The AlAE serves as a major outlet for all professional groups except industry


Although academic researchers rely heavily on basic economic journals, most professional groups make substantial use of them. Academic researchers also rely substantially on lay interchange-almost as much as any other group except extension. Except for academic researchers, Choices is the second most important input medium. The results for the categories of personal experience and discussion with colleagues, especially in comparison to the results reported in tables 2 and 4, suggest that the profession is not making the best use of its resources. There may be too much fOffilalism in the profession as well. The results for sources of repons and forecasts are basically equivalent to those obtained for sources of conceptual thinking with a few minor exceptions, e.g., trade journals become more important for teaching. With respect to the outlets for completed analysis, the biggest surprises are the importance of trade journals as an outlet for academic research and basic economic journals as an outlet for industry and government. The latter outcome may reflect desire rather than actual experience. The professional AAEA and ASSA meetings are a major outlet only for academic research. This is in sharp contrast to the extent that most groups rely on professional meetings as an input. In the case of extension, results are reported to colleagues and lay individuals as expected.The results in table 5 for Choices reject the view that it is not a medium for academics. Most articles are prepared by academic researchers who are simply altering their communications style for this particular medium. This, of course,greenhouse vertical farming suggests that the AAEA can influence the type of research products generated by the profession.

For example, if the profession decided that the case study or problem definition approaches need greater emphasis, the experience of Clwices suggests that this can be acllieved by the media policies that the AAEA implements for its products. Additional results were also -developed to examine changes in professional linkages that occur with professional maturity . For sources of conceptual thinking, the most striking results here were that almost all professional groups increase theu- reliance on trade journals and AAEA and ASSA meetings as they become more mature. On the other hand, almost all professional groups decrease their reliance on basic academic journals and on lay interchange. The latter results are highly significant and a sad indictment of the profession. There is also a tendency to replace reliance on the AlAE with reliance on regional agricultural economics journals and activities, especially for extension, industry, and government groups. For completed analyses outlets, all groups reduce their publication rate in basic economic journals and increase publication in the AlAE. With few exceptions, most groups reduce discussion with colleagues and increase their presentations to lay groups. The latter exceptions were, however, mostly insignificant. With professional maturity, the publication rate in trade journals increases for all groups except academic research but remarkably so for teaching. Most of these resul ts are not surprising.In addition to the quantitative survey results, some additional results were also generated [or the AlAE, Choices, and the AAEA meetings. For each of these media, the respondents were asked to list problems they thought should be but are not addressed. In general, the qualitative responses to problems that should be addressed support the view that the profession has become too technique oriented, too solution rich, and too risk averse in analyzing possible future scenarios. Moreover, there is too little problem solving knowledge generation for which there is value added, and there are a host of specific issues for which problems have not been well articulated.

The common thread that runs through many of these responses is that there are too few conceptual and empirical pieces that address important problems that exist currently or may emerge at some future date. Instead, most of the conceptual and empirical pieces focus on some construct in the literature or are dictated by the standardized solution frameworks that have been previously developed. The call seems to be for more creative, unstructured publications that can be the basis for valuable professional exchange. Many of the responses cover a broad range of concerns that focus on opportunities for innovations in institutional design and collective group policy analysis. The gap between what is currently published in the AlAE and what would bes~ serve the membership is obviously not only due to the policies that are implemented by the editor, the Association or the peer review process. As one thoughtful respondent argued: “After over thirty years of observing the academic process it appears that most scholarly societies have become agents to establish professional credentials for tenure, promotion or a job offer. This is probably as much the fault of the administrators looking for someone else to make their decisions as anything.” Other respondents suggested that the way of dealing with this problem is to revise the academic reward system so as to encourage more problem solving and applied analysis. Positive rewards should be given for well articulated problems and useful results and insights with penalties imposed for just another technical, standardized application. The institutional changes that are required for such a reward and penalty policy structure to naturally emerge is itself a serious area of social science inquiry. Turning to Choices, the qualitative responses are overwhelmingly favorable. Among tile vast array of favorable comments, however, there are some constJ:.lctive suggestions. Since tile subject matter and problem solving knowledge of the profession is multidisciplinary, Choices should expand its disciplinary base beyond agricultural economics and political science. The articulation of important problems has been one of the most positive features of Choices, but too much space is given to personal opinions without any supporting analysis or empirical justification.

Choices does not devote sufficient space to the large payoff areas of analysis, namely, the design of new institutions or the reform of existing institutions. More lay articulation of market failures as well as government failures would dramatically improve the societal contributions of the magazine. For the AAEA meetings, both summer and winter, the same desires that emerge for the AlAE and Choices appear once again. However, there is less dissatisfaction with the professional meetings than with the AlAE . The responses suggest that the membership would prefer more sessions on feedback from users of economic analysis conducted by members I’ of the profession. This could help structure and focus future analysis where the largest payoffs might exist. More visionary sessions requiring ex ante analysis are desired. Specifically, what major problems are likely to emerge down the road that will require fundamental economic analysis?The anecdotal evidence as well as both the quantitative and qualitative responses to the survey presented in this paper imply that the product mix of the AAEA does not sufficiently emphasize problem definition, case studies, and heuristic application of economic principles based on understanding and experience. Note that changing the product mix in these directions does not imply lowering the quality standard imposed by the peer review process but instead expanding the scope of such standards. In the existing portfolio, relatively too much emphasis has been placed on ex post analysis of historical secondary data using formal frameworks and on expression of individual viewpoints . Criteria used in the selection of products,vertical agriculture which is the collective responsibility of the Boa.rd of Directors and both the editors and reviewers of AlAE and Choices, have imposed limitations on the profession which has reduced its ability to tackle forward-looking problems and fostcrinstitutional innovations. Some of the self-imposed limitations include insistence on historical data analysis and “falsification”; imposition of a false sense of objectivity, limitation of technology for empirical research, the emphasis on linear logic, and the presumption that economic understanding is a convergent process. Contrary to many claims that extension, teaching, or industry components are not well related to the profession, the results show that the profession is highly interconnected through its various media channels. However, the AAEA has a serious problem of balance between inputs and outputs. All major groups in the profession -rely on the AAEA and ASSA meetings as a major input in their thinking but no major groups regard the meetings as an important output for their work .

The same statement applies to Choices as well. In contrast, almost all major professional groups place high emphasis on output in the AlAE but almost no group relies on it as an important input in their thinking . The experiment with Choices has demonstrated that the AAEA Board plays a strong role in influencing the product mix of the profession. This fact together with the results in this paper suggest that the AAEA Board should take action to encourage more forward-looking problem definition and heuristic application of economic principles to problems for which adequate data have not yet been generated-not in lieu of the types of products now produced but as an enhancement of the product mix. Changes are needed that will balance the inputs and outputs of the profession by placing higher rewards on those outputs that–have the largest impact and reducing transaction costs incurred in accessing the best information the profession has to offer. Some possibilities include introducing a submission and refereeing process for invited papers at meetings , adding a session on forward-looking problem definition , and broadening the scope of analysis in the AlAE by adding sections for brief, highly readable papers on problem definition and heuristic application of economic principles.Whether Proposition 37 complies with the WTO Agreements requires determining against which WTO Agreement—the Agreement on the Application of Sanitary and Phytosanitary Measures or the Agreement on Technical Barriers to Trade —Proposition 37 must be measured. Proposition 37 must first be classified either as a sanitary and phytosanitary measure or as a technical barrier to trade measure. Once classified, either the SPS Agreement or the TBT Agreement, and it alone, serves as the legal standard by which to evaluate Proposition 37. SPS Agreement Annex A defines sanitary and phytosanitary measures as “all relevant laws, decrees, regulations, requirements and procedures including, inter alia, … packaging and labelling requirements directly related to food safety.” From this Annex A definition, California’s Proposition 37 is a SPS measure if it is “labelling requirements directly related to food safety.” Evidence that Proposition 37 is a label directly related to food safety comes from two sources—its language and its electoral promotion. In its language, Proposition 37 proclaims in five of the eleven paragraphs of Section 1 that its proponents support it because of concerns about adverse health. In addition, if adopted at the November 2012 election, Proposition 37 states that its provisions become part of the California Health and Safety Code. In the documents and articles promoting Proposition 37, proponents regularly proclaim that the voters should support Proposition 37 because Californians are at great risk for their health and safety against which risks labels would provide them protection. At face value from its language and its supporters’ statements, Proposition 37 easily can be classified as a labeling requirement directly about food safety and, therefore, as a SPS measure.SPS Agreement Article 2 states, in paragraph 2.1, that “Members have the right to take sanitary and phytosanitary measures necessary for the protection of human, animal or plant life or health, provided that such measures are not inconsistent with the provisions of this [SPS] Agreement.” Paragraph 2.2 provides that Members can adopt SPS measures “… only to the extent necessary to protect human, animal or plant life or health, is based on scientific principles and is not maintained without sufficient scientific evidence, …” Construing paragraph 2.1 together with paragraph 2.2 means that a SPS measure is not compliant with the SPS Agreement if the measure is not necessary and if the measure fails to be based upon and maintained upon sufficient scientific evidence. If the SPS measure fails the standard set forth in paragraph 2.2, the SPS measure is per se a violation of the SPS Agreement. Proponents of Proposition 37 face a difficult, if not impossible, task of meeting the burden of providing scientific evidence to support it as a SPS measure under Paragraph 2.2. Regulatory agencies around the world have granted regulatory approval to genetically engineered crops, from which the raw agricultural products and processed food ingredients come, after specifically evaluating human, animal, and plant health and safety.