Many management problems faced by government in the public sector can be corrected only by changing the way that central bureaucracies are organized and managed. Recognizing this, developing countries have made numerous attempts to reform their bureaucracies in recent years. Few have succeeded in improving public sector efficiency, partly because of the unfavourable political climate and partly because of the institutions that many countries inherited at independence. Developing strong and efficient public institutions requires considerable investment in human skills, and a readiness to experiment with organizational structures to find those that best fit the societies they are intended to serve. It took industrialized countries more than a century to develop reasonably effective institutions. Many developing countries are attempting to compress that process into a few decades. It is not surprising that disappointments and political strains have often occurred.
Developing countries also have to contend with unprecedentedly rapid population growth and technical change, a more integrated and competitive world economy (which multiplies the consequences of mistakes as well as of success), and high political and economic expectations among their peoples. Creating the managerial skills and institutions to cope with these demands is an exceptional challenge. In countries where general educational levels are low, poverty severe, and institutional experience limited, the task will also need sustained external assistance, although the main contribution will always be a national one.
The eradication of corruption as a feature of public life depends on the gradual creation of a political and public climate favoring impartial institutions, as well as on specific actions by government. Many governments from time to time have initiated anti-corruption drives. However' such efforts tend to be shortlived and ineffective, since they often concentrate on punitive measures and even closer - but still unworkable - controls, instead of designing interventions so as to minimize the opportunities or incentives for corruption. For example, corruption can be limited by avoiding administratively created scarcities (as some centrally planned economies are doing by effectively sanctioning a "second economy"); by reducing controls on international trade and payments; and by improving the incentives and accountability of officials in the areas where regulations or administrative discretion remain. Corruption is usually better fought by a combination of fewer, better-paid officials controlling only what really needs to be (and can effectively be) controlled in the full light of public scrutiny' than by occasional anticorruption "campaigns".
This strategy features in the framework of Agenda 21 as formulated at UNCED (Rio de Janeiro, 1992), now coordinated by the United Nations Commission on Sustainable Development and implemented through national and local authorities. In this context, Agenda 21 recommends establishing effective combinations of economic, regulatory and voluntary (self-regulatory) approaches.
Current institutional problems should not obscure the notable progress made by developing countries. African countries have within a generation established the entire framework of national institutions and staffed those institutions with their own citizens. Some Latin American countries, whose institutional structures long served predominantly rural oligarchies, have expanded their administrations to cope with the demands of rapid urban and industrial growth. East Asia has developed sophisticated economies along with a more modest, but still impressive, growth in government capabilities.
Many countries have brought some of their key industries into public ownership to assert national control; change in their status or their mode of operation is often politically sensitive. Economic efficiency therefore has to be balanced against considerations of practical politics, national sovereignty, and social policy. Similarly, appointment-by-merit in the public service is a principle that might have to be modified to take account of a country's ethnic or religious tensions. In many cases, therefore, "inefficiency" is less the fault of bureaucrats than the consequence of demands, legitimate or otherwise, that the political system places on the bureaucracy's limited capacities.