The above discussion is summarized in Fig 2, which lists the five types of power-wielders that we have discussed, and the constraints that are likely to be most significant for each. Of course, the appropriateness and efficacy of any of our mechanisms for accountability will depend on the particular context. For example, firms that depend on brand name products can be expected to be more constrained by reputational accountability than other firms. Moreover, with respect to states, the actual character of accountability will depend as much on internal institutions, especially those of democracy, as on accountability mechanisms at the global level. The fact that we have emphasized the latter should not be interpreted as implying that we downgrade the significance of internal democracy. Elections are the principal means by which ordinary individuals can have an impact on policy. The fact that global democracy is infeasible does not mean that domestic democracy is unimportant. What we do emphasize, however, is that domestic democracy is insufficient. Even democratic states will act in a biased way towards non-citizens. The mechanisms of accountability that can work so well to make governments responsive to their own citizens, also can work against the interests of non-citizens affected by government policies.
Nothing in our discussion is meant to imply that accountability mechanisms in world politics work optimally, or even very well. Accountability in world politics is inextricably entangled with power relationships. Those who would hold power-wielders to account need power themselves. Weak actors – including small, poor countries in the Global South, and more, their often disenfranchised publics – lack the capacity systematically to hold powerful actors accountable. Furthermore, accountability is relatively haphazard. Nothing guarantees that the issues brought to public view are the