21 Lesson 2: It is constantly repeated that the aims of GA are effectiveness (how useful an expenditure is), effi- ciency (how to achieve the maximum at minimum cost), equity, and transparency. Some of them can be satisfactorily attained at a local level, others are beyond the possibilities of a small admini- stration. The effectiveness of a service (and related expenditure) can be estimated in an "objective" way, by comparison with suitable benchmarks, or "subjectively" by surveys and analysis aiming to evalu- ate whether the service meets local needs. The latter method is usually too costly, while the com- parison with benchmarks is easier and less expensive. The benchmarks may be suggested by a higher-level administration (state, region) or they can be worked out initially by experts, following procedures which can be replicated later by the local administrators. The experience I have been describing brought to light the many merits of the method of "network comparison", the benchmark being the best performance in the network. International standards can also be used. To quote an example from our experience in the province of Siena, it is possible to construct an indicator which puts together the number of cr?che places relative to the population aged 0-2, the opening hours per week, the teacher/child ratio. How to construct it, can be decided together by experts and poli- cymakers. Local administrators can be easily instructed how to replicate the procedure and use the indicator to monitor their performance. Similarly, a checklist of the services for immigrant women can be assembled, to suggest what can/must be done for them (so that, hopefully, local administration will not organize only football matches to improve the relations between natives and migrants!) Efficiency calculations are much more difficult, since a correct calculation implies the inclusion of all costs, visible and invisible. It is not only the problem of determining imputed costs, such as the cost for the use of premises where a service is provided, which are only "opportunity costs" (i.e. forgone income) and therefore do not appear in the budget. It is also necessary to include the in- visible cost of care labour. For example, if we want to determine whether the cost of assisting old people in their homes is higher or lower than in homes for the elderly, we must take into account the cost of the labour of the relatives/ friends who look after the old person at home. Obviously this calculation is beyond the reach of a simple accountant. Equity can be more easily studied, at least whenever there is agreement on what "equity" means. Information about the role of women in the local government and the personnel policies of the local administration is easily available. The collaboration of the administrators is necessary, however, to gather information about their policy when they have to hire a consultant or to appoint a representa- tive in the executive board of a private or public company. Table 1: Division by gender Table 2: Share of remuneration The tables above show that apparently the administration of the Comune of Siena follows an equal opportunity policy, since it hires men and women in exactly the same number. However, men get a much higher share of the remunerations: 65% vs. 35%. DONNE 50% UOMINI 50% DONNE 35% UOMINI 65%