Pension reform and public information in Poland

Chlon, Agnieszka Article Pension reform and public information in Poland Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies Provided in Cooperation with: Duncker & Humblot, Berlin Suggested Citation: Chlon, Agnieszka (2000) : Pension reform and public information in Poland, Schmollers Jahrbuch – Zeitschrift für Wirtschaftsund Sozialwissenschaften. Journal of Applied Social Science Studies, ISSN 1865-5742, Duncker & Humblot, Berlin, Vol. 120, Iss. 3, pp. 489-534, https://doi.org/10.3790/schm.120.3.489 This Version is available at: https://hdl.handle.net/10419/291966 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Schmollers Jahrbuch 120 (2000), 489-534 Duncker & Humblot, Berlin Pension reform and public information in Poland By Agnieszka Chlon1 Abstract Public information is a central part of the process of Pension reform and every stage from design to building support for change and to implementing the reform. An in-depth-analysis of Poland shows that a public campaign was a success: not least because mistakes were quickly corrected. Poland's experience holds some valuable lessons for other countries who want fundamental reform of their Pension systems. Zusammenfassung Wenn kapitalgedeckte private Renten eine größere Rolle in der Altersvorsorge spielen sollen, dann ist es notwendig, dass die öffentlichen Informationen über das neue System verbessert werden. Öffentliche Informationen sollten deswegen Teil einer jeden Reform in Richtung mehr kapitalgedeckter Renten sein. Der Aufsatz untersucht eine breit angelegte öffentliche Informationskampagne in Polen und kommt zu dem Ergebnis, dass diese Kampagne erfolgreich war; nicht zuletzt, weil viele Fehler schnell korrigiert wurden. Insgesamt kann Polens Erfahrungen mit einer öffentlichen Informationskampagne für andere Länder, die die kapitalgedeckte private Altersvorsorge ausweiten wollen, nützlich sein. JEL-Classification: G 18, G 23, H 55 Poland introduced its pension reform in 1999. The need for change was readily apparent both to experts - who pointed to huge system deficits and the rapid demographic transition - and to the public, who, as we shall see, wanted a more transparent and comprehensible retirement scheme. The old scheme had been very unstable, sometimes amended several times in a single year. Change affected every part of the system, including the benefit formula, pension age, eligibility criteria and indexation rules. Even the staff of the social security administration found it hard to keep track: eventually, there were more than ten different laws affecting the system. 1 The author is grateful to Robert Palacios for his support and to Edward Whitehouse for his comments. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 490 Agnieszka Chlon The new pension scheme aimed to offer a more transparent regime, with benefits more closely related to individual contributions in both the public, pay-as-you-go and the private, funded components. This paper begins with a description of the old and new pension schemes from the point of view of the contributors during the working period and after retirement. In particular, it highlights the features of the new scheme that increase the responsibility of individual members. The discussion is deliberately kept brief: the interested reader is referred to Chlon, Góra and Rutkowski (1999) for more details. The subsequent section charts the development of Poland's financial sector. Until the 1990s, this was practically non-existent. Retail banking was undeveloped, as most people preferred to keep their money 'under the mattress', often in foreign currency. Inflation had wiped out savings in domestic currency. Insurance was monopolised by two state providers mainly offering car and home-contents products. During the 1990s, however, the financial sector developed rapidly and the financial literacy of the population improved in its wake. Yet the question remains whether people were ready for a modern pension scheme and the choices they had to make. The following two sections describe the government's public-information campaign both during and after the reform's implementation. They also look at the marketing and sales strategies of the new pension funds. A concluding section draws out the lessons for other countries. 1. The pension system in Poland before and after reform 1.1 Basic features of the old pension system The old pension system in Poland was a traditional, defined - benefit system, financed on a pay-as-you-go basis. Employers paid social security contributions as a proportion of their total wage bill. So workers often had little interest in the mechanics of the financing of the scheme and even whether the employer actually paid contributions. Moreover, the social security administration (ZUS) did not keep records of the contributions made on individuals' behalf. On retirement, the employer submitted a form to ZUS, which calculated the pension entitlement. Pensions were not uprated until the 1980s, when inflation rapidly eroded the purchasing power of benefits. After a period of ad-hoc increases, automatic indexation procedures were introduced in 1991, with the pension uprated in line with earnings. There were many other changes at the same time, some of which were applied retrospectively This resulted in a recalculation of existing pensioners' Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 491 benefits, increasing the average pension from 53 to 61 per cent of average earnings. There were three components to the pension under the new scheme: • a flat-rate element of 24 per cent of average earnings, adjusted by a coefficient ranging from 91 per cent in 1991 to 100 per cent in 1999; • an earnings-related component of 1.3 per cent of pensionable pay for each year of contributions; and • a credit of 0.7 per cent of pensionable earnings for each year without contributions, due to bringing up children, participation in higher education etc. to a maximum of one third of contribution years. The introduction of earnings indexation was not favourable for pensioners in the early 1990s because prices increased faster than wages. But real wages grew after 1993 and, as a result, pension spending increased sharply. The subsidy to the social insurance fund from general revenues reached 5.7 per cent of GDP in 1993. The government was forced to act. Pensionable earnings were reduced from 100 per cent to 91 per cent of actual pay in 1993 and benefits were indexed to prices from 1995. The rapid spending increase in part resulted from the spread of early retirement. For example, people subject to group lay-offs could retire regardless of age if they fulfilled the contribution criterion (40 years for men and 35 years for women). Restructuring branches were also offered early retirement. Almost a quarter of the workforce was covered by some kind of early retirement privilege, which seemed a convenient way of reducing labourmarket pressures. However, there were no projections of the long-term impact of this policy. To conclude, pension legislation changed constantly during the 1990s under two powerful but conflicting pressures: the poor labour-market situation and the increase in the number of pensioners and the value of pensions. 1.2 Public opinion of the old pension system Opinion polls showed a lack of public confidence in the pension system in the mid-to-late 1990s (Figure 1). More than 40 per cent told the survey in 1997 that they thought the system was in a bad state and that change was necessary. Another 24 per cent described the system as quite bad. Fewer than 9 per cent thought the system performed well, and even fewer, that change was unnecessary. Mistrust in the old system correlated strongly with education: 72 per cent of people with higher education favoured change, compared with just 37 per cent of people with only primary education. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 492 Agnieszka Chlon against reform < favours reform • system don't system system is bad: good know quite bad changes necessary don't transparent know not transparent don't good know poor Overall views of the pension system System transparency Knowledge of pension formula 75 50 25 0 25 per cent of respondents 50 75 Note: Survey conduced in October 1997 on a representative sample of 1,168 adults. Source: CBOS (1997). Figure 1: Public opinion of the pension system, 1997 100 per cent of respondents I system bad: changes necessary 8S81IB • iliflSfiPlP system quite bad 8S81IB don't know system good August 1995 April 1997 October 1997 Source: CBOS. Figure 2: Public opinion of the pension system, 1995-1997 Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 493 Confidence appeared to be deteriorating over the period 1995 to 1997 (Figure 2). There was little change in the (small) proportion saying that they thought the pension system worked well. There was a decline in the numbers saying that they did not know about the system or that they found it difficult to express an opinion. The major change, however, was an increase of more than one half in the proportion that though the system was working poorly and that change was necessary. This increase drew mainly from the people who though that the system was quite bad (and in part from the 'don't knows'). The surveys also asked people slightly more specific questions on their views of the pension system. Again, the results were not encouraging. Almost 70 per cent said that the system lacked transparency and nearly two-thirds reported that they did not know well on what their pension entitlement would depend. These proportions, however, were more stable over time. 1.3 Basic features of the new pension system The government began to assess alternative reform options in 1994 but there was little consensus on the appropriate way forward. Nevertheless, the government appointed a Plenipotentiary for Social Security Reform and a supporting secretariat with responsibility for formulating a detailed reform proposal. The result, entitled Security through Diversity, envisaged a 'multi-pillar' pension system. A contribution of 19.52 per cent of earnings pays for the two mandatory components of the new system. Of the total, 12.22 per cent goes to finance a pay-as-you-go, public pension, while the remaining 7.3 per cent goes into an individual pension account. The public scheme is based on 'notional accounts', a pay-as-you-go system designed to tie benefit entitlements more closely to contributions and to adjust payouts automatically with individual choice of retirement age changes in general longevity.2 The second mandatory component is fully funded with member choice of pension fund. Contributions and investment returns accrue in individual accounts and the resulting accumulation is used to buy an annuity on retirement. The existing Social Security Administration (ZUS) continues to collect contributions centrally for both components and transfers contributions, where appropriate, to the chosen pension fund. A Pension Fund Supervision Office (UNFE) has been established to license pension funds and to monitor their performance. UNFE oversees detailed regulations. These set out, inter 2 See Disney (1999) for a description of such schemes. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 494 Agnieszka'Chlon alia, the type of charges that funds can levy and a guarantee of a minimum rate of return relative to the performance of all funds over a two-year period. The legislation also requires both pension funds and ZUS to send members annual statements of their accumulated account values and an estimate of their pension entitlement. The standard pension age remains at 65 for men and 60 for women but, with the withdrawal of early-retirement privileges, this implies an increase of five years in the effective retirement age. The main changes between the old and new systems are presented in Table 1. The new system is relatively simple compared, for example, with the United Kingdom.3 People can choose only one pension fund and the competing funds offer broadly similar products. Nevertheless, the new system requires what might be described as more 'conscious' participation from members. For example, they must monitor their pension account throughout their working lives and people born between 1949 and 1968 had to choose whether to split their contributions between the two pillars or remain in the public, pay-as-you-go system alone. Table 1 A comparison of the old and new pension systems Old system New system Contributors Financing Pension base Record-keeping Pension age First contact with ZUS Pension formula Employer Pay-as-you-go 10-year average earnings Documents collected on retirement Various early retirement privileges: actual retirement ages 59 for men and 55 for women On retirement Defined benefit: flat-rate component (24 % of average wage) and 1.3 % accrual rate Both employee and employer Mixed pay-as-you-go (12.22% of salary) and funded (7.3% of salary) Lifetime earnings Individual accounts with ZUS for each member 65 for men and 60 for women for nearly all workers On starting first job and receipt of annual report from ZUS Based on actuarial annuity calculation, using unisex life tables in PAYG pillar 1.4 Public opinion of pension reform Opinion polls were used to assess the public's views of different reform strategies as well as attitudes to the old system. The Office of the Government Plenipotentiary for Social Security Reform - hereafter, the Office for Pension 3 See Whitehouse (1998, 2000a). Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 495 Reform - used the polls to build support its proposals among key policymakers. Nevertheless, many of the questions are rather general and pollsters did not spell out the financial consequences of the proposed changes. Figure 3 sets out the main results of a survey of 1,200 adults from April 1997. Most respondents agreed that pensions should be accumulated in an individual account: around half agreed strongly and another third agreed. Only four per cent were unconvinced of this change. The strongest support came from the self-employed, people with higher education and white-collar workers. The least convinced were people who were not working, including old-age and disability pensioners and housewives. A second question asked whether pensions should depend on the amount and period of contributions. Nearly three-quarters of respondents agreed, with just one in five preferring that pensions should be more or less equal for everyone. Stronger support again came from better-off groups, including white-collar workers, the self-employed and people with medium or high household incomes. This might be interpreted as support for defined-contribution provision and for the notional-accounts system. However, respondents were offered only two options at the end of the spectrum of possible systems. no/ don't probably not know maybe yes pensions should be the same for ail PAYG don't know don't know pensions should depend on contributions funding I Should contributions go into an individual account? Contributory or universal basis Should pensions be paid from own contributions or by the next generation? 75 50 25 25 50 75 per cent of respondents Source: CBOS, April 1997. Figure 3: Public opinion of pension reform strategies The final question asked whether pensions should be provided on a payas-you-go or funded basis. A total of 70 per cent agreed that accumulations from individual contributions and investments should pay for pensions, Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 496 Agnieszka Chlon while just 13 per cent thought that pensions were best financed from the next generation's contributions. Education affected views significantly, with 80 per cent of people with higher education favouring funding compared with 59 per cent of those with just primary education. 2. Financial market development Support for funding might in part be a result of the rapid development of financial services in Poland in the 1990s. This section looks at the evolution of the banking system, the stock exchange, mutual funds and the insurance market. It also sets out the changing attitudes to saving and investments. 2.1 The Warsaw stock exchange The stock market and its regulator were established in 1991. The Warsaw Stock Exchange, the only licensed securities exchange in Poland, has grown rapidly over the last few years. From nine listed companies at its launch, the exchange in 2000 traded 198 companies. Of these, 117 were listed on the main market and 51 on a parallel, junior market for firms that do not meet capital or other requirements. Figure 3 shows that this resulted in an increase in stock market capitalisation that reached 20 per cent of GDP by 1999. By the end of 1999, there were 40 brokerage houses and eight banks carrying out brokerage activities. 25-, percent of GDP 2015105market capitalisation 5 market turnover 1991 1992 1993 1994 1995 1996 1997 1998 1999 Note: market capitalisation at end of year; turnover: annual. Source: Polish Securities and Exchange Commission. Figure 4: Stock market capitalisation and turnover: Poland per cent of GDP Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 503 Activities during this initial period included: • Surveying public opinion of the old pension system and publicising the results; • Designing a consistent look and feel' for materials issued by the Office for Pension Reform. The pension reform logo-representing the rising sunappeared on all output and was even sported on lapel pins worn by staff and 'friends' of the reform; • Training in communication techniques for Office for Pension Reform and ZUS staff; • Organising seminars for politicians, union members, employers etc.; • Establishing a training programme for people who would then go and present the principles of the new system to people in their workplace; • Publishing a set of brochures, targeted on employers, unions, different age groups etc.; • Media relations, including seminars for journalists, press conferences, press releases, offering interviews with senior architects of the reform; • Developing a pension reform web site (http: Wref ormaemerytalna. mpips. gov. pi). A central plank of the public-information strategy was to talk about the reform as if it were already fact, even though there were many hurdles still to overcome. This, along with opinion-poll results showing popular support, left very little room for opponents of reform to question the basic elements of the new system. The targets of this initial campaign were policy-makers and opinion-formers and mailings were initially aimed mainly at these groups. Explaining the success of reforms overseas was a key goal at this point. Government officials, members of parliament and journalists were taken on study tours of Argentina, Chile, Hungary and Sweden to learn about the pension-reform solutions that had influenced the Polish model.5 The study tours did help to build an understanding of the philosophy of different countries' approaches among policy-makers and opinion-formers. (It was particularly helpful that journalists could later build on this experience in explaining the reform to their readers, viewers and listeners.) Table 3 gives a feel for how the public-information campaign developed through the reform process. It shows, by subject, the press releases that were issued by the Office for Pension Reform. At the beginning, information 5 These study tours were financed from a World Bank Institutional Development Fund grant to Ministry of Labour and from USAID help. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 504 Agnieszka Chlon about international experience of pension reform and developments in the legislative process dominated the press campaign. The press releases were distributed to around 400 journalists in Poland. Table 3 Press releases from the Office for Pension Reform, 1997-1999 1997 1998 1999 Legislative process XX X XX X Opinion poll results XX X X XX Retirement age X Retirement privileges X XX XX New system XXXX XX X Information XX XX campaign Other X X X X Later on, the information campaign turned to the particular solutions adopted in Poland. The Office publicised the results of opinion polls and, in particular, made the case for a reform of early-retirement privileges. It is pertinent to note that little publicity effort went on the initial plan to equalise the pension age for men and women and that this idea was later dropped in favour of retaining unequal ages (65 for men, 60 for women). At the end of 1998, the need to shift the campaign's focus from opinionleaders to the general public was apparent. Only 30 per cent of people told a survey that they had heard of the pension reform and that they knew what changes were proposed. The majority (55 per cent) reported that they had heard about the reform, but that they did not know what changes were afoot. Moreover, 83 per cent of those surveyed felt that they were not sufficiently well informed of the rules of the new pension system, 77 per cent said that they would like to learn more. 3.2 Public information and the implementation of the reform A second, broader PR campaign began in March 1999. Its goal: to explain the new system, the choices that it offered, and the different options open to different age groups. The main discussion at the time was over the choice between remaining in the reformed pay-as-you-go, public scheme or splitting the pension contribution between the public scheme and the new funded pension accounts. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 505 The Office for Pension Reform published simulations showing the circumstances under which members were likely get a better pension under each option. However, the pension funds' marketing campaigns often included contradictory assessments. Some, for example, excluded the effect of administrative charges on the pay-as-you-go component (levied by ZUS) or on annuity values. Consequently, the Office for Pension Reform developed special software to calculate the projected size of the pension under both options. This could be downloaded from the Office's web site. The public information campaign had a series of elements, although it was based around a call centre, which opened in March 1999 until the end of that year. Eleven brochures, covering all aspects of the reform, were prepared. A press and TV campaign promoted the call-centre's services. There were four different TV spots. The first, which advertised the socalled 'four social reforms', covered a broad range of subjects. The other three spots were produced solely for the pension reform campaign. They targeted different age groups affected by the change, and aimed to promote ZUS as a new, customer-friendly institution. These spots, transmitted nearly 400 times, reached almost 96 per cent of adults living in urban areas (i.e., they saw the piece at least once). There were also nearly 24,000 published press advertisements and Polish Telecom sent out 7,500 coupons allowing people to apply for the brochures. The results of the first few months of the media campaign were somewhat disappointing: for example, the number of brochures ordered fell short of expectations. Focus group interviews revealed that people did not differentiate between the pension reform campaign and commercial advertising spots on TV Knowledge of the pension reform was sparse: people could recall the slogans but did not understand what they meant. Reform was associated in the public's mind only with the introduction of the funded pillar. Although some parts - such as the fact that savings in the defined contribution pension accounts could be inherited - were well known, other components of the reform were not, particularly: • the 'pillars' of the new pension system; • changes to the pay-as-you-go scheme; • the concept of 'initial capital', reflecting rights accrued in the old system; • the choice between splitting the contribution between funded and payas-you-go elements or putting the whole contribution into the revised pay-as-you-go scheme. The most popular sources of information on the pension reform were discussions with friends and pension funds' agents, press articles and pension funds' publications. ZUS was perceived as a symbol of the old system: conSchmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 506 Agnieszka Chlon tacting the social insurance institution was seen as something of a last resort. To conclude, the public information campaign at this stage was not very effective, and people relied instead mainly on the pension funds' for information. Recognising these problems, the Office for Pension Reform changed its media plan and the design of the campaign. First, new TV spots were recorded, featuring the Plenipotentiary, Ewa Lewicka describing the call-centre and the availability of brochures at a press conference. Also, the spot was billed as a 'government advertisement' rather than a 'paid advertisement'. There were over 300 transmissions of these new spots, reaching some 96 per cent of Poles over age 30 living in urban areas. Secondly, radio was added to the media list because it is often listened to during working hours and it is cost-effective. Furthermore, the use of radio allowed for different messages to be presented in different regions. The radio campaign was intensified after the holiday season, because the youngest contributors had to choose their pension fund before the end of September. Thirdly, there were also changes in the press campaign. The new formula comprised coupons and leaflets ('The new pension system' and 'Initial capital') inserted into daily newspapers, which included the most popular tabloid. New, full-colour advertisements featured the most frequently asked questions, according to the call-centre operators. The performance of the campaign improved as a result of these changes. The call-centre received an average of 3,100 inquiries a day in September, compared with just 200 a day in August. Focus group interviews showed that people had begun to recognise that the information provided by the pension funds was subjective and too general. Also, the public saw, for the first time, a need for government information on the pension reform, which was seen as objective and reliable. The appearance of the Plenipotentiary was a success: people regarded her as a reliable source of information about the pension reform and a symbol of government activity in providing information. In particular, the focus groups felt that there was a need for clear and impartial information on the following questions: • are pension funds reliable? • what are the differences between specific funds? • how should people decide between splitting the pension contribution or staying in the public scheme? • what are the government guarantees of pension fund performance and pension benefits? Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 507 Perhaps the most disappointing part of the campaign was the relatively poor perception of the call-centre, which was typically viewed as the last resort in the search for information. There are many potential explanations. First, the call-centre had a Warsaw telephone number, which meant that many people had to pay long-distance charges. Secondly, the public was unused to the government providing them with detailed information on any subject and so might not have though it worthwhile using this service. Finally, radio, TV and press are passive sources of information, while inquiring at the call-centre requires activity. Unfortunately, opinion poll and focus group results are insufficiently detailed to evaluate these alternatives. A regional campaign, in addition to differentiated radio broadcasts, centred on a tour of the eight largest cities in Poland (outside of Warsaw) during May and June 1999. The Plenipotentiary headed the road-show, meeting employers' and trade union representatives and the local media. A pension reform web site was launched in August 1998. Press releases and legislative information were added to the site as they become available. By the end of 1999, the site had registered more than 100,000 hits, mostly from Poland. International visitors included people from Canada, France, Germany, Sweden, the United Kingdom and the United States. The campaign saw five million brochures distributed and 200,000 inquiries at the call centre. In total, it cost around $5 million. ZUS also participated in informational campaign: its 2,000 trained employees advised contributors how to send in the necessary information and contributions. Its branch offices distributed free software, enabling contributors to transfer information electronically. However, during the implementation of the reform it became apparent that ZUS was administratively ill equipped for its new responsibilities (especially transferring contributions to pension funds). As a result, ZUS's public relations activities were dominated by crisis management. A new president of the institution, appointed in October 1999, prepared an 'emergency plan' to sort out the most pressing problems. The pension fund supervisory agency, UNFE, ran a limited information campaign, mainly because of budgetary constraints. The agency organised open-days in its offices and published a bulletin, but both of these had a very limited reach. As with the Office for Pension Reform, its officials appeared on radio and TV programmes and took part in debates and conferences. Much of the agency's publicity effort was aimed at the 'third pillar' of voluntary schemes, supporting employers and unions in setting up occupational pension plans. UNFE issued a report, 'Security through Competition', which argued that the public was not aware of the charges levied by different funds and the central impact that they have on pension fund Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 508 Agnieszka Chlon returns.6 UNFE would have done well to publicise this information and producing comparative data - 'league tables' - on different funds' charges. Instead, the supervisory agency proposed a complex and difficult to implement change to the cost structure. Although some have argued that supervisory agencies are best not involved in consumer education7, UNFE's role seems to have been too minimal, leaving a gap in the provision of impartial, comparative information on different pension funds. 3.3 Pension reform and the media The government campaign was, naturally, not the only source of information. For example, TV companies prepared (independently) a series of special features broadcast from late 1998 through 1999, including: • a series of short (five-minute) programmes - 'Which pension?' - transmitted twice a week on the first channel of public TV; • longer, fifteen-minute programmes - 'Questions about the pension reform" - on the second channel of public TV; • a series of programmes on the private Polsat network. Members of the Office for Pension Reform were frequently invited for interviews, to take part in debates and in audience phone-in programmes. However, the case of the Polsat programmes bears further discussion. Polsat owns 40 per cent of one of the pension funds. Initially, the TV programme appeared impartial. But once the Polsat pension fund received its licence, it began to sponsor the programme. The Office for Pension Reform decided that no government official could then participate in the programmes, and they became simply an advertisement for the Polsat pension fund. Press coverage of the pension reform was also broad, beginning with the passage of the first set of legislation in June 1997. The focus of press attention was the 'second pillar', the new funded element in the pension system. In 1997, most (68 per cent) articles were purely informative, with a smaller number offering some commentary on the wisdom of the pension reform (according to surveys by the Profile agency). The majority of these commentaries were favourable, arguing that the reform would improve macroeconomic performance, make the system more transparent and ensure a decent and secure retirement income for current workers. Press interest remained at a constant level throughout the first half of 1998, as there were few new developments in the design or implementation 6 See Whitehouse (2000b,c) and World Bank (2000) for a discussion of strategies on charges and their impact on members' returns on pension contributions. 7 See the discussion and references in Whitehouse (2000a). Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 509 of the reform during that period. Again, most articles were simply informative (79 per cent). Commentary was again largely favourable: 53 per cent of articles were positive, according to Profile. Some 17 per cent of commentaries, however, criticised the reform. Most of these appeared in the Try buna daily newspaper, the successor of the communist daily, Trybuna Ludu. In the second half of 1998, press coverage intensified as did reform activity. The main subjects of press articles were: • motions to establish pension fund managing companies; • the work of the pension fund supervisory agency, UNFE; • guidelines on the pension reform and advice for readers; • the legislative process; • information on ZUS and its readiness to implement the reform' • reports on public opinion polls, particularly those on attitudes to earlyretirement privileges, decisions on participating in the second pillar and choice of pension fund; • pension reform and pension systems in other countries. Again, only a minority of articles in the second half of 1998 were commentary on the reform. Most were also positive about the reform. But this changed in early 1999, with the implementation of the pension reform, one of the so-called 'four social reforms'. Nevertheless, the criticism focused on the other reforms: the changes to the pension system were perceived as the best prepared. Some of the attacks included criticism of the public information campaign and others of the slow transfer of contributions to the new pension funds. The subject matter of press coverage shifted over time. In 1999, most articles were about various aspects of the new funded pension sector: the plans and activities of the pension funds, the new UNFE agency. The next most important subject was the calculation of initial capital in the new public pension scheme (a third of articles in May, for example). Around 15 per cent of articles were about ZUS and its activities. Overall, the media played a very important role in informing the public of the impending changes to the pension system and their impact. 3.4 The results of the public information campaign The effectiveness of the public information campaign can be gauged by opinion polls conducted throughout 1999 on consumers' awareness of the pension reform and the changes that were involved. Figure 10 shows how Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 510 Agnieszka Chlon the public rated their knowledge of the changes. The proportion saying that they felt their knowledge of the system was adequate or better increased from less than half in March 1999 to nearly 80 per cent in March 2000. Just 16 per cent said they though they lacked information at the end of the period, compared with nearly half at the beginning. 50 25 0 25 50 75 per cent of respondents 1 very good/ average adequate don't know inadequate none March 1999 June 1999 October 1999 March 2000 Note: Surveys based on representative samples of around 1,100 adult Poles. Source: CBOS. Figure 10: Consumer ratings of the degree of information they have received about the pension reform A second survey question asked people how accessible they found information about the pension reform. Figure 11 shows the results, which confirm the pattern of Figure 11. In June 1999, around half of respondents thought information was easy to find, increasing to more than 70 per cent in March 2000. Most people - 91 per cent - reported that they were aware of the government information campaign in October. Nearly all of these had seen the television spots, nearly half had heard radio commercials and a similar number had seen newspaper advertisements. Almost a quarter heard about the campaign from their friends. Just 3 per cent said they had used the government call-centre (see above). In June 1999, a survey asked people their opinion of the government information campaign. Some 40 per cent criticised the campaign, although a third assessed it as good. Interestingly, views on the campaign varied hugely with education level. Less than a quarter of people with only primary education thought rated the campaign as good, compared with half of people with higher education. This might perhaps inSchmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 511 dicate that the presentation of the campaign was too complex, despite efforts to set out the reform package in the simplest possible way. 75 50 25 0 -25 easy quite easy don't know quite difficult difficult -50 -75 J June 1999 October 1999 March 2000 Source: CBOS. Figure 11: Assessment of access to information on the new pension system 4. Pension funds' marketing and sales activities in 1999 4.1 Pension funds' marketing campaigns Pension funds were allowed begin their marketing campaigns on 16 February 1999 under the watchful eye of UNFE, the pension fund supervisory agency. UNFE intervened when advertisements were unfair or misleading. Wilczynska (2000) divides the pension funds' strategies into three categories (Table 4). The first group attempted to build their strategy on the existing position of their parent companies. These were all well-known providers of financial services in Poland before the reform. The second group adopted a similar strategy, but these funds were new entrants to the Polish market or were building on a very small presence. The final strategy involves companies already active in Poland. But these firms tried to build a new brand for the pension fund business. Since pension funds were essentially an entirely new industry, the last strategy might not have too many disadvantages: all players are new entrants. However, the most effective approach seems to have been the first, Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 512 Agnieszka Chlon although this was, of course only open to firms already well established in Poland. Experts rated the Commercial Union's as the best campaign among the larger funds (see Stepien, 2000). It was based on testimony from well-known celebrities. The Commercial Union avoided making significant errors - unlike its main competitors - and was rewarded with the largest market share. PZU hired an inexperienced agency, which made uninspiring, easily forgotten advertisements. PZU's media spending concentrated more on the press than its competitors. The press advertisements included interviews with the president of the fund. There was no obvious strategy underlying the campaign, and its direction seemed to change frequently. It did little to change the image of the company, widely perceived as a badly managed giant and a relic of the past. PZU even stopped advertising when it reached its market-share target. Table 4 Marketing strategies of the pension funds (1) (2) (3) Campaign based on parent company brand: already well known Campaign based on parent company brand new to Poland Campaign creates new brand Commercial Union AIG Bankowy Kredyt Banku Allianz Dom Nationale-Nederlanden Arka-Invesco Ego PBK Orzel Norwich Union Epoka Pekao / Alliance Polsat Pocztylion Pioneer Winterthur Rodzina PZU Zlota Jesien Zurich Solidarni Skarbiec-Emerytura Source: Wilczynska (2000). Nationale-Nederlanden got off to a false start, transmitting their first advertisements before 16 February. In response to this transgression, UNFE forbade N-N from using any of these spots, and new advertisements had to be recorded. The company had just three days to design a new concept for their campaign and record the advertisements: and it showed. Fortunately for N-N, the parent company had a good reputation that survived these difficulties, and the firm ended up as the third largest pension fund. The fourth largest fund decided against using the well-known name of its parent - Amplico - and operated under the relatively new AIG brand name. After a relatively poor start, its campaign improved, using linked TV spots and printed advertisements. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 519 12 cumulative number of oarticiDants. million April June August October October December Source: ZUS. Figure 15: Cumulative number of members of private pension funds, 1999 Again, these results can be compared with people's intentions (Figure 14). Around 80 per cent of 31 -40 year olds opted for a private pension. Some 57 per cent of a sample of this age group told the June survey that they intended to take out a private pension and 28 per cent were undecided. Total participation of 41-50 year olds was around 30 per cent, which compares with 20 per cent of the June survey that said they intended to participate and 25 per cent were undecided. This implies that the younger undecided group mainly decided to take out a private pension, while the older undecided group mainly opted to remain in the public scheme. Finally, Figure 16 also shows an interesting difference between the sexes. Up to age 40, women were more likely to join a pension fund than men of the same age. But after age 40, more men than women took out a private pension. Surveys also asked people their reasons for choosing to join the funded component of the new pension system. People could give up to three answers: Figure 17 shows the main results. Participation in the funded pillar was primarily motivated by the desire for a higher pension or for a more secure retirement income. These were, of course, the central themes of the pension funds' marketing campaigns. The third most important factor was a 14 See Palacios and Whitehouse (1998), Disney, Palacios and Whitehouse (1999) and World Bank (1999a). Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 520 Agnieszka Chlon bequest motive: funded pension balances can be inherited unlike entitlements in the public scheme. The problems of ZUS convinced 17 per cent of respondents that a private scheme would be better than the public system (managed by ZUS). membership of private pension scheme, per cent of age cohort age in 1999 Note: based on January 2000 data. Source: Gdansk Institute for Market Economics. Figure 16: Participation of 1949-68 cohorts in private pension funds The second panel of Figure 17 shows the main reasons people gave for not joining a private pension fund. In both cases, one of the most important determinants was the compulsion in the system: people born before 1968 had to join a private pension fund and people born after 1948 had to remain in the public scheme. The largest group - over 40 per cent - believed that there would be too short a period until retirement for an adequate sum to accumulate in a private plan. Worryingly, 35 per cent of the youngest age group - 18 to 24 year olds - also gave this answer, although they were, of course, forced to participate. Another sizeable group said that they wanted to retire early, and this is not possible for people who join the private pension funds. Individual characteristics, apart from age, did not influence the decision to participate in the funded pillar. It is also interesting to compare the numbers who remained in the public scheme (less than 10 per cent) because they Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 521 did not trust the private funds with the numbers who took out a private pension plan because they did not trust ZUS (17 per cent). 17a: People who decided to split the pension contribution Higher pension More secure pension Born after 1968 Inheritance Persuaded by agent fc—^M—I ZUS's troubles | Persuaded by family/friends | Persuaded by funds' ads 0 10 20 30 40 50 17b: People who decided to remain in the public scheme alone Higher pension Born before 1948 Want early retirement Lack of trust in funds Funds' fees too high 0 10 20 30 40 50 per cent of respondents Source: CBOS, March 2000. Figure 17: Reasons given for different pension choices Markowski et al. (2000) modelled the factors influencing the decision to participate in the funded component of the pension system using this data set. The results, consistent with Figure 16, show declining participation with age. This is the dominant effect. Other groups with higher private pension take-up included the urban population, people who already had a life insurance policy and people who expected to live longer than the average. Figure 18 compares Poland with four Latin American reformers and with the United Kingdom. The Figure shows the cumulative market share of penSchmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 522 Agnieszka Chlon sion funds, ranked from the largest downward. The largest three funds in Poland have 53 per cent of members, a very similar proportion to Argentina and Colombia, slightly more than Mexico and rather less than Chile. Chile has the most concentrated market: there are now only eight funds and the three largest have three-quarters of the market. The United Kingdom, in contrast, has a much less concentrated pension market: indeed, the ten largest funds have a similar market share to the three largest in Poland. cumulative market share, per cent 100 Colombia Argentina .j75 Chile ' Poland 50 i UK \ Mexico 25 123456789 10 number of pension funds Source: UNFE (Poland), FIAP (Latin America) and HSBC James Capel (United Kingdom). Note: Market shares measured in proportion of total members, except in United Kingdom (proportion of total assets). Figure 18: Concentration of pension funds: Poland, United Kingdom and four Latin American countries 5.2 Choice of the pension fund Who were the winners in Poland? The 'big-three'-Commercial Union, PZU Zlota Jesien and Nationale-Nederlanden - were all established insurers. PZU has a life-insurance subsidiary, the other two are life-insurance specialists. Their existing networks of insurance agents made up the core of their sales forces. Banks also entered the market but they were not particularly successful. This is a marked contrast to the situation in Hungary, where retail banks own most of the largest pension funds. The rest of the market is divided unevenly. There are several mediumsized funds (with a market share of 3 to 5 per cent) and a tail of smaller Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 523 funds, with less than 300,000 members each. It is this tail of smaller firms that distinguishes Poland from the Latin American reformers, but it seems reasonable to expect some consolidation in the future. A regression analysis of pension fund membership (Table 6) shows that the size of the network of agents was the only significant factor determining different pension funds' success. Regressions on the fund's amount of capital and total media spending are significant, but only because they are correlated with the number of agents. Once that variable is included, capital and media spending become insignificant. Interestingly, if advertising spending is broken down into three different media, spending on television advertising has an insignificant effect, but the amount spent on radio and press campaigns is significant. Indeed, even in the model including the size of the sales network, spending on radio advertising has a significant effect on the number of members at the 7 per cent confidence level. Figure 19 plots pension funds' market share against the number of sales agents. It also includes the predicted values from the quadratic term from the equation in Table 6. Although the quadratic shows a diminishing rate of return on additional agents as the size of the sales force increases, over the relevant range here, extra agents still have a positive effect. The success of Commercial Union, Nationale-Nederlanden and PZU is explained mainly by the size of their sales networks. The first two, however, were more successful than the sales network alone would explain (since they lie above the fitted curve). AIG, in contrast, performed much more poorly than the size of the sales network would imply. This might perhaps reflect the decision of its parent company to develop a new brand name rather than use its own name - Amplico - which was well known in Poland. A further survey asked people if they could spontaneously recall the name of a fund and any advertisements for funds. Market share was very strongly correlated with knowledge of the name of a fund (a coefficient of 0.92). The relationship with spontaneous knowledge of a fund's advertisements was much weaker (correlation coefficient of 0.54). Together these results imply that the sales force and a familiar, already developed brand name were the most important determinants of initial market share. Attitudinal surveys also shed some light on consumers' choice of particular pension funds and the factors they took into account. Figure 20 shows the results of one survey, which asked people to categorise different factors of primary importance, of secondary importance or not important for their choice of fund. The issues are ranked in the chart by the proportion of people rating them as primarily important. Consumers rated the size of the fund' capital as the most important issue, with 72 per cent giving it the top Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 524 Agnieszka Chlon rating of significance. Levels of service were also rated highly (51 per cent), both in general and in the specific area of the provision of information (66 per cent). Table 6 Regression analysis of pension fund membership (1) (2) (3) (4) (5) (6) R2 0.224 0.204 0.712 0.732 0.747 0.874 F-test 0.034 0.037 0.000 0.000 0.000 0.000 constant -374.8 (394.5) -335.8 (229.8) 543.2 (255.4) -174.6 (336.2) -110.7 (449.6) capital 11.08 (9.167) 4.346 (6.643) 3.263 (8.969) capital2/100 -2.630 (3.662) 1.195 (2.441) -1.454 (3.241) F-test 0.0341 0.6861 0.8575 Media spend/1000 75.65 (27.15) -31.65 (21.91) (Media spend)2 /1000000 -1.191 (0.4223) .4579 (.3012) F-test 0.037 0.337 TV spend/1000 8.204 (45.19) -15.71 (34.92) (TV spend)2/1000000 -0.2632 (.7753) 0.2300 (0.6149) F-test 0.3008 0.7748 Radio spend/1000 -768.9 (332.0) -68.18 (244.8) (Radio spend)2 /1000000 157.1 (57.78) -23.36 (49.26) F-test 0.0186 0.0722 Press spend/1000 27.92 (320.6) -90.45 (94.29) (Press spend)2 /1000000 26.58 (40.56) 22.65 (29.13) F-test 0.0083 0.6089 Number of agents 4.531 (1.878) 4.875 (0.2970) 5.147 (1.839) (Number of agents)2 0.0024 (0.0016) -0.0027 (0.0019) -.0026 (0.0016) F-test 0.0000 0.0000 0.0038 Note: Dependent variable: number of members of fund. Sample: 21 funds. Standard errors in parentheses. F-tests are the significance level of the joint test of exclusion of both quadratic terms. Standard errors have been adjusted for heteroscedasticity using standard methods. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 525 number of agents, log scale Source: UNFE, OBOP and AEC data. Figure 19: Size of pension fund sales force and market share Given the discussion of the pension funds' various marketing strategies in section 4.1 above, it is interesting to see that the experience of the pension funds' parent companies - both in the Polish market (63 per cent) and internationally (58 per cent) - were mentioned by many consumers. Some consumers also reported that they stuck with a familiar provider of financial services (29 per cent). Fund performance and strategy had the second highest proportion of people rating this factor as of primary importance (68 per cent), with another 48 per cent claiming that their fund had and/or would give them a better rate of return.15 A second survey allowed consumers to name the three most important influences on their choice of pension fund from a slightly longer list (Figure 21). Again, characteristics of the pension fund - that it was perceived as more secure or better performing - were rated the most important. This survey allowed rather more candid responses than the first. For example, a quarter of people said that one of the reasons they joined their specific fund was simply that this was the first agent they had met. Others reported that they had shopped around: nearly a fifth said they chose the best offer after 15 This is, of course, unlikely: very few studies have found any persistence in the performance of fund managers. These studies are surveyed in Whitehouse (2000a, 2000c). Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 526 Agnieszka Chlon meeting several agents and slightly more than a fifth that they had done some personal research of different funds. A fifth of people said they joined the fund because a friend or relative was an agent. (Although this number seems rather high, recall that there were 440,000 registered agents, around one per cent of Poland's adult population.) Very few people said that either their employer or their trade union16 had influenced their decision. Fund manager's capital Investment strategy Fund information Shareholders' domestic experience Shareholders' international experience Fund service Fund returns Fund size Charges Recommendation Advertisements Personal experience of shareholders Fund office nearby 100 Note: Based on a sample of 1,544 people aged 17 to 50 in April 2000. Source: ARC, OFE Monitor, May 2000. Figure 20: Factors influencing the choice of pension fund from a consumer survey Both surveys show that few consumers claimed that they had been influenced by the pension funds' advertisements. They also show that pension funds' charges influenced only a very few consumers' decisions, despite the potential effect that these levies can have on the net pension value. There is some evidence of potentially illegal activity. Employers were prohibited from influencing their employees' choice of pension fund, agents were not allowed to offer inducements to join a particular fund or to promise higher investment returns than other funds. While the first two transgressions appear to be rare, more than a quarter of people told the second survey that they were convinced their pension fund would offer the best return and nearly half said that the fund's high return was of primary importance in their choice. It must unfortunately remain moot whether people had simply 16 Ghilarducci and Ledesma (2000) analyse and explain the limited role of trade unions in the pension reforms in Argentina and Chile. Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 Pension reform and public information in Poland 527 convinced themselves that their fund would out-perform or had been led to believe this by advertisements or agents. Security Potential performance Recommendation First agent encountered Personal research Related to an agent Followed friends/relatives Met several agents: chose best offer Fund manager's capital Advertisements Agent promised highest pension Charges Personal experience of shareholders Employer influence Inducement to join Trade union influence Note: Based on a sample of 1,116 adults under age 60 in March 2000. Source: CBOS, March 2000. Figure 21: Factors influencing the choice of pension fund from a consumer survey It is also possible to look at the reported influences on fund choice by membership of different pension funds using the second survey (from CBOS). People who joined one of the 'big three' - Commercial Union, PZU, Nationale-Nederlanden - tended to stress the security of their pension. Some 56 per cent of PZU members gave this response, compared with 37 per cent of the whole sample. Another significant result was that one in three Commercial Union members argued that this choice would assure the highest pension, compared with a quarter of all funded pension members. PZU's network of agents - the largest network, nearly a quarter of the total number of agents - seemed to have been a success. Its clients were (slightly) more likely to say that they joined this fund because this was the first agent they met.17 Nationale-Nederlanden clients tended to join that fund because of personal recommendation (32 per cent versus 25 per cent for the whole sample) or after personal market research (32 versus 22 per cent). Unfortunately, the sample sizes were too small to get statistically significant results for specific smaller funds. 17 Note that PZU's network was based mainly on the insurance agents in large enterprises - typically people in the personnel or accounting departments - who were well placed to approach potential members early Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55 528 Agnieszka Chlon 3 o •5b <V J a; tí W 2 ^ g'§) ^ S 2 £ B S 5° cd -g tí V cd r¡ -e w) PK A I 0 s O) T5 1 .2 'S o m S rr* m a u tí a» tí üjn ? « «w tí Ti cd Ü ¡D N tí S^ 0.2x3 •a «¡3 S S S a; PQC0«1 a,« Tí tí ^ £ S > iS Ö PQ.S S tí o 'ü cd rO cd Ol a; .cf o 'S l^s S s « Cd £ -M T) W Ö O > O Ü.Sfc TS O O bJD cd S CT1 -M tí «a cd £ ^ £ O > ü.S £ o tí W cd cd II o cd S " cdTJ 3 cd íí o<u cd otó > •h S sa «'S 'S SP 8 s dl tí a; bß cd +-> w Jh MH r^ ¿B U CD 3 tí w O Schmollers Jahrbuch 120 (2000) 3 OPEN ACCESS | Licensed under CC BY 4.0 | https://creativecommons.org/about/cclicenses/ DOI https://doi.org/10.3790/schm.120.3.489 | Generated on 2023-04-04 12:27:55