蔡鍵
摘要 農(nóng)民長期暴露在農(nóng)藥環(huán)境中,他們或者在使用農(nóng)藥時直接接觸農(nóng)藥,或者在沒有過安全期就直接進入農(nóng)地而接觸到農(nóng)藥殘留,由此也引發(fā)了健康風險問題。對此,一部分學者認為農(nóng)民的農(nóng)藥暴露可能是受農(nóng)民風險偏好所驅(qū)動的行為;另一部分學者認為農(nóng)民關于自身農(nóng)藥暴露的信息是非常有限的,這也正是農(nóng)民長期暴露于農(nóng)藥環(huán)境中的根本原因。而本文則是從農(nóng)戶行為角度著手,運用廣東省274個菜農(nóng)的實地調(diào)研數(shù)據(jù),通過個人保護裝備使用情況來判斷農(nóng)民的農(nóng)藥暴露程度,在此基礎上構建有序回歸模型,計量分析了風險偏好、外部信息失效和個人特征對農(nóng)藥暴露行為的影響作用,并對農(nóng)民農(nóng)藥信息失效的原因進行了研究。研究發(fā)現(xiàn):目前廣東農(nóng)民整體存有較高程度的農(nóng)藥暴露現(xiàn)象;在控制農(nóng)民的年齡、受教育程度和性別后,影響農(nóng)民農(nóng)藥暴露行為的顯著因素是“外部信息失效”,而不是“風險偏好”,即相較于農(nóng)藥信息來自自身經(jīng)驗的農(nóng)民,最有用的農(nóng)藥信息是來自外部信息的農(nóng)民在施用農(nóng)藥時穿戴全套保護裝備的概率更低;農(nóng)藥零售商為了追求利益最大化而向農(nóng)民提供不完全或者不對稱的農(nóng)藥信息,這也正是現(xiàn)階段農(nóng)民外部農(nóng)藥信息失效以及高程度農(nóng)藥暴露行為的主要原因。結論表明,為推進農(nóng)藥信息普及工作,降低農(nóng)民的農(nóng)藥暴露程度,應做到:第一,加強對農(nóng)藥零售商的培訓、管理和監(jiān)督,優(yōu)化市場環(huán)境,引導農(nóng)藥零售商向農(nóng)民提供完全、對稱、準確的農(nóng)藥信息;第二,進一步強化農(nóng)技站、植保站和農(nóng)業(yè)學校等相關部門的農(nóng)藥信息服務功能。
關鍵詞 農(nóng)藥暴露;風險偏好;農(nóng)藥信息;信息渠道
中圖分類號 F323.6 文獻標識碼 A 文章編號 1002-2104(2014)09-0135-06
農(nóng)藥是農(nóng)業(yè)中用于消滅病蟲害的主要手段,其在農(nóng)業(yè)領域的引入與使用,大大提高了農(nóng)作物產(chǎn)量和生產(chǎn)率,但不斷增加的農(nóng)藥使用也導致了有毒化學物質(zhì)的釋放,污染了環(huán)境[1-2]。農(nóng)藥已經(jīng)被認為是環(huán)境中最具毒性的傳播物之一,農(nóng)藥對許多非目標對象造成嚴重的負面影響,并引發(fā)農(nóng)民的農(nóng)藥暴露問題[3]。所謂農(nóng)藥暴露,就是農(nóng)業(yè)工人或農(nóng)民職業(yè)性的暴露到農(nóng)藥環(huán)境中,可能是使用農(nóng)藥時直接接觸農(nóng)藥,或在沒有過安全期時進入農(nóng)地而接觸到殘留在作物中的農(nóng)藥[4]。大部分文獻都認為,農(nóng)場工人和農(nóng)民的健康風險問題,主要來自于農(nóng)藥暴露與農(nóng)藥使用[5-6]。農(nóng)藥暴露對健康的影響作用可能是立即的,包括皮疹、頭痛、惡心和嘔吐、定向障礙、休克、昏迷、呼吸衰竭,甚至死亡;農(nóng)藥暴露對健康的影響也可能是長期的,如癌癥、神經(jīng)和生殖問題[7-8]。另外,還有許多不太常見的疾病都被懷疑是由農(nóng)藥暴露引起的,包括帕金森病和出生缺陷等[3]。
對此,有學者提出農(nóng)民在施用農(nóng)藥過程中可通過正確使用個人防護裝備(PPE)來降低農(nóng)藥暴露程度,然而發(fā)展中國家(包括中國)的農(nóng)民卻較少使用保護裝備,用手直接施用農(nóng)藥是常見的現(xiàn)象[9-10]。那么,究竟是什么原因?qū)е铝宿r(nóng)民的農(nóng)藥暴露行為?大量的研究表明,農(nóng)民的風險偏好在他們的農(nóng)業(yè)生產(chǎn)行為中發(fā)揮著巨大的作用[11]。有學者提出農(nóng)民的風險厭惡程度與采納抗蟲棉的決策密切相關[12];也有學者認為農(nóng)民的風險偏好顯著影響他們的信貸決策行為[13-14];另外還有學者提出了農(nóng)藥施用行為與農(nóng)民風險偏好有關[15]??梢?,農(nóng)民的風險偏好與態(tài)度對于其追求利益最大化的行為決策具有重要的影響作用[12,16-17],發(fā)展中國家農(nóng)民的農(nóng)藥暴露可能是受農(nóng)民風險偏好所驅(qū)動的行為。
另外,農(nóng)民一般是自我雇傭的,他們關于自身農(nóng)藥暴露的信息非常有限[18],而信息是農(nóng)民決策過程中減小不確定性的主要因素[19]。因而,農(nóng)民的信息有效性與信息渠道,也是影響農(nóng)民應對風險、做出決策的主要因素。有學者提出農(nóng)民過量施用農(nóng)藥可能與從出售農(nóng)藥的農(nóng)業(yè)技術推廣部門得到的信息有關[15];也有學者認為干旱信息的有效性顯著影響農(nóng)民關于保險購買計劃的決策[19]。由此可見,農(nóng)民的決策行為,除了受到自身風險偏好的影響,還可能受到外部信息的影響。
那么,中國農(nóng)民的農(nóng)藥暴露行為究竟如何?農(nóng)民的農(nóng)藥暴露行為到底是因為自身的風險偏好所形成的決策,還是由于外部信息失效所導致的,抑或是兩者共同的結果?
2.2.1 農(nóng)民風險偏好農(nóng)民的風險偏好或者說風險態(tài)度是農(nóng)民的主觀意愿,難以直接衡量,對此本研究將用“農(nóng)民是否購買保險”作為代理變量。之所以使用該指標作為代理變量,原因有二:一是農(nóng)民是否購買保險能夠客觀衡量;二是購買保險是農(nóng)民厭惡風險的表現(xiàn),即相較于沒有購買保險的農(nóng)民,購買保險的農(nóng)民更加厭惡風險。盡管“農(nóng)民是否購買保險”并不是農(nóng)民風險偏好的直接衡量,但使用該指標能較為明確的將農(nóng)民劃分為風險厭惡程度較高(購買保險)和較低(不購買保險)的兩個群體,而通過研究這兩個群體農(nóng)藥暴露行為的差異性,則可間接判斷出風險偏好是否顯著影響農(nóng)藥暴露行為。
2.2.2 農(nóng)藥信息有效性如果由農(nóng)民自己判斷農(nóng)藥信息的有效性,則可能出現(xiàn)判斷錯誤或者過度判斷等主觀臆斷現(xiàn)象,對此本研究將用“最有用的農(nóng)藥信息是否來自自身經(jīng)驗以外的其他信息”作為代理變量。如果來自外界的信息對農(nóng)藥暴露程度有正向影響作用,則說明外界農(nóng)藥信息失效,降低了農(nóng)民對農(nóng)藥暴露的正確認識,反之則可認為外界農(nóng)藥信息有助于減弱農(nóng)藥暴露行為。
2.2.3 控制變量
由男女農(nóng)民的農(nóng)藥暴露程度差異可知,農(nóng)民的個性特征可能也會對農(nóng)藥暴露行為產(chǎn)生一定的影響作用,因而本研究將性別、年齡和受教育程度等農(nóng)民個性特征因素列為控制變量。
2.3.1 解釋變量回歸結果詮釋
(1)變量“外部信息”通過5%的顯著性檢驗,回歸系數(shù)為-0.868,由關系式OR=eβ可知,該因素的比例優(yōu)勢系數(shù)小于1。表明相較于農(nóng)藥信息來自自身經(jīng)驗的農(nóng)民,最有用的農(nóng)藥信息是來自外部信息的農(nóng)民在施用農(nóng)藥時穿戴全套保護裝備的概率更低。結合本研究代理變量的內(nèi)涵,該回歸結果的含義是:依賴外部信息的農(nóng)民具有高程度的農(nóng)藥暴露現(xiàn)象的概率高于依賴自我經(jīng)驗的農(nóng)民。由此可以判斷,農(nóng)民在農(nóng)藥方面的外部信息失效,從而導致了部分農(nóng)民的高程度農(nóng)藥暴露行為。
(2)變量“是否購買保險”沒有通過顯著性檢驗,表明購買保險與沒有購買保險的農(nóng)民,在施用農(nóng)藥時個人保護裝備的使用情況并不存在顯著的差異。結合本研究代理變量的內(nèi)涵,該結果的含義是:風險厭惡程度的高低并不是影響農(nóng)民農(nóng)藥暴露行為的顯著因素。其原因可能是:對于農(nóng)民而言,他們并不將農(nóng)藥暴露視為一種風險行為,因而他們選擇該行為時并不受自身風險偏好的影響。這也從側面驗證了外部信息失效是農(nóng)藥暴露行為的主要原因,由于外部信息失效,導致農(nóng)民低估農(nóng)藥暴露的危害作用,從而將其視為一種非風險或者低風險行為,最終也導致農(nóng)民的風險偏好不對農(nóng)藥暴露行為產(chǎn)生顯著影響作用。
2.3.2 控制變量回歸結果詮釋
(1)“年齡”通過1%的顯著性檢驗,回歸系數(shù)為-0.064,由關系式OR=eβ可知,該因素的比例優(yōu)勢系數(shù)小于1。表明年齡越大的農(nóng)民,越可能在施用農(nóng)藥時沒有穿戴任何保護裝備。
(2)“受教育程度”中,僅有“受教育程度為小學”的樣本通過10%的顯著性檢驗,表明與受教育程度為高中或中專的農(nóng)民相比,受教育程度為小學的農(nóng)民施用農(nóng)藥時穿戴全套裝備的概率增大。
(3)“性別”沒有通過顯著性檢驗,表明性別并非影響農(nóng)民農(nóng)藥暴露行為的顯著因素。
綜上所述,在控制了農(nóng)民的年齡、受教育程度和性別后,農(nóng)藥暴露行為是由外部信息失效所引起的,而并非農(nóng)民自身的風險偏好所導致的。
3 外部信息失效原因分析
由前文分析可知,農(nóng)民嚴重的農(nóng)藥暴露現(xiàn)象,并非農(nóng)民風險偏好所引起,而是外部信息失效所導致。那么,農(nóng)民的外部信息主要來自于哪些渠道,這些信息失效的原因又是什么?
3.1 信息渠道
如表5所示,對于農(nóng)民而言,主要的農(nóng)藥信息渠道有四種,由高至低分別為:“農(nóng)藥零售店”、“農(nóng)民自己的經(jīng)驗”、“其他農(nóng)民或鄰居的推薦”、“農(nóng)技站、植保站或農(nóng)業(yè)學?!?。其中,選擇“農(nóng)藥零售店”為主要信息渠道的農(nóng)民有157人,占比高達92.9%;選擇“自己的經(jīng)驗”為主要信息渠道的農(nóng)民有137人,占比為81.07%;選擇“其他農(nóng)民或鄰居的推薦”為主要信息渠道的農(nóng)民有96人,占比為56.80%;選擇“農(nóng)技站、植保站或農(nóng)業(yè)學?!睘橹饕畔⑶赖霓r(nóng)民有28人,占比為16.57%。而將另外十一種渠道視為主要農(nóng)藥信息來源的農(nóng)民都不足10%。
3.2 信息失效原因
如前文所述,農(nóng)民外部農(nóng)藥信息存在失效的現(xiàn)象,進而導致了農(nóng)民的農(nóng)藥暴露行為。而由農(nóng)民信息渠道分析可知,“農(nóng)藥零售店”是農(nóng)民最主要和最有用的信息渠道,超過90%的農(nóng)民將“農(nóng)藥零售店”視為主要的農(nóng)藥信息渠道,接近50%的農(nóng)民將“農(nóng)藥零售店”選為最有用的信息渠道。由此可見,“農(nóng)藥零售店”在銷售農(nóng)藥過程中也為農(nóng)民提供了一定的農(nóng)藥信息,但他們提供的并非完全或者對稱的有效信息。本研究認為,之所以農(nóng)藥零售店會提供不完全或者不對稱的信息,主要原因在于:第一,農(nóng)藥零售店是追求利益最大化的經(jīng)濟主體,他們在銷售農(nóng)藥過程中會過分強調(diào)農(nóng)藥的優(yōu)點及正面作用(如殺蟲效果、低毒性等),并故意弱化農(nóng)藥的負面影響(如對環(huán)境的污染、對農(nóng)民健康的危害等),從而提高農(nóng)民對農(nóng)藥的認可程度,增加農(nóng)藥銷量;第二,目前中國的農(nóng)藥零售市場環(huán)境較差、監(jiān)管力度不足,從而導致零售商對農(nóng)藥缺乏全面的認知,進而影響他們提供信息的準確性和有效性。
4 結論與政策含義
基于以上分析,本研究得到如下結論。①廣東農(nóng)民整體存在嚴重的農(nóng)藥暴露現(xiàn)象。②實證分析發(fā)現(xiàn),控制了農(nóng)民的年齡、受教育程度和性別后,影響農(nóng)民農(nóng)藥暴露行為的顯著因素是“外部信息”,而不是“是否購買保險”。③“農(nóng)藥零售店”是農(nóng)民最主要和最有用的信息渠道,農(nóng)藥零售店為了提高農(nóng)民對農(nóng)藥的認可、增加農(nóng)藥銷量,向農(nóng)民提供不完全或者不對稱的農(nóng)藥信息,由此也造成現(xiàn)階段農(nóng)民外部農(nóng)藥信息失效以及高程度的農(nóng)藥暴露現(xiàn)象。
基于以上結論,本研究認為可以通過如下兩個方面的工作來推進農(nóng)藥信息普及和降低農(nóng)民的農(nóng)藥暴露行為。第一,加強對農(nóng)藥零售店(商)的培訓、管理和監(jiān)督,優(yōu)化市場環(huán)境,引導農(nóng)藥零售店(商)向農(nóng)民提供完全、對稱、準確的農(nóng)藥信息。由前文分析可知,農(nóng)民從農(nóng)藥零售店獲得不完全或不對稱的信息是現(xiàn)階段農(nóng)民高程度農(nóng)藥暴露行為的主要原因。因而,相關部門一方面應該加強對農(nóng)藥零售店(商)的培訓,提高他們對農(nóng)藥的認識程度;另一方面則應該優(yōu)化市場環(huán)境,加強對農(nóng)藥零售店(商)的管理與監(jiān)督,確保農(nóng)民在購買農(nóng)藥過程中能夠獲得準確、對稱、完全的農(nóng)藥信息。第二,進一步強化農(nóng)技站、植保站和農(nóng)業(yè)學校等相關部門的農(nóng)藥信息服務功能。由前文分析可知,農(nóng)民的農(nóng)藥信息渠道主要有“農(nóng)藥零售店”、“農(nóng)民自己的經(jīng)驗”、“其他農(nóng)民或鄰居的推薦”和“農(nóng)技站、植保站或農(nóng)業(yè)學?!?,其中屬于外部正式信息渠道的“農(nóng)技站、植保站或農(nóng)業(yè)學校”僅得到16.57%農(nóng)民的認可。這說明能夠客觀提供有效以及對稱的農(nóng)藥信息的農(nóng)技站、植保站和農(nóng)業(yè)學校等部門的信息服務工作并未在農(nóng)村全面開展。因而必須進一步強化農(nóng)技站、植保站和農(nóng)業(yè)學校等相關部門的農(nóng)藥信息服務功能,進而保證農(nóng)民獲得有效的農(nóng)藥信息,降低農(nóng)藥暴露行為。
(編輯:田 紅)
參考文獻(References)
[1]Jones E, Mabota A, Larson D W. Farmers Knowledge of Health Risks and Protective Gear Associated with Pesticide Use on Cotton in Mozambique[J]. The Journal of Developing Areas, 2009, 42(2): 267-282.
[2]Wilson C, Tisdell C. Why Farmers Continue to Use Pesticides Despite Environmental, Health and Sustainability Costs[J]. Ecological Economics, 2001, 39: 449-462.
[3]Rull R A, Ritz B. Historical Pesticide Exposure in California Using Pesticide Use Reports and Landuse Surveys: An Assessment of Misclassification Error and Bias[J]. Environmental Health Perspectives, 2003, 111(13): 1582-1589.
[4]BlancoMuoz J, Lacasaa M. Practices in Pesticide Handling and the Use of Personal Protective Equipment in Mexican Agricultural Workers [J]. Journal of Agromedicine, 2011, 16: 117-126.
[5]Parrott R, Wilson K, Buttram C, et al. Migrant Farm Workers Access to Pesticide Protection and Information: Cultivando Buenos Habitos Campaign Development[J]. Journal of Health Communication, 1999, 4(1): 49-64.
[6]Szmedra P. The Health Impacts of Pesticide Use on Sugarcane Farmers in Fiji[J]. AsiaPacific Journal of Public Health, 1999, 11(2): 82-88.
[7]Arcury T A, Quandt S A, Russell G B. Pesticide Safety Among Farmworkers: Perceived Risk and Perceived Control as Factors Reflecting Environmental Justice[J]. Environmental Health Perspectives, 2002, 110(2): 233-240.
[8]Tilson H A. Developmental Neurotoxicology of Endocrine Disruptors and Pesticides: Identification of Information Gaps and Research Needs[J]. Environmental Health Perspectives, 1998, 1063: 807-811.
[9]Coffman C W, Stone J F, Slocum A C, et al. Use of Engineering Controls and Personal Protective Equipment by Certified Pesticide Applicators. [J]. Journal of Agricultural Safety and Health, 2009, 15(4): 311-326.
[10]Murphy H H, Sanusi A, Dilts R, et al. Health Effects of Pesticide Use Among Indonesian Women Farmers[J]. Journal of Agromedicine, 1999, 6(3): 61-85.
[11]Liu E M, Huang J K. Risk Preferences and Pesticide Use by Cotton Farmers in China[J]. Journal of Development Economics, 2013, 103: 202-215.
[12]Liu E M. Time to Change What to Sow: Risk Preferences and Technology Adoption Decisions of Cotton Farmers in China[J]. Review of Economics and Statistics, 2013, 95(4): 1386-1403.
[13]王瑜,應瑞瑤. 契約選擇和生產(chǎn)者質(zhì)量控制行為研究——基于農(nóng)戶風險偏好視角[J]. 經(jīng)濟問題,2007,(9): 85-87. [Wang Yu, Ying Ruiyao. Researchs on Food Safety, Contract Choice and Producers Quality Control Behavior:Based on Farmers View [J]. On Economic Problem, 2007,(9): 85-87.]
[14]許承明,張建軍. 社會資本、異質(zhì)性風險偏好影響農(nóng)戶信貸與保險互聯(lián)選擇研究[J]. 財貿(mào)經(jīng)濟,2012,(12): 63-70. [Xu Chengming, Zhang Jianjun. Impact of Social Capital and Heterogeneity Risk Preferences on Farmers InterLinked Credit and Insurance Choice [J]. Finance and Trade Economics, 2012,(12): 63-70.]
[15]黃季焜,齊亮,陳瑞劍. 技術信息知識、風險偏好與農(nóng)民施用農(nóng)藥[J]. 管理世界,2008,(5): 71-76. [Huang Jikun, Qi Liang, Chen Ruijian. Technical Information Knowledge, Risk Appetite and Farmers Use of Pesticides [J]. Management World, 2008,(5): 71-76.]
[16]Humphrey S J, Verschoor A. Decisionmaking Under Risk Among Small Farmers in East Uganda[J]. Journal of African Economies, 2004, 13(1): 44-101.
[17]Bontems P, Thomas A. Regulating Nitrogen Pollution with Risk Averse Farmers Under Hidden Information and Moral Hazard[J]. American Journal of Agricultural Economics, 2006, 88(1): 57-72.
[18]Hoppin J A, Yucel F, Dosemeci M, et al. Accuracy of Selfreported Pesticide Use Duration Information from Licensed Pesticide Applicators in the Agricultural Health Study[J]. Journal of Exposure Analysis and Environmental Epidemiology, 2002, 12(5): 313-318.
[19]Quiroga S, Garrote L, FernandezHaddad Z, et al. Valuing Drought Information for Irrigation Farmers: Potential Development of a Hydrological Risk Insurance in Spain[J]. Spanish Journal of Agricultural Research, 2011, 9(4): 1059-1075.
Abstract Farmers often expose to the environment with pesticides, sometimes they directly contact with the pesticide when using it, sometimes they directly get into farmland and exposure to pesticides when there is no safety period. As a result, they get some problems of health hazard. Some scholars think that farmers behavior of pesticide exposure is likely to be driven by farmers risk preference. While other scholars think that the reason why farmers exposure to pesticide is their limited information about pesticide exposure. However, this paper is from the perspective of peasant household behavior, using the field survey data of 274 farmers in Guangdong Province to determine the level of farmers pesticide exposure by how they used personal protective equipment. Based on this, it builds orderly regression model, carries out econometric analysis on how risk preference, external information failure and personal characteristics affect the behavior of pesticide exposure. And then it studies the reasons of farmers information failure. This paper finds that farmers have high degree of pesticide exposure at present in Guangdong. The significant factor which affects farmers behavior of pesticide exposure is ‘external information failure, but not ‘risk appetite when controlling farmers age, level of education and gender. This means that compared with ones whose pesticide information is from their own experience, the farmers whose most useful pesticide information is from external information would less likely wear full set of protection equipment. Pesticide retailers will provide farmers with incomplete or asymmetry pesticide information for getting maximum interest, which is the main reason of farmers external pesticide information failure and high level of pesticide exposure. Results indicate that there are two aspects of work we should do to promote the pesticide information popularization and reduce farmers pesticide exposure. Firstly, we should strengthen the training, management and supervision of pesticide retailers, optimize the environment of market, and guide the pesticide retailers to provide farmers with completely, symmetry and accurate information of pesticides. Secondly, we also should strengthen the pesticide information service function from agricultural technology station, plant protection station, agricultural schools and other departments.
Key words pesticide exposure; risk preference; pesticide information; sources of information
[16]Humphrey S J, Verschoor A. Decisionmaking Under Risk Among Small Farmers in East Uganda[J]. Journal of African Economies, 2004, 13(1): 44-101.
[17]Bontems P, Thomas A. Regulating Nitrogen Pollution with Risk Averse Farmers Under Hidden Information and Moral Hazard[J]. American Journal of Agricultural Economics, 2006, 88(1): 57-72.
[18]Hoppin J A, Yucel F, Dosemeci M, et al. Accuracy of Selfreported Pesticide Use Duration Information from Licensed Pesticide Applicators in the Agricultural Health Study[J]. Journal of Exposure Analysis and Environmental Epidemiology, 2002, 12(5): 313-318.
[19]Quiroga S, Garrote L, FernandezHaddad Z, et al. Valuing Drought Information for Irrigation Farmers: Potential Development of a Hydrological Risk Insurance in Spain[J]. Spanish Journal of Agricultural Research, 2011, 9(4): 1059-1075.
Abstract Farmers often expose to the environment with pesticides, sometimes they directly contact with the pesticide when using it, sometimes they directly get into farmland and exposure to pesticides when there is no safety period. As a result, they get some problems of health hazard. Some scholars think that farmers behavior of pesticide exposure is likely to be driven by farmers risk preference. While other scholars think that the reason why farmers exposure to pesticide is their limited information about pesticide exposure. However, this paper is from the perspective of peasant household behavior, using the field survey data of 274 farmers in Guangdong Province to determine the level of farmers pesticide exposure by how they used personal protective equipment. Based on this, it builds orderly regression model, carries out econometric analysis on how risk preference, external information failure and personal characteristics affect the behavior of pesticide exposure. And then it studies the reasons of farmers information failure. This paper finds that farmers have high degree of pesticide exposure at present in Guangdong. The significant factor which affects farmers behavior of pesticide exposure is ‘external information failure, but not ‘risk appetite when controlling farmers age, level of education and gender. This means that compared with ones whose pesticide information is from their own experience, the farmers whose most useful pesticide information is from external information would less likely wear full set of protection equipment. Pesticide retailers will provide farmers with incomplete or asymmetry pesticide information for getting maximum interest, which is the main reason of farmers external pesticide information failure and high level of pesticide exposure. Results indicate that there are two aspects of work we should do to promote the pesticide information popularization and reduce farmers pesticide exposure. Firstly, we should strengthen the training, management and supervision of pesticide retailers, optimize the environment of market, and guide the pesticide retailers to provide farmers with completely, symmetry and accurate information of pesticides. Secondly, we also should strengthen the pesticide information service function from agricultural technology station, plant protection station, agricultural schools and other departments.
Key words pesticide exposure; risk preference; pesticide information; sources of information
[16]Humphrey S J, Verschoor A. Decisionmaking Under Risk Among Small Farmers in East Uganda[J]. Journal of African Economies, 2004, 13(1): 44-101.
[17]Bontems P, Thomas A. Regulating Nitrogen Pollution with Risk Averse Farmers Under Hidden Information and Moral Hazard[J]. American Journal of Agricultural Economics, 2006, 88(1): 57-72.
[18]Hoppin J A, Yucel F, Dosemeci M, et al. Accuracy of Selfreported Pesticide Use Duration Information from Licensed Pesticide Applicators in the Agricultural Health Study[J]. Journal of Exposure Analysis and Environmental Epidemiology, 2002, 12(5): 313-318.
[19]Quiroga S, Garrote L, FernandezHaddad Z, et al. Valuing Drought Information for Irrigation Farmers: Potential Development of a Hydrological Risk Insurance in Spain[J]. Spanish Journal of Agricultural Research, 2011, 9(4): 1059-1075.
Abstract Farmers often expose to the environment with pesticides, sometimes they directly contact with the pesticide when using it, sometimes they directly get into farmland and exposure to pesticides when there is no safety period. As a result, they get some problems of health hazard. Some scholars think that farmers behavior of pesticide exposure is likely to be driven by farmers risk preference. While other scholars think that the reason why farmers exposure to pesticide is their limited information about pesticide exposure. However, this paper is from the perspective of peasant household behavior, using the field survey data of 274 farmers in Guangdong Province to determine the level of farmers pesticide exposure by how they used personal protective equipment. Based on this, it builds orderly regression model, carries out econometric analysis on how risk preference, external information failure and personal characteristics affect the behavior of pesticide exposure. And then it studies the reasons of farmers information failure. This paper finds that farmers have high degree of pesticide exposure at present in Guangdong. The significant factor which affects farmers behavior of pesticide exposure is ‘external information failure, but not ‘risk appetite when controlling farmers age, level of education and gender. This means that compared with ones whose pesticide information is from their own experience, the farmers whose most useful pesticide information is from external information would less likely wear full set of protection equipment. Pesticide retailers will provide farmers with incomplete or asymmetry pesticide information for getting maximum interest, which is the main reason of farmers external pesticide information failure and high level of pesticide exposure. Results indicate that there are two aspects of work we should do to promote the pesticide information popularization and reduce farmers pesticide exposure. Firstly, we should strengthen the training, management and supervision of pesticide retailers, optimize the environment of market, and guide the pesticide retailers to provide farmers with completely, symmetry and accurate information of pesticides. Secondly, we also should strengthen the pesticide information service function from agricultural technology station, plant protection station, agricultural schools and other departments.
Key words pesticide exposure; risk preference; pesticide information; sources of information