ATHANASE YOUAN BI1, DAO DAOUDA1, LEHMANN BERNARD2, DUMONDEL MICHEL2
1University of Cocody, Swiss Center for Scientific Research, Ivory Coast
2Swiss Federal Institute of Technology (ETH), Agrifood and Agrienmvironmental Economics Group, Switzerland
In 2001, Ivory Coast counted 1 440 000 bovines, and this production is not sufficient to cover national consumption. Its performance does not make it possible to face external competition. In order to determine and to compare the various systems of bovine stock breeding in terms of efficacity, a study was carried out in the north (Korhogo) and the centre (Toumodi). Ninety five (95) bovine stock breeders were implied.
A classification of stock breeding systems was made according to management criteria, livestock size and individual characteristics' of the stockbreeders. Then, we used the data envelopment analysis (DEA) method to determine their efficiency. The Tobit regression applicated to the ``slacks'' resulting from the DEA calculation made it possible to evaluate the stock breeding systems influence on the inputs. A final DEA was applied to the adjusted inputs of the influences of the breeding systems.
We identified three stock breedings classes:
- nomads class, composed of stockbreeders coming from Burkina, Mali and Niger, was excluded from the data during efficiency calculations,
- community stock breedings with few animals and without care. This class seems to be a new system and contains few stockbreeders,
- individual stock breedings being usual and individually practised extensively.
The managerial efficiency evaluation led us to the determination of the global technical efficiency (69.77%) and pure technical efficiency (18.22%). That permits to know the rate of the inefficiency of scale (16.74%). Community system in Korhogo represents the most favourable environment and the individual system characterises the most unfavourable environment.
We can recommande all initiatives in favour of the popularisation of the community stock breeding system.
Keywords: Bovine, Ivory Coast, tobit regression, DEA