German pension policies: the transformation of a defined benefit system into ... what?
1 Changes in the Retirement Age
Changes in the retirement age add up to very important changes concerning, on the one hand, the raising of the mandatory retirement age and, on the other hand, important reductions in pensions because of early retirement. Both are intended to stop and also to change the incentives for having a low labor force participation for older workers.
The German pension system allows for early retirement within a period of 36 months before the mandatory retirement age. During some important reforms in the late 1980s, the government started to increase the regular retirement age for women, the unemployed, those insured for long periods and severely handicapped persons to 65 years. During some important reforms in the late 1980s, the government started to increase the regular retirement age for women, unemployed, long insured and severely handicapped persons to 65 years. Early retirement was and still is possible, but for the first time it was coupled with a reduction of the pension level of 0.3 percent per month. Using the maximum period of 36 months, it amounts to a diminished pension of 10.3 percent. If workers decide to work longer, the pension will increase by 0.5 percent per month. However, at present only a very small number of pensioners make use of that option. The incentive structures of these regulations are clear: the advantage of working beyond the mandatory retirement age will be greater than the disadvantage of taking early retirement. The law was enacted in 1989 and came into force on 1 January 1992.
In 2007, the grand coalition between the CDU and the SPD under Chancellor Merkel decided to raise the retirement age again from 65 to 67. To be clear, starting in the year 2012 and with the birth year of 1947, the retirement age will be increased by one month per year and per birth year. Only by the year 2029 will the retirement age reach the target of 67 for all those born in the year 1964. That means that people born in 1958 must work until the age of 66. All in all, these gradual increases in the retirement age make it possible for future pensioners to adjust to the changes and to become clear about the consequences for the course of the individual's working life. Working longer will be still fostered and early retirement discouraged by the reductions mentioned above.
A heated debate set in mainly within the Social Democratic Party, but also within the unions and among the general public, while the unions and parts of the Social Democratic Party, primarily the leftist wing, started to protest the measures taken by the government. This, perhaps, will force the party leadership to change their position in the years to come.
The intentions in increasing the retirement age are clear. Firstly, in the face of greater life expectancy, the duration of pension payments is reduced and, as a consequence, the expenditures for pensions will be diminished. Secondly, the number of contributors to the pension system will be higher because future pensioners will have incentives to work longer in order to avoid pension reductions. And thirdly, if pensioners are forced to retire due to bad labor market conditions for older workers or simply decide to do so, their pensions will be lowered by a maximum of 10.3 percent, which in turn cuts down on the costs of the expenditure side of the pension system.
All in all, these measures mark a decisive change in the retirement policies in Germany. During the 80s, early retirement was a very important method for large firms to lay off older workers using the early retirement regulations of the pension and the unemployment systems. In addition, the unions and big enterprises negotiated contracts for early retirement. The situation was exacerbated with the introduction of the so-called PreRetirement Act of 1984 (Vorruhestandsgesetz) which contained several new measures designed to promote early retirement, but made it contingent on the existence of industrial agreements between the unions and the employers' organizations. Early retirement was made possible via collective labor agreements or through occupational retirement plans. As a consequence, a complete institutional regime for early retirement developed and functioned on the basis of an interplay between the pensions and the unemployment systems (c.f. Trampusch 2005).
The act allowed early retirement at the age of 58 and set the minimum benefit at 65 percent of the last gross income. In addition, improvements could be made possible by agreements between unions and companies or by collective bargaining. It was mainly the chemical industry, but also the metal, mining and other industries, which made use of the Vorruhestandsregelung to lay off older workers. Parallel to this, the duration of unemployment benefits was extended for older workers and for workers with a long duration of compulsory contributions. They were now entitled to thirty--two months of unemployment benefits (Trampusch 2005: 210). The various laws now made it possible to retire at the age of 57 because workers could receive unemployment benefits in the form of Arbeitslosengeld for three years and then receive pensions due to unemployment at the age of 60. Because companies often subsidized the loss of income during the period of unemployment benefits, and because pension entitlements due to unemployment were paid out without any reductions, the Vorruhestandsregelung and other regulations developed into a regime of early retirement which was frequently made use of, mainly in the energy and mining sectors, but also in the iron, steel and automobile industries, and which was backed up by the social partners and the main political parties. During German reunification, early exit from the labor market became the most important instrument in dealing with the economic crisis in the Eastern part of Germany. Between 1993 and 2002, about 675,944 East Germans retired early due to unemployment, and in 1992, 30 percent of all withdrawals from the labor market in Eastern Germany took place through pre--retirement, age-bridging, and pensions due to unemployment (Trampusch 2005: 211). However, things began to change as it became increasingly clear that those strategies were lumping the social security systems, especially the pension and unemployment systems, with huge financial burdens. During that time, the contribution rate constantly increased and non-wage labor costs put heavy burdens on German industry. In addition, growing unemployment drove the contribution rate up further, while demographic changes made it clear that the pension system would be burdened with additional costs in the near future. In 1998, the average retirement age in West Germany was 59.7 years for male workers and 60,6 years for female workers. In the Eastern part, or the New Federal States, it was 57.7 and 58.3 respectively (BOrschSupan and Wilke 2005: 583).
2 From Two to One: The Reforms of Invalidity Pensions
From the very beginning, invalidity pensions had been an integral part of the German pension system. To be more precise, under Bismarck, social security in cases of invalidity was the most important part of the pension system. Most people did not reach the regular retirement age and thus gained pensions only in cases of disability. Most workers simply died before reaching the general retirement age.
Since the introduction of the Old Age Security System under Bismarck in 1889, invalidity pensions have been an institutionalized part of the German pension system (for details c.f. Tennstedt 1972). Under the first law of 1889, workers were able to claim invalidity pensions if they were unable to earn more than one third of that which workers with a similar occupational education were able to earn. Only blue-collar workers were eligible for invalidity pensions. In 1911, old age and invalidity insurance was also introduced for white-collar workers, but with important differences. Invalidity was divided into two levels of disability. It introduced for the first time in German history the so-called occupational disability pension (Berufsunfahigkeitsrente) to which persons were entitled if they were only able to earn half of the income which a worker with similar occupational skills would earn. It also--like the old age system for blue-collar workers--provided for invalidity pensions to which persons were entitled if they were not able to earn more than one third of the earnings of a comparable worker. A two-class pensions system thus existed during the Weimar Republic and the Nazi Regime which gave privilege to white-collar workers (Privatbeamte bzw. Angestellte) over blue-collar workers. Only in 1957, together with the great pension reform, was blue-collar invalidity split up into the two levels to match that of white-collar workers.
The German statutory pension scheme (GRV) thus consisted of two levels of invalidity:
The first was the loss of at least 50 percent of the working capability in case of illness or working handicaps (Berufsunfahigkeitsrente). Insured persons were defined as suffering from occupational disability if their earning capacity was reduced to less than half of that possessed by a physically, mentally, and intellectually healthy worker of similar training and comparable knowledge and ability. The range of activities on which the insured person's earning capacity was to be judged included all those activities which were relevant to their occupational training, their previous occupation and particular demands of the previous occupation activity. The worker, however, was expected to earn an additional amount of income within the labor market, which meant working at a part-time job that could be reasonably expected of them considering their occupational training (Berufsschutz). Deciding on the appropriateness of jobs for occupational disability was a severe problem and the social courts started to develop and define a highly complicated scheme. In the end, a hierarchical scheme was used with declining steps of qualifications: (a) superior/highly qualified workers, followed by (b) skilled workers, then (c) semi-skilled workers, and finally (d) unskilled workers (c.f. Viebrok 2003 (1)). If, for instance, a highly skilled worker was unable to find an appropriate job on his level or one level below, i.e. for skilled workers, the pensions system presumed that the labor market was closed. He was not assumed to pick up jobs beneath these qualification levels. As a consequence, occupational disability was then transformed into full invalidity.
The second level was the loss of full working capability, in which case workers were eligible for a full invalidity pension (Erwerbsunfahigkeitsrente). Both pensions were to be paid out until the worker reached the regular retirement age of 65 years. From then on, the payments would be a regular old age pension. The calculation of the two disability pensions is essentially quite simple. Missing periods before the age of 60 are filled in by fictional years. Up to the age of 55, they are based on the relative income position a worker has at the beginning of his/her invalidity. The period from 55 to 60 years is calculated with only 1/3 of this period. For instance, if a worker is entitled to full invalidity at the age of 45, the missing period up to 55 years will be calculated on the basis of 100 percent of his/her earning points, that means for an average earner exactly 1 earning point per year (for that see below 4.). The period between 55 and 60 is calculated on the basis of 1/3, which means only 18 months are credited and this period will amount to 1,5 earning points (instead of a possible 5) (Viebrok 2002: 208). Because the invalidity pensioner will come into the system at the age of 60, the earliest possibility of receiving an old age pension, he/she will accept the deduction of 10.8 percent because of early retirement.
Problems arose mainly on the level of the occupational disability pensions due to the above mentioned jurisdiction of the social courts. In all those cases in which the labor market was unable to provide for appropriate part-time jobs for occupational disability pensioners, those persons had been able to claim a full disability pension. This was called 'concrete treatment' (konkrete Betrachtungsweise). As such, an occupational disability pension was only paid out if the labor market was able to provide for appropriate part-time jobs. If labor markets were 'closed', the pension system took over the labor market risks. At the end of the 1990s, the government introduced major changes. The most important changes were decided on in December 2000 and came into force in January 2001. Two main changes are worth mentioning. Firstly, the introduction of a uniform disability pension, and secondly, the abolition of the so--called "konkrete Betrachtungsweise" and the change to the 'abstract treatment' (abstrakte Betrachtungsweise) (for details see Wollschlager 2001).
To begin with, the SPD-Green government introduced a two-level disability pension which is now called pension for 'reduced working capacity' (Rentenwegen verminderter Erwerbsfahigkeit). Partial disability (teilweise Erwerbsminderung) because of illness or handicap is awarded if an insured individual is not able to work at least six hours a day. The pension is limited to a period of three years and may be extended if the medical conditions are still present. If an insured individual is only able to work less than three hours a day, he/she will receive the full disability pension (Rente wegen voiler Erwerbsminderung). And finally, if a person is able to work more than six hours a day, he/she will receive neither a partial nor a full disability pension. On the contrary, he/she is to be expected to find a job in the labor market in accordance with his/her abilities and training.
Pensions because of full disability are paid out until the affected person reaches the statutory retirement age. However, risk sharing between the pension system and unemployment insurance has changed. If a partially disabled person is unable to find a job for around three hours a day, he/she will not receive benefits from the unemployment insurance because it is not possible to place him/her in a job, but that person is still eligible for a full disability pension. If a person is able to work between three and six hours and the labor market is closed, which means that the unemployment insurance will not be able to place him/her in a job, the current abstract treatment (abstrakte Betrachtungsweise) will make him/her eligible for unemployment benefits rather than for a full pension, as was the case previously. The disability pensions now no longer have the function of substituting wages (Lohnersatzfunktion), but rather of compensating for the reduced ability in one's working capacity. The person is still expected to try to make use of his/her reduced working ability within the labor market. This is why the disability pension is only multiplied with the so-called pensions type factor (Zugangsfaktor) of 0.5, reflecting the remains of the person's working capacity. For old age pensions, the pension type factor is 1, for example.
3 Adjusting Badly or Appropriately? Pensions Adjustments from Gross-Wage Indexation to the Sustainability Factor
Pension indexation is one of the most important elements of every pension system for two reasons. Firstly, the income positions of the pensioners who have often paid contributions for over 40 years must be updated to a value which reflects the current living conditions. Only then is an income oriented pension able to realize the normative principle of guaranteeing the standard of living. Realizing the standard of living simply means that the pension system should only in exceptional cases redistribute income positions beyond the market. As a rule, the pension should reflect the spread of incomes in the labor market within the benefit defined positions in the welfare state. The so-called 'insurance principle' is designed to ensure that nobody would receive an undue income position in the welfare state which would not reflect his or her merits in the labor market. Secondly, every pension indexation formula has to make possible the participation of the pensioners in the current economic development. If pensions were fixed or not index-linked, the income of pensioners would lag behind the incomes of the workers, and thus in the long run expose them to poverty.
In principle, various mechanisms for pension indexation are possible. Pensions are adjusted either on the basis of the wages or on the basis of the development of prices. The first principle may connect the workers' struggles for higher wages with the life situation of the pensioners, and thus may couple the workers' interests for higher wages with those of the pensioners for higher pensions. This was one of the great fears leading to arguments against the so-called dynamic pensions introduced in 1957 under the Adenauer government. Conversely, the second principle may decouple workers' and pensioners' interests because consumer prices may be influenced by various factors. The interests of the pensioners for higher pensions would then stand alone without having any incentives to join interests with others. However, pensions are often adjusted by means of mixed adjustment formula that takes into account both the development of wages and of consumer prices.
Since the introduction of the dynamic pension (dynamische Rente) in 1957, pensions had been indexed by the development of average gross wages. The indexation formula took into account the average of the previous three years. The flat-rate amount in the old pension formula was abolished because of the basic idea that from now on, pensioners would participate in the economic development and the income positions within the labor market would allow for pensions guaranteeing a decent standard of living in old age. During the sixties, financing was changed to a strict PAYG system; before that, financing had been organized on the basis of a strange mixture of capital funding and PAYG financing. In addition, the PAYG system made possible the increase of payments to present pensioners as well, and not only for future pensioners. The adjustment formula took into account the average gross wages of the three years preceding the year in which the pensions had to be adjusted. The consequence of this was that in the economic crisis in the mid 70s, pension adjustments took place on the basis of high growth rates of gross wages in the past, which then had to be paid out in a time of economic crisis and decreasing contribution rates. During that time, the PAYG system ran into a deep crisis and political decisions had to be made to stop the adjustment of the pensions and to prevent increasing contribution rates. In addition, pension adjustments took place on the basis of gross wages; an increase of the contribution rate as well as the tax rates lowered the net income of the employees whereas pensions were adjusted on the basis of gross wages and, in addition, were mostly not burdened by income taxes. Therefore, the ratio of the net earnings of pensioners compared to the net income of employees increased more and more (and would continue to do so in the future). These problems stimulated discussions about a new adjustment formula based on net wages.
On 9 November 1989, the day the Berlin wall came down, the German Bundestag decided on a variety of measures, the most important of which was a new adjustment mechanism. One of the most important-but only technical-reforms was the simplification of the pension formula. It now consists of four factors which account for the value of a pension [P.sub.t,i]:
[P.sub.t,i] = [EP.sub.i] x [RF.sub.i] x [ZF.sub.i] x AW
The first three factors are responsible for the individual pension base, whereas the fourth factor determines the indexation of the pensions and thus the income relationship between the pensioners and the current workers. AW is the most important factor open for the contingencies of politics and which may be used to introduce changes in order to vary the pension level and thus bring into balance the relationship between contributors and pensioners. Hence, the 1989-reform (and the following reforms) introduced major changes into the indexation formula, whereas changes in the first three factors had been rare.
Earning points (EP) reflect the relative income position of an individual within the labor market. If an individual earns within one year the average earnings of all insured workers, he/she receives the value of 1 EP. If a person works part-time and earns exactly half of the average income, he/she receives 0,5 EP; earnings which are twice as high as the average figure amount to 2 EP. To be sure, EPs do not reflect the added value of contributions during a lifelong working period, but simply reflect the relative income position one has gained in the labor market. EP's can also be earned in child-rearing periods, due to military service, or, until 2004, due to higher education. In accepting non-wage periods, the government is able to influence the individual pension level by introducing or withdrawing periods for which EP are credited on a non-contributive base. During unemployment, the unemployment insurance pays contributions to the pension system on the basis of the current unemployment benefits. Compared to pension index formulas in other European countries, workers cannot drop certain periods in their working life and choose others for crediting.
The pension type factor (RF) (Rentenartfaktor) denotes the mode of pensions one receives. For old age pensions (and for fully reduced working capacity) the value is 1, for disability pensions it is 0.5 and for survivor's pensions it is 0.55.
The pensions accrual factor (ZF) (Zugangsfaktor) is simply 1 if a person reaches the mandatory retirement age. However, it is reduced by 0.003 per month for retirement prior to the standard retirement age of 65 years and increased by 0.005 per month of deferred retirement beyond 65 years. It reflects the individuals choice to retire which is often contingent upon the situation of the labor market a person is challenged with and/or one's state of health. As mentioned above, the mandatory retirement age will be increased to 67 years beginning in 2012 and ending in 2029.
The current pension value (AW) (aktueller Rentenwert) is the most important figure determining the pension level. It links together the workers' earnings and the pensioners' benefits and is indexed to annual changes in the wages and salaries achieved in the labor market.
As mentioned above, until 1989, the pension's value changed according to gross wages. From 1992 on, the indexation was linked to the development of net wages in general. This change was financially the most important one in the reform package which was decided upon in 1989 and came into force in January 1992. However, the then incumbent Kohl government quickly realized the shortcomings of a mere after-tax adjustment. If, for instance, the contribution rate in other social security systems decreased (which was the case for unemployment insurance) or taxes decreased (because of political reforms due to the economic crisis), pensions would increase. As a consequence, the government again changed the indexation formula and introduced the modified net adjustment. Now, the pensions decrease when the contribution rate for the pensions increases and vice versa. All other changes in the net income of the population are now strictly out of computation.
In 1998, the outgoing Kohl government introduced an additional factor that changed the current pension value. It was the demographic factor which simply took into account changes in the life expectancy of the population (for details c.f. Rurup 1999). If it increases, for example, the pensions will be reduced, and vice versa. However, the data used in the calculation of the demographic factor were from 1990/1991 and not the data of the year of the introduction of the new law. The reason was simple: In the years after 1995, the degree of the increase of the life expectancy did not amount to that directly after reunification in the years 1990/1991 to 1995. Therefore, the years after reunification reduced the pensions slightly more than those taken from 1999 on (Rurup 1999: 45). As one will see in what follows, those discretionary political decisions became part of the pension policies in the years to come. However, the demographic factor never became operative.
Due to electoral promises, the incoming SPD-Green government skipped the demographic factor and adjusted pensions according to the inflation rate for two years (2000 and 2001). Again, this was a completely discretionary decision taken by the new government and was legitimized with the argument of buying time for working out a coherent pension reform. In the end, concerning the adjustment of pensions, the SPD-Green government eventually introduced the so-called sustainability factor which now indexes the pensions. It was decided upon in 2005 and came into force on 1 January 2006.
Its final version is based on the proposal of an ad hoc commission which was established to work out ideas for a sustainable development of the social security systems. Besides a multitude of other proposals, it favored a new pensions adjustment formula, which was adopted by the government. The basic idea of the new formula is the self-adaptation of the pension system to external changes and contingencies in its social and economic environment. What are the main points of the sustainability factor (for details c.f. German Social Advisory Board 2004; Hain et al. 2004; Reimann 2004)?
Firstly, it takes into account not only changes in the life expectancy (like the demographic factor of the outgoing Kohl Government), but also changes in the entire demographic development, including changes in the birth rate as well as the net balance of immigration and emigration which is reflected in the so-called 'standard-contributor' or the 'standard pensioner'.
Secondly, it reflects major changes in the labor force. Any decrease in the labor force is reflected in the adjustment formula and will lower the pension level because of the decrease in contributors. On the other hand, an increase in the female labor force will boost the financial resources of the pension system and thus lead to higher pension adjustment. However, the relation of contribution payers and pensioners will be calculated on the basis of the so-called--equivalence pensioner". The introduction of this fictional figure was intended to avoid an artificial decrease of insures and pensioners. That could happen, for example, with a simple increase of employed persons whose average income is extremely low because of part-time work, low income or both. On the other hand, that could, for example, occur with an increase in the number of low pensions that result from short insurance time periods. Neither would reflect the real financial situation of the pay-as-you-go system (PAYG). The whole pool of pensioners is converted into 'standard pensioners' where each has a number of 45 insured years with an average income of all those insured. Also, the number of contributors is converted into 'standard contributors' because their number is calculated based on the number of employed persons on an average income. As a consequence, the 'standard contributor' not only reflects the net balance of immigration and emigration but also demographic changes. As a consequence, the 'standard contributor' not only reflects the net balance of immigration and emigration, but also demographic changes.
Thirdly, the complex adjustment formula is multiplied by a factor a which has the value of 0.25. This factor is the result of various computer simulations which were undertaken to find a value that is able to limit the contribution rate to 20 percent in the year 2020 and to 22 percent in 2030. In general, it is an arbitrarily chosen factor which decision makers may easily change in the years to come. This became clear in the statement of the government when introducing the new law. It stated that--the parameter [alpha] (...) will temper the effects of the sustainability factor and will steer the aim of fixing the contribution rate at 22 percent in the year 2030." (BT-Drs. 15/2159: 58; own translation) It only burdens the pensioners with 1/4 of the possible effects of the changes in the relation between the standard contributors and standard pensioners. If a were 0.5, the contribution rate would be at 20.8 in the year 2030; if the pensioners were fully burdened with the effects of these changes, with a thus being 1, the contribution rate would be only 18 percent in 2030 (c.f. Hain et al. 2004: 347, table 7). However, the government stated that it would retain the option of changing the alpha--factor if things change and jeopardize the aims set by the government.
And fourthly, following the introduction of the--RiesterRente" in 2001, the share paid into additional old-age provision schemes (Altersvorsorgeanteil) is now also part of the pensions adjustment formula and consequently lowers the level of the state-run pension (for details see below 4.)
All in all, the sustainability formula (for details see Annex 2) is a nearly perfect realization of the idea of the self-stabilization of the pension system in a troubled, uncontrollable and unforeseeable world. It internalizes external contingencies into its own institutional devices and thus guarantees its survival. In addition, it is a de-politicization of pension politics because of the self-stabilization mechanism of the pension system. However, in view of the recent federal elections in 2009, pensions politics was characterized by a re-politicization. Both parties of the grand coalition, the Social Democrats (SPD) and the Christian Democrats (CDU/CSU) decided to suspend the sustainability formula which would have lowered the pensions during the electoral campaign. A mere vote-seeking strategy thus gave room for political opportunism and the decision to "freeze" the pensions instead of lowering them. Politicians were afraid of the electoral consequences of their decisions made a few years before. In the face of the rising German debt and despite the huge demographic challenges, Chancellor Merkel defended the suspension of the sustainability factor during a debate in the German parliament as--an important measure". (2) However, this measure will lump future contributors with huge financial burdens in the years to come.
Annex 2: The sustainability adjustment formula of the German pensions system:
A[R.sub.t] = ARt - 1 * [[B[E.sub.[t - 1]]]/[B[E.sub.[t - 2]]]] * [[100 - AV[A.sub.[t - 1]] - B[S.sub.[t - 1]]]/[100 - AV[A.sub.[t - 2]] - B[S.sub.[t - 2]]]] * ((1 - [R[Q.sub.[t - 1]]]/[R[Q.sub.[t - 2]]]) * [alpha] + 1)
* [AR.sub.t] = new current pension value to be determined
* [AR.sub.t-1] = current pension value in effect (West)
* [BE.sub.t-1] = average gross wage earned during the last year (if t=current year)
* [BE.sub.t-2] = average gross wage earned during the year before last; here changes in the gross wages (which are liable to contribution from 2006 onwards) per employed person on average income are taken into account, inclusive of recipients of unemployment benefits but excluding civil servants
* [BS.sub.t-1] = last year's pension insurance contribution rate
* [AVA.sub.t-1] = share paid into additional old-age provision scheme in the year t-1
* [AVA.sub.t-2] = share paid into additional old age provision scheme in the year before the year t-1.
* [RQ.sub.t-1] = pensioner quotient = equivalence pensioner/equivalence contribution payer in the last year
* [RQ.sub.t-2] = pensioner quotient = equivalence pensioner/equivalence contribution payer in the year before the last year
* [alpha] = sustainability factor of 0.25
Source: Social Advisory Board 2004: numerical 59.
The introduction of the sustainability factor may have dramatic consequences for the effectiveness of the pension system. It will "socialize" its beneficiaries by reducing the pension levels of average income earners close to or under the poverty line in the years to come. This will undermine the legitimacy of the whole system because it is no longer able to generate the still normatively proclaimed aim: namely to guarantee the standard of living achieved in the labor market. The sustainability factor constantly lowers the current pension value (AW) and, as a consequence, the net pensions level of new pensioners and that of current pensions. Concerning the years between 2005 and 2030, the current pensions value will lag behind the development of the wages by about 0.7 percent per year, which will amount to an absolute value of 17 percent in 2030 (Hain et al. 2004: 341). Concerning the net pensions, it will decrease from about 65 percent in 2010 to about 52 percent in 2030. Within 20 years, the pension level of a standard pensioner will decrease by about 13 to 15 percent, thus lowering the pension accordingly. This general reduction of the pension level (let alone the effects of unemployment and shorter or interrupted working careers) will reinforce the effect that a great number of employees, even after long periods of paying compulsory contributions to the pension insurance, will inevitably receive a pension that may even be below the level of social assistance security (especially in East Germany) (cf. Sachverstandigenrat 2007, 193-194; Schmahl 2006a and 2006b; DIW 2010).
4 A New Public-Private Mix in Pension Policy: The 2001 Pension Reform Act
In November 2000, after more than a year of controversial discussions and substantial alterations, the Ministry of Labor and Social Affairs presented the bill of a two-part pension reform act. (3) It eventually differed strongly from the ambiguous basics of the governmental statement and introduced both a partial privatization of the pension provisions and, simultaneously, a gradual weakening of the role of public benefits. After parliament (the Bundestag) had passed the first part of the bill--the new pension formula (AVmEG)--in January 2001, the second chamber (the Bundesrat) approved the second part-the new financial support and stimulation of private provision for old age (AvmG)--in Mai 2001. (4) The AVmG passed in both chambers in spite of the resistance of the Christian-Democratic opposition who had tried to block it in the Mediation Committee (Vermittlungsausschuss) (cf. Lamping and Rub 2004; 2006).
The cornerstones of the 2001 Pension Reform were the most fundamental changes in German pension policies since 1957 and add up to a system change. The 1957 reform introduced the concept of dynamic pension and-closely linked to it-that of its wage replacement function. Now it is the inseparable combination of both public and private pensions--which are closely connected with each other--that will take over the function of wage replacement. The AVmEG (5) provided a reform within the institutional framework of social pension insurance. It aimed primarily to reduce benefits and costs of the PAYG scheme in order to stabilize the contribution rate. Its core policy was the new formula for the adjustment of pensions which aimed at slowing down or even stopping the increase of the contribution rate necessary to balance the social pension insurance budget. According to rather optimistic estimations made by the Federal Ministry of Finance at that time, the contribution rate should not exceed 20 percent in 2020--despite an increasing ageing of the population--and subsequently should remain below 22 percent up to 2030. Additionally, the government committed itself to keeping the contribution rate below these percentages and to maintaining the promised pension level of 67 percent of average net earnings.
According to this conception, from 2001 onwards pension adjustments will follow the development of net wages only in a restricted manner (for details see above 3.). To keep the complex pension formula simple, two elements will have the effect of deductions or pension-level reduction factors: firstly, the current pension value (aktueller Rentenwert) will be adjusted on the basis of average wages taking only the development of the current level of pension contributions into consideration. One notable consequence is that the adjustment of pensions will be decoupled from other policy areas. Pensioners will therefore no longer profit from positive (or negative) changes in contribution rates for health care, long-term care, unemployment insurance, and income taxes. The new formula will lead to minor adjustment rates relative to the development of (real) average net wages.
Secondly, from 2003 onwards, the pension adjustment will be reduced in proportion to the increase of the new private provision expenses (up to 4 percent in 2008): The fictional assumption about the increase of the private provision share decreases in proportion to the adjustment of public pensions. This makes sense because it directly influences the level of net wages and, from now on, the development of the public pension level. This provision share is a fictional one for at least two reasons. There is apparently no direct relationship between this share and the real expenses of the insured individuals for additional private income. Moreover, this concept of proportional deductions assumes and implies that there will be a comprehensive introduction of private provision which might be a crucial point vis-a-vis the fact that the private pillar is not mandatory.
In short, the first part of the new pension reform puts an end to the net-wage adjustment of pensions which was introduced in 1992 in favor of a so-called "modified net adjustment": an increase in the pension insurance contribution rate and in the fictional savings rate from 2001 on has proportionally reduced the level of pensions. It is not economically induced wage developments but rather politically constructed ratios which adjust pensions. However, as a result of the new alteration of the pension formula, the 'net standard pension level' (after 45 years of contributions) will slowly but gradually decrease to about 67 percent (of average net earnings minus the assumed savings rate for private old-age provision) in 2030. That figure was regarded as the absolute minimum level. (6)
This internal reform of the PAYG scheme was only made possible on the basis of two developments: the first of these was the introduction of a new basic financial security in old-age (a sort of social assistance for needy pensioners) as a response to the steady decrease of the pension level and as a "safety net" for those who are unable or unwilling to make additional private provision. The second was an external supplementary and voluntary financing of old-age security: the encouraging of individuals to make greater private provision for pensions. In this respect, the second part of the 2001 Pension Reform Act-the AVmG (7)-introduced a system change in the structure of financing and providing old-age security. It is"a first step towards the privatization of responsibility for pension provision" (Green et al. 2007, 135). Now, private (and occupational) schemes are supposed to compensate for the cuts in public insurance. While fostering a new shift from social insurance benefits to voluntary private saving plans, German old-age security has become a hybrid system. The new public-private mix combines the concept of a mandatory earnings-and contribution-based defined benefit public scheme with that of a voluntary, defined contribution scheme. In fact, while the public pension scheme will gradually lose its importance for adequate old-age income, private capital-funded defined contribution schemes will increasingly have to fill the provision gap. The two schemes are closely linked via the new pension adjustment formula which implies a fictitious parallelism: the decrease of the pension level and the increase of private provision (plus rising public subsidies).
The new additional private pension provision is not compulsory as originally planned by the Ministry of Labor in June 1999. Instead, it actively motivates employees through subsidies or tax incentives to act in accordance with the overall political aim of increasing private and occupational pension provisions. After signing a private insurance contract (or a functional equivalent) ("Riester-contract") which is certified by a new state agency, those compulsorily insured in the public pension scheme have been entitled to an annual allowance since 2002. The tax-financed subsidies are mainly targeted at low-and average-wage earners, families, and single parents. The financial support is being introduced step by step. In 2008, those persons who invest at least 4 percent of their gross income (including the public subsidy) in long-term savings plans will receive up to [euro] 150 (married couples up to [euro] 300) as a maximum allowance. Additionally, a maximum amount of about [euro] 180 a year for each child is available for those eligible to receive a child allowance (Kindergeld). From that year on, the state intends to support private saving plans with an annual amount of approximately [euro] 11 billion via direct allowances or tax relief. In order to maintain the individual standard of living in old age, entitled persons have to pay both the relatively high contributions (about 20 percent) to the public pension scheme and (from 2008 onwards) 4 percent of their income for private provision. This is the biggest increase in the contribution rate since 1957 but it is a "hidden" increase (Nullmeier 2001: 77).
We conceptualize this radical policy change in old-age security as a fundamental departure from former policy paradigms and policy beliefs (cf. Sabatier (ed.) 1999; Sabatier 1998; Hall 1993).Moreover, we emphasize that the new market (and the new demand) for private pensions is politically constituted and politically regulated. The pension reform gives rise to a new double role for the state in pension policy: a combination of both a re-distributive welfare state and a regulatory welfare state that constitutes and regulates "welfare markets" (Nullmeier, 2001; Hyde et al. 2003; Bode 2007).
Since the 1980s, Western welfare states have been privatizing and marketizing social services and programs. In the wake of these developments of adopting more market-like arrangements in social policy, the multifaceted phenomenon of such welfare markets plays an expanding role. There is no such thing as a welfare market; rather there is a plurality of concepts of welfare markets embracing different levels and degrees of marketization and public regulation alike. Functioning, regulation, embeddedness and cultural underpinnings differ from one welfare state to the other and from sector to sector (cf. Bode 2007). All the types of welfare markets, however, have a common feature, though implemented to a different degree, which is the notion of more (provider) competition and individual (consumer) choice as a mode of resource allocation on these politically constructed and regulated welfare markets: Providers compete for consumers, consumers as buyers dispose of bargaining power and make consequential choices between alternatives, while the market is regulated by public bodies in order to achieve explicit social objectives defined via social law. The "welfare market" is not simply a sort of a regulated market (where governments are typically concerned with market failure, producer power, market entry regulation, etc.); rather it is both a variant and an explicit means of a new social policy. Welfare markets are functional equivalents of state activity. According to this functional perspective, they produce goods and services formerly produced by the welfare state or which are traditionally or culturally perceived as falling within the responsibility of the state.
Since the 1995 reform of the long-term care insurance, and especially since the 2001 pension reform, the German welfare state has experienced what Blomquist (2004), with regard to the privatization of Swedish welfare services, has called a "choice revolution". Yet it is especially in pension policy where the German government bluntly admitted that the partial privatization of old-age income would reduce the financial burden on the pension insurance and public budget alike. The welfare market for additional private provision is a partially subsidized market with low to medium public regulation: Insurance companies have to fulfill certain minimum criteria (market entry criteria) in order to qualify for a "RiesterProdukt", whereas consumers are subsidized by the state (equality of opportunity) and protected by a certain degree of product standardization (rate of interest, deposit guarantee, etc.). The new German policy of fostering additional private provision is rather undetermined in the sense that it relies heavily on one's own initiative and 'good choices'. Regarding this shift in the public--private--mix in German pension policy, it would be misleading to conceptualize the private sector as the mere opposite of the public sector and, in addition, to see its increased role as a simple policy of "retrenchment". On the contrary, the welfare state purposely provides room for the expansion of private markets. The market does not replace the welfare state but becomes an essential part of it. In this respect, the 2001 German pension reform is a striking example of the creation of a new market via social policy. Creating, promoting and regulating this market are now part of the social policy of the democratic state. In other words: the more the welfare state withdraws from guaranteeing collective entitlements and an adequate level of social protection, and the more individual well-being becomes dependent upon private choices and behavior, the more people are obliged to become informed managers of their own social security portfolios--and the higher the necessity is that social policy is flanked by an elaborated and explicit consumer (protection) policy: only if supplied with such a consumer protection policy or framework are consumers able to behave in an adequate and self-responsible manner which, as the idea goes, eventually provides more responsiveness, effectiveness and inclusion.
5 Creating New Consumers in Pension Policy. Creating New Gaps and Cleavages?
Much has now been written on the advantages and disadvantages of PAYG schemes as well on the promises and pitfalls of moving from PAYG to funding or a combination of both (for a good synopsis see Barr and Diamond 2009a). There are important fiscal, financial and individual implications of moving from one system to another which have to be taken into account. Whether or not such a shift in the welfare mix improves welfare depends on the specific context in which funded systems work, on the specific objectives they have to meet, and on the specific institutionalization and implementation of the "third pillar" and its interconnection with the other pillars. Here, the design and regulation matter--but the role and protection of "consumers" matters as well.
The German multitier pension system, in particular its private pillar, is a fairly peculiar one and, to put it bluntly, tends to be part of the problem and not part of the solution. Signing a Riester contract is voluntary, but is subsidized by the state and, theoretically, is attractive especially to the groups mentioned above. However, a private pension provision to substitute the state's provision can only work to the extent that those concerned are both capable and willing to invest in private savings plans. There was a broad implicit consensus within the new government that there is significant potential for private savings among the working population. Apparently, that view does not cover this issue in its entirety.
Generally, due to the political decision to retrench the statutory pension insurance and to preserve the social and income status of the beneficiaries, the state has become the "hostage of the capital markets"(Lenze 2010: 13; own translation) and dependent upon the proper functioning of the "third pillar"--which apparently is a fairly optimistic strategy given the severe and unparalleled financial crisis beginning in 2008 (see, with respect to the impact on social security and pension funds, Pino and Yermo 2010). Apart from that particular problem of funding, Le Grand (2007: 213) claims that extending user choice and provider competition "will provide a higher-quality, more responsive, more efficient and more equitable service than the alternatives". But "equity" is certainly the Achilles' heel of such concepts. If the aim is to extend choice and competition in order to improve public services (in the broadest sense) to serve better the needs not of the well-educated and the better-off, but of the population as a whole, then "empowerment in welfare markets" (North 2007) is crucial and of utmost importance in order to make informed and better choices. Without this empowerment of consumers, the poor and disadvantaged will continue to be losers in an unequal and unfair competition for welfare resources and opportunities. This is most likely the case in situations where decisions which may have serious consequences have to be taken under the conditions of in-transparency, uncertainty, asymmetries of knowledge, poor quality of information (and the like)--i.e. in a situation dominated by providers and professionals. Regarding the transformation of the German pension system, this situation is even more challenging.
Although the take-up rate of Riester pensions is steadily increasing, there is reason to fear that private provision is still far from being broadly accepted, especially by those affected the most by the retrenchment policies. Part of the explanation is that state subsidies and tax deductions certainly represent incentives to take up a Riester pension, but they cover only a small part of the expenses for private provision. Another part of the explanation is the difficulties associated with self-responsibility and consumer choice in German pension provision which are fairly numerous, and may be unacceptably high and hard to overcome (Lamping 2009). Drawing on evidence from behavioral economics (cf. Tapia and Yermo 2007; Tversky and Kahnemann 1992; Thaler and Benartzi 2004), people very often fail to make choices that improve their long-term well-being, and often make no explicit choice at all, which comes as no surprise, given imperfect information, high complexity, a poor sense of (future) risks, passive behavior (and the like) (see also Barr and Diamond 2009b, 9-12).
Employing data from the German Socio-economic Panel 2007, Lamping and Tepe (2009) investigated the question as to which individual determinants influence the decision to enter into a Riester contract. They came to the conclusion that the task of the state is not only to foster the readiness to plan and the corresponding competence in a "pedagogical" manner, but also to strengthen the financial ability of those individuals who are most in need of it. Moreover, Lamping and Tepe argue that a proportion of the state subsidies goes to those people who would have engaged in private retirement planning anyway (or have done so already), even without state-funded incentives, which suggests a bandwagon effect. Therefore, in line with Corneo et al. (2007), they conclude that higher subsidy quotas would hardly increase the proportion of households engaged in saving among low income earners. Furthermore, the consequence of the new German policy of fostering private provision is not only an individualization of one's personal risk, it is also the massive increase of income inequality in old-age and of old--age poverty (growing stratification). In pension policy, it is to be expected that additional private provision will further promote the processes of social selection and lead to problematic distributive effects, precisely because it is voluntary.
Coverage is still uneven across population groups. Younger people, low to mid--income workers, part-time or self-employed workers, the long-term unemployed and workers with discontinuing job careers, migrants, and women with insufficient earnings-related pensions all have particularly low coverage rates (as well as low public pensions). Since a huge number of people still cannot afford to take out private pension contracts, the danger of old--age poverty is particularly high in Germany, as the OECD recently warned. (8) In this respect, depending also on the role of private provision within the overall system, such supplementary systems can become a cause for severe concern about future adequacy, in particular for the most vulnerable groups (see also Social Protection Committee 2008). Moreover, those who make wrong decisions (or make no explicit choice at all) risk being "punished" by the capital market in terms of losing money and/or being dependent on basic financial security in old-age.
In a nutshell, the challenge for German pension policy is to move in the direction of a greater equality of choices (and thus greater equity) in supplementary pensions. Otherwise the consequence will certainly be a further increase of old-age poverty, uncertainty and insecurity--along with rising costs for the tax-financed basic pension.
6 Contributions and Taxes: The Myth of Contribution-Financed Pensions
Since the introduction of the pension system in 1889, the system has been financed by contributions and taxes; it was never a system which was based exclusively on contributions. In addition, the German pension system is not part of the state budget, but is separated from it. All insurance systems--health care, unemployment, long term care and pensions--are financed as separate and independent systems on the basis of a PAYG system. But only from 1969 on was the financing principle a clear PAYG system in which a minimum financial reserve (Mindestrucklage) existed. During that time, it amounted to the sum of three month expenditure. In 1992, the government reduced it to the sum of only one month, in 2002 to that of 0.8 months, and from 2004 on to that of only 0.2 months. To be sure, the real financial reserve is often higher than the minimum reserve. But from 1995 until 2003, the financial reserve was constantly below the legally defined minimum, and in 2005 the actual reserve again dropped below the minimum financial reserve. In such a strictly organized PAYG, the problem is simply that even marginal changes in the economic and financial environment of the pension system display consequences in an immediate imbalance between contributions and pension expenditures. This is always an enormous challenge for a defined benefit system that wishes to guarantee a normatively justified level of welfare benefit. As mentioned previously, there has always been a general state subsidy for the pensions system. However, its meaning has changed significantly. The pension reform in 1989, which came into force on 1 January 1992, reflected the ongoing trouble with the principle of defined benefits still dominant. It became clear that due to changing national and international economic conditions, a defined benefit system could only survive with permanently increasing contribution rates. As a consequence, the government coined a new principle which reflected a phase in which a paradigm shift started to become apparent, but was not fully realized. The principle was revenue-based expenditure policy (einnahmeorientierte Ausgabenpolitik) which was indeed never thoroughly realized. Instead, the government started to bring in new money in the form firstly of increased state subsidies, and secondly of raised taxes. Since 1 April 1998 there has been a supplementary federal lump--sum subsidy which was financed by increasing the VAT by one percentage point. In addition, since the beginning of the year 2000, revenues from a so-called "ecological-tax", which is part of taxes on petrol and other energy sources, have been exclusively running into the pensions system. Both measures are intended to stabilize the contribution rate and, at the same time, still make possible a defined benefit pension.
The general federal subsidies for the pension system increased steadily and rapidly during the post-war period. In 1990, the federal state subsidies amounted to about 22.5 percent of the revenues for the pension system; in 2009 they amounted to more than 31 percent. In absolute numbers this was a total of [pounds sterling]57.3 billion in 2009. However, one has to take into account the fact that the pension level was instantly lowered and, at the same time, that the federal state subsidies have to be constantly raised and that the contribution rate always has to be increased; now oscillating at about 19 percent. Although there was a partial turning away from the defined benefit principle and the change to a revenue-based expenditure policy, the pension system did not get along without supplementary funding. The additional subsidy from the federal budget (zusatzlicher Bundeszuschuss), which was introduced in 1998, was steadily boosted from about [euro] 5 billion to about [euro] 18.2 billion in 2008 (DRV 2009: 222). With respect to the overall amount of taxes pouring into old-age security, one should finally mention the state subsidies for the take-up of a Riester contract: the supplementary allowances paid by the state for those signing a Riester contract amounted to [euro] 3.4 billion from 2002 until mid-2008 (Alterssicherungsbericht 2009).
7 Conclusion: Moving Forward Back to Bismarck?
Germany's pension policies are a masterpiece of political reform. Instead of aiming at major steps or at one great reform, the reforms have proceeded in an incremental manner. Cumulatively, the effect of the various reforms has been considerable. Particularly the measures taken from 2001 onwards have all been contentious and have stirred up public opinion. Pension policy has gradually become a source of individual uncertainty and insecurity, even in the German middle class.
In general, welfare states seem to be less resilient to change than expected in the early "new politics of the welfare state" debates. They move slowly, but once they have changed direction it is hard to bring them back to the former path. Germany is a striking example of that: Within a time period of about 20 years and beginning with the Pension Reform Act in 1989, the pension system moved step by step from a defined benefit system to an "uncertain something else" (Lamping and Rub 2004; Hegelich 2006b). As we see it, it introduced two major paradigm shifts (and four additional factors) which move the German pension system "forward back to Bismarck" (9).
(i) The first and most intensively discussed was the move away from a merely state-organized pension system which aimed at guaranteeing the standard of living. This was done by a defined benefit principle which was the major policy principle in the pensions system and which was commonly shared by all the main political parties and the social partners. However, defined benefit systems are, in principle, confronted with a serious paradox which is now becoming increasingly clear: the normatively justified social rights presume a permanent and stable flow of revenues on the input side of the PAYG insurance systems which allow the constant financing of the politically defined benefits. The policy paradigm works under exceptional conditions which B. Lutz called the "the short dream of enduring prosperity" (Lutz 1984). Under conditions of unemployment, the demographic challenge of growing numbers of pensioners supported by ever-decreasing numbers of wage earners, declining birth rates and increasing life expectancy, and economic and social contingencies, which are intensified and accelerated by globalization, the fiscal instability of PAYG systems becomes the norm and does not display a crisis phenomenon of a PAYG-financed defined benefit system. Given a stable status of social rights, changes in the economic and social environment immediately bring out of balance the fragile equilibrium of contribution-based (and state-subsidized) inputs and benefit oriented outcomes. In addition, given the political will not to raise the contribution rate and with out an appropriate financial reserve which is able to smooth imbalances between inputs and outputs, the dynamics of the globalized market economy make themselves visible in financial problems of the pensions system within relatively short time spans. Today, the German pension system displays a contradictory mixture of a revenue-based expenditure policy and a defined benefit system. However, the defined benefit, the guarantee of the standard of living, is based on premises which are no longer trustworthy.
The reason is simple: The standard of living is calculated based on the fictional "standard pensioner" ("Eckrentner") which is constructed based on the fictional idea of having a 45-year-long working life and a continuous income on the basis of an average earner. These fictional working conditions are in contrast to the real developments in the German labor market. The most important changes are those in the life path which turn the "three box life" of education, work, and retirement into risk biographies (DeVroom and Bannink 2008). The standard working model, which is based on a full--time job (about 35 to 40 hours a week) and full-time employment (more than 40 continuous years of work), is in decline, while atypical work is on the increase. In Germany, the standard working model declined from 82.5 percent of the workforce in 1997 to 74.4 percent in 2007. In absolute numbers, this represents a decline of more than 1.5 million workers. In the same period, atypical work rose from 17.5 percent to 25.5 percent, which amounts to an absolute number of more than 7.7 million workers (Statistisches Bundesamt 2008). In addition, wages increased only marginally and were not able to compensate for the changes in income.
The "Riester-Rente" broke with the previous principle of the state-produced guarantee of the standard of living. Now, the standard of living in old age is expected to be guaranteed by the mix of public and private pensions and were the share of (fictional) contributions to private savings reduces via the new pensions formula the level of the public pension. The state and democratic politics expect that contributors are willing and able to additionally invest a share of their income for their private or occupational old age provisions. However, we have argued that the more the state withdraws as a producer of welfare and the more it switches to regulating welfare markets, the more it should become a "consumer protection state" in social policy. If not, the politics of relieving the state could have the opposite effect--either as re-regulation or as a request for new poverty policies. For workers operating under atypical working conditions, the pension system will no longer be able to guarantee the standard of living; on the contrary, it will produce pensions which are below the official poverty line; this holds true above all for workers in the new German Lander (cf. DIW 2010). Consequently, the pension is--as under Bismarck--a provision against poverty, but no longer one for the standard of living.
(ii) The sustainability formula will decisively reduce the pension level. Although the approximations and the computer forecasts differ to some extent, the net pension (before taxes) will decrease from about 53% in 2006 to about 42% (before taxes) in 2030 (see Ehrentraut/Heidler 2007, figure 1: 5 f.) The whole indexation formula aims at producing a politically defined contribution rate which is expected to be fixed at 22 percent in 2030. Most importantly, the a-factor of 0.25 was explicitly chosen to achieve this very objective. This holds true in principle, although the government was not willing to bind itself to its own laws and suspended the sustainability factor in view of the last upcoming elections. In the years to come, German pension policies will again move forward by incremental steps and further deal with the paradox and contradictory mixture of a revenue-based expenditure policy and a defined benefit system which still exists, but only in rudimentary form. . Because the net pension is calculated on the basis on the fictional "Eckrenter", and because all pensioners are affected, most of the factual pensions are below that level. The pension, then, is no longer able to guarantee the standard of living, but only a contribution--financed basic security. Old age poverty will return and pensioners may be forced to undertake additional work during their retirement phase. As under Bismarck, the pension is a poverty provision and pensioners are expected to work during the period in which they receive their pension.
In addition, there are some other developments from which we might conclude that "moving forward back to Bismarck" is more than just a play on words, but which do not display a paradigm shift: (i) the increase in the retirement age extends the working period in the course of a lifetime and displays a re-commodification justified as a functional reaction against demographic changes; (ii) the implementation of a (means-tested, tax-financed) basic financial security in old-age signals that poverty is expected to be a severe problem in the (near) future and that pensioners with a respective share of paid contributions into the pensions system should be treated differently from those exposed to other forms of poverty; (iii) the increase of individual self-responsibility in effectively (or ineffectively) managing one's old-age income mix in a way reflects the social responsibility of the family to manage the disabled or old workers' income mix between the state pensions and (private) family income; and finally, (iv) the increase of the general federal subsidy, the introduction (1998) of the additional subsidy from the federal budget (zusatzlicher Bundeszuschuss), the federal lump subsidy by increasing the VAT by 1 per cent and the ecological tax on gasoline emblematize the subsidy of the German Reich to the individual pension (Reichszuschuss). These elements taken together are reminiscent to some extent of the initial conception of the pension set in place in Germany in the late 1880s-and favored by the "Iron Chancellor".
Alterssicherungsbericht (2009). Bundesministerium fur Arbeit und Sozialordnung: Erganzender Bericht der Bundesregierung zum Rentenversicherungsbericht 2008, Berlin.
Barr, N., Diamond, P. (2009a). Reforming Pensions: Principles and Policy Choices, Oxford: OUP.
Barr, N., Diamond, P. (2009b). Reforming Pensions: Principles, Analytical Errors and Policy Directions, in: International Social Security Review 62 (2), 5-29.
Blomquist, P. (2004). The Choice Revolution: Privatization of Swedish Welfare Services in the 1990s, in: Social Policy and Administration 33 (2), 139-155.
Bode, I. (2007). The Culture of Welfare Markets. The International Recasting of Pension and Care Systems, London/New York: Routledge Chapman & Hall.
Bonin, H. (2009). 15 Years of Pension Reform in Germany: Old Successes and New Threats. Policy Paper No. 11. Bonn: IZA.
Borsch-Supan, A. H., Wilke, C. B. (2005). Reforming the German Public Pension System. Discussion Paper 226, Center for Intergenerational Studies, Institute of Economic Research, Hitotsubashi University.
Borsch-Supan, A. H., Wilke, C. B. (2006). The German Public Pension System: How It Will Become an NDC System Look-Alike, in: Holzmann, Robert/Palmer, Edward (eds.): Pension Reform: Issues and Prospects for Non-Financial Defined Contribution (NDC) Schemes. Washington: World Bank, 573-610.
Bundesministerium fur Arbeit und Soziales (1994). Ubersicht uber das Sozialrecht, 3rd edition. Bonn.
Bundesministerium fur Arbeit und Soziales (2010). Ubersicht uber das Sozialrecht, Bonn: BW Bildung und Wissen.
Corneo, G., Keese, M., Schroder, C. (2007). Erhoht die Riester-Forderung die Sparneigung von Geringverdienern? (Economics Working Paper No. 2007--30, Christian-Albrechts-Universitat Kiel, Department of Economics).
De Vroom, B., Bannink, D. (2008). Changing Life Courses and New Social Risks: The Case of Old Age Pensions, in: Journal of Comparative Policy Analysis, 10/1, 75-92.
DIW (2010). Deutsches Institut fur Wirtschaftsforschung: Kunftige Altersrenten in Deutschland: Relative Stabilitat im Wes--ten, starker Ruckgang im Osten (Wochenbericht Nr. 11/2010), 2-11.
Doring, D. (1980). Das System der gesetzlichen Rentenversicherung: Eine Einfuhrung, Frankfurt/M./New York: Campus
DRV (2009). Deutsche Rentenversicherung Bund: Rentenversicherung in Zeitreihen 2009. DRV-Schriften Band 22. Sonderausgabe der DRV, Berlin.
Esping-Andersen, G. (1985). Politics Against Markets: The Social Democratic Road to Power, Princeton: Princeton University Press.
German Social Advisory Council, (2004). Report of the on the Federal Government's 2004 Pension Insurance Report, Berlin. www.sozialbeirat.de/files/gsacreport2004englisch.pdf
Gladen, A. (1974). Geschichte der Sozialpolitik in Deutschland, Wiesbaden: Franz Steiner Verlag.
Green, S., Hough, D., Miskimmon, A., Timmins, G. (2007). The Politics of the New Germany, New York.
Hain, W., Lohmann, A., Lubke, E. (2004). Veranderungen bei der Rentenanpassung durch das "Rentenversicherungs Nachhaltigkeitsgesetz", in: Deutsche Rentenversicherung, Heft 6-7, 333-349.
Hegelich, S. (2006a). Reformkorridore des deutschen Rentensystems, Wiesbaden: VS-Verlag.
Hegelich, S. (2006b) Moving from "defined benefit" to "undefined". The interests in the Enhancement of Supplementary Provisions for Old Age, in: German Policy Studies 3 (3): 535-585.
Hinrichs, K. (2001). Elephants on the Move: Patterns of Public Pension Reform in OECD Countries, in: Stephan Leibfried (ed.), Welfare State Futures, Cambridge: Cambridge University Press, 77-102.
Hinrichs, K., Kangas, O. (2003). When Is a Change Big Enough to Be a System Shift? Small System-shifting Changes in German and Finnish Pension Policies, in: Social Policy and Administration 37 (6): 573-591.
Holzmann, R., Palmer, E. (2006). The Status of the NDC Discussion: Introduction and Overview, in: Holzmann, Robert/Palmer, Edward (eds.): Pension Reform: Issues and Prospects for Non-Financial Defined Contribution (NDC) Schemes. Washington: World Bank, 1-15.
Joussen, J. (2002). Die Rente wegen voller und teilweiser Erwerbsminderung nach neuem Recht, in: Neue Zeitschrift fur Sozialrecht, 294-298.
Lamping, W. (2009). Verbraucherkompetenz und Verbraucher--schutz auf Wohlfahrtsmarkten: Neue Herausforderungen an eine sozialpolitische Verbraucherpolitik, in: Vierteljahreshefte fur Wirtschaftsforschung 78 (3), 44-62.
Lamping, W., Rub, F. W. (2004). From the Conservative Welfare State to an "Uncertain Something Else": German Pension Politics in Comparative Perspective, in: Policy & Politics 32 (2), 169-192.
Lamping, W. Rub, F. W. (2008). Introduction: Moving Bulky Goods. How new Ideas and Partisan Politics are Transforming the German Welfare State, in: German Policy Studies 4 (2), 1-18.
Lamping, W., Rub, F. W. (2006). "Experimental Law-Making" and the Politics of German Pension Reforms, in: German Policy Studies (GPS), (3) 3, 446-491.
Lamping, W., Tepe, M. (2009). Vom Konnen und Wollen der privaten Altersvorsorge. Eine empirische Analyse zur Inanspruchnahme der Riester-Rente auf Basis des Sozio-okonomischen Panels, in: Zeitschrift fur Sozialreform 55 (4), 409-430.
Le Grand, J. (2007a). The Politics of Choice and Competition in Public Services, in: The Political Quarterly 78 (2), 207-213.
Le Grand, J. (2007b). Delivering Public Services through Choice and Competition: The Other Invisible Hand, Princeton: Princeton University Press.
Leisering, L. (2007). Privatisierung der Alterssicherung als komplexe Ordnungsbildung. Zur Entstehung von Wohlfahrtsmarkten und regulativer Staatlichkeit, in: Becker, Ulrich/Kaufmann, FranzXaver/von Maydell, Bernd/Schmahl, Winfried/Zacher, Hans F. (Hrsg.): Alterssicherung in Deutschland. Festschrift fur Franz Ruland zum 65. Geburtstag, Baden-Baden: Nomos, 188-219.
Lenze, A. (2010). Abschied von der Solidargemeinschaft, in: Kritische Justiz 43 (2), 132-75.
Newman, J., Kuhlmann, E. (2007). Consumers Enter the Political Stage? The Modernization of Health Care in Britain and Germany, in: Journal of European Social Policy 17 (2), 99-111.
Newman, J., Vidler, E. (2006). Discriminating Customers, Responsible Patients, Empowered Users: Consumerism and the Modernisation of Health Care, in: Journal of Social Policy 35 (2), 193-209.
North, N. (2007). Empowerment in Welfare Markets, in: Health & Social Care in the Community 1 (3), 129-137.
OECD (2007). Closing the Pension Gaps: The Role of Private Pensions (Policy Brief September 2007), Paris.
Pino, A., Yermo, J. (2010). The Impact of the 2007-2009 Crisis on Social Security and Private Pension Funds: A Threat to their Financial Soundness, in: International Social Security Review 63 (2), 5-30.
Reimann, A. (2003). Das Verhaltnis von Pravention, Rehabilitation und Erwerbsminderungsrenten nach neuem Recht, in: Zeitschrift fur Sozialreform, 49 (3), 461-473.
Reimann, A. (2004). Das RV-Nachhaltigkeitsgesetz--Gesamtwirkungen und Bewertung, in: Deutsche Rentenversicherung, 6-7, 318--332.
Riester, W. (2000). Die Rentenversicherung an der Schwelle zum neuen Jahrtausend, in: Deutsche Rentenversicherung, 1-2, 1-14.
Sachverstandigenrat (2007). Sachverstandigenrat zur Begutachtung der gesamtwirtschaftlichen Entwicklung: Jahresgutachten 2007/2008, Berlin.
Schellhorn, W. (2005). Grundsicherung im Alter und bei Erwerbsminderung, in: Familie und Recht, 16 (1), 1-5.
Schmahl, W. (2006a). Die neue deutsche Alterssicherungspolitik und die Gefahr steigender Armut, in: Soziale Sicherheit 55 (1), 397-402.
Schmahl, W. (2006b). Das Soziale in der Alterssicherung--Oder: Welches Alterssicherungssystem wollen wir?, in: Deutsche Rentenversicherung 61 (11-12), 676-690.
Schmahl, W. (2007). Dismantling an Earnings-Related Social Pension Scheme: Germany's New Pension Policy, in: Jnl Soc. Pol., 36/2, 319-340.
Schulze, I., Jochem, S. (2007). Germany: beyond policy gridlock, in: Immergut, Ellen M./Anderson, Karen M./Schulz, Isabelle (eds.): The Handbook of West European Pension Politics. Oxford: Oxford UP, 660-710.
Social Advisory Council (2004). Report of the German Social Advisory Council on the Federal Government's 2004 Pension Insurance Report. www.sozialbeirat.de/files/gsacreport2004englisch.pdf [visited 07.06.2010].
Social Protection Committee (2008). Privately Managed Funded Pension Provision and Their Contribution to Adequate and Sustainable Pensions, Brussels.
Streeck, W., Trampusch, C. (2005). Economic Reform and the Political Economy of the German Welfare State, in: German Politics, 14/2, 174-195.
Trampusch, C. (2005). Institutional Resettlement: The Case of German Early Retirement, in: Streeck, Wolfgang/Thelen, Kathleen (eds.): Beyond Continuity. Institutional Change in Advanced Political Economics, Oxford: Oxford UP: 204-228.
Tapia, W., Yermo, J. (2007). Implications of Behavioural Economics for Mandatory Individual Account Pension Systems (OECD Working Paper on Insurance and Private Pensions, no. 11), Paris.
Tennstedt, F. (1972). Berufsunfahigkeit im Sozialrecht. Ein soziologischer Beitrag zur Entwicklung der Berufsunfahigkeitsrenten in Deutschland. Frankfurt a.M.: Europaische Verlagsanstalt.
Thaler, R. H., Benartzi, S. (2004). Save more Tomorrow: Using Behavioral Economics to Increase Employee Saving, in: Journal of Political Economy 112 (1), 164-187.
Tversky, A., Kahnemann, D. (1992). Advances in Prospect Theory: Cumulative Representation of Uncertainty, in: Journal of Risk and Uncertainty 5 (4), 297-323.
Viebrok, H. (2003). Disability Pensions in Germany, in: Prinz, Christopher (ed.): European Disability Pension. Policies: 11 Country Trends 1970-2002. Aldershot: Ashgate, 197-224.
Wollschlager, F. (2001). Gesetz zur Reform der Renten wegen verminderter Erwerbsfahigkeit, in: Deutsche Rentenversicherung, 5, 276-294.
Statistisches Bundesamt (2008). Atypische Beschaftigungsverhaltnisse auf dem Deutschen Arbeitsmarkt, Begleitmaterial zum Pressegesprach am 09. Sept. 2008, Wiesbaden.
(1) The social courts, mainly the Bundessozialgericht, started to differentiate within the four groups, mainly in the group of semi--skilled workers, and developed a scheme with a total of seven occupational groups. Concerning white collar workers, the social courts decided on a similar scheme with a total of nine occupational groups; c.f. for details Hesse/Filtkath 1993; BMAS (Hg.) 1994: 263--239.
(2) Plenary protocol of the Bundestag, 30th session, March 17, 2010, p. 2712.
(3) Entwurf eines Gesetzes zur Reform der gesetzlichen Rentenversicherung und zur Forderung eines kapitalgedeckten Altersvorsorgevermogens--AltersvermOgensgesetz" (AVmG) and " Gesetz zur Erganzung des Gesetzes zur Reform der gesetzlichen Rentenversicherung und zur Forderungeines kapitalgedeckten Altersvorsorgegesetzes--Altersvermogens--Erganzungsgesetz" (AVmEG).
(4) While the first part of the 2001 Pension Reform law--the new pension formula--introducing internal changes could have been realized by the decision of the Bundestag alone, the tax-financed public subsidies for the new private pension pillar required the consent and approval of the Second Chamber, the Bundesrat (representing the states, the Lander).
(5) BGB1 I, 2001: 403-418.
(6) While in the initial bill the minimum level had been 64 percent, the government (forced by large sections of the SPD and one of the big trade unions) decided during the policy process to accept 67 percent as a standard level, promising optimistically that pensions should and would not fall below this level (BT-Drs. 14/5150: p. 25). In combination with the new private provision, the maximum total income in old age was expected to be 76 percent of average net earnings in the future (BMA, 2001a; BMA, 2001b). But such a relatively high income could only be achieved under specific conditions and would have been impossible for most of the individuals insured.
(7) BGB1 I, 2001: 1310-1343.
(8) Germany is among those countries whose mandatory replacement rate is substantially below the OECD average. According to the OECD, the wage replacement rate of pensions is around 38% in Germany (see OECD 2007).
(9) For a detailed overview of the Bismarckian pension system c.f. DOring 1980; Gladen 1974: 67-70.
Friedbert W. Rub Humboldt--University Berlin Department of Social Sciences (Germany)
Wolfram Lamping Leibniz University Hanover Political Science Department (Germany)