This case study is on a dryland salinity credit trade trial that is one of eleven trials implemented under the Commonwealth Market-Based Instruments Pilot Program, round one. The trial, was developed in response to a specific request by the round one trial selection panel for a credit trading trial focussed on dryland salinity. It was implemented in the upper Bet Bet subcatchment of the Loddon River in north central Victoria.

The fundamental goal of the trial was to set an aggregate threshold for groundwater recharge volumes for each trial participant. Participants can then meet their goals through credits resulting from land management outcomes resulting in recharge reduction or through credit trade amongst participants. This allows individual under-performance if it is compensated by over-performance elsewhere.

This case study contains:

• a brief synopsis of trial work to design salinity impact assessment and accounting protocols, and design credit trade policy implementation arrangements;
• a description of the outcomes of implementation of the pilot to date including:
  • summary of the level of on-ground works undertaken;
  • outcomes of participant performance audits; and
  •  participant debit credit positions as of early 2006;
• an evaluation of Bet Bet Creek, Loddon River and River Murray flow and salinity impacts expected to result from trial implementation;
• An assessment of the benefits and costs expected to result from trial implementation; and
• A summary of key outcomes and broadly applicable learnings from the MBI trial.

Trial design phase outcomes

The first step in this trial involved identifying potential impediments to cost effective, environmentally reliable and politically feasible implementation and trial design options to overcome impediments. The process is described in more detail in Milestone 5 report (Connor, et al. 2004).

Efforts to identify potential impediments to effective on-ground policy design for further investigation involved:

• A review of relevant environmental economics and auction theory literature;
• Local background research and stakeholder input;
• Integrated biophysical/economics modelling.

Six key impediments were identified which could prevent cost-effective and environmentally effective outcomes in the implementation phase if not addressed:

• A lack of fully articulated and enforceable property rights arrangements;
• A lack of a performance based incentive to reduce recharge;
• Capital/cash-flow/time preference constraints;
• Costly information;
• The likely presence of thin markets;
• Non-market motivations.

A social survey was conducted to gain an understanding of what might motivate local landholders to participate in a recharge credit trade trial and what kinds of implementation design might be most effective given community social character. Key findings were that:

• Landholders in the region are quite traditional in that they tend to mostly use similar sheep grazing based farming systems. There was little use of computer record-keeping and decision-tools, little adoption of non-traditional selling methods, and little use of formal farm business planning in the survey sample. This may mean that the majority of landholders lack the skills, inclination and familiarity with markets and trading to participate effectively in an individualistic trading scheme.
• The most probable motive for participation in a salinity recharge-credit trading scheme for landholders in the Upper Bet Bet region is the community spirit and Landcare ethic, and the high degree of acceptance that while salinity is a problem in the district, it can be managed on-farm.

More detail on the survey methodology and results can be found in the Milestone 5 report (Connor et al., 2004) and in Thomson (2004). Conclusions from survey findings were reinforced in the field demonstration with Bet Bet landholders. In these field trial credit trading was simulated using an of experimental economics representation of farming and credit trade economics in the catchment (Connor et al., 2004).

Experimental economics was used to further test the significance of several potential impediments to effective on-ground trial implementation and policy designs to overcome impediments. The experiments took place in the experimental economics laboratory at Griffith University. They involved developing an experimental setting representative of the economic decision making and trading environment landholders in the upper Bet Bet would face under a range of recharge credit trade policy implementations.

Based on scoping meetings with landholders, sociological survey results, field demonstrations and literature review it was decided to use experimental economics to:

• Test the potential of a cap and trade system to reduce cost of capping recharge;
• ompare cost effectiveness of alternative formats of cap and trade;
• Test the cost and recharge reduction reliability of alternative tendering system designs to induce landholders to enter into contracts to provide recharge reduction; and
• Test the potential of collective “social payments” as an incentive to increase willingness to provide recharge reduction.

Key conclusions from the experimental economics (discussed in more detail in Milestone 5 report (Connor et al. 2004), and in Ward, Connor and Tisdell (2006) were that:

1. The more complex market information requirements associated with open call markets as a credit trading mechanism may not result in reliable and substantial trade or cost savings if used in the on-ground trial. The gains from trade observed in experiments with this design were unstable and did not converge toward a predicted, stable equilibrium with repetition. If implemented on the ground, participants could well find the system hard to understand and this could reduce market exchange.

2. The simpler information requirements associated with a closed call market structure for trading recharge credits had more reliable and predictable outcomes. While it took several repetitive periods, experimental outcomes converged toward theoretical “frictionless market” predictions. The results suggest that there is a reasonable chance of reliable, cost saving trade occurring if a closed call trading format is used in the onground trial implementation. The gains from trade from a closed call market were significantly (p < 0.05) higher than a control no trade and open call market treatment.

3. Under current property rights arrangements, land holders are not required to manage recharge or provide water quality to a set standard for effected downstream water users. Changing fundamental property rights to establish legally enforceable obligations to provide water quality across the region was not politically feasible for the trial. The solution to this property rights impediment was to establish contractual obligations for willing landholders to provide recharge reduction in exchange for compensation.

Experiments tested the cost effectiveness and recharge reduction level likely with alternative tendering approaches to establish recharge reduction obligations in the onground trial. An important objective was to test capacity of alternative auction formats to truthfully reveal individual costs of providing recharge reduction. Two tender structures were compared in the experiments: a discriminant price and a uniform price auction.

Auction theory suggests that given the imperfectly competitive market conditions likely in the trial setting, a discriminant price auction is likely to be less incentive compatible than a uniform price auction. Accordingly, the result of uniform price auction should be lower costs for the agency conducting the tender. Results found that this was the case for the simulated upper Bet Bet setting. The conclusion was that given the thin market and the complex informational processing required of experimental subjects in discriminant price auction treatments, participants had difficulty finding optimal solutions.

4. Results showed that in experiments involving a discriminant price tendering format, some subjects had trouble recognising their optimal strategy. This enabled others to recognise and exploit an opportunity to gain large profits. This may have been a result of the thin market simulated in the experiments combined with the complex information processing required of subjects to understand opportunities to profit with the discriminant price tender format. Presumably as a result of the constant price information revealed with the uniform price auction tendering format, subjects were not able to seek excessive profit as successfully as in the discriminate price tender experiments. The cost per recharge unit was significantly higher (p < 0.05) in the discriminant compared to the uniform auction treatment.

5. The results indicate that acknowledging and accounting for the thin markets characterising the Bet Bet catchment will be important for the successful implementation of the trial. The experimental results indicate that a uniform price tender system may provide more cost effective and reliable recharge reduction compared to a discriminate price tender in the initial process of establishing contracts to provide recharge reduction.

Thus, to encourage high levels of commitment to recharge reduction obligation, a uniform price auction format was found to be preferable to a discriminate price auction format.

6. The final set of experiments conducted tested the potential of a form of collective performance based payment. In the experiment part of the payment received by participants was contingent on the group as a whole achieving a defined level of aggregate recharge reduction. The results of these experiments support the conclusion that using such a collective performance incentive payment to establish initial recharge reduction obligations in the on-ground trial may increase participation and reliability of recharge reduction.

Implementation approach

Recharge outcome monitoring and credit accounting protocols were established and a legal agreement that forms the basis for the on-ground pilot trial was drafted. Both are described in more detail in this report.

The results of work on salinity impact assessment and accounting was a set of monitoring protocols and look-up tables that relate monitored land management outcomes to estimated salinity impact. The monitoring approach (described in more detail later in this report) involves assessing plant cover achieved with straightforward field sampling techniques.

Lookup tables to relate expected recharge reduction to monitored cover outcome given to crop type, landscape position were developed based on plant, soil water balance modelling.

The monitoring, credit accounting approach developed allows for policy incentive to manage for recharge reduction as more credits result for participants who are more successful at planting establishment.

The contract is a commitment to either meet a salinity recharge obligation or offset any recharge in excess of this obligation through the purchase of tradeable salinity recharge credits. Those who exceed this obligation will be issued salinity recharge credits and be able to sell them to those who prefer to meet their obligations through cooperation with others. If land management change and trading occurs to the extent necessary to meet a defined collective aggregated recharge reduction target, all participants receive a bonus payment.

The trial implementation involves five stages:

1. Negotiation of landholder contracts
2. One year of land management change
3. Recharge audit Repeat steps 2-4 for three trial years
4. Trading among landholders
5. Bonus paid if trading clears collective obligation at trial conclusion (end of 3rd year)

The main features of the landholder agreement was:

• A voluntary multiple year agreements with landholders for management changes that reduce recharge to agreed levels in exchange for payment. Three types of land management change are possible under the agreement:
  • New pasture establishment and management;
  • A farm forestry establishment, management; and
  • Native vegetation establishment and management.
• An establishment payment with a level that varies based on estimated level of annual recharge reduction and permanence of the management change undertaken and location, disbursed at a $ per ML payment rate.
• Annual “performance payments” based on monitored ground cover and look-up tables relating monitored cover to model estimated recharge reduction given on crop type and location.
• Estimated recharge achieved will be compared with the level of recharge obligation participants contract to provide. To qualify for annual performance payments available at the end of project year 2 and 3 (trial commencement), participants will require credits equal to exceeding their recharge obligation level. The credits can be obtained from management on their farm or trade.
• To ensure that seasonal conditions do not result in an infeasible outcome, banking and borrowing of credits is allowed so that deficit in one year can be made up if the deficit is redressed through increased on-ground work or credit trade before the end of the trial.
• A collective agreement provision to achieve recharge reduction targets will require that funds be withheld unless landholders can reach consensus to meet a minimum aggregate level of reduction.

Implementation results

Implementation on the ground beginning in early 2005 resulted in enrolment at 22 sites to provide establishment of native vegetation on 103.4 hectares and new perennial pasture establishment on 12 sites totalling 257 hectares.

Audits on cover on all projects were undertaken in late 2005 or early 2006. The results for native vegetation sites are shown in the first table on the next page. They lead to the prediction that credit deficits are likely on 5 of 22 native vegetation sites, surpluses are likely on 15 sites, and 2 sites are predicted to break-even.

A predicted overall net surplus of credits on all revegetation sites of 78.4 credits resulted. The results for pasture sites are shown in the second table on the next page. The prediction based on these results is that deficits are likely on 2 of 10 sites audited, surpluses are likely on 6 sites and two sites are predicted to break-even with an overall net surplus of credits on all pasture sites of 8.2 credits.

Under terms of the contracts with participants, they are allowed to trade credits at any time during the three year on ground implementation phase. No trading has taken place to date.

However, given that some sites are in surplus and some in deficit, and given that there are incentive payments at the end of year two and three for any participant who meets their obligation through outcomes of land management on their property and credit trade, trades are likely over the next two years.

Cost benefit analysis

A benefit cost analysis (BCA) of the trial was conducted based on assumptions about effectiveness of currently established cover. BCA outcomes (described in detail in this report) lead to the conclusion that benefits of avoided salinity damage are likely to be substantial, $246,334 on a net present value basis. These estimated benefits exceed the cost of payments to trial participants equal to $119,775.

However, the revegetation resulting from trial implementation is also expected to reduce flow to Bet Bet Creek, the Loddon River and ultimately, the River Murray. BCA results are sensitive to assumptions about the opportunity cost of flow reduction as well as assumption about carbon sequestration benefits of revegetation.

With a high estimate of the opportunity cost of flow reduction (a $50/ML/year temporary water market price), a positive net benefit only results assuming that the dollar value per tonne of carbon sequestration benefits is equal to the current European carbon credit trading market price (29 Euro/tonne of carbon).

If a high opportunity cost of flow reduction and low value of carbon benefits is assumed, the trial is estimated to result in net cost rather than net benefit.

Higher level learnings

Several lessons emerge from this trial that have implications beyond the trial case study.

1. One widely applicable lesson from this pilot is that it is feasible to use individual legal agreements to provide advantages of tradeable credit policy. Tradeable credit or cap and trade policies require individual limits on allowable emissions. Often, in diffuse source emissions setting no limits exist. Defining individual emission limits for all in such settings would involve fundamental changes to legal definitions of environmental property rights.

Given the potential costs to landholders to comply with limits and the transactions and administrative cost it is typically politically infeasible to set individual emission limits for all.

Using a legal agreement, as opposed to legislated basis for environmental limits represents a way to partially overcome more fundamental political economy and transactions cost challenges to more fundamentally defining individual emissions limits.

In Victoria as elsewhere in Australia the only property right with respect to recharge emissions is an implicit right to emit recharge. In the trial individual legal agreements provide obligations to meet environmental limits. The limits are defined as an obligation to provide an agreed level of credits through results of own land management and credit trade. This allows realisation of some of the benefits of an emission trade policy where there are no overarching legal limits on emission.

2. A second lesson is that dynamic incentive can be built into both credit and auction policies with contractual arrangements. Dynamic incentive is created in policy when the policy creates ongoing motivation to continuously seek out lower cost innovative ways to meet environmental goals (Young 1997, Tietenberg and Johnstone, 2004).

Recent evaluations of experience with incentive based environmental policies in the United States and Europe confirm that that tradable credit and charge approaches have significant capacity to produce dynamic incentive and lead to lower than anticipated policy compliance cost (Harrington et al., 2004). A prerequisite for creation of dynamic incentive is a feedback mechanism where parties subject to policy make repeated production decisions and production and policy compliance costs that result are revealed.

The approach trialled creates dynamic incentive contractually in a way that can be applied in other settings. The dynamic incentive is created by specifying :a monitoring approach, a basis for relating monitoring outcome to performance, repeated performance monitored on intervals, and a payment schedule relating level of repeated payments to monitored outcomes.

The trial approach relates groundcover monitoring outcomes to recharge credits based on plant soil water balance models. Incentive for dynamic efficiency and trade is created through second and third year payments in the three year trial, contingent on achieving agreed credit levels through monitored outcome and credit trade.

This is significant because practice in most auctions and other payment policy to date is payment on input or practice implementation rather than monitored outcome. The monitored outcome basis for this trial is replicable in other dryland farming settings in Australia and provides a basis for tradeable credit and other performance based policy.

3. A collective performance incentive feature is included in the trial. The basic idea is that part of the total payment is only dispersed to participants if the sum of individual outcomes reaches an aggregate pre-specified level. The collective performance outcome based payment appeared to increase uptake of incentive payments in the Bet Bet. A third key lesson from this trial is that collective performance payments hold promise as way to increase uptake of incentive payments for on-ground works in similar settings where environmental action is a high priority but voluntary participation in payment programs is low.

Revegetation in the Upper Bet Bet sub-catchment of the Loddon where this trial took place has been a high policy priority for several years prior to this trial. This is because salt loads from drainage in the area are the highest per ML of drainage of all Loddon Sub-catchments.

A result of continuing focus on the area was that most ready adopters appeared to have already undertaken substantial revegetation at the trial outset. Only 5 hectare of revegetation in exchange for payments was undertaken in 2004, the year prior to the trial. In contrast, there has been a high level of enrolment in the trial with 103 hectares of revegetation and 257 hectares of perennial pasture establishment.

The trial outcome suggests that the group performance incentive payment feature and local Landcare administration of the trial increased voluntary enrolment. The significance is that the approach could effectively increase program uptake in other targeted environmental priority areas where enrolment rates in on-grounds programs are insufficient to satisfy natural resource management targets.

There is a considerable body of economic research supporting the notion that incentives for collective performance hold promise as an approach to reducing diffuse source emissions.

Ostrom (1998); Gintis (2000); Tisdell et al. (2004) report willingness to diverge from individualistic profit maximizing behaviour for the public good in small, cohesive communities.

Though other research suggests that a free riding problem can arise with collective incentive policy where there is too little individual incentive and individual behaviour is not easily observed (Poe et al. 2004).

This trial is one of the first demonstrations of the potential of policy with a collective outcome basis implemented in an on-ground setting that we are aware of. How successful a collective performance incentive approach is likely to be in other settings remains an open question.

Results of the social survey carried out in the trial region indicate a high level of social cohesion, a high level of membership in Landcare, the organisation who administered the pilot, and a strong belief that on-farm action can improve salinity outcomes. We suspect that these characteristics of the community are a least a partial explanation for the success of the collective performance incentive in the Trial. The results may not be replicable in settings where there is less social cohesion.