Understanding farmers’ intentions to adopt organic farming in Albania

Organic agriculture in Europe is expanding, yet is still in the early stages in less developed areas of Europe. Understanding the factors of converting to organic agriculture is a key component for both policy design and attaining effective support for the organic sector from the government, donors, or the private sector. Therefore, the study attempts to explore the determinant factors of farmers’ intentions to engage in organic farming based on primary data collected through a farm survey in Albania. Results show that subjective norms, perceived behavior control, favorable attitudes toward organic farming conversion, as well as awareness about risk from conventional farming are positively associated with the probability of converting to organic farming. Farmers’ perceptions of EU policy opportunities and attitudes towards environment protection are negatively associated with farmers’ tendency to convert. The study findings call for the use of financial and non-financial policy instruments for supporting conversion to organic farming and the need for increased information on opportunities and costs expected from the integration into the EU single market.


Introduction
Recently, organic agriculture has been developing worldwide.This is a positive development in Europe (see Willer et al., 2018) which, beyond the advantages of environmental protection and improved human health, has also been spurred by consumer awareness and concerns regarding food safety and quality (Rana and Paul, 2017) and concerns emerging from the use of conventional methods (Biao et al., 2003).In developing countries, organic agriculture provides a promising approach to reducing poverty and resource degradation (Wollni and Andersson, 2014;Qiao et al., 2018).
The adoption of organic agricultural techniques is seen as an important aspect of moving farmers towards a more sustainable agriculture, while creating added value in the sector (Seufert et al., 2019).Therefore, understanding the factors that lead farmers to adopt organic agriculture is a key component for policy design necessary for the sector's development.Examining farmer's decisions regarding production methods helps in obtaining effective support for the organic sector by the government, donors, or the private sector.Considering these key elements of research in exploring the potential of organic agriculture, the paper's objective is to provide a scientific understanding of the factors influencing farmers' decision to adopt organic farming.
In Albania, but also in other South-Eastern Europe (SEE) countries, the organic sector is still in the early stages of development and can be considered a small niche market.In terms of cultivated area, only around 550 ha of land was certified as organic in Albania during 2017, which is significantly lower when compared to other countries in the region (Table 1).However, while the cultivated land occupies a significantly smaller area, the organic sector in Albania is dominated by forest and wild collection areas with a total of 380,612 ha (Table 1).
Official data regarding land in conversion indicate a potential growth of the sector in the near future.The total number of organic operators was 150 in 2017 where majority of them were producers (Table 1).Moreover, medicinal and aromatic organic plants (organic MAPs) make up the majority share of organic production and exports of Albania (Bernet and Kazazi, 2012).In 2017, wild MAPs accounted for 72% of the total production, while other organic products such as strawberries, herbs and fresh vegetables continue to be produced in modest quantities (Cakraj, 2019).Most of the output is export driven (mainly with European and regional destination countries), while there is very limited domestic demand for organic products despite an explicit preference of consumers for organic products (Imami et al., 2017;Skreli et al., 2017).
Given the need to strengthen the organic sector, the Albanian government supports farmers through subsidies under organic farming schemes and acknowledges organic farming as a diversification activity and promising sector which can withstand competitive pressure from external markets.However, the number of organic operators during the recent decade, while increasing, continues to be limited.The failure of integrating farmers into organic agriculture might be due to limitations in terms of knowledge, input and transition costs, concerns regarding plant protection, or market expectations (Curtis and Quarnstrom, 2019;Mzoughi, 2011;Nelson et al., 2015).
Utilizing empirical evidence drawn from a structured survey with organic and nonorganic farmers in Albania, this paper contributes to the existing gap in the literature on the uptake of organic farming in a post-socialist country which is also under the process of integration toward European Union (EU).In order to contribute to the literature, in addition to the frequently sourced factors from the Theory of Planned Behavior (TPB) (see Yanakittkul and Aungvaravong, 2019) the authors also make use of perception-related factors (e.g., EU agriculture policy opportunities).A few studies, such as Zagata et al. (2020), have attempted to provide an overview of the organic agriculture trends in a post-socialist country, and found that organic agriculture is an outcome of policy measures rather than social will.Moreover, similar to  Crisostomo et al. (2012), organic farming policy is predominantly driven by EU policies, as it happens even in European Member countries.Albania, being a candidate country, risk to behave similarly to Member countries (Crisostomo et al., 2012), where compliance with EU agriculture policies is a guiding principle for the sector, while national debate emerges generally scarce.
To the best of the authors' knowledge, there is no study analyzing factors influencing organic farming adoption by Albanian farmers.Different from the findings of Zagata et al. (2020), the Albanian organic sector is driven primarily by foreign market demand rather than policies or social patterns.While the former is important to shift the farmers' perceptions on returns from investments, the latter is important for helping farmers to overcome certification barriers and take the first step toward organic agriculture.
In this paper we explored various factors influencing organic farming adoption, namely: perceived behavior control factors, subjective norms, information awareness, risk perception, attitudes towards protecting the environment, and perceived EU policy opportunities.The results have strong implications for the design of a proper policy environment for promoting organic agriculture.The rest of the paper is structured as follows: the next section describes the theoretical and conceptual framework applied as along with the main hypotheses, section 3 outlines the methodology, section 4 describes the results and, finally, section 5 provides some concluding remarks and discussions.

Theoretical and Conceptual Framework
Over the last few decades, the behavior of farmers has been explored through various theoretical and conceptual bases (i.e.Theory of Planned Behavior (TPB), Prospect Theory, Value-Belief-Norm Theory, Protection Motivation Theory, and Psychological Distance Theory).Specifically, for exploring and explaining farmers' behavior regarding the adoption of organic farming, different researchers have applied the TPB (see Hattam, 2006;Läpple and Kelley, 2013).Therefore, the conceptual framework of this paper will be based primarily on the main concepts used by Fishbein and Ajzen (1975).Despite the TPB framework, the paper will also employ a variety of background indicators -such as farmers' perceptions, social and demographic characteristics, etc. (see Figure 1).This approach seems to include all possible determinants of performing a particular behavior, as the motivational factors alone (i.e.TPB concepts) may apply to some behaviors while the performance of some other behaviors is determined to some extent by factors such as the required opportunities and resources (see Ajzen, 1988).Thus, as long as the farmer has the required opportunities and resources, and intends to convert to organic farming, he or she should succeed in doing so.
Under the TPB, the core determinants of farmers intentions to adopt organic farming are subjective norms, perceived behavioral control, and attitude towards the behavior (see Hall and Rhoades, 2010;Hattam, 2006;Läpple and Kelley, 2013;Mzoughi, 2011;Yanakittkul and Aungvaravong, 2019).The concept of subjective norms is used to indicate the perceived social pressure to convert or not to convert to organic farming, and perceived behavioral control is used to indicate farmers' perceptions of their ability to adopt organic farming.On the other hand, the concept of attitudes refers to the extent of favoring the conversion to organic.Accordingly, the more favorable the attitude and subjective norms towards organic farming conversion and the greater the perceived behavioral control, the stronger the intention to convert to organic farming.
According to Ajzen and Fishbein (2005: 53) "When people believe that others would expect them to perform the behavior or are themselves performing the behavior, the subjective norm will exert pressure to engage in the behavior".Therefore, a farmer's intention to convert to organic farming will be positively influenced if other neighbor farmers will convert as well, and if information and news releases from media is favorable to organic farming conversion (see Table A.1 in the Appendix A).Furthermore, perceived behavioral control is measured through control statements that relate to farmers' self-confidence in regard to carrying out organic farming, certifying, and controlling productivity, as well as being informed about the differences from conventional farming (see Table A.1 in the Appendix A).Thus, one can argue that having high self-confidence and knowledge in carrying out organic farming is positively associated with farmers' intentions to convert.Ajzen and Fishbein (2005) suggest that attitudes towards a particular behavior will affect behavioral beliefs and consequently the intention of carrying out that particular behavior.A number of empirical studies have used farmers' environmental attitudes as factors associated with the adoption of organic farming (namely Burton et al., 2003;Canavari et al., 2008;De Cock, 2005;Genius et al., 2006;Haris et al., 2018;Läpple, 2010;Läpple and Van Rensburg, 2011;Mzoughi, 2011;Rezvanfar et al., 2011).Consequently, environmental attitudes of farmers are operationalized with a set of reversed items (i.e., negative environmental attitudes, see Table A.1 in the Appendix A).Specifically, the more favorable (or unfavorable) the attitude towards environmental values of organic farming practices, the stronger (the weaker) the intention to convert.
Other factors are important too.Sarker et al. (2009) argue that farmers are assumed to hold spe-cific perceptions regarding certain effects of changing pattern of production and these perceptions can be potential determinants of the adoption decision.Namely, if the farmer perceives that organic farming leads to an improved supply of healthy foods and reduces the impact of agrochemicals on the environment, it is more likely that he converts to organic farming (ibid).Thus, this study assumes that farmer perceptions regarding EU policy opportunities (such as technologies, new market opportunities, decreased costs, etc.) and perceived risks from conventional farming will positively affect farmers' intentions to engage in organic farming (see Table A.1 in the Appendix A).
The indicator "perceptions regrading EU policy opportunities" is an innovative control factor added to the literature of EFA regarding organic farming adoption.The question is not used for assessing farmers' preferences (being hostile to pro-EU integration), but to explore their perception of opportunities (or costs) created from the integration into the single market.Depending on the level of information and type of ex- Source: Authors' elaborations based on Ajzen and Fishbein (2005).
periences with the EU market, farmers may be prone to perceiving the benefits, such as price markups.Therefore, we expect a positive relationship with the organic adoption decision.It might also be that the additional rules restricting export to the EU, for instance those related to safety and quality, lower farmers' perceptions of policy opportunities and create a negative association with the decision to convert to organic.However, considering that farmers are not direct exporters, we expect that they see opportunities rather than threats from the EU market.Also, it may be that the relationship is not significant since their ongoing experience with the market (the majority of these farmers are suppliers of organic MAPs for export-oriented value chain) is providing such opportunities currently.
The literature of adoption models suggests that farmers make decisions in order to maximize their expected profits (see Rogers, 1995;Parra López and Calatrava Requena, 2005).For instance, Reimer et al. (2012) shows that farmers, concerned most with profitability, were less likely to adopt organic farming.Having said that, it can be argued that the higher the farmers' orientation towards profit, the lower the likelihood to adopt organic farming; since converting would require a high amount of certification costs, information, time, and possibly present problems with production yields.
The adoption of organic farming systems is an information-intensive process; therefore, a factor related to information awareness would complement the study's conceptual framework (see for example from Mediterranean countries: Cukur, 2015;Läpple, 2010;Läpple and Van Rensburg, 2011).By information awareness the study implies the ability to capture, analyze, evaluate, and use information by receiving the highest level of benefit from it.One can argue that the higher the level of information awareness (i.e., information about legislation, rules, requirements, and standards of organic farming -both national and international) the more likely is that farmers will intend to convert to organic.
Based on the discussion made so far, the hypotheses that emerge are: H1 -The more favorable the attitude and subjective norms towards organic farming conversion and the greater the perceived behavioral control, the higher the farmers' likelihood to convert to organic farming.
H2 -Farmers' perceptions of risks from conventional farming and perception towards EU policy opportunities are positively associated with farmers' tendency to adopt organic farming.
H3 -Conversion to organic farming is positively affected by farmers' information awareness.
H4 -Environmental (non-preferable) attitudes and profit orientation are negatively associated with farmers' likelihood of adopting organic farming.
The complete conceptual framework applied with hypotheses directions are presented in Figure 1.

Questionnaire design
A structured farm questionnaire was designed to collect the primary data from conventional and organic farmers in different regions of Albania.The design of the questionnaire was based on the literature review as well as on the organization of three focus groups held with members of Albanian Organic Agriculture Association (AAOA) and other affiliated farmers in April and May 2019, in Malesi e Madhe (Shkoder), Lushnje and Durres, which represent also the main areas were activities of organic agriculture are carried out.The objective of the focus groups was to identify main challenges of the organic agriculture in Albania as well as to prioritize the main factors influencing the farmers' choice to convert.A SWOT analysis was carried with each focus group.The meetings helped the researchers to select the main factors and introduce in the questionnaire through the use of separate con-structs.In order to guide the discussions a list of factors was presented and discussed with the members in the final phase of the focus group.
The farm questionnaire was designed in four sections.The first section included questions relating to social and demographic characteristics of farmers.The second section included questions on the profile of the farm such as size and main activities.Meanwhile, the third part of the questionnaire included questions that generally relate to information on sustainable agricultural policies and practices and expectations for Albania's potential EU membership.The final section of the questionnaire consisted of questions on incomes that respondents benefit from agricultural activities in general and organic farming.

Data collection
Interviews were conducted in June and July 2019.Interviewers were trained to conduct interviews with farmers.
The area of the survey was selected based on the highest concentration of members of AOAA, namely Tirana, Durres, Malesia e Madhe (Shkoder), Lushnje and Vlora.The survey considered as a population the members of the Albanian Association of Organic Agriculture which have been in the past subject of capacity building activities in the sphere of organic agriculture.Currently, there are no more than 131 farms certified in Albania and a similar number in the process of certification.Considering these farms as well as the number of the members of the AAOA, the overall population is less than 1000 farms.The sample selection was made using a purposive sample, where those farmers who had previously part of trainings and capacity building events with AAOA and attempted to follow organic farming procedures were first asked, and following a snowball approach, other farmers were also identified.The representatives of the AAOA supported the surveying activity by contacting the members and carrying the interviews, under to supervision of the group of researchers.A list of members was used by There were no restrictions on the farming activities, therefore the sample includes a variety of farmers belonging to sectors such as olives, vineyards, orchards, vegetables, MAPs, honey, etc.
Data collected through the questionnaire were subject to controlling and revisions.The surveys were validated by tracing back 10% of the farms through phone calls.Data were entered into excel spreadsheets and processed through SPSS and R.

Sample characteristics
In total, 127 Albanian producers were included in the sample (farmers who are undergoing a certification process, certified (organic), and conventional).Table 2  of the socio-demographic profile of the surveyed farmers.Only 4% of the sample are women, while 1/3 of them have completed secondary education in the field of agriculture, 56% of them are (self) employed in the agriculture sector (agricultural activities being the main source of household incomes).A comparison with the main characteristics of farmers in the Census of Agriculture Holdings in Albania show that there is no evident sample bias.The figures are similar with the Census of 2012 which show that 88% of the farmers have a primary or secondary education, and that only 6% of farm heads are women (MARD, 2013).The number of farmers with University Education is slightly higher, probably due to challenging characteristics in terms of knowledge required by operators in the organic sector.Moreover, the average age of farms head in the Census is 45 years, which considering the time difference and the intergeneration transfer, represent a similar trend 1 .

Data analyses
In order to explore the determinants of the decision for farmers to adopt organic farming, the study makes use of compositions of variables which are constructed through Exploratory Factor Analysis (EFA) (see Table A.1 in the Appendix A).The main variables added are perceived behavior control, perceived EU policy opportunities, information awareness, environmental attitude, farmers perception of health risk from conventional farming, profit orientation and subjective norms influence.In order to develop measures for these variables, all the items representing them are entered in the EFA.As it is shown in Table 3, the EFA revealed 7 factors by using principal component applied to the items of the variables presented above.These 7 factors explain the structure of the data set accounting for 76.891% of the total variation.KMO test of sampling adequa-1 The Census of Agriculture Holdings in Albania carried in 2012 represented similar figures.The next Census of Agriculture Holdings in Albania will take place in 2023.Therefore, no comparisons can be made with more recent surveys of Censuses.cy (0.756) and the Barlett's test of sphericity (χ2=2258.271:df =300; sig<0.001)confirm the appropriateness of the factor analysis (refer to Field, 2009).While it is possible to extract as many factors as there are items, it was decided to extract only those factors that fulfilled the Kaiser's criterion for factor retention were extracted.Kaiser (1960) recommended retaining all factors with eigenvalues greater than one, and as s a result, all factor loadings mark values more than Stevens (2012) recommended value of 0.40, providing evidence of constructs convergent validity.
Next, logistic regression is employed to explore the relationships between the developed latent variables and farmers' decisions to engage in organic farming (the model also controls for socio-economic factors such as age, education, household size, and land ownership availability).This model was selected considering the dichotomous nature of the dependent variable (dummy), which is composed of farmers that are in process of certifying and certified as organic (1) versus conventional producers (0).Table 3 presents the operationalization of model variables and hypotheses.

Results
The results obtained from the logistic regression model are presented in Table 3. Results reveal that perceived behavioral control, information awareness, subjective norms (H1), and perceived risk from conventional farming (H2) positively affect the likelihood of converting to organic farming.On the other hand, perceived opportunities from EU policies (H2) and non-preferable attitudes towards environmental values of organic farming (H4) were negatively associated with the likelihood of converting to organic farming.Surprisingly, the variable of profit orientation was insignificant.
As the level of information awareness (on legislation/standards of organic production) increases, the likelihood of converting increases as well.Similar results (i.e.positive association with adoption of organic farming) are observed for the variables of perceived risk from conventional farming and subjective norms.
Furthermore, among control variables, farmers and farm characteristics are not significantly related to the likelihood of adopting organic farming, with the exception of land ownership availability and vocational education in agriculture.Thus, those farms having a household head with secondary education in agricultural sciences and with higher availability of land ownership, the likelihood of the farmer to convert to organic agriculture rises.

Concluding remarks and discussion
During the last decade, the organic agriculture sector has experienced rapid growth due to market-related factors, resource availability, and consumer preferences.For the case of Albania, similar to other developing countries, the organic sector is export-driven and mainly based on primary production.Under these conditions, the most important market in terms of value created is the production and export of wild MAPs.Thus, there is a need to identify and assess the role of various factors influencing the decision to adopt organic farming in Albania, considering its post-socialist transformation and challenges ahead in the context of the integration into EU market.
The results show that the probability of adopting organic farming is positively associated with farmers perceived behavioral control.Similar results were also obtained from other studies (see for instance Hattam, 2006;Läpple and Kelley, 2013;Sharifuddin et al., 2018;Yanakittkul and Aungvaravong, 2020).Perceived behavior control is based on the self-confidence and know-how of the farmer, as well as on the supportiveness of the surrounding environment.Additionally, subjective norms appear to be also positively associated with the probability of converting to organic farming.Similarly, this finding has also been proven by other studies such as Hall and Rhoades (2010), Hattam (2006), Läpple andKelley (2013), Sharifuddin et al. (2018), Yanakittkul and Aungvaravong (2020).Subjective norms are derived from information collected by media advertising and fellow farmers in the village.The influence by the community (i.e., neighbor farmers) or the media are important sources of relevant information and experience sharing for farmers to adopt organic farming.Same as recent studies of Sharifuddin et al. (2018) and Yanakittkul and Aungvaravong (2020), results show that farmers' family, community, group norms and influential people affect farmers decision to convert to organic farming.
Information awareness positively affects farmers' engagement in organic farming.Farmers that are informed about organic farming tend to engage more in this type of activity.This is similar to the findings of Genius et al. (2006), Hattam (2006) and Cukur (2015).Access to information sources might expand their knowledge on market indices and production technology as the number of agricultural information sources used is a significant factor for adoption of organic practices.
Interestingly, farmers that perceive risk from conventional farming are more likely to adopt organic farming.In addition, farmers who are more prone to attitudes of protecting the environment express a higher probability of adopting organic farming.Consequently, the higher the level of environmental concern the higher the probability to enter into the organic sector.This indicates that Albanian farmers perceive that organic farming is better for the environment.Similarly, studies such as Läpple (2010), Reimer et al. (2012) and Haris et al. (2018) found a positive association with regard to environmental attitude and organic farming conversion.
Surprisingly, perceived EU policy opportunities are negatively associated with the probability to adopt organic farming.It seems that they are more aware of the strict rules associated with access to this market compared to conventional farming.Moreover, by not being higher up in the export chain, farmers have perceived threats (strict rules on safety which increase their costs and losses) rather than benefits (the positive difference of price margins experienced in the EU market compared to the rest).This is supported by the fact that no significant relationship between adopting organic farming and being a profit-oriented farmer has been found.The lack of distinction within the domestic market puts conventional farmers in a vulnerable position when embracing organic farming and joining an expanding single market with more rules and opportunities, such as the EU single market.Results might also be influenced by the type of information and awareness provided in the past regarding EU policies.
These results are evidence for a needed intervention to improve information / awareness on EU opportunities.An information campaign in regard to EU legislation and market information would benefit policy makers and the sector operators.
There is no significant relationship between adopting organic farming and being a profit-oriented farmer.This is different from the findings of Curtis and Quarnstrom (2019) and Läpple ( 2010), but similar with Reimer et al. (2012).There are two major reasons which support this finding.First, in the foreign market (mainly in the Medicial and Aromatic Plants which compose more than 95% of the land covered with organic farming) the exporters do not benefit from high margins in the market rather than from more secure market space.Considering that the highest share of producers is related to MAPs production, the farmers find no clear profit pattern in the current market.In the MAPs sector there is no market mark-up, but rather market consistency.Second, the internal market does not provide clear indication to the consumers about the organic attributes due to weak marketing and weak consumers signals.Moreover, lack of quality standards in the market make the organic farm products not clearly distincted in terms of price.Thus, there are no positive market margins available for farmers in the market for organic products.This is similar to the findings of Zagata et al. (2020), which find that in post-socialist countries organic agriculture is not driven by activities of a social (organic) nature due to the fragile existence of the institutional environment (i.e.civil society, policy, and consumers).
Farmer and farm characteristics, except for land availability and secondary education in agricultural sciences, are not significantly related with the probability of adopting organic farming.As expected, the probability of being engaged in organic farming is closely linked with the farm head having pursued agricultural vocational school.Lastly, farms owning more land (or being wealthier) have a higher probability of adopting organic farming, same as other study cases from Mediterranean countries such as Greece (Genius et al., 2006;Alexopoulos et al., 2010) or other developed countries such as USA (Liu et al., 2019).Since the majority of organic farming is based on MAPs, the share of owned land increases the security of planting perennial crops and to fulfilling contracts in the future which is of high concern in Albania (Zhllima et al., 2010).This finding is actually not in line with other studies, which perceive limited land availability as a supporting factor for conventional farming.The result is acceptable since structural limitations and high fragmentation is a widespread phenomenon in Albania (Ciaian et al., 2018) make land availability a supporting factor for organic, which is not of concern in EU countries.
In conclusion, the study results are helpful evidence for the need to address current market failures in supporting the increased adoption of organic farming and for establishing an effective agricultural policy for supporting the conversion from conventional agriculture.The results implore an increase of information and awareness vis-a-vis the provision of a more intensive and CAP-oriented policy for promoting organic agriculture.As in the case of Crisostomo et al. (2012), organic farming policy in Albania is expected to be predominantly driven by EU policies, therefore compliance with EU agriculture policies should be a guiding principle for the operators, especially those oriented toward exports.A larger focus should be given to overall consumers' knowledge in order to increase knowhow and raise an overall level of civic participation in the national debate.In order to widen organic practices, policymakers might attempt certification in groups, considering the influence of factors related to information and subjective norms.The group approach also makes the diffusion of information and awareness easier.Capacity building activities such as coaching and training should be guided by a precautionary process of identifying farmers and groups of farmers with similar approaches to land resources.Here, the spatial neighborhood effect takes an important role, which supports the findings of Wollni and Andersson (2014).
In overall, access to services, information sources, governmental support policies and continuing trainings are among the key drivers for promoting the sustainable adoption of organic farming.Similar to Papadopoulos et al. (2018), attitudes to organic farming are as-sociated with economic reasons, but to a lesser extent when compared to conventional farming.Therefore, budgetary support policies should act as a guiding signal rather than filling the costs disadvantages compared to conventional farming.Therefore, premium criteria in national support schemes and IPARD programme measures should be introduced in order to incentivize farmers' orientation toward organic farming and enable the sector orientation toward the European Green Deal principles.

Figure 1 -
Figure 1 -Conceptual framework of the study.

Table 1 -
Key data of organic sector in selected SEE countries (2017).

Table 2 -
Socio-demographic characteristics of the sample.
Source: Authors' elaboration based on survey results.

Table 3 -
Definitions of variables included in the model, their operationalization, and hypotheses.