Some thirty years ago, former communist and socialist countries in Central and Eastern Europe (CEEC) began the process of transitioning from a centrally planned to market economy. The major structural transformations, creation of fundamentally different institutions, and privatization and promotion of private-owned enterprises, which followed after the initial economic collapse in the early phases of transition, have gradually led to an increase in productivity and economic growth [1
]. Whilst some of the early adopters became leaders of these systemic transformations and slowly caught up with their developed Western neighbors (CEEC who joined the European Union), the laggards, like Bosnia and Herzegovina (BH) and other Western Balkan countries, still experience sluggish and unstable economic growth [2
]. The transition to a market economy in CEEC involved a substantial reallocation of labor and capital across economic sectors in order to increase overall efficiency. This process was heavily disrupted in BH by the armed conflict (1992–1995), which led to massive devastation, large-scale migrations, ethnic segregation, and a general economic and socio-cultural decline. A large proportion of the population in rural areas and towns previously employed as skilled workers in industry became unemployed after the destruction of the existing state-owned industrial capacities, as well as after restructuring and privatization of the remaining capacities in the post-war time. In the absence of other income options, a significant part of that population shifted to agriculture as a main or as an additional source of income. At present, more than 30% of the total households in BH are engaged in agriculture, while only 5% of them are considered commercial farmers [3
]. This underscores the significance of the agricultural sector in BH in terms of food security, rural poverty reduction, and further means to mitigate the social burdens of economic reforms and restructuring [4
]. At the same time, because agriculture remains one of the main sources of income for most rural households in BH, this also means that the livelihoods of many have become increasingly more vulnerable to current and future climate pressures that directly impact agricultural productivity [5
More than 60% of the population in BH live in rural areas [6
]. Although agriculture as a share of gross domestic product (GDP) is constantly decreasing, it still forms the backbone of the rural economy, employing 20% of the total workforce and constituting 6.4% of the total GDP [7
]. The potential for agriculture in BH is substantial. Of the total 2.1 million hectares of agricultural land, 46.5% is arable. However, only 50% of that arable land is currently utilized in agricultural production [8
]. Favorable climatic conditions, its geopolitical position, abundant freshwater supplies, and relatively cheap labor costs give the agricultural sector in BH a clear, comparative advantage over many other European countries and potentially gives BH an advantage in terms of labor-intensive productions. Livestock production has the highest economic value in the present agricultural production system in BH, with great potential for further expansion and intensification due to the high availability of grasslands and pastures in areas less favorable for intensive crop production. Agriculture in BH, however, suffers from low investment levels and low overall production and productivity, involving rather extensive farming practices and technologies, low levels of financial capital inputs, and productions carried out on small and fragmented farms. This is a problem partly inherited from the past socio-political system, where agriculture was marginalized as a result of industrial development. This was further exacerbated in the post-war period through poor governance in the ongoing processes of transition [8
Poverty in BH is mostly a rural phenomenon—close to 80% of the total poor live in rural areas [9
]. Non-farm employment opportunities in rural areas are limited. Many of these areas have poorly developed infrastructure and services, all of which are basic preconditions for social and economic development. In addition, frail public and governmental policies and resources do little to support development activities. These conditions preceded the massive changes caused by war and the transition process, but the constraining conditions still remain largely unaddressed. This is especially pronounced in remote areas, where a continuous process of out-migration results in the shutdown and degradation of the existing infrastructure and public services.
From a natural resource point of view, BH is considered highly vulnerable to climate change [10
]. Extreme weather events, such as increased intensity of droughts, frequency of heat waves, and heavy precipitation resulting in floods and landslides, are increasingly occurring and have already caused significant economic losses and environmental degradation [8
]. There is a limited adaptive capacity to cope with both climate and other shocks in BH. The situation is actually similar to that found in many developing countries. This inherently low adaptive capacity is the main determinant of vulnerability to climate change in most rural areas of BH, even more so than the degree to which these areas are de facto exposed to significant climatic variations [5
Adaptation to climate change and other shocks and perturbations is crucial both in order to enhance the resilience of both the agricultural sector at large and for individuals to secure and improve their livelihoods. Adequate responses in terms of adaptation to climate change depend on issues such as adaptive capacity, knowledge and skills, robustness of livelihoods and alternatives, resources, and access to appropriate institutions in order to undertake effective adaptation [13
]. While technological development, government programs, and insurance schemes require greater investments from both the public and private sector to be subsequently adopted by farmers [14
], the adaptation to climate change at the farm level does include many possible responses. It could encompass changes in crop and livestock management practices, land use and land management, and a variety of both on-farm and off-/non-farm combined or diversified livelihood strategies [15
]. While climate change has yet to be mainstreamed into national and regional development policies in BH, individuals and communities are already adapting to the factual changing climatic conditions. Such adaptation strategies are mostly reactive and carried out in response to perceived and experienced adverse impacts of climate change and variability [16
]. Perceptions of these issues are key components in on-farm decision-making processes. Observations, and not least experiences over time, shape farmers’ climate change and climate variability perceptions and influence or inform their choices of appropriate adaptation strategies. However, adaptations in agriculture are not carried out with respect to climatic stimuli alone, but one can rather see it as “joint effects of multiple forces” [14
] (p. 92). People also adapt to changes in their external frame conditions in different ways and it can be difficult to assess complex changes in institutional arrangements and explore what can be reasonably linked to climate change and what can be attributed to other frame condition changes. These interactions additionally generate complexities in relation to how people adapt in both the short- and long-term.
The objectives of this study were to: i) assess the livelihoods of rural households in BH, their access to assets, and livelihood diversification strategies; ii) investigate how the households are affected by climate change and how they perceive these ongoing changes; and iii) analyze household adaptation strategies and the degree of adoption of certain practices and technologies. The results were analyzed and discussed in the context of three different agricultural regions and three wealth groups in BH according to total income level of the households.
2. Research Framework
Climate change is widely considered as one of the main environmental challenges of the 21st century. According to the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment Report, the globally averaged combined surface temperature data as calculated by a linear trend showed a warming of 0.85 (± 0.20) Cover the period from 1880 to 2012 [17
]. The increase of global mean surface temperature by the end of this century is predicted to be 1.5–4 °C in most scenarios [18
]. In addition, it is expected that the incidence and duration of heat waves, droughts, floods, hail, storms, cyclones and wildfires, intensified melting of glaciers and other ice, sea level increases and soil erosion, will all increase over this century. This may pose a significant threat to ecosystems and their various services and the vulnerability of many human systems are likely to increase [18
Climate change and agriculture are interrelated processes, both occurring at a global scale. Agriculture is extremely vulnerable to climate change. Increased incidence and duration of extreme weather events cause yield reductions or crop failures [19
], as well as increases in and the emergence, growth, and frequency of weeds and pests [20
]. It may also cause damage to farm infrastructure [21
]. While some regions of the world may benefit by the increase in temperatures, the overall impacts of climate change on agriculture will be negative, threatening global food security [20
Globally, smallholder farmers constitute about 85% of the world’s farmers [22
]. Most of these people are located in the low-income countries of South Asia and sub-Saharan Africa, but they are also dominant in rural areas of many upper middle-income countries [23
], including CEEC, the Western Balkan countries, and BH. The main characteristics of these regions are large rural populations, widespread poverty, and extensive areas of low agricultural productivity due to steadily degrading resource bases, weak markets, and high climatic risks [24
]. Most smallholder farmers rely directly on agriculture for their livelihoods and survival and have limited resources and capacity to cope with the shocks/impact of climate change. Any reductions in agricultural productivity could have significant impacts on their food security, nutrition, income, and well-being [25
]. Climate change is expected to further exacerbate the risks and uncertainties that farmers face. It affects food production directly through changes in agro-ecological conditions and indirectly by affecting growth and distribution of incomes, and thus demand for agricultural produce [26
Changes in climatic conditions will require different adaptation strategies, in terms of both overall livelihood strategies and adjustments in agricultural production itself in order to alleviate the severity of climate change impacts. The empirical research on adaptation across disciplines has identified resource, institutional, informational, and financial constraints as the most significant determinants of adaptation [27
]. Furthermore, adaptations can be either planned (public) or autonomous (private) with the latter being carried out depending on how the perceptions of climate change are translated into agricultural decision-making processes [29
]. However, the extent of autonomous adaptations will likely not be enough to cope with the negative effects of climate change and may even lead to maladaptation due to clashing cultural contexts and social goals [31
]. Thus, the “mainstreaming” of climate change adaptation into policies would be necessary in order to enable and facilitate effective planning and capacity building for adaptation to climate change [32
]. Adaptation to climate change in agriculture can be achieved through a broad range of management practices and adoption of new technologies [14
]. However, there is no “one-size-fits-all” framework for adaptation and adoption of new practices and technologies. Successful adaptation should be based on adequate, local, and scientific knowledge and be continuously updated based on new research findings.
Adaptation, whether analyzed for purposes of assessment or practice, is closely associated with vulnerability, since the extent of sustainable adaptation depends on the magnitude of climate change and its variability, as well as the capacity to adapt to these changes [33
]. The limited access to livelihood assets and capabilities often shapes poverty and consequently the lack of adaptive capacity [34
]. However, people´s adaptive capacity may be underestimated by only looking at access to material resources and one should also involve socio-cognitive variables in order to develop more realistic scenarios for adaptation and important policy implications [35
Livelihood as a concept is widely used in the literature, linked to vulnerability, poverty, and rural development. According to one of the earliest definitions of livelihood by Chambers and Conway [36
], “a livelihood comprises the capabilities, assets (stores, resources, claims and access) and activities required for a means of living. A livelihood is sustainable when it can cope with and recover from stress and shocks, maintain or enhance its capabilities and assets, and provide sustainable livelihood opportunities for the next generation; and which contributes net benefits to other livelihoods at the local and global levels and in the short and long term” (p. 6). In this definition, capabilities are the options one possesses to pursue different activities to generate income required for survival and to realize its potential as a human being. Capabilities are determined based on the portfolio of assets one possesses, based upon which one makes decisions to produce outcomes necessary for sustenance and well-being.
We applied the sustainable livelihoods approach (SLA) as a conceptual framework [36
] to assess the households’ access to livelihood assets, their activities and reported outcomes, and the contextual factors that influence them. This approach is based on mapping people’s access to assets and the way people use and access these assets. Five main categories of capital contribute to livelihood assets: natural, physical, human, financial, and social capital [38
]. Furthermore, access to livelihood assets is mediated by institutions, social relations, and policies (transforming structures and processes), as well as shocks, trends, and seasonality [37
]. Under such circumstances, poor people undertake a range of activities and choices (livelihood strategies) in order to achieve different livelihood outcomes [39
]. In terms of livelihood strategies under the lens of the SLA, this means that rural households with access to agricultural means of production can choose between agriculture and non-agricultural economic activities through diversification processes. Different adaptation strategies within agriculture can also be considered, such as extensification vs. intensification, as well as adaptation as leaving agriculture through “exit options” such as off- and non-farm activities, migration, and remittance strategies.
As much as climate change is one of several drivers for particular courses of action, other frame conditions such as agricultural policies, market conditions, alternative economic options and other factors, situate households with different asset access in different positions, which also influence and complicate their livelihood choices. There is no single most profitable or desired option for all households or individuals; but diversification patterns in terms of adaptation must be understood as a broad specter of opportunities for actors with different initial asset access and for the same households over their lifespan or demographic cycle. It is also useful to see these decision processes as combinations of free choice vs. necessity [39
] and also see decisions in a structure-agency dichotomy. Choices are made, but under such strong preconditions or bearings that the choice is often “given” or placed within. Furthermore, BH is a country with a pronounced heterogeneity in terms of geographical, agro-ecological, and climatic conditions, as well as unique and rather asymmetric constitutional, political, and institutional arrangements and governance structures. This means that households under different socio-economic and agro-ecological conditions have different realities to relate to, as well as different perceptions of climate change and perceived adaptive capacity for agricultural and other income generating potential [35
3. Study Area
BH has, as mentioned, a complex economic, political, and historical background, with substantial differences in agro-ecological and climatic conditions and also varying demographic, ethnic, cultural, and religious compositions (Figures S1 and S2
). The recent war in BH (1992–1995) resulted in devastation, and the death of at least 100,000 people [40
]. There was massive emigration of more than two million people and massive internal displacements and migration, with significant and lasting consequences on the demographics and economy of its local communities. The psychological effects of war are still visible and reflected in political instability caused by the instrumentalization of ethno-nationalism [41
]. Each political decision is carefully reviewed for its potential impacts on the existing three ethnic groups, as well as the balance of power and resources between the state, its two entities, regional and local governments. In order to avoid the potential biases and “politicization” of livelihood assets, activities and outcomes found in the different state entities and ethnic groups which may arise from this study, we opted to interpret the results based on the location (in terms of agro-climatic conditions, instead of administrative units) and household income (total household income of surveyed households). The main features and general distinction between agricultural regions, as well as different challenges faced in them are explained in more detail in Supplementary Note 1
BH is a south-east European country located in the Western Balkan region, with a total surface area of 51,209.2 km2
. The landscape is mainly hilly to mountainous, with an average altitude of 500 m a.s.l (meters above sea level). Of the total land area, 5% is lowlands, 24% hills, 42% mountains, and 29% karst region. The general atmospheric circulation, the dynamic topography, the orientation of mountain ranges, the hydrographical network, and the vicinity of the Adriatic Sea have created conditions for a wide spectrum of climate types and subtypes in BH [42
]. In brief, these include a humid continental climate, represented mostly in the northern and lowland central parts of the territory; a sub-alpine and alpine climate in the mountainous region of central, east, and western BH, and a Mediterranean climate dominant in the coastal area and lowlands of Herzegovina (southern BH). Areas above 800 m a.s.l. are mostly unsuitable for intensive agricultural production due to biophysical constraints.
The country is divided into six agricultural regions based on dominant climate conditions, elevation, share of different types of crops and livestock and the degree of economic development. These can further be roughly grouped in three main agricultural regions: the lowlands, the hilly-mountainous, and the Mediterranean region [43
]. The study sites are situated in these three regions (Figure 1
), referred to as the northern (lowland), central (hilly-mountainous), and southern region (Mediterranean).
The studied rural households in BH are exposed to multiple stressors that constrain their livelihood options. Many stressors are typical for economies in transition, and include factors such as inefficient institutions and policies, inadequate infrastructure and services, imperfect markets, and lack of market access. Under such conditions, continued high dependence on agriculture in rural areas is driven more by a lack of alternative income sources than its inherent profitability.
The results indicate a very high relative dependence on agriculture, especially in the north and south, and irrespective of wealth group. Higher income from agriculture is the key driver for the wealthiest group in our study, and most of them are from the southern region. Low agricultural income and a significant degree of subsistence, coupled with low income from off-farm activities, constrain the livelihoods and income from the poorest group, most of which are from the northern region. The situation in the central region resembles that found in the north, with one difference—low incomes from agriculture are partly complemented by income from non-farm and off-farm activities. Higher agricultural incomes in the southern region can be mainly attributed to more favorable climatic conditions in this region, which enables farmers to time their production in order to exploit market opportunities and also produce more lucrative crops. In addition, households in the southern region invest more in technologies (e.g., irrigation, greenhouses) and are more often part of agricultural associations. Wealthier households show a higher degree of entrepreneurship, reflected in more variable market outlet choices.
The studied households expressed their awareness of the recent climate trends. Most respondents in our study perceive an increase in average temperature in line with the actual climate data. The way changes in precipitation are perceived, both in terms of amount and distribution, varied by region. We attribute this inconsistency between perceived and actual precipitation to a decrease in water availability under the increased frequency and intensity of droughts and higher temperatures.
While it is hard to differentiate whether it is climate change, agricultural intensification, or some other factor that is the main motive behind certain adaptation patterns within agriculture, the number of adopted measures shows that there are signs of an overall intensification strategy of agricultural production in BH, as well as marked adaptations to (perceived) climate changes. Most notable was the application of both organic and mineral fertilizers, changes in crops and crop varieties, and irrigation. Other common agricultural practices were those that require no or little investment. The main reported constraints for further adoption and long-term adaptation within agriculture were lack of funds, knowledge, and labor. Certain agricultural practices adopted were in some cases more region-specific, but overall were similar in relation to wealth group. The results also indicate that increased access to different types of assets in most cases influence the likelihood of adaptation in agriculture. Higher access to social capital was found to have a positive influence on the adoption of a significant number of agricultural practices, along with higher access to technologies (tractors) and human capital. Investment in physical capital, like greenhouses or irrigation systems, can also be seen as a more long-term adaptation strategy. Higher access to natural capital (land) and financial capital had a mixed effect, increasing the likelihood of adoption of some practices, but the opposite effect was found mainly in the case of larger on-farm investments.
Climate change in BH enters a landscape of economic recession and constrained income generating possibilities, especially outside agriculture. In this situation, the main resources at hand for the rural dwellers in the sample are land, skilled labor, and agricultural options, for both cash and subsistence incomes. Climate change does not change the heavy dependence on agriculture but has created additional challenges and hardships for most households. The increased incidences of adverse weather events lead to lower and less predictable incomes from agriculture due to production declines and variations, and as the alternative employment options are limited, climate change may lead to increased poverty and vulnerability for those who lack the capacity to adapt.
Our results indicate that agriculture may be or become the main driver of rural development in BH both in its own right and more passively as the only more or less short-term available vehicle of growth and development. The context-specific and local agricultural adaptation practices identified in this study should be taken into account, assessed further, and integrated in rural development policies and climate change adaptation strategies. A more integrative and participative approach to rural development will likely improve the identification and selection of meaningful adaptation options, which should in turn improve rural livelihood outcomes. Potential increased incomes from agriculture can also lead to investments in other sectors in rural areas and support the ability of households to make strategic long-term decisions and improve their adaptive capacity. However, there are constraints for such developments in terms of production, markets, and institutions. Therefore, future research and rural development policies should be focused on rural infrastructure, availability and accessibility to technologies, communicating climate-related information (as well as case-specific adaptation options), improved irrigation technology and improved access to other inputs, easier land transactions and land consolidation measures, access to markets, farming cooperatives, better agricultural extension, and accessible and more favorable credits, including pilots on crop insurance.