As with any technology development, agricultural automation can have negative social and environmental consequences. Therefore, while the above-mentioned benefits are promised, they may not happen automatically and they depend on good management. Agricultural and economy-wide structural factors can hinder the inclusive, sustainable adoption of agricultural automation. Land fragmentation, for example, is a serious constraint in many regions and can make agricultural automation economically unviable. Lack of enabling infrastructures, such as roads, connectivity and electricity, can also shape adoption and exclude producers operating in more disadvantaged and remote areas. Under certain conditions, agricultural automation can displace rural labour and lead to negative environmental consequences such as land degradation and biodiversity loss. These challenges are introduced in the following sections and are discussed in more detail in Chapters 2 and 4.
Unequal capacity
The promised benefits of agricultural automation may not be evenly distributed among producers and other stakeholders, exacerbating social inequalities and creating new ones by favouring already powerful actors in food production.48, 49 This may be the case in particular if technology companies – already large and with substantial market power – retain and own data, which they may use in a manner not in conformity with data protection policies, leading to the creation of data monopolies.50 Inequalities can also be exacerbated if larger, richer and more educated producers have greater capacities (e.g. finances, rural infrastructure, skills) to invest in new technologies or to retrain and learn new skills. Indeed, many farmers may lack the basic capacity to operate digital automation technologies or understand how they work. A good agricultural practitioner is not necessarily an expert in digital technologies, and the same applies to extension officers and service providers. Capacity building and farming adaptation are essential for the uptake of automated equipment and its correct use; only with capacity can farmers exploit the full potential of automation.15
In this regard, women are often more marginalized than men from education opportunities18 and have less access to finance.51 Men tend to take over the buying and selling of crops, and to own and operate the new equipment, reducing women’s control of income produced and relegating them to the more labour-intensive tasks of weeding and transplanting.52 Similarly, rural youth, especially women, face significant obstacles to obtaining quality education and training, as well as access to land, credit or markets.53
Labour disruption
Emerging evidence from other industries suggests that automation may increase demand for higher-paying jobs requiring secondary education, where humans have a comparative advantage over machines (e.g. data management and analysis), but reduce demand for jobs that involve routine tasks (e.g. planting and harvesting).54, 55 As countries develop, the numbers for total employment in agriculture decline; nevertheless, there are still approximately 300–500 million waged workers who depend on farm jobs.56 In many countries, the percentage of the workforce in agriculture remains high – for example, in Burundi (86 percent), Somalia (80 percent), Malawi (76 percent), Chad (75 percent), the Niger (73 percent) and Uganda (72 percent) – often accompanied by high rates of illiteracy, poverty and gender inequality.
In such countries, a reduction in the direct labour requirements per unit of output may create inequalities or deepen existing ones. For this reason, in some contexts, agricultural automation may be politically unattractive and unfeasible. Ultimately, the impact on labour and wages will be determined by a series of factors, including the capacity to generate new and more attractive jobs or alternative decent employment options outside the agriculture sector. It will also depend on whether scale effects – where farmers expand the scale of their production and increase their income – outweigh substitution effects when labour is pushed out of the sector.57 Yet, with the right policies and legislative and regulatory environment, agricultural automation can create economic opportunities, encourage decent employment that provides a living income and reasonable working conditions, and draw youth back into the agriculture sector.
Environmental concerns
There are concerns that, if not well managed, some types of agricultural automation, especially those relying on heavy, large machinery, may jeopardize environmental sustainability and resilience by contributing to deforestation, farmland monoculture, biodiversity loss, land degradation, soil compaction and erosion, salinity buildup, and drainage system malfunctioning.58 While these concerns must be taken seriously, many can be avoided or minimized with appropriate policies and legislation. Moreover, certain new advances in automation machinery and equipment – especially small equipment that relies on AI – can actually reverse some of the negative environmental impacts of old automation machinery (see Chapter 3).
The potential opportunities, challenges and consequences of agricultural automation depend on the specific technology used, its design, and how well it suits local conditions and is adapted to local realities. In addition, the level of socioeconomic development, as well as institutional and political constraints, determine the mix of suitable technologies likely to be adopted. As a consequence, the effects – positive and negative – of agricultural automation are highly context-specific. It is important to assess whether the environmental, social and political conditions are right in each country or region before proposing specific automation solutions. Not all automation technologies are suited to all contexts, and adapted versions may have to be considered.
