Transaction Costs, Collective Action and Adaptation in Managing Social-Ecological Systems

Graham Marshall

Institute for Rural Futures, School of Behavioural, Cognitive and Social Sciences, UNE

57th Annual AARES Conference, Sydney, 7 February 2013

• Neoclassical economic focus is on adaptation through market transactions; i.e. private good transactions

• Markets assumed to operate mechanistically... only negative-feedback (diminishing-return) dynamics accounted for... any disturbance leads to a single equilibrium adaptation ... and this adaptation can be evaluated through comparative statics

and stochastic decision analysis

Adaptation and neoclassical economics

• World becoming more complex and uncertain... increasing importance of collective action – collective goods

provision – for adaptation (Adger 2003)... ‘environmental’ problems increasingly concerned with social-

ecological systems (SES)... which are types of complex adaptive systems (CAS) rather than

mechanistic systems• CAS:... behaviour ‘emerges’ as parts adapt to each other and evolving state

of system ... driven significantly by positive-feedback dynamics... path dependencies and lock-in... multiple or non equilibria... unforeseen events (surprises) are ubiquitous, not exceptions... Knightian uncertainty and bounded rationality... stochastic decision analysis is ‘naive’ and may mis-identify optimal

adaptation choices

Adaptation in complex systems

• Path dependencies (PD) mean that... past institutional choices increase costs of present policy reform... present institutional choices increase costs of future policy reforms• PD are surprising, highly sensitive to random events... even though some of the general dynamics might be anticipated

(e.g. vested interests, crowding out, sunk costs, mental models)• Added costs from PD (lock-in costs) depend on the

particular surprises that emerge to necessitate policy reform

... can’t be accounted for deductively/ stochastically; only via induction, bounded rationality and heuristics

Path dependencies

• Governance is crucial for large-group collective action... involves the formation and stewardship of institutions (formal and

informal rules) ... increasing uncertainties have led to a greater focus on ‘adaptive

governance’• Transaction costs are the costs of resources used to:

define, establish, maintain, use and change institutions and organisations; and define the problems that these institutions are intended to solve

• Institutional choices typically also affect technological choices, so the transformation (including abatement) cost impacts of institutional choices also need to be accounted for

Governance, institutions and their costs

• Benefit-cost analysis is ideal, but cost-effectiveness analysis (CEA) is typically more feasible for institutional choices

• Institutional CEA (ICEA) involves comparing the total (transformation + transaction) cost impacts of institutional options

... both costs added and costs avoided

... relative to a common do-nothing scenario• Marshall (2005) built on Challen (2000) to propose an

ICEA framework (ICEAF) distinguishing six types of costs

... where institutional and technological transition costs result mainly from the path dependencies arising from prior institutional choices

Evaluating institutional cost-effectiveness

Table 1: Definitions of the six classes of costs comprising the institutional cost effectiveness framework

Class of costs Definition

Static transaction costs The transaction costs incurred in operating under an institutional option.

Institutional transition costs The transaction costs incurred in effecting change from existing institutional arrangements to a new institutional option.

Institutional lock-in costs The additional institutional transition costs incurred by ‘successor’ institutional options (i.e. those eventually chosen as adaptations, transformations or replacements of the option under consideration) due to the impact on institutional path dependencies of the institutional option under consideration.

Static transformation costs The transformation costs incurred in operating under the technologies or practices that are adopted subject to the influence of the institutional option under consideration

Technological transition costs The transformation costs incurred in effecting change from existing technologies or practices to those adopted subject to the influence of the institutional option under consideration.

Technological lock-in costs The additional technological transition costs incurred by ‘successor’ technologies or practices (i.e. those chosen under the influence of ‘successor’ institutional options) due to the impact on technological path dependencies of the institutional option under consideration.

• Two main institutional options implemented... voluntary water buy-back programs... subsidy programs for water-saving infrastructure upgrades • Economic evaluations to date have applied CEA... accounted only for expenditures on buy-backs and infrastructure

subsidies; i.e. for technological transition costs... buy-backs found to be far more cost-effective• Productivity Commission (2010) ‘recognises that

[infrastructure subsidies] can be seen as the price the Australian Government was prepared to pay to make progress... [They were] needed to convince the states to a truly Basin-wide approach to water planning and to elicit the irrigation sector’s support for increasing environmental water allocations’.

• This ‘price’ refers to institutional and technological transition costs ... why weren’t they accounted for?

MDB environmental water recovery: illustrating the ICEAF

Table 3: Illustrating how the extended cost-effectiveness framework applies to the choice between two institutional options for accumulating environmental water for the Murray-Darling Basin

Cost class Fund a water buy-back program Fund an infrastructure upgrade program

Relevant cost items

Static transaction costs

Added costs incurred (a) in administering the program and (b) by private entitlement holders in transacting with the program

Added costs incurred (a) in administering the program and (b) by prospective infrastructure upgraders in transacting with the program

Institutional transition costs

Added costs incurred in establishing the water buy-back program in respect of: research and stakeholder consultation; negotiating program goals and design; lobbying for, legislating and implementing the program; and negotiating and institutionalising the rules and policies for program administration

Costs of administering privately-owned water entitlements that are avoided by buy-back

Added costs incurred in establishing the infrastructure-upgrade program in respect of: research and stakeholder consultation; negotiating program goals and design; lobbying for, legislating and implementing the program; and negotiating and institutionalising the rules and policies for program administration

Costs avoided in administering infrastructure made redundant by the program

Institutional lock-in costs

Costs added due to emergence of parties with vested interests (e.g. public agencies administering the program, prospective sellers to the program) in opposing changes to the program contrary to those interests

Costs associated with path dependencies arising from existing water entitlement arrangements that are avoided by reallocating entitlements from private to public ownershipa

Costs added due to emergence of parties with vested interests (e.g. public agencies administering the program, irrigators and their input suppliers benefitting from the program) in opposing changes to the infrastructure upgrade program contrary to those interests

Table 3 (cont): Illustrating how the extended cost-effectiveness framework applies to the choice between two institutional options for accumulating environmental water for the Murray-Darling Basin

Cost class Fund a water buy-back program Fund an infrastructure upgrade program

Relevant cost items

Static transformation costs

Nil. (Program is concerned with accumulating environmental water, not with influencing the technologies for applying this water.)

Added costs of operating and maintaining the upgraded infrastructure and any new irrigation technologies adopted due to this upgrading

Technological transition costs

Economic rents from irrigated production that are foregone due to sales of water entitlements to the programb

Costs of operating and maintaining irrigation infrastructure and technologies that are avoided when properties shift to less irrigation intensive (or dryland) production due to sale of water entitlements

Added costs of the infrastructure upgrades supported by the programc

Costs avoided in operating and maintaining irrigation infrastructure and technologies that are made redundant by the program

Technological lock-in costs

Costs associated with path dependencies arising from existing investments in irrigation infrastructure and technologies that are avoided when producers shift to less irrigation intensive (or dryland) technologies due to sale of water entitlements

Costs associated with path dependencies arising from investments in less irrigation intensive (or dryland) technologies that are added when producers shift to such technologies due to sale of water entitlements

Costs associated with path dependencies arising from existing irrigation infrastructure and associated technology that are avoided by upgrading this infrastructure and technology

Costs added due to path dependencies arising from investments in infrastructure upgrades and associated adoption of new irrigation technologies.

• Static transformation costs and technological transition costs reasonably easy to estimate

• Static transaction costs and institutional transition costs are harder to estimate, and often neglected

• Institutional and technological lock-in costs are sensitive to surprises, so cannot be accounted for stochastically

• Ignoring lock-in costs constitutes naive CEA when the choice problem involves a CAS

... such naive analysis risks mis-identifying the most cost-effective institutional option

... this risk is greater the more that society values capacities for adaptation / transformation

Challenges in applying the ICEAF empirically

• Quiggin (2007, 2008) observed that decision makers can use heuristics to anticipate whether an option will lead to a domain of favourable or unfavourable surprises

• The two-stage boundedly rational procedure he proposed for decision analysis of surprise-prone choices was adapted for application to the ICEAF

1. Apply ICEAF to options on a naive basis (i.e. ignoring lock-in costs)2. Select as the preferred option the naively highest-ranked one that

is not expected to lead to a domain of unfavourable surprises for either institutional or technological lock-in costs. If no such option exists, invoke the precautionary principle and retain the institutional status quo.

• A domain of unfavourable surprises is one where lock-in costs are higher than under the institutional status quo

• Marshall (2005, 2013) proposed a research strategy for improving the heuristics used to anticipate surprise domains

A boundedly rational procedure for comprehensive empirical application of the ICEAF

Table 5: Illustrative application of the ICEAF to a case with four institutional options

Institutional options available for achieving the on-ground management target

W X Y Z

Static transaction costs ($K) 100 80 60 200

Institutional transitional costs ($K) 20 40 90 70

Institutional lock-in costs (domain of surprises)a – + + –

Static transformation costs ($K) 150 140 160 90

Technological transition costs ($K) 40 40 100 160

Technological lock-in costs (domain of surprises)a – – + –

Naïve total cost impact of option ($K) 310 340 410 520

Naïve cost effectiveness rank 1st 2nd 3rd 4th

Satisfies precautionary principle? No No Yes No

Preferred option √

a Favourable and unfavourable domains of surprises for lock-in costs are indicated by + and –, respectively.

• Transaction costs of institutional change are significant• CEA of institutional choices should account for

transaction costs as well as transformation (abatement) costs

• CEA studies that account only for transformation costs should be qualified accordingly

• Lock-in cost impacts should at least be assessed qualitatively

... and preferably through the proposed boundedly rational procedure• ‘Ignoring important costs, which are obvious to the

agencies involved, [makes] the economics profession … less credible’ (McCann et al. 2005 p. 528).

Concluding remarks

Paper to be soon published online in

Ecological Economics special issue

‘Transaction costs and environmental policy’

http://www.sciencedirect.com/science/journal/aip/09218009

Transaction Costs, Collective Action and Adaptation in Managing Social-Ecological Systems

Graham Marshall

Institute for Rural Futures, School of Behavioural, Cognitive and Social Sciences, UNE

57th Annual AARES Conference, Sydney, 7 February 2013