For algae, the EC50 values are much larger for spinosad and pyrethrins than copper hydroxide , which is the most heavily applied AI in the fixed copper group . The EC50 toxicity value of sulfur is 0.06 and it is known to be an artifact due to the fact that sulfur is almost completely insoluble in water. So the impact of sulfur use on surface water is minimal. The PURE soil index value is estimated based specifically on the toxicity for earthworms, which is measured by the dose that is lethal to 50% of the test population . Similar to the EC50, the four AIs/AI groups impacted by the organic acreage per farm have different levels of toxicity. Sulfur and Copper hydroxide are less toxic to earthworms, with LC50 values of 2,000 and 677, respectively; spinosad and pyrethrins are moderately toxic to earthworm, with LC50 values of 458 and 24, respectively . So larger operations have less impact on soil by using sulfur and copper products more frequently. The PURE index value for air is determined by the level of VOC emissions. Sulfur products have zero VOC emissions as they do not sublimate or evaporate at ambient temperatures. The use of sulfur products reduces VOC emissions, which in turn leads to a decrease in the PURE air index value. For groundwater, bluleberry packing boxes the average index value is almost zero and there is not enough variation across fields to identify a significant impact of the acreage or experience variables.

The impact on organic acreage on pollinators is significant but with a smaller magnitude for a similar reason.This essay identified organic fields from the PUR database using historical pesticide use records, analyzed pesticide use and associated environmental impacts in organic crop production, established the consolidation of organic cropland, and quantified the effect of acreage expansion on pesticide use and its environmental impacts. Our approach provides the basis for future studies to use the PUR database for the analysis of many different aspects of organic agriculture in California. From a spatial perspective, organic fields in California have expanded into new production regions over the past two decades, and the growth of organic acreage for fresh fruits and vegetables has been profound. For example, organic acreage for kale exceeded conventional acreage in 2015. Organic growers’ pesticide portfolio has changed dramatically during the study period. New AIs, such as spinosad and azadirachtin, were quickly adopted once approved for organic use and the shares of usage for sulfur and fixed copper, which were widely applied in earlier years, fell accordingly. This essay applied the PURE index, based on pesticide use, to assess the environmental impacts of organic crop production. We showed that organic agriculture has greater impacts on air and soil than on surface water, groundwater, and pollinators. There have been upward trends in the index values, particularly for the air and soil, which indicates that the changes which have occurred in the portfolio have negative environmental consequences, and previous assessments might have been too optimistic in generalizing the environmental performance of organic agriculture as the sector has grown dramatically. Using the PUR database, we found that large farm operations increased their share of total organic acreage, especially after the launch of the NOP in 2001.

The NOP was designed to convey a reliable signal to consumers that distinguishes organic produce from conventional. Consumers are willing to pay more for organic products, so producers have more incentive to expand their organic production. However, not all crops have followed the consolidation trend. The number of organic farms increased as well as the organic acreage. The consolidation happens when the number of farms increased slower than the acreage. The process of cropland consolidation resulted in shifts in the pesticide portfolio, which can alter the environmental impact of organic agriculture. Namely, growers with larger organic farm size applied more sulfur and fixed copper and less spinosad and pyrethrins per acre. These four major AIs/AI groups have different ecotoxicological properties. Compared to spinosad and pyrethrins, sulfur and copper hydroxide are less toxic to earthworms and copper hydroxide is more toxic to aquatic organisms. Therefore, the pesticide practices of larger organic operations have greater impacts on surface water and less on soil. The impact on air, measured by VOC emissions, was smaller for larger operations because the sulfur products they use heavily have zero VOC emissions. Changes in the crop composition are observed in organic agriculture in California. The acreage share of vegetables increased while the share of grapes and field crops decreased. However, the shift in crop mix does not alter the result that the consolidation affects pesticide use and environmental impacts because this pattern does not change significantly across crops. Primarily due to the lack of field-level data, previous studies focused less on the variation within the organic agriculture sector. Instead, the average performance across numerous organic growers was compared with other farming systems to illustrate the benefit from organic farming practices. The recent trend of cropland consolidation into larger operations has raised the question of how large organic farms behave differently from small ones, and what impacts those differences might have. The results partially answer this question in terms of pesticide use and show that as organic cropland has increased, growers have changed their pesticide portfolios and associated environmental impacts. As observed in conventional agriculture , the consolidation of cropland is almost inevitable. Therefore, new policy tools might be necessary to address the usage of pesticides in organic agriculture. The change in farm size could also alter practices other than pest management, such as fertilizer use. Future studies are needed to deliver a comprehensive analysis on the effect of consolidation on the overall environmental performance of organic agriculture.Why do farms differ in size? Why are farms becoming more specialized? The consolidation of acreage and production has long characterized U.S agriculture . Meanwhile the number of very small farms has continued to grow over the last thirty years, in part because the definition of a farm has not been adjusted for infla-tion.1 Another trend documented by MacDonald et al. is that as production shifts to large farms, specialization also occurs. In 1996, 37% of the value of corn production came from farms that grew fewer than 3 commodities.2 This share increased to 57% by 2015. Similar patterns exist for other field crops as well as livestock. For example, 31% of the value of hog production occurred on farms without any crop harvested in 2015, up from 14% in 1996. Farm size and specialization interact. On one hand, more specialized farms need less acreage to take advantage of economies of scale. On the other hand, large farms can purchase specialized equipment or acquire specialized knowledge, which decreases the cost of producing a small number of crops. In this essay, I will focus only on crop farms and argue that changes in either size or specialization can be explained by changes in the distribution of crop-based knowledge across farmers, which I will call the knowledge distribution from now on. Recent trends in size and specialization can be explained by a model with heterogeneous farm operators whose knowledge evolves over time.3 Farm operators learn from others, package of blueberries and expand acreage of one crop, ceteris paribus, as their knowledge increases for that crop. The knowledge they learn is specialized to a certain crop production process and it cannot be perfectly transferred to produce other crops. As farmers’ specialized knowledge accumulates, the opportunity cost of planting crops that they know less about increases, which results in specialization in a few crops for which they have the largest knowledge stock when everything else is equal. In this essay, I present a multi-good model to study changes in farm size and specialization. For crop farms, size can be measured by acreage, quantity of output, value added, or revenue and in many other ways . Although revenue is a widely used measurement when a farm grows multiple crops, it is affected by other factors beside production decisions. Demand shocks and inflation could alter farm “size” without farmers changing their practices.

The same reasoning applies to quantity produced. Some inputs, such as weather, are not determined by farmers, which makes quantities an inaccurate measure of farmers’ decisions. The value of production per acre varies across crops due to differences in cost and revenue across crops. When measuring size by total acreage, farms that grow different crops are not directly comparable. Albeit imperfect, acreage is the measurement of farm size in this essay. In the model, acreage increases directly as knowledge accumulates. Farm operators have crop-based knowledge, which may increase after meeting and learning from other farmers. Knowledge and land are inputs in production, which makes land demand for each crop a function of knowledge. The more knowledge farmers have about one crop, the more acreage they will allocate to that crop. Total land supply is fixed but the acreage of a specific crop varies based on the evolution of knowledge. To simplify the model, the demand for goods is assumed to be exogenous. Equilibrium prices clear all commodity markets and the land market, thus determining the farm size for each producer given his knowledge level. In other words, the farm size distribution is a transformation of the underlying knowledge distribution, similar to the intuition in Lucas Jr . Because the total land supply is fixed, the growth of farm size is accomplished by farmers exiting the crop sector. Although knowledge does not depreciate over time by assumption, knowledge evolves for everyone, which means individuals need to keep learning in order to maintain their current acreage. This approach of modeling the evolution of knowledge is based on Lucas where agents have opportunities to increase their knowledge by learning from others. Learning comes from imitation. Agents meet randomly with others and copy their knowledge if it is better. The meeting and learning process are costless for agents, which means whether agents learn or not in each period is not correlated with their production.4 Therefore agents who currently have no production can still increase their knowledge and start producing in the future. This implication of Lucas is consistent with observations of U.S. agriculture that show people enter the agricultural sector. The Census of Agriculture collects information about how long the principal operator has operated any farm. In 2017, beginning farmers, who are defined as principal operators with no more than ten years of experience on any farm, operated 25% of total farms which accounted for 16% of total farmland and 15% of total agricultural sales . The evidence of learning from other farmers is well documented in the economic literature. Foster and Rosenzweig first separated learning by doing and learning from others in agriculture. They found that farmers with experienced neighbors earn higher profits. Using data on farmers’ communication network, Conley and Udry showed the significance of social learning in the diffusion of agricultural technology. Other studies have covered the role of learning in specific farming decisions in the U.S. ; Kroma ; Schneider et al. ; Goodhue et al.. Learning as modeled here only covers learning between farmers, which does not necessarily require farmers to meet in person. As long as the knowledge acquired by one farmer is generated by another farmer, the process can be modeled as meeting and learning between farmers. To extend this point, both public and private agencies have facilitated information diffusion among farmers in different ways. If the information they shared, both online and in print, is based on findings in a farmer’s fields, learning from this piece of information can be viewed as learning from that farmer. Examination of the knowledge generated from other sources, such as extension agents or industry dealers, is left for future work. Previous endogenous growth analyses focused on a one-good economy in which a composite good is produced and consumed . Therefore they cannot explain the trend of specialization observed in agriculture. This essay contributes to the endogenous growth literature by modeling the evolution of industry-based knowledge when there are multiple industries . In agriculture, much production and marketing knowledge is crop-based. Some crop-based knowledge, such as the management of pests, can be applied to a large group of crops. Other knowledge, such as the timing of harvest, refers to a single crop or a small number of crops. Special-ized knowledge required in farming can be very similar for plants with similar agronomic characteristics.