The development of resistance by organisms to antimicrobials is a natural phenomenon. Although it is increased by the use of antimicrobials, reduction in use of them will not necessarily reduce resistance, and there is a possibility that eventually most, if not all, antimicrobials will become largely ineffective. However, current estimates of the economic burden are modest – at anything from less than £5 to more than £20,000 in reported additional costs per patient per episode for hospital costs, and anything up to around £10 billion per year in societal costs, they are far lower than estimates of economic burden from other health problems, such as cancers, heart disease and mental disorders. The reason that current estimates of the cost of resistance are modest is that they are based loosely on the ‘incremental’ cost related to the extra treatment of resistant over susceptible primary infection. This masks the most critical economic burden, which is when resistance leads to the loss of many of the advantages in medical care that antimicrobials have enabled. For instance, advanced surgical procedures and cancer chemotherapy might become far more dangerous as rates of associated infection increase and cannot be treated. The effective ‘removal’ of antimicrobials could mean soaring rates of post-operative infection, mortality and morbidity from what are currently considered to be trivial infections, presenting an apocalyptic blow to health system development. The full economic burden of this is not only inestimable at present, but unimaginable. It is therefore urgent that a full system analysis of resistance is undertaken. Rather than continuing the focus solely on the additional cost of treating an infectious disease in the presence of resistance, what is needed now is to look at how health services more broadly might be affected if resistance becomes endemic, and the wider implications of this to society. Incentive mechanisms also need to be developed at a number of levels, and could require radical change, such as moving to more restrictive prescription (for example, through a hospital setting only, removing the ability to prescribe from primary and community care). Clearly the necessity of, and ability to implement, such radical change will need to be informed by the system analysis, and also supported by it when it comes to engaging the wider political and popular support that would be required. In needing to consider a system approach, resistance shares many of the characteristics of climate change. Both provide future significant threats to human well-being, both are subject to considerable uncertainty about their future extent and trajectory, both are global problems where the response differs across nations, and both have as an underlying cause the over-consumption of ‘goods’ that lead to short term benefit. Yet for both there is considerable inertia to major, radical, change because there is a focus on current burden and because the personal incentives seldom match those of society more generally. However, there appears to be increasing scientific consensus about the impact of global warming in a way that is perhaps less clear – or at least less coherent – for antimicrobial resistance. For global warming this has led to clear, simple and consistent messages from the scientific community, underpinning increasing acceptance of the desirability of action by a broader policy-making community including politicians, economists and philanthropists. For antimicrobial resistance there is a clear danger that waiting for the burden to become significant before taking action may mean waiting until it is too late to stop an apocalyptic scenario – the very drive behind the early environmental movement’s advocacy of the ‘precautionary principle’.