盛濟川 曹杰 周慧
摘要 本文通過建立一個簡單支付模型研究了完全信息和不完全信息情景下經(jīng)濟目標(biāo)、環(huán)境目標(biāo)以及福利目標(biāo)對于REDD+機制收益分配的影響。按照政策制定者知道的代理人機會成本的信息,本文設(shè)定了完全信息和不完全信息兩種情景。政策制定者在兩種情景中對于總毀林和潛在造林面積的分布、代理人總收益以及各代理人的毀林或潛在造林面積都擁有完全信息。為了研究不同的政策目標(biāo)對REDD+機制效果的影響,本文設(shè)定三種政策目標(biāo):經(jīng)濟目標(biāo)、環(huán)境目標(biāo)以及福利目標(biāo)。在此基礎(chǔ)上,利用云南生態(tài)固碳造林項目的入戶調(diào)查數(shù)據(jù),對三種政策目標(biāo)的效果進行了仿真研究。通過仿真研究分析了在完全信息和不完全信息條件下三種政策目標(biāo)對于代理人受益、政策制定者收益以及減少毀林或增加造林總面積的影響。研究結(jié)果表明,在不完全信息情景下,政策制定者只能按照相同的補償標(biāo)準(zhǔn)支付給所有代理人,因而三種政策目標(biāo)的產(chǎn)出完全相同。對于經(jīng)濟目標(biāo)的政策制定者而言,完全信息并不會帶來森林面積的增加,但會導(dǎo)致REDD+剩余從代理人轉(zhuǎn)移至政策制定者。相反,對于環(huán)境目標(biāo)政策制定者而言,完全信息會導(dǎo)致森林面積增加而減少代理人的收益。對于福利目標(biāo)政策制定者,完全信息并不會導(dǎo)致總體福利有所差別,且收益仍歸代理人所有,而減少毀林或增加造林的面積大于等于不完全信息。
關(guān)鍵詞 REDD+機制;完全信息;不完全信息;收益分配;政策目標(biāo)
中圖分類號 X196
文獻標(biāo)識碼 A
文章編號 1002-2104(2014)09-0037-08
巴厘島路線圖將“REDD+機制”定義為“采取各種政策方法和積極的激勵措施,以幫助發(fā)展中國家減少砍伐和森林退化,同時還包括森林保護、森林的可持續(xù)經(jīng)營以及增加森林碳匯。 森林對CO2的吸收是碳捕捉和碳儲存的一種重要途徑[1],森林面積大約占全球陸地面積的15%[2],卻儲存了陸地生物圈約25%的碳[3]。因森林砍伐和森林退化所導(dǎo)致的溫室氣體排放目前已成為全球變暖的第二大主因,其總量已占到由人為因素導(dǎo)致碳排放總量的12-20%[4-5]。因而聯(lián)合國氣候變化框架公約(UNFCCC)在2007年提出了旨在減少森林砍伐和退化的REDD+機制,目前REDD+機制已成為最經(jīng)濟的氣候變化減緩措施之一[6]。
在REDD+機制中,一個重要的因素是REDD+的機會成本。當(dāng)取得的收益高于減少毀林或增加造林的機會成本時,減少森林砍伐或增加森林面積便成為有利可圖的選項[7]。因而減少毀林或增加造林的機會成本信息成為REDD+機制得以有效實施的一個關(guān)鍵因素[8],政策制定者對于REDD+機制中代理人的私人機會成本信息的了解程度將直接影響到REDD+的收益分配。另一方面,在REDD+機制實施過程中政策制定者的政策目標(biāo)也是多樣的或多重的,例如在許多發(fā)展中國家環(huán)境和發(fā)展往往是政策目標(biāo)的核心[9] ,而不同的政策目標(biāo)也會對REDD+機制的效果產(chǎn)生不同的影響。為此,本文將設(shè)定三種政策目標(biāo):經(jīng)濟目標(biāo)(即政策制定者收益最大化目標(biāo))、環(huán)境目標(biāo)(即減少毀林或增加造林最大化目標(biāo))以及福利目標(biāo)(即代理人收益最大化目標(biāo)),通過建立一個簡單支付模型用以研究在完全信息和不完全信息條件下不同政策目標(biāo)對REDD+機制收益分配的影響。
1 不完全信息條件下REDD+機制的收益分配
由式(15)可以發(fā)現(xiàn),在完全信息下政策制定者并沒有獲得REDD+剩余,同樣代理人也沒有獲得這部分剩余。在環(huán)境目標(biāo)下,政策制定者將低機會成本代理人的收益用于彌補高機會成本代理人的損失,因而代理人的收益水平并未提高,仍和基線情景下的收益水平相同,并低于完全信息下的收益水平。但是無差異代理人數(shù)量mE是最多的,因而減少毀林或增加造林的面積是最大的。
2.3 福利目標(biāo)的均衡支付水平
一方面,為了實現(xiàn)福利目標(biāo),政策制定者需要最大化加入REDD+機制代理人的收益;另一方面,由于具有代理人機會成本的完全信息,政策制定者可以預(yù)算約束條件采取任意的方式分配REDD+剩余。為了簡化模型,我們假定政策制定者按照羅爾斯的最大化最小值標(biāo)準(zhǔn)[10]進行收益分配,即選擇使最后一個愿意加入REDD+機制的代理人收益最大化的分配方案。由于代理人的機會成本Yi/(Di+Ai)嚴格反映了代理人的收益排序,因此福利目標(biāo)下的代理人總收益水平可以用式(12)來表示。因此根據(jù)羅爾斯的最大化最小值標(biāo)準(zhǔn),應(yīng)按照統(tǒng)一的補償價格支付給所有的代理人,而國際碳信用價格r是政策制定者在不虧本的前提下可以提供給代理人的最高補償價格。只要提供給代理人的補償價格pi高于代理人i的機會成本,那么代理人i的收益水平仍然是好于基線情景下的收益水平。
3 基于云南生態(tài)固碳造林項目的仿真研究
3.1 項目概況
作為REDD+機制中的重要組成部分,生物固碳造林和沼氣建設(shè)對減緩氣候變化具有十分重要和不可替代的地位和作用?!霸颇鲜±梅▏_發(fā)署貸款開展生物固碳造林和沼氣建設(shè)項目”是利用法國開發(fā)署(AFD)貸款在中國實施的第一個生物固碳項目,該項目利用法國開發(fā)署貸款3 500萬歐元,將在曲靖市、西雙版納傣族自治州和普洱市等3個州市的9個縣(市)建設(shè)生物固碳林59,000.0hm2,其中人工造林45 584.7 hm2、低產(chǎn)林改造5 415.3 hm2、思茅松現(xiàn)有林培育8 000.0 hm2。項目建設(shè)期為5年,從2010至2014年,總投資為65 768.37萬元,其中法國開發(fā)署貸款3 500.00萬歐元。為研究完全信息和不完全信息條件下經(jīng)濟目標(biāo)、環(huán)境目標(biāo)和福利目標(biāo)對于REDD+機制收益分配的影響,我們對項目區(qū)域9個縣的279戶進行了問卷調(diào)查,并使用這些入戶數(shù)據(jù)數(shù)據(jù)進行仿真研究,入戶調(diào)查數(shù)據(jù)的收集持續(xù)1個月。
3.2 機會成本和補償價格的確定
面積是相同的。由于政策制定者不具有樣本戶的機會成本信息,因而按照統(tǒng)一的補償價格進行支付,政策制定者無法從REDD+機制中獲益,所有的REDD+剩余都歸樣本戶所有。②在完全信息條件下,采用經(jīng)濟目標(biāo)時,無論采用何種貼現(xiàn)率,政策制定者都無法把所有樣本戶的土地納入REDD+機制之中,此時獲得的REDD+剩余都歸政策制定者所有。當(dāng)采用福利目標(biāo)時按照統(tǒng)一的補償價格支付給所有樣本戶,其結(jié)果與經(jīng)濟目標(biāo)相同,只不過所有的REDD+剩余都歸樣本戶所有。而當(dāng)采用環(huán)境目標(biāo)時,樣本戶和政策制定者都無法獲得REDD+剩余,但在貼現(xiàn)率為5%的情況下,90.64%的樣本戶土地會變?yōu)樾略炝郑N現(xiàn)率為10%和15%時,所有的樣本戶都會選擇參加生態(tài)固碳項目。因此在完全信息條件下,經(jīng)濟目標(biāo)會使得REDD+剩余從樣本戶轉(zhuǎn)移至政策制定者,而環(huán)境目標(biāo)會使REDD+剩余從低機會成本樣本戶向高機會成本樣本戶轉(zhuǎn)移,而福利目標(biāo)下樣本戶的總收益以及增加造林面積和不完全信息是相同的。
4 結(jié)論與啟示
REDD+機制是國際社會為減緩氣候變化而提出的新舉措,通過向發(fā)展中國家提供大量資金以減少森林砍伐和森林退化。本文通過建立一個簡單支付模型研究了完全信息和不完全信息條件下不同政策目標(biāo)對于REDD+機制收益分配的影響。
研究發(fā)現(xiàn):無論采用何種政策目標(biāo),在不完全信息條件下,由于政策制定者不擁有各代理人機會成本信息,政策制定者只能按照相同的補償標(biāo)準(zhǔn)支付給所有代理人。因而政策制定者無法從REDD+機制中獲益,也無法對REDD+剩余進行分配,所有的REDD+剩余都歸代理人所有。相比基線情景,代理人在不完全信息條件下可以從REDD+機制中獲利。仿真研究的結(jié)果表明,所有政策目標(biāo)的產(chǎn)出(即減少毀林和增加造林面積)是完全相同的。
在完全信息條件下:①政策制定者采用經(jīng)濟目標(biāo)所得到的減少毀林量或增加造林量與不完全信息是一樣的,只不過此時的REDD+剩余歸政策制定者所有。當(dāng)采用經(jīng)濟目標(biāo)時,無論采用何種貼現(xiàn)率,政策制定者都無法把所有的土地納入REDD+機制之中。②當(dāng)政策制定者采用環(huán)境目標(biāo)時,由于可以將低機會成本代理人的REDD+剩余用于對高機會成本代理人的補償,因而在完全信息條件下的減少毀林量或增加造林量大于不完全信息,但是代理人總收益是一樣的,只不過完全信息的存在導(dǎo)致了REDD+剩余的再分配。在環(huán)境目標(biāo)下代理人和政策制定者都無法獲得REDD+剩余,但會使得減少毀林量或增加造林面積顯著增加。③當(dāng)政策制定者采用福利目標(biāo)時,代理人的總收益與不完全信息是相同的,而不同在于各代理人的收益分配。在完全信息條件下,各代理人的收益分配主要取決于政策制定者的偏好。如果采用羅爾斯的最大化最小值標(biāo)準(zhǔn),所有代理人會獲得相同的補償價格,這就使得完全信息條件下的減少毀林量或增加造林量以及各代理人的收益與不完全信息是完全相同的。
本文中的REDD+支付模型只是對政策制定者復(fù)雜決策過程的簡化,對于模型的適用性需要進一步研究。本文忽視了REDD+機制中的各種交易成本,特別是獲取代理人的機會成本信息的成本,這些交易成本的存在可能會降低代理人加入REDD+機制的動機[14],因此需要重視REDD+中知識整合的管理能力[15]。此外REDD+監(jiān)測、報告和驗證體系(MRV)的成本以及環(huán)境規(guī)制計劃所帶來的各項成本[16]也被忽視,而這部分的成本也會對REDD+機制的收益分配產(chǎn)生重要的影響。對于經(jīng)濟目標(biāo)和環(huán)境目標(biāo)而言,機會成本信息的獲取對于政策制定者而言是至關(guān)重要的;而對于福利目標(biāo)而言則無足輕重,但是當(dāng)REDD+剩余的分配不再按照羅爾斯的最大化最小值標(biāo)準(zhǔn)時,這些機會成本的信息就變得非常重要了。
(編輯:劉呈慶)
參考文獻(References)
[1]張慧明,周德群,曹杰,等. 溫室氣體減排技術(shù)研究進展[J]. 中國科技論壇,2011,(9):104-109. [Zhang Huiming, Zhou Dequn, Cao Jie, et al. Research Progress of Reducing Greenhouse Gas Emissions [J]. Forum on Science and Technology in China, 2011, (9): 104-109.]
[2]Food and Agriculture Organization of the United Nations. Key Findings of Global Forest Resources Assessment 2010 [R]. Rome: Forest Resources Assessment, 2010.
[3]Bonan G B. Forests and Climate Change: Forcings, Feedbacks, and the Climate Benefits of Forests [J]. Science, 2008, 320:1444-1449.
[4]Sala O E, Ill F S C, Armesto J J, et al. Global Biodiversity Scenarios for the Year 2100 [J]. Science, 2000, 287: 1770-1774.
[5]Houghton R A. Carbon Flux to the Atmosphere from Landuse Changes: 1850-2005 [R]. Oak Ridge, Tennessee: Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, U.S. Department of Energy, 2008.
[6]Stern N. The Economics of Climate Change: The Stern Review [M]. Cambridge, UK: Cambridge University Press, 2007.
[7]Groom B, Palmer C. Costeffective Provision of Environmental Services: The Role of Relaxing Market Constraints? [J]. Environment and Development Economics, 2010, 15 (2): 219-240.
[8]Delacote P, Palmer C, Bakkegaard R K, et al. Unveiling Information on Opportunity Costs in REDD: Who Obtains the Surplus When Policy Objectives Differ? [J]. Resource and Energy Economics, 2014, 36 (2): 508-527.
[9]Bulte E H, Lipper L, Stringer R, et al. Payments for Ecosystem Services and Poverty Reduction: Concepts, Issues, and Empirical Perspectives [J]. Environment and Development Economics, 2008, 13 (3): 245-254.
[10]Rawls J A. Theory of Justice [M]. Cambridge, Massachusetts, USA: Harvard University Press, 1971.
[11]Brner J, Wunder S. Paying for Avoided Deforestation in the Brazilian Amazon: From Cost Assessment to Scheme Design [J]. International Forestry Review, 2008, 10 (3): 496-511.
[12]Dutschke M, Wong J L P, Rumberg M. Value and Risks of Expiring Carbon Credits from CDM Afforestation and Reforestation [J]. Climate Policy, 2005, 5 (1): 109-125.
[13]李亮. 云南省1992-2007年森林植被碳儲量動態(tài)變化及其碳匯潛力分析[D]. 昆明:云南財經(jīng)大學(xué),2012. [Li Liang. The Dynamic Changes and Potential of Forest Carbon Stock in Yunnan: 1992-2007 [D]. Kunming: Yunnan University of Finance and Economics, 2012.]
[14]Anthon S, Bogetoft P, Thorsen B. Socially Optimal Procurement with Tight Budgets and Rationing? [J]. Journal of Public Economics, 2007, 91 (7-8): 1625-1642.
[15]單海燕,王文平. 跨組織知識整合下的創(chuàng)新網(wǎng)絡(luò)結(jié)構(gòu)分析[J]. 中國管理科學(xué),2012,20 (6):176-184. [Shan Haiyan, Wang Wenping. Analysis of the Structure of Interorganization Innovation Network during the Process of Knowledge Integration [J]. Chinese Journal of Management Science, 2012, 20 (6): 176-184.]
[16]張三峰,卜茂亮. 環(huán)境規(guī)制、環(huán)保投入與中國企業(yè)生產(chǎn)率:基于中國企業(yè)問卷數(shù)據(jù)的實證研究[J]. 南開經(jīng)濟研究,2011,(2):129-146. [Zhang Sanfeng, Pu Maoliang. Environmental Regulation, Environmental Protection Investment and Productivity: An Empirical Study Based on Questionnaire of Enterprises in China [J]. Nankai Economic Studies, 2011, (2): 129-146.]
Abstract The impacts of economic object, environmental object and poverty alleviation object on benefit distribution for REDD+ are analyzed by a simple payment model in two scenarios: asymmetric and full information for opportunity cost. According to agents opportunity costs the policy makers known, the scenarios of asymmetric and full information are established. The policy makers have full information about total distribution of deforestation and potential afforestation area, agents benefits, amount of agents deforestation or potential afforestation in both scenarios. In order to study the impacts of different policy objects on REDD+ results, economic object, environmental object and poverty alleviation object are set up in the paper. On this basis, the household survey data of ecological reforestation and carbon sequestration project in Yunnan is used to simulate the effects of three policy objectives. According to the simulation study, the impacts of three policy objects on agents benefits, benefits of policy makers and the avoided deforestation or increased afforestation are analyzed. The results show that policy makers can only pay same compensation to all agents in the scenario of asymmetric information. Therefore, the outputs of three policy objects are the same. Full information may not increase the forest area for the policy makers of economic object, but could lead to a redistribution of REDD+ surplus from agents to policy maker. By contrast, full information increases the forest area and reduces the agents benefits for the policy makers of environmental object. Full information makes no difference to overall welfare for the policy makers of poverty alleviation object, and the benefits remain belong the agents. The avoided deforestation or increased afforestation in the scenario of full information will be more than that in the scenario of asymmetric information.
Key words REDD+; asymmetric information; full information; profit distribution; policy object
[9]Bulte E H, Lipper L, Stringer R, et al. Payments for Ecosystem Services and Poverty Reduction: Concepts, Issues, and Empirical Perspectives [J]. Environment and Development Economics, 2008, 13 (3): 245-254.
[10]Rawls J A. Theory of Justice [M]. Cambridge, Massachusetts, USA: Harvard University Press, 1971.
[11]Brner J, Wunder S. Paying for Avoided Deforestation in the Brazilian Amazon: From Cost Assessment to Scheme Design [J]. International Forestry Review, 2008, 10 (3): 496-511.
[12]Dutschke M, Wong J L P, Rumberg M. Value and Risks of Expiring Carbon Credits from CDM Afforestation and Reforestation [J]. Climate Policy, 2005, 5 (1): 109-125.
[13]李亮. 云南省1992-2007年森林植被碳儲量動態(tài)變化及其碳匯潛力分析[D]. 昆明:云南財經(jīng)大學(xué),2012. [Li Liang. The Dynamic Changes and Potential of Forest Carbon Stock in Yunnan: 1992-2007 [D]. Kunming: Yunnan University of Finance and Economics, 2012.]
[14]Anthon S, Bogetoft P, Thorsen B. Socially Optimal Procurement with Tight Budgets and Rationing? [J]. Journal of Public Economics, 2007, 91 (7-8): 1625-1642.
[15]單海燕,王文平. 跨組織知識整合下的創(chuàng)新網(wǎng)絡(luò)結(jié)構(gòu)分析[J]. 中國管理科學(xué),2012,20 (6):176-184. [Shan Haiyan, Wang Wenping. Analysis of the Structure of Interorganization Innovation Network during the Process of Knowledge Integration [J]. Chinese Journal of Management Science, 2012, 20 (6): 176-184.]
[16]張三峰,卜茂亮. 環(huán)境規(guī)制、環(huán)保投入與中國企業(yè)生產(chǎn)率:基于中國企業(yè)問卷數(shù)據(jù)的實證研究[J]. 南開經(jīng)濟研究,2011,(2):129-146. [Zhang Sanfeng, Pu Maoliang. Environmental Regulation, Environmental Protection Investment and Productivity: An Empirical Study Based on Questionnaire of Enterprises in China [J]. Nankai Economic Studies, 2011, (2): 129-146.]
Abstract The impacts of economic object, environmental object and poverty alleviation object on benefit distribution for REDD+ are analyzed by a simple payment model in two scenarios: asymmetric and full information for opportunity cost. According to agents opportunity costs the policy makers known, the scenarios of asymmetric and full information are established. The policy makers have full information about total distribution of deforestation and potential afforestation area, agents benefits, amount of agents deforestation or potential afforestation in both scenarios. In order to study the impacts of different policy objects on REDD+ results, economic object, environmental object and poverty alleviation object are set up in the paper. On this basis, the household survey data of ecological reforestation and carbon sequestration project in Yunnan is used to simulate the effects of three policy objectives. According to the simulation study, the impacts of three policy objects on agents benefits, benefits of policy makers and the avoided deforestation or increased afforestation are analyzed. The results show that policy makers can only pay same compensation to all agents in the scenario of asymmetric information. Therefore, the outputs of three policy objects are the same. Full information may not increase the forest area for the policy makers of economic object, but could lead to a redistribution of REDD+ surplus from agents to policy maker. By contrast, full information increases the forest area and reduces the agents benefits for the policy makers of environmental object. Full information makes no difference to overall welfare for the policy makers of poverty alleviation object, and the benefits remain belong the agents. The avoided deforestation or increased afforestation in the scenario of full information will be more than that in the scenario of asymmetric information.
Key words REDD+; asymmetric information; full information; profit distribution; policy object
[9]Bulte E H, Lipper L, Stringer R, et al. Payments for Ecosystem Services and Poverty Reduction: Concepts, Issues, and Empirical Perspectives [J]. Environment and Development Economics, 2008, 13 (3): 245-254.
[10]Rawls J A. Theory of Justice [M]. Cambridge, Massachusetts, USA: Harvard University Press, 1971.
[11]Brner J, Wunder S. Paying for Avoided Deforestation in the Brazilian Amazon: From Cost Assessment to Scheme Design [J]. International Forestry Review, 2008, 10 (3): 496-511.
[12]Dutschke M, Wong J L P, Rumberg M. Value and Risks of Expiring Carbon Credits from CDM Afforestation and Reforestation [J]. Climate Policy, 2005, 5 (1): 109-125.
[13]李亮. 云南省1992-2007年森林植被碳儲量動態(tài)變化及其碳匯潛力分析[D]. 昆明:云南財經(jīng)大學(xué),2012. [Li Liang. The Dynamic Changes and Potential of Forest Carbon Stock in Yunnan: 1992-2007 [D]. Kunming: Yunnan University of Finance and Economics, 2012.]
[14]Anthon S, Bogetoft P, Thorsen B. Socially Optimal Procurement with Tight Budgets and Rationing? [J]. Journal of Public Economics, 2007, 91 (7-8): 1625-1642.
[15]單海燕,王文平. 跨組織知識整合下的創(chuàng)新網(wǎng)絡(luò)結(jié)構(gòu)分析[J]. 中國管理科學(xué),2012,20 (6):176-184. [Shan Haiyan, Wang Wenping. Analysis of the Structure of Interorganization Innovation Network during the Process of Knowledge Integration [J]. Chinese Journal of Management Science, 2012, 20 (6): 176-184.]
[16]張三峰,卜茂亮. 環(huán)境規(guī)制、環(huán)保投入與中國企業(yè)生產(chǎn)率:基于中國企業(yè)問卷數(shù)據(jù)的實證研究[J]. 南開經(jīng)濟研究,2011,(2):129-146. [Zhang Sanfeng, Pu Maoliang. Environmental Regulation, Environmental Protection Investment and Productivity: An Empirical Study Based on Questionnaire of Enterprises in China [J]. Nankai Economic Studies, 2011, (2): 129-146.]
Abstract The impacts of economic object, environmental object and poverty alleviation object on benefit distribution for REDD+ are analyzed by a simple payment model in two scenarios: asymmetric and full information for opportunity cost. According to agents opportunity costs the policy makers known, the scenarios of asymmetric and full information are established. The policy makers have full information about total distribution of deforestation and potential afforestation area, agents benefits, amount of agents deforestation or potential afforestation in both scenarios. In order to study the impacts of different policy objects on REDD+ results, economic object, environmental object and poverty alleviation object are set up in the paper. On this basis, the household survey data of ecological reforestation and carbon sequestration project in Yunnan is used to simulate the effects of three policy objectives. According to the simulation study, the impacts of three policy objects on agents benefits, benefits of policy makers and the avoided deforestation or increased afforestation are analyzed. The results show that policy makers can only pay same compensation to all agents in the scenario of asymmetric information. Therefore, the outputs of three policy objects are the same. Full information may not increase the forest area for the policy makers of economic object, but could lead to a redistribution of REDD+ surplus from agents to policy maker. By contrast, full information increases the forest area and reduces the agents benefits for the policy makers of environmental object. Full information makes no difference to overall welfare for the policy makers of poverty alleviation object, and the benefits remain belong the agents. The avoided deforestation or increased afforestation in the scenario of full information will be more than that in the scenario of asymmetric information.
Key words REDD+; asymmetric information; full information; profit distribution; policy object