Learning from Difference: The New Architecture of Experimentalist Governance in the EU

1
Learning from Difference The New Architecture
of Experimentalist Governance in the EU

• Charles Sabel (Columbia Law School)Jonathan
  Zeitlin (University of Wisconsin-Madison)

2
A novel polity without a state

• Despite recurrent crises, the EU is creating a
  single market, while constructing a framework
  within which MS can protect public health and
  safety (broadly understood) in nationally diverse
  ways

3
A new governance architecture

• EUs regulatory successes rest on a new
  architecture of decision making
• A recursive process of framework rule making and
  revision based on networked deliberation among
  European and national actors
• New architecture increasingly widespread across
  EU policy making despite not being explicitly
  grounded in Treaties and directives

4
I. Elements of the new architecture

• Framework goals and metrics for assessing
  progress towards them established jointly by EU
  institutions and MS
• E.g. full employment, good water status,
  unified energy grid
• Lower-level units (national ministries,
  regulatory authorities, collaborating actors)
  given freedom to advance these goals as they see
  fit

5
Elements of the new architecture (2)

• In return for autonomy, lower-level units must
• report regularly on their performance, esp. as
  measured by agreed metrics
• participate in peer reviews where their results
  are compared to others pursuing other means to
  same ends
• Framework goals, metrics, procedures
  periodically revised in light of experience
• By actors who initially established them other
  participants whose views are recognized as
  important for full fair deliberation (e.g.
  civil society orgs.)

6
Experimentalist governance as a functional concept

• 4 key architectural elements as a set of
  necessary functions which can be performed
  through a variety of institutional arrangements
• No one-to-one mapping of governance functions to
  specific institutional mechanisms, and vice versa
• A single governance function (e.g.
  monitoring/review of implementation) can be
  performed by multiple institutional devices,
  singly or in combination
• A single institutional mechanism (e.g. formal
  peer review) can perform several analytically
  distinct governance functions

7
EU governance as directly-deliberative polyarchy
(DDP)

• Deliberative uses argument to disentrench
  settled practices and redefine interests/preferenc
  es
• Directly-deliberative uses the concrete
  experience of actors differing reactions to
  current problems to generate novel possibilities
  for consideration
• Polyarchic a system in which local units learn
  from, discipline, and set goals for each other

8
A machine for learning from diversity

• Well-suited to heterogeneous settings like the EU
• where local units face similar problems, and can
  learn from each others separate efforts to solve
  them, even if solutions can rarely be generalized
  straightforwardly
• Transforms diversity into an asset rather than a
  obstacle to closer integration
• A system of experimentalist governance
• in the pragmatist sense of systematically
  provoking doubt about its own assumptions and
  practices, while treating all solutions as
  provisional and corrigible

9
Multiple routes to the EUs new governance
architecture

• Federated regulation in privatized infrastructure
• Fora councils of regulators in telecoms, energy
• Networked administrative agencies protecting
  public health and safety
• Drug authorization, occupational health safety,
  (environmental protection)
• Open Methods of Coordination (OMC)
• Employment, social protection/inclusion,
  education/training
• Operational coordination among natl authorities
  EU agencies in Justice Home Affairs

10
Extensions of the new architecture

• Responses to catastrophe
• Food safety, maritime safety
• Prudential regulation in advance of failure
• Financial services (Lamfalussy Process)
• Revision of existing centralized regulation
• Competition/antitrust policy, state aid
• From rules to rights
• Anti-discrimination policy, fundamental rights

11
Minimal scope conditions

• Scope conditions for experimentalist governance
  broader than historical contexts of emergence in
  EU
• Strategic uncertainty
• Policy makers recognize that they do not know how
  to achieve their declared goals
• Multi-polar or polyarchic distribution of power
• No single actor able to impose preferred solution
  w/o taking into account views of others
• Opens up possibility of transforming distributive
  bargaining into deliberative problem solving
  through institutional mechanisms of
  experimentalist governance

12
II. Responses to standard objections

• Undermining the rule of law?
• Lack of accountability in recursive rule making
• Ineffectiveness/lack of teeth?
• Deliberative governance as soft law
• At best a secondary complement to hard law
  based on political bargaining
• Subverting democracy?
• Technocratic deliberation among transnational
  expert networks as a threat to democracy

13
The limits of principal-agent governance

• Accountability in representative democracy is
  based on principal-agent model
• But principal-agent accountability depends on
  ability of principals to specify goals precisely
  enough to give clear instructions to agents or to
  recognize reliably when actions do or dont serve
  specified ends
• Model breaks down under conditions of
  uncertainty/volatility when solutions to problems
  can only be identified as they are being pursued
• Recursive redefinition of ends and means in EU
  experimentalist governance as a response to this
  situation

14
dynamic accountability through peer review as
an alternative

• Dynamic accountability not as conformity to a
  preset rule, but as requirement to explain
  reasons for exercising discretion in choosing one
  particular way of advancing a common project,
  subject to review by peers with sufficient
  knowledge to evaluate such explanations
• Peer review of EU experimentalist governance
  through fora, networked agencies, councils of
  regulators, OMCs
• Processes through which EU decision-makers learn
  from and correct each other even as they set
  goals and performance standards for the Union

15
Experimentalist governance is not soft law

• New architecture routinely results in revisions
  of EU law, or elaboration of revisable standards
  mandated by law and principles that may
  eventually be given binding force
• For every decision-making process that produces
  non-binding legal sanctions, there is typically
  another in the same domain where non-compliance
  can have draconian consequences

16
Destabilization regimes

• Sanctions in EU experimentalist governance often
  take the form of what may be called a
  destabilization regime
• A set of mechanisms for unblocking impasses in
  framework rule making and revision by rendering
  the current situation untenable while suggesting
  -- or causing the parties to suggest --plausible
  and superior alternatives

17
Varieties of destabilization regimes

• Destabilization regimes can take many forms
• Requirement to deliberate in public
• E.g. EU food safety conflict clause
• In cases of disagreement over scientific risk
  assessment between EFSA and national authorities,
  both sides are obliged to submit a joint document
  explaining their differences
• But doing so would put both parties at risk
  because either could lose the debate in full
  public view, while political instrumentalization
  of the issue could endanger future collaboration
  and scientific exchange
• This gives both parties strong incentives to
  continue deliberation until an agreement that
  both can defend is public is reached

18
The right to challenge and the obligation to
explain

• E.g. EU competition policy network
• The Commission has the right to take over cases
  from national competition authorities, but must
  formally justify its decision to other members of
  the network
• Not just a vertical but also a horizontal right
  of challenge any member of the network can
  demand a review of another national competition
  authoritys handling of a case

19
Penalty defaults

• At the other extreme are penalty defaults
• Instead of imposing deliberation requirements,
  the central authority creates brutal
  disincentives for refusal to deliberate
• Rules sufficiently unpalatable to all parties
  that each is motivated to contribute to an
  information-sharing regime that allows fair and
  effective regulation of their interdependence
• In a world where standard rule making produces
  such unpredictable consequences as to be
  unworkable, the easiest way to generate penalty
  defaults is to (threaten to) engage in
  traditional rule making

20
An EU example the Florence Electricity Forum

• Unblocking of deliberative impasses on thorny
  issues such as cross-border tarification by
  Commission threats to invoke competition law
  powers (antitrust, merger control, state aid
  rules)
• Could leave obstructive/intransigent actors worse
  off than a compromise reached in the Forum
• Commissions powers of legislative initiative and
  delegated regulation as a means of inducing MS
  and private actors to cooperate in framework rule
  making for fear of unpredictable consequences of
  an imposed alternative

21
Bargaining in the shadow of hierarchy or
deliberating when hierarchy becomes a shadow?

• From this it follows that EU experimentalist
  governance can rarely be understood as regulatory
  cooperation in the shadow of traditional public
  hierarchy
• Bargaining in the shadow of the state (Scharpf,
  Héritier) assumes that public authorities can
  effectively regulate in case of deadlock,
  mismanagement, or policy failure
• But both in the case of the public deliberation
  and penalty default mechanisms, the EU induces
  the actors to provide rules for themselves,
  presumably because it cannot itself provide
  defaults that would do the job as well or better
• Not bargaining in the shadow of hierarchy, but
  deliberating when hierarchy becomes a shadow

22
The Florence Forum revisited

• Competition law as a negative control
  instrument, based on ex post, case-by case
  procedures, which cant produce fine-tuned
  solutions like those developed by the Forum
• Further legislation, even if more stringent and
  detailed, still needs to be interpreted and
  implemented at national level cant address all
  future needs in a rapidly evolving policy area
• Hence continuing need for rule development and
  coordination through networked governance

23
Democratizing destabilization

• Networked, experimentalist governance/DDP not
  intrinsically democratic
• But dynamic accountability through peer review
  subject to penalty defaults has a potentially
  democratizing destabilization effect on domestic
  politics and the EU itself
• Requirement that each national administration
  justify its rule choices publicly, in light of
  comparable choices by others, allows domestic
  actors to contest official proposals on the basis
  of richer information about feasible alternatives
  than would otherwise be available

24
Transparency and participation as procedural
requirements

• Transparency and participation as foundations for
  democratizing destabilization
• EU becoming increasingly transparent and
  participatory
• ECJ case law decisions on administrative
  procedure and access to documents
• Treaty revisions and EU regulations
• Progressive elaboration of legal transparency
  requirements in EU networked governance
• Increasing tendency to establish requirements for
  ensuring active participation by a broad range of
  stakeholders (incl. civil society)
• Directives establishing networked agencies and
  other EU regulatory frameworks (e.g. water)