Agriculture and Employment

1. General considerations
Employment is an all important issue for economic and social policy for, among other things, it is the most unquestionably legitimate way of earning one's livelihood, of sharing in the national wealth and of participating in the economic and social life. All other ways, including transfers through various redistribution mechanisms, are sooner or later questioned. Thus, the concern about securing employment is an autonomous goal of policy, parallel to that of generating income, and trade-offs between the two may be needed. Full employment would be the natural way of having all people sharing in the national product. Instead, we all know that in the last 2-3 decades heavy and persistent unemployment has become a permanent plague and a threat to social cohesion, even in the most developed countries.

Rural areas constitute the largest part of the national territories and are inhabited by a substantial portion of the population. Besides, they are cradles to particular social structures and cultures that are of increasing value to societies because they are at present in danger of decaying and disappearing.

Rural areas with their low density that entails fragmentation of labour, service and consumer markets and with their particular social and cultural characteristics tend to have low competitiveness with respect to most modern activities, which need the advantages of agglomeration. Few of these could be viable in rural areas, especially the remote ones.

Thus it is imperative to look for activities which are suitable for these areas because they can make use of their characteristics. (What is said about the rural areas of the developed countries holds true for the whole economies of the less developed and economically weaker countries).

Agriculture is, par excellence, an activity suited to rural areas. It is in fact peculiar to such areas. It needs the natural factors and makes use of the traditional skills; has a traditional local and national market; possesses a natural differentiation based on the local bio-climatic conditions; it can benefit from the consumers' preferences for familiar traditional varieties; finally the fresh character of a great part of the products makes for local rather than centralised production. Thus, agriculture is a natural source of employment in rural areas.,

Employment in agriculture is of course secularly decreasing because of technological progress and as other activities develop (process of "civilisation"). It is also decreasing because of loss of local competitiveness due to the elimination of barriers to trade, be those natural or institutional. Natural barriers are made ineffective by progress in transportation and means of conservation. Institutional barriers are abolished by acts of will, usually by mutual agreements between countries, in view of the benefits that the "customs unions" or liberalisation of trade generate (enlargement of markets, lowering of transaction costs, specialisation, economies of scale).

However it is well known that the customs unions benefit mostly the strongest countries and regions. In any case, reallocation and redistribution take place, both between branches and between regions. The adverse effects of these reallocations are supposed to be taken care of by the overall gains of the union, through various redistribution mechanisms. However, when the "customs" or "economic union" is not a narrow area with a central authority to take care of the redistribution in a systematic way (like the European Union which has a well defined redistribution system) then this redistribution is left to chance, charity, or political bargaining. And even when such mechanisms are installed, after some time their cause is forgotten and their legitimacy is questioned.

Coming back to agriculture we retain that agricultural employment in our days suffers from both: linkurl_blank¹>> ^{, linkurl_blank}> linkurl_blank²>>

-Fast technological changes, which make labour redundant and standardise inputs, processes and products, reducing thus local differentiation, and

-Progressive unification of markets through the establishment of economic unions, like the EEC, and through international agreements, like the GATT, which for the first time included agricultural products in its previsions.

This would not be so bad from the point of view of employment if the national and regional economies were buoyant and the displaced or new labour could be quickly absorbed in other branches. But in an era of relative stagnation and of generalised and persistent unemployment, losing already existing jobs, in the name of an abstract international efficiency argument, without any well-defined mechanism of redistribution of the gains, is likely to be challenged. Especially by the economically weak countries and regions, which the globalisation does not favour but poses a great strain on. And even by the stronger countries for their weaker regions, which do not offer many alternative opportunities for employment.

The least one could ask for would be the changes to be gradual and not very fast, so that any adjustments are made more natural and less painful; that there continue to be compensatory redistribution between countries, regions and individuals, both to ease individual and regional hardship and to promote alternative activities viable in the hit areas; and finally that agriculture is recognised to not be like any industrial activity, but produces products which are essential for life and health, and which, therefore, all human societies have the right to keep under control. So that the existence of the activity at an adequate level to secure its long-term viability in every country is not jeopardised.

This may mean to have the time and the mental ease to develop new competitive advantages based precisely on the local differentiation of quality, genetic diversity or genuineness, and new consumer preferences, and not just the price and the promotion mechanisms of the strongest. linkurl_blank₃ >>For if the national markets are lost under the pressures of international competition, then the productive know-how will also be lost, and so will the local genetic stock. And much of the agricultural land will be turned to urban or industrial uses and will deteriorate irretrievably. And so the country will lose the possibility to ever compete and will become dependent in this most strategic of all sectors ^. >

The following sections draw heavily on recent and past OECD work for which I am thankful.

2. Some facts on unemployment
Almost 35 million people, more than 8% of the total labour force in OECD member countries, are presently unemployed and searching for a job. Since 1975 their number has more than doubled (OECD 1996, p. 24 fig 1.5a). In several OECD countries every second unemployed has been looking for a job for more than 12 months. In some countries, 1/4 to 1/3 of the young is unemployed. Over the last 20 years the average unemployment rate ranged in the order of 5% to 8% (op. cit. fig 1.5b). (This does not mean that the number of employed has not increased; in fact, since 1975, about 75 million additional new jobs were created, but this was not enough to absorb the total increase in the labour force). Of course the differences between countries are great. The unemployment rates range between 3% for Luxembourg and Japan, to 17% and 23% for Finland and Spain; (with Norway at 6%, USA 7%, and Greece at 8%) (op. cit. fig. 1.5c).

The regional differences are also high within countries, and typically higher and more persistent in Europe than in the USA. The regional distribution is particularly uneven in Finland, Germany, Italy, Portugal and Spain (OECD 1994c, p.11). However all countries, including those with comparatively low national rates, are confronted with substantial differences in regional unemployment. (Table 1).

Between 1980 and 1990, while unemployment increased, territorial disparities as indicated by the regional variation coefficients slightly decreased. But this reflected the fact that in times of rising national unemployment also the better off regions were affected.

Unemployment tends to be higher in rural regions, although not everywhere so. Agricultural employment is almost everywhere declining, not only in relative but also in absolute terms.

3. Rural areas
About one third of the OECD population, 250 to 300 million people, are living in rural communities (i.e. with a population density below 150 inhab/km2, except for Japan where the threshhold is 500) occupying over 90% of the territory. National shares differ, of course, ranging from just less than 10% for the Netherlands and Belgium to about 60% for Finland, Norway and Turkey (OECD 1994b, p. 27).

Table 2 shows the spatial distribution of population by the three types of regions defined by the OECD-Rural Indicators Project according to their degree of rurality, that is:

• Predominantly rural regions (PR) where more than 50% of the population lives in rural communities as defined above,
• Significantly rural (SR), where the rural population is between 15% and 50%, and
• Predominantly urban (PU), where the rural population is less than 15% of the population of the region.

On average, one quarter of the OECD population dwell in PR regions, while a further one third live in significantly rural ones (SR).

For many OECD countries like the Netherlands, Belgium, Germany, the UK and Switzerland the population share increases from PR to PU regions. For Austria and the Scandinavian countries the opposite is true: less than a quarter of the population live in PU regions. In other member countries, like France and Spain, the largest share live in intermediate regions (SR); while Ireland, Iceland, Greece, Portugal, but also Canada and Australia, have a dual structure with greater shares inhabiting either PR or PU regions.

The unemployment situation is worse than average in the predominantly rural regions in most OECD countries, and in some of them considerably so (Finland 1,25, Canada 1,17, Iceland 1,39, Sweden 1,13, as compared to the national average being 1). On the other hand, in some countries the PR regions are doing better than average in this respect (Japan 0,95, Austria 0,95, Czech Rep. 0,92, Switzerland 0,82, Germany 0,75, UK 0,78, and Belgium 0,80). (Table 3).

4. Agricultural employment
The general picture of the OECD, as to be expected, is one of diminishing share for agriculture in total civilian employment and in output. And the former has fallen faster than the latter because of rising productivity of agricultural labour due to technological change and increasing capital intensity. However in all countries it represents a significant share, varying inversely with the overall development. Thus we have a low of 2% and 3% for the UK and the USA and a high of 25% for Greece and 50% for Turkey. (Table 4).

The share is larger in the predominantly rural regions where it represents, on the whole, a very important share (from 37% for Portugal, Greece and Iceland, to 6% for the USA and Sweden). And it is that this aggregation conceals much wider regional disparities, either at regional or at prefecture level. linkurl_blank₄ >These disparities are very important because some regions or prefectures are remote or fragmented and there is no easy unification of their labour markets (mountainous areas). The same holds true for unemployment, where aggregation conceals very different local and regional situations that were not unveiled in the work for the rural employment indicators (OECD, '96).

We all know, of course, that agriculture is not the unique source of income even for the so-called agricultural households. Even for those, it accounts about half of their income. linkurl_blank₅ >Even the heads of the agricultural enterprises engage in other gainful activities when the incomes from farming are not sufficient and when they have the opportunity to do so. The AGRI case study (in OECD '96) shows that although underemployment in agriculture is very large depending on the type of agriculture (intensive or not), the kind of culture (annual or permanent) and the size of the farms, pluriactivity is in all cases smaller. And, although underemployment ranges from a low 31-33% in Ireland, the Netherlands and the USA, to a high of 89% in Italy, and 84% in Japan and Greece, the level of pluriactivity varies much less, ranging between 22-26% for the Netherlands, Finland, France, Italy and Greece, to a high of 42-44% for Canada and Germany. ₅. For the majority of the countries, pluriactive farmers form about 1/3 of the total, but there appears to be no consistent pattern depending on the degree of rurality. (It is possible that a lower level of aggregation is needed for this to appear).

The AGRI case study finds also that in countries where the share of agriculture in the total employment is high, the share of underemployment in agriculture is also high.

The study also confirms the rising share of agricultural employment from PU to PR regions - so the increasing importance of agriculture as a source of employment for the more rural regions - so also and the fact that within the secular decline of the of agricultural labour (annual decrease in the '80s from 1% to 4%) the rural regions are often weaker in either keeping or creating agricultural employment. And, more often than not, in creating non-agricultural employment as well. (AGRI p. 116). The study finds further that even in the few countries in which employment in the primary sector was stable or slightly increasing (-1><+1) this consisted mainly in new jobs in related activities, like fishing, gardening, landscaping, rather than in farming itself. And it was created in PU rather than in PR regions. These countries were Canada, the USA, New Zealand, the Netherlands and the Czech Republic. (Table 6). (In the case of Canada, however, it is estimated that the positive movement in all kinds of areas may reflect inclusion in the data of occupations that were not previously counted as agricultural employment).

Local differences in the changes of agricultural employment are often interpreted as a result of push and pull factors, and empirical tests are consistent with this hypothesis (Efstratoglou-Todoulou 1990). Labour is pushed out of agriculture and pulled into other sectors. Strong push factors are likely to be present in rural areas, especially in those with insufficient agricultural structures, while at the same time the pull factors are reduced by moderate employment opportunities. However national and regional differences are great in this respect, reflecting the dynamism of the area. It appears that, the growth rates of non-agricultural employment is a better indicator of the strength of the pull factor in a region than simply the degree of rurality (AGRI p. 116).

It is recognised further, that the relationship between agricultural labour and the rural context is a complex one, and the aggregation in the three categories adopted in the OECD-Rural Development Programme does not allow valuable information on the full extent of dispersion around the national means to appear.

Apart from the quantitative importance of agricultural employment, there are also qualitative aspects, which are equally important. It has been shown by various studies that agricultural employment possesses a degree of stability with respect to the general economic and employment outlook, and is thus little influenced by the normal business cycles. (OECD, 1994, p. 51).

Indeed, equations linking family and hired farm labour to various agricultural and macro-economic variables for eight countries (Austria, Canada, W. Germany, Italy, Japan, New Zealand, the U.K. and the USA) have provided evidence that both hired and family labour are rather insensitive to changes in macroeconomic variables; neither category seems to be heavily influenced by the business cycle.

Although this work is based on highly aggregated data which mask the underlying categories of the labour force (part time and full time work, younger and older workers, different educational attainment) it nevertheless highlights the fact that short term conditions prevailing in the labour markets do not have a major impact on farm family and hired labour, as measured in the number of people. What indeed may vary is the labour input, i.e. the number of hours or days of work, the degree of full or part-time employment; but there will not be open unemployment, as it is registered in other sectors. In this respect agriculture is a stabilising factor in the economy, precisely because it has the flexibility to absorb short-term changes through the variation in the degree of employment.

This stabilising role has been pointed out also in a study performed by OECD's Directorate of Education, Employment, Labour and Social Affairs (DEELSA), of the farm household's role in rural labour markets in the North-eastern-central regions of Italy (NEC). Here, the importance of the rural family in providing the labour markets with flexibility and resilience is emphasised. On the other hand, the local possibilities of pluriactivity protect the family from various shocks in particular markets of labour or products. The stability of relations in rural areas, based on the farm family's ability to serve as a buffer against various shocks, makes for stability and resistance against all kinds of macroeconomic shocks. The polyvalent farm household, provided it is not dismantled, may provide a buffer against the shocks of agricultural reform, if this proceeds mildly enough.

The relative stability of farm employment has been confirmed also by other empirical work. A review of econometric studies conducted by the OECD has found comparative evidence: Out of four such studies, three tend to confirm the finding that general labour market and macro-economic variables have little influence on the level of farm employment (for Finland, Australia and the European Community). The forth though, for the UK, which however is mush more specific, finds strong relationships between macroeconomic variables and the hired, full-time, male labour in England and Wales.

5. The importance of the agro-food sector
Agriculture does not affect the local economy only directly, through the primary activity itself, but also indirectly through the linkages between the farm sector and related upstream and downstream sectors. Further, the full range of interdependencies between the wider agro-food sector and the rural economy must be recognised.

Although the food sector is not wholly dependent upon the local or national farming activity, common experience shows the strong links between the two, especially in a dynamic or historical perspective. Policy changes that will affect the farming sector will most probably unleash a whole range of chain reactions, which may deeply affect the local and national economy.

Table 7 shows the relative importance of the upstream and downstream sectors in terms of employment, for selected countries for which data were available. The combined agro-food sector has a considerably greater importance for employment than the primary sector alone. And the importance, more often than not, is greater for the rural regions than the predominantly urban ones (Table 8).

And it is important to note that although farm employment is undergoing a long-term decline, employment in closely related activities such as agricultural services, inputs, processing and marketing has remained fairly stable or even increased in a number of countries (OECD '98, p. 54).

The regional distribution of agro-food establishments also bears witness to the wide spatial distribution of the activity and its importance for regional development (Table 9).

A number of case studies have substantiated the importance of the agro-food sector for the rural economy of countries like Canada, the USA and New Zealand.

For Canada it was estimated that up to 18% of the employment in the PR areas and 20% in the SR areas originated from the agro-food sectors in 1991. (OECD '98, p.56).

In the USA, also in 1991, the same sector provided a quarter of the jobs in rural America (op. cit p.56).

In New Zealand, the sector provided employment for about 17.4% of the country's work force in 1996.

A study conducted in the U.K. showed significant links between farms and their locality. Especially the smaller farms, as these have more transactions with rural areas than the larger ones. The industry providing the highest value of produce from rural areas was the feed industry, while the main farm output being sold to rural areas was cereals and milk. Pig and poultry farms had the greatest backward linkages with other firms in rural areas.

The study also estimated that approximately one quarter of the people working in agriculture-related industries is working in rural areas. Also, that the ratio of those working in farming and those working in auxiliary rural industries is approximately two to one. The study concludes that agriculture has indeed strong links with the local economy, perhaps more than had previously been thought.

A survey conducted by the OECD, of econometric studies attempting to measure the linkages and multiplier effects of the agro-food sector to the wider economy as well as to the local rural economy, points among others to the following salient findings:

• The agro-food sector has indeed significant economic linkages to other sectors of the economy and constitutes an important generator of employment in rural areas.
• The primary sector is found to have the largest income and employment multipliers in both PR and SR regions.
• The farm sector is an important generator of employment in downstream industries, particularly food processing.
• There is an asymmetry of leakage to and from the farm sector. Because of the dependence of farming on non-agricultural inputs there are large leakage from the farm sector to the wider economy. While, because of the small elasticity of food products, rises in non-agricultural incomes benefit the farm sector only to a small degree.
• Livestock commodities seem to have the highest backward and forward linkages as they require more intermediate inputs than crops and are relatively income elastic, thus generating higher multiplier effects. (OECD '98 p.57).

However, the contribution of agriculture to sustaining local economies differs widely depending on the structures of the sector, farm types, the size of the region, and the structure of the upstream and downstream sectors. Thus the magnitude of the output, employment and income multipliers differ significantly among the agro-food sectors, and between regions. It is clear that stronger economies, whether regional of national, have more integrated productive systems so that the various multipliers exert their effect mostly within them, rather than outside, as is the case with the weaker economies.

6. The role of agricultural policies
Agricultural policies have evolved over time and are increasingly seen by many OECD countries as a vehicle for economic and social revitalisation of the rural areas and not only as a means for maintaining farm incomes. (OECD ‘98 p. 49).

Their effectiveness and efficiency have been both defended and questioned by theoretical arguments as well as empirical research. And of course all views on the matter can be supported and documented.

For the purpose of this presentation, I am going to pick selectively only some of the points discussed in the OECD ‘98 report, chapter 3.3.6 ''Overall assessment '', which is based on a review of a very extended and varied literature on the subject. And I will interpret them in my own way:

So, although policies have not prevented the long-term reduction in the agricultural employment, nor reversed the tendency of the young to leave the sector and, very often, the country side, it is widely believed that this shrinkage and exodus would have been stronger and more painful in the absence of the policies. Especially during the decades, when rurality and its "culture" were not a social value but quite the opposite, {when peasant, rural, rustic meant uncivil, rough, boorish (Arnold Mandeson lexicon, 1961)}.

This fast agricultural exodus would certainly have had adverse multiplicative effects on the whole economy of all but the strongest rural regions. ⁶>

In the absence of the support policies the production would tend to be more concentrated in the areas and countries which would be strongest in developing competitive advantages - not necessarily associated with favourable bioclimatic conditions and cheap labour but with technology, standardisation, marketing and general economic strength. ₇ >

The intensification associated with technology, together with the tendency to concentration, would certainly have had adverse effects on the environment and on the diversity and quality of the products themselves. These events have already occurred, even in the presence of support policies, because there is a universal tendency towards more technology, which is strengthened by the competitive advantages it confers. And as we have said, international competition was not abolished because of the support policies.

It has been contended and substantiated by empirical evidence that the present structure of price support policies, which are based on output or acreage, have accentuated income inequalities and regional disparities in agriculture (OECD '98 p. 62). linkurl_blank8 This is an obvious result, since large farms and more "efficient" (intensive) farmers, operating in the most favourable areas reap the benefits associated with production quantities. linkurl_blank⁹>>

However these studies examine the inequalities within the agricultural sector and not inside the national economies as a whole. From that point of view, the conclusions might be different. Indeed, from an ongoing review of the situation of inequality and poverty in Greece, which is being conducted at KEPE, (Carabatsou-Pachaki 1998, unpublished) some interesting results emerge concerning the agricultural sector:

• Agricultural incomes influence greatly and negatively the level of inequality within the national economy, because farmers are very numerous and relatively poor: a uniform decrease in their incomes would greatly increase the GINI index of inequality (elasticity of GINI coefficient of -0,042 for incomes from agriculture against +0,044 for wages and salaries and +0,078 for self-employment). The agricultural support payments have a much smaller impact on total inequality (-0,005) (Mitrakos '98). Their progressive redistribution would marginally affect total inequality, and this provided that it would not entail an important fall of the total agricultural incomes. If more equality within the farm sector means much lower total agricultural incomes because of the policy reform, then this will worsen the inequality of society as a whole. Because farmers are still one of the poorest sections of society in many countries.
• In spite of all support policies, the agricultural households still contain the largest portion of poor people: 36% against 24% on average. However this portion was 48,8% in 1974 before the CAP, it fell to 40,4% in '81/'82 during the period when internal prices were gradually raised to the EEC levels; and continued to fall, inspite the fact that agricultural product rose very little after accession, in 1981. The portion of the rich farm households is only 0,34% on the total ('87/'88).
• Agricultural households draw about 2 of their income from agriculture and their total income is 83,0% of the average ('93/'94). And this income comes to a large extent from auto-consumption (25% in '87/'88). Only about 26% of "farmers" have another gainful activity, although only 14% are fully engaged in agriculture. Even in peri-urban areas the pluriactivity of both farmers and their family members is not much greater, because farms tend to be more intensive there (Moysides et al.).
• Agricultural employment has continuously fallen in terms of Annual Work Units (671.000 AWU) but not in number of persons employed (1.500.000).
• The CAP does benefit more the richer and more productive farmers, even per unit of labour. But these differences are much smaller in Greece because the distribution of land is much less unequal. Most of the farms are small or medium small (Tarditi, Zanias).
• Greece has a net inflow from the CAP and all Greek regions have a net benefit from it, even if the burden of the Greek consumers and taxpayers is taken into account (except for three regions where the agricultural sector is very small - Attica, Ionian Islands, Southern Aegean). Thus the agricultural support policies have been beneficial and progressive for Greece as a whole.

7. Policy reform
A continuing policy reform, in the direction of a reduction in the overall level of agricultural support and a shift away from policies linked to production quantities, to measures targeted to specific social, environmental or territorial goals, seems an unavoidable trend, supported by considerations of budgetary unsustainability and partial ineffectiveness of the past policies. linkurl_blank¹⁰>>

The effects of a deep going reform are difficult to foresee and will depend on the pattern and pace of change, the local and regional structures, and the national strengths or weaknesses. Much will also depend on the set of social values prevailing over time, which affect general attitudes and consumer preferences. The same changes may not have the same effects now as in the sixties, or as if the policies had never existed. For now, people are more sensitive to considerations such as environmental protection, healthiness and wholesomeness of food products, quality beyond appearance and packaging, and to notions such as countryside, rurality, heritage, tradition, social and family relations. Consumer and producer preferences and social awareness may in the end protect what each society considers worth protecting, despite of the elimination of economic protective barriers.

Thus, models that have attempted to evaluate the implications of agriculture policy reform and which give a stark picture of adjustments entailed, with agricultural production gravitating to a fairly limited number of low-cost producing regions, where the comparative advantage is a combination of relative factor endowments and technical requirements of production, may prove too pessimistic. Because dynamic aspects of international competitiveness play a most important role (OECD '94, p. 58). In fact, other analyses focus on the fact that primary factors are not qualitatively equivalent across countries; neither are human resources which differ, among other things, in scientific, technical and marketing expertise; which, in turn, is a function of current and past production patterns. Also, local demand conditions, the existing structure of the industry and of other industries (networks, agglomeration effects) all contribute to the shaping or the lack of an ability to compete, which is known as ''competitive '' rather than ''comparative '' advantage. (Dutch horticulture is an example of a development, which was not to be expected on the basis of the factor endowment argument).

But there will certainly be strong pressures, especially on those countries and regions, which have not already developed a strong ability to compete. Countries and regions, which depend heavily on agriculture, especially remote rural regions with not many other linkages to the overall economy, may suffer very heavily. In such areas, the decline in agricultural employment will unleash chain effects on their whole agro-food sector with important implications for the rural and national economies. And these will not be compensated for by any remuneration linked to the non-productive functions of agriculture _{. linkurl_blank}linkurl_blank₁₁

The OECD (1998, p. 73) seems to accept that reform will accelerate the underlying long-term trends towards reduced employment in the sector, and the restructuring of the labour force towards more family labour and flexible hired labour, used mainly as seasonal and casual workers and agricultural contractors. At the same time much of the effect within the family labour is likely to take the form of increased underemployment rather than exit from farming altogether. In some countries, however, where these processes have already taken place to a great extent (ex. Greece, where 88-95% of the labour is family and 2,5 persons offer 1 annual work unit), increased exit of the young and bankruptcy of much of the established farmers, which are heavily indebted, is a very real possibility. Which will much worsen the already difficult situation in the labour markets (present unemployment over 10%) and on the poverty front.

8. Indications from case studies
A number of case studies that have been conducted by the OECD in order to explore the probable effects of various policy reforms in different socio-economic contexts give very interesting insights.

We present the main results of two of them, Ireland and New Zealand, because they deal explicitly with the problem of transition from a context of higher support to one of lower or no support at all, and the effects on rural employment.

Ireland:
Although none of the econometric studies reviewed deals explicitly with the employment implications of reform, from the results for output and from the structure of the farm sector one can infer that hired labour is more at risk while farmers are more immobile but prone to under-employment and pluriactivity. Which again depends on the availability of off-farm employment and on other activities developing a market. The outlook is aggravated by the structural characteristics of the farming community (mostly one single male person). The case study concludes that a reduction in agricultural activity following a deep policy reform is likely to have significant income and employment effects because of the important share of agriculture and of the food industries.

New Zealand:
Agriculture has endured a rigorous period of adjustment since mid-1984, when the Government launched an economy-wide reform programme, by which assistance to agriculture was virtually abolished. As a result there have been shifts in the composition of agricultural output and employment, but in the end this has not entailed an exodus from agriculture. Sub-sectors, which were heavily protected, have faced the brunt of adjustment, where, on top, realignment was required in the upstream and downstream sectors, even stronger than in farming itself. The pressures on agriculture and its labour force were accentuated by the uneven and asymmetrical dismantling of assistance across the other sectors of the economy. However fears of rural collapse never materialised and only few farmers were forced out of the land.

In fact, employment in the agricultural sector has recovered since 1990 with farm levels in 1995 reaching their highest level since 1979, with restructuring, however, in favour of casual rather than permanent hired labour.

9. Profile of at risk farm workers and farmers
The OECD seems to accept the prospect of an increased rate of loss of jobs in agriculture following a continuing reform of the protective policies. He even goes on to paint a profile of at risk farm workers and farmers based on its own analyses (OECD '94, p.63) and on many case studies. So these are more likely to be:

• hired farm workers rather than owner-operators. In some countries these are likely to be immigrants who are anyway vulnerable ¹² >>
• resident of a remote rural area which itself is highly dependent on agriculture. A remote area is almost by definition, isolated from alternative sources of employment, education and training centres.
• employed on a farm which cannot be viable without subsidies and which for various reasons cannot diversify (high-cost production, few alternative crops).
• isolated in socio-economic sense, i.e. cannot benefit from the informal social insurance afforded by the local pluriactive farm family (immigrant, minority).

10. OECD policy recommendations and recent issues
Given this prospect, ongoing OECD work tries to explore the possibilities of generalised rural development, based on alternative activities including commercialisation of the so-called rural amenities (Rural Development Program); of agriculture itself to develop new competitive advantages based on quality, differentiation, value adding, niche marketing; finally of labour market policies to accommodate the inevitable changes, by improving the quality of labour through education and training, and by facilitating the access to jobs through information and counselling. (OECD 1994, p. 67).

Lately, responding to a stream of arguments concerning the multiple functions of agriculture (among which rural employment), the OECD seems to examine the possibilities of "decoupling" the remuneration for these functions from the production of the marketable goods, and of linking it (or "targeting") to certain measurable indicators of these other functions {OECD, 1998 restricted, AGR/CA(98)9}.

Although this discussion is still at an early stage and more elucidation are needed some "caveats" are in order:

1. The "targeting", in the strict sense of linking to "well defined" and "measurable" goals, may be very limited in scope, thus not allowing the adequate fulfilment of the functions; while the "decoupling" may lift any incentive to improve the productive performance, leading to a shrinkage of the activity and of all the social benefits inseparable from it (know-how, productive readiness, genetic stock etc).
2. The contribution of agriculture to such goals as land management or biodiversity cannot be captured by a limited number of indicators. Most of it is yet unknown. Neither would the freezing of any state of affairs, by linking it to a set of indicators, be a desirable situation. The relation of agriculture to these functions is a developing one, which cannot be pinned down and frozen in any particular set of indicators.
3. Any system of remuneration linked to such elusive targets would be very costly to define, operate and monitor, and very bureaucratic. The costs and inefficiencies of bureaucratisation are well known.
4. Any such system would at any time be open to challenge for arbitrariness since in general there are no markets for the public goods produced jointly with the marketable products; or the social costs incurred.

Thus while the concepts of decoupling and targeting are useful to the limited extent that they can be operational, we should be very cautious about accepting them as principles answering the arguments of public goods, externalities, jointness, dynamic competitive advantage, uncertain hazards. For they are not: most of the essential aspects of these cannot be decoupled or pinned down to fixed targets. And as practical methods for dealing with more manageable aspects like the income support to particular target groups they, too, are liable to serious inefficiencies as any other policies, and for that matter the administered prices.

BIBLIOGRAPHY
Carabatsou-Pachaki C.: "Rural problems and policies in Greece", KEPE, Discussion Papers No 18, Athens 1993.

Carabatsou-Pachaki C.: "The quality strategy", KEPE, Discussion Papers No 33, Jan 1994.

Carabatsou-Pachaki C.: "Topics in Agriculture, Rural and Local Development", KEPE, Reports No 23, Athens 1996 (in Greek).

Carabatsou-Pachaki C.: "Creating employment in rural areas" in the above KEPE Report No 23.

Carabatsou-Pachaki C.: "Inequality, poverty, redistribution policies: The primary sector" contribution to the homonymous project in KEPE, 1998 (unpublished, in Greek)

Damianos D., Demoussis M., Kassimis G.: "The empirical dimensions of multiple job holding agriculture in Greece", Sociologia Ruralis, Vol XXXI-1, 1991.

Efstratoglou - Todoulou S.: "Pluriactivity in different socioeconomic contexts", Journal of Rural Studies, Vol 6, No4, 1990.

Efstratoglou - Todoulou S.: "Pluriactivity and off-farm incomes of the agricultural households" in GPA, 1996.

Fennell Rosemary: "Common Agricultural Policy: Continuity and change", Oxford University Press, 1997.

GPA (Agricultural University of Athens): "State and the rural space", proceedings of the 3 ^rd >Congress of agricultural economics, ed. Papazissis, Athens 1996 (in Greek).

Harrison L.: "The impact of the agricultural industry on the rural economy - Tracking the spacial distribution of the farm inputs and the farm outputs", Journal of Rural Studies, Vol 9, 1993.

KEPE (Centre of Planning and Economic Research): "Basic Data of Regional Socioeconomic Development in Greece", Athens, 1997 (provisional circulation)

Maravegias N., Moutsatsos D., Duquenne M.N.: "The GATT and its Impact on Greek agriculture", Greek Centre of European Studies (EKEM) Athens, 1998, (in Greek).

Mitrakos Th., Tsakloglou P.: "The structure of economic inequalities in Greece, OPA, Vol B, 1997.

Mitrakos Th.: "The contribution of individual sources of income to the overall inequality", Economic Bulletin of the Bank of Greece, No 11, June 1998 (in Greek).

Moysides A., Zioga N., Duquenne M.N.: "The rural space in peri-urban areas", in GPA, 1996.

OECD (1994): "Farm Employment and Economic Adjustment in OECD Countries", Paris 1994.

OECD (1994b): "Creating Rural Indicators - for shaping territorial policy", Paris 1994.

OECD (1994c): "The OECD Jobs Study: Facts, analysis, strategies", Paris 1994.

OECD (1995 restricted): "Rural employment indicators", C/RUR(95)5/REV1/PART 1-2 and 3, Group of the Council on Rural Development.

OECD (1996): "Territorial Indicators of Employment - Focusing on Rural Development", Paris 1996.

OECD (1997 restricted): "Agriculture and the rural economy", AGR/CA(97)17, Committee for Agriculture, 125 ^th> Session.

OECD (1998): "Agricultural Policy Reform and the Rural Economy in OECD Countries", Paris 1998.

OECD (1998 restricted): AGR/CA(98)9, 17 Nov. 1998.

OECD case studies:

Ireland, in OECD (1994), p. 77

AGRI, OECD (1996), p. 113

New Zealand, OECD (1998), p. 269

OPA (Economic University of Athens): "Proceedings of the conference on the present and future of the greek economy", 3 volumes, OPA 1997 (in Greek).

Redclift M.R,, Lekakis J.N., Zanias G.P.: "Agriculture and World Trade Liberalization", CABI Publishing, 1999.

Sarris Al.: "Legends, truths and prospects of Greek agriculture", OPA, Vol A, 1997

Tarditi S., Zanias G.: "The impact of the agricultural price policy on the cohesion of the European Union", in Hall, Smithe, Tsoukalis (eds) Competitiveness and Cohesion: An Evaluation of EU Policies, Oxford University Press (forthcoming

1

In fact it has always, up to now, been able to take advantage of the local and national markets, being naturally protected by distance, traditional consumption patterns, and a generalised situation of scarcity which favoured the local procurement of the necessities. Apart from this, institutional protection has always existed, being part of the national sovereignty rights (accordingly, trade concessions were always discretionary and had to be reciprocated). Thus it is unjustified to treat the reluctance of a country to agree on a deep-going disarmament of its agriculture, as having "harmful spill-over effects for other countries" by "causing a loss of trading opportunities" (OECD '98 restricted par.72 and elsewhere); for these opportunities were not actually there, and trade has never been so large as today; and the imaginary loss of opportunities has to be weighted against a very real loss of existing production and of all the trade and non-trade concerns associated with it.
linkurl_blankBack to text.

2

Rural employment is of course a joint outcome of the agricultural activity but so it is for any viable activity. So the important points to be stressed in this respect are that agriculture is one of the few activities viable in the rural areas, especially the remote ones; that it is labour intensive in a time of widespread unemployment; and that it makes use of existing human and physical capital that have few alternative uses in the medium term and some of them even in the long term (agricultural infrastructures) _{.
linkurl_blank}linkurl_blank_{Back to text.}

³ >Also, that the process is monitored so as to see that no country risks to undergo such violent and deep-going changes that would put at risk other extremely important goals of national policy, such as food self-reliance
linkurl_blankBack to text.

₄ The corresponding disaggregation for Greece shows a variation from 12,67% to 40% at NUTS-2 level (S. Aegean and Peloponese regions) and from 7,9% to 59,9% for NUTS-3 level (Dodecanese and Rodopi prefectures). (Data for 1991, with the updated national average of 19,56%. Source KEPE, 1997, pp. 44 and 136,Tables B.2/B.2.3.).
Back to text

₅ In Greece in 1987/88 it accounted for 48,5% of the income of the poor agricultural households, where another 25% came from auto-consumption; for 50.8% of the middle ones with 20,2% from auto-consumption; and for 35,9% of the rich, with 15,8% from auto-consumption. (Sarris, '97). From the same piece of research it appears that the income coming from salaries and enterprises was 7,4% and 3,4% respectively for the poor, 12,1% and 5,8% for the middle, and 11,0% and 11,1% for the rich agricultural households. So, in no case is the monetary income from supplementary employment very substantial.
linkurl_blankBack to text.

⁶ >The effects of such an exodus during the '50s and the '60s in Greece, and probably in other countries, are still very vivid in our minds and were disastrous for the social fabric, the economic and cultural life of many areas. Especially the mountainous and islandic ones, which were vulnerable anyway, where it disrupted the continuity of the social and productive structures and set up a rapid process of depopulation.
linkurl_blankBack to text.

⁷ >Although it is possible that support, being universal and almost uniform for reasons of competition, may have not caused such a great distortion in trade and patterns of production, as it is usually thought, but only a universal transfer of income from consumers and tax payers to the producers of agricultural products; who, for very well documented reasons of inelasticities in production and consumption, cannot follow the income generating capacity of the other sectors.
linkurl_blankBack to text.

⁸ >A study for Austria for the '80s showed that most farmers with low income and especially those in the less favoured areas gained little from the then applying policy of administered prices and export subsidies, in comparison to an equilibrium of domestic market situation (OECD '98, p. 62, Box 3.2), which, however, would not be the case in an open economy. Also, from EEC analyses it becomes clear that the Mediterranean regions receive considerably less than farms in the northern regions of the Community.
linkurl_blankBack to text.

9

>But the CAP was not meant originally as a social welfare policy, but as a development policy for European agriculture coupled with a redistribution policy towards the branches and the countries who could not benefit much from the common market. And it has succeeded in both aims.
linkurl_blankBack to text.

_{10 Several years ago I had pointed out (Carabatsou-Pachaki 1994, p. 13): AAs Community statistics show, agricultural income is hardly maintained at constant levels, thanks to the reduction of the agricultural population, while expenses for sustaining the CAP have been rising disproportionately to other magnitudes of the agricultural sector. This is due, according to analyses made also at Community level, to compressed agricultural prices under the pressure of persistent surpluses, while a large part of the CAP expenditure leaks out to the handling of surpluses, to the distribution and promotion channels, to the subsidization of foreign consumers and, indirectly, to the industries producing agricultural inputs. And the OECD notices (OECD '98, p. 70-71): "work carried out by the OECD Secretariat on the relative efficiency of agricultural policy instruments commonly used in OECD countries for transferring incomes to farmers, concludes that less than one-third of what is spent on support programmes results in additional farm income. Furthermore, the limitations of such policies are increasing over time as the proportion of gross receipts accounted for by purchased inputs is high and growing, so that the impact on net farm income of market price support policies is very low". Thus, the tendency towards reform of these policies is inescapable
linkurl_blank}linkurl_blank_{Back to text.}

11

_{But for the above to have an opportunity to hold, reform must not be so deep-going as to throw whole regional productive systems out of operation. For, in a dynamic perspective, new competitive advantages develop but only for those who manage to maintain a lively sector, present in the markets and keeping track of the developments. And, in any case, the comparative advantages revealed in the markets are based on private costs, and no account is taken of social costs and of unknown or uncertain factors, like environmental and health hazards; and so they cannot be but only one of the factors influencing the allocation of activities.
linkurl_blank}linkurl_blank_{Back to text.}

¹² >The profile does not explicitly include the owner-operators who may be highly indebted, many of which will not be able to use one of the adjustment strategies (i.e. diversification, pluriactivity, cutting on elastic production and consumption expenses) and who, therefore, will run the higher risks of bankruptcy, without necessarily having run an inefficient farm or used inefficient practices. These are not deemed by the OECD to fall into a state of need in the social welfare sense (OECD '94, p. 64), because some analyses in strong countries have shown that average wealth and household income of the farmers were not very low with respect to the national average (USA, Wisconsin survey, OECD '94, p. 29). This however is by far not the case in other countries, where farmers are among the poorest groups of the population. And even in terms of the value of their real estate, this will dramatically fall, in all but some tourist areas, if the agricultural use ceases to be a viable perspective in the region.
linkurl_blankBack to text.