Welcome

Brochure

Current Projects

Research Reports
and Bulletins

Newsletters

Vol.1 No.1
Vol.1 No.2
Vol.1 No.3
Vol.1 No.4
Vol.2 No.1
Vol.2 No.2
Vol.2 No.3
Vol.2 No.4
Vol.3 No.1
Vol.3 No.2
Vol.3 No.3
Vol.3 No.4

Newsletter--Volume 3, Number 2

“Got Milk?” Who hasn’t been tempted by the image of a frothy glass of milk and a plate of chocolate chip cookies? And everyone knows “milk does a body good.” Whether broadcast nightly into your living room or printed in a magazine, these recurring images are courtesy of a producer-supported, national dairy promotion program. U.S. dairy farmers, through their promotion organizations, have increased fluid milk demand and improved retail prices through such advertising. But are producers getting the biggest bang for their advertising buck?

The national dairy promotion program finds its roots in the Dairy and Tobacco Adjustment Act (1983) which authorized the current assessment of 15 cents per hundred pounds of milk marketed in the continental United States. Totaling about $200 million per year, the assessment funds activities such as nutrition research, education, promotion, and fluid milk advertising. The funds earmarked for generic milk advertising have changed over time. On average, $20 million per year (discounted to the value of a dollar in 1982) is spent on generic milk advertising, with spending generally increasing over time. A 1997 study by Kaiser found the advertising of dairy products increased milk demand by 2.14 percent. The increase improved farm milk prices by 2.9 percent.

Advertising dollars are spent on four distinct media outlets--television, print, radio, and outdoor (e.g., billboards). These advertising expenditures are not spread evenly across all outlets; rather, television receives more advertising dollars per quarter than the other media outlets combined. On average, television receives 89 percent of the advertising budget while print receives 5 percent, radio has 4 percent, and outdoor receives 2 percent. The reason for the bias toward television has to do with consumer response. Some media outlets engage the consumer more quickly and the advertising message may be retained longer. This is true for television relative to other media outlets, which is why it receives so much of the advertising budget.

The relative effectiveness of the various modes of advertising may be estimated using an economic simulation model. This econometric model relates the amount of fluid milk sold in each quarter to important factors affecting milk demand including the retail price of milk, consumer income, price of milk substitutes, and, in particular, the amount of television, radio, print, and outdoor advertising. By relating the demand factors to fluid milk sales, the econometric model can approximate net percentage changes in sales for a given amount of change in television, radio, print, or outdoor advertising expenditures. For an additional 1 percent allocated to television advertising, fluid milk demand increases by 0.0022 percent or roughly 299,000 pounds in a quarter. In contrast, print provides only a 0.0004 percent increase, radio a 0.0003 percent increase, and outdoor a 0.0006 percent increase, per 1 percent increase in their respective advertising expenditures. At first glance, television is the most effective media outlet, followed distantly by the other three options.

While it has the greatest impact, television has been overused in dairy advertising. Although it has the largest advertising response to a 1 percent increase in expenditures, television’s expenditure base is substantially larger than that of other advertising modes. For instance, between 1984 and 1993, a 1 percent increase in advertising expenditures (in 1982 dollars) was the equivalent of $33,000 per quarter for television, $2,000 for print, $1,600 for radio, and $900 for outdoor. A 1 percent increase in television advertising would have required an investment that was 16.5 times larger than the same increase in print advertising. Since the advertising responsiveness of television is only 5.5 times larger than that of print, reallocation of expenditures from television to print would clearly be profitable. This crude simple analysis also applies to potential reallocations between television and other media. To determine the exact extent of the optimal reallocation requires a more sophisticated computation, as was done in our research. Intuitively, as expenditures are added incrementally to a particular advertising mode, less and less is gained from subsequent exposure because of the principle of diminishing marginal returns. This means that the most profitable approach to advertising is a diversified one--funds should be distributed among different media types in an optimal mix so that returns can be maximized.

How might the optimal mix of advertising be determined? Once the relationship between demand factors and fluid milk sales has been estimated, we can then consider how the milk supply responds to changes in fluid milk sales. An increase in fluid milk sales leads to higher prices, and higher prices encourage more milk production. The increased production is a supply response. A second econometric model quantifies the supply response by relating important supply factors (like fluid milk price and cost of production) to the amount of milk produced in each quarter. The resulting supply and demand estimates become important components of an optimal control model. The goal of the optimal control model is to achieve the greatest discounted profits (fluid milk sales minus advertising expenditures) for a given time period. To do this, the optimal control model adjusts the advertising expenditure level of each media outlet incrementally while staying within the advertising budget. Information on supply and demand is used to determine the amount of fluid milk sold, the price of milk, and profits. In this manner, the model can control both the level of estimated demand in each quarter and the subsequent estimated supply response. Ultimately, the optimal control model arrives at an advertising mix that maximizes profits for a given time period.

What is the optimal mix of advertising? On average, the best mix of funding is found when television receives 70 percent of the advertising budget--down from the previously mentioned 89 percent. Outdoor spending increases from 2 to 15 percent, radio grows from 5 to 6 percent, and print changes from 5 to 9 percent.

What would a switch to the optimal advertising mix mean to dairy producers? Suppose we go back in time to the period 1984-1993 and switch from the typical mix of advertising to the optimal one. Milk demand increases in each quarter with the switch. Both the typical and optimal advertising mixes share the same “high” and “low” demand quarters. These peaks and valleys reflect the seasonal nature of consumer demand. Yet, demand from the optimal advertising mix is greater than the typical advertising mix. Had this extra milk been sold, dairy producers would have made an additional $265 million for the period, or about $29 million per year. Furthermore, the switch could have been achieved for virtually no cost--it would have simply been a matter of putting the right funds in the right place.

Clearly, dairy promotion programs influence the amount of fluid milk consumed in the United States each year. Television is by far the most effective media outlet--its effectiveness is easily three times that of alternatives. But the possibility of having too much of a good thing exists--the effectiveness of television is subject to diminishing marginal returns. To get the biggest bang from the milk advertising buck requires an optimal mix of media outlets. Our research shows that switching from the typical advertising mix to the optimal mix is worth an additional $29 million per year to dairy producers.

*The economic model used in this study is currently being refined. Consequently, the results reported here may be somewhat different than those reported in the future due to minor modifications.

Data for this article is from the Leading National Advertisers and USDA Dairy Situation and Outlook.

James G. Pritchett and Donald J. Liu are graduate research assistant and assistant professor, respectively, in the Department of Applied Economics, University of Minnesota. Harry M. Kaiser is an associate professor in the Department of Agricultural, Resource, and Managerial Economics, Cornell University, and is also the director of NICPRE.

Editor's Notes
Jennifer L. Ferrero

Can funds be reallocated in such a way as to increase returns without increasing expenditures? This is the question addressed by Pritchett, Liu, and Kaiser in this quarter’s lead article. Harry Kaiser, the director of NICPRE, discusses the Institute and past research projects it has sponsored in his column on page 4.

Nichols, Capps, Davis, and Bessler of Texas A&M University recently evaluated the research and promotion activities funded by the cotton checkoff. A shorter version of the full technical report written by Texas A&M as well as a perspective from David Fraser, Vice President of Communications with the Cotton Board, can also be found in this issue.

As editor of the NICPRE Quarterly, I’m always interested in hearing what our readers would like to see more of in future issues. Is there a topic or issue you feel is missing from the Quarterly? Our staff is always open to feedback and new ideas so call, write, e-mail, or fax us today!

EVALUATING RETURNS TO THE COTTON CHECKOFF PROGRAM
John P. Nichols, Oral Capps, Jr., George C. Davis, and David A. Bessler

Over the past 30 years, Congress has enacted legislation authorizing 16 freestanding research and promotion programs for agricultural commodities. The cotton checkoff, established in 1966, was the first of these. An analysis we recently completed at Texas A&M University has concluded that this program, over the period 1986 through 1995, was successful in increasing domestic consumption of cotton and farm-level demand. In addition, since 1991, the return on investment under the current program, measured at the mill level, was greater than $7 for each dollar of assessment. Measured at the producer level, net returns for the current program were estimated to be $3.2 to $3.5 per dollar invested.

Cotton Checkoff Background
The Cotton Research and Promotion Act was initiated to provide for the development of an industry-wide research and promotion effort. In 1990, Congress amended the Act to eliminate producer refunds and to add a new levy on imports. The 1990 legislation also called for a review by the Secretary of Agriculture to assess the effects of these changes and to determine the need for a continuance referendum every five years. The research described in this article was undertaken in 1996 to provide an economic analysis of the cotton checkoff program with the inclusion of imports subsequent to the 1990 legislation. This analysis, plus USDA’s assessment of program implementation and industry support, provided the basis for the Secretary’s review.

The checkoff program is financed through an assessment of $1 per bale plus 0.5 percent of the value of the cotton. Over the period of this analysis (1986-1995), total funding grew from $18.5 million in 1986 to $55.1 million in 1995. Funds from the checkoff program are collected and administered by the Cotton Board, located in Memphis, Tennessee. All research and promotion activities funded by the checkoff are managed by Cotton, Incorporated (CI) which is under contract with the Cotton Board. CI is headquartered in New York with its research division based in Raleigh, North Carolina.

Research Objectives
Texas A&M undertook the evaluation project in late 1995 with an agreement to examine two related questions: (1) what are the effects of textile research and promotion activities on domestic consumption of cotton and farm-level demand? and (2) what is the rate of return associated with the cotton checkoff program? The analysis was designed to examine separately, to the extent possible, the periods before and after the 1990 amendments took effect.

Cotton Use Trends
Monthly data for research and promotion expenditures were only available back to 1986 so this analysis is based on observations from January 1986 through July 1995. This period provides sufficient observations and covers both sides of the program change, allowing for a sufficiently large sample to carry out the econometric analysis. Also by starting in 1986, the confounding effects of a major change in the 1985 Farm Bill, which altered price support programs, could be avoided. Thus, the ten-year interval used for the analysis constitutes a relatively homogeneous period in terms of farm program shifts.

In the analysis, domestic consumption is the sum of U.S. mill consumption of cotton plus net imports of raw fiber-equivalent cotton textile products (yarn, thread, and fabric; apparel; and home furnishings). Industrial uses for cotton fiber were not included. Farm-level demand for cotton is defined as the sum of U.S. mill consumption plus U.S. exports of raw cotton. These data were all obtained from USDA publications. A more detailed discussion of data sources and trends, as well as additional tables, are available in the full technical report cited at the end of this article.

Research and Promotion Activities
Marketing activities account for the largest expenditures in the CI budget. In all years analyzed, domestic marketing accounted for over 50 percent of expenditures. Other important expenditure categories include international marketing (11 percent in 1995), textile research and development (20 percent), agricultural research (10 percent), and administration (4 percent). Our analysis incorporated the two primary demand-enhancing activities: promotion (U.S. and international) and textile research and development (fiber processing, textile research and implementation, fiber quality). Agricultural or agronomic research was not included.

Promotion activities include media advertising, public relations, and marketing research. These are targeted to consumers of apparel and home fabrics, as well as the fashion and retail industries. International marketing includes a wide array of information, technical, and promotional services.

Textile research and development includes both quality of fiber and processing technology. These activities include substantial expenditures on technical services and the transfer of research findings to mills and textile manufacturers.

Evaluation Approach
Our primary research problem was to determine whether the textile research and promotion expenditures of CI (representing about 86 percent of total expenditures) affected domestic consumption and if so, by how much. This study was unique in that two different, independent modeling approaches (the structural/econometric method and the time-series method) were used to estimate this basic relationship. In each case, statistical models were developed to estimate critical relationships.

The structural/econometric model approach is based on a description of the relationships among all key variables consistent with economic theory about the behavior of firms and markets involved. The time-series model approach does not attempt to impose this structural framework and relies only on a few key statistical restrictions to determine the existence of relationships among variables. Results are based on the same data set, yet were generated from two distinctly different modeling approaches. This dual track effort serves as a check on the “robustness,” or strength, of the results.

The first step of analysis was to identify the factors influencing domestic consumption of cotton at mill-level, farm-level, and imports. Promotion and research expenditures were the primary explanatory variables but many others must be accounted for to minimize the confounding of effects. In general, these include domestic prices of cotton and other major fibers, world cotton prices, government support programs, domestic and rest-of-world stocks, population, inflation, income, and prices of other inputs used in the production of textile products at the mill level. Also, methods were employed to identify and measure the lags in time between research and promotion expenditures and the associated change in demand.

Rate of return estimates were calculated as the ratio of cumulative net returns over cumulative assessments. The net return is the measure of economic gain at a specified level of the marketing channel resulting from combined textile research and promotion activities. The denominator of the ratio is cumulative assessments paid over the relevant time period.

As another check on our analysis, a system of external reviews was incorporated into the project design. Three experienced analysts visited Texas A&M and reviewed our progress during the later stages of model development and initial estimation. These reviews resulted in renovations including better incorporation of international cotton market effects and an improvement in the method of calculating net returns on investment (ROI).

Impacts
Promotion and textile research and development expenditures had a positive impact on domestic cotton demand. To translate into industry terms, for example, a 10 percent increase in promotion expenditures, on average, gave rise to a nearly 2 million pound increase in domestic consumption per month, after a delay of eight months.

Textile research expenditures, interestingly, had a larger effect; a 10 percent increase, on average, translated into a 4 million pound per month increase in domestic consumption, after a delay of nine months. The textile research and development activities of CI include extensive technology transfer and technical servicing in mills. It is apparent that this set of activities paid off in terms of getting rapid adoption of new processing and fiber quality technology.

The U.S. textile industry is characterized by lags between orders and deliveries. Forward contracting is prevalent in this industry. Often distributors and retailers contract for cotton fiber 12 or more months prior to delivery. The impacts of lags on imports are similar to those on domestic consumption. After a delay of 12 months, a 10 percent increase in promotion expenditures translated into a nearly 4 million pound per month increase in farm-level demand. A 10 percent increase in textile research expenditures gave rise to a 6.4 million pound per month increase in farm-level demand after a delay of 14 months.

Return on Investment
These estimates were calculated separately for the period prior to August 1991, when the amended checkoff went into effect, and the period following that time. The ROI calculations could not be started until January 1988 because of the lag procedure incorporated in the model. ROI estimates were calculated at the mill level, at the producer level, and for importers as well.

Econometric/structural and time-series results were very close for each time period and market level estimated. At the mill level, the net return for the current program was in the range of $7.12 to $8.14 per dollar of investment. At the producer level, the net return for the current program was $3.23 to $3.49, and for importers the net return was slightly higher at $3.63 to $4.33.

Several interesting points were revealed by the analysis. First, producer-level ROI was substantial, but lower than when measured at the mill level. This finding may be explained in two ways. First, under the price support programs prior to 1996, a shift in demand may not have translated to a very significant increase in price because of the nature of the target price program. Also, farm-level demand includes exported cotton, the price of which is not as likely to have responded as much in world markets.

It is useful to note the major differences in ROI before and after the checkoff was amended. The greater ROI obtained under the current program may suggest that economies of scale exist in checkoff-funded market development activities. After 1991, a critical threshold was reached which allowed for increases in net returns.

Conclusions
The basic story here is consistent and clear. The cotton checkoff has significantly increased domestic cotton consumption, farm-level demand, and imports. Both textile research and development and promotion activities have contributed to these observed increases. These results, in the form of promotion and research elasticities, are similar to those reported in other research studies.

This demand effect has translated into a net return at the mill level of greater than $7 per dollar of assessment, and $3 to $4 of net return at the producer and importer level. The rate of return estimates were also within the range established by studies done for other commodities. Looked at another way, producers and importers made a 300 to 400 percent ROI because of checkoff assessments.

In addition, the methodological approach of using two separate estimation procedures provided evidence that the results were not just an artifact of the methods used. This robustness should provide additional comfort to producers, importers, and policy-makers that the cotton checkoff is meeting its intended objectives.

References
Capps, Jr., O., D.A. Bessler, G.C. Davis, and J.P. Nichols. “Economic Evaluation of the Cotton Checkoff Program,” Department Technical Report 97-2, Department of Agricultural Economics, Texas A&M University, College Station, Texas, 1997.

Oral Capps, Jr., David A. Bessler, George C. Davis, and John P. Nichols are professors, assistant professor, and professor, respectively, in the Department of Agricultural Economics, Texas A&M University.

Manager's Viewpoint
David Fraser
Vice President of Communications
Cotton Board

Evaluation Confirms Significant Returns to Cotton Producers and Importers

As noted in the cotton checkoff program study also appearing in this issue, funding for cotton promotion and research was originally authorized through the Cotton Research and Promotion Act of 1966 and amended in 1990. The intent of the legislation was to strengthen the competitive position of cotton against primarily man-made fibers in addition to expanding domestic and foreign markets and uses for U.S. cotton.

The Cotton Board contracts with Cotton, Incorporated (CI) to carry out the research and promotion activities authorized by the legislative Acts. Cotton promotion activities include national television advertising campaigns via television commercials, seasonal promotions, and special public relations programs. CI also assists mills, manufacturers, and retailers in marketing cotton and cotton products for domestic and international users. Textile research activities include technical processing and production support to mills, as well as product and textile development.

From all of the market research we have conducted over the last 26 years, we knew our Cotton Research and Promotion Program had grown to be an outstanding success. Cotton Board chairman, Rick Wegis, acknowledged, “Cotton had hit an all-time retail market share low of 34 percent in 1975. But now, after 16 consecutive years of growth, our retail market share has skyrocketed to 59 percent. And the 'Seal of Cotton' is the third most recognized trademark in America with 75 percent of all consumers able to identify it, even without the word 'cotton' as part of the logo test.”

Market research has been extremely positive for a number of years and producer awareness of and satisfaction with the program, as conducted by CI and administered by the Cotton Board, has reached record highs. However, there was a critical piece of research that had never been done for us: an independent evaluation to determine the actual economic return on investment received by U.S. cotton producers and importers funding the program through their assessment payments. We all needed to know if program benefits outweighed costs and if the benefits were greater, by how much? Or as Rick Wegis put it, “In undertaking this study, we wanted to not only evaluate the effectiveness and return on investment of the Cotton Research and Promotion Program in its objective to build demand for cotton, but we also wanted an independent research study that was designed in such a way that the results generated could undergo considerable scrutiny and still be successfully validated and substantiated.”

We first discussed commissioning an independent evaluation in June of 1995. By the spring of 1996, the researchers in the Agricultural Economics department at Texas A&M University had nearly completed their data collection. Ironically, the 1996 Freedom to Farm Act contained a provision that mandated all commodity organizations periodically conduct independent evaluations. Not only were we a step ahead on the new provision, but now we also have the bottom-line proof of the program’s success.

Needless to say, we are extremely pleased with the report produced by Texas A&M. Not only did it meet our criteria of validating benefits vs. costs, it also held up to an intensive peer review and featured a unique methodology using two different analytical approaches that achieved identical results. As Dr. John Nichols and Dr. Oral Capps, Jr. of Texas A&M concluded in the report, “Unequivocally, promotion and research expenditures, adjusted for inflation and seasonality, do significantly stimulate domestic consumption, farm-level demand, and imports of cotton.”

Director's Corner
Harry M. Kaiser

The annual meeting of the NICPRE Advisory and Steering Committees will be in September. One major agenda item will be to establish research directions for the upcoming year. I welcome and encourage any comments, suggestions, or research ideas within the realm of commodity promotion economics from anyone receiving this newsletter. Indeed, research is a two-way street and those of us in this field seriously consider feedback by people in industry and government when selecting research topics.

This fall marks the third year that NICPRE has been in existence. I believe that NICPRE has been very productive in conducting and sponsoring research over this time, resulting in an improved understanding of the economic impacts of commodity promotion and advertising. Since 1994, there have been 30 research projects initiated by NICPRE, 62 papers published in alternative outlets, a major database on generic and branded advertising expenditures for all food developed, seven NEC-63 conferences held, and several NICPRE-organized meetings with industry personnel to tackle current economic challenges and opportunities in commodity promotion. Much of the research undertaken by NICPRE has been summarized succinctly in various NICPRE Quarterly articles.

NICPRE has devoted 35-40 percent of its budget to support research at other universities in the United States. This “leveraging” of research dollars has been effective in increasing the amount of research performed and has been quite cost-effective, since many researchers have combined other sources of support with NICPRE to complete the research. The commodities covered by these research projects include milk, cheese, butter, frozen dairy products, beef, cotton, eggs, almonds, wool, and table grapes. While not exhaustive, this has been a fairly large number of commodities studied in a fairly short period of time. In the future, we hope to expand the list of commodities studied.

In closing, I would like to again encourage you to send us any suggestions for future research topics. Our ears are always open.

NEC-63 FALL 1997 MEETING

The Research Committee on Commodity Promotion (NEC-63) will sponsor a symposium to examine the effectiveness of different nutrition education programs and messages in changing consumers' nutrition knowledge, awareness of diet-health linkages, and attitudes about healthy eating. The meeting will be held in Washington, DC, on October 6 and 7, 1997. If you need more information, please contact Jim Blaylock, ERS, (202) 219-0900, jblayloc@econ.ag.gov or Karen Ackerman, ERS, (202) 501-8511, ackerman@econ.ag.gov.