This presentation tries to summarize the article presented by Luis Moreno Ocampo and myself to be published by Governance, the journal of the Economic Development Institute of the World Bark The article refers to the experience of the Dominican Republic, were the Interamerican Development Bank funded the "Programa para 91 Mejoramiento de Ia Transparencia en la Gestion Publica". The .program was conducted by the Procuraduria General de la Republica--and the Departamento de Prevencion de la Corrupcion. This writing also mentions the work of economists such as Dany Kaufman and Rafael Di Tells who are exploring the economic implications of this model.

The model is a database designed to resister the terms and prices paid by different agencies of the government in their procurement operations. In the case of; the Dominican Republic, we started collecting information an the procurement of public hospitals.

The proposed model has the following features:

• It is not focused on individual behavior, (avoiding the conflicts that arise from moral blaming, prosecutions and long trials).

• It is not focused on a particular model or system of procurement. (it doesn't take into account the formality of the procurement practice and therefore it can be applied to different procedures).

• Its focus is the result or outcome of the procurement process, measured in terms of price, quantity, conditions, etc.

• It helps comparing efficient management, without analyzing the cause whether inefficiency, corruption or others.

• The information provided by the database on the efficiency of the procurement process is helpful not only for the public officials responsible, but also for citizens at large.

• It is based on hard data, rather than perceptions or subjective information,

Taking into account, Robert Klitgaard's well known formula for corruption (C=M+D-T), the database helps reducing monopoly and discrecionality and increase transparency, by openly scrutinizing the result of the agents performance.

When the information contained in the database is made public, it becomes an incentive for honest officials who are rewarded when their results are compared with those who perform badly. In turn, it discourages corrupt behavior, since it increases the chances of some sort of punishment.

The model is based on the idea that, in an environment of hyper-corruption, publishing the result of the procurement process (that is, the price paid for a given product) can reduce significantly the Misuse of public funds.

For example, in the case of the Health Department of the city of Buenos Aires, we found that once the database was implemented, a significant change took place. One of the products included in the bass was of dextrose, a very basic product for hospitals. Once the information was collected, the Secretary of Health showed it to Hospital Directors, sharing the information of the price paid by other hospitals and letting. The result was that In four moths, the average -pnce paid per hospital fell a 25%.

The database can also be used to identify the more efficient public officials and have them train the rest of their peers on how to reach identical results,

The model also helps avoiding the difficulties of using the formal legal system to control corruption. The analysis of corruption simply as an act, which violates the law, obscures the fact that in most countries, the law is not the rule followed by the different.

Political leaders have their own rules to obtain and maintain power, forcing them to protect the members of their group. Businessmen must maximize profits, and if the market in which they act is used to violate the limits of the law to obtain handsome earnings, they have a strong incentive to do so. Therefore, the rules of membership to a given group or maximizing earnings can contradict the law. Henceforth, the problem Is not about an act that transgresses a system of rules, but of different systems of rules that come into conflict.

The experience of the Dominican Republic

The Procuraduria general de Ia Republica has recently conducted the "Programa Para el Mejoramiento de la Transparencia an la Gestion Publica", sponsored by grant of the Interamerican Development Bank

One of the components of the this project was to create a database within the Secretaria do Estado do Salud Publica y Asistencia Social, SESPAS (Ministry of Health). The purpose of this database was to register basic information regarding of public procurement. As said before, the database is a simple and practical method to monitor the variance in prices and conditions for a given product.

The database is designed to register the prices paid in the purchase of products with homogeneous properties. For example, dextrose, cotton, syringes (discardable), fluorescent tubes, writing paper, fuel and items alike fall under this category. Furniture, clothes (unless standardized), vehicles, computers and items alike can vary significantly. Therefore, they are not good samples for the database.

Almost any product that is sufficiently homogeneous and standardized can be included. This allows the database to be used for a wide range of items. Comparisons are always established for single products. Acquisitions of dextrose are only compared to other acquisitions of dextrose, either by the same hospital or any other hospital, and so an with any other product included in the database

The project to establish a database was quickly accepted by the Secretary Of Health, who appointed the Auditing Department as the unit for operating the database.

The Auditing Department was selected because it concentrates all the procurement information. Every month, each hospital sands to this office the documents of supporting their operations. This allowed an easy access to the information to feed the database.

A special group was trained at analyzing the procurement documents, selecting the required information and in the steps required to load the database. The database is a standard software spreadsheet. A computer was set aside at the Auditing Department for loading the software.

Once the software was loaded and the personnel trained, a series of definitions have to be made regarding the different fields of the database. In our case, the database had the following information questions:

a. Identification of the products to be registered

b. Terms for payment

c. Brands of the products to be registered.

d. Name of hospitals included in the sample.

e. Identification of the name of the principal suppliers.

i. Duration of the sample

The first stop consisted in selecting the products to include in the database. As it has been mentioned, these products should be homogeneous and comparable, that is to say, they should not suffer substantial variations according to the brand, the supplier or even the packaging.

For each product the database has a field for quantity, unit of measurement (pound, litter, dozen, ton, etc.), price per unit and total price paid.

The initial idea was to select some items that have already been registered I . n databases for hospital procurement in countries other than the Dominican Republic. We only found one matching element (dextrose), and it was included.

Three more products: 10cc. syringes, "00000" quality rice and bacterodyne were included.

Regarding the conditions of payment for these operations, eight alternatives were selected: 30, 60, 90, 120,.150, 180 and more than ISO days credit and cash, In the chart presented, they are identified with the numbers 0 to 8)

Most of these products are produced by a limited number of manufacturers, at least in the case of the Dominican Republic. The list of the principal suppliers, importers and providers was also included.

The sample was limited to the fourteen largest public hospitals. in the Dominican Republic, The Government has established a special medicine plan named Programs do Modicamentos Esenciales (PROMESE). A special office of the Presidency administers this program. Their job is to buy basic medicines directly from the producers. Then they sell these medicines directly to the people and also to hospitals. They work with a limited list products. Their prices are published widely. Officially, they charge a small commission, which allowed the program to be self sufficient in a short time. We decided to included their prices for syringes, as an additional source of information.

As far as time is concerned, the, sample included procurement operation's performed during July and August 1997 (J and A in the chart).

 The outcome of the experience

The following chart is a condensed and partial example of the database for the 10cc. syringes.

The results show a substantial variation in prices paid. Pay special attention to the sixth column, where the conditions of payment are described, remember that 0 stands for 30 days credit, 1 for 60 and so on. Compare observation # 3 and # 44. In observation * 3, the hospital paid $4.84 for each syringe, payment due on 180 days. In observation # 44, another hospital, one month later, bought a larger quantity of syringes, paying $ 1.135, each and payment due after 160 days,

Other observations show similar variations.

The database was helpful not only for detecting this kind of problem. It allowed the Auditing Department to identify a series of obstacles for properly obtaining and processing procurement information. Thus, they issued a series of recommendations for the Secretary of Halt to implement in improving the existing procedures.

. But perhaps the most important experience was that the Secretary of Health summoned a meeting with the directors and Administrators of all the hospitals included in the sample. During a meeting, the information was shared, pointing out situations that required a corrective strategy.

The use of the Information for economic analysis

Some economists have recently paid attention to this model and are trying to reach further conclusions from the information obtained.

Economists, such as Di Tells and Kaufman, believe that the first consequence is that this system allows for controlling corruption through increases in product market competition. In their view, this is particularly effective in countries where law enforcement is lax.

In Di Tella's words 'a practical way to relate incomes and expenditures is to create competing scenario to achieve 'bureaucratic efficiency'. In the case of public agencies, this can latter be used to re-distribute funds according to achievement. Instruments such as the database can provide economists with the information required to establish a more precise comparison among different agencies.

Many developing countries suffer not only a shortage of funds, In some' cases there is also a problem of efficiency and accountability in the use of the existing funds (sufficient. or otherwise).

This model can help, if expanded in -different countries and agencies, to extract more data on the results of procurement decision, Moreover. this information can be transformed by economists In a new and, perhaps more reliabie, guide to decide on the best option for allocating public funds.

Sources:

"How to Reduce Rapidly Endemic Corruption in Public Procuremenr Luis Moreno Ocampo & Roberto de Michele, Governance, EDI, The World Bank.

Programa para el Mejoramiento de la Transparencia en la Gestion Publica
Procuraduria General do la Republica, ATN/SF-5436-DR.
Banco Interamericano do Desarrollo

"AlIgunas consideraciones economicas sobre le corrupcion y una propuesta para. combatirla' Rafael Di Telia Fundacion Mediterranea.