Desktop version

Home arrow Economics arrow Disaggregated impacts of CAP reforms : proceedings of an OECD workshop.

Source

Methodology of the study on the impact of modulation4

The methodological approach that has been taken to understand the impact of modulation is based on several different types of analysis, which can be divided into two broad categories: a modelling approach and a non-modelling approach (Nowicki et al., 2009). The modelling approach allowed for results to be generated on impacts across the EU27, and for simulations of the likely changes of these impacts under different rates of modulation, while the non-modelling approach allowed for more qualitative, contextspecific insights into the impacts of modulation to be made. The use of models also permitted an exploration of any differences that might emerge from changes to rules relating to franchise levels, co-financing requirements, or allocation of funds within Pillar 2 to specific measures, albeit based on a set of generalised assumptions.

The modelling approach consists firstly of a custom-built budget model, which allows the transfers of money involved from the national cuts in the Pillar 1 through to the expenditure for each measure within member states’ Rural Development Programmes to be tracked. Secondly, a suite of economic models place the Pillar 1 reductions and the additional budget available for Pillar 2 measures within the framework of the world economy, from both a general and partial, or sector-specific (agriculture), perspective. Finally, a land-use model attributes changes in land-use that are calculated by the economic models to particular areas, on the basis of a 1 km grid covering the European Union. The use of economic models to understand the impact of Pillar 2 expenditures was carried out for the first time, and was informed by insights acquired from the nonmodelling approach. The non-modelling approach included a literature review, case studies undertaken in eight member states, questionnaires to member state authorities for agriculture and rural development, and an assessment of standard indicators compiled within the Common Monitoring and Evaluation Framework for EU rural development policy.

A number of difficulties were encountered in identifying the precise impacts of compulsory modulation on the range of themes addressed by this study, some methodological, and some relating to data availability. These are to be expected in a relatively new policy area and included: the lack of empirical studies (ex post), especially on the effectiveness and efficiency of Pillar 2 measures, lack of data, the use of analytical tools that were not in every case specifically designed to accomplish the task required, and the need for complementary research in a context where time and human resources were limited. The quantitative modelling approach was therefore limited to ex ante analyses and based on strong assumptions.

Figure 15.1 demonstrates that the modelling approach is integrally associated not only with the budget model and the other economic and land-use models, but also with the case studies, and the modelling approach also draws on a literature review in order to investigate the exogenous parameters and multiplier coefficients that are used in the modelling approach. Where such information is not available, assumptions with regard to parameters and multipliers have to be made by the modellers, on the basis of the best available expert knowledge. In order to model the economic and environmental impacts of modulation, it is necessary to find a means of linking agricultural commodity parameters with regional / territorial aspects. The global economy-wide dimension is covered by the economic model, LEITAP (Francois et al., 2005; Meijl et al., 2006; Eickhout et al., 2007; Banse et al., 2008). ESIM provides more agricultural detail for the EU25 countries (Banse and Grethe, 2007), CAPRI distributes this impact to the regional (NUTS 2) level (Britz et al., 2008), and FES to the farm level. Dyna-CLUE provides a detailed analysis of land cover change, thereby giving a spatial representation of the economic modelling outcomes (Verburg et al., 2008; Verburg and Overmars, 2009).

Models are shown with their output contributions in this study. Rounded fields indicate national levels and squared fields regional levels. The budget model provides basic information to all models and to the case studies, which, together with literature provide the basis for the assumptions regarding the parameters for human and physical capital that are used in the models.

Figure 15.1. Quantitative impact analysis

Analysis of modulation within the modelling framework

Modelling modulation has been made through a set of linked models. The modelling was carried out in two steps: first, Pillar 1 was reduced, and second, the money was introduced into the Pillar 2. The first step is usually quite straightforward (Table.15. 1), with the main challenge being the modelling of decoupled payments. The second step is more complicated since modelling Pillar 2 has never been done before. Introductory comments regarding the treatment of rural development measures are provided below (Table 15.2). One important aspect of agriculture is its contribution to public goods. The models used in this study are not suited for analysing this aspect, and the current literature in the field does not allow for any consistent implementation in modelling policy interventions.

Table.15.1. Treatment of Direct Payments (Pillar 1) in models

Treated in Model

Implementation of Direct Payments

LEITAP

Farm payments are implemented as primary factor payments in the various agricultural sectors. Coupled payments are directly coupled to sectors. Decoupled payments are implemented as an equal payment rate to land in all eligible sectors, and therefore do not provide an incentive to switch between eligible sectors and between production factors used within the eligible sectors.

FES

Farm payments are directly calculated and implemented at farm level.

CAPRI

Effects of changes in farm payments are analysed at the regional farm and sector level, distinguishing between a large number of premium types. Decoupled premiums such as milk and sugar premiums are distributed over the eligible crops of the regional farm. Coupled premiums are linked to agricultural activities at the regional level.

Table 15.2. Treatment of rural development measures1 in quantitative models

Treated in model

How implemented

(information needed from other models/case studies)

01 - Human Capital Investment [111-115, 131-133]

LEITAP

Payments influencing the total factor productivity in agriculture.

Rate of return on investment is 40% (Evenson, 2001).

Deadweight loss assumed to be zero (sensitivity analysis is done with 25% deadweight loss).

CAPRI

Via link with LEITAP.

FES

Payments on investment at farm level.

02 - Physical Capital Investment [121-126]

LEITAP

Payments which influence the total factor productivity due to capital investments in all agricultural sectors.

Rate of return on investment is 30% (Wolff, 1996; Gittleman, ten Raab and Wolff, 2006).

Deadweight loss is assumed to be zero (sensitivity analyses done with 25% deadweight loss).

CAPRI

Via link with LEITAP.

FES

Payments on investment at farm level.

03 - Less Favoured Area (LFA) Land Use Support [211, 212]

LEITAP

Income payment linked to land in agricultural sector. FADN data are used to distribute payments across sectors.

CAPRI

Regional direct support. Distribution over sectors and regions based on FADN data and CLUE results.

FES

Farms receive LFA or mountain area support when they are in these areas (income support).

Dyna-CLUE

LFA support adds to the relative preference for the location for arable land or grassland (only for current agricultural land within LFA regions).

04 - Natura 2000 [213]

LEITAP

Income support linked to land in agricultural sector. FADN data are used to distribute payments across sectors.

CAPRI

Regional direct support. Distribution over sectors and regions based on FADN data and CLUE results. Conditional on extensive technology being used.

Dyna-CLUE

Agricultural land in Natura 2000 areas receives a higher relative preference (as compared to no support) for agriculture (only for current HNV agricultural land within LFA regions).

05 - Agri-Environment

measures

[214-216]

LEITAP

On the one hand, income support linked to land in agricultural sector, and on the other hand, a yield and labour productivity loss. FADN data are used to distribute payments across sectors.

CAPRI

Regional direct support. Distribution over sectors and regions based on FADN data. 50% of the support directed towards FADN ‘TF8' farm types 1, 2, 3, 4 and 8 is conditional on extensive technology being used; for the remainder, extensive as well as intensive technology is eligible.

FES

Payments linked to land.

  • 06 - Forestry [221-227]
  • 07 - Diversification [311-313]
  • 08 - General rural development [321-323, 331,341]
  • 09 - LEADER [411-413, 421,431]
  • 10 - Technical assistance [511,611]

LEITAP

Investment support for non-agricultural activities that increase productivity. Rate of return on investment is 30%. Deadweight loss is assumed to be zero (sensitivity analysis is done with 25% deadweight loss).

CAPRI

Via link with LEITAP.

Dyna-CLUE

For forestry: conversion of arable land to forestry or grassland in erosion sensitive areas is stimulated by lowering the relative preference of current arable land in erosion-sensitive areas.

1. The RD measure numbers are indicated between square brackets [#].

Following the intervention logics for the rural development measures, the economic models and the land use model employed are able to perform a series of analyses in order to provide insight on the thematic issues in the study, some of which are elaborated upon in Section 4 below. These analyses cannot reasonably be performed separately for each of the 46 rural development measures, and are thus grouped according to fundamental similarities in the economic mechanisms and how these are handled by each of the models. As an elaboration of this principle, Table 2 presents the groupings of rural development measures, the models that are used for their analysis, and the relationships between the models.

 
Source
Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >

Related topics