Livestock Research for Rural Development 26 (10) 2014 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Determining factors of modernization of dairy farming in the Brazilian Amazon

M A S Santos, A C Santana, L C B Raiol and J B Lourenço Júnior1

Federal Rural University of the Amazon (UFRA).
Postal Code: 66.077-530. PO Box 917. Belém, Pará, Brazil.
1 State University of Pará.
Postal Code: 66.095-100. Belém, Pará, Brazil.


The article identifies the factors of the modernization of dairy farming in the Amazon and estimates the index of modernization of dairy farming (IMDF), establishing a hierarchy between the municipalities as regards the technological level of the activity. Fifteen variables obtained from the 2006 agricultural census were submitted to factor analysis. Five factors of modernization were extracted: F1 = food and sanitary management; F2 = specialization of commercial production; F3 = animal reproduction technology; F4 = mechanization of milking; and F5 = rural credit.

The IMDF allowed the framing of 393 municipalities analyzed into four technological levels (high, medium, low and very low), only 15 municipalities were identified with a high technological level. The State of Rondônia has the most modern dairy farming in the region – among the 15 municipalities with greater IMDF, 10 were located in Rondônia. The ranking of the factors indicates that the adoption of food and sanitary management practices is higher than that of reproductive technologies and mechanization of milking. The instruments of agricultural policy, and particularly rural credit, should finance the use of these technologies in order to increase the productivity of the herd and generate a higher quality product that would bring benefits to the whole production chain.

Key words: milk production, technology, tropics


Dairy farming is one of the activities of the regional agricultural sector that has grown the most in the last two decades. Santos et al (2010), highlight that this growth is attributed to the strong insertion in markets caused by the expansion of dairying, the number of agricultural establishments with the possibility of developing in production mixed livestock systems, the ability to generate income relatively continuously throughout the year and the availability of resources for funding, which all made it a significant option for family farming.

According to data from the 2006 agricultural census, 87,732 agricultural establishments produced milk in the northern region, which corresponded to a percentage of approximately 19% of the total. It is an activity that has strong small-producer participation since 79.3% of the total milk produced in the region is obtained from establishments with less than 200 hectares (IBGE 2011).

Martins et al (2008) highlighted that herds are specialized for the production of milk in only a small number of regions. However, he points out that there have been advances in the technological field, especially in the States of Pará, Rondônia and Tocantins, in the light of greater investment and the adoption of health management practices and pasture. These aspects have motivated the development of this study whose basic question is: Which factors are contributing to the modernization of the regional dairy farming?

To answer this question, the objective of this study was to identify the factors of modernization and to estimate an index to measure the level of modernization of dairy farming in the Amazon to establish a hierarchy between the municipalities on the basis of a technological level.

Materials and methods

The database used in this study was obtained from the 2006 agricultural census of the Brazilian Institute of Geography and Statistics (IBGE 2011). Fifteen technological level indicators of dairy farming were determined to be applied to 393 municipalities in the northern region.

The 15 indicators were: X1– percentage of establishments which effect pasture rotation; X2 – percentage of establishments which effect fertilization of pasture; X3 – percentage of establishments which effect disease control; X4 – percentage of establishments which take into account expenses with salt and feed; X5 – percentage of establishments which show animal medicines expenses; X6 – participation percentage of each municipality in the value of regional milk production; X7 – percentage of establishments that sell fresh milk; X8 – participation percentage of the herd of cows milking in relation to effective municipal beef; X9 – index of specialization or locational quotient (IQ) of the municipality in relation to dairy farming; X10 – dairy productivity (litres/cow/year); X11 –percentage of establishments that perform artificial insemination; X12 – percentage of establishments that carry out embryo transfer; X13 – percentage of establishments carrying out mechanized milking; X14 – percentage of establishments which utilize cooling tanks; X15 – participation percentage of it is a town and municipality in the total value of the Fund’s credit financing Constitutional North (FNO), applied in dairy farming.

All variables, with the exception of the X9 variable, were calculated directly as percentages. The X9 variable named ‘specialization index’ or ‘locational quotient’ (QL), is a traditional indicator in studies of regional economy and aims to determine whether a municipality in particular has expertise in a given activity or specific sector. A detailed description of this indicator can be obtained at Haddad et al (1989) and Santana (2005). Applications for the case of dairy farming can be found at Lemos et al (2003) and Sena et al (2010).

The 15 indicators were submitted to factor analysis, with the aim of identifying the factors which affect the process of modernizing dairy farming. Factor analysis is a multivariate statistical method, used for the summarization of data. The goal is to analyze the relationships between a broad set of correlated variables, simplifying them by defining a set of common, termed latent dimensions of factors (Dillon and Goldstein 1984; Hair et al 2006; Manly 2008; Mingoti 2005). Subsequently the factorial scores were used to build the index of modernization of dairy farming (IMDF), aiming to establish a hierarchy among the 393 municipalities analyzed, based on four technological levels – namely (a) high; (b) medium; (c) low and (d) very low – IMDF was established.

Results and discussion

Five factors were extracted with eigenvalue above 1, and the total variance explained 71.2% of the data. Bartlett’s test was significant at 1% probability, rejecting the null hypothesis that the correlation matrix is an identity matrix. The KMO test presented a value of 0.797, indicating that the data sample is adequate for the factor analysis (Table 1).

Factor 1 explained the largest portion of the total data variance (23.7%) and it was associated positively with the variables X1, X2, X3, X4 and X5, which indicate rotation technologies and fertilization of pastureland, pest and disease control, and expenses with salt, food and medicine. Therefore, the dimension may be designated as “food and sanitary management”. The association of these variables in a single factor is justified by the fact that they are technologies with a higher level of adoption by the farmers and, often, are adopted jointly in livestock systems’ dual capability (beef and milk).

The second factor (factor 2) was responsible for 19.5% of the total variance of the data and it was related to the variables X6, X7, X8, X9 and X10, involving locational quotient (QL), the participation in the total cows milked roster, in marketing of fresh milk and level of productivity of the herds. The combination of these variables defines the degree of productive specialization and market insertion, and may be called “specialization of commercial production”.

Table 1 : Factorial loads after orthogonal rotation and their commonalities.

















































































































Variance explained (%)







Variance explained accumulated (%)







Source: survey data.
Notes: Bartlett’s Sphericity test = .27 6,452 (p < 0.01) and KMO = 0.797.
(*) Proportion of the total variance of the variable explained by common factors.
Marked in bold are the biggest factors’ weighing variable.

The third factor explained 12.2% of the total variance and related positively with the variables X11 and X12, which indicate the use of artificial insemination and embryo transfer in dairy herds and was referred to as “technology.” The combination of these technologies is of great importance for the improvement of the genetic pattern of the herd and milk quality. The fourth factor was named “mechanization of milking’ and was positively correlated with the variables X13 and X14 which specify the use of mechanized milking and cooling tanks on the property. The fifth factor was defined by the variable X15, which specifies the total amount of credit of the FNO applied in dairy farming, hence it was called “rural credit.”

The use of the region’s dairy properties insemination is only incipient – 1.40% of establishments adopt this type of practice and between the states this percentage does not exceed 2%. The same pattern is observed regarding the use of mechanized milking, whose regional average is only 0.98% of establishments (IBGE 2011). The expansion of these percentages should be a major goal, because it can ensure important gains in terms of genetic improvement of livestock and the quality of milk.

An important tool that can support the adoption of these technologies is rural credit (F5). In the last 20 years, more than 265,000 operations were contracted, including the modalities of funding, investment and marketing aimed at livestock used only for its milk, which involved resources to the tune of R$ 2.5 billion (BACEN 2011). The participation of the Constitutional Fund for the Financing of the North (FNO), managed by Banco da Amazônia, has been instrumental in this regard, since it accounted for an application of R$ 1.3 billion in the past 10 years (2000–2009).

Beyond traditional financing for the acquisition of arrays, breeders and animals for settlement priority must be given to the acquisition of machinery and equipment for mechanization and milking hygiene. The same support was important in assisting producers to comply with the normative 51 of Ministry of Agriculture, Livestock and Food Supply of Brazil, which deals with the use of cooling tanks. The expansion of the use of these technologies would add to the raw material quality for the dairy, bringing benefits to the entire production chain.

From each factor F1 = food and sanitary management; F2 = specialization of commercial production, F3 = F4 = playback technology, mechanization of milking and F5 = rural credit, the index of modernization of dairy farming (IMDF) was estimated and the segmentation of the municipalities into four technological levels (high, medium, low and very low) was established. It was found that 277 municipalities in the region were framed in the levels low and very low, amounting to 70% of the total. Only 15 municipalities, 5% of the total, exhibited a high technological level, with 10 located in the State of Rondônia, 3 in Tocantins and 2 in Pará.

The average IMDF for the region was only 37.7%, a value that lies in the low technological level. In accordance with the highest average was for the State of Rondônia (62.2%), followed by the Tocantins (44.4%) and Pará (30.8%). The lowest average value was the State of Amazonas (18.9%).

As for the regional average productivity variable this was 838 litres per cow per year, ranging from the very low technological level of 649 litres/cow/year to 1,176 litres/cow/year on a high technological level. Among these, the Rondônia State stands out, with the highest average in the northern region (1,080 litres/cow/year); however, this is lower than the national average (1,595 litres/cow/year) and well below the 2,407 litres/cow/year, obtained by the Santa Catarina State, which has the highest average productivity.



Brazilian Institute of Geography and Statistics (IBGE) 2010 Censo agropecuário 2006. Retrieved june 30, 2012, from

Central Bank of Brazil (BACEN) 2011 Anuário Estatístico do Crédito Rural. Retrieved june 30, 2011, from

Dillon W R and Goldstein M 1984 Multivariate analysis: methods and applications. New York: John Wiley & Sons. 587p.

Haddad P R 1989 Medidas de localização e especialização. In: Haddad PR, Ferreira CMC, Boisier S and Andrade TA. (editors). Economia regional: teorias e métodos de análise. Fortaleza: BNB-ETENE. p 225-247.

Lemos M B, Galinari R, Campos B, Biasi E and Santos F 2003 Tecnologia, especialização regional e produtividade: um estudo da pecuária leiteira em Minas Gerais. Revista de Economia e Sociologia Rural, 41: 117-1373. Retrieved june 20, 2012, from:

Hair Jr J F, Anderson R E, Tathan R L and Black W C 2006 Análise multivariada de dados. Porto Alegre: Bookman, 5th edition 593p.

Manly B F J 2008 Métodos estatísticos multivariados: uma introdução. Porto Alegre: Bookman, 3rd edition 229p.

Martins G C C, Rebello F K and Santana A C 2008 Mercado e dinâmica espacial da cadeia produtiva do leite na região Norte. Belém: Banco da Amazônia, 67p. Retrieved june 30, 2011, from

Mingoti S A 2005 Análise de dados através de métodos de estatística multivariada: uma abordagem aplicada. Belo Horizonte: EDUFMG. 297p.

Santana A C 2005 Elementos de economia, agronegócio e desenvolvimento local. Belém: UFRA; GTZ. 197p.

Santos M AS, Oliveira C M, Almeida R H C and Santana A C 2010 Estrutura e fontes de crescimento da pecuária leiteira no Norte do Brasil. Folha Socioambiental. Belém: UFRA-ISARH, 1: 7-11. Retrieved june 30, 2011, from

Sena A L S, Santos M A S, Santos J C, and Homma A K O 2010 Concentração espacial e caracterização da pecuária leiteira no estado do Pará. In: Congresso Brasileiro de Economia, Administração e Sociologia Rural, 48, 2010, Campo Grande, MS. Anais... Campo Grande, MS: SOBER. Retrieved june 15, 2012, from

Received 22 June 2014; Accepted 10 September 2014; Published 3 October 2014

Go to top