ................................ ................................ ..... 30 New Price Dependent Models ................................ ................................ .......... 30 Benefit Cost An alyses for Quantity Dependent Models ................................ .... 31 4 RESULTS ................................ ................................ ................................ ............... 34 Estimation Results and Quantity Dependent Model Evaluations ............................ 34 Food Safety ................................ ................................ ................................ ...... 34 Imports ................................ ................................ ................................ ............. 35

PAGE 6

6 Promotion & Income ................................ ................................ ......................... 35 Price ................................ ................................ ................................ ................. 36 Interaction Term ................................ ................................ ............................... 36 Benefits from Promotion Efforts According to the Quantity Depen dent Model ........ 37 Estimation Results and Price Dependent Model Evaluations ................................ 38 Food Safety ................................ ................................ ................................ ...... 39 Imports ................................ ................................ ................................ ............. 39 Promotion & Income ................................ ................................ ......................... 39 Quantity ................................ ................................ ................................ ............ 40 Trend ................................ ................................ ................................ ................ 40 Benefits from Promotion with Price Dependent Model ................................ ............ 40 Hypothesis Test Decision Rules ................................ ................................ ............. 42 Hypothesis Test Results ................................ ................................ ......................... 42 5 CONCLUSION AND RECOMMENDATIONS ................................ ......................... 49 Review of Study ................................ ................................ ................................ ...... 49 Recommendations for Future Studies ................................ ................................ ..... 50 APPENDIX ADDITIONAL DATA ................................ ................................ ................................ 53 LIST OF REFERE NCES ................................ ................................ ............................... 62 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 64

PAGE 7

7 LIST OF TABLES Table page 4 1 Regression output from USQ model ................................ ................................ ...... 44 4 2 Regression output from FLQ model ................................ ................................ ....... 45 4 3 Regression output from FLP model ................................ ................................ ........ 46 4 4 Regression output from USGP model ................................ ................................ .... 47 4 5 Hypothesis test decision results ................................ ................................ ............. 48 A 1 Definition and source of regression variables ................................ ........................ 53 A 2 Data set for USQ model ................................ ................................ ......................... 54 A 4 Correlations between Florid a regression variables ................................ ................ 56 A 5 Correlations between U.S. regression variables ................................ .................... 56 A 6 Augmented FLQ model with inter action terms ................................ ....................... 57 A 7 Returns to Florida producers (2005 2010) based on FLQ model .......................... 58 A 8 Returns to Florida produc ers (2005 2010) based on FLP model ........................... 58 A 9 Percent of total fresh tomato production by country ................................ ............... 59 A 10 Other comm odity benefit cost results ................................ ................................ .. 60

PAGE 8

8 LIST OF FIGURES Figure page A 1 Trends in Florida tomato consumption ................................ ................................ ... 61 A 2 Trends in Florida tomato prices ................................ ................................ ............. 61

PAGE 9

9 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Science EFFECTIVENESS OF MARKETING AND PROMOTION EXPENDITURES IN THE FLORIDA TOMATO MARKET CONSIDERING THE PRESENCE OF FOOD SAFETY INCIDENTS By Samuel Joseph Einsohn August 2012 Chair: John VanSickle Cochair. Richard Weldon Major: Fo od and Resource Economics Much research has been performed on the topic of commodity promotion over the past 30 years but few of them have considered food safety as a factor in the creation of relevant price and demand models De spite Florida tomato growe growing practices and prevent salmonella outbreaks the Florida Tomato Committee at tributes national food safety concerns to recent decreases in the consumption of Florida tomatoes This research study addresses this concern while si multaneously e valuating the effectiveness of c ommittee expenditures on education and promotio n activities by utilizing regression analysis to quantify the imp acts of food safety and promotion on the price of and demand for Florida tomatoes Additionally, a n interaction term for promotion and food safety was included in an augmented model to determine whether or not promotion efforts have helped to mitigate the effects of food safety incidents. A benefit/ cost analysis was also performed to quantify the retu rns on investment for producers. Results of the analyses yielded statistically significant results for most variables Promotional r eturns on investment for producers were positive and

PAGE 10

10 the effects of food safety were negative h owever promotion effort s ha ve not mitigated these results

PAGE 11

11 CHAPTER 1 INTRODUCTION The tomato is an American native fruit that was originally named xitomatl by the Aztecs. Spanish conquistadors discovered and brought the crop home with them after visiting and exploring the new world Once in Europe its name became tomate and it began to grow throughout the Mediterranean in sunny climates and was subsequently incorporated into a wide variety of world cuisines. In its early modern history there was much debate over its status as a fru it since it was used like a vegetable in cooking but actually a vine grown fruit. In 1898 the U.S. Supreme Court officially classified it as a vegetable. Until today it remains a commonly misclassified produce item (Florida Tomato Committee, 2010). In the ir basic form tomatoes are a simple homogeneous food commodity with little geographic economic model, when two farmers are selling the same indistinguishable tomato one will have to lower its price in order to increase the value to the consumer In order for the generic commodity prod ucers to take some form of control over the market for their product they must find a wa y to differentiate it from that of their competition By differentiating their agricultural products, grower s are often able to increase both their price and demand. Accordingly, c ertain economic trends and behaviors found in the U.S. market for fresh toma toes must be considered in order to understand the operational nature of the Florida tomato industry. For example, both the supply of and demand for Florida fresh tomatoes is inelastic especially in the non retail market as in the case of restaurants One reason for the inelasticity in supply is that most produce products are highly perishable and

PAGE 12

12 have to be sold at market baring prices at the time the crop is harvest ed In order to effectively influence price and demand, Florida growers must differen tiate their tomatoes from those of other regions by way of product improvement from research and development and/or product education and promotion activities. A Brief History of the Market Order In 1937, as a way to address the plight of the impoverished farme r during the great depression, Marketing Order s were established under an amendment to the 1933 Agricultural Adjustment Act. The With this legis lation farmers were able to create or join a federally mandated group of producers selling the ir same crop in order to stabilize supply and expand the demand for their products. To this day, Marketing O rd er programs are designed by crop growers of a speci fic geographic region who form a committee and are legally enforced by the United States Department of Agriculture (USDA) Growers in a particular Marketing Order are able to benefit from c ommittee activities through the sharing of mutual industry expenses such as the cost s of promotion and research and development. Collectively the members of the committee work to open new market s and distribution channels, regulate trade practices, practice collective bargaining activities, combine production volume to fi ll large orders, etc. The USDA requires all members to contribute annual fees to pay for the cost of the organizational activities listed above. Because the federal government req uires that the members of each O rder pay their share of dues, committee expe nses budgeted by the members must be approved by the Secretary of Agriculture Accordingly, studies must be performed by third party research institution s to determine whether or not the funds spent on marketing activities

PAGE 13

13 have yielded equitable returns fo r the members. In the last 30 years there have been several evaluations performed throughout the U.S. by various Universities and private research firms each explaining the effectiveness of promotion on agricultural commodities. The Florida Tomato Indust ry The statistic and economic information on the development of the Florida tomato industry, from its historical beginnings to present, provides a fundamental background necessary to carry out relevant econometric analysis and modeling. The use of this in formation provides the information essential to understanding the econometric findings and allows for a clear interpretation of the evaluation results. Harvest and Processing Florida, due to its warm and sunny climate is an ideal location f or growing fresh field tomatoes and as such began production e fforts in 1870. C urrently Florida is the largest producer of fresh tomatoes in the U.S. supplying 33 % of all domestically produced fresh tomatoes Harvesting is done primarily from October through June and pe aks during December and January. On average, tomato production reaches almost one third of the annual value of all fresh vegetables produced in Florida. There are approximately 31,500 acres of land currently cultivating Florida fresh tomatoes at an average production cost of nearly $12,000 per acre. At the farm level nearly 33,00 0 laborers are needed to hand pick the crops and producers ship more than 1.1 billio n pounds of their produce throughout the U.S. Canada, and overseas to Japan and other Asian cou ntries for a total cr op value exceeding $619 million (Florida Tomato Committee, 2010).

PAGE 14

14 patterns play a significant role in determining c rop yield through the season as heavy levels of rainfall and cold winters occur during much of the planting and growth months. For example, in August 2004 rainfall delayed growers ability to plant their crops and above average heat in November caused a productivity slowdown during picking. Heavy rain also fell during December of that year leadi ng to both a decrease in quantity harvest ed as well as a quality reduction in some of the crops (Florida Tomato Committee, 2004). Trends in Market Share Imports, and Exports Domestic consumers have shown a preference for tomatoes as a vegetable which is second in popularity only to potatoes a nd lettuce in terms of per capita quantities consumed Throughout the U.S., demand has increased 50% in the last 40 years. Average per capita consumption of fresh tomatoes reached almost 21 pounds per person in 2010 a nd showed a n average growth rate of .25 lbs. per year since 1972. The consumer preference for Florida tomatoes has historically been a matter of physical taste and nutritional value. Rose Research Incorporated conducted a blind taste test in 2009 with a p anel of 1200 taste testers to determine whether U.S. consumers preferred the taste of Florida fresh tomatoes to the taste of Mexican fresh tomatoes. 66% of the panel preferred the taste of Florida t omatoes and of that group 60% said that flavor was the mai n reason for their choice. This finding serves as the basis for recent Florida Tomato Committee efforts to further improve the taste and texture of Florida tomato es through scientific research in cooperation with the University of Florida Institute of Fo od and Animal Sciences (Florida Tomato Committee, 2009). U.S. imports of fresh tomatoes come primarily from Canada and Mexico but smaller portions of aggregate consumption also come from Guatemala, Dominican

PAGE 15

15 Republic, Netherlands, Costa Rica, France, Ital y, Belgium, Columbia, New Zealand, and Israel. An analysis of the 2009/ 2010 data on monthly imports from October to May shows that Mexico provided approximately 71% of total fresh imported tomatoes sold in the U.S while Florida accounted for 25% and non M exican imports accounted for the remaining 4%. Overall, throughout the 2009/ 2010 October to May season, Florida held 33% of the entire U.S tomato market while Mexico provided 54% between their field and greenhouse production while remaining imports account ed for the final 13%. 33% share show s a drastic decline from previous years at 51% and 52% in 2007/2008 and 200 8/2009 respectively. At the same time 2010 total market share of 54% was an increase from 2008 and 2009 levels at 40% and 39% respectively. The Florida Tomato Committee attributes much of this decline in market share to poor weather conditions caused by heavy rainfall which increased the presence of disease and pest regions (Florida Tomato Committee 2010). Market Access Promotion programs (MAP) w ere implemented in 1998 by the Florida Tomato Committee to increase exports to Canada and Asia with a large portion of program fun ding coming from the USDA. On average, the c ommittee has spent $187,015 per year on MAP expenditures since its adoption of the program Considering recent efforts, the c ommittee considers 2009/ 2010 MAP efforts to have been less than effective as a result of latile weathe r patterns Exports to Canada during that season fell drastically due to the limited availability of Florida tomatoes, variability in quality, and higher free on b oard (FOB) prices In the same season, Canada, which is known for its greenhouse production was able to use climate

PAGE 16

16 controls to maintain consistent crop quality maximize yield, and fill the supply gap caused by the decrease in Florida imports Canadian shoppers were able to benefit from lower prices on Mexican and Canadian products as a result o f their respective levels of greenhouse production while Florida tomatoes were sold at nearly $2 more per pound Greenhouse production heavily reduces much of risk inherent to field production and provides a higher level of certainty to crop projections th roughout the season. Trends in Yield and Prices The price of Florida tomatoes has shown a high level of variability over the past 40 years. On average price s have declined by approximately 2.8% per year since 1972 From 2005 to 2010 price s increased and decreased with each season starting with a per pound price of $.50 in 2005 and falling to $.29 in 2007 It then ro se again to $.50 in 2008 and then drop ped back down to $.29 in 2009. Continuin g this pattern, the price rose to its historical peak of $.52 in 2010 The complete price trend from 1972 2010 can be found in F igure A 1 Consumption patterns for Florida tomatoes from 1972 2010 a declining trend in per capita consumption of Florida tomatoes, falling at an average rate of 1.8% per year. In 1972 per capita consumption began to increase from 3 pounds per person to over 6.5 pounds in 1989. It hit its historical peak in 1992 at 7 pounds per person and then fell almost annually until 2010 where it had reached its lowest historical point of 2.25 pounds Th e complete per capita consumpt ion trend can be found on F igure A 2 Per capita consumption is calculated as the quantity of Florida field tomatoes produced divided by the tot al U.S. population in that year. From this, the variation in consumption due to F lorida weather conditions can be well understood. T he perishable nature of fresh produce should also be kept in mind t o explain the variation s in price,

PAGE 17

17 i.e., that most fresh fruits and vegetable s must be sold soon after harvest G rowers are therefore sub ject to month ly market prices and are considered to be Promotion efforts are undertaken to improve these conditions by bol st ering demand and price for Florida tomatoes. Promotion Activities The Florida tomato industry has been using promot ion as means of increasing consumption of their products for several years with programs focusing on categorical market participation. For example, on the retail side of the industry efforts have been focused on building the retail community. In the area of public relations, the industry has published press releases in newspapers and magazines in order to educate consumers on the proper handling and serving of tomatoes. Efforts were also made in the form of trade advertising for those who purchase their tomatoes directly through Florida shipper s. During the 2006 season, the c ommittee launched its first television campaign in an attempt to brand Florida tomatoes uding the Food Network, HGTV, DIY, Discovery Home, and Discovery Health. E ducational information on h andling and serving was also included in the commercials. Also, as mentioned previously, MAP promotion activities were conducted to increase exports to Can ada and Asia with a large portion of program funding coming from the USDA Developments in Food Safety Historically, food safety has been under the auspice of the USDA. Currently however, the responsibilities for supervision and regulatory enforcement are divided among agencies with two groups from the Department of Human Health and Services, three groups from the Department of Agriculture, and one group from the Environmental

PAGE 18

18 Protection Agency Each group performs some aspect of food safety regulation and monitoring Currently there are no federal mandates in place for the implementation of quality control practice s such as the Agricu ltural Practices (GAP), h owever the Florida Tomato Commit tee has been practicing GAP) for the past several years (Ferro 2010). Since t he adoption of these practices, the Florida tomato industry has been able to start distancing itself from the ac cusations that their tomatoes caused reported Salmonella outbreaks throughout the country recent lack of involvement in food safety outbreaks prices for Florida tomatoes have still been negatively affected simply due to their association with the market as a whole There have been approximately 44 outbreaks of salmonella in the United States since 1972 that have been directly tied to the consumption of tomatoes, causing at least 5,148 illnesses across the country Given these recent food safety concerns, it is expected that these outbreaks have had an effect on the consumption of Florida tomatoes. Previous evaluations of the Florida tomat o market have yet to consider food safety as a determinant of consumpt ion. This study however does include this factor by adding a variable which accounts for the presence of salmonella outbreaks associated with tomato consumption in the estimation of a consumer demand model. The results from the model will help define food approximate effects on the demand for Florida tomatoes while also allowing for an evaluation of whether education and promotion activities have acted as a significant hedge against the anticipated negative effects of food safety incidents.

PAGE 19

19 Researc hable Problem Education and promotion expenditures have the potential to increase the price of and deman d for fresh t omatoes however, r raised concerns that the these expenditures may not be as e ffective conside ring issues of food s afety which is thought to have had a negative impact on price and demand over the past several years. Because Florida Tomato Committee members are obligated to pay for the Marketing Order education and promotion expenses an analysi s must be performed which measures the effectiveness of promotion efforts at raising both price and demand for Florida fresh tomatoes. The results of this analysis can then allow for a benefit/cost evaluation to be performed that will express returns to gr owers for their investments in c ommittee promotion programs Relevant Hypotheses Hypotheses were developed to address several quest ions necessary to understand the role of commodity promotion under a federally mandated marketing order. These questions are as follows: What factors influence the U.S. price and demand for fresh tomatoes and more specifically for Florida fresh field grown tomatoes? How do these factors in fluence those prices and demand ? To what extent do education and promotion activities help to bolster prices and demand? What impacts do food safety outbreaks have on the U.S. and Florida tomato markets? To what extent do education and promotion activities help to mitigate the perceived food safety impact? Hypothesis 1: Florida T omato Committe e education and promotion e xpenditures have a p ositive effect on a. the quantity of U.S tomatoes demanded domestically b. the domestic price of U.S. tomatoes c. the quantity of Florida tomatoes demanded domestically

PAGE 20

20 d. the domestic price of Florida toma toes Hypothesis 2: Food safety incidents negatively impact a. the quantity of U.S tomatoes demanded domestically b. the domestic price of U.S. tomatoes c. the quantity of Florida tomatoes demanded domestically d. the domestic price of Florida tomatoe s Hypothesis 3: Florida Tomato Committee educa tion and promotion expenditures effectively reduce the impact of food safety incidents on a. the U.S. demand for fresh tomatoes b. the U.S. demand for Florida fresh tomatoes

PAGE 21

21 CHAPTER 2 LITERATURE REVIEW Ge neral Approaches to Promotion Evaluations Market evaluations use econometric modeling to determine if generic advertising campaigns have had a statistically significant effect on demand. Econometrics can be used to isolate the variables effecting demand fo r a commodity in order to see the approximate portion of variation in demand that can be attributed to each of the independent variables included in the model When performing an empirical evaluation of a marketing program it is necessary to develop an eco nometric model to determine the approximate effect of promotion programs on demand. The model must first be stated in a functional form and then transformed by sample data collected over a well defined period of evaluation. The data can then be analyzed us ing the appropriate statistical tools (Forker and Ward, 1991). Several studies have been performed in the last 30 years for the purpose of evaluating the effectiveness of promotion of agricultural commodities. Some of the most prominent research was perf ormed by Degner (1985), Forker and Ward (1991,1993), and Sexton and Crespi (1999, 2003). Each of these studies came to the same conclusion ; that commodity promotion programs do in fact yield a significantly positive benefit to producers. Simply stated, spe nding money on promotion activities increases the demand for commodities. At the same time, in order for promotion programs to be worthwhile, their returns must be greater than their costs (Crespi and Sexton, p.2). A n effective promotion program offers pos itive net returns to the growers.

PAGE 22

22 Quantity Dependent vs Price Dependent Models Certain considerations must be made in order to fully understand the benefits to promoting commodity generic produce both in terms of the market effect as well as the returns t o the producers of the crop. Inherent problems exist with quantity dependent demand models for fresh produce as those crops are highly perishable Accordingly, f resh tomatoes must be sold close to harvest and at the market price market for that time, i.e., supply is highly inelastic In light of this consideration, it is not always appropriate to measure promotion effects using the results of a quantity based demand model. Price dependent models are therefore sometimes used to measure the price response to promotion activities Previous studies by Sexton and Crespi (1991), VanSickle and Ev a ns (2001), and VanSickle and Ranjan (2006) do not consider the price dependent nature of the tomato market but rather focus on the ability of promotion to drive demand on an annual basis. These analyses therefore demonstrate a more long run approach to the bene fits of commodity promotion. An analysis based on results from a price dependent model however offers a more short run approach as promotion efforts are not targeted at increasing consumption of all tomatoes but rather specifically Florida fresh tomatoes which would logically lead to a faster response in price The theory behind the ability of promotion to effect price is rooted in the concept of product differentiat ion and the subsequent price premium gained from it For example, the market price of a zucchini is determined by the utility it holds for the consumer. Quoting Sherwin Rosen (1980) If two brands [or two different producers of zucchinis] offer the same bu commodity value], but sell for different prices, consumers only consider the less expensive good and the identity of sellers is irreleva nt to their purchase decisions.

PAGE 23

23 Consequently, prior to pro duct differentiation con sumers determine the value of a product according their level of perceived utility. The consumer is less concerned with where their produce comes from and more concerned with the value that product offers. Promotion e fforts directed at product differentiation effectively create heterogeneity and this will lead to consumers discriminating in their purchases and result s in higher prices for the products they perceive as superior. As mentioned previously, the Florida Toma to Committee has dedicated much of their research and development on producing a higher quality product and much of their promotion efforts to help ing consumers recognize the increased value that only com es with Florida fresh tomatoes, hoping higher prices result for their product Recent Price Dependent Evaluations A n evaluation of the California a vocado industry (Carman and Craft, 1998) shows that promotion efforts by the California Avocado Commission have a statistically significant positive impact on p rice. The evaluation states that demand was estimated using the price dependent model because avocado production is essentially predetermined and avocados are a highly perishable good that is sold at a market determined price. A model was develo ped to acco unt for the effect of advertising considering California, Florida, and other import quantities as well as consumer income. The function form of the model is expressed as (2 1) where is the annual price o f California avocados, is per capita sales of California Avocados, is the per capita sales of Florida avocados produced, is per capita imports, is per capita income, and is advertising expenditures. Price was adjusted

PAGE 24

24 for in flation using the Consumer Price I ndex. An econometric regression was run in order to show elasticities using a time series method for production years 1962 1994 Overall the results were almost entirely consistent with a priori expectations and showed sp ecifically positive results for the promotion variable; indicating that advertising has a positive effect on the price of California avocados. Both long and short run benefit/cost evaluations were also performed to show direct impacts on average producer revenue from the promotion investments. Sh ort run returns were show n track ing year to year increases in revenue from a price perspective only and purposefully did not account for annual increases in consumption caused by promotion. The short run returns i n revenue ranged from $.76 to $14.37 per dollar invested on advertising and promotion from 1962 1994 while the long run returns which include the effects of increased consumption, ranged from $.74 to $3 during the same period Recent Quantity Dependent E valuations The two most recent evaluations on Florida tomato promotion programs were conducted by VanSickle and Evans (2001) and VanSickle and Ranjan (2006). These evaluations were conducted with models used as the foundation of this study; however, these models have been redefined and further deve loped for this evaluation. The adjustments made while necessary to include in this study, make direct comparisons to previous studies difficult. Crespi and Sexton (2003) introduced the following model to evaluat e the impact of advertising and promotion on the demand for tomatoes Q=f(rP, rINC, rPROMO ,TREND) (2 2)

PAGE 25

25 where Q is annual per capita consumption in the United States, rP is the inflation adjusted price of the commodity rINC is the inflation adjusted annua l per capita U.S. income, rPROMO is the inflation adjusted value of education and promotion expenditures, and TREND is a linear variable which accounts for the annual stochastic variability in the market that is not captured by the other stated variables. VanSickle and Evans (2001) altered the model to provide a more realistic depiction of the U.S. tomato market. Prior to 2011, country of origin labeling requirements were not mandated for fresh produce and consumers were typically unable to discern Florida tomatoes from those of another source. As a result, it was would influence demand for fresh market tomatoes from Florida as well as other sources. Two new models for evaluating promotion in the Florida tomato industry resulted from that assumption. The first is as follows: USQ=f (rUSGP, rINC, rPROMO, TREND) (2 3) where USQ is the annual U.S. per capita consumption of all fresh market tomatoes, rUSGP is the inflation adjusted U.S. average annual price for fresh tomatoes, rINC is the sam e variable expressed in Model 2 2 rPromo is the inflation adjusted annual dollars spent to promote Florida tomatoes, and Trend is the s ame variable defined in Model 2 2 The second model is specifi ed as FLQ=f (rFLP ,rINC, rPROMO, TREND, USIMP) (2 4) where FLQ is the annual per capita consumption of Florida fresh tomatoes (measured as total annual production in the marketing order area divided by the respective annual

PAGE 26

26 population), rFLP is the inflat ion adjusted per pound price of Florida fresh tomatoes, rINC is the sam e variable expressed in Model 2 2 rPromo is the inflation adjusted annual dollars spe nt to promote Florida tomatoes, TREND is the same variable defined in M o del 2 2 and USIMP is the p er capita consumption of imported fresh market tomatoes in the U.S. The USIMP variable was added to the model in order to show the impact of imports on demand for Florida tomatoes It was not included in Model 2 3 because the USQ variable already contain s imports as part of total U.S. fresh tomato consumption without regard for its origin These two models showed the impact of promotion on total U.S. tomato demand and on Florida tomato demand. The results provided estimates of the benefits from promotion on the consumption of Florida fresh tomatoes as well as the spillover effects received by producers outside of the marketing order area VanSickle and Ranjan (2006) later adjusted the promotion variable by differentiating between U.S and Florida promotion expenditures. The rPROMO variable in Model 2 3 was replaced with rUSPROMO to account only for expenditures on education and promotion which were used to impact the U.S. tomato industry MAP expenditures were added promotion in the Florida model to create r FLPROMO because Florida exports to foreign markets were thought to be included in the Florida consumption variable. Consequently a change in MAP expenditures would affect consumption of Florida tomatoes. FLPROMO has now been proven to be an invalid variabl e as Florida exports are so small relative to total production that the MAP effects are insignificant relative to the regular promotion variable. Therefore, this evaluation has

PAGE 27

27 returned to the pre 2005 promotion variable rPROMO and the values are expressed equally in FLQ FLP, USQ, and USGP models. I t was also concluded that the trend variable found in both Model s 2 3 and 2 4 could not be included in this analysis as trend and income are highly correlated. A correlation of rINC and Trend revealed a 0.989 co rrelation coefficient between the two variables, making it difficult to separate the effects of both variables in the same model. As a result, the income variable already captures the trend effect. Therefore, TREND is no longer being considered as a separa te variable in this analysis, but rather being considered as part of the rINC variable Table A 4 and Table A 5 show correlation s for all variables included in this evaluation for years 1972 2010. I ntroduction of the Food Safety Variable As mentioned pre viously food safety has become of increasingly more concern to Florida growers and as a result a new food safety variable has been included in the estimation of both U.S. and Florida fresh tomato consumption. Previous works by Ferro (2010) demonstrate tha t the presence of food safety incidents in the Florida produce market c ause statistically significant negative e ffects on both price and demand. By 2008 when the salmonella Saint Paul incident had occurred, Florida tomato producers had already adopted and implemented the Tomato Best Management Practic es (T BMT) and the Tomato Good A gricultural Practices (T GAP). The adoption of these practices has allowed the Florida tomato industry to eventually distance itself from the accusations that their tomatoes had caused the outbreak. Nevertheless, prices for Florida tomatoes fell during this period. In 2007 it is estimated that the tomato industry lost more than $18 million due to the presence of food safety incidences (Ferro, 2010).

PAGE 28

28 Benefits and Costs of Promotio n effectiveness at generating returns to producers. The average benefit/cost ratio (ABC) can be used to calculate producer returns resulting from the dollars spent on promotion This is calculated by first estimating the effects of various inputs, such as promotion, on the quantity consumed. The marginal benefit/cost ratio can be used to determine returns on promotion at the margin and can be used further to calculate profits to producers resulting from additional dollars spent on promotion. Previous studies for the Florida Tomato Committee show an ABC value of $ 27.15 (VanSickle and Evans, 2001). This means that each dollar spent on tomato promotion returned $27.15 in additional revenue to the growers.

PAGE 29

29 CHAPTER 3 METHODOLOGY Purpose The purpose of this study is to develop multivariate regression model s which will reflect the effects of education and promotion activities on the U.S. and Florida tomato markets. D ata Collection The data for the analysis were taken from various sources which can be accessed easily by the public. Income, population, and Gross Domestic Product (GDP) data were room 1 while U.S. tomato prices, per capita U.S. consumption, and U.S. imports were collected from the ERS Vegetable and Melon Yearbook (USDA, 2011). Florida tomato prices, Florida production education and promotion expenditures, and Market Access Promotion expenditu reports. All price data is adjusted for inflation by converting all values into 2005 dollars using the ERS reported GDP deflator. Florida consumption was calculated as total Florida product ion divided by U.S. population. All data used in the analysis are shown in their nominal form in Table A 1 and Table A 2. Regressions Analysis A preliminary regression was performed to reproduce the VanSickle and Ranjan (2006) quantity dependent models Th ese demand models were designed to measure MAP expenditures. As mentioned previously, they act as the foundation for this 1 This data is available online at http://www.ers.usda.gov/Data/Macroeconomics/ Accessed 2/1/2012

PAGE 30

30 evaluation with certain changes made to the variable specifications. All data used by VanSickle and Ranjan (2006) were updated and brought current using m any of the same sources of information for each of the variables presented in their mode ls from 1972 to 2010 New Quantity Dependent Models The fo llowing equations reflect the adjustments made to the VanSickl e and Ranjan models USQ=f (rUSGP, rINC, rPromo, ILL) (3 1) FLQ=f (rFLP, rINC, rPromo, USIMP, ILL) (3 2 ) w here ILL is a dummy variable with a value of 1 for years were CDC reported illnesses that were caused by tomato related salmonella outbreaks in the U.S and 0 where no such illnesses were reported. The inclusion of this variable allows for the measurement all fresh tomatoes consumed in th e U.S as specified in Model 3 1 and the demand for Florida fresh tomatoes as specified in Model 3 2 The illnesses variable was also considered in the form of an intensity variable stated as the annual number of illnesses reported, however the results show ed a lower level of statistical significance when compared to the dummy variable. This would indicate that the public is more concerned that there was a tomato related salmonella outbreak than the number of people who became ill from its relative consumpti on. New Price Dependent Models Using a variation of Model 3 1 and Model 3 2 a new model was developed to determine the effects of education and promotion on price. The consumption variables for both the U.S. and Florida models were switched with their res pective price

PAGE 31

31 counterparts and a trend 2 variable was added to capture the remaining effects not attributed to the rest of the specified variables, including the remaining effects not accounted for in rINC as previously specified. The new models are specifi ed as USGP=f (USQ, rINC rPROMO ILL, TREND) (3 3 ) FLP=f (FLQ, rIN C, rPROMO, USIMP, ILL, TREND) (3 4 ) Benefit Cost Analyses for Quantity Dependent Models As mentioned previously, benefit/cost ratios can be used to determine the effective nes s of education a nd promotion programs. This type of evaluation can be performed for both quantity and price dependent models. For the quantity dependent form, the average ratio (ABC) is calculated using the results of the OLS regression in the previous section. To accompl ish this, the rPromo coefficient is first used to calculate the precise increase in demand that result s from promotion expenditures and is calculated as (3 5) This value is calcula ted for each year that promotion expenditures are present and is used to further calculate marginal benefits as (3 6) Marginal b enefit is equal to the change in per capita quantity consumed as a res price minus the per pound cost of production. The average gross benefit (AGB) from promotion is calculated as 2 TREND for Model 3 5 and Model 3 6 represents the same variable expressed in Model 2 2.

PAGE 32

32 (3 7) This value is calculated for each year in which promotion took place and is used to further calculate the average benefit/cost ratio (ABC). (3 8) Benefit Cost Analyses for Price Dependent Models As in the quantity dependent model, Benefit/Cost analysis can be used to determine the effectiveness of investments in promotion and education programs. The average benefit ratio (ABC) is calculated using the results of the OLS regression from the price dependent model. Unlike previously, the rPROMO coefficient is first use d to calculate the change in price resulting from a change in promotion expenditures and is calculated as (3 9) This value is calculated for each year that promotion expenditures are p resent and is used to further calculate marginal benefit as (3 10) Marginal benefit is equal to the change in price as a result of promotion multiplied culated as per capita known as the gross benefit realized from advertising and can be used to further calculate the marginal gross margin benefit/cost ratio (MGMBC) which gi ves the promotion activi ties. The average gross benefit (AGB) from promotion was calculated as

PAGE 33

33 (3 11) This value was calculated for each year in which promotion took place and was used to further calculate the average benefit/cost ratio (ABC). (3 12)

PAGE 34

34 CHAPTER 4 RESULTS Estimation Results and Qua ntity Dependent Model Evaluations Models 3 1 and 3 2 were estimated using the double log form with the ordinary least squ ares (OLS) method of regression By using the double log form of the model, the results show constant elasticities for each variable, i .e., the percentage change in demand caused by a one percent change in each of the explanatory variables. The results of the regression are shown in Table 4 1 and Table 4 2 Standard linear forms of each model were also estimated but the results were found to be in significant The results of the double log models are mostly consistent with a priori expectations and describe most of the variation in the annual consumption of both Florida and U.S. fresh tomatoes. The USQ model yielded an adjusted of .97 while the FLQ model yielded an adjusted of .76 Most parameters in both models yield expected signs and most are statistically significant at the 95% confidence level. Food Safety Parameter estimates for food safety (ILL) were negative in both mod els, indicating that higher number s of illnesses per year result in lower consumption s of U.S. and Florida tomatoes. The coefficient in the USQ model indicates that the food safety elas ticity is .0045 i.e. highly inelastic where a change in the number o f illnesses per year will result in relatively minor changes in the quantity of tomatoes consumed. In aggregate however this does translate to a substantial financial loss. The FLQ model coeff icient for food safety is 0.104 indicating a slightly higher albeit still inelastic, degree of demand responsiveness to changes in the number of illnesses per year in the case of Florida tomatoes. Anecdotal evidence also supports this conclusion. Overall, the

PAGE 35

35 number of reported salmonella related illnesses increas ed from 2006 to 2010. T he consumption of U.S. tomatoes increase d over this same period by only 1.2% while consumption of Florida tomatoes decline d by 19.1 % These results suggest that Florida suffered a significant portion of the U.S. decline in fresh mark et tomato consumption due to food safety issues. This may be due to the fact that Florida is the primary supplier of fresh market tomatoes to the U.S. during the winter months a time when their major competitor is Mexico a country notoriously assumed to be responsible for many salmonella outbreaks Imports The parameter estimate for U.S. imports (USIMP) in the Florida model also yielded negative results with a coefficient of .66, indicating a moderate to high level of elasticity with regard to demand fo r Florida tomatoes, i.e., that a slight change in the number of U.S. imports will result in a relatively large change in the quantity of Florida tomatoes consumed. The remain in g 3 4% may be inconsequential considering the relative differences between Flori da quantities sold and quantities imported. For example, if Mexico increased imports to the U.S. by 1%, Florida a relatively larger producer may only see a .66% decrease in its quantity produced. Promotion & Income Parameter estimates for promotion and income ( rPROMO and rINC ) were positive and significant in both models. The coefficients for promotion were 0.02 in the FLQ model and .005 in the USQ model, indicating inelasticity in both markets, however more in elastic in the U.S. model. The coefficients for income were exceptional ly similar to one another at .75 in the FLQ model and .78 in the USQ model. The variable in the USQ showed an extremely high level of statistical significance with a value of 15.46 and the

PAGE 36

36 FLQ variable showed a much lower, albeit still significant, level of significance with a value of 2.43 With these results it can be seen that income is main indicator of determining consumption in both models. Effectively, a small change in consumer income result s in a relatively large change i n tomato consumption in both the U.S. and Florida markets. Price The price parameters in both models were inelastic and negative at .005 for rUSGP and .40 for rFLP While the coefficient for rFLP was significant at 2.85, the rUSGP was statistical ly in si gnificance at .113 Preliminary models using ILL as an intensity variable rather than a dummy variable had actually caused the results for the r USGP variable to be positive. This would be contrary to a priori expectations which state that an increase in t omato price should cause a reduction in the quantity of tomatoes consumed. accurately negative sign demonstrates an example of the superiority of the dummy expression of the ILL variable in terms of capturing more accurate market behaviors Intera ction Term In order to determine the effectiveness of c ommittee promotion and education at mitigating the negative influences of food safety on consumption, a new interaction variable (ILLx r P ROMO ) was introduced in the FLQ model The interaction terms allo wed for the examination of the promotion variables as a function of presence in the market. It was expected that there would be a statistically significant positive effect on both models. In Model 3 2 the effect however proved to be negat ive with a coefficient of .124 and was s tatistica l l y significant The f act that the result is significant for a negative parameter strengthen s the notion that promotion has not

PAGE 37

37 worked effectively to lessen the effects of food safety concerns In actuality it seems to reveal that efforts education and promotion during outbreaks causes a reaction of consumer backlash. More is discussed on this matter in the next section. Benefits from Promotion Efforts According to the Quantity Dependent Model In this analysis, cost was assumed t o be $.30/LB for each year. The marginal benefit result, also known as the gross benefit realized from advertising is used to further calculate a benefit/cost ratio known as the marginal gross margin benefit/cost ratio (MGMBC). This val ue gives the additional profit to producers gained after recouping the expense of that year A 7 shows the marginal and average benefits to producers for years 2005 benefit/cost analysis for yea rs 2005 2010 shows a range of $1.93 $11.90, averaging ross margin ranged from $ .30 to $2.20, averaging $1.01. On average, the past five years show that every dollar spent on promotion of Florida t omatoes as has yielded just above a 100% return after breaking even, i.e., $1.00 spent on promotion created an average of $2.01 in revenue to the growers. As this is an average value over a 6 year period, it can be misleading to assume such a return for ev ery year. Each value should be considered in light of how it was calculated. In 2009 for example, $1.00 in promotion yielded $0.97 in a dditional revenue, resulting in a loss on in vestment in promotion of $.03. This loss occurred when promotion expenditure s increased from 2008 levels by nearly $13,000 in the following year. T he model suggests there was a statistically significant increase in quantity consumed as a result of promotion. Howeve r, referring back to Model 3 6 it can be understood that this actu ally

PAGE 38

38 price of $.29. More tomatoes were sold but at a $.01 per pound loss The same issue surfaces in the interaction of promotion and food safety. Considering the effect s of food safety in years 2008 and 2009, there was as drastic decrease in the number of food borne illnesses. Such a decrease in illnesses coupled with a $13,000 increase in promotion expenditures should logically yield higher returns to producers. H owever 2009 showed a return of only $.97 on each dollar of promotion expended, while 2008 showed a return of $1.49 despite the high number of reported illnesses It can be concluded from this and the insignificant interaction variable (ILL x rPROMO) that promot ion has not effectively mitigated the negative effect of food safety issues. One possible reason for this is that the Florida Tomato Committee has not specifically targeted their education and promotion programs at informing the public of their safety prac tices, such as their use of Best Tomato Management Practices. These results suggest that doing so in the future may allow the committee to further differentiate their products as safer than the average tomatoes sold in the U.S. Estimation Results and Price Dependent Model Evaluations As before, the double log form was used to show constant elasticities for each variable The results of the regressions are mostly consistent with a priori expectations and describe the most of the variation in the annual mark et price of both F lorida and U.S. fresh tomatoes. T he USGP Model 3 5 yielded an adjusted of .65 while the FLP Model 3 6 yielded an adjusted of .60 Not all parameters in both models yielded expected signs and few of them show statistical sign ificance a t the 95% level of confidence.

PAGE 39

39 Food Safety Parameter estimates for food safety (ILL) were negative in price models indicating that an increase in the numbers of illnesses per year results in a decrease in the average annual price of all U.S. t omatoes sold. This follows a priori expectations that the annual presence of food safety concerns over tomato consumption will cause prices to fall The coef ficient for ILL in the U.S. Model 3 5 was .0065 but had a very low level o f statistical significan ce at .1 In the Florida Model 3 6 food safety yielded a coefficient of 109 and was close to statistical s ignificance at a 95% confidence level with a value of 1. 59 Imports The estimate for imports (USIMP) in the Florida model yielded negative re su lts with a coefficient of .3 3 indicating a moderate level of elasticity with regard to prices, i.e. a 1% increase in the number of tomato es imported to the U.S. from overseas competitors will result in a .33 % decline in Florida tomato prices. This value was statistically significant at 2.05 Promotion & Income Parameter estimate s for promotion showed opposite effects in the U.S. and Florida. In the U.S. model the coefficient for promotion was .0 087 with a statistical significance of 1.36 while the coefficient for the Florida model was 003 with almost no significance at .413. The Florida model yielded a negative coefficient for income at .002 with a .197 level of significance This is contrary to a priori expectations that an increase in per capita income would yield a higher price. In the U.S. model, the coefficient for income was positive at .34 but with a .56 level of significance.

PAGE 40

40 Quantity The quantity parameters were moderately elastic in both models and the coefficients were also both negati ve. For USQ the coefficient was .45 with s tatistical significance at .72 whi le FLQ had a coefficient of .53 and a much higher level of si gnificance at 2.75 T hese results follow a classic supply and demand dynamic wherein an increase in the quantity o f tomatoes suppli ed results in a decrease in tomato prices. This relationship exists in both models however it is important to notice that the relationship is stronger in the Florida model where the coefficient is higher and there is a higher level of stat istical significance. Trend The trend variable was added back into the price model in order to account for the little remaining unspecified variations in price that were not captured by the other included variable s, including income In the U.S. model th e parameter was negative with a coefficient of .12 and was statistically sign ificant at 2.03 indicating that there are significant inelastic factors outside the model which cause a decrease in the average price of U.S. tomatoes. The coefficient for the Florida model also showed inelasticity however it was positive at .036 and was stati stically insignificant at 44 Benefits from Promotion with Price Dependent Model Th e ABC is actua l l y equal to MGMBC in this portion of the analysis as they both measure the annual return on investment from $1 in promotion caused by the resulting increase in price. This value is stated in dollars generated by the producer after the cost s 2005 2010 shows a range of $4.02 $ 8.33 averaging $ 5.87 This is a 587 % return after

PAGE 41

41 breaking even, i.e., $1.00 spent on pro motion created an average of $5.87 in profits to the growers As this is an average value over a 6 year period, it can be misle ading to assume such a return for every year. Like the results of the quantity dependent benefit/cost analysis, these values should also be considered in light of how they were calculated. It should be kept in mind that the presented values are measurement s of change from the In 2009 there was a $ 175 607 increase in promotional expenditures from the previous year which should have yielded a higher market price for Florida tomatoes and therefore a larger return on investment to t he growers relative to the previous year. While the analysis does show a 2009 return of $4.02 Florida tomato prices actually fell that year from $.50 per pound in 2008 to only $.30 per pound in 2009 Conversely, there was a $ 676 717 decrease in promotion expenditures in 2010 which predictably would lead to a decrease in price and a low return on investment, however, i n reality price increased from $.30 per pound to $.52. One possible reason for these results is that during 2009 and 2010 other factors effe cting price, especially quantity produced, which according to the model has a large and highly significant negative effect, substantially influenced the changes in price. In 2009 there was a decrease in price but an increase in quantity produced while in 2 010 there was an increase in price but a decrease in quantity produced. In essence the effect of promotion on price was overshadowed by the effects of demand and therefore careful consideration should be made was planning future investments in promotion b ased on the price dependent model. Average profits to producers from 2005 2010 can be found in Table A 8.

PAGE 42

42 Hypothesis Test Decision Rules A t test of the resulting coefficients from the previous chapter allow for hypothesis testing at a 95% level of confid ence. With 39 observations, the t value is calculated at 1.68 for the use of determining the results of the hypothesis tests. Again, the H ypotheses are stated as follows Hypothesis 1: Florida Tomato Committee education an d promotion expenditures have a pos itive effect on a. the quantity of U.S tomatoes demanded domestically b. the domestic price of U.S. tomatoes c. the quantity of Florida tomatoes demanded domestically d. the dom estic price of Florida tomatoes. Hypothesis 2: Food safe ty incidents negatively impacts. a. the quantity of U.S tomatoes demanded domestically b. the domestic price of U.S. tomatoes c. the quantity of Florida tomatoes demanded domestically d. the dom estic price of Florida tomatoes. Hypothesis 3: Florida Tomato Committee educa tion and promotion expenditures effectively reduce the impact of food safety incidents on a. the U.S. demand for fresh tomatoes b. the U.S. demand for Florida fresh tomato Hypothesis Test Results For H ypothesis 1a we fail to reject that Florida Tomato Committee education and promotion expendi tures have a positive effect on the quantity of U.S tomatoes demanded domesti cally but must reject the H ypothesis 1b that they have a positive effect on the their domestic price. Alternatively, we fail to rejects H ypothesis 1c that Florida Tomato Committee education and promotion expendi tures have a positive effect

PAGE 43

43 on the quantity of Florida tomatoes demanded domestically but reject H ypothesis 1d that they have a positive effect on their price. For H ypothesis 2a we fail to reject that Food safety incidents negatively impacts the quantity of U.S tomatoes demanded domestically but re ject H ypothesis 2b that it has a negative effect on their price of U.S. tomatoes. For H ypotheses 2c we fail to reject that food safety has a significantly negative effect but for H ypothesis 2d we must reject that it negatively impact the price of Florida t omatoes For H ypotheses 3a and Hypothesis 3b we must reject that Florida Tomato Committee education and promotion expenditures effectively reduce the imp act of food safety incidents on the U.S. demand for fresh tomatoes or the U.S. demand for Florida fres h tomato Table 5 1 shows these results according to their t statistics

PAGE 44

44 Table 4 1 Regression o utput from USQ m odel SUMMARY OUTPUT Regression Statistics Multiple R 0.985 R Square 0.969 Adjusted R Square 0.966 S tandard Error 0.034 Observations 39.000 ANOVA df SS MS F Significance F Regression 4.000 1.265 0.316 268.415 0.000 Residual 34.000 0.040 0.001 Total 38.000 1.305 Coefficients Standard Error t Stat P value Lower 95% Upper 95 .0 % Lower 95.0% Upper 95.0% Intercept 5.352 0.604 8.856 0.000 6.580 4.124 6.580 4.124 rUSGP 0.005 0.047 0.113 0.911 0.101 0.091 0.101 0.091 r PROMO 0.005 0.002 2.925 0.006 0.001 0.008 0.001 0.008 Ill 0.0 45 0.016 2.861 0.007 0.077 0.013 0.077 0.013 rI NC 0.782 0.051 15.466 0.000 0.679 0.885 0.679 0.885

PAGE 45

45 Table 4 2 Regression output from FLQ m odel SUMMARY OUTPUT Regression Statistics Multiple R 0.8881 R Square 0.7887 Adjusted R Square 0.7567 Standard Error 0.1242 Observations 39 ANOVA df SS MS F Significance F Regression 5 1.9021 0.3804 24.642 3E 10 Residual 33 0.5095 0.0154 Total 38 2.4115 Coefficients Standard Error t Stat P value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 5.751 3.0091 1.911 0.0647 11.87 0.3706 11.87 0.3706 rPROMO 0.0204 0.0061 3.3394 0.0021 0.008 0.0328 0.008 0.0328 USIMP 0.66 0.0976 6.762 1E 07 0 .858 0.461 0.858 0.461 rINC 0.7467 0.3077 2.4265 0.0209 0.1206 1.3727 0.1206 1.3727 ILL 0.104 0.0599 1.735 0.0921 0.226 0.0179 0.226 0.0179 rFLP 0.397 0.1394 2.85 0.0075 0.681 0.114 0.681 0.114

PAGE 46

46 Table 4 3 Regression o utput from FL P m odel SUMM ARY OUTPUT Regression Statistics Multiple R 0.815 R Square 0.665 Adjusted R Square 0.602 Standard Error 0.141 Observations 39.000 ANOVA df SS MS F Significance F Regr ession 6.000 1.256 0.209 10.572 0.000 Residual 32.000 0.633 0.020 Total 38.000 1.889 Coefficients Standard Error t Stat P value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 1.366 4.637 0.294 0.770 8.081 10.812 8. 081 10.812 FLQ 0.532 0.193 2.752 0.010 0.926 0.138 0.926 0.138 rP ROMO 0.003 0.008 0.413 0.682 0.013 0.020 0.013 0.020 USIMP 0.329 0.161 2.048 0.049 0.656 0.002 0.656 0.002 rINC 0.094 0.479 0.197 0.845 1.069 0.881 1.069 0.881 ILL 0.1 09 0.069 1.589 0.122 0.248 0.031 0.248 0.031 TREND 0.036 0.081 0.444 0.660 0.129 0.200 0.129 0.200

PAGE 47

47 Table 4 4 Regression o utput from USGP m odel SUMMARY OUTPUT Regression Statistics Multiple R 0.835 R Square 0.697 Adjusted R Square 0.651 Standard Error 0.119 Observations 39.000 ANOVA df SS MS F Significance F Regression 5.000 1.081 0.216 15.161 0.000 Residual 33.000 0.471 0.014 Total 38.000 1.552 Coefficients Standard Error t Stat P value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 2.675 4.534 0.590 0.559 11.900 6.550 11.900 6.550 rPROMO 0.009 0.007 1.335 0.191 0.022 0.005 0.022 0.005 ILL 0.006 0.062 0.104 0.918 0.133 0.120 0.133 0.120 rINC 0.3 35 0.594 0.565 0.576 0.872 1.543 0.872 1.543 USQ 0.448 0.625 0.717 0.479 1.719 0.824 1.719 0.824 TREND 0.123 0.061 2.030 0.050 0.247 0.000 0.247 0.000

PAGE 48

48 Table 4 5 Hypothesis test decision r esults Hypothesis Coefficient t stat Result 1a 0.00 5 2.9 3 Fail to Reject 1b 0.0 09 1.34 Reject 1c 0.020 3. 39 Fail to Reject 1d 0.00 3 0. 41 Reject 2a 0.0 45 2.9 3 Fail to Reject 2b 0.00 6 0. 10 Reject 2c 0. 104 1.7 4 Fail to Re ject 2d 0. 109 1.59 Reject 3a 0.005 0.76 Reject 3b 0. 124 2.82 Reject

PAGE 49

49 C HAPTER 5 CONCLUSION AND RECOM MENDATIONS Review of Study This study analyze s the factors which influence both the price of and demand for total U.S. and Florida tomatoes consumed domestically. It also provides an evaluation of the returns on investment from education an d promotion expenditures in the Florida Tomato market. In 1986 the Florida Tomato Committee Marketing Order was amended to permit educ ation and promotion programs to be funded by moneys collected from the Committee members. The purpose of these programs has been to increase demand for Florida tomatoes as well as increase profits to producers. The Marketing Order also began using Market A ccess Promotion (MAP) activities in the 1997/1998 season with partial funding by the U.S. Department of Agriculture. MAP programs were introduced in order to increase Florida fresh tomato exports to Canada and Asia and to increase foreign consumption but w ere not considered in the study as these dollars were invested in order to increase overseas consumption This study uses econometric model s originally created for other commodities which were designed to evaluate the effectiveness of education and promot ion expenditures and adapts it to accurately represent the Florida tomato market. Unlike previous studies, this evaluation considers food safety as a primary variable in the determination of tomatoes consumed on an annual basis as well as in a new price dep endent model designed to show a more short run response to investment for Florida tomato growers Finding s show that education and promotion expenditures have resulted in the increase of both price and demand for tomatoes. For the quantity dependent model, a verage sales revenue increased by $4.75 for each dollar spent on

PAGE 50

50 these programs during years 2005 2010. Additionally, each dollar invested in these programs returned an average of $1.01 in marginal profits to producers during the same time period. Accord ingly, it can be seen that producers are able to realize even more benefit with increases in program expenditures. For the price dependent model, average sales revenue increased by $3.52 for each dollar invested in promotion. F or the price evaluation, aver age revenue is also known as profit to producers. Food safety has become a significant determinant in estimating the quantity of tomatoes that will be sold annually across the country. This study shows that a 1% increase in the number of tomato related sa lmonella illnesses in the U.S. yields a .014% decrease in the quantity of Florida tomatoes consumed and a .005% decrease in price The data does not show that increases in education and promotion expenditures reduced the negative effects of food safety. Th is can be explained by the fact that the Florida Tomato Committee has not focused their education and promotion program Despite the absence of a food safety promotion program, results are conclusive that current education and promotion activities do yield significantly p ositive returns to the Florida t omato growers Producers in Florida realize substantial benefits from each dollar spent on promotion efforts. Recommendations for Future Studies The data used for this study was arranged by annual time series according to the information which was mostly made available by the Florida Tomato Committee, the nual the use of annual data has been effective, the use of monthly data could help to provide more accurate models with a higher level of statistical significance for some of the

PAGE 51

51 explanatory vari ables and could support an even more practical evaluation of the participation, a seasonal analysis may offer additional insights into the operational benefits of promoting he avily during particular seasons An analysis of specific Fl orida Tomato Committee promotion activities, such as their television commercials or their trade publications could be performed to dete rmine to most effective use of c ommittee promotional funds. Monthly data is available for each of the varia bles used in all of the models. H owever specific monthly promotion expenditures, monthly market prices, and monthly yields for Florida tomatoes would need to be identified in order to perform the above specified analyses. The U.S. tomato market has also seen an inc rease in the availability of Greenhouse products imported from around the world. In 1997, 1,471,000 pounds of Greenhouse tomatoes were sold in the U.S. A verage annual production has reached nearly 3,500,000 pounds since then and 2010 data shows that rates have already reached 4,500,000 pounds. Growing at an average rate of 8%, the amount of greenhouse tomatoes sold has begun to pose a threat to the products of the much older Florida field grown method s two main competitors. Mexican greenhouse production alone has increased its market share from only 15% in the 2006/2007 season, when Florida field grown tomatoes held 57% of the total market, to 27% in the 2009/2010 season, when Florida field grown held only 33% of the total market. For this reason, future evaluations may find it necessary to include U.S. consumption of greenhouse tomatoes in the model either in the form of a

PAGE 52

52 Florida greenhouse production variable and/or a foreign import greenhouse consum ption variable.

PAGE 53

53 APPENDIX ADDITIONAL DATA Table A 1 Definition and source of regression v ariables Variable Definition Source FLQ Annual Per capita consumption of Florida tomatoes Calculated as total annual production by Florida Tomato Committee annua l reports (1972 2010) divided by U.S. International Macroeconomic Data Set (2010) USQ Annual per capita consumption of US fresh tomatoes Economic Research Service Vegetables and Specialties Situation and Outlook Yearbook (USDA, 2010) USIMP Per capita consumption of imported tomatoes Calculated as total annual imports of tomatoes reported by Economic Research Service Vegetables and Specialties Situation and Outlook Yearbook (USDA, 2010) divided by U.S. Pop ulation rPROMO Education and promotion expenditures by the Florida Tomato Committee Florida Tomato Committee annual reports (1972 2010) GDP Deflator Base year 2005 Economic Research Service International Macroeconomic Data Set (USDA, 2010) rINC Real per capita Gross Domestic Product for the U.S. Economic Research Service International Macroeconomic Data Set (USDA, 2010) USGP Real average per pound price of U.S. tomatoes in dollars Economic Research Service Vegetables and Specialties Situation and Outloo k Yearbook (USDA, 2010) FLP Real average per pound price of Florida tomatoes in dollars Calculated as Florida cents per pound by the Florida Tomato Committee annual reports (1972 2010), divided by GDP deflator, multiplied by 100 ILL Annual number of sal monella illnesses in the U.S. related to tomato consumption Center of Disease Control Outbreak Net Surveillance Data: Historical Data (1990 1997) and CDC Food Outbreak Online Database (all year, all state, all location, salmonella)

PAGE 54

54 Table A 2 Data se t for USQ m odel Y ear US Q rPROMO rINC rUSG P ILL ILL*rPROMO GDP DEF 1972 12.09 0.00 22188.99 57.77 0 0 25.88 1973 12.48 0.00 23244.08 59.12 0 0 27.17 1974 11.84 0.00 22900.79 61.96 0 0 28.35 1975 12 .00 0.00 22627.34 64.69 0 0 29.94 1976 12.56 0.00 23611 .09 55.06 0 0 32.63 1977 12.36 0.00 24449.85 62.43 0 0 35.72 1978 12.92 0.00 25542.22 57.44 0 0 37.78 1979 12.42 0.00 26050.58 55.08 0 0 40.18 1980 12.83 0.00 25675.03 51.39 0 0 43.01 1981 12.31 0.00 26069.69 50.25 0 0 46.57 1982 12.88 0.00 25320.67 45.43 0 0 50.81 1983 13.47 0.00 26223.74 54.28 0 0 55.57 1984 14.2 0 1054965.06 27866.01 46.34 0 0 58.96 1985 14.85 1023069.33 28762.99 39.10 0 0 61.29 1986 15.8 0 489081.33 29485.76 48.94 0 0 63.59 1987 15.84 539686.41 30158.21 44.44 0 0 65.52 1988 16 .83 581871.45 31114.24 42.88 0 0 66.98 1989 16.84 1074688.16 31922.96 54.02 0 0 68.79 1990 15.52 1531274.23 32156.92 40.87 102 156189971.11 71.15 1991 15.36 1800120.01 31655.74 50.44 600 1080072003.66 73.84 1992 15.45 2126019.59 32278.91 46.68 34 72284 666.06 76.70 1993 16.3 0 2515680.76 32764.87 45.61 143 359742349.28 79.37 1994 16.21 2468093.47 33683.89 35.76 339 836683684.73 81.20 1995 16.84 1660377.67 34122.18 33.71 36 59773596.23 83.07 1996 17.4 0 1441717.29 34989.61 36.87 48 69202429.75 84.83 19 97 17.29 532943.73 36112.64 36.83 34 18120086.70 86.57 1998 18.5 0 777128.42 37247.95 41.31 86 66833044.20 88.21 1999 19.07 1039591.90 38600.18 33.01 0 0 89.67 2000 18.98 915733.96 39748.81 30.40 86 78753120.56 90.66 2001 19.2 0 763749.98 39771.08 39.84 0 0 91.97 2002 20.31 619350.39 40107.17 33.28 679 420538916.32 93.98 2003 19.41 672436.67 40767.91 39.86 61 41018636.96 96.23 2004 19.95 712860.54 41792.11 32.84 566 403479065.05 97.92 2005 20.15 798330.00 42680.86 50.00 277 221137410.00 100.00 2006 1 9.77 937067.14 43333.60 39.79 349 327036430.74 103.24 2007 19.21 1043844.05 43724.86 28.94 149 155532763.79 106.28 2008 18.51 959823.24 43178.16 50.50 1538 1476208142.33 108.60 2009 19.59 1086843.37 41312.85 29.67 21 22823710.77 109.59 2010 20.84 10592 52.53 42189.41 52.02 0 0 110.64

PAGE 55

55 Table A 3 Data set for FLQ m odel Year FLQ rPROMO USIMP rINC rFLP ILL ILL*rPROMO GDP DEF 1972 2.94 0.00 2.80 22188.99 57.77 0 0 25.88 1973 3.00 0.00 3.60 23244.08 59.12 0 0 27.17 1974 2.96 0.00 2.80 22900.79 61.96 0 0 28.35 1975 3.45 0.00 2.60 22627.34 64.69 0 0 29.94 1976 3.62 0.00 3.00 23611.09 55.06 0 0 32.63 1977 2.86 0.00 3.60 24449.85 62.43 0 0 35.72 1978 3.31 0.00 3.70 25542.22 57.44 0 0 37.78 1979 4.01 0.00 3.20 26050.58 55.08 0 0 40.18 1980 4.54 0.00 2.90 25675.03 51.39 0 0 43.01 1981 4.45 0.00 2.30 26069.69 50.25 0 0 46.57 1982 4.80 0.00 2.60 25320.67 45.43 0 0 50.81 1983 4.88 0.00 3.20 26223.74 54.28 0 0 55.57 1984 4.81 1054965.06 3.50 27866.01 46.34 0 0 58.96 1985 5.50 1023069.33 3.60 28762.99 39.1 0 0 0 61.29 1986 5.45 489081.33 4.10 29485.76 48.94 0 0 63.59 1987 5.80 539686.41 3.80 30158.21 44.44 0 0 65.52 1988 6.61 581871.45 3.30 31114.24 42.88 0 0 66.98 1989 6.56 1074688.16 3.50 31922.96 54.02 0 0 68.79 1990 5.27 1531274.23 3.20 32156.92 40. 87 102 156189971.11 71.15 1991 5.46 1800120.01 3.10 31655.74 50.44 600 1080072003.66 73.84 1992 6.99 2126019.59 1.70 32278.91 46.68 34 72284666.06 76.70 1993 6.04 2515680.76 3.50 32764.87 45.61 143 359742349.28 79.37 1994 5.60 2468093.47 3.30 33683.89 35.76 339 836683684.73 81.20 1995 5.20 1660377.67 5.10 34122.18 33.71 36 59773596.23 83.07 1996 4.38 1441717.29 6.00 34989.61 36.87 48 69202429.75 84.83 1997 4.39 532943.73 6.00 36112.64 36.83 34 18120086.70 86.57 1998 4.31 777128.42 6.80 37247.95 41.3 1 86 66833044.20 88.21 1999 5.08 1039591.90 5.66 38600.18 33.01 0 0 89.67 2000 5.14 915733.96 5.14 39748.81 30.40 86 78753120.56 90.66 2001 4.71 763749.98 5.76 39771.08 39.84 0 0 91.97 2002 4.75 619350.39 5.82 40107.17 33.28 679 420538916.32 93.98 200 3 4.39 672436.67 6.16 40767.91 39.86 61 41018636.96 96.23 2004 4.95 712860.54 5.85 41792.11 32.84 566 403479065.05 97.92 2005 4.48 798330.00 5.68 42680.86 50.00 277 221137410.00 100.00 2006 4.01 937067.14 5.85 43333.60 39.79 349 327036430.74 103.24 200 7 4.35 1043844.05 6.11 43724.86 28.94 149 155532763.79 106.28 2008 3.71 959823.24 6.23 43178.16 50.50 1538 1476208142.33 108.60 2009 3.83 1086843.37 6.32 41312.85 29.67 21 22823710.77 109.59 2010 2.25 1059252.53 10.90 42189.41 52.02 0 0 110.64

PAGE 56

56 Table A 4 Corre lations between Florida regression v ariables Variable TREND FLQ rPROMO USIMP rINC rFLP ILL ILL*rPROMO TREND 1 FLQ 0.154 1 rPROMO 0.526 0.559 1 USIMP 0.796 0.316 0.168 1 rINC 0.989 0.131 0.477 0.793 1 rFLP 0.726 0 .417 0.516 0.447 0.719 1 ILL 0.695 0.226 0.643 0.362 0.689 0.621 1 ILL*rPROMO 0.442 0.398 0.868 0.052 0.402 0.446 0.833 1 Table A 5 Correlations between U.S. regression v ariables Variable TREND rUSGP rPROMO ILL rINC USQ ILLXPromo TREND 1 rUSGP 0.75862 1 rPROMO 0.381708 0.52696 1 ILL 0.694651 0.54961 0.607349 1 rINC 0.989058 0.73646 0.333597 0.689026 1 USQ 0.964288 0.75813 0.326735 0.619666 0.973469 1 ILLXPromo 0.371374 0.41014 0.896402 0.775699 0.32409 0.268687 1

PAGE 57

57 Table A 6. Augmented FLQ m odel w ith interaction t erms SUMMARY OUTPUT Regression Statistics Multiple R 0.882 R Square 0.777 Adjusted R Square 0.736 Standard Error 0.129 Observations 39.000 ANOVA df SS MS F Signifi cance F Regression 6.000 1.875 0.312 18.635 0.000 Residual 32.000 0.537 0.017 Total 38.000 2.412 Coefficients Standard Error t Stat P value Lower 95% Upper 95% Lower 95.0% Upper 95.0% Intercept 3.530 3.527 1.001 0.324 10 .713 3.654 10.713 3.654 rPROMO 0.020 0.007 3.099 0.004 0.007 0.034 0.007 0.034 USIMP 0.658 0.102 6.463 0.000 0.865 0.450 0.865 0.450 rINC 0.697 0.338 2.062 0.047 0.008 1.387 0.008 1.387 ILL 0.003 0.024 0.104 0.917 0.047 0.052 0.047 0.052 rFLP 0.371 0.142 2.605 0.014 0.662 0.081 0.662 0.081 ( ILL*PROMO ) 0.005 0.006 0.757 0.455 0.018 0.008 0.018 0.008

PAGE 58

58 Table A 7 Returns to Florida p roducers (2005 2010) b ase d on FLQ m odel Year ABC MGMBC 2005 0.00221 $ 5.07 $2.25 2006 0.00300 $3.55 $1.00 2007 0.00508 $2.08 $(0.10) 2008 (0.00164) $3.99 $1.49 2009 0.00455 $1.93 $(0.03) 2010 (0.01430) $11.90 $1.44 2005 2010 (0.00018) $4.75 $ 1.01 Table A 8 Returns to F lorida p roducers (2005 2010) based on FLP m odel Year ABC MGMBC 2005 0.0003 $7.35 $7.35 2006 0.0004 $5.77 $5.77 2007 0.0007 $5.04 $5.04 2008 0.0003 $4.73 $4.73 2009 0.0006 $4.02 $4.02 2010 0.0082 $8.33 $8.33 2005 20 10 0.0011 $ 5.87 $5.87

PAGE 59

59 Table A 9 Percent of total fresh tomato production by c ountry Country Oct. Nov. Dec. Jan. Feb. Mar. Apr. May Total Canada 13.08 6.42 1.04 0.07 0.03 1.05 6.20 8.79 3.85 Guatemala 0.23 0.18 0.24 0.20 0.01 0.35 0.31 0.19 0.22 Dominican 0.02 0.14 0.14 0.15 0.25 0.20 0.10 0.09 0.14 Belgium 0.12 0.07 0.00 0.01 0.01 0.01 0.03 Netherlands 0.17 0.45 0.27 0.31 0.01 0.00 0.00 0.14 New Zealand 0.02 0.03 0.01 0.01 0.01 Costa Rica 0.03 0.06 0.09 0.07 0.03 0.03 Col umbia 0.00 0.00 Israel 0.03 0.07 0.05 0.03 0.01 0.03 France 0.00 0.00 Total Imports 85.78 62.34 52.17 73.72 94.44 95.48 81.63 63.06 75.46 Non Mexican Imports 13.39 7.37 1.89 0.84 0.81 1.71 6.67 9.08 4.49 Florida 14.22 37.66 47.83 26.28 5.56 4.52 18.37 36.94 24.54

PAGE 60

60 Table A 10 Other commodity benefit cost r esults Commodity Promotion Program Benefit Cost Ratio Author(s) Tomatoes, fresh 27.15 average VanSickle and Evans (2001) Fluid Mil k 4.29 average Kaiser (2000) Beef 5.7 average Ward (1996, 2001) Pork 16 average Davis, et al. (2000) California Avocados 5 6 Carman and Craft (1997) California Prunes, domestic 11 (avg. of five models) Alston, et al. (1998) California Table Grapes, domestic 145 Alston, et al. (199 7) California Table Grapes, domestic 80 Alston, et al. (1997) California Table Grapes, export 8 Alston, et al. (1997) California Table Grapes, export 4 Alston, et al. (1997) Pork 237 marginal Piggott (1997) Beef 15.5 marginal Piggott (1997) Californi a Walnuts, export 6 Weiss, Green and Havenner (1996) Canadian Butter 1 Goddard and Amuah (1989) Canadian Margarine 1 Goddard and Amuah (1989) U.S. Soybeans, export 58 Williams (1985) Washington Apples 7 Ward and Forker (1991)

PAGE 61

61 Figure A 1 Trends in F lorida t omato c onsumption Figure A 2 Trends in Florida tomato p rices 0 1 2 3 4 5 6 7 8 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Per Capita Consumtion in LBS Season Tracking FLQ 0 10 20 30 40 50 60 70 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 Per Pound Price in Cents Season Tracking FLP

PAGE 62

62 LIST OF REFERENCES Alston, J.M., H.F. Carman, J.A. Chalfant, J.M. Crespi, R.J. Sexton, and R.J Venner. The dation Research Report no. 344, March 1998. Alston, J.M., J.A. Chalfant, J.E. Christian, E. Meng, and N.E. Piggot. The California Table Grape 43, November 1997. B.C. French an 581 593. Available at http://www.jstor.org/stable/1242811 Carman, H.F., and R.K. Craf Marketing Programs, 1961 University of California, Davis, February 4, 1997. Center of Disease Control Food Outbreak Online Database: all y ears, all states, all Locations, salmonella. Available at http://wwwn.cdc.gov/foodborneoutbreaks/ Center of Disease Control Outbreak Net Surveillance Data: Historical Data (1990 1997). Available at http://www.cdc.gov/outbreaknet/surveillance_data.html Cr espi, J.M., (2003): 315. Available at http://aepp.oxfordjournals.org.lp.hscl.ufl.edu/content/25/2/294.full.pdf+html Crespi, California, Davis, 1999. Davis, G.C., O. Capps Jr., D.A. Bessler, J.H. Leigh, J.P. Nichols and E. Goddard (2000): An Economic Evaluation of the Pork Checkoff Program, College Station Texas, A&M University. Ding L. and H.W. Kinnucan (1996): Market Allocation Rules for Non Price Promotion with Farm Programs: US Cotton, Journal of Agricultural and Resource Economics, 21, 351:367. 2010. Ferro, G. Food Safety Standards I n Tomatoes and T heir Effects on R isk Mitigation f or Growers. Food and Resource Economics Department University of Flor ida 2010. of American Journal of Ag ricultural Economics. 71 (1989) : 741 749. Kaiser, H.M., D.J., Liu, T. Consignado (2003): An Economic Analysis of California Raisin Export Promotion, Agribusiness, Vol. 19 (2), 189 201.

PAGE 63

63 Kaiser, H.M., and C. Chung (2002): Impact of Generic Milk Advertising on New York State Markets, 1986 2000, Dept. of Applied Economics and Management, College of Agriculture and Life S ciences, Cornell University. Kinnucan H.W. and Y. Zheng (2001): National Cost Benefit Estimates for Dairy, Beef Pork and Cotton Promotion Programs. Murray, B.C., R.C. Beach, W.J. White, C. Viator, N., Piggott, and M. Wohlgenant (2001): An Economic Analy sis of Cotton Research and Promotion Programs, Research Triangle Park, NC. Piggot, N.E. (1997): The Benefit and Costs of Generic Advertising of Agricultural Commodities, PhD Dissertation, University of California, Davis. Rosen S. Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition Journal of Political Economy 82 (1974) : 34 55 .ers.usda.gov/Data/Macroeconomics/ VGS 2016. Economic Research Service, USDA. Pr Promotion Program 2001 rsity of Florida. Commission WAC 91:1, University of Florida, Gainesville, 1991. Ward R.W. (1996): Beef Checkoff Impact on US Beef Demand, UF NC A 1996 No.1. Ward, R.W. (2001): Beef Demand and Its Response to the Beef Checkoff, NCA 2001, No.1. ase Study of Generic Promotion u Commodity Promotion Policies and Programs in the Global Agri Food System, Proceedings of NEC63 Conference, NICPRE. Cornell University, May 1996: 47 80. 243 261.

PAGE 64

64 BI OGRAPHICAL SKETCH Samuel Einsohn was born in 1988 in Miramar Florida. The youngest of three boys, he grew up in Cooper City Florida, graduating in the top 10% of his class at Cooper City High School in 2006 He then attended the Roth berg International Sc hool at The Hebrew University in Jerusalem, Israel until 2007 when he returned home to the U.S. and received an A. A. in a ccounting in 2008 from Sante Fe College H e then earned his B.S. f ood a nd resource e conomics with an emphasis on international e conomi cs from the University of Florida (UF) in 2010 Samuel enrolled in the Master of Science degree seeking program l at er that year at UF, majoring in food and resource e conomics During his time in u ndergraduate studies, Samuel sat of the board of the both th e Hillel and Lubavitch Chabad Student Group s and in 200 8 led Chabad at the University Campus Convention in Crown Heights, New York Throughout graduate school he continued his leadership with the Lubavitch Cha bad Student Center and joined several professional organizations for agricultural economists.