Invasion From Within: Ideas,Power,and the Transmission of Institutional Logics Between Policy Domains

Paradigm Desperation and Aspiration

Logics of action are most likely to invade another policy field when the authority of a policy paradigm in one area is waning and the authority of a different paradigm in another policy area is strong. On a conceptual level, we consider this the conjunction of two marginal processes. The point of departure for change dynamics across domains then is that “gradual yet increasing disillusionment and the slow delegitimization of existing beliefs . . . open[s] up a political space into which new ideas can be inserted” (Berman, 2001, p. 234). The weakening of a policy paradigm is produced through a process fraught with contests for legitimacy. As the anomalies that eat away at the authority of a policy paradigm (Hall, 1993) do not automatically present themselves as such but rather need to be collectively interpreted and acknowledged as anomalies (Blyth, 2013), actors seeking to export a logic of action to a neighboring policy area need to construct a “crisis narrative” (Kuipers, 2010; Matthijs, 2011) and, second, forcefully make a case for the superior problem-solving capacity of the logic of action that they are seeking to export (Carstensen & Matthijs, 2018). Although some ideas and discourses are based primarily on technical and scientific (cognitive) arguments, to make these powerful in persuading the broader public and the organizations representing it, they still need to fulfill a normative function by providing a more generally accessible narrative about the causes of current problems and what needs to be done to remedy them that resonate with the public (Schmidt, 2006, pp. 251-253). The success of such an operation depends crucially on the legitimacy that these ideas as well as their supporters already enjoy within their own policy area and in society at large; a legitimacy that can be leveraged by arguing for their relevance within a new policy area.

It is thus key for the success of “invasion from within” that the dominant paradigm in the exporting domain provides connecting points to help overcome obstacles that may previously have stood in the way of forming a coalition in favor of a certain policy prescription. Consider the case of the German pension paradigm shift. At the surface, the conjunction of financial aspiration and pension desperation is seen. However, as argued below, the underlying reason why the financial discourse of pension privatization managed to successfully penetrate the pension arena is that in the course of the financial liberalization process, a consensus regarding the benefits of the outsider–finance paradigm emerged. This dominant discourse, on a most fundamental level, made away with much of the German skepticism vis-à-vis capital markets, but more germane to our argument, it established the basic sensibility of fostering prefunded pensions as a suitable way to provide for old age. In other words, ideas that muster a firm coalition in one domain enjoy a fundamental legitimacy and political momentum that may propel them to the top of the political agenda as plausible policy options in domains that are in need of new solutions. Also, in the case of German pension privatization, the respective policy prescription was already largely developed “waiting in the wings” (Schmidt, 2002, p. 223), which contributed to its appeal over other reform options.

Networks Across Institutional Domains

To bring in a new interpretation of existing institutions requires that a powerful coalition of actors support the reinterpretation. As noted by Parsons (2016), ideas that generate new coalitions by definition stretch across diverse and previously separate actors and agendas. Working as “coalition magnets” (Béland & Cox, 2016), such ideas may influence policy precisely because they mean different things to different people, making it easier to agree (Jabko, 2006); a process that crucially hinges on actors’ discursive construction of likeness between ideas prevailing and widely considered legitimate in one issue area and the ideas they are seeking to push through in another.

The successful transmission of a logic of action from one institutional domain to another thus requires coalition building structured around a common, but nonetheless often ambiguous, understanding of policy problems and viable solutions (Culpepper, 2008; Palier, 2005). We may imagine multiple shapes that such coalitions can take. One could for example see a winning coalition made up only of actors from the adjacent policy area exporting an institutional logic. In this case, the widespread support that the logic of action enjoys among powerful actors in one area of the economy may enable agents to transmit it to another institutional site, over time leading to gradual institutional change. Alternatively, the import of ideas from one policy area to another may help forge coalitions across institutional divides, that is, between actors placed in disparate institutional sites that coalesce around a joint interpretation of how the two policy areas should work and jointly support the introduction of their shared logic of action in both institutional domains. In this case, incumbent actors seeking to change the dominant institutional interpretation in their policy area may harness the legitimacy their preferred logic of action enjoys in a neighboring policy area, along with the authority of the coalescing actors, based for example on prestige related to professional and educational background or their membership of international regulatory networks (Seabrooke, 2014).

Importantly, the harnessing of legitimacy is not just a matter of providing persuasive arguments that may bring on board the most important political groups and the electorate, but likewise a process that works to shut down competing crisis narratives and alternative policy solutions. From this perspective, the use of ideas to exert power is instrumental, in the sense that the actors who come to accept the new set of ideas and the changed logic of action their institutionalization entails do not necessarily believe in the ideas. Instead, the intersubjective efficacy of the idea—and the communicative discourse employed by the ideational agent—is so strong that the actors concerned are compelled to adhere to the idea (Carstensen & Schmidt, 2016).

It is important to recognize that the effectiveness with which networks crossing institutional sites are able to export a logic of action from one institutional domain to another is not only based on persuasiveness of the ideas in the sense that any set of actors with the “right ideas” can potentially establish such interlinkage. The efficacy of such an operation in large part hinges on other nonideational factors, notably access to significant economic resources—that enable actors to campaign for their ideas and in a broader perspective lend weight to their claims—as well as institutional position that places the actors in the network centrally in the political system and broader societal networks of power. Material resources, however, alone cannot account for the successful movement of a logic of action. As argued above, for incumbent actors to identify and make their interests actionable within the parameters of a different logic of action requires that the supporting network provides a coherent and appealing narrative that may speak to interests that cut across ideological divides and persuades about the problem-solving capacity of a new set of ideas. On a conceptual level, then, this means that the power of ideas oftentimes interacts with other kinds of power, such as structural, institutional, and coercive power (Barnett & Duvall, 2005; Parsons, 2016). In terms of the German pension case, rather than highlighting ideational power at the expense of the structural power of finance, it is thus key to appreciate the interaction between different kinds of power within the case.

Strategic Translation of a New Logic of Action

If a reinterpretation is to be successfully imported to another institutional site to establish synchronization, it requires that it be fitted to its host environment. To make a policy appear persuasive or “coherent,” actors need to engage in a strategic transfer of legitimacy and frames across spheres to show how the logic of action offers normatively acceptable ideas that also appear to be an effective cognitive alternative in the perceived material interest of actors (Schmidt, 2008, p. 307). It is auspicious to conceive of the discursive strategies underlying invasion from within as what Acharya (2004) refers to as “localization.” He proposes to think of localization as a “congruence-building process” during which actors engage in grafting and framing. ⁴ In Acharya’s (2004) framework, “grafting” is a tactic policy entrepreneurs employ to institutionalize a new norm by associating it with a preexisting norm in the same issue area, which makes a similar prohibition or injunction. Such grafting in turn requires framing because “the linkages between existing norms and emergent norms are not often obvious and must be actively constructed by proponents of new norms” (p. 243).

One such strategy is not to advocate to displace the existing paradigm altogether, but rather to modify it. Acharya (2004) finds localization attempts more likely to be successful when “the idea recipient’s chief goal was to strengthen, not replace, existing institutions . . . with the infusion of new pathways of legitimation” (p. 246). In the case at hand, privatization advocates merely sought to displace the logic that the government pillar suffices to secure the living standard of former wage earners into old age. To do so, they made the case that only by integrating funded elements into the old system, will this system be able to survive. This is intertwined with another way in which actors adjust their discourse when aiming at a new domain. Not only did they adapt to the prevalent frames, but also the communicated goals underlying the policy. When pension privatization was discussed in the financial domain, proprivatization arguments revolved around the policy’s positive effects in terms of deepening capital markets and their superior yields and, by extension, increasing economic growth and employment, whereas normative categories of generational justice dominated the debate in the pension arena.

Although this adjustment in terms of vocabulary is hardly surprising, the way in which privatization advocates fused arguments of economic efficiency and social justice is less obvious. Privatization advocates presented the economic superiority of prefunded pension schemes vis-à-vis PAYG varieties as something that can be taken for granted. This is where a policy prescription—or rather its fundamental ideational dimension—that evolved in an adjacent policy area really unfolds its legitimizing power: Advocates do not need to actually put forward the argument anymore; instead mere self-evident side notes à la “as we all know, funded alternatives provide superior yields” suffice. Hence, the axiomatic quality that a policy prescription gained in a neighboring policy has the effect of immunization through the creation of unquestionability (Brettschneider, 2009, p. 192). It is this transfer of legitimacy that enabled actors to bring about an interpretative closing of the “knowledge market” that made pension privatization appear to be a “simple necessity” (Marschallek, 2004, p. 299) as part of what one may refer to as the “social construction of an imperative” (Cox, 2001). This speaks to the taken for granted effect that paradigms exert (Blyth, 2002; Hall, 1993). What previously stood in the way of pension privatization—capital markets skepticism and so on—is now part of the proprivatization argument but relegated to a marginal note.

Research Design

Before proceeding to the case study, it is necessary to set out the specifics of the research design underlying the empirical analysis. Faced with the twin tasks of explaining the role of finance in the pension paradigm shift and crafting a theoretical framework useful for such an endeavor, a theory-building process-tracing approach is particularly useful. According to Beach and Pedersen (2013, p. 16), there are two key steps in developing a causal mechanism based on theory-building process tracing. First, its starting point is inductive with the collection of evidence that may be used as “clues about the possible empirical manifestations of an underlying causal mechanism between X and Y.” The second step involves, “inferring from the observable empirical evidence that these manifestations reflect an underlying causal mechanism that was present in the case” (p. 16). In the context of this article, the inductive collection of evidence was ordered using the existing and quite extensive literature on the PRA 2001 reform, which was combined with a deductive element of moving back and forth between ideational and historical institutionalist theorizing and the available data in an effort to tease out causal connections underlying the developments in the empirical material.

To demonstrate the analytical usefulness of the causal mechanism of “invasion from within” for understanding the multipillarization of the German pension system, requires showing that the three constitutive parts of the causal mechanism are necessary to include to account for the outcome of interest. We do so by analyzing, first, how the two policy fields—finance and pension—develop in the decades preceding the reform. The analysis traces changes in the sociopolitical and institutional environment pertaining to both domains and links them to the preferences, strategic considerations, and policy positions of key actors. Here, we show that initially the two fields remained disconnected in the sense that policy making in each was structured by two largely incommensurate paradigms and that no coherent coalition had formed in finance that argued in favor of pension privatization. We then track the waning authority of the Bismarckian pension paradigm and the simultaneous strengthening of the finance paradigm. Second, we show how previous obstacles toward compromise were unencumbered through the marginal process of financial liberalization. Finally, we trace how privatization advocates leveraged the legitimacy of liberalization in finance to push for the introduction of private schemes in the pension domain, and that actors “localized” their policy prescription in the pension sphere by engaging in discursive strategies of reframing.

Throughout the empirical analysis, these arguments are weighed against relevant alternative explanations. First, one could reasonably conjecture that the structural power of finance alone can account for the shift toward pension privatization. This would entail that once the German financial sector gained enough economic and political clout, financial interests set their eyes on a liberalization of the German pension system and were able to push through the reform. In that case, we do not need the causal mechanism of “invasion from within” to explain the shift toward multipillarization as we can simply point toward material and structural factors. Second, one may imagine that the ideas that inspired the policy shift in German pension politics originated within the pension area itself. In these circumstances, we would be able to claim that both the second and the third part of the proposed causal mechanism, namely, the forging of coalitions within an adjacent policy area and the process of strategic localization, are unnecessary for accounting for the outcome. Put differently, it would be superfluous to study connections between institutional sites as the reform was created within the pension area. Third, and finally, we seek to determine the extent to which ideas concerning pension privatization were introduced from abroad rather than developed within the German polity. If that was the case, again, we do not need the causal mechanism of “invasion from within” to account for the outcome but can instead limit our focus to studying the effect of ideas imported from abroad.

The empirical analysis draws on a wide range of primary and secondary sources, including parliamentary debates, various newspaper archives, and expert interviews. Expert interviews have been employed to validate the findings from documentary research, enhance the reconstruction of political processes, and discern actors’ motives and beliefs (Tansey, 2007). Interview subjects have been selected based on case knowledge (purposive sampling) and recommendations (snowball sampling). ⁵

Obstacles of Pension Privatization: Powerful “Pension Men” and German “Insider Finance”

The 1957 pension reform enshrined the basic principles of the Bismarckian, one-pillar pension paradigm. Contrary to the consensual character often attributed to the process leading up to this seminal reform, key aspects pertaining to the pension–finance nexus were subject to extensive political battles along fault lines that would reemerge in pension debates throughout the remainder of the century. The economically liberal parties, Free Democratic Party (FDP) and German Party (DP); the liberal wing of the governing Christian Democratic Union (CDU), and the entire camp of organized business opposed the dynamized PAYG approach, ⁶ yet for reasons of political expedience, as well as the perceived failure of funded alternatives in recent history of wars and inflation, a near ideal-type conception of a “Bismarckian” pension system reigned supreme (Nullmeier & Rüb, 1993, pp. 93-115). In turn, voices fundamentally questioning the pension configuration were increasingly marginalized in what Babel (2001) refers to as a “silent revolution” (p. 51).

By the mid-1980s, a formidable liberal coalition had begun to question the one-pillar paradigm encompassing inter alia the FDP, the economic and financial wings of the CDU, the Bundesbank, and many academics. In the context of what was widely perceived to a be a “pension crisis,” the period between 1985 and 1989 saw a battle between the agents of whose policy preferences largely fall within the multipillarization paradigm and agents of continuity defending the principles of the one-pillar pension paradigm. The PRA 1989 explicitly adhered to the traditional order as multipillarization advocates neither managed to undermine the authority of the one-pillar paradigm nor seriously challenged the discursive-political power wielded by the so-called Rentemänner (pension men) that undergirded the old pension order, in particular, organized labor, conservative and social democratic “social politicians,” and administrators of the public pension scheme (Nullmeier & Rüb, 1993, p. 301). This coalition put concerted efforts in finding policy solutions that explicitly remained within the parameters of the 1957 reform. The importance of this “coalition of path dependents” (Conrad, 1998, p. 112) for continuity in the German pension system can hardly be overstated (Hinrichs, 2005, p. 53).

Although the one-pillar paradigm proved resilient despite mounting material pressures, “pension men’s” efforts in fending off attacks were also facilitated by a lack of cohesiveness among agents of change. Even though the policy solutions advocated by this coalition all fell within the category of pension multipillarization, they differed widely. ⁷ Note that the main line of division did not run between political parties but across the conservative and social democratic people’s parties, between “social politicians” on one hand and “financial” and “economic politicians” on the other (Berner, 2009). Consider how the heterogeneity among liberal agents of change played out within the CDU, which is widely held to be the decisive battleground of pension politics at this point (Nullmeier & Rüb, 1993, p. 353): One camp led by Kurt Biedenkopf advocated a systemic shift toward a Beveridgean type of system, whereas Christian Schwarz-Schilling’s proposal envisaged introducing a capital stock within the public pillar (Biedenkopf, 1985; Miegel, 1981; Schwarz-Schilling, 1987a, 1987b). When the CDU voted on their pension program at the party convention in 1988, these two visions of multipillarization competed with one another, somewhat canceling each other out and facilitating labor minister Blüm’s victory. ⁸ This internal heterogeneity stands in stark contrast with agents of continuity, who displayed remarkable cohesion in their positions and lines of defense (Hegelich, 2006, p. 234). For example, when Schwarz-Schilling presented his proposal, various “pension men” dispelled it in remarkably similar ways (Nullmeier & Rüb, 1993, p. 388). Agents of continuity rejected multipillarization policy ideas out of principle without an “internal audit,” as two of the most central “pension men” confirmed during interviews. ⁹ One “pension man” recalls that “we truly have built a concrete wall” to fend off multipillarization initiatives. ¹⁰

Just when policy makers within social policy thought to have had sustainably repaired the pension system, reunification added to the long-term marginal pressures the system had been facing. At this point, the long-standing globalization–competitiveness discourse (Standortdebatte) lamenting strenuous “social costs” grew stronger (Seeleib-Kaiser, 2002), and even social democrats began to adopt positions that higher pension contributions ought to be precluded. Multipillarization advocacy was focused on fiscal retrenchment and reducing nonwage costs in the name of competitiveness (Banchoff, 1999; Hegelich, 2006, pp. 217-232). As Cox (2001) writes, “[b]ecause enhancing competitiveness is a concept alien to the construction of the German welfare state, it does not offer a legitimate basis for welfare reform” (p. 498). Although effective in pushing for budget cuts, such a discourse failed to “shape paths,” which at that point is reflected in comparatively steep retrenchment but limited restructuring in the German pension system (Schludi, 2005). By 1995, multipillarization advocacy was one dimensional and internally heterogeneous.

Reconnecting Private and Public Pensions

One may be tempted to draw a straight line from generous public pension systems to a private pension regime that contributes little to capital formation (Jackson & Vitols, 2001), but such a link is difficult to empirically corroborate (e.g., World, 1994, p. 307). Notwithstanding its dominance, it is not as if generous public pensions simply “crowded out” private (old-age) saving in Germany. German private savings rates actually increased after the 1957 pension reform and remained high (Hockerts, 1980, p. 383). ¹¹ Economists have referred to this as the “German savings puzzle.” ¹² In light of the parallel quantitative development of public pensions and overall private savings, we need to consider both spheres as “qualitatively drifting apart” (Wilke, 2016, p. 47) rather than functional derivatives of one another. The 1957 pension reform instigated a remarkable dissociation of pension politics into “separate worlds of old-age provision” (Berner, 2009, p. 21). As the one-pillar approach meant that the public pension pillar alone should suffice to maintain one’s standard of living throughout retirement, the purpose of occupational pension schemes altered from topping up meager public pensions to a tool of employee retention and motivation during times of full employment (Hegelich, 2006, p. 122). Wealth formation policies aiming at a higher share of “productive capital in the hands of employees” were also decoupled from the old-age provision context (Dietrich, 1996, pp. 222-228). Company pensions and private wealth more generally became “de-socialpoliticized” (Berner, 2009, p. 125) and thus disjunctured from what was considered “pension politics” or the “pension system.”

Whereas during private capital formation parliamentary debates of the 1960s, the implications of the recently established pension system were, if only occasionally, raised by policy makers of various stripes as early as in the 1970s the two issues were not conflated anymore. Perusing the parliamentary debates and media reporting pertaining to the private capital formation policy initiatives during that period ¹³ suggests that the context of old-age provision did not factor into these debates. This separation held throughout the 1980s, when the financial solidity of the government pension was increasingly put in question. None of the parliamentary debates, motions, or law proposals by the government and opposition parties on private capital formation from that decade contain a reference to old-age provision. ¹⁴ Perhaps more surprisingly, the goals of private capital formation policies were also remarkably separated from private old-age provision. Equity-based products were considered overly speculative and thus inappropriate for old-age saving, whereas life insurances were deemed unsuitable to foster capital formation. ¹⁵ As the favorable tax treatment of housing and life insurances was the “strongest pillar of German saving policy” (Börsch-Supan & Eymann, 2000, p. 17) and efforts at fostering private capital formation through savings subsidies remained comparatively unsuccessful, saving patterns of German private households come as little surprise: In 1991, 75% of household assets were held in currency, deposits, insurance claims, and pension rights (Detzer, Dodig, Evans, Hein, & Herr, 2012, p. 64) and even by 1998, a mere 6% of the adult population directly owned shares (Jürgens, Naumann, & Rupp, 2000, p. 57).

This political pillarization of the three pension pillars has two main upshots. First, it highlights the perceived incommensurability of policy goals pertaining to old-age provision and capital formation as a crucial obstacle toward capital market–oriented pensions. Second, it suggests that one needs to understand the evolution of politics pertaining to the second and third pillar of the German pension system in the context of the German postwar financial paradigm. Widely held to be the ideal-typical bank-based and insider-oriented variety up until the 1990s, the German financial system constituted an institutional arrangement that shielded firms from short-term market pressures and thus ensured “patient capital” (Deeg, 1999; Hall & Soskice, 2001; Vitols, 2004). Most company pension commitments were held as internal book reserves (Direktzusage) with such “organizationally embedded schemes” constituting a key source of market-independent liquidity (Jackson & Vitols, 2001) and thus a core element of the patient capital paradigm (Röper, 2018).

Especially, the recession between 1979 and 1982, however, delegitimized what was perceived as an overreliance on bank and internal finance in light of a wave of corporate bankruptcies (Story, 1996, p. 385). Corporate Germany’s “equity problem” became more severe throughout the 1980s. ¹⁶ Upon the decision to form a Single European Market, the financial system increasingly became part of the German globalization–competitiveness debate (Standort) in the second half of the 1980s (Beyer & Höpner, 2003, p. 191). Rather than a “big bang,” however, marginal reorientations toward finance capitalism ensued (Deeg, 2005; Jürgens et al., 2000, p. 67). A widely shared consensus in the political field of finance began to emerge that saw the deregulation of the financial sector as a factor of competitiveness, source of economic growth and employment, and essential in bringing “equity culture” to German households. By 1995, finance capitalism had reached a tipping point of dominance (Röper, 2018). The shifting political winds encouraged banks and investment companies to challenge incumbent insurers with “financialized” pension policy proposals (Röper, n.d.). Investment companies (BVI) proposed the first investment fund (AS-funds) ¹⁷ officially labeled as a pension product (Laux, 1995) and Deutsche Bank’s research arm for the first time made concrete proposals to establish Anglo-American type of pension funds in Germany (Deutsche Bank Research, 1995).

The BVI connected two issue spheres that had long been separated: private capital formation and fostering private old-age saving. It is therefore not too aggrandizing that BVI-chief Laux referred to this “combination of capital formation based on equities with the idea of old-age provision” as a “revolution” when first presenting AS-funds on German television (TV interview “Telebörse,” September 15, 1995). Deutsche Bank Research’s Axel Siedenberg similarly underscored a dual strategy: “Our initiative aims to strengthen occupational pensions and at the same time broaden and deepen the Finanzplatz Deutschland” (Siedenberg, 1995, p. 3). These policy initiatives are expressive of bricolage (Campbell, 2004; Carstensen, 2011) as organized finance and financial policy makers reformulated private capital formation policies and company pensions as solutions to the problems facing the pension system within the context of the wider finance capitalism program—and thus broke up the long-standing incommensurabilities separating the three pension pillars. Investment companies explicitly took aim at the “political taboo that wealth formation policies and old age provision policies be conceptually kept strictly separated” (Laux, 1995).

This recombination did not generate entirely new ideas but rearranged long-standing policy goals of fostering “equity culture” and private capital formation, boosting capital markets, and privatizing parts of the pension system in novel ways. BVI-chief Laux accordingly described his policy innovation as “killing three birds with one stone”: private capital formation among employees, deepen capital markets, and improving old-age provision for employees (TV interview “Telebörse,” September 15, 1995). This constitutes the integration of the finance capitalism program and multipillarization advocacy: Notions of efficiency (rates of return), long-standing but abstract arguments about increased capital formation and thus economic growth through funded pensions, and the need to privately save for retirement were recombined and translated into concrete policy solutions. The forging of a connection worked on both a discursive level—which facilitated “path shaping” in “establishing new grounds for evaluating the legitimacy of policy proposals” (Cox, 2001, p. 474)—as well as on a political level, where these reformulations meant concrete policy solutions that could be fought for. Two marginal processes coincided: pension desperation and financial aspiration. The stage was set for invasion from within.

Notwithstanding the central role of international organizations in spreading ideas about pension reform (Orenstein, 2013), with the Organisation for Economic Co-operation and Development (OECD) and World Bank often pointed to as key sources of ideas concerning multipillarization, in the case of the German pension paradigm shift, the ideas and discourses behind the push for pension privatization were “homegrown.” For one, interviews with elite decision makers at the time shows that few were aware of the existence of the studies of international organizations such as the OECD or the World Bank, let alone had read it. In the case of the World Bank report, Lamping and Rüb (2006) corroborate this: “We found no single evidence that the World Bank’s concept played any role in the development of the reform ideas” (p. 465). Second, even if policy ideas of multipillarization were in some form imported from abroad, it is unlikely that their initial recipients were located in the sphere of social policy (especially given the abovementioned German compartmentalizing of economic and social policy making). It seems more plausible that actors in other policy areas or societal spheres were susceptible to such ideas at an earlier stage and then advanced them domestically, turning this into a process “from within.”

Localization: Leveraging Financial Consensus in the Pension Domain

The reformulations of policy goals and the conjunction of marginal change processes created a new political space in which privatization advocates had to focus on a different audience. Unlike the “quiet” financial domain (Culpepper, 2011), pension politics are among the loudest. This means that rather than a small technocratic circle the public at large needed persuading. Once investment companies and banks had developed concrete policy solutions to open up the field of pension politics, they allocated tremendous resources to campaigns pushing for partial privatization (Wehlau, 2009). One manifestation of these efforts by financial firms was to found a research institute solely dedicated to the issue of old-age provision, the Deutsche Institut zur Zukunft der Altersvorsorge, in 1997. The financial sector, liberal think tanks and academics produced and disseminated a myriad of academic and pseudoscientific studies, especially, after 1995, in what one may refer to as “wallpapering” of public discourse.

A significant part of this effort was aimed toward the perhaps most obvious and long-standing point of critique vis-à-vis the one-pillar pension system, namely, its vulnerability to societal aging. Warnings about the “exploding social costs” that had long been part of the fiscal sustainability and competitiveness discourse experienced a notable amplification post 1995 in what amounted to a “pension panic.” We can pinpoint the increased participation by organized finance through the proliferation of the terms “pension gap” (Rentenlücke) and “provision gap” (Versorgungslücke), which clearly aim to point out insufficiencies in the public pension system (see, for example, Bundesverband deutscher Banken [BdB], 1997, 1998; Bundesverband Deutscher Investment-gesellschaften [BVI], 1996, 1997, 1998; Gesamtverband der Deutschen Versicherungswirtschaft [GDV], 1996, 1997, 1998). What used to be specialist jargon in the financial sector was transferred to the wider pension debate. A former CEO of an investment company and president of the BVI recalls: “We mentioned provision gaps in every third sentence . . . . We mentioned it again and again, we brought it to the fore again and again.” ¹⁸ Winfried Schmähl, the most prominent academic among the “pension men,” laments “an instrumentalizing of demographic change by conjuring up crisis scenarios,” (Roland, 2001, p. 53) beginning in the mid-1990, in the context of which he attributes a key role to banks and insurers in linking “demography and the insecurity of pensions” (p. 54).

By engaging in what one may refer to as “demographic pessimism” (Skidmore, 1999), pension privatization advocates sought to dislodge the logic that the government pillar alone would suffices to secure the living standard of former wage earners into old age (Lebensstandardsicherung). They made the case that system-internal solutions could not deal effectively with the “ticking demographic time bomb” and that the first pillar alone will no longer guarantee Lebensstandardsicherung, thus rendering multipillarization “inevitable” (see, for example, BdB 1997; BVI, 1996; GDV, 1996). Although a seemingly intuitive strategy, this reformulation contrasts starkly with multipillarization advocacy in the 1980s. During the 1989 pension reform debate, agents of continuity defended the old order by arguing that Lebensstandardsicherung may not be endangered. At that point, multipillarization advocates such as Schwarz-Schilling dared not to put in question this policy goal of the first pillar (Schwarz-Schilling, 1987a, 1987b). During the 1990s, by contrast, privatization advocates absorbed Lebensstandardsicherung as the overarching policy goal but reformulated it in arguing that all three pillar combined shall ensure it. That is, they maintained the policy goal but altered the means to attain it as attain it as Lebensstandardsicherung through the three pillars.

As noted above, Acharya (2004, p. 246) finds recombination strategies more likely to be successful when the idea recipient’s goal is to bolster existing institutions by providing new raw material for legitimation. Partial privatization advocates did just that when making the case that only by integrating funded elements into the old system will this system and its promise to secure the living standard of wage earners into old age be sustained. This line of reasoning became widely used thereafter. Katrin Göring-Eckardt (Green Party), for example, argued during the parliamentary debate of the PRA 2001 that only once generational equity is re-established—crucially by means of pension privatization—“will the acceptance of PAYG old-age provision rise again” (Plenarprotokoll 14/46). This “implementing the new merely to save the old”-framing conceivably contradicts the notion of the private pillar successively crowding out the state pension.

Feeding into the construction of likeness, privatization advocates understated the paradigmatic extent of the changes they were proposing by insisting that the bigger role played by private and occupational old-age saving be a “supplement” to rather than a “replacement” of the public pension pillar (see, for example, BdB, 1997; BVI, 1996). This marks an interesting contrast to the dramatic tone employed when characterizing the dire financial situation of the public pillar. A former CEO of an investment company and president of the BVI confirms that this was a deliberate strategy of understating. ¹⁹ This has been widely criticized by agents of continuity and one of their key representatives even refers to this as “deception.” ²⁰ Labor Minister Riester attests to this nondisplacement framing regarding the introduction of the private pillar during the first reading of the law when characterizing it as “not a replacement, but a supplement to the state pension” (Plenarprotokoll 14/133).

Another discursive strategy of undermining the old pension paradigm and introducing new cognitive and normative concepts as superior solutions by privatization advocates can be found in the context of presenting the PAYG system as generationally unjust. “Generational accounting” (Generationenbilanzierung) provides the economic tool underlying much of this debate (Schmähl, 2004). On a most fundamental level, generational accounting allows comparing benefits that different types of pension systems provide for individual cohorts. Financial sector actors primarily used generational accounting to compare rates of returns between the public pension pillar and their respective policy alternatives, such as investment funds (Hockerts, 2011, p. 312). In addition to presenting the public pension as inefficient in terms of rates of return, especially, partial privatization advocates used generational accounting to warn of a “sustainability gap” (Nachhaltigkeitslücke) in the public pension system, which unduly burdens younger generations (Leisering, 2004; see, for example, BdB, 1997; BVI, 1996; GDV, 1996).

Tracing the career of terms, arguments, and concepts of partial privatization from a before-and-after 1995 perspective for the four main political parties and their respective key agents leads to the conclusion that by 1998, partial pension privatization had become a perceived imperative among all main political parties. And not because agents of continuity grew convinced of its advantages or because they had to resign but because the pressure created by partial privatization advocates forced them to make concessions and thus redraw the contours of the pension debate. The two key agents of continuity—the long-time social minister Norbert Blüm (CDU) and the long-time social policy spokesperson Norbert Dreßler (Social Democratic Party [SPD])—felt coerced to adapt both their language and their positions as a strategic move to prevent further reaching steps toward partial privatization. ²¹ Examples of this include the fact that both entertained ideas of introducing a capital stock in the public pension pillar to appease multipillarization advocates, their concrete battles with privatization advocates about interpreting Lebensstandardsicherung, or Blüm’s attempt to shift the meaning of “generational equity” toward his leitmotiv in an debate with the FDP (Torp, 2015, p. 375). In 1994, Dreßler still flat-out rejected any necessity to partially privatize (Dreßler, 1994), by 1998, he had reluctantly supported concrete policy proposals toward that end and begun to defensively adapt the multipillarization nomenclature (Dreßler, 1998).

One may interpret this convergence among change and continuity agents as a learning process by the latter. One is hard pressed, however, to find evidence of such importing of ideas or adjustment of preferences. During a public hearing for the Third Financial Market Promotion Act, for example, financial policy maker Wolfgang Steiger tellingly asked the Goldmann Sachs representative: “What kind of arguments may help win over employees and social politicians to introduce pension funds in Germany?” (Public Hearing Finance Committee, November 12, 1997, protocol no. 93, 111-112). A prominent social politician from the 1990s recalls how distinct the spheres of financial and social policy making remained in parliament:

those were separate playing fields, so to speak. It wasn’t as if financial politicians and social politicians would get together and ponder what to do. Realistically there was no exchange [prior to the PRA 2001] . . . . Politics was strongly pillared back then. ²²

Further substantiating this, he describes a pronounced polarization within the CDU/CSU: “90 percent of social politicians supported the Blüm model and 90 percent of financial and economic politicians opposed it”. In the run-up to the PRA 2001 most of the core agents of continuity such as Rudolf Dreßler or the chairman of the Social Council, Winfried Schmähl, were ostracized (Wehlau, 2009, pp. 147-158). Within the SPD, the PRA 2001 also caused a rift: During the internal vote of the parliamentary group, 19 MPs voted against the reform, which was also the number of members of the committee for social affairs. ²³ When the Bundestag passed the legislation, 48 SPD MPs expressed their “fundamental social political concerns” about the reform in an attachment to the protocol and justified their voting in favor as merely the lesser evil (Plenarprotokoll 14/147, 14458). Taken together, this substantiates an invading export by financial actors rather than a learning import by social actors.

Outlook Beyond 1998

Taken together, by 1998, we can attest the “social construction of an imperative” (Cox, 2001, p. 463) to partially privatize the German pension system. The smaller political parties, FPD and Greens, converged on partial privatization (Torp, 2015, p. 379) coming from different points of departure. Social politicians within the people’s parties found themselves forced to make strategic concessions to defend the old order. ²⁴ Organized business gladly used the ideational power that AS-funds and pension funds wielded as government policies to justify their more radical privatization advocacy. ²⁵ Organized labor increasingly fragmentized over the issue of fighting multipillarization versus adapting it to gain influence through collective occupational pension schemes (Schludi, 2005, p. 161). And in a context of favorable stock market performances, the media almost unanimously and often euphorically called for partial privatization (Schmähl, 2012).

By 1999, when the discussion of the PRA 2001 began, the economic effects of pension privatization were hardly ever mentioned anymore. If at all, they were alluded to in passing, which is remarkable, given how complex and debatable the economic case in favor of pension privatization is (Marschallek, 2004). Hall (1993) likens ideas to subatomic particles, for they “do not leave much of a trail when they shift” (p. 290). In the case at hand, what is left unsaid is arguably most instructive to assess the financial paradigm shift’s impact on the pension domain. The government’s proposal for the PRA 2001 does not contain a single reference to the economic effects of pension privatization and the counter motion by CDU/CSU merely states in the most general terms: “The expansion of the first and second pillar would disburden the state pension and strengthen the business and financial location” (BT-DS 14/1310). During both readings of the law in the Bundestag, not a single speaker of any party specified this notion of economic effects other than vague allusions to the overall integration of social and financial policies.

Although we surely observe a combination of both deliberate and subconscious absences during the parliamentary debates of the PRA 2001, the latter appear to outweigh the former. In the only parliamentary discussion of privatization’s economic effects, Kurt Biedenkopf addressed potential doubts about pension privatization’s concomitant double payment problem when submitting in the Bundesrat (upper house): “But due to the compound interest-effect, private capital formation is more economic than the collective PAYG old-age provision . . . . Consensus about this basic idea prevails across party lines” (Plenarprotokoll 758). Upon proposing the long-term goal of a 40% share of old-age provision coming from private savings, Biedenkopf further argues, “Economic reasons, which I cannot explain further here, support this as well. Private old-age savings create capital formation, which increases productivity.” These quotes teeming of self-evidence combined with the observation that not a single MP even bothered to mention superior yields to make the case for funded schemes—let alone address potential risks stemming from fluctuating capital markets ²⁶ —support the hypothesized invasion from within-effect of the financial paradigm in the pension domain.

Having followed the process through which pension privatization became the favored reform option, we are now in a better position to determine whether in fact a more simple structural explanation—focused on the increased economic and political clout of German finance—may suffice as explanation for the paradigm shift. At first, such an argument seems persuasive, considering that German organized labor has lost ground in terms of coverage, level of centralization, and public legitimacy throughout the 1980s and 1990s (Hassel, 1999; Schroeder & Keudel, 2008) and actors from the financial sector increasingly began to penetrate social policy discussions during the 1990s (Wehlau, 2009). Although the benefits of the PRA 2001 for the financial sector are undoubtedly tremendous—Lamping and Rüb (2004, p. 182) refer to it as a “gigantic capital market extension program via the creation and motivation of private demand”—two main factors shed doubt on a structuralist argument. First, beyond reasonable doubts whether structural power is a plausible and testable causal factor in the case of pension policy making to begin with (Röper, n.d.), the structural development of the German economy hardly displays substantive financialization. For example, the percentage of GDP represented by the financial sector has increased by merely 0.5% between 1991 and 2013 (Statistisches Bundesamt, 2014). Second, pension preferences have been markedly heterogeneous within the financial services industry, which led to extensive infighting (Röper, n.d.). If key fractions of finance only reluctantly got on board with “financialized” pension policy solutions, the purchase of a structural power argument seems limited. In other words, and as shown above, to make their interests and material resources actionable, political actors within the finance area needed a common interpretation of pension reform and a discourse that could legitimize the paradigm shift, and only once this was established did their structural power come to bear on pension politics. Financial liberalization unloosened the underlying Gordian Knot in the form of the insider financial paradigm and redirected the idea of private capital formation toward old-age provision.