Avoiding carbon leakage from nature-based offsets by design

With nature-based offsets emerging as a core strategy for meeting near-term climate targets, it is essential they deliver real and verifiable mitigation gains. However, the interventions that generate can have unintended effects cause carbon leakage ultimately reduce mitigation. Although "old news" various anti-leakage measures been considered, there little evidence current practices to address actually work. In this perspective, we present vastly underestimated in practice argue efforts improve accounting methods are unlikely accuracy required. We therefore propose elaborate an alternative approach by design, based on new conceptual framework understanding interventions. further outline three principles offset developers, certifiers, consumers implement now credibility of offsets, without negating ambition investment solutions. Amid enthusiasm decentralized, market-led change solutions, ecosystems widely seen linchpins global strategies. Land use (notably forests) provides quarter planned under Paris Agreement,1Grassi G. House J. Dentener F. Federici S. den Elzen M. Penman The key role forests targets requires science credible mitigation.Nat. Clim. Chang. 2017; 7: 220-226https://doi.org/10.1038/nclimate3227Crossref Scopus (277) Google Scholar COP26 (Glasgow) signaled willingness allow international transfers Nature-based already feature most market emissions pricing schemes2Webb C. Zakir Z. What Agreement Means Carbon Pricing Natural Climate Solutions: A Business Guide.2019Google central corporate net-zero pledges.3Kreibich N. Hermwille L. Caught between: feasibility voluntary post-2020.Clim. Pol. 2021; 21: 939-957https://doi.org/10.1080/14693062.2021.1948384Crossref (29) Driven commitments,4The World BankState Trends 2021. Bank, 2021Crossref markets neared US$2 billion traded value 2021, with 67% originating forestry land-use projects.5Donogrio Maguire P. Myers K. Daley Lin State Voluntary Markets 2021: Installment 1 (Ecosystem Marketplace).2021Google Support broad (projects offer both low-cost environmental benefits), estimates total potential high (20%–30% needed keep warming 1.5°C).6Roe Streck Beach R. Busch Chapman Daioglou V. Deppermann A. Doelman Emmet-Booth Engelmann et al.Land-based mitigate change: country.Glob. Biol. 27: 6025-6058https://doi.org/10.1111/gcb.15873Crossref PubMed (49) Not all will (or should7Anderson C.M. DeFries R.S. Litterman Matson P.A. Nepstad D.C. Pacala Schlesinger W.H. Shaw M.R. Smith Weber Field C.B. solutions not enough.Science. 2019; 363: 933-934https://doi.org/10.1126/science.aaw2741Crossref (78) Scholar,8Griscom B.W. Lomax Kroeger T. Fargione J.E. Adams Almond Bossio D. Cook-Patton S.C. Ellis P.W. Kennedy Kiesecker need natural energy stabilize our climate.Glob. 25: 1889-1890https://doi.org/10.1111/gcb.14612Crossref (24) Scholar) substitute reductions, but rapidly expanding. Most pathways depend altering state coupled ecological-economic systems.9Seddon Harnessing mitigating adapting change.Science. 2022; 376: 1410-1416Crossref (33) Scholar,10Griscom Houghton R.A. Miteva D.A. Shoch Siikamäki J.V. al.Natural solutions.Proc. Natl. Acad. Sci. USA. 114: 11645-11650https://doi.org/10.1073/pnas.1710465114Crossref (1214) effect such be tracked credits framework, which become when used other actions. If underlying inaccurate, offsetting may prove dangerous distraction. Recent work11West T.A.P. Börner Sills E.O. Kontoleon Overstated emission reductions from REDD+ projects Brazilian Amazon.Proc. 2020; 117: 24188-24194https://doi.org/10.1073/pnas.2004334117Crossref (63) Scholar,12Badgley Freeman Hamman J.J. Haya B. Trugman A.T. Anderegg W.R.L. Cullenward Systematic over-crediting California’s forest program.Glob. 28: 1433-1445https://doi.org/10.1111/gcb.15943Crossref (22) Scholar,13Guizar-Coutiño Jones J.P.G. Balmford Carmenta Coomes evaluation effectiveness at reducing deforestation degradation moist tropics.Conserv. 36: e13970https://doi.org/10.1111/cobi.13970Crossref (6) has highlighted widespread overestimation degree altered world (i.e., their additionality), another source inaccuracy—carbon leakage—merits equal attention. Leakage occurs some intervention fall outside boundary track (e.g., action causing one place also increases elsewhere). These beyond-boundary extremely difficult measure,14Meyfroidt Garrett Gardner Godar Kis-Katos Soares-Filho B.S. Wunder Focus spillovers: informing governance tele-coupled world.Environ. Res. Lett. 15090202https://doi.org/10.1088/1748-9326/ab7397Crossref (32) Scholar,15Schwarze Niles J.O. Olander Understanding managing forest–based greenhouse–gas–mitigation projects.Philos. Trans. Math. Phys. Eng. 2002; 360: 1685-1703https://doi.org/10.1098/rsta.2002.1040Crossref (51) particularly economic implicated (so-called leakage). Today’s compliance routinely transact premise account sufficiently accurate,16Chagas Galt H. Lee Neeff Close Look Quality Credits.2020Google Scholar,17Espejo A.B. Becerra-Leal M.C. Aguilar-Amuchastegui Comparing integrity REDD programs renewable projects.Forests. 11: 1360https://doi.org/10.3390/f11121360Crossref (9) Scholar,18van der Gaast W. Sikkema Vohrer contribution credit addressing challenge.Clim. 2018; 18: 42-48https://doi.org/10.1080/14693062.2016.1242056Crossref (43) often framed tractable problem “sophisticated robust tools”2Webb “policy levers manage risks”7Anderson available. Yet main currently (adjusting issued using discount factor) recognized insufficiently rigorous “in long run,”19Atmadja Verchot review research, policies strategies (REDD+).Mitig. Adapt. Strateg. Glob. 2012; 17: 311-336https://doi.org/10.1007/s11027-011-9328-4Crossref (57) how best deal controversial.20Peterson St-Laurent Hagerman Hoberg Barriers development offsetting: insights British Columbia, Canada.J. Environ. Manage. 203: 208-217https://doi.org/10.1016/j.jenvman.2017.07.051Crossref Scholar,21Streck leakage: debunking myths promoting integrated solutions.Clim. 843-852https://doi.org/10.1080/14693062.2021.1920363Crossref (13) Among specialists, problems well known: Richards Andersson22Richards Andersson leaky sink: persistent obstacles sequestration program individual projects.Clim. 2001; 1: 41-54Crossref (2) argued over 20 years ago theoretical practical challenges prevent accurate measurement associated specific intervention, concluding “either reliability project analysis low or costs high, quite possibly both.” Concern about followed first early 1990s,23Putz F.E. Pinard M.A. Reduced-impact logging carbon-offset method.Conserv. 1993; 755-757https://doi.org/10.1046/j.1523-1739.1993.7407551.xCrossref Scholar,24Bastos Lima M.G. Persson U.M. Meyfroidt boosting governance: analysis.Environ. 14105006https://doi.org/10.1088/1748-9326/ab4551Crossref Scholar,25Brown Neil Adger Economic political offsets.For. Ecol. Manag. 1994; 68: 217-229https://doi.org/10.1016/0378-1127(94)90047-7Crossref (17) decades work thoroughly explored issue: several reviews exist,19Atmadja Scholar,26Hertel T.W. perspectives land leakage.Environ. 13075012https://doi.org/10.1088/1748-9326/aad2a4Crossref (28) Scholar,27Henders Ostwald Forest quantification suitability assessing REDD.Forests. 3: 33-58https://doi.org/10.3390/f3010033Crossref research interest remains high.9Seddon Scholar,14Meyfroidt Scholar,28Atmadja S.S. Duchelle A.E. De Sy Selviana Komalasari Angelsen How do contribute goals Agreement?.Environ. 17044038https://doi.org/10.1088/1748-9326/ac5669Crossref (1) proposed either politically intractable scaling up frameworks include effects21Streck Scholar,29Andersson K.R. Implementing program: sink fixed?.Clim. 16Google Scholar,30Wunder leakage?.in: Moving ahead REDD: Issues, Options, Implications. Center International Forestry Research, 2008: 65-76Google Scholar,31Boucher Elias From deforestation-free supply chains: scale.Carbon 2013; 4: 473-475https://doi.org/10.4155/cmt.13.47Crossref Scholar,32Schwartzman Lubowski R.N. S.W. Keohane N.O. Kerr Oppenheimer Hamburg S.P. Environmental depends scale systemic changes, sector origin.Environ. 16091001https://doi.org/10.1088/1748-9326/ac18e8Crossref (7) Scholar,33Keith Vardon Obst Young Mackey Evaluating conservation comprehensive accounting.Sci. Total 769144341https://doi.org/10.1016/j.scitotenv.2020.144341Crossref (31) poorly understood inconsistently applied.15Schwarze absence viable alternative, continue retired ad hoc unknown accuracy. aim develop dealing offsets. focus leakage, reviewing its basis, empirical literature, practice. suggests today systematically underestimating effects, attempts refine then novel designing avoid problem: thinking enables identification arises Leveraging available through design-based solution preclude (currently unachievable) precision accounting. Our conservative immediately implemented within today’s decentralized regimes, clarifying ongoing misunderstandings context. While number classifications leakage14Meyfroidt Scholar,19Atmadja Scholar,34Meyfroidt Roy Chowdhury de Bremond E.C. Erb K.-H. Filatova R.D. Grove J.M. Heinimann Kuemmerle al.Middle-range theories system change.Global Change. 53: 52-67https://doi.org/10.1016/j.gloenvcha.2018.08.006Crossref (267) exist (including impacts via connected ecological systems,15Schwarze information, motivation, institutions,24Bastos spatial interactions35Delacote Robinson E.J. Roussel Deforestation, avoided policies: analysis.Resour. Energy Econ. 2016; 45: 192-210https://doi.org/10.1016/j.reseneeco.2016.06.006Crossref Scholar), two canonical types involving agents: “direct” “market” leakage. Direct “activity”) agents targeted shift activities boundary, whereas non-targeted adjust behavior response incentives. activity relatively (targeted known, actions observable), not. Teasing out background difficult, since responding changes incentives much due unobservable motivations. mechanism operates information transmission price. model single equilibrium, output producer causes prices rise, moving equilibrium incentivizing producers respond adjusts along path. net result adjustments interconnected complex, theory general guidance.36Murray B.C. McCarl B.A. H.-C. Estimating programs.Land 2004; 80: 109-124https://doi.org/10.2307/3147147Crossref (170) Scholar,37Gan Measuring transnational conservation.Ecol. 2007; 64: 423-432https://doi.org/10.1016/j.ecolecon.2007.02.032Crossref (115) All else being equal, lower if demand elastic respect price products substitutable, higher more price, substitutable goods footprint, restrictions small (Figure 1). conditions raise concerns typically operate globalized “food, fuel, fiber” commodity where highly inelastic price,38Turner J.A. Buongiorno income elasticities imports panel data.Scand. For. 19: 358-373https://doi.org/10.1080/02827580410030127Crossref (35) Scholar,39Buongiorno Income time dependence product implications forecasting.Silva Fenn. 2015; 49: 17https://doi.org/10.14214/sf.1395Crossref Scholar,40Roberts M.J. Schlenker Identifying agricultural commodities: US ethanol mandate.Am. Rev. 103: 2265-2295https://doi.org/10.1257/aer.103.6.2265Crossref (173) very large relative size approaches measuring each limitations. Partial models arguably suitable because developed specifically capture interdependencies (equilibrium effects) simulating agents. Unfortunately, subtle parameters substantially alter results39Buongiorno (as demonstrated principle below), building specialized personnel abundant data. Accounting-based input-output material flow analysis) provide compelling circumstantial evidence41Meyfroidt Lambin E.F. transition Vietnam displacement abroad.Proc. 2009; 106: 16139-16144https://doi.org/10.1073/pnas.0904942106Crossref (256) Scholar,42Meyfroidt Rudel T.K. transitions, trade, use.Proc. 2010; 107: 20917-20922https://doi.org/10.1073/pnas.1014773107Crossref (397) cannot separate causal impact intervention. Quasi-experimental econometric techniques can, rely assumptions independence pseudo-controls violated presence effects. Many researchers apply simple zone-based methods,43Alix-Garcia Shapiro E.N. Sims K.R.E. slippage: Mexico’s national payments ecosystem services program.Land 88: 613-638https://doi.org/10.3368/le.88.4.613Crossref (213) Scholar,44Roopsind Sohngen Brandt Evidence reduces tree cover loss cover, country.Proc. 116: 24492-24499https://doi.org/10.1073/pnas.1904027116Crossref quantify activity-shifting protected areas45Ford S.A. Jepsen Kingston Lewis E. Brooks T.M. MacSharry Mertz O. Deforestation undermines tropical subtropical areas.Global Biogeogr. 29: 2014-2024https://doi.org/10.1111/geb.13172Crossref (23) Scholar,46Robalino Pfaff Villalobos Heterogeneous local spillovers areas Costa Rica.J. Assoc. Resour. 795-820https://doi.org/10.1086/692089Crossref (25) Scholar). This assumption known unaffected incentives, untenable well-functioning large-scale indirect concern. innovations ongoing,47Moffette Gibbs H.K. Agricultural legal Amazon.Land 97: 155-179Crossref Scholar,48Groom Palmer Sileci Indonesia’s moratorium concessions cost-effective yet pledges.Proc. 119e2102613119https://doi.org/10.1073/pnas.2102613119Crossref (10) accessible, replicable, method estimating found. Thus far, assessments literature focused primarily “stop-harvest” projects, frequently leveraging established partial models. countries these least 70% reduced output, measured production37Gan Scholar,49I Kallio A.M. Solberg harvest open economy: case Norway.Scand. 33: 502-510https://doi.org/10.1080/02827581.2018.1427787Crossref (15) Scholar,50N Wear Murray Federal timber restrictions, interregional spillovers, softwood markets.J. 47: 307-330https://doi.org/10.1016/S0095-0696(03)00081-0Crossref (75) Scholar,51Hu X. Shi Hodges China’s policies.Forests. 2014; 5: 2613-2625https://doi.org/10.3390/f5112613Crossref Scholar,52Kallio A.M.I. Käär Päivinen setting reference levels sinks EU European sector.For. 92: 193-201https://doi.org/10.1016/j.forpol.2018.04.010Crossref (37) stocks.53Nepal Ince P.J. Skog K.E. Chang S.J. benefits, reserve scenarios United States.J. 286-306https://doi.org/10.1016/j.jfe.2013.06.001Crossref (14) Lower (50% less) found price,54Sun Set-asides sequestration: permanence leakage.Climatic 96: 409-419https://doi.org/10.1007/s10584-009-9628-9Crossref (21) developing country context deemed negligible55Sohngen Brown economies: stop harvesting Bolivia study.Can. 34: 829-839https://doi.org/10.1139/x03-249Crossref (73) Kuik56Kuik food markets: CGE analysis.Mitig. 641-655https://doi.org/10.1007/s11027-013-9527-2Crossref (16) 0.5%–11.3% countries, methodological choices contradicted evidence47Moffette Afforestation produce than conversion ≤43% vs. ≤ 92% estimate36Murray productivity differences availability underutilized land. non-forest less studied. Kim al.57Kim M.-K. Peralta Land-based greenhouse gas discounting.Ecol. 105: 265-273https://doi.org/10.1016/j.ecolecon.2014.06.009Crossref find 15% crop conversion, while studies reserves (also “slippage”) suggest important58Lichtenberg Smith-Ramírez Slippage cost sharing.Am. Agric. 2011; 93: 113-129https://doi.org/10.1093/ajae/aaq124Crossref (39) low59Fortmann Southgate Assessing group heterogeneity community Guatemala’s Maya biosphere reserve.Land 503-526https://doi.org/10.3368/le.93.3.503Crossref (20) Scholar: 4% (measured loss)43Alix-Garcia 20% farm area,60Wu program.Am. 2000; 82: 979-992Crossref although criticisms estimate61Roberts Bucholtz spurious cor