16. Just as governments are facing up to the challenge of hiving off their public enterprises to private ownership and/or management, at the same time they are examining their own expenditures, at central and local levels, with a view
to identifying activities that could be contracted out to private firms at lower cost to the public purse and higher quality to users. "Alternative delivery" of public services opens up options which many governments are exploring. Services to the government such as office cleaning, canteen services, buildings and grounds maintenance, printing and legal services and the contract leasing of vehicles are common examples. Services by government agencies which are widely contracted out include the collection and disposal of solid waste, street cleaning and lighting, road maintenance and almost all infrastructure construction. Less frequently contracted are the "preventive" services, such as prison detention and corrective services (which are contracted out in Australia, the United Kingdom and some states of the United States of America), environmental monitoring (e.g. water standards in Chile) and tax collection (Chad, Indonesia). Some hard-pressed governments have contracted out even defence to mercenary forces, though most democracies would prefer not to have their national security left to private armies.
17. If the public goal is simply economic, that is, to provide quality public services at the lowest possible cost, the decision to contract out a service depends on the supplier's price, plus contract administration costs, being less than the total costs of in-house production as before. This simple calculus implies that contracting out should be decided on a case-by-case basis, even if there is a general policy in favour. The rule is often forgotten, particularly the need to include contract administration costs in the equation. Even in the industrialized countries, few government agencies have any idea of what they are spending on contract letting, administration and monitoring, though one survey found that these costs added up to an average of 16 per cent of contract cost, with wide variation from one service to another (Stevens 1984). A more sophisticated analysis would also take into account the disposal of assets that may no longer be needed and the costs of personnel resettlement (Prager 1992). For many governments, the personnel implications are all-important, since long-term contracting out implies a movement of personnel out of the government service to independent contractor or contractor employee status. Governments may make it a condition of contracting out that the contractor absorbs both the fixed assets and employees formerly in government service. There may be retraining costs, terminal payments and compensation for loss of job security.
18. In countries which have made the greatest use of contracting out, such as the United States of America and the United Kingdom, the studies are almost unanimous in their finding that costs are saved without loss of quality, and even with gain in quality. Cost savings are typically in the range of 20 to 30 per cent of the former cost of government provision (Savas 1992), despite the requirement (in the United States) that contractors pay their workers no less than prevailing wages. Developing countries and countries in transition, however, should not expect the same magnitude of savings unless there is keen competition among potential suppliers. The less the competition, the less the potential savings by a change to competitive bidding. Gains could be in quality rather than cost: in Jamaica, the contracting out of hospital dietary services actually led to an increase of government expenditure as standards were raised substantially.
19. One alternative delivery option is to improve the existing service and to "contract in". In-house providers should always be given the opportunity of reviewing and re-engineering their own operation so as to make themselves more efficient, if necessary by separating themselves from administrative red tape in order to be more flexible, as in the "executive agencies" of the United Kingdom, Australia, New Zealand and Singapore. The prospect of losing work to an outside agency brings out innovative thinking and willingness to reform obsolete work practices. In practice, competitively bid contracts are often won by the in-house provider. In the United States, federal government agencies win contracting-out competitions 40 per cent of the time (report of the President's Commission on Privatization, 1988). In the United Kingdom, where local authorities are required by law to put services to competitive tendering, in-house providers are now getting the majority of total contracts.1 The efficiency and savings come, not from privatization per se, but from the introduction of competition. Savings are sustainable for as long as competition is maintained.
20. A major advantage accruing from the preparatory work for contracting is that government agencies are forced, often for the first time, to define and specify, in a monitorable form, exactly what services they are providing, to whom and to what standards. Thus, even if the service is not ultimately contracted out, valuable data is provided for better planning and supervision.
21. There are five main constraints on the development of government contracting which are commonly experienced in developing countries:
(a) Political leadership to envision such public-private collaboration and to push for its realization, often against political and labour opposition and vested interests in the civil service;
(b) Lack of sufficient firms to establish competitive bidding. This may be due to a government-imposed monopoly, which can be removed by deregulation, or it may be a simple consequence of a past tradition of government provision, such that no entrepreneur has tried to bid. The first step is an analysis of capacity of potential suppliers. Where there is insufficient competition, the government may divide the work (e.g. geographically) between outside contractors and inside workforces, as is done for instance by the city of Bogota, Colombia, where solid waste collection is shared between public and private agencies, with a private firm made responsible for monitoring the operations of all the agencies.2 Continued government provision of at least part of the service provides in-house knowledge of costs and technology changes and thereby some protection against inflated contracts;
(c) Scarce administrative capacity to draw up, negotiate and monitor contracts in an impartial and non-corrupt manner, to structure partnerships and to ensure that they operate in the public interest, while maintaining independence and avoiding conflicts of interest or regulatory capture;
(d) An organizational implication of contracting is the need to make the contract unit independent of the operations unit (the Chinese wall problem); if these are branches of the same ministry or involve former colleagues, they may act in collusion. Even without collusion, it is necessary to ensure that in-house units do not underbid to get contracts, secure in the knowledge that a loss on the contract will be met by the government;
(e) Lack of accounting data or unreliable data to make good decisions, e.g. on the economics of contracting out. As government accounting is almost everywhere on a cash basis, not a full-cost accrual basis, the cost of in-house operations is likely to be understated.