There is one way of reforming electoral arrangements that shows remarkable success at a relatively low cost to the authorities that introduce them. This is by using electoral quotas and reserved seats to ensure greater representation of social groups on elected bodies that would otherwise be under-represented. Quotas specify a target percentage of seats to be filled by a given section of the population by election. Reserved seats set aside a specific number of places to be filled by the target population either by election or appointment (Reynolds 2005). Maoris have had reserved seats in the New Zealand parliament since 1867 (Archer 2003), but at least ten other countries reserve seats for specific ethnic, religious or linguistic groups (Meier 2007). Quotas for women are found in one form or another in more than 100 countries, some introduced as early as 1954 (Pakistan) but most in the 1990s or later (Dahlerup 2007b).
If quota rules are clear, adapted to the electoral rules of the country, and backed by sanctions, they can result in a rapid and large increase in women sitting in national and local parliaments - as they have in such diverse places as Costa Rica, South Africa and Rwanda (Matland 2007: 278-80). Fourteen of the seventeen countries with more than 30 per cent of parliamentary seats held by women in 2005 had either legal or party quotas (Dahlerup 2007b: 18). Moreover, the evidence shows that they can have a substantial effect on increasing women’s representation in a short time, sometimes in the first election to which the quotas rules are applied. While most quota systems do not manage to meet their target figures, they still manage to increase the representation of the target group substantially and quickly.
The message is clear: quotas work. So it is worthwhile spelling out briefly the reasons for their efficacy, if they are correctly applied in the first place. First, although some are voluntary, some are backed by the legislative force of the state and enforced with sanctions. They are mandatory obligations: not the kind of voluntary options that can be ignored. Second, they are targeted on a specific portion of the population, rather than the general electorate. Third, female quotas are defined and demarcated in a precise way. This is not possible for many other social groups in society that might try to claim a quota, such as ethnic, religious, class or age groups. In their case, quotas would provoke huge difficulties about who exactly is covered by the legislation and who is not. Fourth, they are relatively costless for governments to introduce, if not for the organizations obliged to respond to them. Governments simply have to pass an act, which is just part of their daily routine. They do not have to set up special budgets and departments to organize complex activities or experiments and they do not have to create special agencies to monitor and control the application of policy or the spending of money. Fifth, female quotas in most countries seem to arouse little controversy. And sixth, unlike the technology of the new electronic communications that is not easily transplanted to developing countries (Heeks 2003), quotas are easily understood and, in principle, relatively easily applied.
Like any other political practice, quotas may also have unanticipated consequences. For example, if a national legislature is to represent the voting preferences of a large part of the electorate, there may be room for only one quota. To discriminate in favour of one group, however, is necessarily to exclude others from this preferential treatment and this may cause political problems and resentment on the part of the excluded.