Trop Anim Health Prod (2015) 47:1271–1278 DOI 10.1007/s11250-015-0859-7
REGULAR ARTICLES
Work organization on smallholder dairy farms: a process unique to each farm Nathalie Hostiou 1 & Nathalie Cialdella 2,4 & Vincent Vazquez 3 & Artur Gustavo Müller 4 & Pierre-Yves Le Gal 2
Received: 22 October 2014 / Accepted: 18 May 2015 / Published online: 2 June 2015 # Springer Science+Business Media Dordrecht 2015
Abstract The way smallholder farms organize and carry out work impacts their ability to secure their livelihoods and meet growing demand for agricultural products. This study investigates the way dairy family farms in Brazil manage their workforce to achieve their objectives of production and income. Fifteen smallholder farms were surveyed using the QuaeWork method to understand the work organization on each farm. A high diversity of workloads was found, but these do not appear to be strictly related to the farms’ production systems. The high variability of workloads is linked to the available workforce, technical choices, and the delegation of tasks to an external workforce. Farmers can decrease their workload by adopting milking mechanization, silage, hiring labor, and increasing the duration of the work day. Work organization depends on a farmer’s personal choices, rendering the whole issue of workforce management a process unique to each farm.
Keywords Labor . Brazil . Task duration . Workload . Variability
* Nathalie Hostiou [email protected] 1
INRA, UMR 1273 Métafort, Theix, 63122 Saint Genès Champanelle, France
2
CIRAD, UMR Innovation, 34398 Montpellier, France
3
Montpellier SupAgro, 2 Place Viala, 34060 Montpellier, France
4
EMBRAPA Amazonia Oriental, Trav. Dr. Eneas Pinheiro s/n, CP 48, 66095-100 Belém, PA, Brazil
Introduction More than 150 million farms, and about 750 to 895 million people, are involved in dairy production in the world. Family farms provide three quarters of global milk production (Hemme and Otte 2010). In Brazil, where milk production increased by 75 % between 1990 and 2006, 50 % of the 25.4 billion liters of milk produced in 2006 came from family farms (Guilhoto et al. 2011). Agricultural tasks on these farms are mainly carried out by family labour (Wiggins et al. 2010), particularly when the farms have little equipment (Sraïri et al. 2013). This family-based workforce renders work organization, defined as the way farmers allocate labor resources to the various tasks to be carried out throughout the year, flexible and resilient, allowing farms to respond to changes in demand for agricultural products (Astigarraga and Ingrand 2011). However, it can be a disadvantage when the family is small or when the workload is heavy (Novo et al. 2012), or according to the type of production system involved (Wright et al. 2011). On specialized farms, the workload can fluctuate over the course of the year, resulting in a potential overload of work at specific periods (Dedieu 2009). In contrast, diversified farms with a range of on or off-farm activities can keep a family workforce occupied on a relatively steady basis over the entire year. Few studies have examined how farmers manage their workforce, either to improve their livelihoods or to meet market demands (Darnhofer et al. 2010). In this regard, smallholder dairy farms in Brazil are an interesting case because many are simultaneously intensifying production while diversifying their activities to varying degrees (Bernard et al. 2011). This study aims to (i) characterize work organization on family farms inserted in dairy markets and (ii) identify factors explaining the diversity of work organization found within
1272
Trop Anim Health Prod (2015) 47:1271–1278
the sample. The consequences of this diversity on the ability of family farms to meet growing demand for milk and vegetables are then discussed.
Material and methods Study area and sampling The study was conducted in the municipality of Unaí (State of Minas Gerais). The subhumid tropical climate includes a dry season from April to September. Fifty-five percent of the smallholder farms have dairy activities, and 60 % belong to assentamentos, settlements created after the 1996 land tenure reforms. A local dairy cooperative provides regular income and technical support to these farms, which supply up to 25 % of the 200,000 L collected daily by the cooperative and which account for 80 % of the cooperative’s suppliers (Altafin et al. 2011). Over the past decade, an increasing number of smallholder farms have specialized in dairy; in 2014, 21 % owned an individual cooling tank and 15 % a milking machine. Meanwhile, Brazilian government programs are providing them opportunities to diversify their production through market garden produce and fruit pulp. In order to achieve a detailed understanding of the way each farmer manages his workforce, the sample was limited to 15 dairy farmers reflecting the diversity of Boa União Table 1
Characteristics of the 15 farms studied
Name of farmer
Type
Number of dairy cows
Ro Ed Ge Wa DiP Ar Ad Fe Js
SLIF SLIF SLIF SLIF SLIF SIF SIF SIF SIF
20 14 12 8 15 23 12 28 22
Di
SIF
Am Zr Wal Eds Va
Dairy production (L/year)
assentamento farming. The farms were selected based on their level of specialization and intensification of dairy production, and the organization and size of their workforce. They were clustered in three groups to assess how their degree of specialization and intensification impacted their work organization: specialized farms with little intensification (SLIFs), specialized farms with intensification (SIFs), and diversified farms (DIFs) (Table 1). SLIFs have adopted the dominant model of animal feeding in the region based on cut sugarcane in the dry season (Bernard et al. 2011). They have no equipment and a low stocking rate (0.64 cows/ha). The farmer works all year, sometimes helped by his spouse and children, and more rarely by hired seasonal workers. SIFs have introduced maize or sorghum silage into their cropping plan. They are equipped (cooling tank, trap pail), have a higher stocking rate per hectare (0.84 cows/ha), and a higher number of permanent workers (two people: the farmer, plus his spouse or sons). DIFs combine milk production (stocking rate of 0.76 cows/ ha on average) with vegetables and fruit pulp produced on 0.5 to 1 ha. The volume of vegetables sold is low (for example 100 to 200 heads of lettuce) while the volume of fruit sold can be high (from 120 kg to over a ton a year). There are up to four permanent workers on the farm (a couple with two children). The considerable variability of productivity per cow observed within the sample (from 1120 to 4630 L/cow/year) is not linked to either the production system or the breed of cow involved. The three farm groups show similar values (around
Average production per cow (L/year)
MFA (ha)
Load/ha of MFA
Basic groupa
Outside basic groupb
Milk unit equipment
67,300 64,800 18,000 16,500 27,000 50,400 28,300 91,000 35,500
3370 4630 1500 2060 1800 2190 2360 3250 1610
48.0 15.0 16.0 20.5 22.5 20.0 33.5 33.5 20.5
0.4 0.9 0.8 0.4 0.7 1.2 0.4 0.8 1.1
1 1 1 1 1 2 1 1 2
One salaried Spouse Spouse+two children Spouse – Child Child Child Children
– – – – – Cooling tank – Trap pails –
40
133,100
3330
59.0
0.7
2
–
DIF DIF DIF DIF
14 17 11 14
32,400 48,900 44,000 32,200
2310 2880 4000 2300
20.0 25.0 20.0 22.0
0.7 0.7 0.6 0.6
4 2 2 2
– Child Child –
Trap pails, cooling tank, tractor, silo Cooling tank – Cooling tank Cooling tank
DIF
20
22,400
1120
16.5
1.2
1
Spouse+two children
–
All of the farmers own a cart and a cane shredder SLIF specialized farms with little intensification, SIF specialized farms with intensification, DIF diversified farm, MFA main fodder area a
Permanent workers organizing the work on the farm
b
Workers who do not spend a preponderant amount of time on livestock activities but whose participation can be critical for the realization of a task
Trop Anim Health Prod (2015) 47:1271–1278
2500 L/cow/year) and all but two farms (DiP and Eds) have herds of BGirlondo^, a cross between zebu cattle (Bos indicus) and Prim’Holstein.
Data collection and analysis Data were collected on the farms’ production structure, labor force, equipment–buildings available, combinations of activities (on and off farm) and livestock management practices (feed/diet, reproduction, and batching). The QuaeWork method was adapted to describe work organization over the year (Hostiou and Dedieu 2012). The method analyzes the evolution of the type of task (routine or seasonal) and workers involved throughout the yearly calendar. A routine task, quantified in hours per day, has to be done almost every day and can be neither aggregated nor postponed (milking, feeding). Seasonal work, quantified in days per year, includes tasks that can be postponed more easily and/or grouped over a given period (silage, weeding crops, or making fruit pulp). Two categories of farm labor are differentiated. The basic group (BG) comprises workers who devote most of their time to, and derive the bulk of their income from, agricultural work (farmer, couple). The nonbasic group (NBG) includes people who contribute to farm work but are uninvolved in farm management decisions, such as family help, hired labor, or informal arrangements between farmers to help each other. The year is split into homogenous periods, each characterized by a standard set-day. Each set-day is split into three time slots: (i) nonworking slot (meal, rest) (Tn); (ii) routine work slot (Tr); and seasonal work slot (Ts). The duration of each slot is defined for each worker in each period of the year; each total daily work duration is thus specific to each worker. The calculated time available for seasonal work (CTA), i.e., the time left to the BG worker (BGw) once the listed farming tasks have been performed is estimated in hours following (Eq. 1). Xn CT A ¼ ðDStdDayi N i −T ri −T si Þ ð1Þ i¼1 With CTA i DStdDayi Ni Tri Tsi
1273
These indicators were combined with qualitative information about workers’ choices and constraints regarding work.
Results A variable workforce and workload The workforce on 13 out of 15 farms consists mainly of family labor, with the BG composed of one to two people (Table 1), i.e., the husband, with the help of his children (boys) and wife in some cases. Physically demanding tasks such as daily milking and fodder cane feeding, and seasonal cane weeding and pasture maintenance, are conducted only by men. SIFs share silage tasks with other farmers or contract firms equipped with harvesters. DIFs have a larger workforce because women are full members of the basic group. They assume routine market gardening tasks (weeding, irrigating the kitchen garden) and seasonal work related to fruit (harvest, making fruit pulp). Despite the apparent homogeneity of tasks conducted, the total duration of routine work varies greatly between farms, from 1860 to 5280 h/year (3300 h/year on average) and from 710 to 3610 h/year/BGw (2000 h/year/BGw on average). The duration of seasonal work also is very variable, from 320 to 2370 h per year (810 h/year on average). A scale factor appears, with a strong correlation between the time dedicated to fodder crops and the total surface area on the farm (r=0.88), with an average of around 14 h/ha. The DIFs with a fruit pulp unit dedicate from 60 to 90 % of the total seasonal work to this activity, which takes a central place in the occupation of the family labor force. The total hours of work dedicated per cow is not linked to cow productivity (r=0.13; Fig. 1). Labor productivity, i.e., the amount of milk produced per hour dedicated to milk production, varies greatly from one farm to another (8 to 40 L of milk per hour) (Table 2), and appears to be linked to variations in milk and herd feeding practices. Impact of practices on dairy workload
Calculated time available for seasonal work Period i Duration of standard day for period i Number of days for period i Total duration of routine work for period i Total duration of seasonal work already carried out for period i
An Excel® file was created to store quantitative data regarding work durations per worker type. CTA was considered as a quantitative indicator to evaluate a farm’s ability to increase production. Work efficiency was estimated for dairy production (Tr/cow/year) and for sugarcane weeding (Ts/ha/year).
Milking (including the transport of milk up to the cooling tank) and herd feeding are the most demanding routine dairy tasks, accounting for respectively 46 and 39 % of the routine work during the dry season. These tasks are the most variable from one farm to another, with the total duration of routine herd work ranging from 105 to 279 h per cow. Economies of scale related to herd size are not systematic (Fig. 2). The total daily duration of milking varies from 1.8 to 8.0 h depending on the number of cows (r=0.67), but also the number of milkings per day and the number of workers involved. The use of trap pails cuts milking time by half and allows a greater volume of milk without requiring increased labor.
1274
Trop Anim Health Prod (2015) 47:1271–1278
Fig. 1 Variation of cow productivity (L/cow/year) according to the total work hours spent per cow (h/cow)
5000 4500
Litres per cow/year
4000 3500
R² = 0.1343
3000 2500 2000 1500 1000 500 0
100
150
200
250
300
350
hours per cow
Cane fodder feeding in the dry season presents a similar variability, ranging from 1.5 to 8.0 h/person/day. Heavy weed infestation doubles the time required to cut a cartful of cane. The duration of transport increases with the distance between the fodder cane field and the corral where cows are fed (between 0.5 and 2 h/day). Silage used in SIFs requires less time as the silo often is located near the corral. Delegating routine dairy work is rare since the BG carries out nearly all (94 %) of the work. 25 % of the seasonal work is delegated to NBG workers. Tasks which can be mechanized (ploughing, sowing, silage) are contracted out, using the tractor of the assentamento. Manual weeding of pastures is delegated to unpaid family labor (children) or to day workers. Three farmers also rely on mutual aid arrangements between neighbors for pasture maintenance and silage harvesting.
Table 2 Main work variables per farm
Market gardening activities mainly compete with routine dairy herd work during the dry season, while fruit pulp work competes with forage maintenance at the beginning of the rainy season. When dairy production is the first priority, this competition may restrain a farm from diversifying. Women are more involved in vegetable and fruit activities, and consequently in related decision making, than in milk production. Their involvement is thus fundamental for diversification strategies. Evaluating farm work abilities to expand production CTA gives an estimate of a farmer’s ability to increase labor inputs on his farm. A low CTA indicates that a farm’s workforce is operating at close to peak capacity, while a high one
Name of farmer
CTA
Total work dedicated to milk production (h)
Labor productivity (L milk/h)
Ro Ed Ge Wa DiP Ar Ad
1140 730 770 910 1760 1480 1640
3390 4640 1510 2070 1820 2210 2370
20 14 12 8 15 23 12
Fe Js Di Am Zr Wal Eds Va
1260 3490 3100 1770 710 1050 1160 1550
3280 1640 3370 2330 2890 4010 2310 1140
28 22 40 14 17 11 14 20
Trop Anim Health Prod (2015) 47:1271–1278
1275 300 Duration of routine dairy work per cow (hour)
Fig. 2 Variation of routine dairy work per cow according to the herd size per farm
280 260 240 220 200 180 160 140 120 100 5
indicates that extra work capacity is available for expanding agricultural activities. The CTA in the sample range from 710 to 3490 h/year (1500 h/year/farm on average). The size of the BG does not really explain this variability because the CTA per person in the BG also varies from 360 to 1740 h/year (980 h/year on average). Three groups emerge (Fig. 3): (i) four farms with little extra work capacity (CTA under 600 h/person); (ii) six farms with considerable extra work capacity
10
15
20
25 30 Number of cows
35
40
45
(CTA over 1100 h/person); and (iii) five farms with an intermediary extra work capacity. These differences are partly related to the type of production system due to its consequences on routine work in the dry season. SIFs with silage complementing cane have an intermediary or even high extra capacity because they have reduced the time dedicated to feeding. DIFs often have little room to maneuver because they juggle at least two activities
2000
Fig. 3 Variation of total available seasonal time according to total duration of current routine and seasonal work per farm
1800 1600
CTA/BGw (h)
1400 1200 1000 800 600 400 200 0 0
1000
2000
3000 4000 (Tr+Ts)/BGw (h) SLIF
SIF
5000
6000
DIF
Tr: Routine time duration; Ts: Seasonal time; BGw: Basic Group worker; CTA: Calculated time available for seasonal works SLIF: Specialized farm with little intensification SIF: Specialized farm with intensification DIF: Diversified farm
7000
1276
Trop Anim Health Prod (2015) 47:1271–1278
Fig. 4 Variation of calculated time available per day per person according to the duration of the work day in the dry season
requiring daily routine work. SLIFs are found in all three groups depending on their fodder feeding system. Daily routine work in the dry season can be nearly three times longer (ranging from 5 to 14 h) in cane-based systems than in silagebased systems. Consequently, the CTA per BGw is highly variable (0 to 9.9 h) depending on the type of farm and the size of the BG. Adjusting daily work duration Farmers may increase their daily work duration when confronted, regularly or episodically, with a heavy workload. This work duration varies from 7 to 14 h per day in the dry season, a period when labor constraints are felt most deeply due to trough feeding of animals and market gardening activities. Four types of situations were identified (Fig. 4): (a) five farmers have decided not to exceed 11 h per day but have a reduced CTA per BGw of less than 3 h/day. These cases show a good fit between what the BG is prepared to do and the work demand; (b) three farmers working less than 11 h per day and with a CTA above 4 h/day. The labor demands are relatively low, which allow the BG to have a moderate workload. This
case shows a considerable extra capacity to increase production activities; (c) six farmers ready to work more than 11 h/ day, but with a low CTA. These farms have little room to maneuver because they cannot reasonably increase the daily duration of work; (d) one farmer ready to work more than 11 h per day with a CTA above 4 h/day. This distribution is only partially related to farm type: while DIFs generally are in the group with high labor constraints (c) and SLIFs are in group (a), SIFs are divided between the three groups. This fragmented distribution indicates that choosing the duration of the working day on these smallholder farms remains a personal decision which is directly related to the amount of time each farmer wishes to dedicate to production activities.
Discussion This study shows the great diversity of situations regarding labor force and work organization that can be encountered using a limited sample and relatively similar production systems in a rapidly growing dairy basin. The initial assumption
Trop Anim Health Prod (2015) 47:1271–1278
made in the sampling process of a link between a farm’s production system and work organization is not strictly upheld. Indeed, the high variability of workloads on these smallholder dairy farms is linked to a combination of many factors: available workforce and its allocation to given tasks, technical choices such as mechanization made for each task, possibility to increase daily work duration, delegation of tasks to hired labor, mutual aid or contractors. The numerous interactions between these factors render workforce management a complex process where individual choices prevail, making each farm a special case. These choices typically are related to each individual’s personal vision (Martel et al. 2008): faced with increasing demand for labor, one farmer will solicit his spouse or children, another will lengthen his working day, and another will hire a paid worker. Strategies to increase milk or fruit production to meet growing market demands, for instance by increasing herd/ garden size or improving cow/fruit productivity, could increase workloads during some periods of the year and create tension in relation to labor supply, in particular on diversified farms. Some farmers may prefer to reduce their focus on dairy production and spread their workload throughout the year by diversifying their activities. This process expands the workforce involved in agricultural tasks by allowing women to participate in farm production decisions. However, it can lead to higher workloads and competition with dairy production (Lemaire et al. 2014). Other farmers can increase the supply of labor when needed by extending their daily work duration, hiring labor, or developing mutual assistance networks. However, these alternatives require specific human resource management skills (Stup et al. 2006). Technical changes may also contribute to reducing workloads while maintaining or increasing milk production. For instance, switching from sugarcane cutting to maize silage reduces the time spent feeding cows (Ferris et al. 2006). While such a change does not involve much investment, it does require a certain degree of equipment and technical knowhow to ensure feed quality. Switching from twice to oncedaily milking also could directly impact work efficiency without entailing extra costs, but it would decrease cow productivity (Clark et al. 2006). The benefits of a milking machine (Wagner et al. 2001) remain to be verified since only two farms in the sample had one, and both had high milking times. Moreover, as farmers would require cash or credit to invest in such machines, their farms would become more tightly linked to the dairy industry.
Conclusions This study shows that on smallholder dairy farms which are quite similar in structure and modes of functioning, work task durations and time available vary greatly. Factors contributing
1277
to this variability include the structure of the farm (notably the size of herds and fodder crop areas), production system adopted (diversification of activities, cow feeding system), but also the position of the farmer in relation to work (attitude toward and acceptability of the workload). The combination of this set of factors renders the management of the workforce a process unique to each farm. In a context of growing demand for milk, it would be useful to develop support tools to help farmers consider how their work organization can be affected by strategic changes such as increasing herd size, changing the feeding system, or introducing new activities or equipment. A simulation tool based on the QuaeWork framework is currently under development to serve this need. Acknowledgments The authors would like to thank the 15 livestock farmers from Unaí-MG who participated in the study and Grace Delobel for translating the paper into English. This work was funded by the Agence Nationale de la Recherche under the Systerra Program: ANR08-STRA-10 (Ecological, technical and social innovation processes in Conservation Agriculture). Conflict of interest The authors declare that they have no conflict of interest.
References Altafin, I., Pinheiro, M.E.F., Valone, G.V., and Gregolin, A.C., 2011. Produção familiar de leite no brasil: um estudo sobre os assentamentos de reforma agrária no município de Unaí (MG). Revista UNI 1, 31–49. Astigarraga, L., and Ingrand, S., 2011. Production Flexibility in Extensive Beef Farming Systems. Ecology and Society 16, 1. Bernard, J., Le Gal, P.Y., Triomphe, B., Hostiou, N., and Moulin, C.H., 2011. Involvement of small-scale dairy farms in an industrial supply chain: when production standards meet farm diversity. Animal 5, 961–971. Clark, D.A., Phyn, C.V.C., Tong, M.J., Collis, S.J., and Dalley, D.E., 2006. A systems comparison of once- versus twice-daily milking of pastured dairy cows. Journal of Dairy Sciences 89, 1854–1862. Darnhofer, I., Bellon, S., Dedieu, B., and Milestad, R., 2010. Adaptiveness to enhance the sustainability of farming systems. A review. Agronomy for Sustainable Development 30, 545–555. Dedieu, B., 2009. Qualification of the adaptive capacities of livestock farming systems, Revista Brasileira de Zootecnia 38, 397–404. Ferris, C.P., Frost, J.P., Binnie, R.C., Patterson, D.C., 2006. Dairy cows performance and labour inputs associated with two silage feeding systems. Grass and Forage Science 61, 304–314. Guilhoto, J., Azzoni, C.R., Silveira, F.G., Ichihara, S.M., Diniz, B.P., and Moreira, G. R. In : PIB Da Agricultura Familiar: Brasil – Estados. 2011. http://ssrn.com/abstract=1803225 orhttp://dx.doi.org/10. 2139/ssrn.1803225. Accessed on 18 Apr 2015. Hemme, T., and Otte, J., 2010. Status and Prospects for Smallholder Milk Production. A Global Perspective. FAO-PPLFP, Rome, Italy. Hostiou, N., and Dedieu, B., 2012. A method for assessing work productivity and flexibility in livestock farms. Animal 6, 852–862. Lemaire, G., Franzluebbers, F., Carvalho, P., and Dedieu, B., 2014. Integrated crop–livestock systems: Strategies to achieve synergy between agricultural production and environmental quality. Agriculture, Ecosystems & Environment 190, 4–8.
1278 Martel, G., Dourmad, J.Y., and Dedieu, B., 2008. Do labour productivity and preferences about work load distribution affect reproduction management and performance in pig farms? Livestock Science 116, 96–107 Novo, A.M., Jansen, K., and Slingerland, M., 2012. The sugarcanebiofuel expansion and dairy farmers’ responses in Brazil. Journal of Rural Studies 28, 640–649. Sraïri, M.T., Bahri, S., and Kuper, M., 2013. Le travail et sa contribution aux stratégies d’adaptation de petites exploitations agricoles familiales mixtes d’élevage bovin/polyculture au Maroc. Biotechnologie, Agronomie, Société et Environnement 17, 463–474.
Trop Anim Health Prod (2015) 47:1271–1278 Stup, R.E., Hyde, J., and Holden, LA. 2006. Relationships between selected human resource management practices and dairy farm performance. Journal of Dairy Science 89, 1116–1120. Wagner, A., Palmer, R.W., Bewley, J., and Jackson-Smith, D.B., 2001. Producer satisfaction, efficiency, and investment cost factors of different milking systems. Journal of Dairy Science 84, 1890–1898. Wiggins, S., Kirsten, J., and Llambí, L., 2010. The Future of Small Farms. World Development 38, 1341–1348. Wright, I.A., Tarawali, S., Blümmel, M., Gerard, B., Teufel, N., and Herrero, M., 2011. Integrating crops and livestock in subtropical agricultural systems. Journal of the Science of Food and Agriculture 92, 1010–1015.
REGULAR ARTICLES
Work organization on smallholder dairy farms: a process unique to each farm Nathalie Hostiou 1 & Nathalie Cialdella 2,4 & Vincent Vazquez 3 & Artur Gustavo Müller 4 & Pierre-Yves Le Gal 2
Received: 22 October 2014 / Accepted: 18 May 2015 / Published online: 2 June 2015 # Springer Science+Business Media Dordrecht 2015
Abstract The way smallholder farms organize and carry out work impacts their ability to secure their livelihoods and meet growing demand for agricultural products. This study investigates the way dairy family farms in Brazil manage their workforce to achieve their objectives of production and income. Fifteen smallholder farms were surveyed using the QuaeWork method to understand the work organization on each farm. A high diversity of workloads was found, but these do not appear to be strictly related to the farms’ production systems. The high variability of workloads is linked to the available workforce, technical choices, and the delegation of tasks to an external workforce. Farmers can decrease their workload by adopting milking mechanization, silage, hiring labor, and increasing the duration of the work day. Work organization depends on a farmer’s personal choices, rendering the whole issue of workforce management a process unique to each farm.
Keywords Labor . Brazil . Task duration . Workload . Variability
* Nathalie Hostiou [email protected] 1
INRA, UMR 1273 Métafort, Theix, 63122 Saint Genès Champanelle, France
2
CIRAD, UMR Innovation, 34398 Montpellier, France
3
Montpellier SupAgro, 2 Place Viala, 34060 Montpellier, France
4
EMBRAPA Amazonia Oriental, Trav. Dr. Eneas Pinheiro s/n, CP 48, 66095-100 Belém, PA, Brazil
Introduction More than 150 million farms, and about 750 to 895 million people, are involved in dairy production in the world. Family farms provide three quarters of global milk production (Hemme and Otte 2010). In Brazil, where milk production increased by 75 % between 1990 and 2006, 50 % of the 25.4 billion liters of milk produced in 2006 came from family farms (Guilhoto et al. 2011). Agricultural tasks on these farms are mainly carried out by family labour (Wiggins et al. 2010), particularly when the farms have little equipment (Sraïri et al. 2013). This family-based workforce renders work organization, defined as the way farmers allocate labor resources to the various tasks to be carried out throughout the year, flexible and resilient, allowing farms to respond to changes in demand for agricultural products (Astigarraga and Ingrand 2011). However, it can be a disadvantage when the family is small or when the workload is heavy (Novo et al. 2012), or according to the type of production system involved (Wright et al. 2011). On specialized farms, the workload can fluctuate over the course of the year, resulting in a potential overload of work at specific periods (Dedieu 2009). In contrast, diversified farms with a range of on or off-farm activities can keep a family workforce occupied on a relatively steady basis over the entire year. Few studies have examined how farmers manage their workforce, either to improve their livelihoods or to meet market demands (Darnhofer et al. 2010). In this regard, smallholder dairy farms in Brazil are an interesting case because many are simultaneously intensifying production while diversifying their activities to varying degrees (Bernard et al. 2011). This study aims to (i) characterize work organization on family farms inserted in dairy markets and (ii) identify factors explaining the diversity of work organization found within
1272
Trop Anim Health Prod (2015) 47:1271–1278
the sample. The consequences of this diversity on the ability of family farms to meet growing demand for milk and vegetables are then discussed.
Material and methods Study area and sampling The study was conducted in the municipality of Unaí (State of Minas Gerais). The subhumid tropical climate includes a dry season from April to September. Fifty-five percent of the smallholder farms have dairy activities, and 60 % belong to assentamentos, settlements created after the 1996 land tenure reforms. A local dairy cooperative provides regular income and technical support to these farms, which supply up to 25 % of the 200,000 L collected daily by the cooperative and which account for 80 % of the cooperative’s suppliers (Altafin et al. 2011). Over the past decade, an increasing number of smallholder farms have specialized in dairy; in 2014, 21 % owned an individual cooling tank and 15 % a milking machine. Meanwhile, Brazilian government programs are providing them opportunities to diversify their production through market garden produce and fruit pulp. In order to achieve a detailed understanding of the way each farmer manages his workforce, the sample was limited to 15 dairy farmers reflecting the diversity of Boa União Table 1
Characteristics of the 15 farms studied
Name of farmer
Type
Number of dairy cows
Ro Ed Ge Wa DiP Ar Ad Fe Js
SLIF SLIF SLIF SLIF SLIF SIF SIF SIF SIF
20 14 12 8 15 23 12 28 22
Di
SIF
Am Zr Wal Eds Va
Dairy production (L/year)
assentamento farming. The farms were selected based on their level of specialization and intensification of dairy production, and the organization and size of their workforce. They were clustered in three groups to assess how their degree of specialization and intensification impacted their work organization: specialized farms with little intensification (SLIFs), specialized farms with intensification (SIFs), and diversified farms (DIFs) (Table 1). SLIFs have adopted the dominant model of animal feeding in the region based on cut sugarcane in the dry season (Bernard et al. 2011). They have no equipment and a low stocking rate (0.64 cows/ha). The farmer works all year, sometimes helped by his spouse and children, and more rarely by hired seasonal workers. SIFs have introduced maize or sorghum silage into their cropping plan. They are equipped (cooling tank, trap pail), have a higher stocking rate per hectare (0.84 cows/ha), and a higher number of permanent workers (two people: the farmer, plus his spouse or sons). DIFs combine milk production (stocking rate of 0.76 cows/ ha on average) with vegetables and fruit pulp produced on 0.5 to 1 ha. The volume of vegetables sold is low (for example 100 to 200 heads of lettuce) while the volume of fruit sold can be high (from 120 kg to over a ton a year). There are up to four permanent workers on the farm (a couple with two children). The considerable variability of productivity per cow observed within the sample (from 1120 to 4630 L/cow/year) is not linked to either the production system or the breed of cow involved. The three farm groups show similar values (around
Average production per cow (L/year)
MFA (ha)
Load/ha of MFA
Basic groupa
Outside basic groupb
Milk unit equipment
67,300 64,800 18,000 16,500 27,000 50,400 28,300 91,000 35,500
3370 4630 1500 2060 1800 2190 2360 3250 1610
48.0 15.0 16.0 20.5 22.5 20.0 33.5 33.5 20.5
0.4 0.9 0.8 0.4 0.7 1.2 0.4 0.8 1.1
1 1 1 1 1 2 1 1 2
One salaried Spouse Spouse+two children Spouse – Child Child Child Children
– – – – – Cooling tank – Trap pails –
40
133,100
3330
59.0
0.7
2
–
DIF DIF DIF DIF
14 17 11 14
32,400 48,900 44,000 32,200
2310 2880 4000 2300
20.0 25.0 20.0 22.0
0.7 0.7 0.6 0.6
4 2 2 2
– Child Child –
Trap pails, cooling tank, tractor, silo Cooling tank – Cooling tank Cooling tank
DIF
20
22,400
1120
16.5
1.2
1
Spouse+two children
–
All of the farmers own a cart and a cane shredder SLIF specialized farms with little intensification, SIF specialized farms with intensification, DIF diversified farm, MFA main fodder area a
Permanent workers organizing the work on the farm
b
Workers who do not spend a preponderant amount of time on livestock activities but whose participation can be critical for the realization of a task
Trop Anim Health Prod (2015) 47:1271–1278
2500 L/cow/year) and all but two farms (DiP and Eds) have herds of BGirlondo^, a cross between zebu cattle (Bos indicus) and Prim’Holstein.
Data collection and analysis Data were collected on the farms’ production structure, labor force, equipment–buildings available, combinations of activities (on and off farm) and livestock management practices (feed/diet, reproduction, and batching). The QuaeWork method was adapted to describe work organization over the year (Hostiou and Dedieu 2012). The method analyzes the evolution of the type of task (routine or seasonal) and workers involved throughout the yearly calendar. A routine task, quantified in hours per day, has to be done almost every day and can be neither aggregated nor postponed (milking, feeding). Seasonal work, quantified in days per year, includes tasks that can be postponed more easily and/or grouped over a given period (silage, weeding crops, or making fruit pulp). Two categories of farm labor are differentiated. The basic group (BG) comprises workers who devote most of their time to, and derive the bulk of their income from, agricultural work (farmer, couple). The nonbasic group (NBG) includes people who contribute to farm work but are uninvolved in farm management decisions, such as family help, hired labor, or informal arrangements between farmers to help each other. The year is split into homogenous periods, each characterized by a standard set-day. Each set-day is split into three time slots: (i) nonworking slot (meal, rest) (Tn); (ii) routine work slot (Tr); and seasonal work slot (Ts). The duration of each slot is defined for each worker in each period of the year; each total daily work duration is thus specific to each worker. The calculated time available for seasonal work (CTA), i.e., the time left to the BG worker (BGw) once the listed farming tasks have been performed is estimated in hours following (Eq. 1). Xn CT A ¼ ðDStdDayi N i −T ri −T si Þ ð1Þ i¼1 With CTA i DStdDayi Ni Tri Tsi
1273
These indicators were combined with qualitative information about workers’ choices and constraints regarding work.
Results A variable workforce and workload The workforce on 13 out of 15 farms consists mainly of family labor, with the BG composed of one to two people (Table 1), i.e., the husband, with the help of his children (boys) and wife in some cases. Physically demanding tasks such as daily milking and fodder cane feeding, and seasonal cane weeding and pasture maintenance, are conducted only by men. SIFs share silage tasks with other farmers or contract firms equipped with harvesters. DIFs have a larger workforce because women are full members of the basic group. They assume routine market gardening tasks (weeding, irrigating the kitchen garden) and seasonal work related to fruit (harvest, making fruit pulp). Despite the apparent homogeneity of tasks conducted, the total duration of routine work varies greatly between farms, from 1860 to 5280 h/year (3300 h/year on average) and from 710 to 3610 h/year/BGw (2000 h/year/BGw on average). The duration of seasonal work also is very variable, from 320 to 2370 h per year (810 h/year on average). A scale factor appears, with a strong correlation between the time dedicated to fodder crops and the total surface area on the farm (r=0.88), with an average of around 14 h/ha. The DIFs with a fruit pulp unit dedicate from 60 to 90 % of the total seasonal work to this activity, which takes a central place in the occupation of the family labor force. The total hours of work dedicated per cow is not linked to cow productivity (r=0.13; Fig. 1). Labor productivity, i.e., the amount of milk produced per hour dedicated to milk production, varies greatly from one farm to another (8 to 40 L of milk per hour) (Table 2), and appears to be linked to variations in milk and herd feeding practices. Impact of practices on dairy workload
Calculated time available for seasonal work Period i Duration of standard day for period i Number of days for period i Total duration of routine work for period i Total duration of seasonal work already carried out for period i
An Excel® file was created to store quantitative data regarding work durations per worker type. CTA was considered as a quantitative indicator to evaluate a farm’s ability to increase production. Work efficiency was estimated for dairy production (Tr/cow/year) and for sugarcane weeding (Ts/ha/year).
Milking (including the transport of milk up to the cooling tank) and herd feeding are the most demanding routine dairy tasks, accounting for respectively 46 and 39 % of the routine work during the dry season. These tasks are the most variable from one farm to another, with the total duration of routine herd work ranging from 105 to 279 h per cow. Economies of scale related to herd size are not systematic (Fig. 2). The total daily duration of milking varies from 1.8 to 8.0 h depending on the number of cows (r=0.67), but also the number of milkings per day and the number of workers involved. The use of trap pails cuts milking time by half and allows a greater volume of milk without requiring increased labor.
1274
Trop Anim Health Prod (2015) 47:1271–1278
Fig. 1 Variation of cow productivity (L/cow/year) according to the total work hours spent per cow (h/cow)
5000 4500
Litres per cow/year
4000 3500
R² = 0.1343
3000 2500 2000 1500 1000 500 0
100
150
200
250
300
350
hours per cow
Cane fodder feeding in the dry season presents a similar variability, ranging from 1.5 to 8.0 h/person/day. Heavy weed infestation doubles the time required to cut a cartful of cane. The duration of transport increases with the distance between the fodder cane field and the corral where cows are fed (between 0.5 and 2 h/day). Silage used in SIFs requires less time as the silo often is located near the corral. Delegating routine dairy work is rare since the BG carries out nearly all (94 %) of the work. 25 % of the seasonal work is delegated to NBG workers. Tasks which can be mechanized (ploughing, sowing, silage) are contracted out, using the tractor of the assentamento. Manual weeding of pastures is delegated to unpaid family labor (children) or to day workers. Three farmers also rely on mutual aid arrangements between neighbors for pasture maintenance and silage harvesting.
Table 2 Main work variables per farm
Market gardening activities mainly compete with routine dairy herd work during the dry season, while fruit pulp work competes with forage maintenance at the beginning of the rainy season. When dairy production is the first priority, this competition may restrain a farm from diversifying. Women are more involved in vegetable and fruit activities, and consequently in related decision making, than in milk production. Their involvement is thus fundamental for diversification strategies. Evaluating farm work abilities to expand production CTA gives an estimate of a farmer’s ability to increase labor inputs on his farm. A low CTA indicates that a farm’s workforce is operating at close to peak capacity, while a high one
Name of farmer
CTA
Total work dedicated to milk production (h)
Labor productivity (L milk/h)
Ro Ed Ge Wa DiP Ar Ad
1140 730 770 910 1760 1480 1640
3390 4640 1510 2070 1820 2210 2370
20 14 12 8 15 23 12
Fe Js Di Am Zr Wal Eds Va
1260 3490 3100 1770 710 1050 1160 1550
3280 1640 3370 2330 2890 4010 2310 1140
28 22 40 14 17 11 14 20
Trop Anim Health Prod (2015) 47:1271–1278
1275 300 Duration of routine dairy work per cow (hour)
Fig. 2 Variation of routine dairy work per cow according to the herd size per farm
280 260 240 220 200 180 160 140 120 100 5
indicates that extra work capacity is available for expanding agricultural activities. The CTA in the sample range from 710 to 3490 h/year (1500 h/year/farm on average). The size of the BG does not really explain this variability because the CTA per person in the BG also varies from 360 to 1740 h/year (980 h/year on average). Three groups emerge (Fig. 3): (i) four farms with little extra work capacity (CTA under 600 h/person); (ii) six farms with considerable extra work capacity
10
15
20
25 30 Number of cows
35
40
45
(CTA over 1100 h/person); and (iii) five farms with an intermediary extra work capacity. These differences are partly related to the type of production system due to its consequences on routine work in the dry season. SIFs with silage complementing cane have an intermediary or even high extra capacity because they have reduced the time dedicated to feeding. DIFs often have little room to maneuver because they juggle at least two activities
2000
Fig. 3 Variation of total available seasonal time according to total duration of current routine and seasonal work per farm
1800 1600
CTA/BGw (h)
1400 1200 1000 800 600 400 200 0 0
1000
2000
3000 4000 (Tr+Ts)/BGw (h) SLIF
SIF
5000
6000
DIF
Tr: Routine time duration; Ts: Seasonal time; BGw: Basic Group worker; CTA: Calculated time available for seasonal works SLIF: Specialized farm with little intensification SIF: Specialized farm with intensification DIF: Diversified farm
7000
1276
Trop Anim Health Prod (2015) 47:1271–1278
Fig. 4 Variation of calculated time available per day per person according to the duration of the work day in the dry season
requiring daily routine work. SLIFs are found in all three groups depending on their fodder feeding system. Daily routine work in the dry season can be nearly three times longer (ranging from 5 to 14 h) in cane-based systems than in silagebased systems. Consequently, the CTA per BGw is highly variable (0 to 9.9 h) depending on the type of farm and the size of the BG. Adjusting daily work duration Farmers may increase their daily work duration when confronted, regularly or episodically, with a heavy workload. This work duration varies from 7 to 14 h per day in the dry season, a period when labor constraints are felt most deeply due to trough feeding of animals and market gardening activities. Four types of situations were identified (Fig. 4): (a) five farmers have decided not to exceed 11 h per day but have a reduced CTA per BGw of less than 3 h/day. These cases show a good fit between what the BG is prepared to do and the work demand; (b) three farmers working less than 11 h per day and with a CTA above 4 h/day. The labor demands are relatively low, which allow the BG to have a moderate workload. This
case shows a considerable extra capacity to increase production activities; (c) six farmers ready to work more than 11 h/ day, but with a low CTA. These farms have little room to maneuver because they cannot reasonably increase the daily duration of work; (d) one farmer ready to work more than 11 h per day with a CTA above 4 h/day. This distribution is only partially related to farm type: while DIFs generally are in the group with high labor constraints (c) and SLIFs are in group (a), SIFs are divided between the three groups. This fragmented distribution indicates that choosing the duration of the working day on these smallholder farms remains a personal decision which is directly related to the amount of time each farmer wishes to dedicate to production activities.
Discussion This study shows the great diversity of situations regarding labor force and work organization that can be encountered using a limited sample and relatively similar production systems in a rapidly growing dairy basin. The initial assumption
Trop Anim Health Prod (2015) 47:1271–1278
made in the sampling process of a link between a farm’s production system and work organization is not strictly upheld. Indeed, the high variability of workloads on these smallholder dairy farms is linked to a combination of many factors: available workforce and its allocation to given tasks, technical choices such as mechanization made for each task, possibility to increase daily work duration, delegation of tasks to hired labor, mutual aid or contractors. The numerous interactions between these factors render workforce management a complex process where individual choices prevail, making each farm a special case. These choices typically are related to each individual’s personal vision (Martel et al. 2008): faced with increasing demand for labor, one farmer will solicit his spouse or children, another will lengthen his working day, and another will hire a paid worker. Strategies to increase milk or fruit production to meet growing market demands, for instance by increasing herd/ garden size or improving cow/fruit productivity, could increase workloads during some periods of the year and create tension in relation to labor supply, in particular on diversified farms. Some farmers may prefer to reduce their focus on dairy production and spread their workload throughout the year by diversifying their activities. This process expands the workforce involved in agricultural tasks by allowing women to participate in farm production decisions. However, it can lead to higher workloads and competition with dairy production (Lemaire et al. 2014). Other farmers can increase the supply of labor when needed by extending their daily work duration, hiring labor, or developing mutual assistance networks. However, these alternatives require specific human resource management skills (Stup et al. 2006). Technical changes may also contribute to reducing workloads while maintaining or increasing milk production. For instance, switching from sugarcane cutting to maize silage reduces the time spent feeding cows (Ferris et al. 2006). While such a change does not involve much investment, it does require a certain degree of equipment and technical knowhow to ensure feed quality. Switching from twice to oncedaily milking also could directly impact work efficiency without entailing extra costs, but it would decrease cow productivity (Clark et al. 2006). The benefits of a milking machine (Wagner et al. 2001) remain to be verified since only two farms in the sample had one, and both had high milking times. Moreover, as farmers would require cash or credit to invest in such machines, their farms would become more tightly linked to the dairy industry.
Conclusions This study shows that on smallholder dairy farms which are quite similar in structure and modes of functioning, work task durations and time available vary greatly. Factors contributing
1277
to this variability include the structure of the farm (notably the size of herds and fodder crop areas), production system adopted (diversification of activities, cow feeding system), but also the position of the farmer in relation to work (attitude toward and acceptability of the workload). The combination of this set of factors renders the management of the workforce a process unique to each farm. In a context of growing demand for milk, it would be useful to develop support tools to help farmers consider how their work organization can be affected by strategic changes such as increasing herd size, changing the feeding system, or introducing new activities or equipment. A simulation tool based on the QuaeWork framework is currently under development to serve this need. Acknowledgments The authors would like to thank the 15 livestock farmers from Unaí-MG who participated in the study and Grace Delobel for translating the paper into English. This work was funded by the Agence Nationale de la Recherche under the Systerra Program: ANR08-STRA-10 (Ecological, technical and social innovation processes in Conservation Agriculture). Conflict of interest The authors declare that they have no conflict of interest.
References Altafin, I., Pinheiro, M.E.F., Valone, G.V., and Gregolin, A.C., 2011. Produção familiar de leite no brasil: um estudo sobre os assentamentos de reforma agrária no município de Unaí (MG). Revista UNI 1, 31–49. Astigarraga, L., and Ingrand, S., 2011. Production Flexibility in Extensive Beef Farming Systems. Ecology and Society 16, 1. Bernard, J., Le Gal, P.Y., Triomphe, B., Hostiou, N., and Moulin, C.H., 2011. Involvement of small-scale dairy farms in an industrial supply chain: when production standards meet farm diversity. Animal 5, 961–971. Clark, D.A., Phyn, C.V.C., Tong, M.J., Collis, S.J., and Dalley, D.E., 2006. A systems comparison of once- versus twice-daily milking of pastured dairy cows. Journal of Dairy Sciences 89, 1854–1862. Darnhofer, I., Bellon, S., Dedieu, B., and Milestad, R., 2010. Adaptiveness to enhance the sustainability of farming systems. A review. Agronomy for Sustainable Development 30, 545–555. Dedieu, B., 2009. Qualification of the adaptive capacities of livestock farming systems, Revista Brasileira de Zootecnia 38, 397–404. Ferris, C.P., Frost, J.P., Binnie, R.C., Patterson, D.C., 2006. Dairy cows performance and labour inputs associated with two silage feeding systems. Grass and Forage Science 61, 304–314. Guilhoto, J., Azzoni, C.R., Silveira, F.G., Ichihara, S.M., Diniz, B.P., and Moreira, G. R. In : PIB Da Agricultura Familiar: Brasil – Estados. 2011. http://ssrn.com/abstract=1803225 orhttp://dx.doi.org/10. 2139/ssrn.1803225. Accessed on 18 Apr 2015. Hemme, T., and Otte, J., 2010. Status and Prospects for Smallholder Milk Production. A Global Perspective. FAO-PPLFP, Rome, Italy. Hostiou, N., and Dedieu, B., 2012. A method for assessing work productivity and flexibility in livestock farms. Animal 6, 852–862. Lemaire, G., Franzluebbers, F., Carvalho, P., and Dedieu, B., 2014. Integrated crop–livestock systems: Strategies to achieve synergy between agricultural production and environmental quality. Agriculture, Ecosystems & Environment 190, 4–8.
1278 Martel, G., Dourmad, J.Y., and Dedieu, B., 2008. Do labour productivity and preferences about work load distribution affect reproduction management and performance in pig farms? Livestock Science 116, 96–107 Novo, A.M., Jansen, K., and Slingerland, M., 2012. The sugarcanebiofuel expansion and dairy farmers’ responses in Brazil. Journal of Rural Studies 28, 640–649. Sraïri, M.T., Bahri, S., and Kuper, M., 2013. Le travail et sa contribution aux stratégies d’adaptation de petites exploitations agricoles familiales mixtes d’élevage bovin/polyculture au Maroc. Biotechnologie, Agronomie, Société et Environnement 17, 463–474.
Trop Anim Health Prod (2015) 47:1271–1278 Stup, R.E., Hyde, J., and Holden, LA. 2006. Relationships between selected human resource management practices and dairy farm performance. Journal of Dairy Science 89, 1116–1120. Wagner, A., Palmer, R.W., Bewley, J., and Jackson-Smith, D.B., 2001. Producer satisfaction, efficiency, and investment cost factors of different milking systems. Journal of Dairy Science 84, 1890–1898. Wiggins, S., Kirsten, J., and Llambí, L., 2010. The Future of Small Farms. World Development 38, 1341–1348. Wright, I.A., Tarawali, S., Blümmel, M., Gerard, B., Teufel, N., and Herrero, M., 2011. Integrating crops and livestock in subtropical agricultural systems. Journal of the Science of Food and Agriculture 92, 1010–1015.
