PESERA-DESMICE: Land management evaluation model

Table of Contents

1. Authors
2. Summary
3. Acronyms
4. Introduction
5. Coarse-scale model description
   1. The PESERA model: summary
   2. Alternative approaches to assessing soil erosion
   3. Process representation in PESERA
   4. Extensions of PESERA
   5. Adapting PESERA
   6. Assessment at catchment scales
6. Cost-effectiveness modelling
   1. Step 1
   2. Step 2
   3. Step 3
   4. Step 4
   5. Step 5
   6. Step 6
   7. Step 7
   8. Step 8
   9. Step 9
   10. Step 10
   11. Step 11
   12. Step 12
7. Agent-Based modelling in Spain
   1. Phase 1
   2. Phase 2
   3. Phase 3
   4. Phase 4
   5. Phase 5
   6. Phase 6
8. Input-Output modelling in Spain
   1. Model construction
   2. Scenario analyses
9. References

Cost-effectiveness modelling

This section outlines the socio-economic modelling approach developed for application across all DESIRE study areas. The approach is developed to integrate with the PESERA model, which is used to evaluate the biophysical consequences of alternative remediation strategies. According to the WOCAT terminology applied in WB3, remediation strategies consist of technologies and approaches. A technology can consist of a single or multiple of four types of measures: structural, vegetative, agronomic and management measures, respectively (WOCAT, 2007). The cost-effectiveness modelling methodology consists of twelve steps which will be described below. These twelve steps form the logical modelling sequence and include both steps delivering relevant intermediate output and technical steps to allow progression to subsequent analyses. The intermediate outputs correspond to the following topics and methodologies:

1. Applicability limitations and spatial variation of investment costs (steps 1 and 3)
2. Evaluating effectiveness of technology investments using cost-effectiveness and financial cost-benefit analysis (steps 4-7)
3. Adoption of technologies and diffusion of innovations (step 8)
4. Economic (including wider economic effects) Cost-Benefit Analysis (steps 10-11)
5. Policy scenario analysis (step 12)

Before describing the twelve steps, we will first touch upon some key-issues of the above points. The first point deals with the planning and design of conservation technologies. All technologies, whether based on indigenous knowledge and dating back centuries or the result of recent scientific experimentation, are designed for specific environmental and socio-economic conditions. The design of conservation technologies is due to its practical aspect probably the most studied aspect of land management. Much of the design literature can be associated with large-scale government-led project interventions initiated to tackle land degradation problems, as guidelines and manuals needed to be prepared for training field technicians and providing them with rules on how to lay out selected measures (Wenner, 1981; US Bureau of Reclamation, 1987; Alaya et al., 1993; WDLUD, 1995). Of later date (except perhaps early anthropological and historical accounts) are contributions documenting indigenous land management practices (e.g. Reij et al., 1996). Many of those studies were inspired by the apparent lack of effectiveness of large-scale soil and water conservation campaigns that were rolled out across the developing world - one reason for which was concluded to be a lack of fit of the imposed 'solutions' to the realities of the farmers on who's land they were implemented (Hudson, 1991). Of recent origin is the idea to share success stories in conservation across the globe to increase chances of cross-pollination (matching tradition with innovation). The WOCAT initiative (www.wocat.net) is the most well-known of such network efforts, and its approach was also adopted by the DESIRE project (Schwilch et al., 2009).

A first step in tailoring conservation technologies to a specific environment is to establish the preconditions necessary for their implementation. WOCAT uses lists of environmental and socio-economic variables to label characteristics of localities where technologies and approaches followed for their dissemination have been effective. As the WOCAT methodology is designed to facilitate knowledge exchange, such relatively broad labels suffice. However, in spatially explicit modelling, a further refinement of applicability is necessary.

Cost-effectiveness analysis and cost-benefit analysis are economic evaluation methods used to select the best among several alternatives. In the case of cost-effectiveness analysis, (biophysical) effects resulting from the alternatives considered are evaluated as is, while cost-benefit analysis implies that all effects are translated in monetary units. They are distinct methods in that cost-effectiveness analysis needs an explicit (policy) objective against which to evaluate performance of alternatives, whereas cost-benefit analysis will select the best alternative given a series of cash flows of monetary costs and benefits for each of the alternatives and a discount factor. Cost-effectiveness analysis has been criticized for being arbitrary with regard to the subjective element of setting targets (de Graaff, 1996), while cost-benefit analysis has generated discussion over the possibility and desirability of attributing monetary value to all impacts of any government initiated project (e.g. intangible effects on biodiversity, human lives saved, etc.) and philosophical and technical discussions over what discount rate should be applied (e.g. Pearce and Turner, 1990; Arrow et al., 1996; Almansa Sáez and Calatrava Requena, 2007). The latter is especially relevant when a societal perspective is taken (economic Cost-Benefit Analysis (CBA), as opposed to financial CBA), and particularly when decisions have to be made about environmental sustainability (discounting is essentially incompatible with long-term decision-making, leading to discussions over inter-generational equitability).

Even for the relatively straight-forward application of financial CBA, more frequently than not, ex-post analyses have shown that predicted rational adoption behaviour (based on profit maximising) has more often than not poorly been correlated to actual land user's behaviour. Land users face several challenges that are either difficult to incorporate or have often been neglected in financial CBA. Among such challenges are elements of risk (e.g. land tenure, climate, pests and diseases, price fluctuations), lack of access to knowledge, labour and/or capital resources, and socio-cultural and psychological factors (e.g. ineffective decision-making structures, power relations, inappropriate technology, cultural norms and values). There is a large body of research on the factors influencing adoption of soil and water conservation measures, which appears to come up with context-specific determining factors (Lapar and Pandey, 1999; Shiferaw and Holden, 2001; Tenge et al., 2004). Several studies also indicate differences in factors determining initial and sustained adoption (Paudel and Thapa, 2004; Amsalu and de Graaff, 2007). Apart from the question what determines adoption, the issue how adoption processes take place is a pertinent one, dealt with broadly in the social theory of diffusion of innovations (Rogers, 2003). If we consider a sustainable land management technology to be an innovation, how will it, after its introduction, disperse among agents, in time and in space? Rural sociologists have since the 1940s extensively studied the diffusion of innovations in agriculture (including the pioneering study by Ryan and Gross, 1943), but its integration with GIS offers the potential to put much more emphasis on the spatial dimension then has hitherto been possible.

Notwithstanding the difficulties associated with the methods introduced above, policy-makers face an acute level of urgency in dealing with land degradation. Important questions are how to motivate land users to adopt more sustainable production methods, what policy instruments to use and where to focus attention. These complex issues can only be resolved by making assumptions and simulate decisions in scenario studies using modelling approaches. Hence, we depart from the assumption that although other factors may limit adoption, a positive expected return to investment (as calculated with CBA) is a precondition for a technology to be taken up by land users. By applying CBA, an upper boundary for potential adoption can thus be inferred. More realistic adoption dynamics can subsequently be included by using additional models (e.g. ABM, Section 4.1). Policy-makers may employ a range of instruments to stimulate adoption; however, their criteria to do so will depend on various measures of cost-effectiveness. By combining cost-effectiveness analysis, CBA and ABM in one model (or, as in our case, several coupled models), it becomes possible to evaluate a multitude of 'what if' options. As validation of scenario studies is notoriously difficult, the possibility to evaluate the effect of assumptions made is an essential feature of complex models. Moreover, as our models are embedded in a participatory approach (information will be taken up and results evaluated in workshops with stakeholders), they will receive an additional validation or reality check.

The following text describes each of the 12 steps in the socio-economic modelling approach developed for application across all DESIRE study areas. For clarity, a summary is provided at the end of each step, showing the required data inputs and intermediary outputs that are used in other steps. Figure 3.1 presents a graphical representation of the cost-effectiveness model and its interrelations with other models.

Figure 3.1: The twelve steps of cost-effectiveness modelling and interrelations with other models and project work blocks