Information and Advsiory Note Number 68, December 1996
This report was commissioned by SNH from Doyle et al. of Scottish Agricultural College, Auchincruive. It seeks to predict the likely effect of replacing the current mechanism of support with natural heritage incentives. It provides some challenging conclusions, which are based on a number of debatable assumptions.
The executive summary of the study as provided by the consultants is followed by an SNH commentary.
2.1 Background and aims of study
There has been a widespread consensus of opinion that changes are required in agricultural support payments to better reflect the needs of the Community. Thus, the recent reports initiated by the Ministry of Agriculture, Fisheries and Food, the National Farmers’ Union of England and the Scottish Landowners’ Federation have recognised that environmental considerations will need to be increasingly incorporated into agricultural policy and that farmers should be paid for providing environmental services. However while there is general recognition that support methods must change as a consequence, there is a shortage of proposals regarding how a change in emphasis from production- to environmentally-linked payments might be effected. Even less information is available on the consequences of any such change on the social and economic structure of rural areas. Accordingly, this study was commissioned by SNH with the objectives of:
1. considering ways in which current public support for agricultural production might be re-directed into environmental support;
2. estimating the social and economic consequences of such a shift on the Scottish rural economy in terms of regional incomes and employment;
3. assessing the validity of the concerns expressed by the farming sector that a shift to environmental- as opposed to production-based agricultural support might adversely affect the wider rural economy of Scotland; and
4. injecting greater understanding into the whole debate surrounding the socioeconomic impacts of a shift from production-related to environmental payments for farmers.
2.2 Possibilities for redirecting support for agriculture towards environmental objectives
In so far as there is any a shift in public financial assistance for agriculture towards linking support to environmental objectives, the underlying assumption made is that it will be fiscally neutral, though not production neutral. Thus, the changes can be expected to lead to both a redistribution of support payments among farmers and landowners and changes in agricultural output, but not in the total cost to the exchequer. As regards the precise form that any environmental support scheme might take, it is open to considerable speculation, but it is likely to have some or all of the following elements: (i) financial assistance for environmental preservation or improvement; (ii) incentives to reduce the intensity of livestock farming; and (iii) inducements to reduce the use of chemical inputs in arable farming.
As regards assistance for environmental improvement, this is most likely to be provided in the form of direct payments for specific actions on individual farms. Each farm would have its own development plan and payments would be made for specific actions, e.g. restricted grazing at given times of the year and maintenance of hedgerows. As indicated, the second element of any modified support regime is likely to involve controlling stocking rates for livestock, at least in selected areas. Livestock stocking densities are readily acknowledged as having an impact on the bio-diversity of a holding and are a useful and easily understood means of directing support payments in an environmentally sensitive way. Finally, the other predictable element of any modified support regime is a concerted effort to reduce the use of fertilisers and crop sprays, which are acknowledged to be production enhancing, but also environmentally damaging.
Based on these general principles, six possible scenarios were considered. The first three involved the complete withdrawal of the existing support system, in which assistance is primarily linked to agricultural production, and its replacement by a programme of environmental payments. In this case, the possible options were considered to be:
1. A compulsory support regime, in which public assistance to farmers was explicitly linked to compliance with agroenvironmental guidelines. While the historical distribution of support between livestock and arable farming might be retained, the basis for assistance would be changed. In respect of livestock farming, all support for dairying, beef rearing and sheep production, including Less Favoured Area payments, might be converted to an area payment linked to stocking rate, with the payment being inversely related to stocking density. In the case of arable farming, all crop aid could again be converted to an area payment and be paid on the condition that certain management practices in respect of chemical and fertiliser usage are adopted. Alternatively, the scheme might operate along the lines of a compulsory set-aside, with yield and husbandry practices being permitted to continue unchanged on the land left in cultivation (Proposal I).
2. An alternative, but more radical, variant along the lines of the proposed by the Scottish Landowners’ Federation in a recent report. This involves withdrawing existing production-based support and replacing by a scheme involving support payments linked to land management, environmental and social objectives. Under this proposal there would be three of four tiers of payment for the different environmental and social components. In particular, the land management component might be linked in some way to the production capability of the land, while the environmental element of support could be dependent on the type of actions undertaken. Finally, the social component could be linked to employment levels in farming. The ultimate balance of support between the different elements would a matter of political judgement (Proposal II).
3. A voluntary scheme, in which farmers are compensated for physical actions and expenditure to conserve or re-instate environmentally positive features, such as hedgerows, woodland and natural habitats. Payments could be made on a mixture of a flat rate per business, absorbing, say, 25 per cent of the funds allocated, with additional payments being made on a per hectare basis with different rates for, say, rough grazing, permanent grassland, arable cropping and woodland (Proposal III).
The alternative to a complete withdrawal of existing production-linked support is a partial retention of such support, with any funds released being used to introduce one of the three proposals outlined. For the purposes of analysis, it was assumed that 50 per cent of existing public expenditure on agriculture would be released for environmental support programmes. This gives a further three scenarios (Proposals IV, V and VI).
2.3 Socio-economic impacts of a shift in support towards environmental objectives
To estimate the impact of the proposed scenarios on income and employment, a series of regional models were developed for the twelve ‘former’ administrative regions of Scotland. These models were designed to estimate the consequences that a change in support would have directly on agricultural output, incomes and employment and indirectly on those industries who supply inputs to agriculture or process its output. In this study, the socio-economic impacts were then estimated in three steps. The first step involved forecasting the consequences for income and employment regionally and nationally of withdrawing the existing programme of production-based support for agriculture. The next step was to assess the gains from introducing a support programme along the lines outlined for each proposal. Finally, the projected socio-economio consequences of a switch in the method of support for farming was derived by subtracting the gains in step two from the losses in step one. In practice, because the issues being studied were ones for which there was limited direct evidence, the models involved a number of assumptions, which could not be properly tested. Accordingly, it was necessary to test the sensitivity of the estimated impacts to changes in assumptions. For this reason, it has been more realistic to indicate a range of outcomes rather than a single estimate of the impact. The expectation is that the actual consequences of a change in the method of providing agricultural support will lie somewhere within the projected range.
The models were specifically used to address three questions:
1. Will a shift away from linking support to production, to linking it to environmental actions bring socio-economic gains or losses nationally and regionally?
2. Is a partial replacement of existing production-based support by environmental incentives (Proposals IV, V and VI) better from a socio-economic viewpoint at either a regional or national level than a total replacement (Proposals I, II and III)?
3. Which of the proposals would appear to be most beneficial in terms of the regional distribution of the impacts?
From a national (all-Scotland) viewpoint, the models indicated that a withdrawal of existing support for agriculture could produce a reduction in overall income of between £1000 and £1500 per year and a decline in employment of up to 40 to 60 thousand jobs. That is to say, every £1 reduction on agricultural support means a £2 to £3 loss in income for the Scottish economy and every job lost in agriculture is expected to reduce overall employment by 1.8 to 2.7 jobs. The introduction of a system of ‘environmental’ payments in place of existing production-based support was estimated to produce income and employment gains, but not sufficient to compensate for those lost through the withdrawal of existing agricultural support. Thus, the effect of switching to a new payment system, along the lines of those outlined in Proposals I, II and III, is to produce a net loss to the Scottish economy of between £0 and £760 million per year and 0 to 18000 jobs.
This is despite the fact that the overall level of support expenditure was assumed to be unchanged. Where there is only a partial shift in support, with 50 per cent of public assistance still being channelled as an aid to production (Proposals IV, V and VI), the projected effect is to reduce the losses from the shift towards environmental incentive schemes by around 30 per cent. The corresponding regional effects of the various policy proposals are presented in Tables A and B.
The best and worst outcomes for each proposal are given, with a positive value indicating a net gain and a negative value indicating a net loss, compared to the existing system of farm support. Compared to the national picture, the position is more complex, with some areas, such as Shetland, Highland, Grampian, Strathclyde and Central tending to show economic benefits from a switch to a system of ‘environmental’ payments, In the main, however, the impression left is still one of projected socio-economic losses from a move to environmental-based farm support.
For each region, the six proposals were ranked according to the projected net gains or losses from implementing them in place of the existing agricultural support scheme. This comparative analysis indicated that proposals IV and VI appeared to represent the best switches, in terms of giving the greatest socioeconomic benefits for the majority of the regions. Under proposal IV, fifty per cent of total support for Scottish agriculture would continue to be channelled through the current system of production support, while the remaining assistance for both arable and livestock farming would be made conditional on actions aimed at reducing chemical usage and extensifying production in a variety of ways. Similarly, proposal VI involves a partial retention of existing production-based support, with the remaining assistance being provided in the form of discretionary payments made to farmers to compensate for specific physical actions, designed to conserve and reinstate environmentally positive features. Nonetheless, although these proposals are the most beneficial from a regional standpoint if a shift towards environmental incentives is to be made, they are still projected to cause a decline in regional income and employment in many cases, compared to the existing system of farm support.
2.4 Validity of concerns about income and employment losses from a shift towards linking farm support to environmental objectives
Although this analysis of the socio-economic consequences of re-directing support for agriculture towards environmentally-driven objectives has tended to re-affirm the fear that there could be adverse income and employment effects, the projections are surrounded with considerable uncertainty. In particular, depending on the assumptions used, it was found from sensitivity analysis that the projected socio-economic impacts could vary from being very substantial losses to modest gains. Critical to the conclusions were:
1. the response of farmers to cuts in production support;
2. the effects of changes in agricultural output and incomes on associated businesses, who supply inputs or process farm output; and
3. the spatial distribution of the impacts.
As regards the response of farmers to a partial or total shift in public assistance for agriculture towards environmental objectives, the size of the projected decline in agricultural output is a key element in the conclusions that the impacts of the switch will be economically adverse. Given this, it is important to recognise that the regional agricultural models developed are really only suitable for exploring marginal shifts in support. Thus, they are being used to study impacts outside the range within which confidence can be attached to their predictions. Certainly, it is impossible to validate their projections and therefore they are open to the criticism that they exaggerate the decline in farm output that would occur in reality. A look at studies conducted of two existing environmental schemes (the Countryside Stewardship and Tir Cymen Schemes) have indicated comparatively small impacts on farm output. However, the findings of these two studies are not directly relevant to the issues being explored. First, the two schemes do not involve a withdrawal of existing support, simply an offer of additional payments if certain environmental actions are undertaken; whereas the study is concerned with examining a re-allocation of existing funding support to agriculture and not additional assistance. Second, the schemes are marginal in terms of the totality of support being given to agriculture.
The second critical issue is the estimated impacts of changes in agricultural output and incomes on patterns of farm household expenditure and the consequent effects on income and employment in industries associated with agriculture. These were derived by estimating economic ‘multipliers’, which provide forecasts of the effects of changes in agricultural output on incomes and employment in the rest of the local and national economy. As indicated in an investigation of the uncertainties attached to the projections, the economic multipliers used to predict the consequences of a withdrawal of existing production-based support for agriculture on regional income and employment were apparently high. There is some evidence to suggest that they might exaggerate the losses by as much as 50 per cent. This does appear to have a significant impact on the conclusions regarding the gains and losses from a shift in the method of support. Against this, the economic multipliers derived from the regional input-output matrices only strictly capture the upstream effects, associated with industries supplying inputs to agriculture. The downstream consequences of a change in agricultural output on industries which process its output are not captured. At one level, it is possible to argue that these industries will not experience any contraction, since they will merely switch to sourcing agricultural raw materials from external suppliers. This may be partially true, but it is probably unlikely that they would be able to replace totally all requirements in this way.
However, the overall conclusions regarding the net impacts of replacing existing farm support by various ‘environmental’ schemes are just as dependent on estimates of the income and jobs created by the new system of environmental payments to farmers. The problem here was that it was very difficult to be specific about the expenditure patterns and consequent upstream and downstream effects of the environmental payments. This was because the proposed schemes were essentially ‘generic’ ones and did not contain specific details about the type and level of agri-environment payments. As a result some rather general assumptions, regarding the economic multipliers used to project the gross socio-economic impacts of the proposed environmental incentive schemes, had to be made. While the estimated multipliers are not inconsistent with those obtained in the Tir Cymen and Countryside Stewardship Schemes, it is obviously of concern that they cannot be regarded as confident estimates.
However, to make any progress, it is necessary to formulate in detail the precise level and type of agri-environment payments, so that it is possible to predict expenditure flows as was done in the Countryside Stewardship study. Again this necessitates predicting how farmers will respond to a shift in the method of support. At the same time, there is also need to identify those expenditure flows, which may be short-lived, from those that can be regarded as fairly permanent. In particular, in both the Tir Cymen and Countryside Stewardship studies a significant proportion of the income and employment benefits were associated with capital expenditures by the farmer. In many cases, it is difficult to see how these capital expenditures will be sustained over many years; rather they might be expected to be high at the outset of the scheme and decline with time. Unfortunately, neither of the schemes has been running long enough to assess this. Equally, where the environmental schemes are based on voluntary participation, there remains the problem of forecasting uptake.
Finally, the effectiveness of any agricultural support regime as regards promoting social and economic development cannot simply be measured in terms of overall increases in income and employment. The spatial distribution of the gains and losses is just as important. From an admittedly cursory analysis for two of the proposed environmental incentive schemes in three administrative regions in Scotland, it appeared that a shift in public assistance to agriculture away from production towards environmental support would not bring about a re-distribution of regional wealth and income towards economically less favoured areas. However, again the problem is that these projections, which are based on a theoretical model of regional expenditure flows, cannot be validated.
2.5 The way forward
Arguably the study has opened up more questions than it has solved. However, it has underlined the central importance of getting accurate estimates of the impact of any reforms in agricultural support on
Only then can some of the socio-economic impacts be projected with greater confidence. At the same time, to project accurately the impact of replacing existing production-based support by a system of environmental payments requires more detailed specification of the type and level agri-environment payments. Only then will it be possible to synthesise accurate economic multipliers for these proposals.
Accordingly, to prepare convincing forecasts of the impacts of a major reform of agricultural support payments, involving the replacement of existing support by natural heritage and environmental incentives, three things are needed:
1. More accurate estimation of the impact of changes in the levels of production support on agricultural output. This might be achieved, by constructing a series of representative farm models, using mathematical techniques such as linear programming, and studying the projected changes in farm output. Using Parish and regional data from the June Agricultural Census, it should be possible to aggregate up to the regional and national level.
2. More accurate estimation of the economic ‘multipliers’, which forecast the impact of changes in agricultural output on regional incomes and employment. This may be achieved by combining non-survey techniques of the type used in this study, with field surveys to refine the estimates of the size of the economic linkages between specific agricultural enterprises and the rest of the economy.
3. More accurate estimation of the economic ‘multipliers’ associated with agrienvironment support schemes. This basically involves preparing more specific proposals, than the ‘generic’ schemes, used in this study. The danger in doing this is that it is always less easy to secure agreement about the specific details of any scheme. However, ultimately, if the proposals are to form the basis of policies, then this will have to happen.
3.1 The model used is a comparative static one, and therefore does not take into account the possibility that the agriculture sector would adjust its structure radically in the event of the removal of all production support. Even in the event of the removal of 50% of production support, one expects that radical restructuring of agriculture would occur.
3.2 The results of the work are, in some ways, frustrating to the policy designer, since the model cannot discriminate between different approaches to agri-environmental policy. The modelling process is too crude, and the data too sparse to allow this.
3.3 The results of this model should not be read as facts.
3.4 The predicted decline in the rural economy is brought about by the difference-between the upstream economic multipliers for present agricultural activity and those for the household sector, of which the farm household sector is assumed to be a part. In other words farmers are predicted to spend less money in the local economy if it is given to them through a means other than agricultural production support. The study highlights just how little is known about the economy of farm households.
3.5 It should be noted that this study excludes the effects of economic changes downstream from the farm. An example of downstream economic activity is that of slaughterhouses in response to a downturn in aggregate farm production.
3.6 Despite these misgivings about the strength of the study’ findings, the study has many uses.
3.6.1 The study challenges the reader to consider how they would react if the policy predictions made were to come true.
3.6.2 It shows that the portfolio of secondary data for undertaking agricultural policy analysis has considerable shortcomings when one wishes to discriminate between rural policies in terms of their effects in the environmental and social spheres.
3.6.3 It illustrates that support for economic regeneration will be needed in areas currently dominated by agriculture in the event of the withdrawal of agricultural production subsidies. Any savings from reduction in the agricultural subsidy budget should go into economic renewal as well as environmental care and that care should be taken to seek joint benefits of environmental care and the promotion of rural economic activity from expenditure in rural areas.
3.6.4 The multipliers, if true, indicate that there may be a trade off between increasing farm household incomes and increasing economic activity in the rural economy.
3.6.5 It shows the importance of the design and the inclusion of capital elements within any agri-environmental scheme are of critical importance, if that scheme is to provide local socio-economic benefits along the lines that Tir Cymen has done.
3.6.6 The study shows the need for further work on the broad socio-economic effects of agricultural and environmental policy change.
It is arguable that all these conclusions could have been drawn a priori, but the value of the modelling exercise is to bring home the implications of the potential conflicts between objectives of agricultural, environmental and rural economic and social policies and to make policy designers aware of the potential scale of policy conflicts in financial terms.
Doyle C., Ashworth 5, Mitchell M. and Topp K., 1997. Socio-Econornic Impact Of Replacing Agricultural Commodity Support with Natural Heritage Incentives. SNH Research, Survey and Monitoring report - in press. SNH, Battleby.