At the Dawn of Swiss E-Government: Planning and Use of a Unique Identifier in the Public Administration in the 1970s

Sources

The study is based on archival evidence from the Swiss Federal Archives in Bern and focuses on the early phases of ICT use in public administration from the late 1960s to 1980. First, all documents around the revision of the legal basis to social security number use were gathered. In a second step, further documents around the use of an identifier were followed in the different dossiers. To capture the institutional dynamics, interventions in the parliamentary debate on the issue of the unique identifier, data protection, and coordination of ICT in public administration were gathered. A key element of the debate is the social security number (called OASI-Number in this article). The number was created and used for social security purposes, but its use as an identifier for all government purpose has repeatedly been suggested, even though the number was not invariant.

Within this subject-based approach, the records of the four agencies and organizations mainly involved in these discussions were consulted. The files of Federal Social Insurance Office (FSIO) hold records on all matters regarding the insurance number of the Old Age and Survivors Insurance (OASI, in German AHV), including the correspondence of the Central Compensation Office (CCO) in charge of managing the OASI contribution and the OASI number. The Central Office for Organizational Tasks in the Federal Administration (Zentralstelle für Organisationsfragen der Bundesverwaltung [ZOB]) was in charge of ICT coordination and represented the Federal Government in the Swiss IT Conference (Schweizerische Informatikkonferenz [SIK]), hence the holding of the ZOB contains all reports and protocols of the SIK. The SIK, with the cantons and the federal administration as well as communes as its member, aims at improving IT in public administration by common purchase, training, and infrastructure provisioning.

Based on the rich archival material, elements of an administrative history with a focus on organizational development have been elaborated.

Political Interventions

The interaction of politicians and public services needs to be discussed for the specific case; the theories on the executive character of the public service do not fully apply (see Jenkins, 2008). For Switzerland, two main factors need to be considered. On one hand, the public administration has extensive power in the preparatory works for policy changes. On the other hand, general political tendencies may influence the administrations’ choices (Ladner, 2013). On a political level, the promises and risks of ICT use in public administration were brought up in Swiss Parliament. The interventions were made by members of the parliament who had previously dealt with the subject of ICT use in the role as members of city governments. In 1969, the liberal national councilor Alfred Hummler (cf. Mayer, 2012), who had been actively involved in the founding of one of the first regional state-owned IT service centers in St. Gallen, focused on the challenges of coordination and asked the government in a parliamentary postulate what steps it intended to take to coordinate the development of a national information system and a national database (Hummler, 1969). Hummler drafted a long-term vision of a national IT architecture and a central database with regard to its efficiency, the high cost of IT, and the shortage of experts and programmers. He suggested that the national government and administration should take a leading role in the coordination of the different needs. In addition, he asserted that the national coordination should also lead to legislative action to ensure privacy and data protection to mitigate risks arising. The postulate was passed, and in reaction, the Federal Council commissioned a detailed study of the subject, which will be discussed later in this article.

At the same time, in 1971, due to the rapid developments in IT and the use of databases, the Social Democratic National Councilor Alfred Bussey (cf. Clavien, 2007) asked the government in a parliamentary motion to take all the necessary legislative steps to protect citizen’s privacy. The legislation should cover all database use, whereby he assumed that private business use must be the primary concern (Bussey, 1972). In response to his demand, the government came forward with extensive references to the situation in other countries and the current research on the matter and agreed to evaluate possible measures. Nonetheless, the debate on data protection legislation went on for another 20 years before legislation on data protection was put into force in 1993 (Schaaf, 2010).

With only two interventions, ICT use in public administration has been a highly peripheral issue in parliament. Nevertheless, both interventions have triggered the legislative and planning works of the government and administration. They refer to both the benefits regarding cost efficiency and the risk of IT concerning data protection and thereby show the ambivalent position toward IT in the period. Furthermore, argumentation in both cases includes references to the discussion on the subject in Germany and the privacy debate in the United States. These references lead to the assumption that the international debates and developments have been closely followed in Switzerland.

The Role of Public Administration Agencies

To address coordination issues and develop overarching strategies for efficiency gains, the Central Office for Organizational Tasks of the Federal Administration (ZOB) was founded in 1954 in reaction to a popular initiative to oppose the growth of the federal administration (Koller & Schüpbach, 2016). Initially, it was set up as an internal consulting and training unit for all government agencies; since 1960, the ZOB was additionally in charge of central IT coordination, at the time called “automation” (Fehr, 2011). In response to Hummler’s postulate and to a motion by the Committee on Financial Affairs of the National Council, an expert committee was installed to propose reforms of the ZOB to strengthen its role and increase the impact. The expert committee under the lead of Professor Willi Geiger, an academic in the field of public management, focused on the role of the ZOB in general and ordered a subreport on the question of IT systems to the former head of the of the data center at the Federal Statistic Office, Kurt Steiner.

The report, published in March 1973 (Kommission Geiger, 1973), first explains the term “information system.” The author defines it as integrated use of information, which consists in the automated reuse of the same piece of information for different tasks within one agency or across different organizations. The report referred to first experiences in IT system development which revealed the great complexity and resulted in high costs and demanding coordination tasks. Hence, it argued for partial, subject-related information systems, which would help to limit the complexity, and questioned the feasibility of a single overarching information system.

The report then sketches out the impact of information systems use in the organization and argues for a balanced approach. This balanced approach is underlined with a critique of extreme positions described as “technocrats” and “pragmatist” both putting either technology or the current organizational structure above other requirements. The report suggests three categories of action:

Overall, the report stresses that a national information system cannot consist of one database or one system but needs to be based on an architecture which allows interoperability and multiple uses of data supported by the definition of roles.

Based on the report, in 1974, the Federal Council changed the Ordinance on the OASI and stated in the yearly activity report that he intended to allow a wider use of the OASI number for further applications (Bundesrat, 1975). However, the CCO, in charge of managing the insurance system and issuing the OASI number, and the FSIO as an owner of the social security number were rather reluctant to open up their key tool to further users. First and foremost, they feared other usages would increase errors and lead to misuse, endangering their core mission as to administrate social welfare.

Following the newly changed ordinance, the Social Insurance Office issued provisions concerning the use of the OASI number. Most importantly, the use of the OASI number had to be communicated to the Office and would only be granted if third party use would not interfere with the core mission. Second, only numbers taken directly from a person’s insurance card could be officially labeled as OASI number. Usage based on self-derived numbers (the 11-digit social security number in use until 2008 consisted of a code referring to the first letters of the surname, date of birth, sex, and nationality of a person) needed to be labeled otherwise. Records show that the Social Insurance Office fiercely defended their key tool. An anecdote illustrates the way of action: A commercial publication of an office furniture company (Rüegg-Naegeli und Cie. Büro-Mitteilungen 89/1971) published an article which showed the (analog) registry of a car dealership, organized after the (self-derived) social security number of their customers. The Social Insurance Office asked the editor of the publication to correct the term “OASI number” in the next edition because it did not adhere to the rules (Möll, 1971).

The electronic use of the number was a recognized need by some cantonal and communal government agencies and the first potential users approached the Social Insurance Office to use their electronic number registry. Among the most advanced IT users were the city of Lausanne and the canton of Vaud, which wanted to use the social security number in their registries. They, however, faced practical problems as the number was only issued to the population covered by the insurance, the working or retired population. A solution needed to be found on how municipalities could obtain social security numbers for children and people who were not part of the workforce (Chevallaz & Gavillet, 1973). Technical agreements were sought to handle the issues. By 1976, the city of Lausanne indicated that they had used other identification means in their system and that further studies on the matter would be taken over by the working group of the Swiss IT Conference (SIK; Petoud, 1976). We can conclude that the lengthy process and the high technical and financial demands of the CCO in charge toward the potential data recipients blocked further progress. Some technical details of the agreements show that data transfer was indeed a costly process involving the mailing of magnetic tapes. Recurring references from the CCO and the Social Insurance Office regarding their concerns on the use of “their” number show a limited commitment to the project.

In this early attempt of data use by different government agencies, we can observe strong cultural limitations. Shared data would require a minimum of coordinating the requirements between the various use cases of the government bodies involved. As the data owner emphasized its ownership in a silo logic, the effort and the risks for further users were too high, as they had no possibility to influence the long-term development of this key resource. The benefits of a unique identifier in an overarching administration perspective were not valued by the Social Insurance Office. Organizational transformation as reflected in the maturity stage model was not acknowledged.

Coordination as Alternative Approach

The shared use of information was first and foremost a definite need of the cantonal administrations. In parallel to political and administrative activities on the federal level, the cantonal authorities concluded that their efforts in the field of IT would need coordination. An institutional framework was sought, and the Swiss IT Conference (SIK) was founded in 1975. The system of intercantonal coordination conferences had been proven to be successful in other policy fields before. This newly established organization encompassed several state levels. The ZOB represented the federal administration, cantonal authorities mostly by their IT departments, if existing, and communal representatives who participated in some of the workgroups. These workgroups were essential for the coordination of activities as the resources of the organization were limited to two staff members (SIK, 1975).

The question of a unique identifier was considered to be essential. Therefore, one of the first working groups of the newly established organization dedicated itself to the use of the OASI number. Representatives of the cantons, as well as representatives of the ZOB and the CCO, participated in the workgroup. One of their main activities was to survey the current use of identifiers in the cantonal administrations. In all, 35 of 42 responses indicated the use of ICT to fulfill their tasks. About half of the respondents confirmed the use of the OASI number; the remaining half was using their own identification numbers. Complicated handling was reported to be the main reason for nonuse. The report concluded that the lack of use should be addressed by simplifying the procedures. The group achieved a further tangible result: They defined a process to ensure that inconsistencies detected in social security numbers between the Compensation Office and the central register of foreigners (another early example of an electronic registry) could be corrected (Arbeitsgruppe AHV-Nummern Verwendung, 1978).

It seems that some participants were, due to the complex handling, in favor of creating a new identification number. Therefore, a new workgroup was formed in 1977 to analyze different options for a unique identifier. In addition to the parties represented in the first working group, a representative from the private sector (insurance company) and an academic joined, indicating that private sector use should also be considered.

The workgroup’s final report (Arbeitsgruppe Personenkennzahl, 1979) mentions three major IT projects with the potential use of a unique identifier: a database with members of the armed forces, a central register of foreigners, and a general ID card for the entire population to be issued for defense purposes. Regarding the use of a unique identifier for various purposes within the cantonal and communal administration, the workgroup compared two options: the use of the social security number and the use of a new identifier. Both options acknowledged the transforming power of ICT use. A technical shortcoming of the social security number was the changes in certain life events like a name change after marriage. While the primary owner’s business process could easily deal with these restrictions, all other users saw the invariability and uniqueness as an essential requirement for their tasks. The workgroup proposed a workaround by storing the old and new numbers in case of change. Data protection considerations, however, lead to the conclusion that only the new number should be used in data exchange to protect information on critical life events regarding privacy such as naturalization, adoption, or sex reversal. This workaround solution shows the reluctance to fundamental changes in organization and tasks based on ICT needs.

The second option, the creation of a new identifier, was seen as a technically better solution as it would provide an invariant and anonymized number and solve some of the data protection issues. As a downside, the task of creating a legal basis and installing a central entity to manage the number was seen as critical as it would only serve the internal purpose of public administration. Overall, the report argued in favor of the OASI number, as the workgroup assumed that the primary objective of the number secured the support of the general public. The proposed workaround would in the long term eliminate the risks emerging from the variability. In addition, the CCO had announced simplification in the process with remote access to its database by 1982. Technical innovation would help to simplify the collaboration between remotely located administrative bodies.

Most importantly, the question of responsibility was discussed within the workgroup. The Social Insurance Office agreed to take on a consulting role of the other users regarding major developmental steps of the numbers and its use. It was, however, made clear that the office remained the sole owner of the tool with the power to take decisions regarding the use and further development of the OASI number.

The primary purpose of the SIK, however, was to foster coordination and collaboration in the field of ICT in public administration (SIK, 1975). A number of workgroups have been created within the SIK which have paved the way for collaboration and data exchange by creating standards for name and address records in electronic files and later name conventions for legal persons. This approach to standardization combined with the voluntary implementation in the different agencies was the preferred way for collaboration, as it created an immediate impact. An investigation, conducted by the managing director of the SIK, Ronald Zürcher after 3 years of activity of the SIK, revealed that standardization proved to be a highly valuable activity. In his report, he also pointed out two tasks to resolve the shortcomings: First, he described standardization of key terminology as a practical actions item. Second, he aimed at strengthening the role of ICT and ICT professionals in the political process. Zürcher complained that laws and ordinances were drafted without involving the ICT professionals. This would often lead to a situation in which the data processing centers in the cantons were confronted with implementation requests once the legal grounds had already been defined. The standardized terminology and the collaboration options, he continued, were at this point difficult to implement. Therefore, his primary goal was to widen the circle of the SIKs influence. Once the SIK get involved in lawmaking projects with implication on ICT by all federal offices, they could ensure the compatibility of the law with ICT solutions needed to implement regulations (SIK, 1978). We can assume that the ICT professionals did not only see coordination as an implementation challenge but also seek to include the logics of informatics into the legal basis of their actions. The report, however, stressed the ethics of public service behind their argumentation: “Automatization of the public administration only makes sense if it leads to an improvement of the relationship between the administration and the administrated people (SIK, 1978, translation from German by the author).

Conclusion

The case study shows that the change from analog to digital procedures has challenged and still challenges the public administration’s traditional procedures and ways of interaction. One of the key reasons for these difficulties is the practice and the strong culture of bureaucratic organization with a functional separation of duties. The idea of shared data use has, in the period studied, on one hand been promising in terms of effectiveness; however, on the other hand, it has been seen as a necessity given the high costs for IT and the alleged shortage of qualified personnel. Shared data use requires organizations to adjust their structures to allow efficient use of ICT. At the conceptual level, specialists have described the need for an overarching architecture and standardization of certain key definitions; the cultural change toward distributed responsibilities for core services beyond use in the own agency could not easily be achieved . While studies on IT architecture point out the slow process to adopt organization structures and allocation of budgets, this study presents evidence for cultural barriers to structural changes. While the Social Insurance Office slowly opened up to coordination, it did not share the OASI vision of the number as a national infrastructure for a new age of electronic administration. In contrary, key decision makers in this agency firmly insisted in remaining the exclusive owners of the number with sole power regarding decisions on change and adaptations. The creation of new organizational entities to manage and offer shared services was seen as difficult to justify in parliament.

The Swiss case shows that coordination and collaboration within the existing organizational structures had a smaller reach, but these actions allowed channeling and limiting the transformational impact of IT use. Within the cooperation structures of the SIK, standardization of the main definitions has been achieved, but major changes like the use of an OASI number as a unique identifier failed due to the lack of structures and commitment in defining and implementing underlying working instrument for shared data use. The limitations of technology were also a decisive factor. As data exchange in our example was, until the beginning of 1980, based on a physical exchange of magnetic tape, decentralized data handling and reuse were burdensome, and therefore, data transfer was expensive and time-consuming. The vision of a shared database had serious technical limitations.

The coordination efforts were also used as a vehicle to trigger a shift of power toward the ICT managers. Their aim was to change their role from service providers to experts of ICT-compatible lawmaking. To which extent this goal has been reached could not be established in the course of this case study.

From today’s perspective, we can observe that institutional change enabling a shared use of information and data across government agencies is still a key challenge for E-government in Switzerland. Coordination instruments have been strengthened (Brugger, Neuroni, & Fraefel, 2012), but a cross-agency use of data has only been successfully implemented in a few cases. The need for cross-agency coordination has further grown with the increased use of ICT in public administration. The cantonal administrations still emphasize the benefits of this key identification instrument. In 2014, the cantonal Directors of Finance renewed their interest in a unique identifier and asked the Federal Government to initiate the creation of legal ground for the use of the number in public administration (Finanzdirektorenkonferenz, 2014). The use of the OASI number for identification purposes would even be easier today as a 13-digit invariant and random number was introduced in 2008.

From a general point of view, the IT-model introduced shows that the organizational capacities for cross-silo collaboration and the role definition of IT and business are essential factors for the speed and the reach of transformation by ICT use in the public administration. While the challenge is universal and consists of a far-reaching cultural change, the strategies to deal with this change may vary from country to country. The primary challenge of adapting the structure of the public administration to IT use has been met in different ways and has derby created path dependencies due to legacy systems, traditionally grown roles and established patterns of problem-solving. These challenges prevail today. A comparative study of different countries would allow a better understanding of the fundamental change to public administration development which arose from the use of ICT.