PrOPara prject Sci-ecnmic impacts f alternative GIN cntrl practices Sheep & gat systems in France and Sctland Prject deliverable 11. (WP4) Sylvain Quiédeville, Simn Makes 21.09.2018 Research Institute f Organic Agriculture FiBL Ackerstrasse 113 Pstfach 219 5070 Frick Switzerland Phne +41 62 865 72 72 inf.suisse@fibl.rg www.fibl.rg
Acknwledgements: We wuld like t thank Hervé Hste, Yves Le Frileux, Felix Heckendrn, Lu Ralph; Spiridula Athanasiadu and Aileen Buchanan fr their assistance in develping the typical farms and rganising the fcus grups in Tuluse (INRA) and Aberdeen (SRUC). Furthermre, we thank Matthias Stlze (FiBL) fr his input in develping the prtcls. Cver pht surces: (left) Wikipedia, (right) www.scttish-blackface.c.uk.
Table f Cntents 1. Intrductin... 1 2. General methdlgy... 2 Farm mdel... 2 Wrkshp apprach... 2 3. Specific methdlgy & Results... 7 Gat system in France (Occitanie and Auvergne-Rhne-Alpes)... 7 3.1.1 Specific methdlgy... 7 3.1.2 Farmers bjectives and evaluatin criteria... 10 3.1.2.1 Tuluse regin... 10 3.1.2.2 Valence regin... 11 3.1.3 Ecnmic impacts f GIN cntrl practices... 14 3.1.4 Ecnmic impacts f ther new GIN cntrl practices (nt mdelled)... 16 3.1.4.1 Tuluse regin... 16 3.1.4.2 Valence regin... 16 3.1.5 Drivers and barriers t innvatin uptake... 17 3.1.5.1 Drivers... 17 3.1.5.2 Barriers... 23 3.1.6 Imprtance f recent GIN cntrl practices... 23 Sheep system in Sctland... 26 3.2.1 Specific methdlgy... 26 3.2.2 Farmers bjectives and evaluatin criteria... 28 3.2.3 Ecnmic impacts f alternatives mdelled... 29 3.2.4 Drivers and barriers t innvatin uptake... 30 3.2.4.1 Drivers... 30 3.2.4.2 Barriers... 37 3.2.5 Imprtance f recent GIN cntrl practices... 37 4. General Discussin & Cnclusin... 38 5. References... 41 Financial supprt fr this prject is prvided by funding bdies within the FP7 ERA-Net CORE Organic Plus, and with cfunds frm the Eurpean Cmmissin.
1. Intrductin Parasitism is well recgnized as a majr challenge t the health and welfare f rganic livestck. In rganic small ruminant prductin systems, endparasitic disease is accepted as the mst imprtant multifactrial syndrme, resulting in high negative effects n animal health, expressed by a lack f appetite, diarrhea, anaemia and in extreme cases, by death (Crwin 1997). In rganic cattle prductin, despite the rather lw stcking densities and use f imprved grazing management practices, helminth infectins are still a significant issue. These may alter the prcess f prductin as well as causing a decrease in the prductin level (meat and/r milk), the quality f feedstck, the daily weight gain and the reprductive rates; thus affecting ecnmic returns (Furichn, Seegers et al. 1999, Lpes, Niclin et al. 2015). The ecnmic return is als affected by higher management csts due t drenching, additinal labur and the implementatin f new techniques (Lpes, Niclin et al. 2015). These explain why it is necessary t assess ecnmic impacts f animal health management practices (Lpes, Niclin et al. 2015) and t quantify the expected cst benefit. Furthermre, since the 1950s, agricultural farms in Eurpe have changed in nature, frm a family type t larger businesses. This shift in the nature f farming implies a greater fcus n the ecnmic aspects (Fetrw, Cady et al. 2005). Mrris (1969) was a pineer in applying the cncept f marginal cst in veterinary decisin-making prcesses; he placed mre emphasis n the ecnmic dimensin, arguing that the inputs used fr disease cntrl shuld be increased t the level where the cst f an additinal input unit equals the supplementary value generated. Many ecnmic impacts studies have been cnducted in the last decades n animal disease (Bennett 1992, Dijkhuizen, Huirne et al. 1995, Huirne, Dijkhuizen et al. 1997). Hwever, several methds and criteria have been used s far, reflecting the farm system cmplexity with n straightfrward impact pathway as well as the absence f a wide cnsensus within the scientific cmmunity n hw t evaluate ecnmic impacts f animal diseases (Lpes, Niclin et al. 2015). The high cmplexity f a farm system requires an understanding f the whle system and nt nly individual cmpnents f it, resulting in a cmplex implementatin f impact assessment studies as well as the adptin f apprpriate disease cntrl strategies (Hwe and McInerney 1987). Since the prfit margin has becme mre critical in livestck systems in the last decades, it is f utmst imprtance t better understand drivers f prductin efficiency in relatin t parasite cntrl, in rder t take mre suitable decisins and ptimise prfits accrdingly (Lpes, Niclin et al. 2015). This reprt aims t deal with that challenge by lking at the ecnmic impacts f a selected range f alternative parasite cntrl strategies in small ruminants, namely gats and sheep. The reprt als lks at the scial drivers and barriers t the adptin f alternative practices. The first sectin presents the methds develped and used. Results are then presented in a secnd sectin and the reprt cncludes with a general discussin and cnclusin sectin. 1
2. General methdlgy The methdlgy is cmpsed f tw distinct parts. The first cncerns the farm mdel that was develped t estimate the ecnmic impacts f GIN cntrl practices. The secnd is n the structured apprach develped and implemented during the stakehlder wrkshps. Results frm a survey f rganic small ruminant farmers, tgether with an ex-ante analysis with the farm mdel are used during wrkshps, which als address scial factrs explaining the uptake and acceptance f GIN cntrl practices t cntrl parasites. Farm mdel A farm mdel was develped in rder t estimate the ecnmic impacts f current and alternative GIN cntrl practices. The farm mdel was designed t be flexible and allw parameter changes in a live setting such as the wrkshps. Farm mdels als allw a relatively lw cst methd f assessing the ptential impacts f management changes. Cntrlled experimentatins prvide mre detailed results, but require extensive financial and human resurces, therefre mdelling can help fcus research n mre prmising areas. The farm mdel develped and used in this study was cnstructed in Micrsft Excel (2013), allwing visualisatins and flexible data entry. Mdelling was restricted t the enterprise level, i.e. sheep r gats, and prvided a representatin f inputs (specifically, feed, GIN cntrl and labur) and utputs (milk and meat) t generate a grss margin per head figure. The mdel was specifically adapted t include precise figures fr meat and milk withdrawal perids, as well as the ability t estimate prductin lsses frm parasitism and variatins in labur input. Fr each system a typical farm descriptin was prvided by the scientific and extensin teams in the tw fcus regins in France and the UK. The typical farms cmprised an rganic gat system in France (typical in Occitanie and Auvergne-Rhône-Alpes Regins) and tw rganic sheep systems in Sctland (lwland and upland). Wrkshp apprach A structured wrkshp apprach was adpted t adequately address bth the scial and ecnmic factrs related t adptin f alternative GIN cntrl practices by farmers. T this purpse, we adapted the Structured Decisin Making (SDM) apprach cmmnly used fr decisins taking (Gregry and Keeney 1994, Cnry, Barker et al. 2008, Ogden and Innes 2009, Gregry 2012, Jhnsn, Eatn et al. 2015, Fatrić and Seekamp 2017). Althugh the purpse f this study was nt frmally abut taking decisins, the varying GIN cntrl ptins fr farmers, allws a similar apprach t SDM. Hwever, fr this purpse the final SDM step n decisin taking was nt cnsidered and was replaced by a general discussin n pssibilities t adpt the different ptins cnsidered and their drivers as well as n research needs and next plans. 2
Furthermre, the SDM apprach excels in finding and analysing alternatives t current practices, hwever, it des nt address factrs n innvatin uptake and farmers acceptance twards thse innvatins. In rder t better address these elements, we cmplemented ur apprach with theries n innvatin, namely the thery f innvatin diffusin by Rgers (Rgers 1995) and the thery f planned behaviur (Terry, Hgg et al. 1999, Armitage and Cnner 2001, Ajzen 2002). These theries have been extensively used empirically t the purpse f innvatin assessment (Sctt, Pltnikff et al. 2008, Talukder 2012). The adapted apprach based n SDM is iterative and cmpsed f 8 steps. The structured participatry wrkshp must invlve as least 4 t 6 farmers, 1-2 cnsultants/extensin fficers and up t 3 scientists (parasitlgist and ecnmist). One f the farmers shuld als be external t the prject t prvide a different perspective. Mre generally, a diversity f views n GIN cntrl practices, sustainability issues and ways f managing farms, must be reached t make the prcess mre reliable and rbust. The wrkshp prcess cmprises: First step: ratinale and bjectives f the wrkshp The first step cnsists f intrducing the wrkshp t the participants and cmprises f three key elements, namely (1) the presentatin f the bjectives t the participants, (2) the expsitin f the expected utcmes fr participants, and (3) asking feedback frm the attendants t clarify pints f cntentin and btaining agreement n the agenda f the wrkshp. Secnd step: t define stakehlders bjectives The secnd step aims at defining bjectives, in the same way as fr SDM. The gal is t make clear what matters fr the participants, what they want t achieve, in rder t better cnsider the alternatives afterwards. Fr the purpse f this study, the bjectives t crrespnd t the general gals f farmers regarding their business unit (e.g. maximising revenue). These bjectives are nt necessarily linked t the tpic f parasite cntrl (they can be if parasite cntrl is cnsidered as an issue n the farm) but will serve as a basis fr further reflectins. This step is structured in a way that examples f bjectives are first presented t the attendance, befre allwing farmers t think n their bjectives and discussing them in a plenary sessin. These bjectives are then srted t eliminate duplicates, and participants are finally asked t rank them accrding t their level f imprtance. This ranking is based n a swing weighting preferences apprach (e.g. Jacbs, Dysn, and Stcktn 2013), accunting fr the number f times each bjective is ranked first and cnverting this int a scre f imprtance (see Table 1). 3
Objective / Ranking (e.g. with 10 participants and 3 bjectives) 1 2 3 Scre f imprtance Objective 1 0 2 8 0 (0/10) Objective 2 3 5 2 0.3 (3/10) Objective 3 8 2 0 0.8 (8/10) Table 1 Example f table f preferences fr bjectives Third step: T transfrm bjectives int evaluatin criteria The third step cnsists f defining the criteria (maximum 4 criteria per bjective) t measure the level f fulfilment f the bjectives. A cnsensus between participants is nt necessary here; but each farmer shuld reflect n his/her wn criteria f imprtance, which will frm the basis fr cmparing perfrmance f different GIN cntrl practices. Furth step: T identify alternatives and tackle scial factrs In each wrkshp, the rganizers present up t five alternative GIN cntrls, in accrdance with results frm the general survey; but farmers have the pprtunity t cmment n these alternatives and identify ther nes they cnsider imprtant in relatin t their bjectives and evaluatin criteria. Then, in rder t tackle scial factrs explaining the uptake and acceptance f alternative practices t cntrl parasites, farmers are asked t fill ut a shrt questinnaire. The alternatives addressed are different depending n the system cnsidered (gat r sheep) and the regin r cuntry targeted: Gat system in France: - The Targeted Treatments (TT) and Targeted Selected Treatments (TST) prcedures; - The strategic use f anthelminthic treatments (Eprinmectin in cmbinatin with Levamisle); - The nn-access t pasture fr kids (up t ne year ld); - Changes in the pasture system (in general); - The use f biactive plants (including the use f Sainfin). Sheep system in Sctland: - Reduced stcking rate - Drenching part f the flck e.g. mre susceptible grups - Targeted selected treatment e.g. individual assessment & treatment - Increased prtein - Biactive feeds e.g. Sainfin, chicry The questinnaire cmprises clsed questins, mainly based n a Likert scale (Brwn, 2010). The Likert scale used is as fllws: strngly agree, agree, neither agree nr disagree, disagree, and strngly disagree. Table 2 specifies the questins addressed, which apply fr the five alternative practices mentined abve. 4
N Questin Pssible answer 1 When thinking abut the relative advantage f these different practices, d yu think that they are mre effective than yur current practices r past practices (if the practice has already been adpted)? Based n the Likert scale Please als specify whether yu adpted r nt these different practices If yu already adpted these practices, wuld yu say that yur decisin was influenced by the surrunding scial cntext (neighburs wh already adpted, etc)? 2 When thinking abut these different practices, wuld yu say that they are in line with yur persnal beliefs and values? 3 When thinking abut these different practices, wuld yu say that (a) They are easy t use/t implement? (b) They are easy t understand? 4 When thinking abut these different practices, wuld yu say that: (a) They can be tested withut requiring an extensive invlvement (capital, labur, training etc ) (b) They can be adapted/mdified t suit yur wn needs (c) They are nt necessary t adapt/mdify? 5 When thinking abut the bservability f these different practices, wuld yu say that evidence n their ptential benefits is available (t ensure a fair judgment f them)? 6 When thinking again abut these different practices, wuld yu say that (a) They are/wuld be useful in yur case? (b) Yu have already a lt f experiences n similar practices? Table 2 Questins asked t farmers n innvatin uptake Yes/N Based n the Likert scale Based n the Likert scale Based n the Likert scale Based n the Likert scale Yes/N Based n the Likert scale Based n the Likert scale Fifth step: T analyse ecnmic impacts f alternative GIN cntrl practices Similarly t the SDM apprach that reflects n cnsequences f alternatives, in this step we cnsider ecnmic impacts f varius GIN cntrl practices either already applied by sme farmers r envisaged fr adptin. Scial factrs are already reflected in the furth step and further in the next step. Factrs f uncertainties and risks are an integral part f the analysis. The discussin arund the ecnmic impacts f different GIN cntrl practices is mainly based n the farm mdel previusly discussed. The farm mdel cnsiders three main GIN cntrl practices (the same as thse f the previus step) but farmers are als given the pprtunity t discuss the ecnmic impacts f the ther practices they cnsidered t be imprtant at the previus step. 5
Sixth step: T cnsider trade-ffs This step aims at identifying trade-ffs between bjectives (in relatin t evaluatin criteria) and eventual implementatin f alternative GIN cntrl practices. These tradeffs are determined (1) n the basis f the evaluatin criteria specified in the third step, (2) n results and discussin f ecnmic impacts in the fifth step, (3) n results frm the shrt questinnaire addressed in the furth step, and (4) n barriers t adptin f different GIN cntrl practices. Result frm the shrt questinnaire (prcessed in the backgrund) are first presented and discussed. Then, general barriers t adptin are discussed, cnsidering the scial, ecnmic, envirnmental, as well as plitical and institutinal dimensins. On this basis, farmers are then asked t reflect n pssible trade-ffs between their bjectives (& related evaluatin criteria) and eventual implementatin f alternative GIN cntrl practices. Seventh step: T rank alternatives In this seventh step, the different alternative GIN cntrl practices are ranked by farmers in terms f their preferences (scre f imprtance). A scre f imprtance is calculated in the same way as fr the bjectives in secnd step. Eighth step: Cnclusin and feedback The purpse f this final step is t cnclude the wrkshp by making a shrt summary f the results, ask feedback abut them and mre generally n the wrkshp, and finally discuss the next steps in terms f research needs and s n. 6
3. Specific methdlgy & Results Gat system in France (Occitanie and Auvergne-Rhne- Alpes) 3.1.1 Specific methdlgy The typical farm used t mdel impacts f changes in the rganic gat system in France cmprised the fllwing characteristics: - Surface f 65 hectares under permanent grassland and grazed frest; - Herd f 120 dairy gats; - Prductin f apprximately 55,000 litres f milk per year, (458 litres/gat/annum); - Use f arund 600 grams f cncentrated feed per day per gat, cmprised f barley, maize, faba bean, and dehydrated alfalfa. T establish a mdelling baseline, we entered the practices that the typical farm applied fr parasites cntrl 5 years earlier. These cmprised systematic treatment with Fenbendazle (FBZ), 2 t 3 times per year. It must be emphasized that at that time there was a minimal withdrawal perid fr milk f 2 days. Hwever, since the regulatin was changed, the baseline nw als takes accunt f the current rules, with a withdrawal perid f 12 days fr rganic milk (the withdrawal perid is nly 6 days in cnventinal). The GIN cntrl practices nw applied by the typical farm were mdelled as the alternative GIN cntrl, which the farm mdel estimated ecnmic impacts: - The develpment f Targeted Treatments (TT) and Targeted Selected Treatments (TST) prcedures, that includes a gap between primiparus and multiparus prducing gats; - The annual use (n average) f 2 Eprinmectin treatments during lactatin, with a withdrawal perid f 2 days fr rganic milk. - The use f Levamisle during the dry perid. - N access t pasture fr gat kids up t ne year ld. Furthermre, the typical farm is interested t use Sainfin t better cntrl parasites but culd nt test it s far because given the limited availability f dried Sainfin pellets n the rganic market. Althugh this practice is nt applied yet, we cnsidered it in the mdelling since it might be an interesting perspective fr farmers. The mdelling f the ecnmic impacts f the alternatives mentined abve as cmpared t the baseline (riginal situatin) cmprised six elements: (1) the cst f drenching, (2) the cst f the milk withdrawal perids, (3) the labur cst, (4) the turnver fr milk 7
(linked t the prductin level), (5) the cst f feed (linked t changes in the ratin), and finally (6) the grss margin (including direct labur cst). Figure 1 shws the mdel interface and sme f the parameters tested. Traitement anthelminthique Référence de base Optimal (évitant l'apparitin de résistance) Matière active Fenbendazle 10% Eprinmectin 0.5% Levamisle 3% (tarissement) Dse (mg) 10 mg/kg 1 mg/kg 12 mg/kg Dse (ml) 1 ml/10kg 2 ml/10kg 4 ml/10kg Cût 0,03267 /ml 0,07268 /ml 0,01408 /ml Myenne. pids (Kg) ml/chèvre /dse ml/chèvre /dse ml/chèvre /dse 65 Chèvre adulte 6,5 0,21 13 0,94 26 0,37 60 1ère lactatin chèvre 6 0,20 12 0,87 24 0,34 80 Buc 8 0,26 16 1,16 32 0,45 50 Chevreau (8-12mis) 5 0,16 10 0,73 20 0,28 15 Chevreau (3-7mis) 1,5 0,05 3 0,22 6 0,08 Dses ml Dses ml Dses ml Chèvre adulte 3 19,5 0,64 1,5 19,5 1,42 1 26 0,37 1ère lactatin chèvre 3 18 0,59 1,5 18 1,31 1 24 0,34 Buc 3 24 0,78 1,5 24 1,74 1 32 0,45 Chevreau (8-12mis) 1 5 0,16 0 0 0,00 0 0 0,00 Chevreau (3-7mis) 1 1,5 0,05 0 0 0,00 0 0 0,00 Traitement Cût ttal Traitements 81,23 178,23 46,02 224,25 Retrait du lait (jurs) 16 2 Péride tarrissement 1 Retrait du lait (adultes) 125 Référence 5991 159 Référence 478 1 Prd Lait/jur (1ère lac) 18 844 22 67 Lait retiré Cût ttal Lait 6835 545 0 6917 724 46 Cûts cmbinés 6917 770 Figure 1 Illustratin f the interface f the farm mdel Cst f drenching The dse used at each applicatin, per 10kg f animal weight, is respectively f 1, 2 and 4ml fr Fenbendazle (baseline), Eprinmectin, and Levamisle (farmacy.c.uk, 2018). The cst f each prduct as well as the average weight f the gats (distinguishing adult gat, first lactatin, billy, dairy KIC and kids) was taken respectively frm (farmacy.c.uk, 2018) and Agridea (2017). The mdel then calculates the cst per dse fr each type f gat. Then, the mdel specifies the number f dses each type f gat is receiving a year (herd average), and an verall annual cst is calculated fr each type f gat and fr altgether. Cst f milk withdrawal perids The cst f the milk withdrawal perid depends n the number f withdrawal days, the milk price and the level f prductin. The withdrawal perid in rganic gat system is currently f 16 days fr Fenbendazle and 2 days fr Eprinmectin (Hste, 2018). The withdrawal perid fr Levamisle is 28 days (Hste, 2018) but is nt cnsidered in the mdel as it is nly used in the dry perid and generally at the very beginning f it (because the withdrawal perid is very lng). Since n milk is prduced during that perid, n financial are be accunted as regard f the withdrawal perid. The mdel then calculates the verall annual cst due t the milk withdrawal perids. 8
Labur cst T accunt fr the cst f labur, the average annual number f hurs per wrker was defined as 2400 hurs and then divided by the number f livestck. The assumed labur cst per hur was 11.26 in France based n statistical data frm Agreste (2018). The mdel calculates the number f wrking hurs per gat as an indicatin. T calculate ecnmic impacts f changes in the labur requirement caused by the implementatin f mre labur intensive practices, the additinal labur can be specified as a percentage. In rder t present a metric that was easier t assess, the mdel calculates hw many minutes f additinal wrk per day n the farm. Milk incme In the mdel, the milk incme directly depends n the prductin level, which can differ depending n the practices being adpted. The weight f the gats and the effect f parasites are tw imprtant elements that are cnsidered. The weight (and specifically the liveweight gain) f yung gats is cnsidered since the fact f keeping kids indr until they are ne year ld (a practice assessed) can influence their grwth and thus their 1 st lactatin prductin level (Alberti et al. 2012). Changes in grwth is expressed as a percentage difference and the weight at 1 st kidding is adjusted accrdingly. Milk yields reflect reality n the mdelled farm, with a lwer yield assumed fr 1 st kidders, related t bdyweight. Therefre the mdel reflects a lwer 1 st year milk yield with a lwer bdyweight at kidding due t parasitism. Fr the effect f parasites n milk prductin, a cefficient f efficiency is applied t bth adult gats and first year dairy gats. A cefficient f 100% means there is n infectin r at least n immediate impact n prductin frm parasites; while a cefficient f 70%, fr instance, indicates a decrease f 30% in the prductin level due t parasites. We assume a cefficient f efficiency f 85% in the baseline (when using Fenbendazle). Cst f feed In the baseline, we assume the use f 600 grams f cncentrates per gat annually with a cst f 436 per tn, based n data frm the typical farm. As per the ther key variables and the feed used and price changes are expressed in a percentage relative t the baseline. Grss margin (including labur cst) The grss margin is a simple difference between the milk turnver and bth variable csts (cst f drenching, cst f feeds) and labur csts. 9
3.1.2 Farmers bjectives and evaluatin criteria Tw fcus grups were held in France in Spring 2018; ne in Tuluse, Occitanie and ne in Valence, Auvergne-Rhne-Alpes. Participants were primarily farmers, tgether with 1-2 extensin staff and 1-2 scientists wrking in the field f GIN cntrl. 3.1.2.1 Tuluse regin The general bjectives expressed by rganic dairy gat farmers in the regin f Tuluse are (1) maximise the revenue, (2) reduce envirnmental impacts, (3) enhance the health f the herd, and (4) ensure the resilience f the system and better manage pastures. Table 3 reprts the relative imprtance farmers gave t these bjectives, and specifies the evaluatin criteria and main cmments given. The mst imprtant bjectives f farmers were ecnmics (maximising the revenue and better resilience), cnfirming ur interest in assessing the ecnmic impacts f the alternative practices. It is als expected that farmers will favurably cnsider alternatives with direct and clear psitive ecnmic impacts. N Objectives Scre f imprtance Criteria f evaluatin 1 Maximising 0.67 Net margin ecnmic Cncentrates expenses results Investment and depreciatin amunt 2 Better resiliency and pasture management 0.67 Quantity f cncentrates, hay and frages bught externally Quantity f cncentrates used per litre f milk prduced Decrease in the number f infectins per year? (in relatin t the quality f pasture) Cmments (actrs statements) Optimizing the pasture allws buying fewer hay and thus t have a better resilience f the system T me, the ptimizatin f the pasture invlves a similar quantity f milk prduced while reducing the quantity f cncentrates used What d we call ptimisatin f pasture? If it s agrnmic, this is the prductin f herbs [ ] and n the ther side there is practices limiting the infestatin I disagree, this is bth in gat system (prductin and limited infestatin) 10
3 Reducing envirnmental impacts 4 Amelirating the herd health 0.33 The nn-systematic use f prducts Adptin f practices limiting the resistance The abandn f classical treatments 0 Signs f caprine arthritis encephalitis virus (CAEV) Rate f mrtality Cull rate due t health issues Overall health status (thinness, etc) Veterinarian expenses Table 3 Farmers bjectives and evaluatin criteria in Tuluse regin But I dn t knw (abut the criteria: decrease in number f infestatins) T find an equilibrium in all f this 3.1.2.2 Valence regin The general bjectives expressed by rganic dairy gat farmers in the regin f Valence are t (1) ensure viability and decent ecnmic results, (2) have an ergnmic wrking place and ptimised labur wrkfrce, (3) limiting envirnmental impacts, (4) fd autnmy and (5) limiting parasite pressure. Table 4 reprts the imprtance granted by farmers t these bjectives and als specifies the evaluatin criteria and main cmments made. The mst imprtant bjectives fr farmers are f ecnmic nature. Even the secnd bjective, which is mre n scial aspects at first sight, invlve the net revenue as a criteria f measurement. These elements reinfrce the interest f assessing the ecnmic impacts f GIN cntrl practices. 11
N Objectives Scre f imprtance Criteria f evaluatin Cmments (actrs statements) 1 Ensuring viability and decent ecnmic results 0.4 Revenue As I just tk ver a farm this is very imprtant. [ ] The balance between prductin lss and cst f the prducts, we d nt really questin it (given the lw cst f prducts) Pasture management Ecnmically this is k if I use pastures well Labur efficiency We have an imprtant cnstraint n labur s that labur efficiency is imprtant Mechanisatin csts We try t limit mechanisatin csts. [ ] we wrk with a CUMA 2 Ergnmic wrking place, ptimised labur wrkfrce and animal welfare 3 Limiting envirnmental impacts 4 Fd autnmy 0.4 Labur quantity These are details maybe Drudgery f wrk and wrkplace adjustment but this is an investment in labur making us happier when ging t the wrk Net revenue Hw t measure animal welfare? 0.2 Number f treatments, and dsage applied Mlecule txicity 0.2 Quantity f cncentrates, hay and frages bught externally Hw shall we evaluate the welfare, is it t put animal utside r nt? Nbdy has the answer whether utside is better Ok but when I pen the dr they g ut, they g in frests, etc 12
5 Limiting parasite pressure 0 Visual aspect (clur f eyelid, general state, raised hairs r nt, etc) Faecal tests Cst f treatment Decrease in milk prductin Table 4 Farmers bjectives and evaluatin criteria in Valence regin It is a risk t take accunt f raised hairs, when we thrw milk away, we thrw everything r nt [ ] s I find Targeted Selected Treatments are nt that evident Results are very variable depending n the gat, the lab, etc We are nt sure if the sample sent will arrive the day after r 15 days later This is als a questin f interpretatin. Sme veterinarians are stricter than thers 13
3.1.3 Ecnmic impacts f GIN cntrl practices Cst f drenching The ttal cst f drenching in the baseline (cst f the prducts used) was calculated t be 79.10 fr the herd, (using Fenbendazle). There is a nn-negligible increase in that cst when using alternative anthelminthic treatments. The use f Eprinmectin and Levamisle fr the whle herd csts respectively 170.32 and 43.98, thus a ttal cst f 214.31. Accrding t the mdel, the implementatin f the alternative drenching practices thus increases the treatment cst by 135.21 eurs n an annual basis fr the whle herd. But since this cst difference applies n the whle herd, it remains a very small amunt at farm level. Participants in the tw wrkshps agreed n this cst difference between the GIN cntrl practices that were applied 5 years ag and thse applied since tw years n the typical farm. The participants agreed n the number f dses prpsed in the mdel: 1.5 dses f Eprinmectin and 1 dse f Levamisle fr adult, first lactatin, and billy gats (alternative practices); as cmpared t (reference) 3 dses f Fenbendazle fr adult, first lactatin, and billy gats as well as 1 dse f Fenbendazle fr dairy KIC (8-12mnths) and kids (3-7mnths). Cst f milk withdrawal perids The withdrawal perids are fixed (regulatin), but since the rules ften changed in the last few years, there was a questin mark as regard f the reference t take int accunt fr the calculatin. Indeed, the milk withdrawal perid after using Fenbendazle in rganic gat systems was f 2 days 5 years ag but is currently at 16 days. It was stressed by the wrkshps attendants that the current withdrawal perid shuld be used as a baseline althugh they als stressed that Fenbendazle shuld nt be used anymre because f resistance prblems. If Fenbendazle was currently used fr gat systems in France, this wuld imply an annual cst f 8,215 due t the milk being withdrawn. The current use f Eprinmectin has an annual milk withdrawal cst f 521. Thus, the annual cst assciated t milk withdraw decreases by 7,694 when implementing the recent GIN cntrl practices selected. Wrkshps participants had n specific cmments r cncerns n this result. Labur cst It was hypthesised that implementing the recent GIN cntrl practices implies additinal wrk, mainly due t the need t bserve gats in a mre individual way. In the baseline, it was assumed that tw full-time emplyees, wrking each 2 400 hurs a year n the farm, represents an annual cst f 54 572. When implementing the new GIN cntrl practices, it was hypthesized that an increase f 5% in the number f hurs required being tested. This represents extra wrk f 39 minutes per day and an additinal annual cst f 2 734 (ttal cst f 57 406 ). The participants stressed that 39 minutes f extra wrk per day is prbably an verestimatin. 14
It was highlighted that there is prbably a need t wrk 5% mre but that the reference f 2 400 hurs a year per wrker is prbably t high. We reduced the reference t 1 800 hurs, reducing the extra daily wrk t 30 minutes. This result was fine fr the attendants. With this new hypthesis, the labur cst fr the baseline decreases t 41 265. The labur cst fr the recent GIN cntrl practices als decreased t 43 328. With this new hypthesis, the labur cst difference decreases t 2 063 (as cmpared t 2 734 ). Mrever, it must be emphasized ne farmer managing a particular system (nt the ne tested in the mdelling) estimated an additinal wrk f half a day per mnth as cmpared t his ld system. In fact, this farmer uses electric fences in pastures t perate rtatinal grazing, implying significant wrk fr the installatin and shifting f fences. Milk incme Tw elements were discussed with respect t the milk turnver: pssible changes in the grwth f gats (and thus in their end weight) and in the effect f parasites (the efficiency ), ptentially impacting the level f prductin and thus the milk turnver. Fr the wrkshp, it was hypthesised that there wuld be n change in the grwth nr in the efficiency. When participants were asked whether keeping kids indr favurs their grwth, they answered that there was n clear evidence. One farmer even said that keeping kids indr is wrse because the first time they g utside their immune system is less well develped and they are thus mre affected by parasites. This als answers the secnd questin abut changes in the efficiency: there are als n clear evidence n this and the level f prductin might even decrease in certain places. Participants stressed that there are many factrs influencing the level f prductin and that it is therefre difficult t islate ne factr r anther. Nne f the farmers clearly bserved an increase in the prductin level. Hwever, ne farmer mentined that n average (when using Fenbendazle), 5 t 6 gats were remved annually with the past system while nly ne is remved per year currently. That said, the ther farmers did nt ntice any significant difference. Therefre, accrding t stakehlders, the milk turnver is nt affected by the adptin f the recent GIN cntrl practices being tested here, and n changes in the mdel are required. Cst f feed In the mdel, we hypthesised n change in the quantity f cncentrates used. Fr the pssible adptin f Sainfin, we hypthesised an increase f 5% in the cst f cncentrates, thugh this is estimated due t a lack f reliable infrmatin. Sainfin is indeed mre expensive given its limited availability n the rganic market. The pssible use f Sainfin was nt reflected n in-depth at the wrkshps but farmers believe there is n clear ecnmic impact n the prductin system. We kept the hypthesis frmulated as there was n clear disagreement frm farmers. 15
Grss margin With the implementatin f the recent GIN cntrl practices being tested (and assuming a withdrawal perid f 16 days fr Fenbendazle), the annual grss margin n the typical farm increases by arund 4 918 (41 /gat). This is mainly due t savings frm fewer quantity f milk withdrawn. There is als a slight increase in the cst f feed due t the use f Sainfin but the latter culd nt be reflected in-depth in wrkshps. 3.1.4 Ecnmic impacts f ther new GIN cntrl practices (nt mdelled) 3.1.4.1 Tuluse regin The stakehlders highlighted tw ther GIN cntrl practices they cnsider imprtant as alternatives t anthelminthic treatments: (1) the genetic selectin fr parasites resistance as well as (2) reseeding f pastures at times in rder t limit infestatin. The alternative practice changes in the pasture system, addressed in the questinnaire but nt mdelled, was nt cnsidered here as it is a very general practice (ecnmic impacts might vary cnsiderably depending n the specifics). It was said it is difficult t assess the impacts f these tw practices withut knwing much n the situatin f reference. That said, ne farmer stated that genetic selectin fr parasites resistance is nt necessarily mre expensive: why wuld it be mre expensive as instead f selecting n the prtein cntent [f the milk], we select n parasites [resistance]. Anther farmer said that reseeding pastures is bviusly mre time cnsuming (tillage, etc). 3.1.4.2 Valence regin The stakehlders highlighted 5 ther GIN cntrl practices they cnsider imprtant as alternatives t anthelminthic treatments: (1) the genetic selectin fr parasite resistance (2) the eliminatin f infected gats, (3) limiting the width f the passages t 2.50 meters t avid gats staying lng and building a parasite reserve, (4) using essential il (e.g. strng dsage f garlic) and (5) using fresh ak leaf. It was said that it is difficult t assess the impacts f these practices. Farmers highlighted that these GIN cntrl practices are mre preventive than curative s that it is very difficult t assess the difference. Als n faecal egg measurement was undertaken befre and after, hwever farmers envisage such testing culd be dne and they call fr systematic experiments. Hwever, the high cst f such experiments was mentined, as they need t be meticulusly implemented and cntrl grups must be used. Mst f these alternatives limit the level r risk f infestatin but are nt intended t eradicate r significantly decrease parasite strains. This particularly applies fr the use f essential ils and ak leaf. One farmer said ne year we had a lt f hassles with parasites. 16
We tried essential ils that stabilized parasitism but this is nt curative. Where essential ils was nt used, the parasitism level cntinued t increase and I had t undertake chemical treatments. 3.1.5 Drivers and barriers t innvatin uptake 3.1.5.1 Drivers Highlights frm the questinnaires cmpleted by farmers at the tw wrkshps are shwn belw. There were n majr differences between the tw wrkshps, s the data was merged. In a secnd sectin, detailed results are shwn. Highlights: Questinnaire results N access t pasture fr kids is nt mre efficient than current (r previus) practices accrding t 3/5 f the farmers. Fr the ther strategies, frm 6 t 7 ut f 7 farmers agree r strngly agree that they are mre efficient than their current (r ld) practices. 5/5 farmers said the practice nn-access t pasture fr kids is nt in line with their beliefs and values. This practice is als nt easy t understand fr 4/7 f the farmers (cmpared t 0 t 1 ut f 9 farmers fr the ther practices); still, nly 1/5 f the farmers agree that this strategy can easily be adapted t suit their system (cmpared t 6 t 8 ut f 6 t 8 farmers fr the ther strategies). The adptin f the practice nn-access t pasture fr kids is nt influenced by the surrunding scial cntext accrding t 4/6 f the farmers. By cntrast, frm 5 t 6 farmers ut f 6 t 7 agree r strngly agree that TST and the strategic use f anthelmintic are influenced by the surrunding cntext (these tw practices are actually adpted the mst). 6 t 7 ut f 7 t 8 farmers agree r strngly agree that TST and strategic use f anthelmintic are easy t use. This is nly 3/7 fr nn-access t pasture fr kids, 3/8 fr changes in pasture system (with 4 farmers disagreeing), and 3/7 fr the use f biactive plants (with 2 farmers disagreeing). 6/8 f the farmers disagree r strngly disagree that changing the pasture system is a strategy that can be easily tested withut requiring an extensive invlvement (labur, resurces ). 5/6 t 6/6 f the farmers affirm t have already many experiences n similar practices as TST and the strategic use f anthelmintic, which might partly explain the higher adptin fr these tw practices. Detailed results frm the questinnaire Figure 2 t Figure 11 shws the results frm the questinnaire cmpleted by farmers in the wrkshps. 17
Figure 2 Effectiveness f alternative GIN cntrl practices Figure 3 Scial cntext and alternative GIN cntrl practices 18
Figure 4 Persnal beliefs & values and alternative GIN cntrl practices Figure 5 Practical use f alternative GIN cntrl practices 19
Figure 6 Understanding f alternative GIN cntrl practices Figure 7 Alternative GIN cntrl practices and required invlvement 20
Figure 8 Adaptability f alternative GIN cntrl practices Figure 9 Evidences n benefits f alternative GIN cntrl practices 21
Figure 10 Usefulness f alternative GIN cntrl practices Figure 11 Experiences n similar practices 22
3.1.5.2 Barriers General ecnmic and scial barriers t innvatin uptake were expressed by farmers in the wrkshps. These are as fllws: Ecnmic: Scial: The ecnmic results maybe, if we take accunt f the number f wrking hurs (ne farmer) Partly, this is starting frm unknwn, we take a risk that is nt measurable (ne farmer) I did nt adpt the strategy f Targeted Selected Treatment as I just have taken ver a farm and the ecnmic risk is t high [ ]; systematic treatments are less risky (ne farmer) The additinal time (ne farmer) Farmers are mre and mre trained while technicians have less time t d s, therefre we d nt make substantial prgress; there is insufficient infrmatin n research, n what is ging n (ne farmer); anther farmer expressed a similar view The interpretatin f lab results n faecal samples (number f eggs) is smetimes very different between veterinarians and we d nt knw where t stand (ne farmer). There is a prblem t cnserve faecal samples; we tested small bags f silica t absrb xygen in the samples, which wrks quite well (an adviser) If when we g t the veterinarian, he/she des nt knw what a gat is, this is annying (ne farmer) 3.1.6 Imprtance f recent GIN cntrl practices Table 5 and Table 6 indicate the scre f imprtance f the different GIN cntrl practices cnsidered in Tuluse and Valence regins respectively, based n the ranking made by farmers. We can see that changing the pasture system was cnsidered as the mst imprtant GIN cntrl practice as alternative t anthelminthic treatment. The practice genetic selectin btained an imprtance scre f 0.33 in Tuluse but was nt acknwledged in Valence. It is interesting t see a certain cherence between farmers bjectives (and related evaluatin criteria) and their ranking f the alternative practices. The mst imprtant bjectives f farmers are ecnmic in nature and all GIN cntrl practices withut clear ecnmic benefit were nt cnsidered well here (e.g. nn-access t pasture fr kids). This ranking can als be partly explained by the different barriers and scial factrs t innvatin uptake fr the different alternative practices. GIN cntrl practices that were nt acknwledged well in the questinnaire are als nt well cnsidered here. This is particularly the case fr the practice nn-access t pasture fr kids, which faced quite 23
sme limitatins: (1) prblem f efficiency; (2) nt in line with farmers values and beliefs; (3) nt easy t understand; (4) lw level f adaptability; (5) lw influence f the scial cntext fr adptin; and (6) nt easy t implement. Furthermre, the strategy f changing pasture system was cnsidered by farmers as nt easy t test withut an extensive invlvement. But this practice was ranked first (largely), meaning that labur requirement is nt directly a very imprtant issue fr mst farmers (it is an issue when additinal peple have t be hired). The ranking was mainly driven by future perspectives in terms f ecnmic gains. The level f risk invlved is als an issue as it was raised t be a barrier fr 2 farmers ut f 9. N Alternatives Scre f imprtance Cmments 1 Changes in the 0.67 Seen as a ptential fr the future pasture system 2 Genetic selectin fr 0.33 parasites cntrl 3 Targeted treatments (TT) and targeted selected treatments 0 Cnsidered as the current standard, hwever tw farmers ranked it in secnd place (TST) 4 Strategic use f 0 anthelminthic treatments 5 Nn-access t pasture 0 Unclear impacts fr kids 6 Using biactive plants (including Sainfin) 0 N scientific evidences yet Table 5 Imprtance f GIN cntrl practices in the Tuluse regin 24
N Alternatives Scre f imprtance Cmments 1 Changes in the pasture 4 system 2 Using biactive plants 0 Tw farmers ranked it in secnd place (including Sainfin and ak leaf) 3 Eliminatin f infected 0 Tw farmers ranked it in third place gats 4 Genetic selectin fr 0 parasites cntrl 5 Targeted Treatments 0 (TT) and Targeted Selected Treatments (TST) 6 Strategic use f 0 anthelminthic treatments 7 Nn-access t pasture 0 fr kids 8 Using essential ils 0 9 Waiting area f 2.50 meters 0 Table 6 Imprtance f GIN cntrl practices in the Valence regin 25
Sheep system in Sctland 3.2.1 Specific methdlgy In Sctland, tw typical farms were used t mdel impacts f changes in bth upland and lwland beef & sheep farm. They present the fllwing characteristics: Upland beef & sheep system - 50 ha f cereals - 150 ha f rtatinal frage (grass & rts) - 890 ha f hill grazing fr sheep - 900 ewes and finishing lambs (100% lambing) - 140 suckler cws and finishing beef Lwland beef & sheep system - 50 ha f cereals - 150 ha f rtatinal frage (grass & rts) - 100 ha f hill grazing fr sheep (and 100 ha f envirnmental areas) - 400 ewes and finishing lambs (175% lambing) - 70 suckler cws and finishing beef T establish the baseline f the mdelling, fr each f these systems, we cnsidered the practices the typical farms applied fr parasite cntrl 5 years earlier. These were an annual applicatin f 2 dses with white (benzimidazle) drench per ewe as well as 2 t 3 treatments fr lambs. The GIN strategies that are currently r culd be applied in the near feature in these tw different systems (and f which the farm mdel estimates their ecnmic impacts) are: - Reductin f the stcking rate - Drenching part f the flck e.g. n the mst susceptible animals - Targeted selected treatment (TST) e.g. individual mnitring and drenching - Increase prtein level fed (particularly fr ewes at lambing) - Use f biactive feeds like Sainfin r Chicry The mdelling f the ecnmic impacts f the alternatives mentined abve was cmpared t the baseline (riginal situatin) cmprising six elements: (1) the cst f drenching, (2) the labur cst, (3) the prductin effect, (4) the cst f feed (linked t changes in the ratin), and finally (5) the grss margin (including the labur cst). Figure 12 is an illustratin f hw the interface f the mdel, where changes in parameters can be tested, lks like. 26
Anthelmintic Treatments Baseline Optimal Chemical name Panacur 10% Panacur 10% Dsage fr sheep 5 mg/kg 5 mg/kg Prduct dse 0,55 ml/10kg 0,55 ml/10kg Cst 0,032936 /ml 0,032936 /ml Avg. bdweight (Kg) ml/sheep /dse ml/sheep /dse 70 Breeding ewe 3,85 0,13 3,85 0,13 80 Rams 4,4 0,14 4,4 0,14 50 Ewe lambs 2,75 0,09 2,75 0,09 35 Lambs 7mths 1,925 0,06 1,925 30 Lambs 5mths 1,65 0,05 1,65 25 Lambs 3mths 1,375 0,05 1,375 0,05 Dses ml Dses ml Breeding ewe 918 2 7,7 0,25 918 1 3,85 0,13 Rams 26 2 8,8 0,29 26 1 4,4 0,14 Ewe lambs 206 1 2,75 0,09 206 0 0 0,00 Lambs 7-8mths 877 1 1,925 0,06 877 0,5 0,9625 0,00 Lambs 5-6mths 916 1 1,65 0,05 916 1 1,65 0,00 Lambs 0-4mths 964 1 1,375 0,05 964 0 0 0,00 Anthelmintic tr Ttal cst f treatments 408 120 Ewe prductivity effects Baseline Optimal (efficiency thrugh targeted GIN cntrl) Ewes/ha 2,5 0,369846 Sheep LU/ha 2,5 100% 0,369846 Sheep LU/ha Lambing % 105% 964 Lambs 105% 100% 964 Lambs Lambs 0-4mths 3,5 3,5 Weaning weight 22,0 0,15 DLWG 22,0 0,15 DLWG Lambs weaned % 95% 916 Lambs 95,00% 100% 916 Lambs Figure 12 Screensht f the interface f the farm mdel Cst f drenching The dse used at each applicatin, per kg f animal weight, is respectively f 5mg (farmacy.c.uk, 2018). The cst f each prduct as well as the average weight f the sheep (distinguishing breeding ewe; rams; ewe lambs; lambs f 3, 5 and 7 mnths) was taken frm farmacy.c.uk (2018) and SAC Cnsulting (2017), respectively. The mdel then calculates the cst per dse fr each type f sheep and the number f dses each type f sheep receives n average per year, and subsequently the verall annual cst per class f sheep and in ttal. Prductin effect The prductin effect is mainly fcused n the ewes. The mdel cmputes the number f ewes per hectare n average with the lambing percentage, thus prviding the number f lambs frm 0 t 4 mnths. We assumed identical values fr bth lwland and upland system. The mdel effectively tries t identify, if we dse less, whether the numbers f lambs brn and then weaned is likely t be very different. If we dse less, it is likely that the ewes prduce less milk because f mre parasites. There wuld be, in turn, increased mrtality in the lambs and a lwer grwth rate. This wuld then impact the incme frm lambs. The mdel assumes n difference between the baseline (ld system with mre treatments) and the new system. Stakehlders were asked in wrkshp t specify their views n this. 27
Labur cst Labur csts per hur were assumed n the same basis as the gat farms, except fr a wage rate f 10.26 in Sctland based n SAC Cnsulting (2017). Cst f feed In the baseline, neither farm fed cncentrates t their sheep. Grss margin (including labur cst) The grss margin is a simple difference between the turnver and bth variable csts (cst f drenching, cst f feeds) and labur csts. 3.2.2 Farmers bjectives and evaluatin criteria A single fcus grup was held in Sctland, UK in spring 2018 at the SRUC Aberdeen campus. Participants were primarily farmers, tgether with 1-2 extensin staff and 1-2 scientists wrking in the field f GIN cntrl. The general bjectives expressed by rganic sheep farmers in Sctland are; (1) t increase the farm ecnmic viability, (2) have a resilient system, (3) t ptimise the wrklad and (4) have an envirnmentally balanced system. Table 7 reprts the imprtance given by farmers t these bjectives and als specifies the evaluatin criteria and main cmments prvided. The mst imprtant bjectives f farmers are the ecnmic viability and system resilience, reinfrcing the interest f assessing the ecnmic impacts f the different GIN cntrl practices. We als expect that farmers will better cnsider alternatives with direct and clear psitive ecnmic impacts. N Objectives Scre f imprtance Criteria f evaluatin Cmments 1 Ecnmic viability 2 Resilient system 3 Optimising wrklad 0.4 Revenue / Prfit 0.4 Lng term prductivity / Adaptability; Diversified / Mixed enterprises; Lwer risks; Preventative healthcare 0.2 Number f wrking hurs I want as much preventative measures as pssible. I d nt enjy drenching s I have been ding sme selective drenching T diversify as well, as I m struggling having enugh clean pastures 28
4 Envirnmen tally balanced system 0 Minimising the use f inputs; adaptatin t climate change Table 7 Farmers bjectives and evaluatin criteria in Sctland I am advisr My primary bjective is t have enugh time Earning enugh t live but nt maximising revenue 3.2.3 Ecnmic impacts f alternatives mdelled Cst f drenching In the lwland system, the ttal cst f drenching (cst f the prducts used) calculated with the farm mdel was 221 (per flck) fr the reference (use f fenbendazle) and 210 fr the tested practice (use f Cydectin). Fr the upland system, the ttal drenching cst fr the reference and the tested practice was 408 and 120, respectively. Participants in the wrkshp had n specific cmments n this but mentined that the additinal cst is actually mre with the labur. Prductin effect In terms f the stcking rate, the Scttish farmers said that reducing the stcking rate is maybe nt the right thing t d fr lwering the number f dses applied. It was reprted that keeping the grass at the right stage fr grazing is the mst imprtant and that it s better t achieve an ptimal stcking rate (rather than reducing it) that allws keeping a gd grass quality fr the year. One farmer highlighted that if we assume the system is ptimised fr using clean grazing, it shuld be pssible t achieve the same prductin level by treating lambs nce instead f twice. Anther farmer said he has pretty clean grazing and that he nly applies ne dse n average. Hwever, ne farmer said he is struggling t keep the grass clean and that is trying chicry with sme sheep t see the effect. But the farmers had difficulties t estimate the effect f reducing the dsage fr lambs given the high cmplexity f the system and the need t understand all the cmpnents f it. The cntrl f anther pasture parasite, liver fluke can als influence GIN cntrl strategies as sme prducts will treat bth issues. We thus still assume n effect n the prductin level. Labur cst It was hypthesised that implementing the alternative GIN cntrl practices tested des nt imply additinal wrk requirements n farm, either fr lwland r upland systems. In the baseline, it was assumed 2 full-time emplyees, wrking each 2 400 hurs a year n the farm, representing an annual cst f 43,200. The wrkshp attendants highlighted that the mre targeted apprach t treatments d require mre skills and time, althugh they encuntered difficulties in estimating the difference with a cnventinal system. Furthermre, ne farmer reprted t have invested in an autmatic 29