In April 1965, a young engineer, Gordon Moore, published an article in Electronics Magazine, predicting that the complexity of integrated circuits would double every year, and the cost would continue to drop, and as a result, information technology will rapidly spread to every corner of the world.
This prediction, known as Moore’s Law, has only proven to be all too accurate with time. After the 1980s, with the rapid development of information, major changes have taken place in the domestic economy and taxation structure. The population has grown significantly, and the degree of urbanization has increased. Along with economic development, taxation data has increased significantly. Tax automation operations must respond to demand and be greatly innovated.
Beginning in September 1987, the FIA, the Taxation Administration, and the Department of Statistics within the MOF, along with the Taipei City National Taxation Bureau, the Kaohsiung City National Taxation Bureau, the Tax Bureau of Taiwan, the Taipei City Tax Collection Office, the Kaohsiung City Taxation Department, and other units jointly formed a development promotion group to plan the “Ministry of Finance’s tax automation (five-year) development plan”. It is divided into three sub-items: national tax, local tax, and tax management.
During the preliminary stage of establishing the prototype for tax automation, the FIA established information systems for different tax categories, and the local tax collection offices also individually installed computer hardware equipment. The tax category systems and programs were not integrated under one unified standard, with varying operation methods among the tax collection offices; therefore, tax automation still required plenty of improvement.
In the tax automation (five-year) development plan, the concept of overall planning and holistic design is specifically emphasized, and the business system planning method (Business Systems Planning, BSP) is used to analyze various tax data processing structures, and classify different tax categories into “processing levels” and “management levels”.
The processing level is divided into two methods: centralized processing and local processing. Taking the individual income tax as an example, the filing system, tax assessment system, tax query service, and regional statistics system are processed locally by various tax collection agencies; in the decision-making information, case selection, assessment, national statistics, and analysis are done by the FIA. The information processed locally is mostly at the operational level, while the information at the management and decision-making level is processed to integrate information from different sources across the country.
From 1988 to 1993, a total of 140 systems were integrated and planned by the development promotion group, with 6,914 programs. A total of 120 tax systems were extended to the tax collection agencies for local processing, and centralized operations were carried out in 41 systems through the FIA.
In addition to completing the system development and maintenance work, the tax automation plan also clearly stipulates the information workflow. The fiscal and taxation data center and the tax collection agency jointly compiled 14 “financial and taxation data processing manuals” for management, and “fiscal and taxation information operation specifications” for operations. A total of 11 types are used as the basis for the implementation of the work of the taxation agency’s electronic work units and business units.
In addition to the completion of individual program systems, how to connect information with each other became the focus of the promotion team. After analysis and discussion, the team members divided the tax information processing units across the country into the FIA responsible for central collection and national taxation bureaus for regional operations. There are 3 levels including the National Taxation Bureau and the tax collection offices of counties and cities that operate locally. The national tax information processing, such as checking, handing over, case selection and checking, household registration, and national statistics, etc., is centrally processed on the mainframes installed by the FIA.
All national taxation bureaus and the revenue service offices of each county and city use the small and medium-sized computers installed to handle the filing, correction, query, tax assessment, issuance of written notice, sales numbers, arrears check, and information operations, such as supplementary collection, tax rebate, regional statistics and local information utilization.
Each tax collection sub-bureau (department) or tax collection office has set up a personal computer workstation, and through the connection with the head office (department), it can provide convenient services of real-time inquiry. The FIA and various taxation collection agencies are connected through the fiscal and taxation network, and a large amount of information can be queried online in real time, and files can be transmitted to meet the requirements of the taxation agencies.
The basic element of the tax automation (five-year) development plan is that “all tasks that can be done by machines are handled by machines, and ones that can be operated on a single track are no longer dual-track.” Through automated operations, processes are precise, manpower is reduced, and efficiency is improved.
Originally, the issuance of certificates requires more than half an hour of manual work, but it can be completed in less than 10 minutes through computer processing. According to FIA estimates, this reduces staffing by more than 4,000 every year, and effectively prevents tax evasion and tax avoidance. Tax revenue has increased approximately NT$78 billion, enabling taxpayers to enjoy fairer, more convenient, and faster tax services.