Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander department, Colombia Received for publication: 04 October, 2016. Accepted for publication: 20 July, 2017 Doi: 10.15446/agron.colomb.v35n2.63972 1 Faculty of Agricultural Sciences, Universidad Nacional de Colombia. Bogota (Colombia). 2 Department of Rural Development, Faculty of Agricultural Sciences, Universidad Nacional de Colombia. Bogota (Colombia). yvarandac@unal.edu.co 3 Department of Agronomy, Faculty of Agricultural Sciences, Universidad Nacional de Colombia. Bogota (Colombia). Agronomía Colombiana 35(2), 247-255, 2017 Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander department, Colombia Tipificación de un sistema de producción de tabaco (Nicotiana tabacum) en la provincia de Guanentá, departamento de Santander, Colombia Jennifer Jiménez C.1, Yesid Aranda C.2, and Enrique Darghan3 ABSTRACT RESUMEN In the province of Guanenta (Santander), lack of technical as- sistance received by producers is one of the main obstacles to the production of snuff. The objective of this research was to characterize the snuff production systems in the province, con- sidering technical, socioeconomic and environmental variables. The sample was determined from census by producers supplied by Coltabaco, using stratified random sampling; the data were processed by multiple correspondence analysis. The results are shown as discriminant variables: type of snuff, the percentage of sharecropping, type of leaf curing, implementation of irriga- tion system, development of alternative crops, crops rotation and membership of producers to associations. Four types of production systems were found: i) conventional - snuff type burley, ii) technified - snuff type burley, iii) conventional alter- native - snuff type burley, iv) conventional - snuff type black. Finally and aiming improve the production of snuff, this study propose improvement actions for technical assistance programs received by producers, alternatives for access improvement and availability of water. En la provincia Guanentá (Santander), la asistencia técnica productiva es uno de los principales obstáculos para la produc- ción de tabaco. El objetivo de esta investigación fue caracterizar los sistemas de producción de tabaco en la provincia, se indagó acerca de variables técnicas, socioeconómicas y ambientales. La muestra se determinó a partir del censo de los productores suministrados por Coltabaco, por medio de muestreo alea- torio estratificado; los datos se procesaron mediante análisis de correspondencias múltiples. Los resultados muestran como variables discriminantes: tipo de tabaco, el porcentaje de apar- cería, tipo de curación, implementación de sistema de riego, desarrollo de cultivos alternativos y la rotación, y la pertenencia de productores a asociaciones gremiales. Se encontraron cuatro tipos de sistemas de producción: i) convencional - Tipo de tabaco burley, ii) tecnificado - Tipo de tabaco burley, iii) alter- nativa convencional - Tipo de tabaco burley, iv) convencional - Tipo de tabaco negro. Para mejorar la producción de tabaco se proponen acciones de mejora a los programas de asistencia técnica que reciben los productores, y mejorar el acceso y la disponibilidad de agua. Key words: technification, sharecropping, contract agriculture, multivariate analysis. Palabras clave: tecnificación, aparcería, agricultura por contrato, análisis multivariado. USA (3%) (FAOSTAT, 2016). The European Union stands as the largest importer of crude snuff, followed by Russia, the United States and Japan (Traub, 2009). According to official statistics, for 2014 in Colombia the blond snuff harvested area reached 6.336 ha with a produc- tion of 11.757 t, registering an increase of 56% and 50.2% in planted area and crop production respectively, compared to 2013 in which were harvested 4,061 ha with a produc- tion of 7,825 t (Agronet, 2016). Despite the increase, area and production of snuff in Colombia does not reach the levels presented for this product in 2005 when 10,892 ha produced 20,925 t of tobacco. Introduction The production of snuff (Nicotiana tabacum) stands out worldwide, because the profitability of snuff production turns out to be high, compared to other agricultural prod- ucts (Peña and Rivera, 2011). Commercially, prices are stable to the producer, and depend mainly on the quality of snuff produced, the tobacco types are Virginia, Black, Burley and Oriental, the last two are the most marketed as this types are employed for cigarettes production (Castel- lanos and Montanez, 2006;Castellanos et al., 2009). By 2013 the country with the highest production of snuff was China (29%), followed by India (9%), Brazil (8%) and http://http://dx.doi.org/10.15446/agron.colomb.v35n2.63972 mailto:yvarandac%40unal.edu.co?subject= 248 Agron. Colomb. 35(2) 2017 Between the 2005 - 2015 period the decline in harvested area is explained by the problems of performance and quality of the regions where domestic production tobacco is located, Nowadays farmers are replacing snuff produc- tion areas by other crops. The tobacco production in the country is concentrated in the departments of Santander, Norte de Santander, Huila, Guajira, Boyacá, Cauca, Valle, Tolima, Cesar and Quindio (Castellanos et al., 2009). In recent years the snuff production in Colombia has reached 1.3% in the agricultural GDP and 0.53% of manu- facturing GDP. In the first quarter of 2016 the performance of the national agricultural industry shows an increase of 0.7% over the same period of 2015, particularly unpro- cessed snuff increased its production by 40.7% (DANE, 2016). The national tobacco industry has been highlighted in the national economy by creating jobs (CNCPT, 2010), it is estimated that the sector generated 17,000 jobs in the production phase and 8,000 jobs in the industrial phase; it is also an important source for income generation for the departments through taxes and foreign exchange, generated by exports of raw and processed snuff products (Agrocadenas, 2005). Regarding the sector competitiveness, the trade balance for 2014 reports that imports of snuff and its derivative products amounted USD $66’340 million, while exports reached USD $54’155 million, showing a negative balance of 19%; usual trend since 2012 (SIEX, 2016). This ref lects a loss of competitiveness of the tobacco industry regarding its performance in the past decade, since the 2002-2009 period showed a positive balance for the country. Santander department is located in the Andean region of Colombia, Locally the snuff production is considered one of the most important economic lines in agriculture (Mojica and Paredes, 2005); its production is concentrated in 35 municipalities. By 2014 about 5,400 ha were cultivated, rep- resenting 56% of the national area, reaching a production amount of about 7,555 t of snuff leaf (Fig. 1), with average yields of 1.68 t ha-1 (MADR, 2015) and an average acreage per farmer of 1.7 ha. In Santander farmers cultivate two types of snuff mainly: blond type tobacco and black type, performing 56.7 and 43.2%, respectively. Farmers produce mainly burley snuff in the first half of the year, because the high environmental availability of water and higher purchase prices versus the black type snuff, but black typo snuff is considered complementary, because the tobacco production is a tradition in some municipalities of the department. Among the benefits some of the major difficul- ties facing the production of snuff in the department are the training for technical improvement of the system and high production costs (Gobernación de Santander, 2010). In the province of Guanenta (municipalities of San Gil, Aratoca, Barichara, Jordan, Curiti and Villanueva), the pro- duction of snuff is mainly related to rural economy systems with an average of 2.5 ha per production unit (Castellanos et al., 2009), production is related to smallholdings, intensive family labor and sharecropping arrangements, where the owner offers the land and, as return, receives a percentage of gross production (Quintero, 2001). In the province the major constraints facing the producer of snuff are related, among others, to the low adoption of technologies by some farmers, land not suitable for cultivation, limited access to irrigation water, low use of inputs. This condition itself generate low yields and low product quality (Mojica and Paredes, 2005). The production of snuff is performed by contracted farming systems, with the presence of two com- panies i) Coltabaco and ii) British American Tobacco; H ar ve st ed a re a (1 ,0 00 h a) P ro du ct io n (1 ,0 00 t) 0 5 10 15 20 25 30 0 5 10 15 20 45 40 35 30 25 50 20 13 20 12 20 11 20 10 20 09 20 08 20 07 20 06 20 05 20 04 20 03 20 02 20 01 20 00 20 14 19 99 19 98 19 97 19 96 19 95 19 94 Production of Santander National production National harvested area Santander harvested area FIGURE 1. Harvested area and tobacco production, national and participation of the department of Santander. Source: The authors, based on Agronet (2016). 249Jiménez C., Aranda C., Darghan: Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander, Colombia who established agreements of purchase depending specific qualities and quantities, with agreed prices; in turn, these companies provide services back to the farmer as the provi- sion of agricultural inputs and technical assistance (Eaton and Shepherd, 2002). The objective of this research was to typify the snuff production system in the Province Guanenta, Santander, to generate information and propose actions aimed to improving the competitiveness of this production system located in the national territory. Materials and methods The Guanenta province is located south east of the depart- ment of Santander (Fig. 1) in the Eastern Cordillera of Andeans (Colombia). Total length is approximately 3,800 km2 and its formed by 18 municipalities, among which Vil- lanueva, San Gil, Curiti, Barichara, Aratoca and Jordan are municipalities with more snuff cultivated area. According to producer’s census by Coltabaco, in the province are 985 farmers that signed purchase agreement of snuff blond production and / or snuff black with this company in 2015. the census was made following an stratified proportional random sampling (Schaeffer et al., 2006), taking as strata the municipalities where the company had purchase agree- ments, The sample size was determined by formulates n = (Np(1-p)) / ([(N-1) (ε / z )2] + p(1-p)), where N is the number of producers who signed purchase agreement with the company in 2015 (985 Producers); z (score of the normal distribution) is the level of confidence (95%); ε is the sam- pling error (4.7%); p is the proportion of producers located in the province and with signed purchase agreements with the company (0.9) and q=1-p. The sample size estimated was 92 Producers (n). The methodology to estimate the proportional weight of each stratum (municipality) ni was obtained with wi = Ni / N, where wi corresponds to the weight of stratum i; Ni correspond to the number of snuff producers located in stratum i (municipality) and N is the number of producers who signed purchase agreement in each municipality 2015. The numbers of surveys to imple- ment each layer are in table 1. It was built an analytical model with 98 variables related to four criteria (general, technical, socioeconomic and en- vironmental), a questionnaire was developed, it validated with 12 producers. Later, between February and April 2016 the questionnaire survey to the producers according strati- fied sampling was applied. The variables were predominantly quantitative and con- tinuous; some of them are discrete, it was necessary to reclassify using the optimal method of the accumulated frequency root (Schaeffer et al., 2006). 24 variables were constants and 14 variables complementary. Multiple cor- respondence analysis (MCA) was applied incorporating only the discriminant variables for the population (60 vari- ables), distributed as follows: i) General: related to the site where production takes place, possession of it and informa- tion of respondents (3 variables); ii) Technical: related to the production process of snuff, other crops and livestock production (39 variables); iii) Socioeconomic: related to the financing of production, the products destination, the amount and type of labor required for production, quality of life, and the farmer household (15 variables); and iv) Environmental: related to climatic factors affecting pro- duction, the use of environmental resources, availability and access to water and climate change (3 variables). Multiple correspondence analysis (MCA) was used as technique to detect the common information between variables by new variables, which are called factors (Jol- liffe, 1986). A matrix of individuals containing qualita- tive or nominal variables was achieved to categorize the population, synthesizing information and obtaining individuals similar in function on the association between categories of respondent’s variables. The variables formed a Burt matrix containing all possible crosses between the surveyed variables (Greenacre, 2007), and through Chi- square statistical was established interdependence by the cumulative frequency, which explains the percentage of variability in dimensions. The multivariate analysis technique allowed establish, in first place, the relationship between the variables of each criterion with an individual way (general, technical, socioeconomic and environmental); and secondly, the relationship between all the criteria and variables ac- cording to the proposed analytical model, determining the variables used to establish the production systems of snuff in the territory. TABLE 1. Sample size with proportional affixing. Stratum Municipality 1 San Gil 85 0.09 8 2 Jordan 22 0.02 2 3 Villanueva 372 0.38 35 4 Curiti 272 0.27 26 5 Barichara 234 0.24 22 Total 985 1 92 250 Agron. Colomb. 35(2) 2017 Data processing was performed using free software (R studio), obtaining plots in two dimensions that explain the greater variability of the components, which were used for the discussion of results. Results and discussion General characteristics of the tobacco production system in the province of Guanenta Variables that became constant (with variance equal to zero) when collecting data were: crop association, imple- mentation of seedbed, weed control, leaves classification, record of activities, technical assistance, production fi- nancing, washing water, kitchen area, working baths, and electric power. The production system in the territory is characterized by: i) Technical component: tobacco producers do not practice associate farming models in the tobacco crop, they rely on technical qualified assistance, use seedbed in trays, man- age weeds control, classify leaves during packing and do not make any record of activities or costs of production. ii) Socio-economic component: contracts of purchase and sale are the financing source of tobacco production, the producer establishes the number of plants or the area to sowing, the expected production, the financial credit, the supervision of the crop and technical consulting and prices of the leaf are related to the tobacco classification of every company in the zone. There are two leading industries in the country, Coltabaco and British American Tobacco, which produce primary tobacco, establishing contract farming agreements with producers and developing the activity of agribusiness of cigarettes (Castellanos, 2009). The product is mainly for sale, family members work in the farm, and eventually wages are paid. All properties surveyed have kitchen area, functional bathrooms and electric power. Analysis by criteria General criterion According to general information variables, the variability of this component is explained by four dimensions with 65.9% (Fig. 2); in order of contribution: municipalities (24.4%) standing out Villanueva and Jordán; Land tenure (15.4%) for the development of productive activity, where both owners and sharecroppers are those who develop tobacco production (Agrocadenas, 2005). The third di- mension is explained again by the variable municipality, in particular San Gil which contributes with 21.6%. It is inferred that in Villanueva municipality there is a major trend to find lands with areas upper to 5 ha with predomi- nance of proprietary farmers of the land. In contrast with the municipalities of San Gil and Curití where predominate systems under sharecropping in lands with areas of less than 5 ha, where land owners prefer to distribute small areas to diverse sharecroppers in order to minimize the risk and to improve the conditions related to the develop- ment and benefit every sharecropper can add to the crop, improving its profitability. Technical criterion The technical component explains 64.2% of variability across 9 dimensions (Fig. 3). In the first dimension stands out for its importance, rotating crops (4.8%) and planting (3.3%); in the territory prevail small farm areas in order to take advantage of the productive land and consequently developing small areas of tobacco production. Nowadays, it is observed the phenomenon of parcels of farms mainly in the municipality of Barichara, generating the arrival of new owners, who acquire the land of build condomi- niums or rest homes. The second dimension explains the 9.4% of the variability and highlights the implementation of irrigation systems (5.2%), characteristic of technical productive systems that improve performance. The third dimension is explained by the type of curing (7%) due to the fact that some farmers have initiated the implementation of the curing system (hanging all the tobacco plant in a driing facility known as “caney”) that is more efficient in terms of the labor requirement, reducing it up to 30%, which improves the quality of the product. Currently, it is necessary to train the producers about benefits generated by this practice, since they perform their work inadequately lowering their massive implementation. FIGURE 2. Correspondence map for the General criterion. La nd te nu re Municipality -1.0 0.5 0 -0.5 1.0 -0.5 0 0.5 1.0 251Jiménez C., Aranda C., Darghan: Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander, Colombia The territory is characterized by low-tech production sys- tems; however, in the third quadrant are located productive systems that have implemented technical improvements for agricultural production. Social and economic criterion According to the analysis of socio-economic data; 60.2% of the variability is explained by four dimensions. In the first dimension, 18.7% of the variability correspond to the wage earner (13.9%). In the territory, it is not common to find this situation, due to the fact that production is related to family farming systems; In there, activities out of the farm are scarce, because labor force is used almost exclusively for own production. The second dimension is explained by acordds with loans entities (5.3%) because of the high presence of credit and financial cooperatives offering these services, It is important to add, the department of Santander is renown for having an honest economic and social management (Davila, 2004). The third dimension is explained by the size of the family nucleus (9.2%), which is fundamental for family farming systems where the develo- pment of agricultural activity derives from family members work force whose assume all farming tasks. In the correspondence chart (Fig. 4), the sample tends to be very homogeneous, in terms of the mentioned criteria; however, there is a close relationship between the people who exercise an activity as employees and sharecroppers who pay the sixth part to the owner of the land. However, the growth of San Gil in last years has created a greater demand of labor in the urban zone, which can be generating the connection to non-agricultural activities. Environmental criterion According to the environmental criteria, 68.9% of the variability is explained by three dimensions; the first dimension (Fig. 5) explains 38.1% of the variance and incorporates the difficulty for water use. The category does not present difficulty for water using (45.3%) and the variable environmental phenomenon in the category does not present environmental phenomena with the capacity to affect production (40.4%). Those are atypical data, due to the drought phenomenon faced in the second half of 2015. The second dimension accounts for 17.8% of the variability and highlights the variables of envi- ronmental phenomenon - hail (35.1%) and water source in its broken category (27.4%). It is important to rescue the environmental phenomenon of hail, which affects the municipalities of San Gil and Curití, and generates pro- duction loss and affects the quality of the leaves (Mojica and Paredes, 2005), becoming a social problem, currently harvest insurance ensures performance yet not quality of product, which is punished in the purchase price to the producer. The third dimension explains 13% of the variability with the variable difficulty for water use and its relation to drought. In the environmental component, the climatic seasonality of the province stands out, with little rainfall during the year; there are prolonged water deficits, mainly in the phase Ir ri ga tio n sy st em s Rotating crops and planting -0.8 0.4 0 -0.4 0.8 -0.4-0.8 0 0.4 0.8 Lo an s Wage earner activities -0.8 -1.2 -1.6 0.4 0 -0.4 0.8 -0.4-0.8-1.2-1.6 0 0.4 0.8 FIGURE 4. Correspondence map for the Socio-economic criterion. FIGURE 3. Correspondence map for the Technical criterion. W at er s ou rc e Environmental phenomenon 0.1 -0.6 -1.3 0.8 1.5 -0.5 0 0.5 1.0 FIGURE 5. Correspondence map for the Environmental criterion. 252 Agron. Colomb. 35(2) 2017 of rapid growth of the vegetative period, which significantly affect production and quality (Ledesma, 2012). Typification of tobacco production systems in the province of Guanenta From the main component analysis incorporating all va- riables of the model, nine dimensions are identified that explain 60.6% of the variability (Tab. 2). The first three dimensions explain 31.57% of the variance. The first dimension (12.8%) explains for the category of activity as employee (2.6%), followed by the sharecropping in which the owner obtains the sixth part of the produc- tion (2.6%), and in the third order the production of black tobacco (2.3%). The second dimension with 10.6%, group together the type of irrigation by drip (3%), additional crop-tomato (2.8%); the declaration of farmeŕ s member- ship into the National Federation of Producers of Tobacco (Fedetabaco) (2.3%), and the structure of storage (2.2%). In the third dimension (8.2%) emphasize the type of cur- ing- hatchet (8.6%) and no external sales (2.2%). The first dimension is formed by socio-economic criteria emphasizing the payment of the sixth part to the owner and executing activities as employee. Also the variable type of tobacco in its black category becomes relevant due to the change of production towards blond tobacco (Burley) as consequence of a greater demand of blond cigarettes at tobacco companies (Mojica and Paredes, 2005). However, black tobacco production in the province is located in areas with higher temperatures, mainly in the low-lying areas of the municipalities of Barichara and Jordán, while in other municipalities of the province it is produced as comple- mentary product to Burley type tobacco as a rotation crop during the second semester (Mojica and Paredes, 2005). The second dimension is formed by the technical variables, where the use of technology on crops becomes relevant. The productive system has a drip irrigation system, stor- age structures and area is used for other crops; Showing a greater degree of use of technology that allows the farmer to improve performance. At the socio-economic level the recognition of being part of associations or guilds, allows to identify farmers who are more immersed in the dynamics of the sector and committed to improve production, they belong to producers associations which are affiliated to Fedetabaco (Castellanos, 2009). In the third dimension are relevant technical criteria, es- pecially the variables of curing-hatchet and additional sales of tobacco, which are dependent on each other. Performing the cutting of the whole plant requires eliminating f lower buds and applying inhibitors, and it is necessary to adapt or construct structures different from the traditional ones for the drying of tobacco leaves. Nowadays, this practice has a little use in the territory, mainly because of the practice of the producers to carry out traditional curing (leaf-to-leaf healing). They believe that high-quality butchering curing tobaccos is less heavy thus the production is lower, and also by the large percentage of sharecropping producers that are limited to construct this type of structures for profit. Types of productive systems According to the correspondence map (Fig. 6), four types of productive systems of tobacco in the province of Guanenta are identified. Type 1. Conventional production system - Burley type tobacco This type of production is used by about 67% of the provin- ce producers . These are conventional production systems that technically are producers of Burley-type tobacco, who- se main rotation crop is bush bean. They make additional sales of tobacco leaf to external buyers, so the predominant type of curing is leaf-to-leaf type. Thirty seven percent TABLE 2. Variability explained by dimensions for each evaluated criterion. Dimension Value singular Inertia Chi-Square Percentage Accumulated percentage Histogram 1 0.1422 0.0202 6467.6744 12.7989 12.7989 **** 2 0.1293 0.0167 5343.0991 10.5735 23.3724 **** 3 0.1139 0.0130 4144.3055 8.2012 31.5736 *** 4 0.0992 0.0099 3148.9104 6.2314 37.8050 ** 5 0.0948 0.0090 2873.0020 5.6854 43.4904 ** 6 0.0895 0.0080 2558.6953 5.0634 48.5539 ** 7 0.0843 0.0071 2271.3055 4.4947 53.0486 ** 8 0.0781 0.0061 1949.2689 3.8574 56.9060 ** 9 0.0763 0.0058 1860.5165 3.6818 60.5878 ** 10 0.0736 0.0054 1731.4459 3.4264 64.0141 ** 253Jiménez C., Aranda C., Darghan: Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander, Colombia (37%) of the producers are livestock owners (mainly cattle), nevertheless, tobacco continues being the main production activity. At the socio-economical level, these are produc- tive systems that in the case of being sharecroppers, the percentage share is 20% of payment to the owner; they do not recognize the membership to Fedetabaco, and are mainly engaged in the development of a productive system in their properties; for this reason those are considered typical family farming systems (Forero et al., 2013). At the environmental level, the main phenomenon that affects production is drought, so that despite having reservoirs for water in summer seasons there is no availability to be used in production. Type 2. Technified production system - Burley type tobacco This type of production system is formed by technical production systems, used by about 16% of producers in the province. At technical level, implement irrigation systems, mainly by dripping; have warehouses for storage of agricul- tural supplies; and additional to the production of tobacco they farm other crops as: tomato and coffee. In these farms, crops are rotated mainly with beans and corn. Land -ow- ning farmers have long- or perennial - cycle crops; however, tobacco cultivation is important for the productive system. At the socioeconomic level, there are no significant diffe- rences with other producers in the province, besides they recognize their affiliation to Fedetabaco, as a sample of the embeddedness or agreements (Sonnino, 2007; Renting et al., 2008), that in particular, producers in this group have with the tobacco industry in the province. Type 3. Conventional production system - Black type tobacco They are conventional production systems, which are used by about 10% of the producers of the province. The- se are characterized by the production of black tobacco, located in the municipality of Jordán and the lower area TABLE 3. Variables contribution to dimensions. Variable Contribution to dimensions 1 2 3 Activity as employee (AC _ AS) 0.026 0.013 0.022 Shareholding percentage- sixth part (AP) 0.026 0.006 0.003 Type of tobacco - black (T _ T) 0.023 0.001 0.003 Type of irrigation-drip (T _ R) 0.004 0.03 0 Alternate crop tomato(O _ C) 0 0.028 0 Storage structure (ALM) 0.016 0.022 0 Farmer membership (AS _ AG) 0.023 0.023 0.001 Healing and hatchet Type (T _ C) 0.004 0.007 0.086 No external sales (BR) 0 0.004 0.022 Te ch ni ca l c om po ne nt ( te ch ni fic at io n) Socio-economic component -0.2 -0.6 -1.0 0.6 0.2 1.0 -0.3-0.6 0 1.20.90.60.3 1.5 TYPE 1 TYPE 3 TYPE 2 TYPE 4 FIGURE 6. Map of multiple correspondences, types of tobacco produc- tion systems in the Guanenta-Santander province. of Barichara where high temperatures and lower rainfall are predominant. For the development of this system it is necessary to carry out a greater number of collections and consequently it is necessary to link more labor throughout the productive cycle. Sowings are made in the second half of the year. At the socio-economic level are farmers with large families (on average 8 members per family) linked to the production in the farms. Type 4. Conventional production system - Burley type tobacco, alternate It is formed by conventional tobacco production systems in which about 5.6% of the province’s producers partici- pate; are basically differentiated by the socio-economical component, with farmers located in the municipality of San Gil, mainly sharecroppers, whose percentage is lower than that of other farmers (payment of the sixth part). Producers combine agricultural production on their farms with activity as wage earners on other farms; unlike other production systems in this municipality, it is declared that production is affected by climate change, specifically by the phenomenon of hail. Conclusions The tobacco production system in Guanenta province has a basic technical level. It has qualified technical as- sistance for the cultivation of tobacco, does not carry out association of cultivation of tobacco with another crop, makes classification of the leaf in the packing phase and do not record activities or production costs. At the socio- economical level, its main source of financing for tobacco production is tobacco purchase and sale contracts with tobacco companies present in the area, the main purpose of most of the production is aimed at sale, labor is mainly family, occasionally making external contracts hiring for 254 Agron. Colomb. 35(2) 2017 labors, and have kitchen area, working bathrooms, electric power and water for washing. The main relevant variables to characterize the tobacco production system are: tobacco t y pe, percentage of sharecropping, activity as wage earner, type of curing, alternative crops, rotation and association of the farmer with guilds. The methodology used, allowed us to establish four types of production systems: i) Conventional production system - Burley type tobacco, ii) Technified production system - Burley type tobacco, iii) Conventional production system - Black type tobacco, iv) Conventional production system - Burley type tobacco, alternate. Currently an issue is the pressure exerted by transforma- tion of the rural envirnment given the proximities with urban area, related by the multifunctionality of agricultural activities, which generate the displacement of agricultural activity and the decrease of labor for the agricultural sector due to the requirement of labor in other economic sectors of the region. The lack of support from the governmental and municipal entities is recognized as an important problem at social level in the region, in essence, the process of coaching and technical assistance to the different crops in the province are developed by private companies. An outreach process is required with a greater participation of public entities in order to support the efforts of private companies and thus seek to solve production problems in the province of Guanenta. Also, it is necessary the commitment of these entities to establish technical committees that tackle the real needs of the rural population. Outreach processes in the Guanenta province should be focused on improving tobacco production conditions in the area, based on programs designed to improve access to water and implement efficient production techniques, improving not only performance but also the quality of tobacco. Future research that could be developed to contribute to the improvement of the competitiveness of the system of snuff in Guanenta province corresponds to characterize with greater level of detail the groups included in this research, specially the analysis of production systems with being sharecroppers in the territory, to evaluate the effects by the change of land uses, multifunctionality of agricultural activities and the different logics of production and the profits for the different agents involved in the production system of snuff Literature cited Agrocadenas. 2005. La cadena del tabaco en Colombia. IICA, Bo- gota, Colombia. Barrientos, J.C., G. Plaza, and J. Rojas. 2012. Comparative analysis of Flue-cured tobacco production costs in Santander and Huila (Colombia). Agron. Colomb. 30(2), 289-296. Castellanos, O., L. Torres and D. Rodríguez. 2009. Desarrollo tecnológico e innovación de la cadena productiva del tabaco. Universidad Nacional de Colombia, Bogota, Colombia. CNCPT - Consejo Nacional de la Cadena Productiva del Tabaco. 2010. Acuerdo de competitividad cadena productiva del tabaco. Consejo Nacional de la Cadena Productiva del Tabaco. Retrieved from: http://sioc.minagricultura.gov.co/templates/ sioc_cadenas/docs/1109.pdf; consulted: October, 2016. Dávila, R. 2004. Innovación y éxito en la gerencia cooperativa, casos exitosos de cooperativas rurales de ahorro y crédito. Pontificia Universidad Javeriana, Bogota, Colombia. DANE - Departamento Administrativo Nacional de Estadística. 2016. Boletín técnico: Cuentas Trimestrales – Colombia Pro- ducto Interno Bruto (PIB) Primer Trimestre de 2016. Retrieved from: https://www.dane.gov.co/files/investigaciones/boletines/ pib/bol_PIB_Itrim16_oferta_demanda.pdf; consulted: Octo- ber, 2016. Eaton, C. and A. Shepherd. 2002. Agricultura por contrato alianzas para el crecimiento. FAO. Retrieved from: http://www.fao.org/ docrep/004/y0937s/y0937s00.htm#toc; consulted: October, 2016. FAO. 2016. Estadística de producción de tabaco FAOSTAT. Retrieved from: http://faostat3.fao.org/download/Q/*/E; consulted: October, 2016. Forero, J., L.J. Garay, F. Barberi, C. Ramírez, D.M. Suárez, and R. Gómez. 2013. La eficiencia económica de los grandes, medianos, y pequeños productores agrícolas colombianos. pp. 69-111. In: Garay, L.J. (ed.). Ref lexiones sobre la ruralidad y el territorio en Colombia. Problemáticas y retos actuales. OXFAM, Bogota, Colombia. Gobernación de Santander. 2010. Plan prospectivo provincia de Gua- nenta. Gobernación de Santander and Fundacion Universitaria de San Gil - UNISANGIL, San Gil, Colombia. Greenacre, M. 2007. Correspondence analysis in practice.Cha- pman and Hall and CRC Press, Boca Raton, FL, USA. Doi: 10.1201/9781420011234 Grupo de Investigación y Desarrollo BioGestión. 2006. Estudio prospectivo en la cadena productiva del tabaco colombiana. Universidad Nacional de Colombia, Bogota, Colombia. Jolliffe, IT. 1986. Principal component analysis. Springer Series in Statistics, Springer-Verlag, New York, USA. MADR - Ministerio de Agricultura y Desarrollo Rural. 2015. Bullets cadena del tabaco- marzo-2015. Ministerio de Agricultura y Desarrollo Rural. Retrieved from: http://sioc.minagricultura. gov.co/index.php/opc-documentoscadena?ide=19; consulted: January, 2016. Ministerio de Hacienda y Crédito Público. 2015b. Resultados PIB cuarto trimestre 2014. Ministerio de Hacienda y Crédito http://sioc.minagricultura.gov.co/templates/sioc_cadenas/docs/1109.pdf http://sioc.minagricultura.gov.co/templates/sioc_cadenas/docs/1109.pdf https://www.dane.gov.co/files/investigaciones/boletines/pib/bol_PIB_Itrim16_oferta_demanda.pdf https://www.dane.gov.co/files/investigaciones/boletines/pib/bol_PIB_Itrim16_oferta_demanda.pdf %20http://www.fao.org/docrep/004/y0937s/y0937s00.htm%23toc %20http://www.fao.org/docrep/004/y0937s/y0937s00.htm%23toc http://faostat3.fao.org/download/Q/%2A/E https://doi.org/10.1201/9781420011234 http://sioc.minagricultura.gov.co/index.php/opc-documentoscadena%3Fide%3D19 http://sioc.minagricultura.gov.co/index.php/opc-documentoscadena%3Fide%3D19 255Jiménez C., Aranda C., Darghan: Typification of the production system of snuff (Nicotiana tabacum) in the Guanenta province, Santander, Colombia Público. Retrieved f rom: ht tp://w w w.minhacienda.gov. co/porta l/page/porta l/HomeMinhacienda/sa ladeprensa/ Presentaciones/2015/2015-03-17-PIB-IV-2014-oferta-3.1.pdf; consulted: January, 2016. Mojica, A. and J. Paredes. 2005. Características del cultivo de tabaco en Santander. Centro Regional de Estudios Económicos de Bucaramanga. Retrieved from: http://www.banrep.gov.co/ docu mentos/publicaciones/reg iona l/ESER / buca ra ma n- ga/2005_diciembre.pdf; consulted: January, 2016. Peña, L. and H. Rivera. 2011. Análisis estratégico del sector taba- calero en Colombia 2005-2010. Universidad del Rosario, Bo- gota, Colombia. Quintero L. 2004. Costos de producción de tabaco en Colombia. Documento de trabajo No 43. Observatorio de Agrocadenas IICA, Ministerio de Agricultura y Desarrollo Rural, Bogota, Colombia. Renting, H., H. Oostindie, C. Laurent, G. Brunori, D. Barjolle, A. Jervell, and M. Heinonen. 2008. Multifunctionality of agricul- tural activities, changing rural identities and new institutional arrangements. Int. J. Agr. Res. Governance Ecol. 7(4-5), 361- 385. Doi: 10.1504/IJARGE.2008.020083 Scheaffer, R.L., W. Mendenhall, and L. Ott. 2006. Elementos de muestreo. Paraninfo, Madrid, Spain. Sistema Estadístico de Comercio Exterior (SIEX). 2016. Estadísticas de comercio exterior, DIAN. Retrieved from: http://websiex. dian.gov.co/; consulted: January, 2016. Sonnino, R. 2007. Embeddedness in action: Saffron and the making of the local in southern Tuscany. Agric. Hum. Values 24(1), 61-74. Doi: 10.1007/s10460-006-9036-y Traub, A. 2009. El tabaco en un mundo de contradicciones. Oficina de Estudios y Politicas Agrarias (ODEPA), Ministerio de Agri- cultura, Gobierno de Chile. Santiago de Chile, Chile. http://www.minhacienda.gov.co/portal/page/portal/HomeMinhacienda/saladeprensa/Presentaciones/2015/2015-03-17-PIB-IV-2014-oferta-3.1.pdf http://www.minhacienda.gov.co/portal/page/portal/HomeMinhacienda/saladeprensa/Presentaciones/2015/2015-03-17-PIB-IV-2014-oferta-3.1.pdf http://www.minhacienda.gov.co/portal/page/portal/HomeMinhacienda/saladeprensa/Presentaciones/2015/2015-03-17-PIB-IV-2014-oferta-3.1.pdf http://www.banrep.gov.co/documentos/publicaciones/regional/ESER/bucaramanga/2005_diciembre.pdf http://www.banrep.gov.co/documentos/publicaciones/regional/ESER/bucaramanga/2005_diciembre.pdf http://www.banrep.gov.co/documentos/publicaciones/regional/ESER/bucaramanga/2005_diciembre.pdf https://doi.org/10.1504/IJARGE.2008.020083 https://doi.org/10.1007/s10460-006-9036-y