Value‐ and Risk‐Based Governance for GEOs‐ Finding Middle Ground
Jennifer KuzmaGoodnight‐NCGSK Foundation Distinguished Professor
School of Public and International AffairsCo‐Director Genetic Engineering and Society Center
NAS‐ILAR RoundtableDecember 8, 2015
regulation
Regulation, oversight, governance
• Governance– Complex set of norms, values,
and processes, and institutions
• Oversight– Watchful and responsible care
under governance
• Regulation– Authoritative rules dealing with
details or procedure having the force of law
Oversight
Regulation
Governance
Kuzma, Environmental Law Reporter 2006
Coordinated Framework for the Regulation of Biotechnology Products (1986‐present)
Agency Jurisdiction Laws
US Dept. of Agriculture (USDA)
Plant pests, plants, veterinary biologics
Federal Plant Pest Act (PPA)—1957, 2000
Food and Drug Administration(FDA)
Food, feed, food additives, vet. drugs, human drugs, medical devices
Federal Food, Drug and Cosmetic Act (FFDCA)--1958
Environmental Protection Agency(EPA)
Microbial and plant pesticides; novel microbes
Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA)--1947; Toxic Substances Control Act (TSCA) 1976
Must be “science based”No “new risks”, no new laws needed, “product not process”
How did this system keep up with technology?The Pacing Problem
Phases of CFRB• Evolution (1950s‐1986)
– Establishment of “pacing through executive branch agency policy‐making”
• Implementation (1986‐circa 2002)– “pacing through rules”
• Adaptation (2002‐circa 2009)– “pacing through guidance”
• Revolution (circa 2009‐present)– “pacing through power (fundamental political
shift)”
Cycle of oversight for GEOsConflict and Reaction
Pacing through Policy
Pacing through Rules
Pacing through Guidance
Pacingthrough Power
(Revolution)
External pressure
External pressure
External pressure
Phases over Time
What next!???
Next generationof
GEOs oversight
Contested System in United StatesPaced, but “properly paced?”
• Policy processes are contested in key ways:– Buying of natural, local, organic foods– NGO legal suits concerning GM crops – Recent labeling propositions on state ballots and state bills– Transgenic salmon approval and controversy
• Delays in technology deployment– LONG (over decade) approval for GM salmon (1st animal‐derived food)– Previous down turn in VARIETY of GM plants submitted for regulatory approval– UNTIL….gene editing!
• Companies are starting to use GM methods that stretch definition of “genetic engineering or rDNA
– Active Regulatory Avoidance– Promoting innovation around regulation– Arguments not to regulate based on “process”
• USDA deciding many GM crops outside of regulatory scope
Integrated Oversight AssessmentNSF grant 2007‐2011
How was the oversight model developed ?
What are its attributes ?
What are its outcomes ?
How do the attributes evolve over time ?
How was the oversight model developed ?
What are its attributes ?
What are its outcomes ?
How do the attributes evolve over time ?
How was the oversight model developed ?
What are its attributes ?
What are its outcomes ?
How do the attributes evolve over time ? Kuzma,, Paradise, et al Risk Analysis(2008)
Complex System of GEOs Governance:
Intertwined Normative, Empirical, and Institutional Criteria
Public InputD4
Data requirements
A9
Incentives A14
Public InputA19
Public Confidence
O24
p<0.05, p<0.0029
GM Oversight: Proper pacing?(Kuzma et al JLME 2009)
• High flexibility
• Clear subject matter (although changing…)
• Weak legal grounding allowing for multiple interpretations
• Complex institutional structure
• Little transparency
• Low level of informed consent
• Few opportunities for public input
• Low capacity
More controversy, delay, rejection?
Too much uncertainty for developers of new GM products?
• Mapping of the Scientific Landscape of Gene Editing
• Interpretation & Analysis of Oversight System
• Expert Visions of Governance
Early work on genome editing (Tagmo)
• NSF DBS Award 0923827 (2009‐2013 )• Precise Engineering of Plant Genomes using Zinc Finger Nucleases & Societal implications
Voytas U of MN PIDrena Dobbs, ISU co‐PIJennifer Kuzma, U of MN (NCSU), co‐PIKen Wang, Harvard, co‐PI
• Document Analysis• Bibliometrics• Interviews & Surveys (content analysis)
• Workshop
Key findings (2010‐2011):
• Variety of new plant genetic engineering techniques• Many may fall outside existing regulatory triggers• Uncertain how oversight systems will address TagMo• Challenge definition of genetic engineering and plant pest
Sept. 2011
• Concentrated • Focusing on biomedical problems • Few partnerships with LDCs • Different foci in DCs and LDCs• Little Collaboration among U.S. funders• Related to History of problems of ag biotech• Little safety research compared to development
funding
• Risk of repeating socio=political context of conflict and contention
Expert technology understandings for do not map neatly onto governance policy preferences
Concerns about gene editing Attitudes about oversight
Subject Matter Stakeholder‐Experts
Gene editing is an Incremental Technology ‐‐
• Maybe it doesn’t change technology concerns dramatically
• It doesn’t FORCE a governance change, but gives us OPPORTUNITY to re‐examine and change governance.
• Diversity of changes suggested from more relaxed to more rigorous
Gene Editing is a Revolutionary Technology
• Gene editing is a dramatically different technology that forces a change in governance: How?
Relaxes need for oversight
Intensifies need for oversight
Hyper tech‐Hypo RegProcess based argument
Systems View
Pragmatist/Opportunist
Key Narratives
Two gene editing researchers
Points of agreement—quantitative survey
• Majority of the subject matter experts (SMEs) agree that mistakes in governance were made with 1st generation GMOs
• Majority would prefer some level of premarket review by government
• Majority acknowledge that more stakeholders and citizens need to be informed/engaged
• What are we doing in U.S.?
U.S. Stakeholder Approaches
• Government– Decide not to capture several GM
plants, including genome edited plants, at all (USDA) under plant pest regulations
– Making decisions largely n private, lack of advisory or stakeholder committees
– Until recently…CFRB meetings.
– U.S. working outside of international processes in some cases (CBD, etc.)
• Developers:– Innovating around regulations– Appealing to “sound science” – Appealing to Product not process as
why gene editing shouldn’t be regulated
– Code for limiting oversight considerations to direct human health and environmental harms
–• NGOs
– Continue to fight decisions in federal court
– Continue to support labeling initiatives
Harmonization?International Policy Process Approaches
• U.S.– Let It Go‐‐behind closed doors– if not plant pest or pesticide
• Japan– think and publish, international treaty (Cartegena—”modern biotech”)– SDN‐1, ‐2, and ‐3 distinctions
• EU – EFSA deliberative expert group– SDN‐1, ‐2, and ‐3 distinctions
• New Zealand – Let it go – SDN‐1 (2?) exemption contested in law suits
• Trade? Public Perception? Health and Environmental Safety?• Respecting values of other countries?
1) Different Scientific Conceptions of Risk (Renn, 1992, et al)
• Different types of harms that need to be considered in risk analysis
• They are “scientific”
• Harms and damages that can occur with “exposure”
– 1st order physical health and environmental– 2nd order physical health and environmental– Social structure harm– Ethical affronts (without choice, voice, or consent)– Psychological well‐being– Financial impacts (direct)– Economic impacts (indirect)– Cultural disruption
• Only a wide range of perspectives and voices can “assess” these
2) Values in the RA
– Endpoints chosen– Interpreting Risk Relevant Data– Temporal and geographical timescale– Choice of baseline(pristine nature?, ten years ago?, conventional
breeding?, past biocontrol agents?)– Decision options and scenarios (are non‐technological or more
“natural” options fully considered)– Ecological economics (are non‐use and existential values being
incorporated)
Thompson 2007; Kuzma and Besley 2008
“Science” alone cannot spur social actionIt cannot tell us what to do
• Places for values– Interpretation of data– Standards for safety– Choice of endpoints
3) Risk Management is always value laden(so is drawing regulatory boundaries)
• There is always risk and always uncertainty• Precaution is not anti‐science, it’s a belief about risk governance• Anti‐precaution is too.• Product versus process , binary framing not helpful
– Regulate all or none, as RA comes into the product review, so how can you know the risk beforehand to capture the products in a purely objective way?
• Huge debates stem from differences in world view about role of government, regulation, and level of precaution that is warranted.
– Progress of technology is often pitted against government regulation.– Government regulation seen as prohibiting technological development
Technological RiskAlways “risk”, never zero
• As with any technology, there are trade‐offs– Desirability should depend on
• careful risk/benefit analysis• life cycle analysis• socioeconomic impacts analysis (+/‐)• jobs lost/jobs gained• systems analysis• cultural harms/benefits• distribution of harms and benefits• comparison of above to other technological or socio‐cultural options
Gene Editing and Drives and Ethical Dimensions
• Issues
Power Voice Autonomy
Informed Consent Equity issues Integrity
“Do no harm” principle
Utilitarianism (Costs and Benefits)
Conceptions and desires for “nature”
Rights of non‐human species
Inter‐generational
equity
• Ways of thinking about ethics– Utilitarian– Bioethics Principles – Rights‐based– Procedural/Structural– Ethics of care– Deep Ecology – Etc.
Esvelt et al. 2014
https://www.youtube.com/watch?v=Cy69C6vnFCQ\
Gene Drives Spread Themselves
Draft typology gene drives for analyzing Intergenerational EquityKuzma and Rawls in review 2015
What about the animals!?• Do Animals engineered with gene drives have rights?
• To Integrity of being in their “natural” or unengineered state?
• To exist in the face of “killing systems” and gene drives? (kill themselves essentially)• At individual level• At population level• At species level
• Would gene drives provide least amount of suffering? • (as opposed to rat poison on islands for example)
– Gene editing—harmful off‐target mutations (CRISPR 0.1% to over 60%, Joung et al) greater or less than genetic engineering? Conventional breeding?
4) People are not “irrational”They can get it
• Amazed by people’s ability to “get” the science and ask important questions– Our focus group studies– My and other’s participation in multiple public engagement events
• They base their views on a variety of factors JUST LIKE WE (as experts) DO
Psychometric theory of Risk Perception (Slovic 1987)Features of technology matter
Unknown
Uncontrollable
Reasons why people do not view risk of driving a car the same as risk of air travel(or GEOs) even thought fatalities per year per person are much, much greater
Public perception of emerging technologies: Altruism and Skepticism(Brown, Fatehi, Kuzma JNR 2015)
U.S. Consumer priorities in labeling(Brown & Kuzma in prep)
Public perception Summary• Consistent in literature of Willingness to Pay for non‐GM
• Consistent desire for labeling
• Acceptance depends on product
• Different “cultural” consumer groups
• Complex set of perception factors including trust, cultural world view, demographics, risk and benefit.
Move from pure “science based” to “science informed, value‐attentive,
public‐respectful ” oversight
• It’s not all about messaging or understanding public attitudes
• It’s about listening, dialogue, and mutual social and bidirectional learning
• Analytical‐deliberative processes (NRC 1996)
Problem
What public perception studiesand engagement
tell us about oversight(what the public cares about)
TransparencyMandatory systems
Opportunities for InputKnowledgeChoice
How emerging technology
oversight systems (non‐medical products)
developand operate
Little Transparency (CBI)Voluntary labeling
Few opportunities for input(Fed Reg, and
Public Meetings)
Will gene editing and gene drives in GEOs deployed in nature prompt a change in governance?
• Will the policy window in the multiple streams crack? It has.
• Greater recognition of problem of GEOs governance due to gene drives?– (problem stream)
• Plenty of policy options proposed – (policy stream)
• But Contention in the politics stream…
• Need for strong Policy Entrepreneurs to negotiate the integration of the three streams, and likely a Focusing Event
Policy Streams (Kingdon)
Paradigms for better governance are not the problem…
Need
• Values are implicit in all types of assessment
• Science is uncertain
• Never enough for “sound science” to make decisions
• Industry and Govt. Scientists are not the only ones who should have a “voice”
Challenges
• Technological Elitism
• “Science Based” system in face of uncertainty/ambiguity leads to biased interpretations of evidence on all sides.
• Sticking to the science marginalizes other world views, local and specialized knowledge
• Creates distrust, skepticism
“Only products matter and impacts”CON: Not enough risk science, hypocrisy, lack of trust
“Only process matters and we don’t like GM process”CONS: Preclude potentially safer and cheaper technology development
A Theoretical “Middle Ground” Forward
Post‐normal science in Decision making
Funtowicz and Ravetz (2006)
“Open dialogue, extended peer communities”“Research as object of critical scrutiny”
Structure for Stakeholder InvolvementIRGC 2006
Reaching out to NGOs from developers
Responsible Research & Innovation
• Owen & Von Schomberg (2013).
• “I categorise here four types of irresponsible innovation: Technology push, Neglectance of fundamental ethical principles, Policy Pull, and Lack of precautionary measures and technology foresight.”
• “Responsible Research and Innovation is a transparent, interactive process by which societal actors and innovators become mutually responsive to each other with a view to the (ethical) acceptability, sustainability and societal desirability of the innovation process and its marketable products( in order to allow a proper embedding of scientific and technological advances in our society).”
SofterApproaches
Voluntary data-sharingCodes of conductVoluntary consultation with agency reviewGuidelines
Harder Approaches
Ban, moratoriumStandardsStringent pre-market testingEnforceable fines
Coordinating Entity or Process*
Public Engage-
mentand Input
Agency Imple-
mentation
* with citizen, governmental, academic, industry, tribal, and NGO representation
Spectrum of Oversight
Dynamic Oversight:An example “practical” way forward
Ramachandran, et al. 2011
Principles Anticipates convergence Inclusive Public empowerment Learning among groups Respectful Multiple iterations Preparedness at all stages ◦ (including post‐market)
Transparent Adequate resources Continuous Evolving Information‐generating Information‐ and value‐based
Policy Delphi for Anticipatory Governance of SBSocietal Risk Evaluation Scheme (Cummings and Kuzma in review)
• Frames• Policy Sciences• Future Studies• Upstream Technology Assessment
• Cases of medium to longer‐term development • Mental models (mapping) of diverse SMEs• Risk Analysis, and Governance questions• Policy Delphi process in 4 rounds
• Interviews• Survey• Workshop• Survey
48
49
(Responsible Innovation) Recommendations
Deploy SRES in Anticipatory Governance at early stages of product development (before investment)
As product developing, gather more information and data in high SRES areas—dialogue or interviews for specific research needs
These activities as foundation for resource allocation and funding for information to inform decision making
SRES as screening tool at R&D stage
Dialogue for more specific needs
Resources For Data & Analysis
Acknowledgments