Decomposition of growth factors of production

To determine whether the lack of increase in the orange yield in Veracruz is because of the lack of adoption of innovations, due to technical obsolescence or to the increase in surface, the ^{Venezian and Gamble (1969)} formula was applied, modified by ^{Contreras (2000)} and ^{Zarazúa et al. (2009)}, with data from ^{SIACON-SAGARPA (2015)}, which indicates that:

where P _t : Total increase of production for the period of analysis; Y ₀(A _t -A ₀): Quantifies the contribution of the surface; A ₀(Y _t -Y ₀): Quantifies the contribution of the yield; (A _t A ₀)(Yt-Yo): Quantifies the combined effect of surface and yield; A _o : Area of orange harvested during the five-year-period of 1980-1984; A _t : Area of orange harvested during the five-year-period of 2010-2014; Y ₀: Average yield of orange in ton /ha during the five-year-period of 1980-1984; Yt: Average yield of orange in ton /ha during the five-year-period of 2010-2014.

Survey of initial and final baseline

The producers were surveyed directly in their productive units, before and after the interventions, with the following information being collected:

Location data. Information was collected with regard to the municipality, locality and telephone of the producer.

Producer’s profile. The variables of sex, years of experience in the activity, age and schooling were recorded.

Characteristics of the productive unit. Information was requested about the surface destined to citrus production that the interview respondents had.

Adoption of innovations. For the calculation of this indicator, the methodology proposed by ^{Muñoz et al. (2007)} was used, who suggest defining the innovating capacity of producers through the calculation of the Innovation Adoption Index (InAI). The InAI is obtained when quantifying the number of practices carried out by the producer in a specific moment compared to the number of total practices that must be carried out to reach efficiency. The catalog of innovations is elaborated by experts. Given that the InAI estimates the amount of innovations adopted compared to a possible total, its calculation can be expressed in a general percentage, by category of group of innovations, by producer and by each innovation included in the catalog.

The InAI is one of the most important variables in this study, since it allows understanding which innovations are being used by the producer, both before and after extension work and, based on this information, being able to quantify the impact of the intervention.

Application of the integral extension work methodology: Use of logbooks

Integral extension work was applied during 2012, according to the methodology described by ^{Almaguer and Ayala (2014)}, which basically consists in applying a diagnosis and, based on the needs of the apprentices, making recommendations, offering workshops and courses, and providing extension work with andragogic techniques. A logbook is used with the aim of collecting information that refers to the implementation of the recommendations of the technical adviser and the workshops, costs, incomes and practices performed every fortnight.

As has been explained, a Survey of Initial Baseline (Encuesta de Línea Base Inicial, ELBI) was applied before the intervention and one year after its operation, which was called Survey of Final Baselinne (Encuesta de Línea Base Final, ELBF), which had the same design and which were helpful to cover the objective of the research of identifying the impact of the intervention in the adoption of innovations.

Increase of the net income

To obtain the net income generated by the sale of orange to the producer who had technical assistance, sections in the design of the logbook were included to gather information related to the costs of activities recommended, which were part of the catalog of adoption of innovations. The total costs of the orange crop included the fixed and variable costs. The net income was obtained from the difference between the total income and the total production costs.

Factors related to the adoption of innovations

To determine the influence of the producer’s profile and his production units in the adoption of innovations, simple regression models were generated, using as dependent variable the increase in the innovation adoption index (InAI) of producers after technological intervention, and as independent variables the following were used: age, years of experience of the producer, and total income of the producer. The information obtained from the surveys was captured and analyzed with the Microsoft Office Excel software.

Decomposition of the production growth factors

Table 1 shows the calculation of the decomposition of orange production growth factors in Veracruz. When the production growth of a region or country is intensive, it means that it is based primarily in the use of technology or adoption of innovations, since the yields increase by surface unit. Instead, when the production growth is based on the increase of the surface, it means that technology is practically not being used, which implies technological obsolescence (^{Contreras, 2000}; ^{Zarazúa et al., 2009}).

Orange production in Veracruz is characterized by a reduced application of technological innovations, since the production increase is given in 74.66 % from the effect of the increase in surface, 13.01 % from a combination of the increase in surface and the use of some technology or innovation, and only 12.31% exclusively from the application of technology or innovations.

Innovation adoption index (InAI)

The average increase in the adoption of technologies for the communities studied was 30. 5%, starting from an InAI of 16.3 %. The analysis of the InAI of the ELBI compared to the ELBF gave the same results by locality:

^{Muñoz et al. (2007)} mention that adoption quantifies the result of the decision of producers to use or not to use a technology and, therefore, it is a measure of the innovating ability. In this regard, the locality that developed the highest innovating ability was Buena Vista Molino, with 68.6 % of adoption, because the group participants were interested in the theme; there were strong blood ties between group receptors (family members or trustworthy friends); high schooling (87.5 % are literate); the amount of surface that they have (7.33 ha); and the citrus income is among the highest of the zone ($5,77.70 per year per hectare).

Instead, the locality of Vara Alta obtained an InAI of 9.8 % in the increase of technology adoption. In this case, it was a large group that showed lack of interest in participating in integral extension work, as well as difficulty for self-organization. Toaco also had low adoption of innovations, attributable to its reduced experience in orange cultivation, advanced age and small surface.

When strawberry agro-businessmen from Michoacán were diagnosed in terms of their adoption of technological innovations (^{Zarazúa et al., 2011}), it was found that their initial InAI was 55.56 % (there was no extension work), which is considered very high in relation to other groups of producers since, for example, ^{Almaguer and Ayala (2014)} quantified the initial adoption of innovations for lime in Veracruz in 15 %. In another study about adoption of innovations in maize, small-scale producers had an InAI of 12.55 % (^{Zarazúa et al. 2012}). ^{Muñoz et al. (2007)} found that the maize network of Estado de México was quite disarticulated because it had 30 % loose nodes, an innovation adoption index (InAI) of 13.3 %, and a centralization index of 14 %, which is why they concluded that there were several structured actors and diffusers with low influence ability.

Such great differences are because strawberry is a crop that is very demanding of technology, since it is mainly for export. In the cases of maize and lime producers, the initial percentage of adoption was similar to the one obtained with orange, although the difference was the application of an efficient extension work methodology.

It should be mentioned that in Veracruz, Mexico, the initial yield in lime was 5.24 t ha^1 and the benefit/cost rate of 1.55, attributable to the reduced percentage of adoption of innovations, was 15 % (survey of initial baseline). After starting a process of integral extension work for two years, the percentage of adoption of innovations increased 45 %; the income, 45 %; and the benefit/cost rate, 68 %, in addition of forming a network for input purchase (^{Almaguer and Ayala, 2014}).

Factors that influence the adoption of technology

Results from the simple linear regression models showed that some aspects of the producer’s profile and of the characteristics of their production units influence the adoption of innovations.

In the Table 2, factors are presented that affected the adoption of technological innovations.

Age of producers. The average age of producers was 56.1 years. The locality of Adalberto Tejeda was the one that had the oldest producers, of 61.8 years, while Ampliación Reforma had the lowest average, of 50.8 years. Age can be a decisive factor in the adoption of innovations (^{Rogers, 2003}).

Figure 1 indicates that there is much variation in terms of the correlation between adoption of innovations and age, since there were young producers with a low InAI, compared to older producers who had a higher acceptance of the new practices. It can be observed that age (for this region) is not an important factor; the correlation was 15 %.

Figure 1 Behavior of the InAI regarding age in Álamo, Veracruz. 

In this regard, ^{Rogers (2003)} pointed out that in approximately 114 studies performed there was evidence that age influences the attitude that older adults show when acquiring new knowledge and practices, and taking on risks. In our case, age did not influence the adoption of innovations, possibly because of the availability of resources to invest, low human capital or because most of the producers are older and there are practically no young people.

Experience of the citrus producer. The results showed that orange producers from Veracruz have been devoted to citrus culture for 16.8 years in average. The locality of Lucio Blanco is the one that presents most experience in citrus production, with 23.5 years devoted to this activity compared to the community of Llano Grande with only 10 years.

The years of experience as citrus producer was not a key factor for the adoption of innovations, since there was great dispersion between both variables; in addition, the correlation found was, same as in age, of only 15 %; the producers need to understand and see results to evaluate the possible adoption of technology (Figure 2).

Figure 2 Relation of the InAI compared to years of experience as citrus producers in Álamo, Veracruz. 

^{Allub (2001)} mentions that the decisive force that directs the decision making process of peasant units is aversion to risk or uncertainty, and not the principle of maximization of utilities or experience. ^{Lipton (1968)} states that this is why attachment to “traditional techniques” are not irrational attitudes, but rather proven ways of minimizing uncertainty to avoid total loss and, as consequence, their disintegration as productive units (^{Cáceres, 1994}). During this whole process, economic agents (peasants) behave as people who minimize risk and not as stubborn players, despite the potential benefits that they could obtain if they were to choose them (^{Feder et al., 1985}).

Being able to test things (experience) gives it points in their favor to be able to make accurate decisions that benefit them; on the other hand, the more attached they are to their traditional practices, it is more difficult to instill new ways of doing things, unless it is proven to them that they work.

A factor that may have influenced the deficient correlation of experience and adoption of innovations was the low schooling of orange producers. Most (55.5) did not conclude primary school, while 24.2% of the producers did not study any school grade; the producers with complete or incomplete secondary school represent 16 %, and 3 % of producers have high school. Only 0.7 % of the producers interviewed have a technical degree.

Various studies point at variables that influence the adoption of technology; age, schooling, relation with change agents (^{Cuevas et al., 2013}), and when they are scarcely developed there cannot be a high innovation index (^{Feder et al., 1985}; ^{Reimers and Klasen, 2013}).

Income and adoption of innovations. The income that orange producers have in the region is variable. The regional average is $3478.20 per year; however, the inhabitants of Buena Vista Molino presented a higher income with $5777.00 annually, while Vara Alta is the community of lowest citrus income with $1461.5 annually (Figure 3).

Figure 3 Income of citrus producers in the region of Álamo, Veracruz. 

The citrus income was a significant factor for the adoption of innovations. The correlation for this case was statistically significant, of 0.36. Figure 4 shows that producers who had more income showed higher adoption of innovations. ^{Allub (2001)}, ^{Binswanger and Sillers (1983)}, ^{Feder and O’Mara (1982)} mention that economic resources determine the scale or level of adoption of a technology.

Figure 4 Behavior of the InAI with regard to the citrus income in Álamo, Veracruz. 

On the other hand, resistance to change in individuals in order to have a higher technological level will depend on the benefits that the change entails compared to the costs that it implies; that is, the individual will be motivated to innovate while the benefits exceed the costs of opportunity. Testing a new technology entails incurring in time, energy, money and land costs, which may be assumed only by those farmers that have sufficient economic resources for small trials, adjusting the scale upwards or downwards, in the direction toward non-adoption, to the extent where there is advancement in knowledge and trust about the results obtained.

Adoption of technology according to the surface destined for citrus production. The locality with a greater surface of orange is Adalberto Tejeda, with an average of 11.9 hectares per producer; there is a high income, although schooling is low. Tumbadero del Águila is the locality with lowest average surface (3.42 ha) destined to this crop; they present low schooling and low income. The average for Veracruz is 3.75 hectares (^{INEGI, 2009}).

The localities of Adalberto Tejeda, Vara Alta and Buena Vista Molino exceed the regional average (5.8 ha). At the state level the average surface is five hectares per producer.

Regarding the relation between adoption of technology and surface, the tendency is for the surface not to influence the adoption of technology, since the correlation is low (3 %); in this case, the surface is more related to the income of the citrus producer (48%) and not to the adoption of technology (Figure 5). It should be mentioned that, through a study carried out with 164 citrus producers in the northern zone of Veracruz, ^{COVECA (2002)} indicates that 32% of the orange producers are traditional, they use only family labor to carry out the farming tasks, and they have less than five hectares; 34 % corresponds to intermediate producers, who use family and paid workforce to carry out farming tasks and they own between 5 and 20 hectares, and the other 34 corresponds to entrepreneurial producers who use only paid workforce and have more than 20 hectares.

Figure 5 Behavior of the InAI with regard to the surface destined to citrus production in Álamo, Veracruz. 

Increase of the net income. Finally, a calculation of the net income of the producer was made. The costs of the practices that the producer performs are considered, such as soil fertilization once a year, unplanned sanitary pruning, weed control, and application of agrichemical products like insecticides or fungicides (Figure 6). The calculations carried out show that the net income increased by 19 % from 2012 to 2013.

Figure 6 Net income per hectare by orange producers, with survey data. The lines at the end of the bar represent the standard deviation.