An unintended key indicator: The emergence of the Königstein Key in German federalism

Abstract

This article conceptualizes key indicators as institutionalized measures that are used collectively, have acquired a conventional symbolic meaning, and are of structural relevance for at least one social field. While such indicators are often assumed to be intentionally designed, we highlight unintentional emergent formation processes as a plausible and underexplored explanation. Emergent indicator formation is a process through which numbers gain practical relevance in decision-making first and are attributed symbolic meaning only afterward in public disputes. Empirically, the article presents the case of the Königstein Key, which is technically an additive index of intuitively intelligible states’ features that has become a key indicator in diverse allocation schemes of German federalism. We propose an explanatory model of an emergent, unintended formation process based on three arguments: predisposing features of the indicator itself, proven practical use, and the accrual of normative interpretations in political and public discourse. Its initial adoption in contentious negotiations about horizontal federal burden-sharing later facilitated its pragmatic adoption in various new tasks. In the course of its institutionalization as a standard instrument, interpretations in highly politicized public situations translated it into a rather generalized symbol of territorial justice at the national level. Several test situations during its trajectory reveal constitutive features of this key indicator that could also inform intentional indicator design.

Introduction

During the last decade, intergovernmental allocation schemes have been publicly debated and have received significant media attention. One prominent example at the European level is the 2015 quota scheme used for the relocation of asylum seekers (Grech, 2017). The scheme was designed to relieve arrival countries in Southern Europe, and though it largely failed in this task, very similar systems at the national level have been put in place in various European countries. In Germany, for instance, the reallocation of asylum seekers among federal states is based on the so-called Königstein Key. Its formula calculates the mandatory reception quotas of the federal states based on their tax revenue and population shares. The Königstein Key was created in 1949 by the German states (Länder) as a fiscal equalization scheme for horizontal cooperation in research policy. Later, its use was generalized for a variety of policy tasks, becoming a default mode of distributing burdens and benefits between federal states, such as in asylum policy. It often has the reputation of being a particularly fair solution. At first glance, the Königstein Key resembles fiscal allocation schemes in other countries, such as the Block Grant formula in the United States (Innes and Stoddard, 1988; Melnick, 2002) or the Barnett formula in the United Kingdom (Edmonds, 2001). However, unlike these intentionally designed and hierarchically implemented formulas, its current broad use was not intentional but instead emerged out of federal practice. How has the Königstein Key emerged rather than having been designed as a standard formula of allocation schemes in a range of policy fields and a symbol of territorial justice in German federalism?

We conceptualize the Königstein Key more broadly as an indicator (Gallopín, 1996) and its current version more specifically as a key indicator. This distinction allows us to trace the historical trajectory of this formula from an initial practical relevance to an increasingly abstract meaning, including significant structural consequences. The term indicator usually refers to numbers that are relevant to collective decision-making (Gallopín, 1996: 102). As numerical signs, indicators are instruments of measurement and, consequently, of commensuration. They define rules on how to convert qualities into quantities and make the measured objects comparable (Espeland and Stevens, 1998). However, the societal relevance of indicators is contingent upon them becoming institutionalized as criteria for information or allocation in processes of collective decision-making. Many indicators have been designed to inform decision-making but remain marginal in practice (e.g. Dorn, 2019). In several fields, a plethora of competing indicators exist without meaningful distinctions, and, hence, all of them remain toothless paper tigers. Since collective decisions highlight the contingency of the choices made, it is often crucial for practically relevant indicators to be imbued with a relatively consensual meaning to be justifiable in public debate. Once indicators have become institutionalized as relevant criteria for collective decision-making, they have structural consequences on at least one social field. We refer to indicators as key indicators

if they refer to a social phenomenon, if they are used collectively, if they are attributed a relatively consensual meaning, and if their production, publication, and use have significant consequences for the constitution, reproduction, or transformation of a particular social field. (Bartl et al., 2019: 13)

Key indicators are a relational concept distinguishing structurally more important indicators from less important ones. The Königstein Key analyzed in this article, for instance, is characterized today by having structural consequences on not only one but several policy fields. This increasingly broad application requires an explanation.

Academic literature on indicators implicitly assumes that key indicators—as we define them—are always the product of an intended, strategic design process based on conceptual clarity and methodological rigor and that indicators will have structural consequences because they match policy purposes efficiently (Sébastien et al., 2014). This can be exemplified by recent proposals to replace the Königstein Key with indicators that would enable a more efficient allocation of asylum seekers in Germany (Brücker et al., 2022; Reinhold et al., 2025) or by proposals to establish a similar quota system at the level of the European Union (Behringer et al., 2019; Hagen, 2024; Thym et al., 2013). This instrumental approach actually constitutes one ideal-typical model for explaining the creation of key indicators. We will refer to this model of key indicator creation by using the terms “instrumental,” “intended,” “strategic,” and “designed” synonymously. Unlike the instrumental model of key indicator creation, the Königstein Key acquired its status as a key indicator for German federal policy through emergent administrative practices and a gradual process of normative interpretation. It is thus an example of an emergent, unintended formation of a key indicator. Unlike scientifically designed indicators that focus on conceptual coherence, it first proved its political usefulness in more or less contentious negotiations before its conceptual validity became a disputable issue. The aim of this article is to trace the process through which the Königstein Key gradually acquired the status of a key indicator in German federalism and to identify its crucial mechanisms. Analyzing processes of unintended indicator formation might provide reflexive knowledge for scholars working on the design of policy indicators. Key indicators formed in an emergent process may be scientifically weaker than those intentionally designed, but they are politically more practical (cf. Elster, 1992).

We contribute to the literature on the sociology of quantification (Diaz-Bone and Didier, 2016; Espeland and Stevens, 2008; Mennicken and Espeland, 2019), which shares some common assumptions (Mennicken and Espeland, 2019: 238–239). Most relevant to our case are two assumptions: first, quantification can address problems of coordination and control because it provides quickly accessible information that conveys a sense of impartial transparency beyond direct physical involvement. This is why numbers are considered useful for processes of transsituational decision-making and democratic control (Porter, 1995; Rose, 1991). Second, the use of numbers in new situations of decision-making might nevertheless face critique, which requires proponents to justify their position by drawing on rather generalized principles of valuation (Boltanski and Thévenot, 2006: 40–41). In such test situations, the value of objects (here: indicators) is confirmed or rejected.

By tying together these two arguments from the quantification literature, the article advances a primarily conceptual contribution. We develop an explanatory model for the emergent and unintended formation of key indicators. The model accounts for three mutually reinforcing dynamics: (1) the constitutive features of the indicators themselves, (2) their stabilization through practical use, and (3) the gradual accumulation of normative interpretations in political and public discourse. We demonstrate the analytical leverage of this perspective by reconstructing the institutionalization process of the Königstein Key. Conceptualizing key indicators in this way has implications for empirical observation. Processes of unintended institutionalization rarely appear in explicit programmatic texts but unfold across dispersed arenas and extended periods of interpretation and generalization. To make these dynamics observable, we combine quantitative, computer-assisted topic modeling with nuanced qualitative analysis in a large corpus covering from 1949 to 2022. The research design, therefore, does not constitute a methodological innovation in itself; rather, it operationalizes the analytical requirements implied by our conceptualization and shows how established methods can be aligned to capture the emergence of indicators over time.

The article is structured as follows: first, we develop our theoretical argument. Then, we outline our research design and describe the data and methods we used. In the next section, we present our empirical findings. Finally, we discuss these empirical findings and deliberate about possible conclusions that can be drawn from our research for indicator design and for future research.

Key indicators and models of key indicator formation

How can the formation of key indicators be explained? We distinguish between an instrumental and an emergent model of key indicator formation. In the instrumental model, the conceptual framework of the indicator is explicitly formulated from the outset. In the emergent model, the meaning of the indicator arises from its repeated pragmatic use and the recursive interpretations that accrue. While concepts are theoretically crucial for constructing valid indicators (Goertz, 2020), practitioners often do not systematically elaborate on the meanings attached to an indicator but prioritize its usefulness in decision-making.

Strategic and emergent key indicator formation

The concept of an intended, strategic indicator design resulting in a (scientific) instrument key to policy formulation and implementation has been summarized, for example, by Markku Lehtonen (2017). Indicators render problems more manageable by simplifying them through quantification. They are believed to be “useful in bridging theory and empirical research, as well as the traditional gap between policymaking and social theory” (Frønes, 2007: 5). According to this instrumental view, indicators are basically scientific inventions that rationalize policymaking. Indicators have been attributed this role because they are produced through rule-based procedures that convey mechanical objectivity (Porter, 1995). We highlight two features of this model: indicators are conceived of as being designed to serve a specific purpose in an efficient way, and they are designed for that purpose in advance. Most of the literature does not distinguish between indicators and key indicators because conventionalized symbolic meaning, relevance to collective decision-making, and structural consequences are assumed to be a quasi-natural by-product of the instrumental design.

Many indicators have been created from the outset as policy tools with a particular and rather abstract meaning, such as fiscal federalism formulas in the United States (Innes and Stoddard, 1988; Nathan, 1987). However, this instrumental design does not necessarily explain their actual influence. While the GDP, for example, was scientifically designed, the enormous influence it gained in public policy cannot be accounted for by a strategic approach but was instead contingent upon emergent historic constellations (Lepenies, 2016). Hospital rankings, in contrast, although intentionally designed to influence decision-making, did not acquire the influence they were expected to (Dorn, 2019).

The complementary process of emergent key indicator formation has received much less attention. Its mechanisms have significant overlap with mechanisms of reactivity, which postulate that numbers do not depict the world but instead intervene in it (Espeland and Stevens, 2008: 412). While we build on this idea, our argument is more specific: a combination of predisposing features of the indicator and two reactive mechanisms contributes to a “successful” trajectory of key indicator formation.

First, indicators emerge from a practice of decision-making that is historically situated and in which certain numbers are likely to trump expert judgements, especially in controversial situations when they are seen to represent a form of mechanical objectivity. Official statistics are typically regarded as objective information and are therefore often linked to collective decision-making (Desrosières, 1991). Indicators that are built into decision-making are more likely to become taken for granted than others (Espeland and Stevens, 1998: 329). Second, ad hoc quantitative criteria used to implement a local decision are not attributed with a more abstract meaning until they are justified to a larger public (cf. Boltanski and Thévenot, 2006). The meaning attributed to practically relevant indicators accrues in a piecemeal way and might change during their life cycle. Producing and embedding numbers in larger projects is a form of distributed cognition and accomplished by largely invisible institutional work (Espeland and Stevens, 2008: 411–412). Tying indicators to particular (new) networks of meaning is likely to change some but not all of their features, depending on the contexts into which they are translated (Callon, 1984; Latour, 1987). To the contrary, previous investments are likely to be considered (Thévenot, 2023). Third, if we take the materiality of indicators and their agentic capabilities seriously, their constitutive features can be expected to resist immediate change. Hence, despite the agency that can be ascribed to human actors, the process that we reconstruct in our analysis was probably not envisaged by the actors who translated the Königstein Key into ever-new contexts and may be attributable in part to enabling and constraining features of the indicator itself.

According to convention theory, the arguments of persons are evaluated according to a plurality of universal principles in prearranged test situations (Boltanski and Thévenot, 2006: 133–137). While Latour understands tests to be trials of the strength of facts, in Boltanski and Thévenot’s approach, objects in test situations prove the legitimacy of moral claims (Guggenheim and Potthast, 2012). We assume that public and parliamentary debates are of special relevance for testing the legitimacy of territorial allocation criteria. If such tests are passed, locally created allocation criteria might acquire a more general symbolic meaning. While in models of intended indicator creation, the meaning of indicators is determined by experts or scientists; in the model of emergent indicator formation, meaning accrues through administrative action and through public and political debate.

However, the models of strategic and emergent indicator formation overlap when key indicators are codified in a legal framework. While legal codification might be the intended form of indicator stabilization in the first model, in the emergent model of key indicator formation, legal codification is not aimed for from the outset. Nevertheless, legal codification might turn out to be (the final) part of an emergent process of key indicator formation that aims to stabilize the use of the indicator in the long run. While an emergent process of key indicator formation can largely unfold without the elaboration of a consistent concept of what the indicator stands for, this changes when it comes to the indicator’s legal codification. Legislative processes in a democracy are a strong way to test political arguments, as they force participants to make their normative and factual claims explicit consistently if they want to succeed. While the use of indicators is precarious as long as it depends upon the discretion of decision-makers, legal codification fundamentally alters this situation and provides strong premises for administrative decision-making and, hence, consequences.

The choice of allocation criteria

Although indicators are often conceived of as scientific measures of a coherent concept, it is important to note that in our model, decision-making requirements come first and conceptual issues only arise later. Our model applies mainly to situations of political negotiations instead of scientifically rigorous testing. Negotiations about the allocation of burdens and benefits to political territories are notoriously contentious when resources are scarce. Jon Elster’s work on local justice provides us with a classification of allocation criteria for goods that analytically describes the choice set of decision-makers: egalitarian criteria, queuing mechanisms, status-based categories, and criteria that are based on the recipient’s individual situation and performance (i.e. desert, need, and efficiency). Equality is often seen as the baseline for allocation as long as there are no substantive reasons against it. Point systems lend themselves particularly well to compromises in situations of coalition formation and bargaining (Elster, 1992: 103). Apart from the choice of allocation criteria themselves, political compromise also depends on how goods and burdens are to be allocated and how the group of beneficiaries is defined (Elster, 1992). Modifying such implementation specifications affects the outcome while leaving the main allocation criteria unchanged.

Similar to Porter (1995), Elster (1992: 169) postulates a historic trend from expert judgement toward more mechanistic allocation criteria because of their time efficiency, low information costs, and low susceptibility to corruption. Furthermore, the results of allocation schemes must be justifiable to the public (Elster, 1992: 175). If we consider that different orders of worth can apply in public debates (Boltanski and Thévenot, 2006), this sensitizes our empirical analysis to identify test situations in which the status of the Königstein Key is controversial.

Allocation formulas by policy design

With notable exceptions (Innes and Stoddard, 1988; Nathan, 1987), little analysis has been conducted on the institutionalization of political allocation formulas in the literature on quantification. This contrasts with empirical relevance, as reflected, for instance, in institutions of fiscal federalism (OECD, 2021), formula funding of schools (Levačić, 2008), or quota-based refugee dispersal policies (Darling, 2022).

Existing case studies have been conducted mainly on indicators explicitly designed for certain allocation schemes. In the United States, federal grants to local governments and states increased from 10% of their total revenue in the mid-1950s to 30% in the mid-1980s (Nathan, 1987: 331). This increase posed serious administrative problems to the system of categorical, special-purpose grants that had originally been in place. Furthermore, the proliferation of conditional grant programs also became rather nontransparent for the targeted jurisdictions. In this situation, the incipient use of information and communication technology made a transition to formula-based federal grants viable. Over 140 federal grant programs used formula funding in the mid-1980s (Innes and Stoddard, 1988: 95). Richard Nathan (1987) analyzed the use of numbers in the revenue-sharing program (running from 1972 to 1986) and the Community Development Block Grant (CDBG) (in place since 1974).

The revenue-sharing program was meant to allocate grants to all local governments but specifically target those in need. Initially, a formula was proposed by an expert committee; however, this was modified during the legislative process. In the end, the

Senate and House passed different formulas for allocating shared revenue to state areas. The Senate adopted a three-factor formula that favored rural and smaller states. The House preferred a five-factor formula more favorable to urban states. The compromise struck at the last minute consisted of having each state receive its allocation under whatever formula was more favorable to that state. (Nathan, 1987: 336)

Although experts perceived there to be serious problems in understanding the formula and its distributive effects, it was never changed so as not to reopen the political debate (Nathan, 1987: 334, 337). The revenue-sharing program required data on population, income, and revenues, which mainly came from the decennial census and increased pressure on the census’s accuracy (Nathan, 1987: 335).

Eligible jurisdictions for the CDBG were originally the central cities of Standard Metropolitan Statistical Areas (SMSA) and urban counties (over 200,000 inhabitants). The program used three main factors: population, overcrowded housing, and poverty (double weighted) for the allocation of funds (Nathan, 1987: 339). However, after an initial transition period from categorical grants, the program was assessed as lacking the urban focus that was seen as the political goal during legislation. To fix the problem, the Brookings Institution proposed a second formula “including the number of pre-1940 housing units that would more strongly reflect physical development needs” (Nathan, 1987: 340). In the end, both formulas were introduced, which was attributable to a mismatch of the original formula and the policy goals, the additional funding that became available, and the existence of official data needed for implementing the second formula.

These empirical studies confirm the increasing relevance of mechanical allocation criteria and show that point systems seem to be preferred over single indicators because of their ability to facilitate compromise. While all analyzed point systems include population as an egalitarian element, they also include more targeted variables that correspond to the explicitly stated policy goals. In the first case, the intense political controversy at the onset led to a path-dependent formula that remained unchanged despite criticism. In contrast, in the case of the CDBG, the political debate surrounding the consistency of policy targets and instruments catalyzed change. These diverging trajectories remind us of the historical contingency of an indicator’s life cycle. The allocation formulas that persevered in the end in these two cases can be regarded as key indicators for their policy field because they are used in collective decision-making, they refer to a rather abstract policy purpose, and they have structural consequences. However, unlike the Königstein Key, neither of these formulas was translated to other policy fields. Furthermore, while the choice of allocation criteria was subject to political negotiation, the formulas were implemented by expert-informed design and not as part of an emergent process that was unintended from the outset.

Data and methods

To analyze the Königstein Key within both its practical and its symbolic dimensions, we combine two types of material: archival documents and newspaper coverage. Archival documents provide insight into the administrative and political practices through which the indicator was first developed and institutionalized; they provide insights into its practical genesis. Newspapers, by contrast, reflect how these practices were perceived, communicated, and problematized in the public sphere, capturing the discursive formation of meaning. Following the logic of methodological triangulation (Flick, 2010), juxtaposing these sources allows us to examine how practices of use and public representations evolved in relation to each other. Divergences between the two are particularly informative for understanding different stages in the emergence of a key indicator.

A second analytical strategy for reconstructing the generalization of the Königstein Key is to compare test situations in which the use of the indicator was questioned or confronted with alternative allocation criteria. This approach follows the logic of minimally contrasting comparisons, allowing us to identify the two crucial mechanisms underlying its emergent formation, as well as the predisposing features that repeatedly facilitated and stabilized the status of the Königstein Key as a key indicator. While these features became visible in the polity of German federalism, they also point to more general situations of coordination and legitimation in other polities. Our approach, therefore, aims at analytic generalization (Yin, 2013), seeking to specify how particular configurations shape the emergence of key indicators.

The reconstruction of the practical origin of the Königstein Key draws on various archival documents (Ventresca and Mohr, 2002), such as minutes of meetings, reports, and printed matter that were produced as part of the negotiations between the Länder in various areas of joint policymaking. This includes its early historical formation phase, when the Königstein Key was not referred to as such but rather in terms of a “distribution according to the Königstein Agreement.” Furthermore, the more recent use of the Königstein Key is documented in (probably not complete) lists provided by public administration (GWK and ZDL, 2016) and one state parliament (Hamburgische Bürgerschaft, 2013).

The public discourse on the Königstein Key is captured through a text corpus of newspaper articles. Politicians usually have a strong interest in newspaper reporting, as it provides the public with essential information, both shaping and reflecting their opinions. Conversely, the newspaper articles themselves are often based on the statements and actions of the political protagonists and take them as an essential point of reference or reproduce the ideas of functional elites (DiMaggio et al., 2013: 573). We analyze a total of 561 articles from the national German newspapers Die Zeit, Frankfurter Allgemeine Zeitung (FAZ), and Süddeutsche Zeitung (SZ), published between 1949 and 2022, which contained (in German) the keywords Königstein Key or Königstein Agreement. The selected newspapers have the advantage that they were published regularly (daily or weekly) during the long period examined by the study; they have a continuous archive and are characterized by a national focus on German politics. The three newspapers we analyzed are usually regarded as quality media that closely cover political events, especially when it comes to very specialized subjects such as, in our case, horizontal federalism. They have more or less detailed regional sections that also cover events on a local level. The FAZ and SZ have been the most widely read newspapers by members of the German parliament and are considered to be influential for the reporting of less prestigious newspapers (Gerhards, 1999). We therefore see the selected newspapers as crucial for the analysis of the emergent formation of consensual meanings of indicators.

The newspaper articles are analyzed using topic models. Topic model analysis is a semi-automated, computer-assisted text mining technique using an iterative algorithm that sorts texts and the terms they contain according to thematic structures (so-called topics) that can be interpreted as contexts of meaning (Papilloud and Hinneburg, 2018). This method enables a complete and detailed evaluation, especially for large amounts of text, without unconsciously omitting or subjectively neglecting parts (DiMaggio et al., 2013: 577). According to common linguistic and cultural sociological understanding, the meaning of a text does not lie within the terms but is instead determined relationally (DiMaggio et al., 2013: 578). In our case, non-negative matrix factorization (NMF) with six topics provided the best results (Figure 1).

Figure 1.

Topics of articles referencing the Königstein Key in national newspapers, from 1949 to 2022.

Computer-assisted categorization has high reliability and intersubjective reproducibility (Nelson, 2020: 6); however, to be applied in the social sciences, the inclusion of external information and the contexts of text production is often important, as is a subsequent close reading (Nelson, 2020: 25). Accordingly, we followed up the topic modeling with a qualitative content analysis. Using an inductive approach, we focused on recurring patterns of representing the Königstein Key and the Königstein Agreement (Kuckartz and Rädiker, 2022). This allowed us to identify dominant narratives and ways of reasoning associated with these terms. While the topic modeling primarily explores the thematic structures of the entire corpus, the qualitative content analysis enables a nuanced interpretation regarding arguments and figures of thought. By integrating computational and interpretive approaches, we leveraged the strengths of both methods: the scalability and rigor of machine learning techniques with the contextual sensitivity and reflexivity of qualitative analysis.

The emergence of the Königstein Key

The central, striking feature of the Königstein Key, which distinguishes it from many other key indicators, is the breadth of its use across policy fields. An initial impression of the policy fields in which the Königstein Key is used can be gained from our topic modeling of national broadsheets. Figure 1 shows the evolution of topics over time. The labelling of the topics is not a result of the algorithm but instead our own interpretation. The topics labelled Research, national and Research, local are related to research funding, the first policy area to which the Königstein Key was applied. Both topics have a strong link to the Königstein Agreement. While the first primarily depicts the debate about the division of tasks between the federal government and the Länder governments in research and cultural policy, the second reflects the local perspective of the particular research and cultural institutions as the recipients of funding. The remaining four topics are linked to the second search term defining the corpus: the Königstein Key. The topic labelled Federal Finance is a very heterogeneous cluster, referring to various policy fields in which the Königstein Key has come to be used. The significance of the Königstein Key in asylum policy is demonstrated by as many as three topics relating to debates about the dispersal of refugees with temporal and scalar subdivisions (Asylum Policy 2015, Local Reception, Ukrainian Refugees).

During the early phase, there are only the topics on research and cultural policy constituted by their reference to the Königstein Agreement. In the 1970s, the overall frequency of newspaper articles decreased. The number of mentions in newspaper articles virtually exploded after 2014, catalyzed by the perceived crisis of asylum reception, long after it had been established as a standard instrument for cooperation between the Länder across various policy fields. It is likely that the significant role suddenly attributed to the Königstein Key in asylum policy also spotlighted the various other fields of application. A second peak in attention occurred in 2022, again linked to a refugee topic, this time the admission of refugees from Ukraine.

It is striking that the peak phase of diffusion of the Königstein Key from the late 1960s onwards, which can be reconstructed from the official lists of use cases (GWK and ZDL, 2016; Hamburgische Bürgerschaft, 2013), is basically not echoed in the media coverage. The policy of horizontal cooperation between the federal states is an area that generally tends to receive little public attention (Krumm, 2015: 48). In this period, the Länder establish numerous administrative institutions aimed at nationwide coordination, for example, in education, in the training of the police and judges, or in the accreditation authorities. In most of these cases, the Königstein Key is applied to sharing costs between the Länder, which is close to its original purpose. It is a phase when the Königstein Key is deliberately transferred into policy fields beyond research and culture. The cultural ministers still have their share in these decisions, but it is also established as the favourite allocation mechanism for the ministers of the interior, health, or justice. The fact that this widespread practical use did not coincide with media attention and that it still lacked the later symbolic meaning indicates an early stage in the emergence of the Königstein Key as a key indicator.

Today, the Königstein Key is the allocation criterion in more than 100 administrative tasks and different policy fields; however, the exact number of uses is difficult to obtain. It has not only been applied to horizontal cooperation in burden-sharing but has also started being used for the implementation of vertical fiscal transfers. The shift from the distribution of costs to benefits is particularly relevant for generalized notions of justice, creating situations in which it is precisely the rich states that benefit more than the poor. A further modification of the “good” to be allocated can be observed in the translation of the Königstein Key to asylum seeker dispersal among the Länder. This implies a shift from fiscal to physical responsibility-sharing.

Origin and stabilization in the administrative routines of federalism

Indicators tend to be perceived as intentionally designed policy tools, but the early history of the Königstein Key clearly illustrates the significance of the distinction we make between intentional design and emergence. The various decisions for creating and establishing it are not the result of a coherent strategy but of balancing interests and finding pragmatic solutions.

Administrative creation

What later became known as the Königstein Key initially only defined a specific quota scheme for the financing of non-university research institutions in a mode of horizontal cooperation among the German Länder as part of the Königstein Agreement. This agreement between heterogeneous Länder was achieved in March 1949 in Königstein, in a period immediately preceding the foundation of the Federal Republic of Germany. After the Second World War, German non-university research institutions, including prestigious ones like the Max Planck Society, faced substantial funding challenges. The Länder, newly established by the Allied forces, saw the necessity to shoulder this task, but the burden varied significantly among the states due to the relocation of research institutions away from heavily bombed urban areas to less populated regions (Osietzki, 1984: 77; Pfuhl, 1959: 285). The Länder wanted to cooperate for two reasons: first, there were concerns that some Länder would not be able to cope with the high expenditures on their own (HHStAW 504/5771). Second, the Länder governments were concerned about the upcoming foundation of the Federal Republic, manifest in the deliberations on a new constitution (Osietzki, 1984: 240). The hitherto strong political role of the Länder was threatened by the potential increase in influence of a central government. The Länder strived to preserve their jurisdiction in as many areas as possible and were eager to prove their capacity to cooperate (Pfuhl, 1959: 288). The Königstein Agreement was thus considered to be a strong signal of functioning horizontal cooperation that protected them against centralizing tendencies.

The allocation of costs was deliberately placed at the end of the negotiations leading to the Königstein Agreement. They focused first on the research institutes that required collective funding, and only once the funding requirement was clear could the cost distribution be determined (BArch Z4/588). This situation is reminiscent of Elster’s point that in allocation schemes, not only are the allocation criteria subject to decision-making but also the objects and recipients of the benefits. A central contention was how to relieve financially overburdened states while achieving the approval of all participants. An early recommendation by negotiators was to use tax revenue as an indicator for cost allocation (HHStAW 504/5771: 13). The final agreement in Königstein on a point index based on taxes and population (in a ratio of 2:1) shows a clear weakening of the redistributive effect compared to the original proposal. The relatively prosperous city-states of Hamburg and Bremen were thus able to reduce their shares by around 20%, respectively. In contrast, more rural and poorer states not only hosted most of the relocated research institutes but also had to contribute more due to this adjustment. Lower Saxony, Rhineland-Palatinate, Bavaria, and Schleswig-Holstein recorded 9% to 18% higher shares. In other words, the formula was, in the end, less targeted than the policy’s purpose would suggest.

The distinctive form of this allocation scheme is thus the result of a compromise based on the bargaining power of participants in a specific historical context (cf. Elster, 1992: 135–183). By this time, statistical information on taxes and population had already become sufficiently black-boxed to serve as icons of states, indicating their (lack of) purchasing power and crude political size. The negotiations translated the veto power of heterogeneous “small” and “large” states into a formula that they would accept because it made them commensurable on an equal scale. While commensuration is often mentioned as one mechanism of reactivity (Espeland and Stevens, 1998), we think it is noteworthy to pay attention to the particular form of quantification that was applied. The creation of an additive index allowed rich states to use their bargaining power to undermine the redistributive effect while still presenting the solution as a quasi-natural choice in terms of (proportional) equality. Consistently, there was no explicit justification for adding the population criterion. The translation of the official statistical categories into the allocation formula of the Königstein Agreement—which did not yet have a proper name or even a justification—is therefore only the first stage of an emergent indicator formation.

Administrative diffusion

The Königstein Agreement was extended several times while the funding volume increased significantly (HHStAW 504/5257-9). The original intention of the Länder to keep the federal government out of research funding failed (Scharpf, 1989: 40). The defensive stance of the Königstein Agreement thus became obsolete. The distribution key, which originally represented more of an ad hoc solution, was, conversely, to become a standard instrument for cost allocation in horizontal federalism. The Standing Conference of the Ministers of Education and Cultural Affairs (Kultusministerkonferenz) continued to be particularly defensive toward the influence of the central government in its policy field (Schulz-Hardt and Fränz, 1998: 182–184). One can assume that the cultural ministers, many of whom had already been involved in the negotiations leading up to the Königstein Agreement, also regarded the former success to be partly of their own making. Subsequently, the cultural ministers resolutely pushed ahead for the establishment of cultural institutions under the responsibility of the Länder. To fund these institutions, the Länder resorted first to an allocation that was proportional to the population share. This can be seen as an expression of conventions for representing political equality, which had emerged just decades earlier (Heintz, 2012). However, in 1954, they switched to the Königstein Key as well (BArch B304/5289). It is in this instance of translating the allocation formula to a new policy field that the formula is coined “the Königstein Key.” It explicitly marks the new interpretation of the formula of the Königstein Agreement as a symbol in its own right, which was arguably less intuitively intelligible than the population indicator it replaced. We identify this as the second stage of emergent indicator formation, which is less situational and more generalized due to the name it has acquired and the test through which it superseded a competing indicator. Both arrangements—the Königstein Agreement and the joint cultural tasks of the Länder—were open to the addition of new items to be financed, and this represents the beginning of the Königstein Key’s trajectory. For the first 20 years, this was the testing ground for the horizontal funding of Länder projects according to the Königstein Key, which then significantly diffused into other policy fields from the late 1960s onwards.

For the bulk of the articles that were identified through the search term Königstein Key, the origin of this instrument is of little importance. Instead, the articles emphasize that the Königstein Key is a standard procedure that is widely used in horizontal cooperation and that has a high level of acceptance. Wordings like “usual” or “proven distribution” appear particularly frequently. The continuous use of the Königstein Key becomes a form of justification in and of itself and reinforces further use. The precise formula, that is, the components of tax revenues and population size, is usually not mentioned, let alone discussed. Thus, a cursory view dominates public perception in which only the function of the indicator as a simple solution to distributional conflicts matters. Public interpretations of the key are no longer situational or related to the actual depiction of Länder attributes. Instead, the key is conceived as a quasi-natural point of agreement in problematic negotiations between the Länder. Discussing the adequacy of the formula’s base indicators as allocation criteria for particular policy targets would have made it more time-consuming and also more difficult to reach an agreement.

Accruing normative meaning in disputes about territorial justice

Even though today there is no official conceptual underpinning of the resulting quota, the insistence on the Königstein Key by various actors derives much of its power from explicitly formulated references to notions of justice that have solidified over time. While it has long been a rather quiet policy instrument, awareness of it has increased, especially in more conflictual policy fields, evoking explicit justifications and critique in the name of justice. We consider this to be a third stage of emergent indicator formation as we identify its translation into a policy tool with normative framings that arise especially when the key is applied to new policy tasks that are of a different nature than the original fiscal burden-sharing: physical responsibility-sharing in asylum policy and the allocation of financial subsidies. Physical responsibility-sharing is different from fiscal burden-sharing because it is based on allocation in kind, and it aims to share intangible (political) responsibility, which makes it more contentious. Financial subsidies are different from cost-sharing, as, in these cases, the direction of the redistributive effect is inverted, which is even more in need of an explanation.

Normative generalization

The first indication for our claim about piecemeal attribution of meaning is a notable speech on the distribution of asylum seekers in the Bundesrat in 1981 in which the notion of justice was explicitly linked to the Königstein Key for the first time: “When it comes to fairly distributing burdens that must be borne jointly among the Länder, we usually apply the Königstein Key” (Deutscher Bundesrat, 1981: 72). It marks a situation where, after severe conflicts between the Länder in the 1970s, the national distribution of asylum seekers was highly politicized. Proposing the Königstein Key as a way to determine mandatory reception quotas for the Länder was intended to depoliticize the situation and turn years of stalemate into policy implementation. Soon afterwards, the Königstein Key became legally codified as the allocation criterion for the dispersal scheme in asylum policy. This example shows that the narrow circle of politicians and administrators using the Königstein Key had come to infuse the formula with a more profound meaning that went beyond a pragmatic sense of fiscal burden-sharing.

This is further enhanced when the Königstein Key becomes a benchmark for just distributions between territories, regardless of where burdens or benefits are subject to distribution and regardless of which subnational territories are concerned. For instance, several newspaper articles report “skewed” distributions of resources between the Länder because they deviate from the envisaged shares of the Königstein Key (SZ, 1985). By referring to the Königstein Key as a benchmark, alternative allocation schemes are delegitimized by the territories that perceive themselves to be disadvantaged by them. Remarkably, this also takes place at political scales to which the key had not been applied until then. The vertical distribution of subsidies by the federal government (SZ, 2021) or even the distribution of resources between counties (FAZ, 2015a) is compared to the Königstein Key in this manner. This interpretation in public goes hand in hand with the actual development of the legislative practice of using the Königstein Key for the allocation of federal funds to the Länder as a means of spatial policy. In this framing, the Königstein Key appears to be a key indicator of justice between different German territories in general.

Justification and critique in public debates

In the context of the repoliticization of asylum policy after 2015, dispersal turns out to be much more controversial for the public than the conflicts over fiscal federalism. We therefore consider these public debates as a fourth stage in the translation process as the Königstein Key becomes known to entirely new groups of actors, some of whom question and criticize its use. With the increased influx of asylum seekers in 2015, the use of the Königstein Key is highlighted by various actors as a central problem-solving approach to burden-sharing and the creation of “orderly conditions” in the reception system: “The most important key to this success is one that the Länder have long agreed upon: The Königstein Key [. . .]. But the Federal Republic has probably never applied it as consistently as it does now” (SZ, 2015b). For the Länder, the role of the Königstein Key is once again to act as a stable, solidarity-based instrument that facilitates their cooperation. The Königstein Key provides a transparent and precise public benchmark that ensures solidarity by clearly spelling out what contributions federal states are expected to make in hosting asylum seekers. Own efforts and those of others are judged and justified by referring to the quotas of the Königstein Key. Munich’s mayor, Dieter Reiter, for example, argues:

If all the federal states were similarly full as we are in Munich, which would have to take five percent according to the Königstein Key, then 200,000 would have arrived in Germany in one day. I would have noticed that. (Zeit, 2015b)

For the Länder and subsequently for municipalities, the distribution quota is relevant as an indicator for planning capacities. Moreover, by referring to the non-negotiable obligation to take in the amount specified by the quota and by pointing out the proportional equality principle implemented by the quota, politicians defend the reception system in the face of local criticism (Zeit, 2014). These public debates imbue the Königstein Key with a rather general notion of territorial justice.

In 2015, the Königstein Key was supported by opposing sides: Bavarian border regions were calling for faster redistribution and solidarity based on the established allocation scheme (FAZ, 2015b). But also, the Eastern German states, affected by lower direct inflows of refugees, argued for the retention of the established allocation scheme (SZ, 2015a). Proposals by politicians for adjustments in the distribution scheme tend to be cautious, often also suggesting financial compensation in return. Reopening the dispersal debate that dates back to the mid-1970s endangers the consensus among the Länder as well as a working reception system. Nowadays, the Königstein Key has become unquestioned among politicians as a way to link physical responsibility-sharing. This becomes particularly evident in newspaper reports on the response to the influx of Ukrainians in 2022. At first glance, the political reactions differ considerably from the standard procedure for asylum seekers, as it formally does not apply to them due to the triggering of the European Union’s Temporary Protection Directive. Ukrainians were allowed to stay in Germany for 90 days without a visa and, upon registration, are immediately granted a temporary residency permit. Nevertheless, Berlin, as the center of in-migration, soon complains of being overloaded and uses the Königstein Key to justify this claim even though it is formally not applicable. For politicians, it goes without saying that territorial distribution according to the “indispensable” Königstein Key is required (Zeit, 2022).

At the municipal level, even though the Königstein Key formally does not apply, it has become a symbol for the difficulties of asylum policy. The latter becomes clear, for example, in a report on local willingness to welcome refugees: “The initiators of the citizens’ petition only want to take in as many refugees in Moosach as a calculation based on the Königstein Key stipulates for the municipality” (SZ, 2016). Hence, explicit references to the Königstein Key as a benchmark for territorial justice in situations beyond its formal jurisdiction probably give the strongest indication that it has acquired the status of a key indicator, although it was never designed as such.

In the debate over the physical distribution of responsibility in asylum reception, there are also prominent voices outside of administration and politics that sharply criticize the Königstein Key. Researchers and policy advisors have denounced it as “inappropriate” in the context of asylum policy, referring most often to requirements of the housing market or the labor market (Zeit, 2016). The voices of volunteers and refugees are rarely represented in newspaper articles. They fundamentally oppose a “rigid” mechanism of physical allocation, which is seen as a severe state intervention in personal lives (Zeit, 2015a). While the explicit justice frame is debated most vividly in physical responsibility-sharing, critics in other fields who are concerned with the distribution of financial funds, for example, in education policy, bemoan that the Königstein Key prevents an efficient redistribution to weaker states by favouring Länder with higher tax revenues. Strikingly, newspaper reports no longer recount that the Königstein Key was originally designed for allocating fiscal obligations, not resources. A representative of the federal government correspondingly expresses their resignation: “If something is to be implemented in financially weaker Länder, then, in certain cases, we must also accept that we give money to better-off Länder” (SZ, 2022).

Limited success of alternative indicators

In contrast to the rather general application of the Königstein Key, proposed alternatives are issue-specific and designed to fit a particular policy purpose. One of the structurally most significant examples emerged only recently, which is the first national program for compensatory funding of schools in deprived areas. It was suggested, for example, that the share of recipients of social assistance should be used as the criterion of allocation instead of the Königstein Key (FAZ, 2022). In the end, alternative criteria were codified only for a part of the program—and only after significant investments in these alternative forms. First, elaborate reports by scientific experts criticized the Königstein Key as unfair because it would counteract the main purpose of the program (Fickermann et al., 2022; Helbig, 2023). Second, pressure from the national government (zwd Politikmagazin, 2023) and careful moderation by a leading and experienced education policymaker for the Länder facilitated that compromise (Wiarda, 2025). Unlike the Königstein Key, these alternative allocation criteria have been designed to a large extent by academic scholars and quickly gave rise to demands for evaluation, even before the program has actually been fully implemented (Sendzik et al., 2025). It remains to be seen if evaluation studies will confirm or delegitimize the use of these alternative allocation criteria. The Königstein Key has never been subject to formal evaluation.

Predisposing features of the key indicator

Until now, we have analyzed the administrative institutionalization of the Königstein Key and the public ascription of symbolic meaning at the various times it was applied. We now turn to the predisposing, formal properties of the Königstein Key, which have paid off well, particularly in Germany’s federalized system, and favoured its generalization. These are most visible during recurring, often conflictual test situations in which the Königstein Key prevailed against its rivals. In these situations, the key was selected and sometimes adapted to facilitate cooperation.

Low information costs

In the early phase of its use, the Königstein Key, as a composite index, was primarily in competition with its own components. In 1949, it prevailed against the use of tax revenue as a single indicator for burden-sharing in research funding, and in 1954, it prevailed against the use of the population share of states as a single indicator for burden-sharing in cultural policy. Both statistics—population and taxes—are readily available and simple to understand. Official statistics appear to be a “natural” choice for gathering information because state variables have come to represent the state itself, thereby creating quasi-natural spaces of equivalence that can be combined with each other to form indices (point systems). Composite indices preserve the original sense of equivalence while translating it to a new, more abstract dimension. Population and tax statistics are figures that are inherently significant tools used by governments and, thus, also by the negotiating ministers of the Länder. Due to a century-old investment in forms, allocation formulas based on existing statistics do not produce high information costs and are relatively easy to interpret for those acquainted with them. This is reminiscent of Elster (1992: 145), who points to the importance of professional norms as the first best conceptions of allocators when choosing an allocation mechanism. Policy indicators are assumed to be more legitimate when their content can be intuitively interpreted (Atkinson et al., 2002: 21–24). While the concepts of population and taxes are probably also intelligible to the wider public, this is less clear when these components are integrated into an index. However, compared to single statistics, composite indices have the additional advantage of allowing for the formation of compromises between diverging interests (Hofstee, 1990).

Facilitating compromise

Finding compromise is crucial in negotiations between the Länder as they operate in a mode of unanimity most of the time, mainly because that is the only way that their policy can exert leverage (Scharpf, 1989). Thus, a tool to facilitate agreement and thereby expand the collaboration was repeatedly needed. Compared to discretionary, situational criteria of allocation, mechanical allocation criteria, when generally accepted, can alleviate negotiations not only with respect to their information costs but, importantly, also to their transaction costs.

We exemplify this with a thought experiment that is based on counterfactual alternatives to the Königstein Key. In 1949, during the constitutive negotiations on research policy, one possible alternative for burden-sharing remained beyond debate: burden-sharing based on expert judgement or political pledging of absolute Länder contributions. At the time, it was not unusual to devise such discretionary solutions. For instance, responsibility-sharing of refugees from Central and Eastern Europe was based on political pledging of absolute numbers of people to be relocated among the Länder (Bartl, 2021). This meant that similar lengthy negotiations had to be taken up periodically as long as that problem persisted. Another example can be found right next to the key: the research institutes to be included in the shared funding scheme of the Königstein Agreement were to be decided upon on a case-by-case basis as part of arduous negotiations because each of them was considered unique and “incommensurable” (Espeland and Stevens, 1998). Hence, resorting to official statistics for cost-sharing—instead of pledging cost contributions—provided relief from arguments of incommensurability based on “mechanical objectivity” (Porter, 1995).

A closer look at the exact implementation of allocation between the Länder up until the present day reveals that the Königstein Key is often at the mere surface of the actual scheme applied. Often, it is the last allocation mechanism following various steps of detailed specifications, such as the scope of allocation, which is quite decisive for allocative effects. The prior determination of a distribution model for costs, which can sometimes have a rather symbolic value, allows the main discord to be shifted to the definition of the goods to be allocated, where conflicts are resolved more easily by issue-related arguments, overlapping benefits, and tie-in deals (Scharpf, 1992). This mode of achieving consensus without repeatedly bringing up the issue of allocation criteria, but still having a certain leeway to maneuver, was very suitable for achieving the agreement of all Länder and gradually expanding it into new areas.

The prior determination of a model for burden-sharing also allows the flexible adaptation of the targeted actor group (cf. Elster, 1992). The Königstein Key sometimes underwent specific, discretionary modifications of the actors involved. The number of German Länder has changed a few times since 1949, most notably with the German Reunification. In these situations, the Königstein Key turned out to be a very flexible instrument because the calculation of shares does not depend on the number of participants, so no renegotiations of the criteria are required. There have been cases, for example, in which the federal government was integrated into the allocation scheme with a discretionarily fixed share or at a level that was equivalent to the largest state. On other occasions, Länder-specific fixed shares were added to the key in a discretionary way to avoid redistribution effects that were perceived as excessive. Although the Königstein Key is a standardized allocation scheme, it is not understood by policymakers to be a restriction, as it might first appear; it can be and has been adapted pragmatically to particular situations.

Taming uncertainty

Uncertainty is a challenge for the design of allocation schemes, for which we identify two potential sources: first, with fixed allocation criteria in place, it is questionable to what extent they reflect changes that are relevant to those affected by the allocation mechanism. Second, in order to be a credible source of “mechanic objectivity” (Porter, 1995), the indicators used as allocation criteria should be relatively resilient against manipulation (cf. Elster, 1992). The indicators on which the Königstein Key is based appear to fulfil both of these conditions. Annual recalculation provides a certain safeguard against the uncertainty of the future, as defining relevant indicators provides for dynamic adjustment, taking changing economic and demographic conditions into account. For instance, between 1982 and 2004, the Königstein Key was used for national responsibility-sharing in asylum policy with fixed shares for states that were determined based on the values of the key’s indicators at the time of its legal codification in 1982. After reunification, the inclusion of the new German Länder was implemented in a discretionary way, with their collective share fixed at 20%, which roughly corresponded to the population share of the new German states. This soon led to a delegitimization of the fixed quota because of the demographic decline in East Germany in the decade after reunification. However, in general, population and taxes are rarely subject to short-term fluctuations. Thus, despite dynamic adjustments, the allocation scheme is also characterized by a relatively high degree of stability and predictability. This is particularly important for the budgetary and administrative planning of governments. In addition, allocation criteria that can be manipulated by the target group are likely to generate unintended behavioral incentives (Elster, 1992). Population and taxes are variables for describing states that are hard to corrupt in the short run when statistical offices are institutionally autonomous (Prévost, 2023).

A symbol of territorial justice?

While balancing interests as part of political negotiations initially had the upper hand when it came to the meaning of the Königstein Key, this gradually changed as it became more successful, and explicit justifications in the name of territorial justice became increasingly more important. Concepts of justice tend to be universalistic (Boltanski and Thévenot, 2006), and the Königstein Key proved to be compatible with several arguments across a wide range of situations. When the key is challenged by alternative modes of burden-sharing in the horizontal cooperation of the Länder, counterarguments always present situational justifications of alternatives that are irrelevant for other cases. The Königstein Key, in turn, is not only backed by habituation but also as a justifiable solution to repeated conflicts about horizontal solidarity. The Königstein Key reflects the constellation of German federalism in which there are states that differ greatly in size and in their capacity to fulfil their tasks. Horizontal cooperation in German federalism builds on the fundamental idea that neither poor states should feel unable to meet their obligations and see themselves constantly disadvantaged nor that rich states should reject what they see as inappropriate redistribution. The double-weighted component of tax revenues reflects the ability of the states to contribute and represents a principle of solidarity. Population size, in turn, tends to be linked to a principle of civic equality that does not take economic differences into consideration. It might well be that, because it encouraged the repeated agreement of all participants, the Königstein Key came to be seen implicitly as a symbol of procedural justice well before it was explicitly justified as a symbol of distributional territorial justice. However, in the face of benefit allocation, the adequacy of this instrument is no longer clear. As with the cost allocations, we found no explicit justifications as to why it was adequate in the name of justice. In fact, the critique it faced in debates about physical responsibility-sharing in asylum policy and in the allocation of fiscal benefits to the Länder signals that its generalization also has limits. While it remained in place in asylum policy, it had to succumb to an alternative in the case of formula funding for schools.

Discussion and conclusion

Why has the Königstein Key become a key indicator of territorial justice in German federalism? Despite the plurality and interdisciplinarity of social studies of quantification (SSSQ), the formation and use of indicators have mainly been conceptualized as a top-down product of scientific innovation. Our article, in contrast, has highlighted a process of emergent indicator formation: The Königstein Key was not invented as a matter of purposeful design but as a rather ad hoc solution in difficult negotiations between the Länder. Only afterwards did it gradually become a key indicator that was given specific, explicit interpretations. Summarizing our analysis, we considered four stages in the Königstein Key’s trajectory in becoming and remaining a key indicator for territorial justice in Germany (Table 1). At first, the established state indicators of population and taxes were integrated into a new framework, which, for the time, was meant to be purely situational. The unquestioned interpretability of the chosen variables made them highly attractive as policy instruments. The transfer of the allocation scheme to cultural policy is the second stage of its trajectory. Coining the label Königstein Key for the allocation formula of the Königstein Agreement laid the groundwork for further diffusion. There was a growing understanding in the negotiations of the Länder that the key was a functional and simple solution for reaching consensus and effective cooperation. It increasingly bore the trait of signifying procedurally established federal justice. While originating in a niche of the political system, its silent and depoliticized application also enabled it to be transferred to areas outside horizontal federalism, including to rather politicized fields, such as asylum policy. Remarkably, the Königstein Key was legally codified in this policy field and hence became stabilized as a symbol of territorial justice. We identify this normative generalization as the third stage in the trajectory of the Königstein Key. It is precisely because of the enormous breadth of application that explicit references to territorial justice were able to claim validity. For a long time, the Königstein Key was a quiet instrument embedded in political operations, but in 2015, a comparatively exorbitant level of interest and dispute arose, primarily as a result of the increased influx of asylum seekers. We identified this as the fourth stage in the trajectory of the Königstein Key, representing its public critique and restabilization. The Königstein Key has become a symbol in public debates of the fair and appropriate contribution of each territorial unit. Territories have referred to it when justifying their own efforts and have used it as a benchmark to measure the perceived lack of solidarity by others. Since that time, questioning about the adequacy of the key’s use has arisen significantly.

Table 1.

Stages in the emergent formation of the Königstein Key as a key indicator.

Phase	Key events	Description
Administrative creation	Königstein Agreement 1949	Establishment of the key as part of a situational compromise; incorporating the rather intuitive tax and population indicators
Administrative diffusion	Translation to culture 1954,Silent diffusion since 1968	Labelled the “Königstein Key” and proving to be a quasi-natural point of agreement between the Länder
Normative generalization	Translation to asylum policy1978–1982	Notions of justice made explicit; used as a benchmark for new purposes (territorial justice); used for new objects of allocation (subsidies, people)
Justification and critique in public debates	Peak of asylum seekers 2015,Ukraine War 2022	Many new actors are involved; the focus is on measuring efforts and solidarity in refugee reception. Critics become louder, but there is also consolidation

We argue that two mechanisms and predisposing features of the indicator itself have been crucial for the emergent institutionalization of the Königstein Key as a key indicator: first, its repeatedly proven usefulness in administrative routine; second, normative attributions that have accrued in test situations of political and public discourse; and third, predisposing features of the indicator itself matching the political requirements of German federalism. We start our discussion of this emergent process with the latter.

The essential prerequisite for an indicator to be successful is that it fits with contexts based on its formal features. The Königstein Key is an intuitively rather interpretable additive index that was flexible enough to accommodate the diverging interests of heterogeneous actors of German federalism by translating them into an allocation scheme that the participants could consider just and efficient. This is achieved through a formula that translates rather intuitive attributes of states (tax revenue and population numbers) into a more abstract quota. The features of the indicator allowed it to persist through changing circumstances and its transfer to new policy fields while still offering stability and predictability. The stabilization of the Königstein Key as a standard instrument has been related not least to the necessity of obtaining the consent of all parties involved in the negotiations, which creates clear obstacles for alternative suggestions. With every new application, the number of examples of successful cooperation increased and proved it to be a well-tried method. While this could be portrayed as a self-reinforcing process (cf. Mahoney, 2000), the translation to new policy fields was highly contingent and depended on processes of sensemaking. Furthermore, we also highlighted the limitations of its diffusion to new policy fields. The few, but sometimes utterly significant cases in which the Königstein Key was suggested but rejected prove this contingency particularly well.

Furthermore, the Königstein Key has been able to maintain its status as a symbol of territorial justice because the concrete specifications of its implementation have mostly been neglected in public debate. Little attention has been paid to decisions regarding scope or recipients, which Elster regards as room for maneuver that leaves the main symbol of an allocation scheme unchanged while adapting it, in fact, to a situation’s practical requirements. In asylum policy, the Königstein Key is only part of a much more complex multilevel system for the spatial allocation of persons (Schmid, 2024). Reform efforts can be more successful if they do not alter the fundamental symbol of territorial justice and try to integrate these structures. A notable example is the recent MatchIN project (Reinhold et al., 2025), which does not involve changing the Königstein Key at the state level but, within this framework, aims to build a matching mechanism that considers both the preferences of local governments and of asylum seekers. Even though our analysis shows that the territorial focus in public debates on the Königstein Key leaves little space for the voices and rights of asylum seekers, characterizing the Königstein Key as mainly an obstacle to an effective integration policy is one-sided. As the indicator is the result of a long-standing emergent formation process, it has fostered territorial cohesion and administrative efficiency, which are also beneficial for the reception of asylum seekers.

What can applied researchers and policymakers learn from this? We will first deliberate about conclusions for the German context. In the recent decade, there has been a tendency in German politics to demand a stronger link between allocation criteria and the actual problems at hand, such as long-term integration in asylum policy. The logic behind these criticisms is particularly evident in policy papers, which propose well-justified dispersal schemes based on different indicators, most importantly relating to the labor market or housing capacity (Altemeyer-Bartscher et al., 2016; Brücker et al., 2022; Büchsel and Schneider, 2016; Geis and Orth, 2016; Simons and Weiden, 2016). However, their instrumental approach tends to pay less attention to how the consensus of federal states and municipalities can be achieved. While the proposed indicators may be efficient for pursuing policy targets from a scientific point of view, they do not necessarily stand for solidarity and territorial justice, principles that have proven to be crucial for political actors. At the same time, criticism of the Königstein Key is certainly warranted when it is pointed out that it is not an efficient allocation criterion for all policy purposes.

While it is difficult for indicator designers to foresee how political actors will react to their proposals, the features we identified might be useful for enhancing the actual implementation of indicators. Counterproposals to the Königstein Key often share some of these features, like low information costs due to the use of official statistics and sensitivity to temporal changes. In contrast, however, meticulously designed multidimensional indices are often difficult to interpret, especially for nonexperts. Furthermore, whereas the Königstein Key strikes a balance between redistribution among the Länder and its limitation, scientific inventions that are specifically targeted at stronger redistribution do not appeal to rich states. The recent program for compensatory funding of schools in disadvantaged areas demonstrates that establishing such alternatives requires not only the mobilization of elaborate arguments but also of promoters among those affected, in this case, the Länder. Hence, indicator designers might want to consider the trade-off between an accepted, albeit less targeted indicator and one that is more fit for purpose but runs the risk of being rejected or remaining unimplemented.

What travels beyond the case? Indicators at the intersection of science and politics reflect a compromise between scientific rigor and political feasibility. If key indicators bridge multiple policy fields, they are likely to be scientifically less robust because they serve several policy purposes at once, but they are probably more politically practical. This makes them particularly difficult to design and diffuse intentionally. Considering that we still know little about such cases, our generalizations are limited. We expect that the predisposing features of the indicator we identified are necessary, if not sufficient, conditions for the formation of key indicators more generally. When indicators are used in the political sphere, information costs always play a crucial role, because time is limited and there is usually little willingness to commit additional resources. In politics, rigor is not the main priority. Instead, indicators help to gain legitimacy. With a growing tendency to use indicators in politics, information costs are likely to become more of an issue. The abilities of an indicator to facilitate compromise and to tame uncertainty are also likely to be important to political allocation in general. Indicators often formalize compromises and facilitate agreement based on commonsense criteria (e.g. equality) or point systems that integrate diverging interests. Interestingly, allocation formulas that have become a symbol of legitimate compromise can still be flexibly adapted in their implementation details, affecting the net effect of the mechanism (e.g. allocation volume, group of beneficiaries). For allocation formulas to work in the long term, they have to deal with the uncertainty of changing circumstances over time, as well as the need of participants to plan accordingly. If an indicator manages to reconcile stability with dynamics, it is likely to become more enduring. The variables that enable such a balance, however, depend on the specific context of application.

For a compromise formula to be considered a legitimate symbol of justice, institutional work in terms of justification is of fundamental importance. Indicators cannot ignore the claims that dominate a given social field if they are to have a realistic chance of practical implementation. The extent to which such justifications take place is very much contingent upon the agency of participants and not generalizable. The requirement to find a formalized compromise of territorial justice is probably less salient in unitary states (or other rather hierarchical settings) than in multilevel systems. The autonomy of subnational governments in unitary states is rather limited, which means that their ability to politicize questions of territorial justice is more restricted as well. Comparisons of reception policies for asylum seekers, for instance, show that central governments in unitary states more often act unilaterally in the erection of reception facilities, bypassing local governments or reaching a compromise in a nonformalized and nontransparent way (Bartl and Lutz, 2025).

The mechanisms of emergent formation of key indicators have received little attention in debates about the use of indicators as policy tools (cf. Lehtonen, 2017). Close to our case, the federal allocation schemes in the United States, as described in “Allocation formulas by policy design”, were purposefully designed and their significance remained within the limits of the policy fields they were intended for. Our analysis suggests that this relatively narrow use might be attributable to the fact that these formulas were implemented with an explicit concept from the outset, whereas the emergent process we analyzed started with administrative practice and acquired a conceptual background only afterward. The gross domestic product, a key indicator par excellence, has arguably received a lot of scholarly attention (Lepenies, 2016). However, the fact that its diffusion into various policy fields was not intended has hardly been fleshed out to make a more general argument about emergent processes of key indicator formation. We tried to make such an argument by pointing out the predisposing features and two mechanisms operating during the emergent process of the key indicator formation that we analyzed. While we compared the Königstein Key with directly competing indicators, future studies could try to conduct systematic comparisons with more dissimilar cases.