Evaluation Report Series
Evaluation Report 3
The Value of Prospective Improvements from Research and Development in the Victorian Grains Industry
December 2001 Agriculture Division
Authors: Rukman Wimalasuriya, Dale Grey, Bill Fisher, Phil Haines, Mark Eigenraam
Other reports in the Evaluation Report Series are:
Department of Natural Resources and Environment 2000, "Beneficiaries and Funders of Research and Development in Agriculture: Science Technology and Innovation Initiative 1999/2000", Evaluation Report No: 1, Melbourne, June.
Department of Natural Resources and Environment 2000, "The Value of Prospective Improvements from Research and Development in the Victorian Dairy Industry", Evaluation Report No: 2, Melbourne, August.
© The State of Victoria, Department of Natural Resources and Environment, 2001
ISBN: 0 7311 5196 8
In 1999, the Victorian Department of Natural Resources and Environment's Agriculture Division implemented a rigorous economic evaluation system for projects funded under the Victorian Government's Science, Technology and Innovation initiative.
Our aim was to improve the quantity and quality of information available to support investment and planning decisions about research and development projects and to help ensure that public resources continue to be allocated to areas of highest return. The evaluation system also makes research and development expenditure more transparent and accountable.
Extended in 2000 to projects funded under the State Government's Naturally Victorian initiative, the system uses contemporary economic theory to quantify the potential benefits from alternative areas of research expenditure, either within or between industry sectors. Computer models developed by the Department's Economics Branch are used to estimate productivity improvement from different research and development projects. The economic benefits from an improvement in productivity are then estimated. This report shows the results for grains research.
Other reports are planned and will be progressively released.
The work reported here is part of a wider economic evaluation system for research and development that has been developed for the Department of Natural Resources and Environment's Agriculture Division. Described in Economics Branch (1998b), this system draws on the research and experience of economists from the Department of Natural Resources and Environment (NRE) and external organisations.
Loris Strappazzon, Gary Stoneham and David Goldsworthy provided helpful advice and comments on earlier drafts.
The authors also wish to acknowledge the help of Annette Wilson (NRE) with statistical data.
Table of contents
- Executive summary
- 1.1 Introduction
- 1.2 Overview of the Victorian Grains Industry
- 1.2.1 The importance of productivity in promoting economic growth
- 1.2.2 Productivity in the grains industry
- 1.4 Resource Allocation in NRE
- 2 A method for estimating the benefits of research
- 2.1 The Farming Systems Models
- 2.2 Targeting Key Areas of NRE's Grains Program
- 2.3 Applying Model Results to the Regions
- 2.3.1 The Wimmera Region
- 2.3.2 The Mallee region
- 2.3.3 The North East region
- 3 Estimating the gross benefits
- 3.1 Key Project: Variety Development
- 3.1.1 Awheat variety tolerant to Pratylenchus nematodes
- 3.1.2 Chickpeas resistant to Ascochyta blight
- 3.1.3 Reducing grain protein content of barley in the Mallee
- 3.2 Key Project: Farming Systems
- 3.2.1 Rotations with grass-cleaning of the pasture phase in the Mallee
- 3.2.2 Changing from pasture-fallow to continuous cropping in the Mallee
- 3.2.3 Reducing the impact of weeds on cereals in the Mallee
- 3.3 Key Project: Natural Resources
- 3.3.1 Changing from subclover to lucerne-based crop rotations
- 3.3.1 Changing from subclover to lucerne-based crop rotations
- 3.4 Key Project: Service Delivery
- 3.4.1 Increasing the grain protein content of barley in the Wimmera
- 3.4.1 Increasing the grain protein content of barley in the Wimmera
- 3.5 Summary of Estimated Productivity Improvements
- 3.6 Summary of the Value of Productivity Improvements
- 3.7 Discussion
- 4 Benefit-cost analysis of research projects
- 4.1 Canola Breeding Program in Victoria
- 4.1.1 Introduction
- 4.1.2 Details
- 4.1.3 Results
- 4.2 Breeding A Chickpea Variety Resistant to Ascochyta Blight
- 4.2.1 Introduction
- 4.2.2 Details
- 4.2.3 Results
- 4.3 Changing from Subclover to Lucerne-based Crop Rotations
- 4.3.1 Introduction
- 4.3.2 Details
- 4.3.3 Results
- Appendix 1
Using NRE's Farming Systems Models to Estimate Productivity Change
- Appendix 2
The Impact of Technology Change
With many competing demands for limited public funds, choices must be made between alternative projects and activities. To assist with the resource allocation process, the Agriculture Division of NRE has developed an economic evaluation system for proposed and existing research and development (R&D) projects. This system is designed to improve the quantity and quality of information used for making decisions about allocating funds for R&D.
Four broad categories of information are needed for R&D allocation decisions: the quality of the science, the quality of project management, the potential returns from R&D projects and the role of government. This report focuses on the potential returns from R&D projects.
Estimating the potential return from an R&D project involves estimating the change in on-farm productivity arising from that project. Productivity is essentially the amount of outputs (goods and services) produced from a given amount of inputs. Improvements in on-farm productivity can come from many sources, including improved genetic material, better rotations, new agronomic techniques and proper timing of farm activities.
Increased productivity often means lowering the cost of each unit of production. For example, a new, high-yielding wheat variety may increase on-farm productivity by producing more grain for the same cost as an existing variety. This also implies that it may be more profitable for one type of farm enterprise to expand at the expense of others. Whole-farm models that are representative of farming systems within a region can take account of on-farm input changes within a farm enterprise and output substitutions between farm enterprises that occur as a result of introducing a more efficient production technology.1
This report uses whole-farm computer models to estimate productivity improvements for areas of research and development within NRE's Grains Industry Strategy. Potential improvements in productivity have been estimated from advice provided by NRE scientists for eight areas of research as shown in Table 1. These improvements will have the effect of reducing the cost of production. Thus the economic value of each potential improvement is estimated by multiplying the productivity improvement by the total variable cost of the relevant farm types of the target region. These estimates are the maximum annual gross benefits that could be expected, before project costs, probabilities of success and adoption issues are considered.
Table 1: Summary of productivity improvements and estimated value
|Project and region a||Farm-level productivity improvement (%)||Economic benefits to the region ($m)|
|Wheat tolerant to Pratylenchus nematode||Wimmera||14.3||1.7|
|Chickpeas resistant to Ascochyta blight||Wimmera and Mallee CC||6.1||3.9|
|Reducing the protein content of barley||Mallee CC||13.4||1.1|
|Grass-cleaning the pasture phase of rotations||Mallee MF||2.0||10.0|
|Changing from pasture-fallow to non-cereals as break crops||Mallee MF||14.3||12.4|
|Reducing weed impact on cereals||Mallee CC||1.0||0.5|
|From subclover to lucerne-based rotations||North East||14.9||0.4|
|Increasing the protein content of barley||Wimmera||15.4||1.6|
a: In the Mallee agricultural region, there are two distinct types of farming systems, namely continuous cropping (CC) and mixed grain/sheep farming (MF).
The estimated economic benefits range from $0.4 million, for changing from subclover to lucerne-based crop rotations in North Eastern Victoria, to $12 million for changing from pasture-fallow to non-cereals as break crops in the Mallee.
The final section of this report outlines three benefit-cost studies using information generated from the whole-farm models. The studies cover the canola breeding program in Victoria, a new chickpea variety resistant to Ascochyta blight, and the introduction of lucerne in crop rotations in North Eastern Victoria. The results are summarised in Table 2.
Table 2: Summary of benefit-cost studies
|Project||Net benefits ($m)||Benefit-cost ratio|
|Canola breeding program (1980-81 to 1994-95)||5.57||1.62|
|New chickpea variety resistant to Ascochyta blight||1.61||2.80|
|Changing from subclover to lucerne-based crop rotations*||-0.24||0.14|
* The benefit-cost analysis reported here measures economic benefits only. Possible environmental and other benefits, such as salinity mitigation, have not been measured. This is because the techniques needed to assess this type of impact are under development by the Department. Thus the benefits from this type of project may be understated.
There are many competing demands for public funds within NRE and between NRE and other government agencies. With limited public funds available, choices have to be made between alternative projects and activities. To assist the resource allocation process, the Agriculture Division of NRE has developed a rigorous and comprehensive system for analysing R&D investments. This system has been developed to raise the quantity and quality of information used to make allocation decisions and to improve the ability of NRE to inform Parliament and the Victorian public about the benefits of government-funded research, development and extension activities. The framework used in this system is explained in section 1.4.
This report estimates the relative improvements in on-farm productivity that could be expected from changes in selected areas of the grains production system. Information generated in this way can be used by NRE staff in assessing the potential net gains from possible work in the selected areas. The report also includes benefit-cost studies for three large projects. The objective throughout is to help senior staff and other stakeholders make decisions about future R&D investments.
1.2 Overview of the Victorian Grains Industry
Crop volume and value of production are strongly influenced by the weather and market prices. Table 3 shows the gross value of production (GVP) of grain crops in Victoria and Australia for the last five years.
Table 3: Gross value of production (GVP) of grains for Victoria and Australia
|Victoria as a % of national production||12.6||10.3||9.2||12.5||17.3|
Sources: ABS (2001), ABS (2000). Notes: Grain crops include wheat, barley, oats, triticale, maize, sorghum, rice, lupins, field peas, chickpeas, faba beans, lentils, canola, sunflowers, safflower, soybeans and other oilseeds.
p = preliminary.
While GVP for Australia has been reasonably stable, Victoria's share of GVP has fluctuated from around nine per cent to around 17 per cent. Australia achieved a record harvest in 1996/97. In Victoria, 2000/01 was a record harvest. The Victorian cropping industry is divided into two distinct farming types: farms that predominantly grow crops and mixed farms incorporating cropping and livestock. Mixed crop-livestock farms comprise 60 per cent of the total number of grain farms. Mixed livestock-crop farmers and specialist grain farmers grow crops on about 2,565,000 hectares or 11 per cent of Victorian land.
Four cereals dominate grain production: wheat, barley, oats and triticale. The oilseed crop canola has increased in production in recent years due to its relative profitability and agronomic benefits. Chickpea production has dropped dramatically (from a five-year average to 1997/98 of 126 kilotonnes to an annual production of 23 kt in 1999/2000) due to the ravages of Ascochyta blight. However, this loss has been offset by increased production of other pulse crops. Canola and pulses provide valuable disease breaks in crop rotations. Pulses also provide nutritional benefits to the cereal crops that follow in a rotation.
Victoria's major cropping areas are in the lower-rainfall, alkaline soil districts in the state's north-west. This area is loosely bordered by the 300-500 mm annual rainfall zones, as shown in Figure 1. In the higher rainfall zones, smaller cropping areas exist in the North East and South West of the state. South Western Victoria has recently experienced large increases in crop area due to the low profitability of grazing and the use of improved drainage technology that alleviates waterlogging by raising crop beds above the level of the surrounding land.
Figure 1: Major cropping areas of Victoria
1.2.1 The importance of productivity in promoting economic growth
To promote economic growth, as governments seek to do, it is important to understand the drivers of economic growth. A report by Monash University's Centre for Policy Studies for NRE (Adams 1999) noted five drivers of economic growth: productivity improvement, market access, wage rates, return on capital, and government consumption. NRE can influence two of these drivers, namely:
- productivity improvement (through R&D that leads to new technology, and through micro-economic reform); and
- enhancing market access (which promotes agricultural exports). Of these two, productivity improvement is the more important. Adams (1999) revealed that a five per cent improvement in productivity in the food and agriculture sector would increase gross state product (GSP) by $488 million a year. A similar improvement in foreign demand for food would increase GSP by $63 million a year.
Box 1: Productivity Explained
Productivity is essentially the amount of outputs (goods and services) produced from a given amount of inputs. As productivity improves, fewer resources are consumed to produce the desired goods and services. Any activity that raises the ratio of outputs to inputs increases productivity. In a developed economy, investment in knowledge and innovation is believed to offer the most potential for sustained, long-term growth. However, other factors such as interest rates, trade and tariff policies and exchange rates also influence outputs and inputs. Because short-term factors influence the measurement of productivity, a long-term data series is used to gauge improvements.
1.2.2 Productivity in the grains industry
In a recent Australian Bureau of Agricultural and Resource Economics (ABARE) report by Knopke et al. (2000), annual productivity growth in the Australian grains industry was estimated to have increased by an average of 3.2 per cent a year over the 22 years to 1998-99. Productivity growth in Victoria during this period increased by 3.2 per cent a year on specialist cropping farms but by less than two per cent a year for crop-livestock farms.
The increase in productivity growth for the grains industry compares favourably with estimates for agricultural industries in competing countries, and is higher than that recorded by Australian broadacre livestock industries.
It has helped grain growers remain internationally competitive at a time when their terms of trade have declined by around three per cent a year (Knopke et al., 2000).
Through a series of workshops, Knopke et al. attempted to identify some of the key drivers and likely impediments to future productivity growth. Workshop participants identified the following factors as important contributors to productivity gains: advances in plant breeding, better farm management, improved crop rotations with better pest and weed control, development of new herbicides, larger scale farming, more efficient fertiliser use and advances in tractor and equipment design.
NRE is currently working to identify which technologies are likely to have the greatest impact on improving productivity in the grains industry. A supplementary survey has been prepared for use with ABARE's annual survey program in 2001. The areas being examined are crop breeding, natural resource management, agronomy, farm management and marketing.
1.3 NRE grains program
The grains program consists of four key projects that reflect government and industry priorities and enable effective management of the project portfolio. Each project produces outputs intended to achieve, primarily, improved economic outcomes. A small number of projects achieve improved economic and environmental outcomes at the same time.
Key Project 1: Variety Development
The project develops new breeding technologies, enhances germplasm and delivers improved cereal, pulse and oilseed varieties that meet grower requirements and market specifications. Breeders released four new crop varieties (one wheat, two oilseeds, one field pea) in 2001. The latest molecular technologies are used to fast track the breeding process.
Key Project 2: Farming Systems
This project delivers best-practice recommendations for Victorian cropping systems including integrated pest management and economic analysis of cropping systems. Constraints to cropping are identified and solutions examined to maximise crop yield from available rainfall. This involves soil improvements and crop rotation to minimise fungal, insect and nematode attacks on all crop species.
Key Project 3: Natural Resources
This project aims to reduce the impact of cropping systems on the natural resource base. It addresses improved water use and management and soil maintenance. The impacts of the cropping process on soil and water are measured and strategies developed to minimise them. NRE is investigating the development of an environmental management system for cropping farmers. The term environmental management system (EMS) is "a generic term to describe any systematic approach used by an enterprise or organisation to manage its impacts on the environment. The system identifies environmental impacts and legal responsibilities, then implements and reviews changes and improvements in a structured way." (Agriculture, Fisheries and Forestry-Australia 2001, p 6).
Key Project 4: Service Delivery
This project integrates new technical knowledge into cropping practices. The national TOPCROP® program is the vehicle for NRE's major cropping extension effort. It emphasises group discussion and farmer-owned trials to facilitate changes in farm practices and to improve productivity. Project staff also collate variety and agronomic information into management packages for grain growers.
Around $14 million a year is spent on grains research and extension in Victoria. The relative contributions from the Victorian Government, Grains Research and Development Corporation (GRDC) and other sources are shown in Table 4. Other funding sources include commercial interests involved in the breeding of new varieties, and agencies such as the Murray Darling Basin Commission (MDBC).
Table 4: Funding allocation to grains projects (2000/01)
|Project Area||State Sources ($)||GRDC ($)||Other Sources ($)|
Source: Haines, P. pers.comm. (2001)
Note: The figures reported here exclude some overhead costs of conducting grains research at NRE locations. Not included is an estimated $1 million received by NRE in royalty revenue as a result of varieties released by the breeding programs. Industry also provides direct and in-kind support estimated at $1-2 million.
1.4 Resource allocation in NRE
The key problem with allocating resources between competing projects is the limited information available to support rational choices. Important information required to make public sector investment decisions includes:
Project development and planning information on the quality of the science and the project management;
Beneficiaries and funders information on the role of government and government funding as opposed to industry funding; and
Potential benefits of research information on the likely impact of productivity improvements, and returns relative to costs.
Figure 2 illustrates the system developed by NRE to generate this information. The objective is to produce the information needed to support investment decisions and to share this information between research funders (the demand side of the market) and providers (the supply side).
This report uses the quantitative component (shown as 'Farm systems models', 'Total variable costs of agricultural production' and 'Benefit-cost' in Figure 2) of the evaluation framework to provide information on the potential benefits of research to the grains industry. This quantitative component is explained in Section 2.
The 'Beneficiaries and funders' criterion, a qualitative component, is described in Economics Branch (1998b). It has been applied to Science, Technology and Innovation initiative funding applications in NRE (Department of Natural Resources and Environment 2000).
Figure 2: NRE research and development investment evaluation framework
Source: Economics Branch (1998b)
2 A method for estimating the benefits of research
2.1 The farming systems models
NRE's Economics Branch uses a linear programming framework to analyse the effect of technological changes on the farm. The Branch has developed a series of computer models for estimating how different research and development projects might affect on-farm productivity in Victoria's main dairy, beef, wool and grains production regions.
The computer models dubbed 'complex activity budgets' (CABs), are used to estimate the productivity change associated with the adoption of a new technology. The CABs take into account any on-farm adjustments that may be necessary to accommodate the new technology. APPENDIX 1 describes the models in more detail.
The approach developed by the Economics Branch is based on work by Alston et al (1995). Most research-induced technology changes have the effect of reducing the unit cost of production. The CABs are used to measure this improvement in productivity as explained in APPENDIX 1. Further details on the theory underlying the approach are shown in APPENDIX 2.
2.2 Targeting key areas of NRE's grains program
The two different categories of grain production systems in Victoria are the continuous cropping type, as in the Wimmera, and the mixed pasture-crop type, as in the Mallee, North East and North Central regions. Since productivity is the ratio of outputs to inputs, productivity gains could be achieved either by reducing inputs for a constant output, or by increasing outputs without changing inputs. Input reductions may include new, disease-resistant crop varieties that reduce or eliminate the need for a routine spray. Increased outputs may include new crop varieties that yield more for the same amount of inputs.
Productivity can also be improved without changing the quantity of either inputs or outputs - by improving output quality so that a higher price is received. This is considered as an increase in 'effective' quantity, and is measured by considering the output with and without the quality improvement as two different farm outputs.
2.3 Applying model results to the regions
Productivity improvements measured using CAB models are based on a single cropping or crop-pasture rotation, representing the top 25 per cent of farmers in a region. However, within each region there may be several common rotations. For each region, the major categories of rotations have been identified in consultation with extension agronomists. In each analysis, the measured productivity improvements are then weighted according to the proportion of the crop affected by the particular technical change in all categories of rotations.
2.3.1 The Wimmera Region
Three broad categories of rotations prevail in the Wimmera (Bedggood, 2001). Continuous cropping predominates on the grey cracking clay soils that characterise about one-third of the region. These farmers commonly rotate cereal crops with two different pulse crops and canola, for example, canola-pulse1-cereal-pulse2-cereal. The cereal crops are mainly wheat, sometimes barley. The pulses are mainly lentils, field peas and faba beans.
A second category of rotations involves rotating crops with long fallow without livestock as a significant component of the farm. A common rotation is fallow-cereal-pulse-cereal. Some farmers grow only wheat while others grow both wheat and barley as the cereals in the rotation. A third category involves rotating crops with pasture running mainly sheep, for example, pasture-pasture-cereal-cereal.
As we are considering improved barley production technologies, splitting the three categories of rotations into with and without barley results in six farm types for the Wimmera.
A theoretical Wimmera region was established using the six farm types. This is shown in Table 5 and was constructed as follows. The area under each farm type was calibrated so that the region's total gross revenue from wheat, barley, total pulses and canola equals the five-year average of its GVP for each crop (ABS, 2001). The gross revenue from each crop was calculated from the yields and prices used in the CAB model. The total variable cost (TVC) of each farm type was obtained from the CAB model. TVC as average dollars per hectare was multiplied by the estimated area under each farm type to calculate the TVC of that farm type in the region. This calculation is needed to value the gross benefits of the proposed technical changes described in Section 3.
Table 5: Major farm types considered in establishing the theoretical Wimmera region
|Wheat in rotation||2||1||2||1||2||1|
|Barley in rotation||1||1||1|
|Pulse 1 in rotation||1||1||1||1|
|Pulse 2 in rotation||1||1|
|Canola in rotation||1||1|
|Fallow in rotation||1||1|
|Pasture in rotation||2||2|
|Total revenue (average $/ha)||414||414||262||262||281||281|
|Total variable cost (average $/ha)||164||161||109||106||119||115|
|Farm gross margin (average $/ha)||249||252||153||156||162||166|
|Area under farm type ('000 ha)||125||250||32||80||50||134|
|Total variable cost (million $)||20.50||40.25||3.49||8.48||5.95||15.41|
C: cereal crop; Ca: canola; F: long fallow; P: pasture; Pl1: pulse crop 1; Pl2: pulse crop 2; Pl: pulse crop.
2.3.2 The Mallee region
Three broad categories of rotations prevail in the Mallee (Sonogan, 2001). Continuous cropping predominates in the eastern and southern parts of the region. These farmers commonly grow cereal crops in rotation with a pulse crop and canola, for example, pulse-canola-cereal-cereal. The cereal crops are mainly wheat, sometimes barley. The pulses are mainly field peas, lentils, lupins and faba beans. Some farmers include a one-year pasture phase (without sheep to graze) between the two cereal crops.
However, canola occupies only four per cent of the cropping area of the Mallee (ABS, 2001), showing that some continuous cropping farmers rotate cereal crops with only pulse crops. A common rotation on these farms is pulse-cereal-cereal. The cereal crops are mainly wheat, sometimes barley. Some farmers include a long fallow phase between the second cereal crop and the pulse crop.
A third category of rotations involves rotating cereal crops with pasture and/or long fallow with livestock as a significant component of a mixed farm, for example, either pasture-cereal or fallow-cereal. Some mixed farmers grow only wheat while others grow wheat and barley alternately as the cereals in the rotation. Some mixed farmers use pasture-fallow-cereal or pasture-fallow-cereal-cereal. However, ABS data on pasture and long fallow area in the Mallee show that around 30 per cent of the cropping area of mixed farming systems do not include a pasture phase and would not have livestock as a major component of the farm. Therefore, only 70 per cent of the estimated area under this category was considered for the two projects involving pasture.
Splitting the three categories of rotations into with and without barley results in six farm types for the Mallee. A theoretical Mallee region established using the six farm types is shown in Table 6. The area under each farm type was estimated based on the GVP of wheat, barley, total pulses and canola following the method described for the Wimmera in section 2.3.1.
Table 6: Major farm types considered in establishing the theoretical Mallee region
|Wheat in rotation||2||1||2||1||2||1|
|Barley in rotation||1||1||1|
|Pulse in rotation||1||1||1||1|
|Canola in rotation||1||1|
|Pasture &/or fallow in rotation||2||2|
|Total revenue (average $/ha)||311||311||277||277||283||283|
|Total variable cost (average $/ha)||139||137||125||123||129||128|
|Farm gross margin (average $/ha)||172||174||152||154||154||155|
|Area under farm type ('000 ha)||28||84||75||183||220||740|
|Total variable cost (million $)||3.89||11.51||9.38||22.51||28.38||94.72|
C: cereal crop; Ca: canola; F: long fallow; P: pasture; Pl: pulse crop.
2.3.3 The North East region
Three broad categories of rotations prevail in the North East, all based on rotating crops with pasture. Pasture is mainly subclover-based annual pasture of varying length. Most farmers restrict their pasture phase to a minimum of two years because of the relatively low profitability of livestock compared to cropping.
The basic difference between the three categories is the mix of crops in the cropping phase. Some rotations contain a cropping phase comprising canola and cereals, for example, pasture-pasture-canola-cereal-cereal. The cereal following canola is usually wheat; the second cereal can be triticale or barley.
The second category of rotations has a cropping phase comprising pulses (mainly lupins) and cereals. A common rotation on these farms is pasture-pasture-cereal-pulse-cereal. The third category of rotations has a cereal-only cropping phase, for example, pasture-pasture-cereal-cereal-cereal. Wheat is the major cereal, but one or two wheat crops would generally be followed by either triticale or barley. The project considered for North Eastern Victoria did not involve any change to a specific crop, so the three categories of rotations were considered as three farm types to represent the grain-sheep farm systems in the region. A theoretical North East Victoria region established using the three farm types is shown in Table 7. The area under each farm type was estimated based on the GVP of wheat, barley, total pulses and canola for the region.
Table 7: Major farm types considered in establishing the theoretical North East region
|Farm type 1||Farm type 2||Farm type 3|
|Cereals in rotation||2||2||3|
|Pulse in rotation||1|
|Canola in rotation||1|
|Pasture in rotation||2||2||2|
|Total revenue (average $/ha)||416||344||356|
|Total variable cost (average $/ha)||148||104||130|
|Farm gross margin (average $/ha)||269||239||225|
|Area under farm type ('000 ha)||4.75||6.90||11.50|
|Total variable cost (million $)||0.70||0.72||1.50|
C: cereal crop; Ca: canola; P: pasture; Pl: pulse crop.
3 Estimating the gross benefits
Productivity change is measured in terms of a reduction in inputs for a given quantity of outputs as explained in section 1.2.1 (estimation is explained in APPENDIX 1). The productivity change is multiplied by the total variable cost (TVC) for the relevant farm types for a particular region to give the gross annual benefits. The TVC used is that for the whole-farm system under consideration. For a continuous cropping system, all crops are considered. Likewise, all crops, pasture and livestock are considered for a mixed livestock-crops system. Section 2.3.1 explains how the TVC of each farm type is estimated for a region using gross revenue and variable cost data from CAB models and five-year average GVP (ABS, 2001).
Gross benefits are generally estimated for the whole of an agricultural region such as the Wimmera or for the proportion of cropping area within a region that is affected by a particular technology. The gross annual benefit is the maximum annual benefit that could be expected before project costs, probability of project success and adoption issues are considered.
The eight projects analysed using the CAB models are outlined in Table 8. Sections 3.1 to 3.4 describe the technologies and the process used to estimate benefits for the eight projects.
Table 8: Description of the projects assessed
|Key project area||Project and region||Description|
|Variety development||Wheat tolerant to Pratylenchus nematode in the Wimmera||Pratylenchus nematodes reduce wheat yields in some paddocks in the Wimmera. A wheat variety tolerant to this nematode would alleviate the problem.|
|Variety development||Chickpeas resistant to Ascochyta blight in the Wimmera and Mallee||After the 1998 epidemic of Ascochyta blight, chickpeas became unprofitable. A chickpea variety resistant to this fungal disease would make this crop a profitable option.|
|Variety development||Reducing the protein content of barley in the Mallee||A new malting barley variety with a protein content below 11 per cent would reduce the incidence of barley being downgraded to feed-grade quality.|
|Farming systems||Grass-cleaning the pasture phase of rotations in the Mallee||Medic pastures are rotated with cereal crops by mixed farmers in the Mallee to break disease and weed cycles. Using a selective herbicide to kill grasses appears to increase the yield and protein content of the following wheat crop.|
|Farming systems||Changing from pasture-fallow to non-cereals as break crops in the Mallee||Traditionally, Mallee farmers used pasture and fallow as disease and weed breaks between cereal crops. Low wool prices have made pasture and sheep less profitable. Continuous cropping with non-cereal crops as breaks instead of pasture and fallow may improve profitability.|
|Farming systems||Reducing weed impact on cereals in the Mallee||Weeds that are difficult to control reduce cereal yields. This hypothetical project would reduce yield losses by 10 per cent without increasing the cost of growing cereals in the Mallee.|
|Natural resources||From subclover to lucerne-based rotations in northeastern Victoria||Mixed grain-sheep farmers in North Eastern Victoria rotate crops with subcloverbased pasture. Using lucerne-based pasture would improve farm profitability and reduce groundwater recharge.|
|Service delivery||Increasing the protein content of barley in the Wimmera||Around four per cent of the Wimmera malting barley crop is downgraded to feed quality due to protein less than nine per cent. This could be alleviated by testing deep-soil nitrogen and applying fertiliser nitrogen.|
3.1 KEY PROJECT: VARIETY DEVELOPMENT
3.1.1 A wheat variety tolerant to Pratylenchus nematodes
One in five wheat paddocks in the Wimmera harbours Pratylenchus thornei while four in five harbour Pratylenchus neglectus (Hollaway, 2001). However, these nematodes reduce yields by approximately 20 per cent on only about 10 per cent of all wheat paddocks in the Wimmera. The benefit to currently-affected Wimmera farmers of adopting a wheat variety tolerant to Pratylenchus was analysed.
Without scenario: Because the only wheat option available to the EMAR (Economic Model of Agronomic Rotations)-Wimmera had a 20 per cent yield loss due to Pratylenchus, the optimum rotation derived by the model as being most profitable did not include wheat. The model was then restricted to selecting wheat as part of its rotation. This analysis only considered farmers in the region whose wheat yields are affected by the nematode.
With scenario: The 'with' case introduced a nematode-tolerant variety of wheat with no yield loss due to Pratylenchus. This wheat variety used the same inputs as the non-tolerant variety. The optimal rotation derived by the EMAR-Wimmera model included the tolerant wheat variety.
The five-year rotation used in this analysis resulted in a 14.3 per cent improvement in on-farm productivity, based on farm type 1.
If all farmers grow wheat, the target population can be represented by taking 10 per cent of the TVC of all grain crops in each farm type (see Table 5). The resulting figure was then multiplied by the productivity improvement to estimate the potential benefit of breeding a wheat variety tolerant to Pratylenchus. Because the productivity improvement was measured using farm type 1 but the proportion of wheat in the rotation differs between farm types, the measured productivity improvement was weighted according to the proportion of wheat in each of the six rotations. The resulting gross benefit of breeding this new variety in the Wimmera region was $1.7 million.
3.1.2 Chickpeas resistant to Ascochyta blight
Chickpeas used to be the most profitable pulse crop grown in the Wimmera. In Victoria, the area sown to chickpeas reached 135,000 hectares in 1996, at an average yield of 1.3 tonnes per hectare (ABS, 2001). After the 1998 epidemic of the fungal disease Ascochyta blight, the area sown fell to around 28,000 ha in 1999 and then to an estimated 5,000 ha in 2000. While chickpea growers still achieve average yields of 1.3 tonnes per hectare, these yields have been maintained through increased use of chemicals costing around $150/ha. Through conventional breeding and the Single Seed Descent method2, molecular markers will be used to breed a chickpea variety moderately resistant to Ascochyta blight. Resistance genes are being inserted into Lasseter, a highly susceptible variety that has desirable quality attributes sought by overseas markets. This will decrease the number of protective sprays required at the pod-filling stage from six to two (at $25 per application), but is not expected to affect yield.
Without scenario: In the absence of a new chickpea variety, the EMAR-Wimmera model selected lentils (instead of chickpeas) and field peas as the two pulse crops in its optimal rotation.
With scenario: The EMAR-Wimmera model selected the new chickpea variety (with average yield and two fungicide sprays at a total cost of $50/ha) and lentils in its optimal rotation.
Using a five-year rotation, this improved productivity by 6.1 per cent. When chickpeas peaked in Victoria in 1996, 102,000 hectares was sown in the Wimmera, or 41 per cent of the region's total pulse area. We applied this same ratio, i.e. 41 per cent to the area of legumes predicted by the model as the maximum area sown to chickpeas. The model predicted 178,000 ha of legumes would be sown. Thus, 41 per cent of 178,000 ha is 73,838. This is the maximum area that could be sown to chickpeas if the new variety is adopted. This area is approximately 72 per cent of the area sown to pulses in the theoretical Wimmera and Mallee regions (see Tables 5 and 6). In estimating gross benefits, we considered 72 per cent of the TVC of only the farm types with pulse crops in the Wimmera and Mallee regions. These two regions comprise 95 per cent of the chickpea area of Victoria (ABS, 2001). The resulting gross benefit for these two regions was $3.9 million.
3.1.3 Reducing grain protein content of barley in the Mallee
Barley represents 36 per cent of the total cereal area in the Mallee. Barley with a protein content between nine and 11 per cent is classified as malting-grade. Protein levels above 11 per cent cause around 23 per cent of the Mallee barley crop to be downgraded to feed-grade quality (Moody, 2001). The five-year average price of malting barley is $195/t (delivered port) while that of feed barley is $150/t.
We looked at the potential benefit of breeding a low-protein variety of malting barley. Mixed grain-sheep farming (MF) and continuous cropping (CC) systems were considered separately, but both on a four-year rotation.
Without scenario: The EMAR-Mallee model included the current barley variety sold as feed grade. The optimal rotation derived for both the MF and CC systems did not include feed-grade barley because of its low price, so we forced the model to replace wheat with feed-grade barley. Only 23 per cent of the cropping area of all farms growing barley was considered in this analysis.
With scenario: Options available to the EMAR-Mallee model included a new malting-grade barley with a lower grain protein content than the previous feed-grade variety. The optimal rotation derived for each system included the new barley.
The productivity improvement as a result of the new variety was 16.4 per cent for the MF system and 13.4 per cent for the CC system.
In calculating the gross value of this improvement, we multiplied the productivity improvement by 23 per cent of the TVC of barley-growing farm types in each system. The resulting gross benefits for the MF and the CC system were $3.6 and $1.0 million, respectively. The total gross benefit to the Mallee region was $4.6 million.
3.2 Key project: Farming systems
3.2.1 Rotations with grass-cleaning of the pasture phase in the Mallee
In the Mallee, medic pasture and fallow are commonly rotated with cereal crops to break disease cycles. Predominantly, these medic pastures regenerate voluntarily and are not managed well. Using a selective herbicide for 'grass-cleaning' (spraying medic pasture to remove grassy weeds) has been shown to increase the yield and protein content of subsequent wheat crops. We used the EMAR-Mallee model to assess the introduction of grass-cleaning of medic pastures in the Mallee. Grass-cleaning at $32 a hectare plus application costs each year for two years would be expected to increase the yield of the subsequent wheat crop by 24 per cent and its protein content by 12 per cent (Grey, 2001).
Without scenario: The optimal rotation derived by the EMAR-Mallee model included wheat with the existing yield and protein content. This analysis is relevant only to mixed farming (MF) system in the Mallee. 70 per cent of the cropping area of the MF system was considered as the target population as explained in Section 2.3.2.
With scenario: The EMAR-Mallee model selected wheat with 24 per cent higher yield and 12 per cent higher protein content immediately following pasture and fallow in the optimal rotation. The cost of buying and applying grass-cleaning chemicals was included in this scenario.
Introducing two years of grass-cleaning for medic pastures in the Mallee improved productivity by 12.0 per cent. The gross benefit of adopting grass-cleaning was estimated by multiplying the productivity improvement by 70 per cent of the TVC of MF farm types. The resulting gross benefit for the Mallee region is $10 million.
3.2.2 Changing from pasture-fallow to continuous cropping in the Mallee
Traditionally, the majority of dryland farmers in the Mallee grow cereals rotated with medic pasture and fallow as disease breaks. Sheep are reared mainly to consume the herbage available in pasture and fallow phases and from the stubble that remains after crops are harvested. Despite their poor profitability relative to cropping, sheep are also an insurance against crop failure.
However, there are some farmers who follow a continuous cropping program with non-cereal crops such as field peas and canola as disease breaks between cereals. The EMAR-Mallee model was used with and without non-cereal crops to evaluate the benefits of changing from mixed grain/sheep farming to continuous cropping.
Without scenario: Options available to the EMAR-Mallee model included a sheep (lamb) enterprise but no non-cereal crops. The optimal rotation selected by the model included pasture, fallow and cereals. As in the previous analysis, our target population was 70 per cent of the cropping area of the region's MF system.
With scenario: The EMAR-Mallee model selected a continuous cropping rotation with non-cereals and no sheep enterprise as optimal. Non-cereal crops are more profitable than sheep and pasture.
The optimal rotation with non-cereals is $36/ha (26 per cent) more profitable than the optimal rotation excluding non-cereals. Changing from pasture-fallow to continuous cropping raised productivity by 14.3 per cent.
This figure was multiplied by 70 per cent of the TVC of MF farm types to estimate the gross benefits of changing from MF to CC. The resulting gross benefit for the Mallee region is $12 million.
Eliminating the fallow phase would also benefit the environment by reducing groundwater recharge and soil erosion, which are inherent problems of fallow. However, these environmental benefits have not been quantified in this report.
Whether farmers adopt this change in the farming system depends to a great extent on their attitude to risk. Although continuous cropping with non-cereals as break-crops is more profitable and productive than the conventional pasture-fallow-cereal rotations, the former involves a higher degree of income variability (or risk) during low-rainfall years. Further discussion and assessment of this situation are beyond the scope of this analysis.
3.2.3 Reducing the impact of weeds on cereals in the Mallee
This hypothetical analysis assesses the profitability of undertaking research, development and extension activities to achieve a 10 per cent reduction in the impact of weeds on cereal crops grown in the Mallee, without additional costs to the grower.
Hypothetical examples provide useful information to NRE scientists. Project cost information can be added and a benefit-cost analysis completed for each alternative. Using models in this way allows alternative project ideas to be assessed and the results used to guide priority setting.
Without scenario: The settings for the EMAR-Mallee model include the crop yields that would be expected after each possible two-year paddock history. As in this scenario, the model can allow for reduced yields due to 'difficult to control' weeds.
With scenario: A hypothetical new weed-control technology was assumed to reduce the yield loss due to weeds by 10 per cent without changing any crop inputs. The impact of this technology was analysed separately for the Mallee MF and CC farming systems. The optimal rotation did not change for either system.
The new technology improved productivity by 1.0 per cent and 0.99 per cent for the MF and CC systems, respectively.
The gross benefits were estimated for each system by multiplying the TVC of all farm types of the system by the productivity improvement. The resulting gross benefits for the MF and CC systems are $1.2 and $0.5 million, respectively. The total benefit to the Mallee region is $1.7 million.
3.3 Key project: Natural resources
The project examined here has environmental outcomes as its primary objective, but also improves farm profitability. We used the CAB model to estimate only the economic benefits.
3.3.1 Changing from subclover to lucerne-based crop rotations
The standard practice among mixed grain-sheep farmers in North Eastern Victoria is to rotate crops with subclover-based annual pasture. Lambing is generally scheduled for autumn. Growers who introduce lucerne pasture into their crop-pasture rotations generally maintain a mixture of subclover and lucerne and adopt spring lambing. We used the PRISM-North East (Profitable Resource Integration, Southern MIDAS – North East) model to evaluate the on-farm benefits of changing from subclover to lucerne pasture. (MIDAS stands for Model of an Integrated Dryland Agricultural System.) Without scenario: This scenario included subclover pasture and an autumn lambing sheep enterprise. Lucerne pasture was not included as an option. Subclover pasture rotated with canola and cereals was selected as the optimal solution. The mixed grain/sheep farming system of the whole of the North East region was the target population for this analysis.
With scenario: Options available in this scenario included lucerne and subclover pasture with a spring-lambing sheep enterprise. The PRISM-North East model selected a combination of two rotations as the optimal solution. The two rotations had, respectively, subclover and lucerne pastures rotated with canola and cereals.
The resulting productivity improvement of 14.9 per cent was multiplied by the TVC of all three farm types in the region (see Table 7) to produce a potential gross benefit of $0.4 million for the whole region. Lucerne also benefits the environment by reducing groundwater recharge and hence dryland salinity, but the potential environmental benefits have not been quantified.
3.4 Key project: Service Delivery
The project considered here extends an existing technology to solve a problem prevailing in the Wimmera.
3.4.1 Increasing the grain protein content of barley in the Wimmera
Barley constitutes 41 per cent of the area of all the cereal grains grown in the Wimmera (ABS, 2001). On average, eight per cent of the Wimmera barley crop is downgraded from malting to feed quality by ABB Grain Ltd (Moody, 2001). When the protein content is less than nine per cent, the market price is reduced by around $40 a tonne. One strategy for overcoming this problem involves testing deep-soil nitrogen and applying nitrogen fertiliser (Moody, 2001) late in the season. We used the EMAR-Wimmera model to estimate the benefits of improving the protein content of Wimmera barley by testing soil nitrogen and applying nitrogen.
Without scenario: Options available to EMAR-Wimmera included barley downgraded to feed quality due to low protein content. Because the model did not include barley in its optimal rotation, we then forced it to select barley instead of wheat. The target population considered for this analysis is the eight per cent of the cropping area of barley growers in the Wimmera region whose barley is downgraded to feed quality due to low protein content.
With scenario: Options available to the model included malting-quality barley produced using deep-soil nitrogen testing and aerial application of 87 kg of urea per hectare. After including these costs and new prices, the optimal solution chosen by the model included wheat as well as barley. The resulting productivity improvement was 15.4 per cent. This was multiplied by eight per cent of the TVC of all barley-growing farm types in the Wimmera. The resulting gross benefit was $1.6 million.
3.5 Summary of estimated productivity improvements
The productivity improvements due to the implementation of the selected projects explained in sections 3.1 to 3.4 of this report are summarised in Figure 3.
Figure 3: Summary of estimated productivity improvements
The estimated productivity improvements range from around 16 per cent for reducing the protein content of barley on mixed farms in the Mallee to one per cent for reducing the impact of weeds on cereal crops in the Mallee. The results provide a guide to the potential size of the benefits from proposed research in these areas.
However, when assessing the feasibility of proposed research, other factors such as science quality issues, and the probability of success of the proposed research also needs to be considered.
3.6 SUMMARY OF THE VALUE OF PRODUCTIVITY IMPROVEMENTS
Figure 4 shows the value of the productivity improvements arising from the implementation of the selected projects, that is, the additional returns to grain growers if they are able to realise these productivity improvements. The results are based on the assumption that all farmers in the target region are able to achieve the productivity improvements.
Figure 4: Summary of the value of productivity improvements
The two most valuable projects are grass-cleaning and the use of non-cereals as break crops. The values of these productivity improvements are related to the high TVC of the target population and are the maximum annual gross benefits that can be expected without taking into account project costs, adoption levels and probabilities of success.
The lowest value project involves changing from subclover to lucerne-based rotations in North Eastern Victoria. Grain/sheep mixed farming occupies only a very small proportion of the region. Low values for several other projects are due to relatively small productivity improvements or target populations for these technologies.
A summary of target regions, productivity changes and economic benefits for all the selected projects are presented in Table 9.
Table 9: Summary of economic benefits
|Project||Agricultural regiona||Target population||Productivity improvement (%)||Gross economic benefits ($m)|
|Wheat tolerant to Pratylenchus nematode||Wimmera||10% of total cropping area||14.3||1.7|
|Chickpeas resistant to Ascochyta blight||Wimmera||72% of total cropping area of pulse growers||6.11||3.9|
|Reducing the protein content of barley||Mallee CC||23% of total cropping area of barley growers||13.4||1.1|
|Grass-cleaning the pasture phase of rotations||Mallee MF||70% of total cropping area||12.0||10.0|
|Changing from pasture-fallow to non-cereal break cropsb||Mallee MF||70% of total cropping area||14.3||12.4|
|Reducing weed impact on cereals||Mallee CC||100% of total cropping area||1.0||0.5|
|From subclover to lucerne-based rotationsc||NE||100% of total cropping area||14.9||0.4|
|Increasing the protein content of barley||Wimmera||8% of total cropping area of barley growers||15.4||1.6|
a: In the Mallee agricultural region, there are two distinct types of farming systems: continuous cropping (CC) and mixed grain/sheep farming (MF);
b: This system reduces soil loss due to wind erosion, and ground water recharge. The environmental benefits from these effects have not been included in this analysis;
c: This system can reduce groundwater recharge and dryland salinity. These environmental benefits have not been included in this analysis.
The economic analyses contained in this report provide valuable information to those charged with making decisions about the funding of research, development and extension activities.
A decision to invest involves an assessment as to whether there are net benefits. This requires a careful assessment of the design of the proposed research. Next, information is needed on project costs, project risks, adoption profiles and the time frame for the proposed research. Alternative research approaches may also need to be considered. A formal benefit-cost study allows these issues to be taken into account in a structured way.
4 Benefit-cost analysis of research projects
NRE has developed a spreadsheet, known as Appraisal, to prepare benefit-cost analyses for research, development and extension projects (see Appleyard, 1996). The benefit-cost analyses reported in this section used Appraisal with a discount rate (or time preference for money) of eight per cent. Appleyard (1996, p 6) explains the selection of the discount rate.
The assumptions used in the three benefit-cost studies and a summary of the results are presented in Tables 10 and 11. The assumptions used in these studies are based on knowledge from earlier studies and the experience of field staff. The details of each study follow.
Table 10: Assumptions used in benefit-cost studies
|Parameter||Canola breeding program||New chickpea variety||Lucerne in crop rotations|
|Probability of project success||100%||95%||100%|
|Probability of successful implementation by growers||100%||90%||80%|
|Depreciation decay rate of the adopted technology||0%||10%||0%|
|Year adoption begins – with project||4||5||3|
|Year adoption begins – without project||6||7||5|
|Target population of the regions considered a||Total cropping area of those growing canola in Victoria||35 per cent of total cropping area of Wimmera & Mallee CC||Total cropping area in North Eastern Vic.|
|Maximum adoption – with project||100%||100%||10%|
|Maximum adoption – without project||100%||100%||10%|
|Years to maximum adoption – with project||12||12||12|
|Years to maximum adoption – without project||14||14||14|
a: CC is continuous cropping.
Table 11: Results from benefit-cost studies
|Parameter||Canola breeding program||New chickpea variety||Lucerne in crop rotationse|
|Present value of benefits $m||14.5||2.50||0.04|
|Present value of costs $m||8.92||0.89||0.28|
|Net present value (NPV) $m||5.57||1.61||-0.24|
|Benefit-cost ratio (BCR)||1.62||2.80||0.14|
e:The benefit-cost analysis reported here measures economic benefits only. Possible environmental and other benefits, such as salinity mitigation, have not been measured as the techniques needed are not yet available. The Department is developing methods to estimate the impact of this type of research. Thus the benefits from this type of project may be understated.
4.1 Canola breeding program in Victoria
An economic review of the Victorian canola breeding program from 1980 to 1994 was undertaken in 1996 using the PRISM-Wimmera model (Jones and Soligo, 1996). The benefits were estimated as changes in whole-farm profit due to the increase in canola yield during the period.
In the following analysis, we have estimated the benefits separately for the continuous cropping and mixed grain/sheep farming systems, using the methods and models described in this report. Using two farming systems models, namely EMAR-Wimmera and PRISM-Bendigo, we have estimated benefits as the value of productivity changes due to the increase in canola yield between before and after the breeding program. Our study covers the same period as Jones and Soligo (1996).
Victoria produced a total of 84 kilotonnes (kt) of oilseeds in 1994, up from just 3.4 kt in 1983. Victorian production of canola was 74.5 kt in 1994 with a gross value of production (GVP) of $26.3 million, or more than 70 per cent of the tonnage and value of the total Victorian oilseed crop.
Canola yield and oil content were increased as a result of improved agronomic practices and a breeding program. The Victorian canola breeding program also aimed to minimise the adverse effects of diseases, overcome losses from seed shattering, control pests and weeds, and develop new varieties with desirable oil characteristics. A locally bred canola cultivar, Marnoo, was released commercially in 1980. Over the period 1980 to 1994, ten canola varieties have been released from the Victorian program.
The 1996-97 Agricultural Census (ABS, 1998) identified Victoria's major canola-growing areas as the Wimmera (60 per cent of production), southern Mallee (15 per cent), Loddon (six per cent), Goulburn (six per cent) and Central Highlands (six per cent) statistical divisions.
The EMAR-Wimmera farming system model represents the continuous cropping system in the Wimmera statistical division, while PRISM-Bendigo represents the mixed grain/sheep farming system in the Loddon and Goulburn statistical divisions. The models were used to evaluate economic benefits of the increase in canola yield from 1980 to 1994. It was assumed that the yield increase could mainly be attributable to the release of new varieties with improved characters throughout the breeding program. Each model was optimised, using average (of 1980 and 1994) yields for all crops other than canola. Prices for all crops and sheep products used in the models were averages for the period. The technical change between 'without' and 'with' scenarios was the increase in canola yield from the level of 1980-81 production year up to the yield level of 1994-95.
Productivity change due to the increase in canola yield was estimated using the inputs and outputs between the 'without' and 'with' scenario run of each model. These productivity increases were multiplied by the total variable costs of all crops, pasture and livestock in the regions to estimate the annual benefits to the grains industry. The benefits are confined to the farmers who are growing canola in the regions. Therefore, the resulting benefits were multiplied by the proportion of farmers growing canola in each region (estimated from ABS, 2001). The distribution of benefits over the period from 1980 up to 1994 was according to the adoption profile specified (see Table 10).
The Wimmera, Loddon and Goulburn statistical divisions account for 72 per cent of all canola grown in Victoria. The benefits for these regions were extrapolated to cover the whole of Victoria. The annual costs of the canola breeding program were obtained from Jones and Soligo (1996). The adoption of the newly-bred canola varieties was considered to commence at year four of the breeding program and achieve 100 per cent uptake by year 12.
Using the 1980-81 canola yield, the EMAR-Wimmera model selected faba beans-cereal-cereal-field peas-cereal as its optimal cropping rotation. With the canola yield increased to the 1994-95 level, the model selected canola-field peas-cereal-canola-cereal. The resulting productivity increase was 8.9 per cent. The PRISM-Bendigo model selected its optimal rotation as pasture-pasturepasture-wheat-lupins-wheat using the 1980-81 canola yield. Increasing the canola yield changed the rotation to pasture-pasture-canola-wheat-lupins. The resulting productivity increase of 14.1 per cent is higher than the 8.9 per cent achieved for the continuous cropping system. The main reason for this is the shortening of the pasture phase in the rotation from 50 per cent to 40 per cent. Pasture (and livestock) are relatively less profitable than cropping. Project costs and estimated benefits to Victoria over the 15-year period are presented in Table 12, together with the results of the benefit-cost analysis.
Table 12: Benefit-cost analysis of the Victorian canola breeding program
|Year||Project costs ($)||Benefits to Victoria ($)|
|Present value (discounted)||8,925,744||14,493,236|
|Net present value (NPV)||$ 5,567,492|
|Benefit-cost ratio (BCR)||1.62|
4.2 Breeding a chickpea variety resistant to Ascochyta blight
Before the fungal disease Ascochyta blight became a serious problem in Victoria, the area sown to chickpeas had reached 135,000 hectares in 1996-97. The area sown to chickpeas has since fallen to around 5,000 ha (2000). The major variety grown is the more tolerant and high-valued kabuli type chickpea. In southern NSW, and South Australia, plantings also dropped dramatically during this period. Western Australia, NSW and Qld have increased or maintained production, as these states have had fewer problems with the disease. The Wimmera and Southern Mallee are the main areas affected in Victoria. It is believed that with the advent of suitable varieties, chickpeas will again become competitive with other crops and be included in rotations.
Introducing a new, resistant variety would reduce spray costs by around $100/ha. The annual benefits have been estimated to be $11 million (see section 3.1.2). It was assumed for the benefit-cost analysis that these benefits are distributed throughout the period of the project according to the adoption profile specified in Table 10. The estimated costs of developing this variety are shown in Table 13.
There is a 95 per cent probability of success because germplasm is available with resistant genes. Given the relative profitability of the new variety, it is assumed that all growers would adopt the new variety. Adoption is assumed to commence a year after the end of the breeding phase and to reach 100 per cent nine years after the end of the breeding phase.
The target region is 72 per cent of the total cropping area of pulse growers and the results of the benefit-cost analysis are presented in Table 13.
Table 13: Benefit-cost analysis of breeding a disease-resistant chickpea variety
|Year||Project costs ($)||Benefits to Victoria ($)|
|Present value (discounted)||893,491||2,503,672|
|Net present value (NPV)||$ 1,610,181|
|Benefit-cost ratio (BCR)||2.80|
4.3 Changing from subclover to lucerne=based crop rotations
This project was conducted from 1993 to 1998 at Agriculture Victoria, Rutherglen. The work was undertaken to improve the efficiency of cropping systems to reduce the amount of water lost from the roots of annual crops and pasture in higher rainfall areas. The inclusion of perennial plants such as lucerne, which can dry the subsoil to depth and provide greater storage capacity for winter rainfall, was tested in a 'phase farming' system. (A phase farming system is one where pastures are rotated with annual crops on the same land.) Lucerne is likely to be a more attractive proposition for farmers than other perennials such as trees because it produces high-quality forage and provides economic returns in the short term.
A formal extension phase was not envisaged at commencement and has not been undertaken. However, the results of the project have been disseminated to the farming community by way of field days and seminars. Our analysis assumed that the increased area of lucerne sown by grain growers after completion of the project was attributable to the project. However, this assumption may overstate the economic benefits of the project, as other factors may have been responsible for the increased lucerne plantings in the target region.
The project aimed to develop improved cropping systems incorporating perennials for increased water use, crop production, grain quality and sustainability in high rainfall areas; and to increase the use of rain where it falls, reducing groundwater accessions and the risk of developing salinity. Plot experiments were conducted to compare subclover (annual) pasture with lucerne (perennial) pasture. Both types of pasture were grown for different lengths of time followed by cropping for different lengths of time. The project measured crop yields, grain quality and soil-water dynamics under pastures and crops. The results showed that lucerne was valuable in this respect and that rotations should include three to four years of lucerne followed by three to four years of cropping.
The proportion of dryland lucerne in the cropping areas of north-eastern Victoria increased from 1.5 to three per cent between 1998 and 2000 (Grey, 2001). Assuming that lucerne is grown in 50:50 rotation with crops, the area under lucerne pasture-crop rotations appears to have increased from three to six per cent. Based on these observations, the maximum adoption of cropping rotations with lucerne was determined as 10 per cent of total cropping area of mixed farming systems at year 12 after the commencement of the project. Adoption was assumed to have commenced in the third year of the project. As the project has been successfully completed, the probability of research success was assumed to be 100 per cent.
This project was designed primarily to examine the role of lucerne in cropping rotations, especially in terms of water use. The project is, in many respects, experimental work, and some of the costs are higher than if the work had simply aimed to implement known technologies for economic benefit alone. The potential gross benefit arising from growers changing from a subclover to a lucerne pasture phase in their cropping rotations has been estimated to be $0.4 million (see section 3.3.1). These benefits were estimated for the Ovens-Murray statistical division of North Eastern Victoria, even though the technology has application in adjoining areas. Other areas where this technology is suitable include high rainfall regions (above 450 mm) such as southern NSW and northern Victoria. This would indicate that the project benefits are understated.
Project costs, annual benefits and the results of the benefit-cost analysis are shown in Table 14. The low rate of adoption (10 per cent) means that the work is not a profitable investment on economic grounds alone, but our analysis does not account for environmental benefits, such as salinity mitigation. This has not been done as methods to estimate this type of impact are not yet available. Methods are under development in the Department.
Table 14: Benefit-cost analysis of changing to lucerne-based cropping rotations
|Year||Project costs ($)||Benefits to Victoria ($)|
|Present value (discounted)||276,065||39,489|
|Net present value (NPV)||$ -236,576 (environmental benefits have not been estimated)|
|Benefit-cost ratio (BCR)||0.14|
ABARE (2000a), Australian Commodities Forecasts and Statistics, Australian Bureau of Agricultural and Resource Economics (ABARE), Canberra.
ABARE (2000b), Australian Farm Surveys Report 2000, Australian Bureau of Agricultural and Resource Economics (ABARE), Canberra.
ABARE (2000c), Australian Grains Industry 2000: A Supplement to the Australian Farm Surveys Report 2000, Australian Bureau of Agricultural and Resource Economics (ABARE), Canberra.
ABARE (2001), Australian Commodities Forecasts and Statistics, Australian Bureau of Agricultural and Resource Economics (ABARE), Canberra. ABS (1998), Agriculture, Victoria 1996-97, ABS Catalogue 7113.2, Australian Bureau of Statistics (ABS), Canberra.
ABS (2000), Value of Principal Agricultural Commodities Produced Australia Preliminary, ABS Catalogue 7501.0, Australian Bureau of Statistics (ABS), Canberra.
ABS (2001), Value of Principal Agricultural Commodities Produced Australia Preliminary, ABS Catalogue 7501.0, Australian Bureau of Statistics (ABS), Canberra.
ABS (2001) AgStats database, Australian Bureau of Statistics (ABS), Canberra.
Adams, P. (1999), "Options for Growth", A paper presented to the NRE Growing Horizons Expert Committee, Centre for Policy Studies, Melbourne.
Agribusiness Unit (2000), Victorian Food and Fibre Export Statistics: 1990-1999, Department of Natural Resources and Environment, East Melbourne.
Agriculture, Fisheries and Forestry Australia (2001), Towards a National Framework for the Development of Environmental Management Systems in Agriculture", A public Discussion Paper prepared for the Environmental Management Systems Working Group, November.
Alston, J.M., Norton, G.W. and Pardey, P.G. (1995), Science Under Scarcity, Cornell University Press.
Appleyard, L. (1996), "Appraisal. A Spreadsheet for Evaluating the Returns to Agricultural Research", Economics Branch Working Paper 9609, Department of Natural Resources and Environment, East Melbourne.
Bedggood, A. (2001), personal communication, Victorian Institute of Dryland Agriculture (VIDA), Horsham.
Department of Natural Resources and Environment (2000), "Beneficiaries and Funders of Research and Development in Agriculture: Science Technology and Innovation Initiative 1999-2000", Evaluation Report No.1, Agriculture Division, East Melbourne.
Economics Branch (1998a), "Explaining the Role of Complex Activity Budgets in NRE's Investment Assessment System", Economics Branch Report, Department of Natural Resources and Environment, East Melbourne.
Economics Branch (1998b), "NRE's Economic Evaluation System: Methods used to evaluate research and development projects", Economics Branch Report, Department of Natural Resources and Environment, East Melbourne.
Economics Branch (2001a), EMAR-Wimmera Manual. EMAR: Economic Model of Agronomic Rotations, Department of Natural Resources and Environment, East Melbourne.
Economics Branch (2001b), EMAR-Mallee Manual. EMAR: Economic Model of Agronomic Rotations, Department of Natural Resources and Environment, East Melbourne.
Fisher, I. (1923), The making of index number, a study of their varieties, test and reliability, Second Edition, Houghton Mifflin Co., Riverside Press, Cambridge.
Grey, D. (2001), personal communication, Agriculture Victoria, Rutherglen.
Haines, P. (2001), personal communication, Agriculture Victoria, Rutherglen.
Hollaway, G. (2001), personal communication, Victorian Institute of Dryland Agriculture (VIDA), Horsham.
Jones, D. and Soligo, J. (1996), "Economic Review of the Victorian Oilseed Research Program" Working Paper 9610, Department of Natural Resources and Environment, East Melbourne.
Knopke, P., O'Donnell, V. and Shepherd, A. (2000), "Productivity Growth in the Australian Grains Industry", ABARE Research Report 2000.1, Australian Bureau of Agricultural and Resource Economics (ABARE), Canberra.
Moody, D. (2001), personal communication, Victorian Institute of Dryland Agriculture (VIDA), Horsham.
Sonogan, A. (2001), personal communication, Department of Natural Resources and Environment, Swan Hill.
Tisdell, C.A. (1972), Microeconomics: the theory of economic allocation, John Wiley and Sons, Brisbane.
Wimalasuriya, R (1999a), PRISM-Bendigo Manual. PRISM: Profitable Resource Integration, Southern MIDAS, Department of Natural Resources and Environment, Bendigo.
Wimalasuriya, R (1999b), PRISM-NE Manual. PRISM: Profitable Resource Integration, Southern MIDAS, Department of Natural Resources and Environment, Bendigo.
Using NRE's farming systems models to estimate productivity change
The NRE Economics Branch has developed a series of computer models for estimating how different research and development projects might affect on-farm productivity in Victoria's main dairy, beef, wool and grains production regions. The models use a linear programming framework.
Information on the proportion of farms affected in the region being examined is then used to estimate the gross benefit, in dollar terms, of the potential productivity improvement.
Known as 'complex activity budgets' (CABs), the computer models can take into account many different factors, including the cost and application rates of farm inputs such as fertiliser and feed, market prices for farm products, the inherent production potential of different livestock and crop varieties, rainfall, water use efficiency, and the effect of weeds, pests and diseases. Data used in the models are drawn from a range of sources, including the Australian Bureau of Agricultural and Resource Economics (ABARE), the Australian Bureau of Statistics (ABS) and NRE.
Because of their complexity, the models estimate the productivity change associated with the adoption of a new technology taking into account any on-farm adjustments that may be necessary to accommodate the new technology. For example, a farmer might devote more land to a new disease-resistant crop variety and reduce the area sown to other less profitable crops. The CAB models developed for the grains industry covers a range of crops and other farm options from which it can select the most profitable combination (in terms of gross margin) for a three, four, five or six year rotation. It can also take into account how each crop is affected by the previous two years of cropping or other activities on that land.
To estimate how a particular R&D project might affect on-farm productivity, the model is first run without the new technology that the project will produce, such as a new plant variety , improved agronomic practices, or reduced use of chemicals. This is the 'base' run or the 'without' scenario. In some cases the model will choose not to include the current practice or disease-prone crop because it is less profitable than other options – and must then be restricted in its choice of options so that the 'with and without' comparison can be made. The old technology option is then replaced by the new technology option and the model is run again in the 'with' scenario. The calculation of the change in productivity is described as follows. A Fisher index (Fisher, 1923) is used in a simple four-part procedure:
- solve the LP for its optimal solution (the 'without' scenario) – report the inputs and outputs in an 'input-output table';
- alter one or more parameters in the LP according to the technical change being evaluated;
- re-solve the LP for its new optimal solution (the 'with' scenario) – report new input-output table; and
- calculate the productivity change with input-output information, holding the output level constant. In general, the more significant an input is in terms of costs, the larger the overall productivity change that is likely to arise from reducing the use of that input.
NRE currently has four grains CAB models representing different farming systems in Victoria. The EMAR-Wimmera (Economic Model of Agronomic Rotations – Wimmera) model represents the continuous cropping system in the Wimmera region and the EMAR-Mallee model represents the medic pasture-crop rotation system in the Mallee. PRISM-Bendigo (Profitable Resource Integration, Southern MIDAS – Bendigo) and PRISM-NE represent the subclover pasture-crop rotation systems in the North Central and North Eastern regions of Victoria, respectively. (MIDAS stands for Model of an Integrated Dryland Agricultural System.) The models have been extensively reviewed with the assistance of NRE and external grains industry experts.
Each model is calibrated for the top 25 per cent of farmers in that industry. This is because these farmers are more likely to be at the forefront of farm technology and more interested in considering and adopting new technologies.
More information about the CAB models, including a full description of all assumptions and data, can be found in Economics Branch (2001a, 2001b) and Wimalasuriya (1999a, 1999b).
The impact of technology change
Most research-induced technology changes have the effect of shifting the aggregate supply curve downwards as shown in Figure 53. Every producer is able to produce each unit of output at a lower cost, due to the new technology.
Figure 5: Downward shift of the commodity supply curve due to technical change
In Figure 5, the initial equilibrium before the adoption of a cost reducing technology is E0Q0. Following the introduction of the cost-reducing technology, the new equilibrium is E1 as the cost reduction now allows quantity Q1 to be supplied.
Following Alston et al. (1995), the cost reduction from a technological change is denoted by the letter k. The k associated with the first round of a technology impact, that is, before any alterations to the farming system take place, is denoted k1. It is the impact that occurs when a technology is adopted, but the use of (other) inputs remains constant. Since the purpose behind research is to prompt a change in on-farm practices, a more satisfactory description of the technology's impact is one that takes account of the altered producer behaviour. This is measured by k2, the cost reduction after input use has changed. In Figure 5, Q1 represents the new equilibrium after input and output substitution has occurred. Therefore, the downward shift in the supply curve is of the k2 type.
1 NRE has developed whole-farm models to estimate on-farm productivity improvements from new technologies. Known as Complex Activity Budgets (CABs), these models are explained in Appendix 1 and in Economics Branch (1998a)
2 In the fast-track Single Seed Descent method, large numbers of small chickpea plants with two to four seeds per plant are grown in a glasshouse. By screening three generations a year, the method produces pure breeding varieties with desirable traits significantly faster than conventional breeding methods.
3 For an explanation of demand and supply, see Tisdell (1972). The demand curve shown in Figure 5 is illustrative only.