Milk quality and financial management at different scales of production on dairy farms located in the south of Minas Gerais state , Brazil

Submitted on April 22 , 2014 and accepted on April 19 , 2017. This study is part of the first author’s Master’s dissertation. 2 Universidade Federal de Lavras, Departamento de Ciência dos Alimentos, Lavras, Minas Gerais, Brazil. marcel.paixao@yahoo.com.br; luizronaldo@dca.ufla.br; sandra@dca.ufla.br 3 Universidade Federal de Lavras, Departamento de Medicina Veterinária, Lavras, Minas Gerais, Brazil. malopes@dmv .ufla.br; gmcosta@dmv .ufla.br 4 Embrapa Gado de Leite, Juiz de Fora, Minas Gerais, Brazil. guilherme.souza@embrapa.br 5 Universidade Federal Fluminense, Departamento de Saúde Coletiva V erinária e Saúde Pública, Niterói, Rio de Janeiro, Brazil. gu herme.souza@embrapa.br *Corresponding author: marcel.paixao@yahoo.com.br Milk quality and financial management at different scales of production on dairy farms located in the south of Minas Gerais state, Brazil 1


INTRODUCTION
Dairy activity is present all over Brazil, and it generates income, taxes and jobs.Minas Gerais state deserves special attention in this activity, and in the third quarter of 2013 it produced 1,553,256 million of liters of milk, representing 25.9% of this country total supply within this period, a slightly higher production as compared to the same period of 2012 within this state (IBGE, 2013).However, the Brazilian milk production chain has undergone a series of transformations, or a restructuring process, which had generated a series of consequences for the different agents involved.It can be observed that the primary production has become more dependent on industry, losing autonomy over its production processes and scale of production (Bortoleto, 2000).
The globalization of the economy and its follow competitiveness, have turned milk quality traits (somatic cell count and total bacterial count) in the main concern for the production chain in order to obtain good economic results (Vieira, 2010).In this scenario, dairy industries started to establish new conditions for the acquisition of bulk tank milk, by remunerating farmers that produced milk with quality and in larger volumes, in order to obtain dairy products with higher industrial yield and better sensorial aspects, reduction of transport costs, and optimization of milk supply routes.
Several reasons contribute to explain these previous cited changes, e.g., the lack of milk quality standardization and volumes of raw milk offered to industry (Martins, 1999).These transformations in the milk production chain and market have concerned the dairy farmers on the need of a good management of the activity, which leaded to a more efficient and competitive production; and thus farmers have been assuming a businessman position and leaving behind the amateurism, by establishing permanent financial management regardless of their production system size (Lopes et al., 2010).
Recent studies have studied some milk quality traits (milk composition and hygienic-sanitary aspects -SCC and TBC) and its association with different production scales (Borges et al., 2013;Bodenmüller et al., 2010), while others have studied technical and management efficiency of dairy farmers and its association with different scales of production (Fassio et al., 2005;Gonçalves et al., 2008;Lopes et al., 2008).However, there are no studies associating both milk quality traits and financial management of dairy farms within different scales of production.It is well known that to raise productivity and bulk milk quality (e.g., adequate levels of total solids and hygienic-sanitary traits) in a herd, the income of dairy activity is, at least in part, the main key to achieved this goal; because it allows investments that can raise productivity with a rapidly payback period, and also contributes to the whole milk production process as investments in herd sanity, feed, and establishment of good agricultural practices for milk quality (Paixão et al., 2014).
The objective of this study was to analyze the relationship between some aspects related to financial management and scale of production with milk quality outcomes of 100 dairy farms located in six counties in the south of Minas Gerais state, Brazil.
A semi-structured questionnaire with questions related to general aspects of production and financial management of the farms was elaborated and applied from January to April 2012 in the selected farms.The data was transferred to an electronic spreadsheet and answers were grouped by categorization and frequency.Categorical (qualitative) data were transcribed according to its category code, while open questions and quantitative data were categorized before transcription.Farmers were grouped according to daily milk scales of production as previously described by Lopes et al. (2008): small (lower than 151 liters per day); medium (151 to 500 liters per day); and large (higher than 500 liters per day).
Data from bulk milk tank analyses [somatic cell count (SCC), total bacterial count (TBC), fat content, protein, and lactose] and daily production were acquired from the dairy industry milk quality extension team, and means were calculated from an interval period of 12 months of analyses (at least one bulk tank milk analyses per month -six months prior and after the beginning of the study, respectively).In cases where farmers no longer sold the milk to the dairy industry during this interval period, data from the previous 12 months prior to their exit were used for calculus.In order to obtain a normal distribution of the data for TBC and SCC means calculus, a logarithmic transformation (Log 10 ) were performed before analyses of variance (ANOVA).
The bulk milk tank samples were collected by the dairy industry milk quality extension team, properly packaged and sent to Clínica do Leite, at "Luiz de Queiroz" School of Agriculture (ESALQ/USP) monthly.Milk composition and SCC analyses were performed on Bentley Combi System 2300 ® electronic equipment and TBC analyses were performed on IBC BactoCount ® electronic equipment (Bentley Instruments Incorporated, Chaska, United States of America).
Analyses of variance and means tests were performed using SISVAR (Statistical Analysis System) software by the Scott-Knott test (Ferreira, 2008).Chi-square tests and relative risks between variables were performed using SPSS Statistics ® 17.0 software (SPSS, 2008).Each outcome were classified in two categories (problematic and control) and scales of production in three categories (small, medium, large) prior to the chi-square tests.Factors associated at P < 0.05 were considered significant if the inferior confidence interval (ICI) and the superior confidence interval (ICS) had values higher than 1.

RESULTS AND DISCUSSION
The descriptive characteristics of the 100 enrolled dairy farms located in six different counties in the southern region of the state of Minas Gerais classified by scales of  According to Lopes et al. (2008): small (lower than 151 liters per day), medium (151 to 500 liters per day), and large (higher than 500 liters per day); 2 Considering 305 days for the lactation period.
production are described in Table 1.It can be observed that the majority of farmers produced less than 151 liters daily, accounting for only 13.42% of the company's total daily milk supply.The average daily production was 269.84 liters per farm (Table 1), higher than previously reported in Brazil (52 L/ farm/day) and in Minas Gerais state (88 L/farm/day) in the year of 2006 (Zoccal, 2008).According to this author, although an increase was observed in the last decade in the daily milk supply from Brazilian farmers, the production per farm still lower as compared to international standards, because countries such United States, New Zealand and Australia production vary from 2,000 to 3,000 L/farm/day.
The average number of lactation cows per farm was 27.33.The overall average for daily milk production per cow during the study period was 8.9 L (Table 1), higher than the Brazilian average (5.1 L/cow/day) in the year of 2011 (SEAB, 2013).The estimated milk production per cow/ year, considering a lactation period of 305 days, was 2,689.57(± 1,399.22)L; lower than previously described by Fassio et al. (2006) (4,157 L/cow/year) in the state of Minas Gerais, but higher than this country average (1,394 L/ cow/year) (Embrapa, 2010) and similar with values from higher producing regions (2,628 L/cow/year) such as north of Rio Grande do Sul, west of Santa Catarina, and southwest of Paraná, Brazil (Zoccal et al., 2011).
Means of bulk tank milk composition analyses from the 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period are presented in Table 2.We identified that all composition traits were in compliance with the current normative of milk quality in Brazil (Normative Instruction No. 62, of 2011) which standardize the minimum values for protein, fat and non-solids fat of 2.9 g/100g; 3.0 g/100g; and 8.4 g/100g for refrigerated raw milk (Brazil, 2011) respectively.Milk composition traits had not differed within daily production scales (P < 0.05).
Protein means in our study were higher as compared to Borges et al. (2013) (ranging from 3.0 to 3.15g/100g) that studied milk composition at different production scales in 183 dairy farms located in Central Mineira and Minas West mesoregions in the state of Minas Gerais.These authors also had not identified statistical differences in protein means within different scales of production.According to Stelwagen (2011) protein levels in bulk tank milk is associated to genetic and non-genetic factors (diet, endocrine factors, environmental temperature, milking frequency, lactation stage and mastitis).
Fat mean values in our study (3.68 g/100g -Table 2) were similar to results previously reported by Bodenmüller et al. (2010) (3.72 g/100g) of 1,196 dairy farms bulk milk tank analyses from a dairy industry located in Londrina, Paraná, Brazil.Fat content had the higher variation among milk components, and according to McGuire et al. (2011) this variation is associated to breed, diet, lactation stage, season, environment temperature, and body score condition of the animals.
Our lactose means ranged from 4.49 to 4.52 g/100g (Table 2) and it was slightly higher than those previously reported by Borges et al. (2013) (4.37 to 4.47 g/ 100g).According to Bodenmüller et al. (2010) somatic cell count and total bacterial count has a positive association with milk lactose, thus, this milk component is impaired when milk sanitary-hygiene aspects are poorly handled.
Our total and non-fat solids results were similar than previously reported by Ribas et al. (2004) (12.32 and 8.62% for total and non-fat solids respectively) in a study that enrolled 257,540 bulk milk samples from 3 Brazilian states (Paraná, Santa Catarina and São Paulo) during January 1999 and November 2001.Paiva et al. (2012) stated that an increase in the raw milk solids content is crucial for the dairy industry in order to raise its products yield, and thus, increasing the competitiveness both in national and international dairy products market.
Means and standard deviations of the total bacterial count (TBC) and somatic cell count (SCC) analyses of bulk tank milk from 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period  (2008): small (lower than 151 liters per day), medium (151 to 500 liters per day), and large (higher than 500 liters per day).

Variable CV (%) 2
All (n = 100) are presented in Table 3.Our TBC results of small farmers was higher than the minimum standards (300,000 CFU/mL) from the current legislation (Normative Instruction number 62 of December 29, 2011, in the South, Southeast and West-Center regions, with dates from September 1, 2013) (Brasil, 2011); but in accordance with the same legislation at the time of the analysis in such region (600,000 CFU/mL).Our Log 10 TBC means of small farmers were greater as compared to medium and large farmers (P < 0.05) (Table 3).This results suggest that small farmers probably has smaller incomes, in concordance with Alves et al. (2004) which affirms that small farmers faces many financial difficulties in order to optimize production and many times they remain outside the process of modernization of agriculture, and thus directly impairing the quality of milk produced.Small farmers also had a low contribution in the total volume of milk supplied to the dairy industry in study (13.42%), but its contribution to the total milk TBC rate (calculated by multiplying the volume of milk collected on each farm by the number of its CFU/mL, and divided by the sum of the total milk volume multiplied by its CFU/mL) was high (34.71%)(Table 3).This results show the importance of strategic plans in order to improve hygienic aspects during milking routine and refrigeration of milk.A proper milking routine should include the correct sanitization of utensils, milking equipment, milking clean cows, and a periodical check for the bulk tank temperature (Paixão et al., 2014), because a small volume of milk with a very high load of microorganisms can influence the quality of the milk truck.
The SCC mean within all farmers scales were not in compliance with the current legislation (500,000 cells/mL) but were in compliance at the period of collection (2011) (600,000 cells/mL) (Normative Instruction No. 62) (Table 3).The contribution rate of SCC of each daily production scale in the total SCC was similar to the percentages of contributions in the volume of milk collected daily.Besides lactose means within all daily production scales farmers did not statistically differ (P < 0.05), medium farmers that had a greater SCC means also presented a slightly lower lactose means as compared to small and large farmers, in agreement with Borges et al. (2013) who identified a similar situation.According to Shuster et al. (1991) the lower synthesis of lactose is caused by destruction of the secretory tissue, leading to a leakage of lactose from the alveolus to the bloodstream due to an increased permeability of the membrane, and also by lactose fermenters intramammary pathogens.
The frequency of farms and relative risks of financial management aspects associated to bulk tank milk total bacterial count (TBC) and somatic cell count (SCC) higher than the standard values of the legislation from the 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period are presented in Table 4.We identified that 40.4% of the total of small farmers had TBC means higher than the legislation limits.Similar proportions within daily milk production scales farmers had SCC above the legislation limits (varying from 46.8% to 50.0% for small and large farmers, respectively) (Table 4).In agreement with our study, Mattioda & Bittencourt (2010) studying different production scales milk farmers in the region of Fernandes Pinheiro, Pará state, Brazil, also identified that smaller farmers had greater bulk tank milk TBC as compared to the others categories of farmers.
We identified that the majority of farmers had no control over incomes (including raw milk vending, animals and manure), but small farmers had the highest absence of incomes control (Table 4), which suggest the lack of professionalism within these farmers classes.According to Oliveira & Pereira (2009), the sustainability of a particular activity, or, in other words, its long-term survival capacity, is directly associated with its economic management.
Most farmers had no control over their costs, and the greater proportion of them was classified under the small farmer category.Spreadsheets were not used for costs control in one-third of the farmers which declared that had it under control.The use of spreadsheets, both electronic or a simple field notebook, are essential for financial costs control, because it facilitate the division of expenses into groups and assist milk quality extension teams and farmers to make future decisions (Lopes & Lopes, 1999).The majority of farmers (77%) reported that they were unaware of their average cost of production; and one farmer reported that knew their costs of production, but do not use a spreadsheet to record it.According to Gomes & Alves (1999), the correct determination of the dairy farm activity production cost is complex, due to its complexity, such as incomes that proceed from meat vending along with milk, family labor, and continuous production.
The majority (55%) of the farmers reported that do not know how much the dairy industry pay for milk quality bonuses neither how much this extra income can represents monthly; and a higher proportion was identified among small farmers as compared to medium and large farmers.Paixão et al. (2014) studying different hypothetical scenarios in order to calculate the implementation costs of different levels of good agricultural practices (GAP) (low, TBC (> 300 † vs. < 300) (CFU/mL of milk) x 10 3 SCC (> 500 † vs. < 500) (cells/mL of milk) x 10 3 medium and high) associated to milk quality improvement in a dairy farm, identified that, despite its higher initial investment, producers under the high GAP scenario had a higher and faster economic returns due to higher SCC and TBC bonuses paid from the dairy industry, as compared to the others hypothetical scenarios.
Financial management factors had no association with SCC levels above the current regulations in the chi-square tests.However, farmers that had no information about how much the dairy industry pay for milk quality bonuses had 2.95 times more likely (P = 0.048) of TBC means to be above the current regulation as compared to farmers who had information about theses bonuses (Table 4).Our results suggest that simple information about monetary values paid for milk quality bonuses has an impact on farmer's attitude, and thus, an indirect positive impact on the milking routine hygiene.
Chi-square tests and relative risks of different scales of productions associated to financial management aspects and hygiene-sanitary quality of bulk tank milk in a 12months analyses period from 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil are presented in Table 5.Small farmers had a greater negligence of costs control and hygienic production of bulk tank milk (TBC), as compared to medium and larger farmers, because from all chi-square tests performed for "small vs. medium" and "small vs. large" farmers, small farmers had an positive association with all negative category (P < 0.05), with an exception for the negative categories of the questions "Do you know how much is your production cost?" (no vs. yes) (P = 0.214) and for SCC (> 500,000 cells/mL vs. < 500,000 cells/mL) (small vs. medium and small vs. large, P = 0.874 and 0.818, respectively) (Table 5).Greater odds ratios (relative risks) were identified for "small vs. large" as compared to "small vs. medium" farmers for all problematic categories tested (with an exception for SCC); in addition, small farmers had 4.1 times more likely to had no information about their production costs (P = 0.05) as compared to larger farmers, evidencing that farmers with higher scales of production had greater concerns about management control of their dairy activity as compared to small farmers.An increase on farms productivity without a rapidly investment in technologies or changes in production system -for a better yield of the land, labor and costs -is only possible if the management of the dairy activity is sturdily improved (Ferreira Junior & Cunha, 2004).
No significant association (P < 0.05) was identified between medium and large farmers within all problematic categories tested.Our results indicate that, unlike the small farmers, medium and large scales of production farmers had greater concerns about hygienic quality of milk (TBC) and costs management of their farm.

CONCLUSIONS
Aspects related to financial management, bulk tanks milk composition and quality traits of 100 dairy farms located in six different counties in the south of Minas Gerais state, Brazil, were studied.Bulk tanks milk composition were in agreement with the current legislation; although, hygienic aspects of milking routine and udder health status from studied herds still requires attention from farmers and milk quality extension teams in order to reduce bulk tank milk TBC and SCC to legislation standards.
Negligence about financial management aspects related to the dairy activity increased as farmer's daily scale of production decreased.Our results suggest that small farmers, which produces milk volumes lower than 150 liters per day, requires special attention from milk quality extension teams in order to improve productivity.Farmers that were aware of milk quality bonuses payment produced milk with a greater hygienic standard, and our results suggest that this awareness has a direct impact on farmer's attitude, and an indirect impact on milking routine hygiene.Field professionals should draw farmer's attention to the benefits that this bonuses has on their monthly income and also consumer's health.

COMMITTEE OF ETHICS ON RESEARCH
This study was approved and performed in accordance with the technical standards of biosafety and ethics from Plataforma Brasil and registered under protocol number 132.569.

Table 1 :
Descriptive characteristics of 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of milk production in a 12

Table 2 :
Means followed by the standard deviations of bulk tank milk composition analyses from 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period Means within a line with different superscripts differs in ANOVA test (P < 0.05); 2 CV = Coefficient of variation. 1ccording toLopes et al.

Table 3 :
Means and standard deviations of total bacterial count (TBC) and somatic cell count (SCC) analyses of bulk tank milk from 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period Scale

Table 4 :
Frequency of farms and relative risks of financial management aspects associated to bulk tank milk total bacterial count (TBC) and somatic cell count (SCC) higher than reference values from current legislation (Normative Instruction number 62, 2011) of 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil, classified by daily scale of production in a 12-months analyses period

Table 5 :
Chi-square tests and relative risks of different scales of productions associated to financial management aspects and hygiene-sanitary quality of bulk tank milk in a 12-months analyses period from 100 enrolled dairy farms located in six different counties in the south of Minas Gerais state, Brazil