Payment system evolution: transitional economies. Transitional economies are moving from a paper-based mono-banking system with local/regional clearing and settlement. The transition is accompanied with a shift from state to private ownership, the disaggregation of the mono-banking structure, serious legal uncertainties, a rise in credit risk, and expanded payment float which significantly affects enterprise trade and the money supply (Table 4). These and other problems are addressed by simultaneously: (a) expanding the legal structure to support private ownership and payments; (b) speeding up the processing and settlement process for existing paper-based payments; (c) developing a same-day or next-day payment network restricted to enterprise and international large value payments; and (d) determining the degree that the banking and payment system will be automated and centralized in the future. This is an exceedingly complex undertaking and it is helpful to have a list, such as that provided in Table 5, of the various issues one should be aware of in this process.

Points of excessive payment delays should be identified and the value of float estimated. Relatively low-cost procedures can be implemented to alleviate this problem in the short-run. Existing "off-the-shelf" payment technology can usually, with some modification, be adapted to local conditions and perform with similar effectiveness in most environments. Overall, payment system change should be evolutionary rather than revolutionary; successful improvements to a payment system depend to a large degree on identifying important unmet market needs and working closely with users to find solutions they can accept. These conditions for success recommend proven payments procedures rather than the newest technology. Indeed, it would not be an exaggeration to say that all short-run and long-run changes to a payment system in a transitional economy can be more than adequately handled with current technology and procedures: the newest technology is not needed and will not be cost-effective if used.

User needs and payment system design. A detailed knowledge of how payment systems in developed market economies are structured and operate is quite helpful to accurately identify areas where payment systems may be improved in emerging market economies. A properly structured payment system should have a number of attributes, driven by enterprise and household need for acceptability, timeliness, and reliability. After this comes safety, efficiency, and ease of use. In restructuring a payments system, the goal is to look for effective ways to improve these attributes for both current and future users.

The initial step in improving a payment system should be to accurately determine user needs which, in turn, will form the basis for appropriate payment system design, system requirements, legal structure, and risk controls. An Appendix provides information needed for this task. This involves an assessment of the current physical, legal, and information infrastructure, a knowledge of the structure and technology of the financial system, an inventory of currently used payment instruments and how they are processed, and the existence of incentives to guide payment use toward the most cost-effective outcomes.

Enterprise needs for payments primarily concern the ability to quickly and safely transfer large values among firms distant from one another within and between countries. As well, enterprises require financing which is typically obtained by issuing debt and equity securities in financial markets. This also requires the transfer of large payment values. These enterprise needs are largely met through the development of a single secure electronic large value transfer network which connects a country's major trading centers and ties into existing international funds transfer arrangements. Such a network may be publicly owned and operated if the private sector is insufficiently developed to handle such a task. In any case, settlement can be most cost-effectively handled through the central bank. Risk is minimized with RTGS using either idle or interest-earning balances or posted liquid collateral. Fees should be set to fully recover costs and the legal rights and liabilities of the various parties to the payment transaction should be clear and enforceable.

While most consumer needs for payments are largely met through the provision of cash for smaller value retail transactions, this is usefully supplemented with noncash payment instruments which allow for larger value consumer payments both at and away from the point of sale (e.g., for bill payments). In relatively aggregated banking systems, which is the typical case, this can be cost-effectively realized through a centralized paper-based or electronic GIRO system developed, owned, and operated either by the banking system or some government entity. GIRO payments, once made, are final payments and settlement is internal (so central bank settlement is not needed). A check-based system is really only a viable alternative when the banking system is highly disaggregated and when there are effective institutional, legal, or communication infrastructure barriers to electronically connect large numbers of institutions to a centralized clearing house. Also, checks cost more than GIRO payments to process, clear, and settle and, because of return items, represent a provisional rather than a final funds transfer.

The issue of what entities - banks, non-bank financial institutions, or enterprises - may have direct access to the payment system is the same one faced in most countries. The typical outcome has been to allow direct access to the payment system, and hence to central bank settlement, almost exclusively to banks and other types of deposit-based financial institutions (savings banks, credit unions, co-operative banks, etc.) Direct access is not usually given to non-deposit financial institutions (insurance companies, securities firms) or to enterprises and their financial subsidiaries. Although the payment access issue has never been adequately investigated and seemingly rests on a desire to separate risky commercial enterprises from less risky banking activities, scale economies are likely realized at the clearing house and payment network level when the number of direct payment participants are small rather than very large.

Ownership of payment processing centers and payment networks is usually in the private sector. The exceptions are the U.S. where about one-fourth of all payments are processed by central bank offices (due to a disaggregated banking system) and various countries' GIRO centers (owned by the postal service). The ownership issue is tied in with how payment services are priced. If the private sector owns these facilities, then pricing should cover all costs, including a reasonable return on invested capital. When a central bank owns and provides these services, they have been either been subsidized (as in the U.S. prior to the 1980s) or priced as if they were privately owned where all costs are covered (the U.S. currently). In general, resource allocation is enhanced if payment services are priced to recover all the costs of their production.