Landuse Change, Productivity & Development - Projections
Final Report of Theme 5.1 to the National Land & Water Resources Audit
August 2001
6. Projections
Summary
Forecasts are assessments of the likely consequences of projecting causal factors but are about an inherently unknowable future.
Many analysts expect agriculture based on food is likely to suffer declining prices and individual incomes in real terms. In response farmers will leaving the industry, and reallocating resources to more profitable uses leading to fewer, larger farms.
A simple rule of thumb projection assumes that the next 20 years are going to be similar to the last 20 with, cattle and sheep numbers and crop areas fluctuating in response to climatic conditions and market prices within the present bounds. This is most likely for the gross pattern of land uses although there will be ongoing forces for change and some opposing inertial forces at the local and regional scale.
An extrapolation of a linear regression of the increase in total DSE with time indicates a rate of 3 million DSE per year to 2020 would give a total productive capacity of about 520 million DSE compared to a level for 2000 of about 495 million DSE.
Long term perspectives of the changing patterns of land use in broadacre industries were undertaken for several scenarios of productivity change in Australian agriculture relative to that in the rest of the world. The projected land use changes under these cases can also be large with strong regional differences. This is because a key factor determining the responsiveness of land use to changing relative returns to land is the land supply conditions for different activities, which in turn differ by region. These differences are due to factors such as the production risk associated with different climatic and physical factors, hence the premium landholders require before they are likely to change land use varies between activities and regions.
From a management perspective, sustainable agriculture is mostly about processes and practices. Productivity increases imply some intensification, considered to be a speeding up some of the processes in agricultural and natural systems. In turn this has implications for land degradation and its effect upon productivity, which will vary with local conditions. To continue to improve management for sustainable agriculture, it is likely that there will need to be broader planning in several spheres - in government, industry and on farm - in order to avoid escalation of unintended consequences of such intensification.
Just as the present is the result of past actions so the future will be develop as the result of actions from many individual decisions being made now and which will be influenced by perceived pressures and challenges. Because of the scale and uncertainties that relate to future pressures and challenges, only some broad directions can be noted here and the influence of any major chaotic events that will occur has to be ignored.
Forecasts are assessments of the likely consequences of projecting causal factors. They were undertaken as aides to targetting policies for sustainable land mangement and rural adjustment, and strategic planning for organisations.
Forecasts are about an inherently unknowable future because of inexact data, imperfectly understood responses to known causes, and many unknown causes.
For instance, we are confident that there will be at least two major droughts in Australia during the next 20 years. However, we are not able to be confident about when they will occur or where they will concentrate. Similarly with forecasts about floods and market downturns. We can also be confident that land uses will change in response to challenges and improvements in technology. Therefore any long-term strategic planning should incorporate various scenarios and include responses for different contingencies.
A number of groups have conducted exercises investigating possible futures for agriculture:
International Food Policy Research Institute (IFPRI) - 2020 Vision for Food, Agriculture, and the Environment which is an ongoing exercise since 1994 with an emphasis on food security for developing countries;
• the OECD Forum for the Future in 1997 (OECD 1998) which produced a "baseline" scenario subject to uncertainties in demand, supply and policy; and
• FAO prospective studies of world agriculture eg World Agriculture: Towards 2010 in 1995 describing a most likely development path rather than extrapolations of trends. FAO released a Technical Interim Report of an upgrade of projections on Agriculture: Towards 2015/30, as at April 2000. The final report was released in 2003.
Their major conclusions are included here.
6.1 Factors influencing forecasts on the future of agriculture
Most forecasts assume that food supply will be highly responsive to price signals in the two next decades so much of the emphasis is on estimating the potential demand first.
Given the caveats about the accuracy of forecasting mentioned above, the long-term prospects for food and agriculture are likely to be determined by these major factors:
demand
the demand for food is likely to grow in response to continuing increases in population (de Haen et al 1998).
• through changes in consumption patterns where the meat sector is particularly uncertain and influenced by cultural preferences, concerns over food safety and health-related issues (de Haen et al 1998)
• through increasing attention that consumers pay to the quality and safety of food, a concern that extends to the environmental friendliness of the entire production and transformation process (de Haen et al 1998, FAO 2000); and
• development rates in some countries particularly recovery in the former CPEs of Europe that will influence consumption per head and consumer preferences (de Haen et al 1998, Lahidji et al 1998)
supply
• biotechnologies which are not expected to have much influence before 2010 but thereafter through hybrid rice production, and apomictic production (de Haen et al 1998).
• Information technology which however could influence before then (de Haen et al 1998).
• Productivity gains from traditional techniques (Lahidji et al 1998, White and Walcott 2000).
externalities to agriculture
• to safeguard natural resources where markets fail (eg land use planning, water use efficiencies, conservation and use of genetic resources) (de Haen et al 1998);
• further changes to international agricultural trade rules (de Haen et al 1998) personal responses
• which are rarely discussed in projections.
personal responses
Emerging trends
These trends are based on expectations that present changes and indications will increase in the future. International trends are important for considerations about land use in Australia because the EC and the USA are major markets for Australian products as well as being competitor producers and are often leading indicators of future trends for Australia. They often are influences on future actions that are not catered for in projections. For instance, Henderson (1998) claims that food consumption patterns around the world are converging and that changes in food consumption in some countries (eg USA) lead changes in others, aided by population mobility and globalisation.
Many analysts (Lahidji et al 1998, Edwards 2000, McGauchie 1998) expect that agriculture will become increasingly integrated with the food industry. It is likely changes in the agro-food sector will be dramatic over the next twenty years driven by changes in consumer tastes, in technologies and in the underlying competitive positions. The market could become more differentiated, necessitating the use of more sophisticated technologies, contract production, and complex responses such as identity preservation and quality assurance.
Major trends identified that are likely to have a bearing on agriculture, include:
Through demand
Through supply
And through external factors
• Continued risk of disaster-related and complex emergencies (FAO 2000);
• Increasing impact of information and communications technology on institutions and societies (FAO 2000);
• The concern by many countries to take the multi-functionality of agriculture (economic, social, environmental etc) into consideration and shift policy from agricultural viability into broader rural development (Lahidji et al 1998);
Since those comments were made other issues have arisen such as: challenges to globalisation through popular demonstrations at international meetings; the further outbreaks of BSE and Foot&Mouth Disease in Europe sharpening consumer concerns about food safety; and sharp increases in energy prices similar to the oil shocks of the 1970s.
Possible responses
To these challenges there may be set a range of possible responses. McGauchie (1998) attests to the ability of the farm sector, given their record in the past 20 years, to organise itself to meet new challenges and changing circumstances and adopt new technology. He considers a move is likely towards specific product farming and marketing, designed to service emerging niche markets with consistency of quality as important. He also foresees that large, capital-intensive corporate-owned industrialised enterprises will be one development, particularly in intensive forms where processes can be controlled and climate variability minimised (wine, pig, poultry, feedlot and irrigation crops). The family farm is more likely to prevail in broadacre agriculture where a premium applies to individual management skills. He sees too, that farmers will need to be part of more formal quality assurance procedures In turn this will necessitate smarter farming and better-trained farmers.
Henderson (1998) predicts that, in the procurement of agricultural commodities for food processing in the United States, the use of contracts and vertical integration is particularly noticeable with broilers, eggs, milk, processing vegetables, potatoes and citrus fruit, but is increasing rapidly for lamb, fruit, vegetables, and pigs. Similar responses are suggested for Europe by Viatte and Shmidhuber (1998).
Innovations, particularly biotechnology’s ability to change the emphasis from selection on phenotype to selection directly on genotype (Paillotin 1998, Peacock/Edwards 2000), could lead to moves to change or control the environment to fit the genotype as in factory farming (Edwards 2000).
Globalisation can be countered (Bailey et al 2000) by productivity growth in agriculture and/or by moving into more non-tradeable and elastic demand products such as services. Agriculture based on food is likely to suffer declining individual incomes in real terms. This might be countered by farmers leaving the industry, and reallocating resources to more profitable uses, or the farm business might expand to take advantage of economies of scale in production. Both factors lead to fewer but larger farms. Other options include using land for production of non-food crops, providing environmental goods in return for transfer payments and the supply of services to the rest of the economy. Examples of services include rural accommodation, eco-tourism, sporting and recreational facilities ( eg equine services) or products such as timber, biofuel, Christmas trees, industrial oil crops, and fibre.
6.2 Projections
Simple rule of thumb
The first method of projecting uses a rule of thumb and assumes that the next 20 years are going to be similar to the last 20. In other words, cattle and sheep numbers and crop areas will fluctuate in response to climatic conditions and market prices within the present bounds. For instance, Cocks (2000) expects little change in the broader landscapes in Australia during the next few decades for the gross pattern of land use although, at the local and regional scale, there will be ongoing forces for change and some opposing inertial forces.
Some coarse calculations for the whole of Australia were made by converting broadacre productivity to dry sheep equivalents (DSEs) as detailed in Appendix2-Methods.
Viewing the last 25 years of productivity within the broadacre industries in Figure 6-1 could suggest that a limit to productive capacity has been reached. This would support the rule of thumb projection.
Extrapolation
Alternatively, taking a longer viewpoint shown in Figure 6-1 indicates there is still a trend for improved total productivity, albeit with severe aberrations occurring at times. A simple linear regression of the increase in total DSE with time indicates a rate of 3 million DSE per year with a coefficient of variation of 90 per cent. Were this to continue at the same rate until 2020 then total productive capacity would be about 520 million DSE. This extrapolation is based on coarse assumptions, including that innovations will continue to contribute to expanding productive capacity, that there is sufficient slack in the total capacity within Australia to deliver it, and that market demand increases for those products. Nevertheless, since the present level for year 2000 was about 495 million DSE, this target does appears feasible. It does not mean that the present mix of products remains the same however. With the ‘water cap’ in the Murray-Darling Basin, there is likely to be further adjustment in the products that come from irrigation to those of higher value, which are not included in these estimates.
Figure 6-1 Total biological productivity calculated as Dry Sheep Equivalents (DSEs) for sheep, cattle, grains and total in Australia since 1860. (source ABS)
Modelled forecasts
The next systems is to use models, that incorporate multiple factors and interactions, to make forecasts. Such efforts take up the rest of this section.
Forecast to 2005
Modelled forecasts necessarily ignore the impacts of the uncertainties of climate variability, such as El Nino, on outcomes.
The OECD Agricultural Outlook (OECD 2000) provides a medium term projection of future trends and prospects in the major agricultural commodity markets. These projections are a “plausible medium-term future”, made by commodity experts based on a number of assumptions concerning policies, macro-economic environment and developments in non-OECD countries, and use the OECD’s Aglink model to ensure consistency and generate scenarios. It forecasts:
• farm production in the OECD may stagnate in the near term;• world production of cereals is forecast to show the strongest gains between 1999 and 2005, followed by oilseeds and sugar. A major growth is seen in feed use of cereals and oilseeds;
• larger total supplies of meat are projected to 2005 especially for pork and poultry, with a slackening in demand for red meat in developed countries;
• the main growth in dairy product is expected to be cheese, followed by whole milk powder and butter.
| COMMODITY | World ave projected 95-99 05/06 | OECD exports (kt)94-98 2005 | Australia production (kt) 95-99 05/06 | |||
|---|---|---|---|---|---|---|
| Wheat mt | 566.4 | 648.8 | 81241 | 99035 | 18.1 | 22.7 |
| Coarse grains mt | 876.3 | 997.2 | 77894 | 90697 | 9.4 | 9.8 |
| Rice mt | 376.3 | 422.2 | 4889 | 4860 | 0.9 | 1.1 |
| Oilseeds mt | 199.9 | 248.7 | 28942 | 33377 | 1.0 | 2.4 |
| Beef & veal kt cwe | 26443 | 27399 | 4697 | 5641 | 1845 | 2015 |
| Pork kt cwe | 32363 | 35996 | 2054 | 2972 | 354 | 378 |
| Poultry kt rtc | 28665 | 36439 | 3277 | 4572 | 539 | 692 |
| Sheep meat kt cwe | 2741 | 2581 | 868 | 856 | 604 | 557 |
| Butter kt pw | 6629 | 7221 | 631 | 696 | 144 | 190 |
| Cheese kt pw | 13624 | 16141 | 1023 | 1354 | 270 | 438 |
| Skim milk powder kt pw | 3320 | 3157 | 881 | 903 | 212 | 280 |
| Whole milk powder kt pw | 2462 | 2957 | 1008 | 1231 | 115 | 159 |
| Sugar mt | 119.2 | 139.2 | 5.4 | 6.1 | ||
Table 6-1 provides a summary of world production, developed country (OECD) total export, and Australian production for 1995-99 average and projected to 2005-06. Such projections assume that the competitive export positions for its members remain as at present. The main features are big rises in all dairy products, poultry and beef. More modest increases for grains and sugar, but a decline for sheep meat.
Forecast to 2010
The FAO “most likely” development path for food and agriculture is based on the medium rate projection for world population and modest increases in per capita food supplies as shown in Table (de Haen et al 1998). FAO projects that total domestic demand will rise by 0.5 per cent in the developed countries while production could grow by about 0.7 per cent in the developed countries.
Table 6-2 Comparison of key drivers of population and food consumption in long-term projections (FAO )
| Year | Population (billion) - medium rate projection | Per Capita food supplies (cal/day) |
|---|---|---|
| 1970 | 3.86 | 2,420 |
| 1980 | 4.43 | 2,560 |
| 1990 | 5.28 | 2,710 |
| 2000 | ||
| 2010 | 6.89 | 2,900 |
| 2020 | 7.8 |
Trade in agricultural commodities may well remain at about 10 per cent of world production, although trade in processed food may grow faster (Lahidji et al 1998). The US, Australia and New Zealand might well gain substantial market shares in world exports, particularly of livestock products and processed foods.
FAO “most likely” development path for food and agriculture is for total world demand to increase by 45% from 1990 until 2010 (de Haen et al 1998). Although most of this is expected to occur within developing countries there is likely to be further growth of net imports of cereals from the developed countries, which may grow from 90 million tons in 1988/90 to about 160 million tones in 2010. Meeting such productivity goals increase the pressure on sustainable farming systems.
The consensus from three studies undertaking projections of food supply (FAO, IFPRI and World Bank) was that world food supply in 2010 would probably meet global demand but that regional problems could occur. The FAO study does not project prices but the IFPRI study does (de Haen et al 1998), and concludes that the long-term downward trend in real prices will continue. An indication of the potential range in the projections is that net trade from developed countries in cereals by 2010 was estimated to vary from 151, 157 and 195 million tons by IFPRI, FAO and World Bank respectively.
Forecast to 2020
Beyond 2010 strong caveats apply as projections become more problematic (de Haen et al 1998) because there is sufficient time for tectonic shifts, or disturbances, to occur eg collapse of the USSR in 1989 and the Asian economic crisis of 1997. Even at this time frame however, climate change per se was not considered an important factor. Considerations not easily accounted for could include:
• the freedom of movement of people and food trade;
• concerns about the capacity of natural resources to support projected populations;
• government action in various parts of the world particularly with respect to the protection of agriculture; the protection of natural resources (pesticide and fertiliser use, greenhouse gas emissions); attitude to scientific innovations; and access to resources particularly water.
• investment in the innovation process and its distribution of benefits.
An example of the influence of unknown factors is that of the Asian economic crisis of 1997. Changes to its effects on agricultural food supply and prices were conducted by Rosegrant and Ringler (2000) using the IFPRI IMPACT model. Per capita demand for livestock products was little affected in developed countries, but reduced by 3 kg/capita under severe crisis (but still an increase over 1993 levels) and by 1 kg/capita with moderate crisis compared to baseline of no Asian financial crisis.
FAO conclude that it should be technically possible to increase world food supplies by as much as required by the growth of effective demand (Agriculture: Towards 2015/30, Technical Interim Report, April 2000).
IFPRI projections presented by Pinstrup-Andersen and Pandya-Lorch (1998) have global food production exceeding demand growth and thus real prices on all food commodities likely to decrease by 2020. This could be as much as 5 per cent for beef and 13.5 per cent for wheat. However, supply by developed countries could still increase by 1 per cent per annum for wheat, and 0.75 per cent for beef.
Modelled productivity changes
This component provides long term perspectives of the changing patterns of land use that are independent of the short-term influence of seasonal conditions. The Australian Bureau of Agricultural and Resource Economics (ABARE) modelled projections of future land use patterns in broadacre industries using several scenarios of productivity change in Australian agriculture relative to that in the rest of the world. The scenarios ignored effects on the semi-intensive and intensive land uses.
While the model and its assumptions are given in some detail in Appendix2-Methods, some brief description is provided here. The model is based on competitive markets for agricultural inputs and outputs. That is, for each land use activity in each region of Australia the prices of inputs and outputs are independent of the level of the activity. Instead, prices for all outputs were set to equal world prices less any transport and other margins. The exception is wool for which Australian production affects the world price.
For each agricultural activity in the model, output is assumed to depend on land and non-land inputs according to a constant-returns-to-scale production function. It assumes that there are unlimited supplies of non-land inputs at the given prices. However, for the land input, producers are assumed to require an annual rental that rises proportionally with the quantity of land supplied to the activity. That is, as more land is used in a particular activity landholders are assumed to require a gradually increasing premium before they will choose to allocate an extra hectare to that activity. This assumption prevents land use change to the extent where there is complete specialisation in a region.
The model is based on five commodity classes and five regions. These classes cover the five main types of broadacre activities: cereals, oilseeds, other crops, sheep and beef cattle. The regions (Figure 6-2) include the pastoral zone, high rainfall zone, and a breakdown of the wheat-sheep zone into three regions: the northern, southern and western wheat-sheep zones. The base year for the model was selected as 1996-97 to agree with the base year used by the Audit. Further details are given in the Appendix2-Methods.
Four cases that compare the productivity of Australia’s agricultural industries with that in the rest of the world were selected for analysis. A higher Australian productivity level was assumed in all cases - if lower levels of productivity were considered, the land use changes would be approximately the same in percentage points but of the opposite sign. A higher level of productivity may be obtained, for example, because a new innovation is introduced in Australia that can not be readily adopted overseas. Similarly, a lower level of productivity could occur because of more land degradation in Australian agriculture, or a disease outbreak may damage agriculture in Australia but not overseas. However, this framework is not appropriate for analysing other shocks such as changes in Australian input prices or demand.
The base case assumes that the productivity of Australia’s agricultural industries changes in the same way as the rest of the world. In that case, Australia is assumed not to face any change in terms of trade and therefore there is no consequent change in the proportion of area allocated to the production of each activity (Table 6-4).
Figure 6-2 Major regions used for productivity and land use projections
Table 6-3 . Area estimates (thousands of hectares) in 1996-97 for each commodity classification by region
| Region | ||||||
|---|---|---|---|---|---|---|
| ABARE Commodity Classification | Pastoral zone | High rainfall zone | Northern wheat sheep | Southern wheat sheep | Western wheat sheep | Australia |
| Cereals | 175 | 569 | 2018 | 5649 | 5409 | 13820 |
| Oilseed | 0 | 52 | 149 | 314 | 236 | 751 |
| Other Crops | 138 | 775 | 544 | 1738 | 1469 | 4664 |
| Sheep | 83319 | 7715 | 4874 | 17978 | 9817 | 123703 |
| Beef | 247074 | 15684 | 22268 | 6710 | 471 | 292208 |
| Total | 330706 | 24796 | 29852 | 32389 | 17402 | 435146 |
Table 6-4. Proportion of land in each region used for each activity - base case 1996-97.
| Region | Cereals | Oilseeds | Other crops | Sheep | Beef |
|---|---|---|---|---|---|
| Pastoral zone | 0.0005 | 0.0000 | 0.0004 | 0.2519 | 0.7471 |
| High rainfall zone | 0.0230 | 0.0021 | 0.0312 | 0.3111 | 0.6325 |
| Nothern wheat sheep | 0.0676 | 0.0050 | 0.0182 | 0.1633 | 0.7459 |
| Southern wheat sheep | 0.1744 | 0.0097 | 0.0537 | 0.5551 | 0.2072 |
| Western wheat sheep | 0.3108 | 0.0136 | 0.0844 | 0.5641 | 0.0271 |
| Total | 0.0318 | 0.0017 | 0.0107 | 0.2843 | 0.6715 |
The four productivity assumptions selected for analysis were:
• Case 2 - Sheep: the index of total factor productivity in Australia's sheep industry is assumed to be 10 per cent higher than in the baseline in 2010, and 20 per cent higher in 2020.
• Case 3 - Beef: the index of total factor productivity in Australia's beef industry is assumed to be 10 per cent higher than in the baseline in 2010, and 20 per cent higher in 2020.
• Case 4 - Crops: the index of total factor productivity in Australia's cropping industries is assumed to be 10 per cent higher than in the baseline in 2010, and 20 per cent higher in 2020.
Obviously, it is unrealistic to expect a constant increase in productivity over 20 years, and a gain of 1 per cent per annum is overly optimistic, but the scenario outputs do provide some boundaries for possible changes.
National level projections
For Case 1 - all broadacre sectors increase by 1 per cent per annum - by 2020 a 20 per cent higher productivity in all activities results in a shift, at the national level, in land use from sheep production (down from 28 to 10 per cent of total area) to beef (up from 67 to 83 per cent of total area) and crop production (up from 4 to 7 per cent), compared to the base case (Table 6-5). The prices received for beef and crop production remain the same in this case because Australia is a price taker in the world market for these commodities. However, as Australia is a large producer of wool the higher productivity of the Australian sheep activity results in a lower world price of wool. This lower price of wool offsets, to some extent, the higher productivity that leads to higher land rentals in sheep production. Therefore, returns to beef and crops are more favourable relative to sheep production so that resources have moved to these activities from the sheep industry.
Table 6-5. Effects of projection to 2020 for the base case and for 20 per cent increases of productivity in a range of of activities on the proportion of land used for those activities.
| Proportion of land in each region used for each activity | |||||
|---|---|---|---|---|---|
| Case | Cereals | Oilseeds | Other crops | Sheep | Beef |
| Base | 0.0318 | 0.0017 | 0.0107 | 0.2843 | 0.6715 |
| Case 1 - all increase | 0.0456 | 0.0102 | 0.0170 | 0.0990 | 0.8282 |
| Case 2 - sheep increase | 0.0316 | 0.0017 | 0.0107 | 0.4402 | 0.5159 |
| Case 3 - beef increase | 0.0312 | 0.0015 | 0.0106 | 0.0757 | 0.8810 |
| Case 4 - crop increase | 0.0465 | 0.0105 | 0.0173 | 0.3220 | 0.6038 |
This discussion will emphasise the beef industry to illustrate changes in land use as it is the most widespread of the industries considered. An interpretation of the geographic distribution of present land use activities devoted to beef production is given in Figure 6-3. As might be expected, the lowest proportions of beef cattle are in the western wheat-sheep zone and the western parts of the southern wheat-sheep zone. High uses occur in northern Australia and also in pockets in the high rainfall zone of eastern and western Australia. An example of projected changes is given in Figure 6-4 for Case 1 at 2020 for the beef industry land use area of a uniform 1 per cent per annum increase in productivity for all activities. It projects an expansion southwards of the tropical beef production area, but less cattle in the southern wheat-sheep zone and little change in the high rainfall zone.
For Case 2 - only the sheep industry achieves increases of 1 per cent per annum -the area used for sheep production is estimated to be larger than in the base case the expense of a correspondingly smaller beef area (Table 6-5). There is virtually no change in the area of crops at the national level. The returns to using land for sheep production are improved as the increase in productivity outweighs the associated fall in the world price of wool. With greater returns to land in the sheep activity relative to alternative activities, more land is used in this activity.
Case 3 - only the beef industry increases by 1 per cent per annum. The main effect (Table 6-5) of the relatively higher productivity in the Australian beef industry than in the base case by is a shift in land use from sheep into beef production. There would only be a small change in land use from cropping activities to the beef activity.
Case 4 - only the cropping industry increases productivity by 1 per cent per annum. A 20 per cent higher productivity of the Australian cropping activities relative to the base case by 2020 would result (Table 6-5) in a larger area used for cereals, oilseeds and other crops (from 4.4 percent to 7.4 percent). It also leads to a slightly larger area used for sheep production. This is at the expense of beef production, which falls from 67 per cent to 60 per cent.
Figure 6-3. Distribution of the proportion of agricultural land used for beef production in 1997 used as the base case for projections. (source ABARE)
Figure 6-4. Projection to 2020 of likely proportion of land use for beef production where all activities increase productivity by 1 per cent per annum. (source ABARE)
Regional level projections
Results for some of the regions reflect the movements observed at the national level. However, there are different responses between the regions for some land uses. This is because of the different input mixes used in each region as well as differences in land supply conditions between regions and between activities within regions. That is, changes in land rentals and land use can vary widely between regions and between activities.
Within the Western wheat-sheep zone, the impact of the different cases on the major products (cereals, oilseeds and other crops were combined into crops) is shown in Figure 6-5. Beef cattle (indicated by solid lines), which begin at a low level, do not change under Case 2 (sheep productivity increases) and Case 4 (Crop productivity increases), but more than doubles under Case 1 (All increase in productivity) and Case 3 (beef productivity increases). Sheep (indicated by dotted lines) are at present the dominant land use but remain at this level only in Case 2 (sheep productivity increases). In all other cases the proportion of sheep declines though only by 13 per cent in Case 3 (beef productivity increases) but by half in Case 1 (All increase in productivity) and 40 per cent in Case 4 (Crop productivity increases). In contrast, the area of crop (shown in dashed lines) is unlikely to contract, with little change projected for Case 2 (sheep productivity increases) and Case 3 (beef productivity increases), but increases of more than 20 per cent for Case 1 (All increase in productivity) and Case 4 (Crop productivity increases).
Within the Southern wheat-sheep zone (in Figure 6-6) the area devoted to beef cattle production (solid lines) appears the most vulnerable to change (a potential range from 3 to 25 per cent of total agricultural area). The area devoted to beef only increases under Case 3 (beef productivity increases). In all other cases the area declines, most noticeably in Case 1 (All increase in productivity) when cattle activities almost cease. The area devoted to sheep production (dotted lines) appears likely to remain the dominant land use under all these scenarios. The area is projected to increase in Case 1 and Case 2 (sheep productivity increases), to decline under Case 3 but to remain relatively static in Case 4. The area under crops (dashed lines) increases under Case 1 and Case 4 (Crop productivity increases), but to remain almost unchanged under Case 2 and Case 3.
Figure 6-5. Projections in the Western wheat-sheep zone to 2010 and 2020 for the proportion of agricultural land devoted to different land uses under 4 different cases presented in the text. (source ABARE)
Figure 6-6. Projections in the Southern wheat-sheep zone to 2010 and 2020 for the proportion of agricultural land devoted to different land uses under 4 different cases presented in the text. (source ABARE)
In the Northern wheat-sheep zone (Figure 6-7) beef cattle (solid lines) is easily the largest land use under all Cases. In Case 1 (All increase in productivity) and Case 3 (beef productivity increases) the area increases to a maximum by 2010, whereas in Case 2 and Case 4 the area declines slightly to 2020. The area devoted to sheep production (dotted lines) declines to zero by 2010 under Case 1 and Case 3, does not change appreciably in Case 4 but does increase in Case 2 (sheep productivity increases). The area under crop (dashed lines) appears little affected by productivity changes under any Case.
Figure 6-7. Projections in the Northern wheat-sheep zone to 2010 and 2020 for the proportion of agricultural land devoted to different land uses under 4 different cases presented in the text. (source ABARE)
The High rainfall zone (Figure 6-8) presents some interesting possibilities because of the different input mixes used and different land supply conditions from the wheat-sheep zones. The area used for beef cattle production (solid lines) remains the major land use under all scenarios. However, although under Case 1 (All increase in productivity) it declines most, almost down to the level of sheep area by 2020. This is because the higher productivity for all activities, with the associated effect on the wool price, result in higher land use for sheep and crops at the expense of beef. Under Case 3 (beef productivity increases) beef areas stay essentially the same. Here, the nation-wide large land use change out of sheep production leads to a higher world wool price, which in the high rainfall zone is sufficient to outweigh the higher beef productivity. In Case 2 (sheep productivity increases) beef area declines only marginally with only a slight land use change to sheep production, which is restricted to this zone due to it's particular input mix and land supply conditions. Sheep production area (dotted lines) is projected to either remain the same (Case 3 and Case 4) or to increase under Case 1, but only marginally under Case 2 (sheep productivity increases). For cropping areas (dashed lines) there is no change projected under Case 2 and Case 3 but substantial increases under Case 1 and Case 4 (Crop productivity increases). It should be noted here that physical constraints to the adoption of a particular land use are not specifically considered in these projections.
Figure 6-8. Projections in the High rainfall zone to 2010 and 2020 for the proportion of agricultural land devoted to different land uses under 4 different cases presented in the text. (source ABARE)
Conclusions
The productivity cases considered in this analysis are for quite large differences in productivity growth between different broadacre sectors, both within Australia and compared with the rest of the world. The projected land use changes under these cases are also large. A key factor determining the responsiveness of land use to changing relative returns to land is the land supply conditions for different activities, which in turn differ by region. These differences are due to factors such as the production risk associated with different climatic and physical factors, hence the premium landholders require before they are likely to change land use varies between activities and regions.
The results also highlight the importance of accounting for the effect of changed supply on prices for commodities where Australian is a large producer. For example, Australia can affect the world price of wool by the volume of wool it produces.
The area of land used for beef production at the national level seems the most sensitive to the assumed productivity levels. Land use changes into and out of sheep production are moderated by the counterbalancing effects on world wool prices and a substantial extra premium is generally needed before large areas of land are moved into cropping.
However, there are some strong regional differences. For example, in the western wheat-sheep zone the proportion of land used for cropping and sheep production is very sensitive to changes in returns to land in cropping activities.
Industry strategies
The wine industry has produced a Strategy 2025 which included the vision of achieving $4.5 billion in annual sales by 2025. The First Five Year Plan (1997-2001) emphasises establishing volume growth and set targets for wine exports of 259 ML and $896 million by 2001, and also boosting domestic consumption from 18.0 L/head to 18.6 L/head. To achieve this wine grape supply needed to be 1 m tonnes.
The Winemakers' Federation of Australia estimated a total intake for 1999 of 1,177,610 tonnes of grapes and exports of 216.2 ML worth an estimated A$991.2 million, so the industry appears on target for its goals.
The Meat Industry Strategic Plan of 1995 aimed to add $3.8 billion in revenue to beef industry and $0.8 billion to sheep meat industry during 1995-2001. For beef, it emphasised increasing exports to Japan to $5.2 billion pa, improved market share to Korea, and increased domestic consumption per capita. For sheep meat the emphasis was to increase total returns to $2.0 billion by 2000 including $0.5 billion in export. By the 1999-2000 year exports of beef were valued at $3.1 billion (with 39 per cent to the USA and 36 per cent to Japan, but only 8 per cent to Korea).
The Australian Dairy Industry Council commenced a Dairy Vision 2010 project in November 1999 to be patterned on that undertaken by the Wine Industry. Although it aimed at preparing the industry for deregulation, and integration along the value chain and international impacts such as the New Zealand restructure, the initiative appears to have been submerged in the course of events around deregulation and processor rationalisation.
6.3 Discussion on projections
Forecasts are assessments of the likely consequences of projecting causal factors but are about an inherently unknowable future. For instance, we are confident that there will be at least two major droughts in Australia during the next 20 years. However, we are not able to be confident about when they will occur or where they will concentrate. Similarly with forecasts about floods and market downturns. Therefore any long-term strategic planning should incorporate various scenarios and include responses for different contingencies.
Many analysts expect that agriculture will become increasingly integrated with the food industry with more use of contracts and vertical integration. These changes in the agro-food sector could be dramatic over the next twenty years if some anticipated changes in consumer tastes, in technologies and in the underlying competitive positions come to pass. Agriculture based on food is likely to suffer declining individual incomes in real terms. In response farmers will leaving the industry, and reallocating resources to more profitable uses leading to fewer, larger farms.
A simple rule of thumb projection assumes that the next 20 years are going to be similar to the last 20 with, cattle and sheep numbers and crop areas fluctuating in response to climatic conditions and market prices within the present bounds. This is most likely for the gross pattern of land uses although there will be ongoing forces for change and some opposing inertial forces at the local and regional scale. A simple extrapolation of a linear regression of the increase in total DSE with time indicates a rate of 3 million DSE per year to 2020 would give a total productive capacity of about 520 million DSE compared to a level for 2000 of about 495 million DSE.
The Food and Agriculture Organisation (FAO), the International Food Policy Institute (IFPRI) and the Organisation of Economic Co-operation and Development (OECD) have made projections about possible futures for world agriculture for times up to 2030. Although demand for food is expected to increase, because of increasing population and consumption per head, most of the expected increase in production will not occur in developed countries. Production could increase at about 0.7 per cent per annum. The major growth areas are expected in wheat, oilseeds, beef, poultry, and processed dairy products. Sheep meat is projected to decline slightly. The long-term trend for prices to decline is expected to continue. However, considerable uncertainties pertain to these projections with 3 projections indicate total cereal production by 2010 ranged between 151 and 195 million tonnes, nearly a 30 per cent difference.
Long term perspectives of the changing patterns of land use in broadacre industries that are independent of the short-term influence of seasonal conditions were gained from ABARE modelling of several scenarios of productivity change in Australian agriculture relative to that in the rest of the world. From the base level of no change from world parity four productivity scenarios were projected to 2020. The cases selected for analysis were to increase the index of total factor productivity, by approximately 1 percent per annum, to be 10 per cent higher than in the baseline in 2010, and 20 per cent higher in 2020 in Australia for all broadacre activities, sheep industry only, beef industry only, and crops industry only.
The projected land use changes under these cases are large. A key factor determining the responsiveness of land use to changing relative returns to land is the land supply conditions for different activities, which in turn differ by region. These differences are due to factors such as the production risk associated with different climatic and physical factors, hence the premium landholders require before they are likely to change land use varies between activities and regions.
The results also highlight the importance of accounting for the effect of changed supply on prices for commodities where Australian is a large producer. For example, Australia can affect the world price of wool by the volume of wool it produces. The area of land used for beef production at the national level seems the most sensitive to the assumed productivity levels. Land use changes into and out of sheep production are moderated by the counterbalancing effects on world wool prices and a substantial extra premium is generally needed before large areas of land are moved into cropping.
However, there are some strong regional differences. For example, in the western wheat-sheep zone the proportion of land used for cropping and sheep production is very sensitive to changes in returns to land in cropping activities.
>From a management perspective, sustainable agriculture is a mostly about processes and practices. Intensification can be considered as speeding up some of the processes in agricultural and natural systems. To continue to improve management, it is likely that sometime there will need to be broader planning at farm landscape scale, in some cases in close collaboration with neighbours in a catchment for some issues to be resolved (Hamblin 2000). Similarly industries will need to undertake serious strategic planning and setting measurable targets.