A spatial stochastic simulation model was used to compare the efficacy of different badger control policies and to determine the theoretical requirements for the control of endemic bovine tuberculosis in badger populations in southwest England. Culling-based strategies for controlling endemic disease were compared with strategies employing a yet-to-be-developed oral vaccine which would provide uninfected badgers with immunity to the infection. A comparative assessment was made of the efficacy of previous and proposed culling-based strategies employed by the Ministry of Agriculture, Fisheries and Food for the control of localized disease, and the potential for an oral vaccine-based strategy for the control of localized disease was examined. For endemic bovine tuberculosis, to achieve a reasonable probability (p > 0.70) of successful control with a strategy involving a single culling operation, a very high proportion of the badger population (> 90%) must be culled. Single vaccination would not be successful in combating endemic disease. However, strategies involving repeated annual vaccination would have a very high probability of eradicating endemic disease, even with a relatively low (40-50%) annual vaccination efficiency.The most successful culling-based strategies for the control of localized disease were the gassing and clean ring strategies. Compared with no control at all, the interim strategy only offered benefits of a lower probability of disease spread and persistence in populations with low disease-free equilibrium group sizes or low initial prevalences of infection. In all other instances the benefits were negligible. The live test strategy will offer an improvement over the interim strategy, but will not be as effective as either the gassing or clean-ring strategies. In addition, it is likely to necessitate the culling of approximately four times as many badgers each year as the interim strategy, and the proportion of those killed that are infected will be approximately half that under the interim strategy. The efficacy of a strategy involving annually repeated oral vaccination of the badgers within a similar area to that covered by the live test depended on the efficiency of vaccination. A vaccination efficiency of 20-60% represented an overall improvement in efficacy over the interim strategy, being equivalent to the live test strategy. However, only vaccination efficiencies of 60-80% or greater achieved similar results to the gassing strategy, and none were so successful as the clean-ring strategy.Recommendations for future management are provided. Reactive strategies based on culling or vaccination will not solve the problem of bovine tuberculosis in badgers. Proactive strategies directed in those areas with a recent history of bovine tuberculosis in badgers should be considered as an alternative short-term control measure. The only strategy likely to eradicate bovine tuberculosis from badger populations in the long term is the use of repeated vaccination in proactive control operations in areas with a history of bovine tuberculosis in the badger population. Analyses should be conducted to evaluate whether the economic benefits of the live-test strategy are likely to outweigh its economic and ecological costs and whether continued research into the development of a vaccine for badgers is likely to offer any significant long-term economic benefits.