La convergence des revenus agricoles et non-agricoles : mythe ou réalité ?
Beyond their diversity, development strategies are all based on the same paradigm, more or less explicit. They follow the model historically observed in high-income countries: increasing agricultural productivity to reduce food prices and encouraging labor migration to sectors where productivity is higher—industry, then services—in a context of increasing urbanization. In this model, which has been the subject of abundant literature, agricultural labor productivity is supposed to converge with that of other sectors, which theoretically leads to a gradual equalization of agricultural and non-agricultural incomes. But has this convergence of productivity actually occurred? How has the convergence of agricultural and non-agricultural incomes, when it has occurred, been achieved?
Here we examine the relative evolution of agricultural and non-agricultural labor productivity in different regions over the past twenty years, and draw some lessons for public policies with a view to reducing income inequality, one of the 17 United Nations Sustainable Development Goals.[1]A promising option in the Global South is to provide direct aid to farmers, similar to what is done in the Global North. This aid should be linked to good environmental practices, such as restoring soil fertility or reducing deforestation.
Poor countries lagging behind
Overall, in high-income countries and so-called emerging middle-income countries, labor productivity, measured by gross domestic product (GDP) per worker, has grown faster in agriculture[2] than in other sectors. However, the gaps remain considerable, since over the period 2016-18, agricultural labor productivity represented on average only 38% of non-agricultural labor productivity in high-income countries and 20% in middle-income countries. In low-income countries, agricultural labor productivity deteriorated in relative terms, falling from 23% to 22% of that obtained on average in other sectors of the economy (table 1).
These results should be viewed with caution. Indeed, a large proportion of the population employed in agriculture is not employed full-time, which tends to underestimate, probably significantly, the real GDP per agricultural worker.[3]However, it can be assumed that the underestimation of agricultural labor productivity has relatively little effect on the gaps in agricultural labor productivity between countries, as well as on the relative evolution of agricultural and non-agricultural productivity within each country.
The situation in low-income countries, where nearly 60% of the workforce still works in agriculture, is particularly worrying. In addition to the widening gap between agricultural and non-agricultural productivity, GDP per agricultural worker in these countries is, on average, in 2010 dollars[4], 39 times lower than that of rich countries and 3.5 times lower than that of emerging countries. This results in a very high rate of extreme poverty (45 % in 2017, probably increasing with the Covid-19 crisis) and poor competitiveness of farmers, both on export markets and with regard to imported products.
Large sectoral gaps in South Asia and sub-Saharan Africa
Average agricultural and non-agricultural labor productivity has evolved in contrasting ways across regions. North America has both a high GDP per agricultural worker and a strong convergence of agricultural and non-agricultural productivity, with the ratio between the latter reaching 78% compared to 33% in the European Union (55% in France).
Conversely, in South Asia, GDP per agricultural worker is very low and has grown less quickly over the past twenty years than GDP per non-agricultural worker. Thus, in India, agricultural labor productivity is now around 1,700 2010 dollars, half that of China, and equivalent to 20% of non-agricultural labor productivity, compared to 24% over the period 1995-97.
Sub-Saharan Africa has the lowest GDP per agricultural worker ($1,300 in 2010). Despite a relative improvement, this represents only 17.1% of GDP per non-agricultural worker. (table 2).
Dark African outlook
In many low-income and emerging countries, the prospects for convergence of agricultural and non-agricultural labor productivity in the coming decades are bleak. Perhaps the most problematic case is sub-Saharan Africa, given the projected sharp increase in the rural population in the region over the next twenty years.[5].
If GDP growth per agricultural and non-agricultural worker in the sub-Saharan region continues at the same rate as that recorded between 1995-97 and 2016-18, i.e. +68 % and +13 % respectively in constant dollars, labor productivity in agriculture by 2040 will be around 2,300 2010 dollars, a level significantly lower than that recorded today in China (3,500 2010 dollars). It would then represent only 25 % of the labor productivity obtained on average in other sectors.
A substantial acceleration in the increase in agricultural labor productivity in sub-Saharan Africa faces several obstacles. While the potential for increased yields is enormous, farmers still need access to credit, inputs, and extension programs, and must be integrated into efficient supply chains. Moreover, many farmers struggle to increase the size of their farms due to demographic pressure. On the contrary, in many African countries, as in India, we are seeing a decrease in the size of farms and the available land per agricultural worker. In these countries, agriculture is caught in the "Lewis trap," as Bruno Dorin puts it.[6] : the development model of high-income countries, based on both high land productivity and the availability of a large area per agricultural worker, is only applicable there at the cost of a massive rural exodus that would empty the countryside. However, with one major difference: unlike what has happened, for example, in Europe since the 19thth century, a large part of the displaced workforce would probably not be absorbed by industry and services or by emigration. Add to this the fact that while the growth of cultivated areas in industrialized countries has historically been achieved without concern for environmental impacts, their expected doubling in sub-Saharan Africa by 2050 raises fears of a continued reduction in forests and pastures, detrimental to biodiversity and the climate.[7].
From productivity to income: the case of France
Labor productivity is the main determinant of farmers' income levels, but it is far from the only component. In all regions of the world, farm households—the farm manager, their spouse, and possibly other household members—earn part of their income from off-farm activities. On the other hand, cash transfers—social benefits and government support under agricultural policies—supplement earned income, mainly in high-income countries.
The example of France is instructive. The multi-activity of farmers[8] or their household[9] is a source of significant income and promotes the maintenance of agricultural activity through various channels: easier access to credit, reduction of the variability of farm income, etc. In addition, public subsidies for agriculture under the CAP or national policy measures massively support the income of French producers. According to our calculations, the amount of direct payments and other aid to agriculture (reduction of social security contributions, tax exemptions, etc.) is equivalent on an annual average, over the period 2016-2018, to some $19,500 per agricultural worker, or almost half of the gap observed between GDP per agricultural worker and GDP per non-agricultural worker[10]These results are probably broadly extrapolatable to many other high-income countries.
Three levers for public policies in the South
Data are lacking to assess the extent of farmers' multiple occupations in emerging or less developed countries. Situations are a priori very diverse. Active rural development policies, creating non-agricultural jobs in the countryside, would allow farmers to either maintain their agricultural activity thanks to the additional income from outside, or to abandon agriculture for a more stable and better-paid job.
Strengthening social protection in agriculture, aimed at combating poverty, improving the health of rural populations and indirectly supporting agricultural investments, was vigorously promoted by the FAO under the general leadership of former Brazilian Minister José Graziano da Silva (2012-2019). Despite the strength of the arguments put forward[11], it does not appear that significant progress has been made in this area.
Finally, paying direct aid to farmers is a promising option. This is the path chosen in most high-income countries and, to a limited extent, in some emerging countries such as China. Digital tools and mobile telephony make feasible what, until recently, was still considered a practical impossibility in low-income countries: paying money to millions of farmers, the majority of whom do not have bank accounts. This aid would help reduce poverty and reduce income inequality between cities and the countryside, while also encouraging farmers to invest in ensuring food security. It is also justified on environmental grounds.[12] : encouraging farmers to preserve meadows and forests, stimulating the adoption of economically efficient agroecological practices with positive impacts on human health, biodiversity, and the climate. This is the shift that support for agriculture is taking, timidly but resolutely, in Europe. It deserves to be seriously considered in developing regions. The budgetary obstacle is not insurmountable: the funds devoted to input subsidies, widely criticized in particular for their ineffectiveness, offer room for maneuver. Isn't it time to think outside the box?
[1] SDG 10 aims to “reduce inequality within and among countries.”
[2] "Agriculture" here includes the agricultural sector itself, as well as fishing, hunting and forestry.
[3] Conversely, however, some agri-entrepreneurs (notably retired civil servants) may not be considered agricultural workers even though they contribute to agricultural GDP.
[4] One US dollar, valued in 2010, is worth 1.19 dollars today.
[5] Between 1995-97 and 2016-18, the number of agricultural workers increased by 50% in sub-Saharan Africa and by 2% in South Asia and Latin America.
[6] Dorin Bruno, 2021 (forthcoming). “Theory, practice and challenges of agroecology in India”, in Hubert Bernard, Couvet Denis (Dir.), “The agroecological transition. What prospects in France?”, French Academy of Agriculture, Presses des Mines, Paris.
[7] Jean-Christophe Debar, “2050 Outlook: Strong Pressure on Land in Africa,” FARM Blog, February 24, 2020.
[8] In 2017, 11 % of self-employed workers in agriculture in the strict sense (excluding forestry and logging) taxed under the actual system, and around 20 % of those taxed under the agricultural micro-profit system, combined their activity with salaried employment. In agriculture in the broad sense (including forestry and logging), the average self-employed income of a single-earner amounts to €1,510 per month, compared to €680 for a multiple-earner. However, multiple-earners receive almost twice as much income from all their activities as single-earners, i.e. €2,790 per month on average, since their salaried activity represents three-quarters of their income. Source: "Employment and income of the self-employed", Insee References Collection, 2020 edition.
[9] Non-agricultural income represents 47 % of the total income of farm households with a “medium or large” farm, according to the criteria of the RICA (Farm Accounting Information Network). This share rises to 63 % if all farm households are taken into account. Source: Piet L., Benoit M., Chatellier V., Dakpo K H., Delame N., Desjeux Y., Dupraz P., Gillot M., Jeanneaux P., Laroche-
Dupraz C., Ridier A., Samson E., Veysset P., Avril P., Beaudouin C., Boukhriss S. (2020). “Heterogeneity, determinants and trajectories of French farmers’ income”. Agr’income project report, Ministry of Agriculture and Food.
[10] This amount does not include certain public expenditures that nevertheless benefit farmers, such as agricultural research, nor state support for the social benefits system (pension, health, old age, maternity, etc.) for farmers. In 2019, the tax resources allocated to this system reached 4.9 billion euros, compared to 14.9 billion for public assistance to agriculture.
[11] See in particular: FAO, 2013. “The interaction between social protection and agriculture. A review of evidence”, and FAO, 2015. “The State of Food and Agriculture. Social protection and agriculture: breaking the cycle of rural poverty”.
[12] Dorin B., 2021, op. cit.
