On the evening of October 6, 2008, Iceland’s prime minister addressed the nation, ending with the unthinkable phrase in Icelandic political discourse: “Guð blessi Ísland” (God bless Iceland).Footnote 1 In the following days, Iceland’s three largest banks collapsed, bringing down with them 97 percent of the financial system’s assets.Footnote 2 The state intervened to salvage domestic operations while leaving foreign liabilities behind.
Some observers described the collapse as a failure of deregulation and weak supervision,Footnote 3 while others pointed to overconfidence and the transmission of the shock through global markets.Footnote 4 This article departs from both accounts, and from histories of Icelandic banking that treat the crisis chiefly as a failure of formal institutions.Footnote 5 The collapse was the culmination of a longer trajectory in which financial liberalization reconfigured, rather than replaced, existing networks of elite power. The central question is how Iceland’s banking system evolved from a state-led system into a liberalized financial sector that remained deeply politicized, and why that combination produced such persistent fragility. I argue that political capitalism, driven by elite networks, repeatedly turned institutional reform into control until the 2020s, when reprivatization encountered its first sustained external check.
I use political capitalism to describe a system in which political and economic elites cooperate for mutual benefit so that regulation, appointments, credit allocation, and state authority reinforce one another rather than operate through impersonal market rules alone.Footnote 6 The term is preferred here to crony capitalism,Footnote 7 which implies the corruption of an otherwise impersonal market order. Political capitalism captures an arrangement in which elite coordination is constitutive of the order itself. Elite networks refer to the small set of political and business actors who recur across board appointments, ownership vehicles, and subsequent private-sector careers. The case is valuable not because it is unique but rather because it exposes dynamics less easily observed in larger economies: how fragility emerges when formal reform leaves elite networks intact.
Analytically, I treat Icelandic banking as a field in which authority over credit and bank governance emerges from the interplay of institutions, networks, and cognitive frames. Following Beckert, I focus on three elements that condition one another over time: formal rules and organizations, durable ties among political and economic insiders, and legitimating understandings of what banks are for.Footnote 8 To specify how the institutional dimension of this triad changes, I draw on Mahoney and Thelen’s account of gradual transformation: layering, drift, conversion, and, where relevant, displacement describe the modes through which ownership regimes, supervisory mandates, and governance rules change without a clean rupture with prior settlements.Footnote 9 Beckert supplies the anatomy of the field and keeps networks and legitimating ideas central; Mahoney and Thelen supply the mechanism of change within the institutional dimension. They are preferred to path-dependence accounts, which locate change at critical junctures, and to capture theory, which presupposes the regulator-regulated separation Iceland never had. Together, they help show not merely how the rules changed, but how the same configurations of people and justifications persisted.
Iceland’s small size is part of the explanation, not merely background context. In a society of fewer than 400,000, the same individuals recur across institutions that are more separated in larger economies, making elite persistence easier, regulatory distance harder, and informal coordination more likely. Iceland is an extreme case in Flyvbjerg’s sense: atypical intensity makes latent dynamics easier to observe.Footnote 10
The analysis draws on Icelandic-language primary sources: parliamentary records, the report of the Special Investigation Commission (SIC) founded in 2008, the National Audit Office’s review of the 2022 bank sale, banks’ annual reports and governance records, and contemporary press. The SIC report is read critically rather than taken at face value, but it remains an unusually authoritative source: a statutory commission with powers to compel testimony, whose findings Iceland’s political spectrum accepted, and whose account can be triangulated against parliamentary records, court judgments, and bank disclosures. Reading these sources in the original is aided by having lived in Iceland for much of the post-2000 period.
The Formation of Iceland’s State Banking System (1885–1986)
Iceland had under 100,000 people until the late nineteenth century and remains under 400,000 today, with an economy limited to fishing and agriculture, fish exports generating most foreign income, and output a fraction of metropolitan Denmark’s.Footnote 11 A Danish territory since 1380, it obtained home ruleFootnote 12 in 1874, expanded it in 1904, and became independent in 1944. Coalition government has been the norm; the conservative Independence Party has been mainly dominant and typically allied with the agrarian Progressive Party.Footnote 13 Since no party could govern alone, banking and politics evolved through coalition negotiation rather than arm’s-length institutional design, and the same individuals and families recurred across public office, finance, and politics.
Until the late nineteenth century, Iceland had no formal banking institutions. Credit was informal, provided by wealthy merchants using goods as collateral.Footnote 14 Savings could be remitted to Denmark via state bonds, but Iceland lacked a system to mobilize capital or allocate credit. Elites already held political power: from the early nineteenth century, the Stephensen-Finsen family controlled the colonial administration and the Church, connected through shared education and control of the country’s only printing press.Footnote 15 The founding of Landsbanki in 1885, which opened the following year, provided the elite with a new institutional lever rather than creating a new elite.Footnote 16
Landsbanki, established by AlþingiFootnote 17 legislation, was Iceland’s first national bank, issuing banknotes and lending to merchants and farmers. For decades it had no real domestic rival of similar size, which made access to formal credit synonymous with development. Lacking sovereign monetary authority, it nonetheless acted as a quasi-central bank, limiting note issuance to align with gold-backed Danish notes and so maintaining stability within a colonial monetary order,Footnote 18 a means for the state to demonstrate economic control and legitimacy.
Landsbanki was born a political entity: its charter depicted it as a driver of national development, but ownership and control remained with the Ministry of Finance. By 1904, after Home Rule, the bank was embedded within the state’s distributive mechanism, linking credit to political priorities.Footnote 19 Intervention was overt. Ministers influenced the bank to direct credit to favored groups like rural cooperatives and merchant families aligned with the government.Footnote 20 Progressive Party-linked fishing and agricultural cooperatives were regularly refinanced despite poor repayment records, a pattern criticized in opposition media and parliamentary debate.Footnote 21 Access to capital depended as much on political loyalty as on financial merit. This reflects Lamoreaux’s description of insider lending in early U.S. banking, where lending to directors and connected borrowers was common in banks run as private ventures.Footnote 22 The parallel is limited in a way that sharpens the Icelandic case. In New England, insider lending was disciplined because the lenders’ own capital stood behind it.Footnote 23 Iceland’s lending banks were state-owned, so losses fell on the public balance sheet rather than on the insiders who authorized them. Insider allocation here was not an early stage of banking disciplined by risk but an enduring political settlement.
A rival private institution, Íslandsbanki, opened in 1904 with Danish capital; despite foreign ownership, its governance preserved Icelandic political oversight.Footnote 24 The two competed unevenly, Landsbanki retaining its privileges, until Íslandsbanki collapsed in 1930 under speculative foreign borrowing.Footnote 25 The Alþingi declined intervention, with critics asking why Icelanders should absorb the debts of foreign financiers.Footnote 26 This was not a refusal to act but a refusal to absorb foreign shareholders’ losses: the state moved readily once the institution was reconstituted on domestic terms as the Fisheries Bank of Iceland (Útvegsbanki Íslands) in April 1930.Footnote 27
The same year, the Alþingi created Búnaðarbanki (Agricultural Bank), a Progressive Party vehicle for rural credit and rural political support. Each new institution reproduced Landsbanki’s governance model of party-nominated boards and three governors.Footnote 28 The sectoral bodies that followed, agricultural credit funds, housing loans, and regional development banks, multiplied patronage opportunities without improving access or accountability. Governance stayed opaque, oversight weak, and credit under the control of whichever party held office; opposition parties inherited the system and perpetuated it.
Between the wars, banking became an extension of parliamentary power. Landsbanki staffed its board and regional branches with party-aligned lawyers, officials, and civil servants whose loyalty mattered as much as their financial expertise.Footnote 29 According to Grímsson, bank managers and boards were “in practice party appointed”Footnote 30 throughout the 1930s and postwar decades. The same individuals served as MPs, ministers, bank managers, and senior judges, reinforcing the link between credit and partisan obligation.Footnote 31 The Independence Party, formed in 1929, became the core of elite cohesion. The Thors family was the party’s main power center for decades, succeeded in the 1960s by the Benediktsson brothers: Bjarni, later prime minister; Sveinn, who ran the family’s business arm; and Pétur, who directed Landsbanki from 1956 to 1969 and married Marta Ólafsdóttir Thors.Footnote 32
The overlap was institutionalized, not merely informal: as late as the 1960s, five of the sixty members of the Alþingi simultaneously managed the country’s major banks.Footnote 33 Iðnaðarbankinn, the Bank of Industry (1953), and Verslunarbankinn, the Bank of Commerce (1961), added complexity without reducing political discretion.Footnote 34 The Ministry of Finance and the sectoral ministries kept de facto control over loan approvals, and ministers consistently bypassed formal criteria. Attempts to professionalize governance through credit scoring, external audit, or independent oversight met resistance or were quietly circumvented.Footnote 35 Board appointments rewarded partisan service, and governments channeled fiscal discretion through the banks rather than through direct spending.Footnote 36 Participants treated this allocation of credit and jobs not as corruption but as ordinary political and commercial practice.Footnote 37
The 1957 reform set the template, separating central-banking functions from Landsbanki’s commercial business; the Central Bank became a fully independent institution in 1961.Footnote 38 The remaining ties were severed, and the Central Bank took charge of foreign exchange and financial supervision. Each of its three governors informally represented one of the main parties, and the Alþingi continued to negotiate appointments to commercial bank boards.Footnote 39 The episode is conversion: a formal modernization, a supposedly independent Central Bank, redeployed to sustain political networks rather than replace them, and with it the political capitalism in which elites rather than impersonal rules mediate access to finance.Footnote 40
By the 1970s, the arrangement was fully institutionalized: banks selected leaders for political ties,Footnote 41 and oversight remained limited and lending criteria weak, justified by elite norms and a shared sense of fairness in which partisan appointment counted as proper governance rather than interference.Footnote 42
Útvegsbanki’s 1985 collapse exposed the model’s fragility. Hafskip, a major Icelandic shipping company, had borrowed from the bank at levels that in some years exceeded the bank’s equity. When Hafskip failed, Útvegsbanki followed.Footnote 43 The entanglements ran through the Independence Party rather than through family or friendship. Hafskip’s chief executive, Björgólfur Guðmundsson, had organized the party’s youth wing in the 1960s before entering business; his co-manager, Ragnar Kristjánsson, was a former party executive; and Albert Guðmundsson (no relation, despite the shared patronym),Footnote 44 then minister of industry and a former finance minister, chaired both Hafskip and Útvegsbanki. He resigned after the disclosure of undeclared payments from Hafskip. The circuit closed: Björgólfur’s company paid the minister who oversaw the bank lending to it.Footnote 45 Parliamentary records and the official investigation report (þskj. 153) document the scale of the exposures and the case’s political sensitivity;Footnote 46 the proceedings ended in a suspended sentence for Björgólfur.Footnote 47 The episode exposed the fragility without dislodging the logic of elite coordination that produced it, a logic inseparable from what critics called the Independence Party’s octopus Footnote 48: a network of interlocking families and business interests reaching across banking, politics, and media.Footnote 49
In Alþingi debates, Albert Guðmundsson repeatedly invoked formal institutional boundaries, in particular the separation between bank council oversight and managerial lending decisions, to deflect responsibility.Footnote 50 In 1986, he also described his own appointment to Útvegsbanki’s bank council as a product of coalition bargaining rather than technocratic selection. The boundary was real enough to serve as a defense, while the network that had placed him there sustained precisely the discretion the boundary was supposed to prevent. Rule, network, and legitimating story were not separate forces.
By the mid-1980s, the system had reached its limits: inflation persisted, defaults rose, and the banking sector increasingly appeared to obstruct rather than enable modernization.Footnote 51 Iceland’s banking history did not move through a single, linear mode of change. It began with institutional creation, as Landsbanki was grafted onto an existing elite network rather than established against it. It shifted through layering, as new banks and credit bodies accumulated without dislodging the logic of political control. The 1957 separation of central banking functions was a case of conversion: formal modernization that preserved the same personnel and the same distributive logic under a different organizational form. Across these phases, the change agents were what Mahoney and Thelen term symbionts, party-appointed managers, board members, and governors whose careers depended on survival of the arrangements they administered, and who exploited institutional discretion while defending the institutions themselves.Footnote 52 Iceland’s early banking system was not a neutral financial architecture later captured by politics. It was political, and later reforms preserved that settlement.
Liberalization and Privatization (1986–2003)
By the mid-1980s, Iceland’s banking system was outdated, inefficient, and politically compromised. Interest rates were set administratively, credit allocated at the minister’s discretion, and bank boards populated through parliamentary negotiation. Macroeconomic pressure, domestic reform agendas, and international liberalization made that arrangement unsustainable.
Liberalization is often an elite project rather than a solution to elite power: those who benefit from lifting restrictions tend to be those bound by them. The relevant question is not why Icelandic liberalization failed to dissolve old networks, but who drove it, what constraints it removed, and what tools it gave them. Posed that way, the Icelandic sequence is exemplary rather than anomalous: a liberalization designed and executed by substantially the same elite it ostensibly disciplined.
The first significant step came in 1986 when the Act on Interest and Indexation (No. 38/1986) removed government control over interest rates, followed by the gradual dismantling of sector-specific lending programs, especially in agriculture and housing.Footnote 53 The Independence Party, then in coalition, drove the agenda and deepened it under Davíð Oddsson’s premiership after 1991.
The network that drove liberalization had formed a decade earlier. In 1972, young Independence Party men, mostly law students or recent graduates of the University of Iceland, took over Eimreiðin and ran it until 1975 as a market-liberalism platform; by one member’s own account, the circle continued to meet regularly after the magazine folded. It included Davíð Oddsson, Þorsteinn Pálsson, Geir H. Haarde, Kjartan Gunnarsson, Jón Steinar Gunnlaugsson, Brynjólfur Bjarnason, Baldur Guðlaugsson, and Hannes Hólmsteinn Gissurarson. Oddsson, Pálsson, and Haarde became party leaders and prime ministers; Kjartan Gunnarsson ran the party apparatus for twenty-five years while sitting on Landsbanki’s board; Jón Steinar Gunnlaugsson joined the Supreme Court; and Baldur Guðlaugsson became permanent secretary of the Finance Ministry.Footnote 54 In a country without a Tokyo University or grandes écoles,Footnote 55 one law faculty, one party youth group, and one magazine shaped the elite.
Accession to the European Economic Area (EEA) in 1994 required the transposition of EU directives on capital flows, competition, and financial supervision,Footnote 56 and the regime incorporated international banking standards, including the 1988 Basel I Accord and its 8 percent capital adequacy requirement.Footnote 57 These reforms did not displace incumbent elites but equipped them with instruments to scale activities and reallocate risks through cross-border lending and wholesale funding, while maintaining dense ties to political actors at home.
The Financial Supervisory Authority (Fjármálaeftirlitið, FME), established in 1999, was severely under-resourced: a founding budget of 265 million ISK (about USD 3.4 million) and fewer than twenty employees, rising to only forty-seven by 2007, with an average tenure of four years,Footnote 58 which left junior staff facing the country’s most prestigious employers poorly positioned to resist what Kwak terms cultural capture.Footnote 59 Despite formal autonomy, the agency depended on government funding, ministers appointed its directors, and it struggled to assert oversight over an increasingly assertive banking sector.Footnote 60 Conti-Brown and Vanatta’s account of American supervision before 1933 identifies the mechanism: an information regime rather than an enforcement regime, where supervisors could criticize but not discipline, and whose credibility drained away because forbearance was the only option. Supervisory legitimacy requires both technical judgment and the political authorization to act on it. The FME had neither.Footnote 61
Through the first decade of liberalization, banks still ran on the old logic. Three bankastjórar headed Landsbanki simultaneously, directorships distributed by convention with an eye to the parties, and the Alþingi elected its supervisory board.Footnote 62 All three, Sverrir Hermannsson, Björgvin Vilmundarson, and Halldór Guðbjarnason, resigned on a single day in April 1998 after a National Audit Office inquiry into the bank’s travel and entertainment expenses (Table 1).Footnote 63 The pipeline between parliament and the bank directorate ran in both directions.
Governance continuity across Iceland’s banking phases: selected trajectories
Table 1. Long description
The table consists of six columns: Individual, Earlier position, Later position, Continuity across, Mode, and Agent type.
* Davíð Oddsson: From Prime Minister (1991–2004) to Central Bank governor (2005–09) and editor. Continuity from liberalization to aftermath via Conversion as a Symbiont.
* Kjartan Gunnarsson: From Independence Party secretary-general and Landsbanki board (1991) to shareholder via privatization. Continuity from state to private ownership via Conversion as a Symbiont.
* Sverrir Hermannsson: From Minister to Landsbanki bankastjóri (1988) to founding a new party in 1998. Continuity from state ownership to incorporation via Layering as a Symbiont.
* Friðrik Sophusson: From Finance Minister to Chairman of Íslandsbanki (2010). Continuity from liberalization to post-crisis governance via Layering as a Symbiont.
* Björgólfur Guðmundsson: From Hafskip C E O to principal of Samson Holding and Landsbanki chairman (2003). Continuity from pre-liberalization to boom via Displacement of ownership and conversion of criteria as an Opportunist.
* Ólafur Ólafsson: From S-group principal to second-largest owner of Kaupthing. Continuity from privatization to boom via Conversion as an Opportunist.
* Sigurjón Þ. Árnason: From Búnaðarbanki head of operations to joint C E O of privatized Landsbanki and later conviction (2015). Continuity from state ownership to prosecution via Conversion as an Opportunist.
* Sigríður Elín Sigfúsdóttir: From Landsbanki director (2003) to C E O of new Landsbanki (2008) and later conviction. Continuity from boom to re-nationalization via Displacement in form as a Symbiont.
* Gunnar Þ. Andersen: From Landsbanki division head to Director-general of the F M E (2009). Continuity from boom to post-crisis supervision via Drift as a Symbiont.
* Þorsteinn Már Baldvinsson: From Glitnir chairman at collapse (2008) to purchasing Íslandsbanki shares in 2022. Continuity from collapse to reprivatization via Displacement of ownership as an Opportunist.
* Benedikt Sveinsson: From principal shareholder of Íslandsbanki-Glitnir to allocation of Íslandsbanki shares in 2022. Continuity from formation to reprivatization via Displacement of ownership as an Opportunist.
Notes: Individuals are listed in approximate chronological order of the trajectories recorded, and are drawn from the sources below, informed by a prosopography of 216 board, executive, and management individuals between 1998 and 2025. The modes follow Mahoney and Thelen’s typology of gradual institutional change; agent types follow their classification of change agents. Assignments summarize the analysis developed in the corresponding sections of the text rather than constituting an independent coding.
Source: Report of the Special Investigation Commission; Alþingi records; National Audit Office review of the March 2022 Íslandsbanki sale; bank annual reports; court judgments.
The most significant changes were in ownership, not regulation. In 1997, both major state banks became limited companies, effective from January 1998. Successive public offerings between 1998 and 2002 reduced the state’s holding in both banks. On December 31, 2002, the government agreed to sell its 45.8 percent stake in Landsbanki to Samson Holding. In January 2003, a matching stake in Búnaðarbanki went to the S-group.Footnote 64
Political entanglement persisted through board continuity. After taking office in 1991, Prime Minister Davíð Oddsson appointed his close associate and the Independence Party’s chief, Kjartan Gunnarsson, to Landsbanki’s board; he stayed through privatization, became a shareholder via a holding company, and kept his directorship.Footnote 65 Elsewhere, the new boards drew from entrepreneurial and political circles rather than from banking, under qualification and independence requirements that the FME applied loosely.
Twice the government sought a foreign owner, and twice it retreated. In 1998, the Ministry of Commerce under Finnur Ingólfsson explored selling a substantial stake in the newly incorporated Landsbanki to the Swedish bank SEB,Footnote 66 alongside a proposed merger of Íslandsbanki with Búnaðarbanki; the talks collapsed amid political resistance to selling “one of the national jewels.”Footnote 67 The objective was an experienced foreign cornerstone investor (kjölfestufjárfestir), but despite approaches to over twenty foreign investors between late 2001 and early 2002, none would acquire a controlling stake on terms acceptable to the Icelandic authorities.Footnote 68 Shares went instead to newly formed domestic investment groups with strong personal and political ties to the ruling parties, several of whose principals had little or no experience in finance. The process resumed in 2002 when the domestic Samson Group expressed interest: Björgólfur Guðmundsson, his son Björgólfur Thor Björgólfsson, and Magnús Þorsteinsson, whose wealth came from selling the Russian brewery Bravo to Heineken earlier that year.Footnote 69 This was the same Björgólfur Guðmundsson who, as chief executive of Hafskip, had received a suspended sentence after the 1985 collapse that brought down Útvegsbanki. Seventeen years later, he returned as prospective owner of one of Iceland’s largest state banks, his earlier downfall recast as entrepreneurial success rather than financial failure.
Björgólfur Guðmundsson testified that the group contacted the privatization committee chaired by Ólafur Davíðsson before submitting its proposal, and met with Prime Minister Davíð Oddsson, a meeting Oddsson confirmed before the Commission. Early government memos noted the group lacked banking experience, which conflicted with the goal of recruiting an experienced banking owner. When the final evaluation framework was developed, however, HSBC assessed bidders not on banking expertise but on their “knowledge and experience of financial markets,” and the accompanying guidance made clear that this category also encompassed “the likely ability of the investor to provide a network of useful contacts within the financial markets.”Footnote 70 Direct banking experience became just one means to legitimacy; success in international business, financial reputation, and influential networks also met the standard.
The two sales did not only favor specific groups; they reflected the coalition bargain known as helmingaskipti, the division into halves by which the Independence and Progressive Parties had shared appointments and economic gains since the mid-century. Samson, backed by offshore wealth and linked to Independence Party biographies, received Landsbanki; the S-group received Búnaðarbanki, a consortium of four entities tied to the Progressive Party: Egla hf., VÍS, the Cooperative Pension Fund, and the Cooperative Insurance Holding Company. Steingrímur Ari Arason, a finance ministry representative since 1991, resigned after the Samson decision and testified that party leaders, not the committee, had made the choices. The Commission found no documents to confirm this, and he admitted he had none. The commerce minister, Valgerður Sverrisdóttir, testified that the Landsbanki decision originated with the prime minister. A declining cooperative bloc converted its assets into bank ownership while a rising network of new money entered through party channels: liberalization redistributed influence between these networks rather than eliminating them.Footnote 71
Where the Landsbanki sale turned on a loosened banking-experience criterion, the Búnaðarbanki sale emphasized foreign participation. Originally, Société Générale was expected as the foreign partner, but Hauck & Aufhäuser, a small German private bank, took that role, presented as a shareholder in Egla.Footnote 72 Doubts about the German bank’s involvement arose quickly, with Sigurjón Þ. Árnason, then head of Búnaðarbanki’s operations and soon joint chief executive of the privatized Landsbanki, voicing concerns.Footnote 73 Fourteen years later, a parliamentary investigation would establish that those doubts had been well founded.
Selling banks is inherently difficult and requires no bad faith. Issuers seek proceeds; buyers need capital; the wealthy and institutions have more of it. The government twice searched for alternatives, in the 1998 SEB talks and the 2001–2002 foreign investor search, both of which failed on acceptable terms. The concern is not that domestic groups owned the banks but deviations in process: banking expertise was reinterpreted mid-process until political and commercial networks satisfied it; the evaluation model was co-produced with the seller’s own staff; one bank went to each coalition party’s network, attested by the finance ministry’s committee representative and partially corroborated by the responsible minister; and a ministerial committee kept no minutes.Footnote 74 Distribution to the wealthy is what offerings do. Rules that bend around specific buyers are what capture looks like.
Ownership shifted, while influence networks stayed. Bank boards still included individuals from political law firms, ministries, and investment groups, with informal influence over appointments through the Ministry of Finance.Footnote 75 Credit decisions, now justified by market risk language, still relied on trust and perceived fit: the new banking language allowed old networks to maintain access under a modern appearance.Footnote 76
Liberalization did not break with the older order so much as rework it. The reforms of the late 1980s and 1990s layered market mechanisms onto an institutional environment still shaped by party influence, weak supervision, and dense personal ties. Supervisory capacity drifted as the banks expanded. Privatization brought a more decisive institutional change in ownership, but the criteria used to allocate the banks were converted during the process itself: banking competence was broadened to include financial reputation and access to useful networks, allowing politically connected buyers to satisfy standards that were initially framed more narrowly. Hence, privatization displaced state ownership without displacing the allocation logic that had long governed Icelandic banking. Where the century of formation rested on symbionts, opportunists seized privatization, exploiting the ambiguity of the criteria to convert market reform into elite recomposition. The outcome was not the end of political capitalism, but reorganization under market language.
The Rise and Collapse of the Privatized Banks (2003–2008)
By 2003, privatization was complete, and unprecedented financial expansion followed. Within a few years, Kaupthing (which merged with Búnaðarbanki that year), Landsbanki, and Glitnir (Íslandsbanki, renamed in 2006) grew from small domestic lenders into aggressive international institutions.Footnote 77 They acquired subsidiaries across Europe, raised billions on foreign markets, and fueled a domestic credit boom that reached into nearly every part of Iceland’s economy.Footnote 78 To many observers, Iceland was a success story: a small, agile, globally competitive nation transformed by liberalization.Footnote 79
Beneath the surface, the system was structurally fragile. The privatized banks carried forward the informal traits of their state-owned predecessors, with continuity in leadership recreating pre-privatization networks inside new private banks.Footnote 80 Personal and political channels filled board seats rather than financial expertise. In 2003, the state’s representatives withdrew from Landsbanki’s board, replaced by Björgólfur Guðmundsson, who became chairman; he and his son were the bank’s largest shareholders. Only Kjartan Gunnarsson of the Independence Party kept his seat.Footnote 81 Despite these conflicts, the FME did not intervene to block the appointments.Footnote 82
Influence shifted from ministerial discretion to private ownership and owner-led governance. Kaupthing’s case is notable: Exista, controlled by Bakkavör founders Lýður and Ágúst Guðmundsson, was the bank’s largest shareholder and, with connected parties, its second-largest credit exposure. At collapse, it owed the bank over ISK 200 billion, routinely collateralized by Kaupthing’s own shares. Its largest exposure was Robert Tchenguiz, a shareholder and board member of Exista. Ólafur Ólafsson, the second-largest owner, borrowed via Kjalar. In 2008, Kaupthing repaid Kjalar’s debt to Citibank and took its shares, held by Egla Invest B.V., as collateral, directly financing its second-largest owner’s stake.Footnote 83 Egla was not a new name: it had served as the vehicle through which Ólafsson’s S-group acquired Búnaðarbanki in the 2003 privatization. The continuity is evident in the organizational charts. Among the twelve directors overseeing Landsbanki under Samson’s ownership in 2003, five (Brynjólfur Helgason, Haukur Þór Haraldsson, Stefán H. Stefánsson, Sigurjón G. Geirsson, and Tryggvi Tryggvason) had held management roles at the state-owned bank in 1998. The two most senior newcomers, co-CEO Sigurjón Þ. Árnason and HR head Atli Atlason, transferred directly from Búnaðarbanki.Footnote 84 The state-banking complex did not staff the privatized bank by accident; it was the only pool there was.
Icelandic banks exploited international markets and limited domestic oversight for rapid growth, issuing Eurobonds, launching foreign deposit platforms, and expanding quickly. Kaupthing acquired Singer & Friedlander in the UK; Landsbanki launched Icesave, drawing billions in foreign deposits.Footnote 85 Between 2003 and 2007, the three banks’ combined assets grew from 100 percent of Iceland’s GDP to over 900 percent.Footnote 86 Iceland accelerated the post-1980 financialization of the economy further than any comparable wealthy country, and its collapse formed part of the 2008 transatlantic banking crisis, in which reliance on short-term dollar funding revealed a major fault line.Footnote 87 Iceland’s distinction lay in the sheer scale of the banks relative to the state and the absence of a lender of last resort in their borrowed currencies. The direction of the traffic is itself evidence. Iceland’s financial elite internationalized by acquisition, buying Heritable Bank in London, FIH in Copenhagen, Singer & Friedlander, a private bank in Luxembourg, rather than by career integration into global institutions. The cohort that ran the banks had formed almost entirely at home, and faced no competing professional frame and no external reputational discipline.
Banks acted like investment banks, using their balance sheets to underwrite leveraged acquisitions and finance equity stakes tied to their networks. Major shareholders and affiliates were often among the biggest borrowers, merging ownership, governance, and credit risk into a relationship-based system. The SIC shows exposures to controlling groups reaching high levels relative to bank capital: Landsbanki’s lending to Björgólfur Guðmundsson’s group peaked over €900 million, about 30–35 percent of its capital, while exposures to Björgólfur Thor Björgólfsson and related entities exceeded €1 billion by 2008.Footnote 88 Connected lending and cross-holdings embedded concentrated counterparty risk within elite networks, a risk largely uncontested amid abundant liquidity.
Domestically, the credit boom inflated asset prices, private consumption, and speculative optimism. Politicians and business leaders praised banks as national champions, and media celebrated Iceland’s Viking bankers, entrepreneurs conquering Europe. This confidence faced no counterweight in collective memory: Iceland had seen no major banking crisis between 1930 and the 2000s. In a society with no memory of financial collapse, politicians and commentators dismissed concerns about systemic risk as pessimism or as unpatriotic.Footnote 89
Behind the euphoria, institutional safeguards lagged. Supervisory staff increased from forty-seven in 2007 to sixty-two in 2008 while balance sheets grew 30–50 percent yearly, and core experience remained only three to four years. Capacity limits forced triage, making routine inspections difficult.Footnote 90 Its leadership’s political profile complicated the Central Bank’s crisis role: Davíð Oddsson, the prime minister who had overseen privatization, became a governor in 2005 and so supervised the banks his government had sold. The bank communicated poorly with ministers, providing neither written assessments nor response proposals even when conveying serious information.Footnote 91
Oversight failures were both structural and material. Regulators were embedded in the same elite network that controlled the banks. Many staff leaving the FME moved to supervised firms, including the commercial banks, blurring the boundary between regulator and regulated and fostering deference.Footnote 92 Supervisory institutions were modernized, yet enforcement remained uneven: board appointments went unchallenged, insider lending persisted, and disclosure rules varied. Criticism of banking was muted, and skepticism risked reputation. Confidence blended with complacency as elite commonsense came to be perceived as expertise, legitimizing risky decisions and limiting institutional and public challenge.
After privatization, banks introduced bonuses, shifting compensation toward pay-for-performance and boosting incentives for balance-sheet expansion. By 2007, bonuses in key units often made up over 80 percent of total pay. Middle managers received large bonuses, sometimes funded by loans against shares, linking personal wealth, stock prices, and risk. The SIC found no clear or transparent framework for incentive pay.Footnote 93 In some cases, banks paid salaries through foreign branches their own auditors did not recognize, showing how informal workarounds persisted inside modern institutions.Footnote 94
Banking culture grew increasingly extravagant, and rising executive pay and conspicuous lifestyles were part of the growth dynamic rather than incidental excess.Footnote 95 By 2007 at Kaupthing, top earners took half of wages, with the CEO earning nearly 159 times the average; the highest-paid executive earned around ISK 2.5 billion (USD 36 million) between 2004 and 2008.Footnote 96 By global banking standards, this pay was typical rather than exceptional: Goldman Sachs paid its CEO USD 68.5 million in 2007, and CEO-to-worker ratios at US firms reached about 246 to 1 that year.Footnote 97 What was exceptional was the pay’s relation to society: a 159-to-one ratio arrived within a decade in a small economy with compressed pay norms, and its form: loans secured on employer shares that tied managers’ solvency to the bank’s stock. Icelandic bankers copied the global model in a country the size of a European city.
In 2007–2008, when global liquidity dried up, Iceland’s banks, reliant on short-term wholesale funding and exposed to foreign liabilities, were among the first to fail. They could not withstand the interbank market freeze, and by October 2008 all three collapsed within days. The government took control of domestic operations and invoked emergency powers but left foreign liabilities outside the rescue. GDP fell about 10 percent in two years.Footnote 98
The Special Investigation Commission (SIC), established in the wake of the collapse, concluded that the crisis reflected systemic weaknesses in governance. Managerial and oversight practices were repeatedly inadequate across the major financial institutions, and supervisory follow-through remained limited even as risks concentrated.Footnote 99 Bank directors failed to understand the scale and interdependence of risks, regulators could not or would not impose constraints, and ministers promoted liberalizations without sufficient oversight capacity.
The boom years introduced only limited institutional change, and both changes reinforced the settlement created by privatization. The first was drift: the 1999 supervisory framework remained formally intact but did not keep pace, and the actors positioned to veto the erosion were embedded with those who benefited from it. The second was conversion: performance pay, adopted as a modern corporate device, was redeployed to bind managers to the share price and to loans secured on their own equity. The visible signs of the boom, foreign acquisitions, Icesave, and connected lending, were therefore not new institutional forms but intensified uses of an already permissive order: the opportunists who had acquired the banks in 2003 now used them. A discourse of national exceptionalism, reinforced by the absence of living memory of crisis, transformed balance-sheet growth into a sign of success and narrowed the space for dissent. The collapse of October 2008 then marked the limit of gradual change: not another mode, but the exogenous rupture that Mahoney and Thelen’s framework brackets by design. Its modes explain how Iceland built a fragile system across a century, not the days in which that system broke.
Re-Nationalization and the Politics of Crisis Management (2008–2010)
The collapse of Iceland’s banking system in October 2008 was the country’s most severe crisis. Within a week, its three major banks became insolvent, with liabilities nearly nine times GDP and no external lender of last resort, forcing the state to respond quickly.Footnote 100 While the crisis created a chance for reform, it mainly facilitated elite adaptation.
The immediate trigger was Glitnir’s impending default. In late September, the bank’s management sought emergency support from both European and domestic actors. On September 25, 2008, Glitnir’s chair requested an emergency collateralized loan of €600 million from the Central Bank of Iceland. Instead of extending liquidity support, the Central Bank and government partially nationalized the bank on September 29, announcing a plan to inject €600 million for a 75 percent stake, a decision made with little parliamentary consultation and under tight time constraints.Footnote 101 Key officials sometimes learned of decisive crisis-management steps through media rather than through formal institutional channels.Footnote 102 Crisis management proceeded through rapid executive action, with parliamentary scrutiny, rule-bound supervision, and arm’s-length governance bypassed at the moment of decision. A proposal to merge Glitnir, Landsbanki, and Straumur with government capital was considered and abandoned; by October 4 the Central Bank warned that failure was imminent.
Central Bank Governor Davíð Oddsson held high-level talks with the prime minister and officials. Ministerial responsibility was uneven. Minister of Commerce Björgvin G. Sigurðsson, in charge of banking, was often excluded from key discussions.Footnote 103 He criticized not being involved in the Glitnir matter until Sunday evening. SIC documents restricted inner-circle meetings, described as a “super-minister group” excluding him.Footnote 104 Earlier in 2008, he was neither invited to nor informed about meetings between the prime minister, Central Bank governors, and bank executives.Footnote 105 Informality was not Iceland’s anomaly; the global crisis was managed worldwide through improvisation, weekend rescues, and personal networks, from Lehman to swap lines.Footnote 106 Reykjavík in October 2008 resembled New York and London. Two aspects distinguished Iceland. First, the informal crisis network and the rescued network were largely the same people, formed at the same sites: the Central Bank governor negotiating the banks’ fate was the former prime minister whose government had chosen their owners. Second, Iceland improvised without the resource that made other improvisations survivable: a lender of last resort in the currencies the banks borrowed.
On October 6, Alþingi passed the Emergency Act (Act No. 125/2008), granting the FME sweeping powers to intervene in financial institutions, override shareholder rights, and prioritize domestic depositors over foreign bondholders.Footnote 107 That evening, the God bless Iceland TV address signaled a failure of political language amid systemic collapse. Within hours, Landsbanki’s London branch closed.Footnote 108 On October 7, the FME took control of both Landsbanki and Glitnir, dismissing their boards and appointing resolution committees to oversee the banks.Footnote 109 The episode was a legal rupture but executed through concentrated executive discretion and informal coordination at the top, not through routine procedures.
The following day, UK authorities froze Landsbanki’s assets under the Anti-Terrorism, Crime and Security Act 2001 (the Landsbanki Freezing Order 2008).Footnote 110 Later that day, the UK Treasury seized control of Kaupthing Singer & Friedlander.Footnote 111 The FME followed suit domestically, taking control of Kaupthing and appointing a resolution committee on October 9.Footnote 112 With the failure of all three major banks, the Icelandic financial system collapsed: some 97 percent of the banking sector by assets was wiped out.Footnote 113 To stabilize the Central Bank and recapitalize the state, the government injected emergency funds equivalent to nearly 20 percent of GDP.Footnote 114 Iceland’s debt-to-GDP ratio more than doubled, rising from about 38 percent in January 2008 to over 80 percent by year-end and in subsequent years over 100 percent of GDP.Footnote 115
Beginning November 15, 2008, protesters gathered outside Alþingi demanding the government’s resignation and the departure of senior financial officials. The protests escalated into mass marches and pots-and-pans demonstrations, and by January 2009 had turned violent, a first in Iceland’s political history. The Social Democratic Alliance broke from the Independence Party and formed a new government with the Left-Greens, promising transparency and structural reform, while key officials, including the trade minister and senior financial supervisors, resigned as the outgoing government ended.Footnote 116
To restore credibility, the newly created successor banks, Arion Bank, Íslandsbanki, and Landsbankinn, appointed foreign board members with international finance experience and no ties to Icelandic elites. State control had been reasserted, but informal power continued to circulate along familiar lines.Footnote 117 Oddsson remained in office until 2009,Footnote 118 when the incoming government amended the Central Bank Act to remove the governors. Within months he was appointed editor of Morgunblaðið, the newspaper long aligned with the Independence Party. Influence was not eliminated but redeployed to a new arena. Jackson traces, across three centuries of European financial crises, the recurrent afterlife in which accountability concentrates on a few figures while the architects of the system circulate onward into new positions of influence.Footnote 119
Reform was incremental. The FME received enhanced statutory authority, and the Central Bank’s tripartite governance structure was revised. However, no independent commission was created to redesign financial supervision, and there was no systemic replacement of top officials. The SIC published a 2,000-page report documenting what it characterized as a “failure of governance at every level.” Implementation of its recommendations was at best partial, as responsibility for follow-up remained largely within the same political and administrative framework that the Commission had scrutinized.Footnote 120 What collapsed was a financing model, not the underlying repertoire of elite coordination.
The 2008 collapse created a moment in which displacement was possible, and the emergency response did overturn ownership and creditor hierarchies. Yet crisis management itself was concentrated at the apex of the same narrow network that had overseen liberalization and privatization. Informal meetings, executive improvisation, and personal channels substituted for routine institutional procedure, and the architect of earlier reform presided over the resolution from within the Central Bank. Foreign directors briefly signaled a break, but the deeper pattern survived because the post-crash state lacked both time and a separate pool of personnel from which to reconstruct authority. The result was displacement of the failed banking model, combined with conversion of many of the same actors into new positions across the crisis apparatus.
Post-Crisis Reform and Institutional Continuity (2010–2020)
The Independence Party had been ousted from government for the first time in decades, and mass protests had forced the sitting prime minister’s resignation; the post-crisis period began under the banner of transparency, justice, and reform. An IMF-backed economic program, a sequence of new financial statutes, and the establishment of the SIC marked a significant shift from the past. However, while the rhetoric of rupture was strong, the institutional trajectory that followed largely reflected continuity through adaptation.Footnote 121
The state restructured the failed banks by separating domestic and foreign assets and liabilities. New institutions, Arion Bank, Íslandsbanki, and Landsbankinn, took over domestic operations, while residual claims and cross-border exposures were left in the old banks’ estates for liquidation and recovery of value.Footnote 122 The new Landsbanki’s first chief executive, Sigríður Elín Sigfúsdóttir, appointed on October 9, 2008, came from the failed bank’s own corporate banking division (Table 1); she held the post until February 2009.Footnote 123
The newly created Icelandic State Financial Investments (ISFI) managed the state’s stakes in the banks. It implemented formal vetting procedures but relied on a relatively homogeneous recruitment pool, figures regarded as professionally credible and politically safe within Iceland’s tightly knit administrative and advisory circles.Footnote 124 Arion, majority-owned by Kaupthing’s creditors, was chaired from 2010 by Monica Caneman, a Swedish banker formerly of SEB, and on her departure in 2017 by another Swede, Eva Cederbalk: foreign ownership produced foreign governance.Footnote 125 Íslandsbanki took the opposite signature. The bank rebuilt at arm’s length from the old system was chaired for its entire reconstruction, from 2010 to 2020, by one of that system’s principal architects, Friðrik Sophusson (Table 1).Footnote 126 When the creditors exited, and appointments passed to the state’s holding company, board composition reverted toward domestically embedded appointees, legal advisers, former civil servants, and business professionals whose careers were intertwined with ministries, state agencies, or the pre-crisis financial ecosystem; the foreign interlude closed as a parenthesis rather than a transformation.
Was this continuity a matter of political connection or necessity? The question deserves a direct answer because the practical constraint was real. A payments system cannot pause while replacements are found; operational knowledge of borrowers and counterparties is not substitutable overnight; and regulators anywhere in 2008 kept most incumbent managers in place while replacing chief executives and boards, in the United States as in Iceland. In a country of about 320,000, the constraint binds even harder: the pool of people qualified to run a bank and untouched by the old system was nearly empty. The initial continuity of 2008–2010, Sigríður Elín’s appointment among it, is fully explained by necessity. But necessity is a hypothesis, and it carries a prediction: as the emergency passed, governance should have converged toward ordinary international practice, and the imported outsiders should have become unremarkable. The opposite happened. Iceland did import outsiders at the moment of maximum need: Sven Harald Øygard as interim Central Bank governor, and foreign chairs at the creditor-owned banks, Monica Caneman at Arion and later Eva Cederbalk. As normality returned, this was unwound and appointments re-domesticated through the state’s ownership channels. Necessity explains why the network was retained in 2009; it cannot explain why the network was restored by 2020. That asymmetry, not the initial continuity, is the evidence of elite adaptation.
The reforms were substantial in scope but uneven in institutional effect. The FME was granted expanded statutory powers. Capital controls were implemented and gradually lifted.Footnote 127 The empowered regulator had to be staffed from somewhere, and the somewhere was the old banks. In April 2009, the new director-general became Gunnar Þ. Andersen, a veteran of Landsbanki (Table 1).Footnote 128 In November 2010, the FME’s board examined his earlier work for Landsbanki’s offshore companies and saw no grounds to act. Fifteen months later, on new information, two external assessors concluded that the same history cast doubt on his fitness to lead the agency, and in March 2012 the board dismissed him.Footnote 129 In 2013, he was convicted of procuring confidential Landsbanki documents for the press. It demonstrates that even the reformed watchdog could not be recruited from outside the web it was built to watch.Footnote 130
Criminal accountability was also unusually extensive: by the mid-2010s, prosecutions had produced multiple Supreme Court convictions, including long prison sentences for several CEOs and senior chairs.Footnote 131 Sigurjón Þ. Árnason, the Búnaðarbanki operations chief who had privately doubted the German participation in his own bank’s privatization, became joint chief executive of the privatized Landsbanki in April 2003 and was indicted by the Special Prosecutor on March 15, 2013, together with Sigríður Elín. The Reykjavík District Court acquitted both on June 5, 2014. On October 8, 2015, the Supreme Court reversed, sentencing Sigurjón to three and a half years and Sigríður Elín to eighteen months for the Ímon transactions. In 2016, Sigurjón received a further eighteen months for manipulation of the bank’s own share price.Footnote 132 The prosecutions were real, but criminal liability falls on individuals for specific transactions, not on the arrangements that generate them.
On December 5, 2016, Icelandic media revealed that Supreme Court justices had held undisclosed shares in the collapsed banks; of the five who convicted Sigríður Elín, three had held such investments. One of them, Justice Viðar Már Matthíasson, had bought Landsbanki shares in 2007 for ISK 14.8 million and lost them in the collapse, a loss the European Court of Human Rights later called substantial and nearly three times the domestic disclosure threshold. Iceland’s Committee on Reopening accepted Sigríður Elín’s request in May 2019 on this ground, and in February 2020 Strasbourg held that “the impartiality of the Supreme Court could have been open to genuine doubt.” In March 2021, the Supreme Court reheard the cases, upheld the convictions, and suspended the sentences.Footnote 133
The crisis did not dissolve elite cohesion; it destabilized elite legitimacy and redirected the terms on which cohesion could be rebuilt. Political parties continued to shape appointments through the state’s ownership and oversight channels, including ISFI. At the same time, former officials and politically connected professionals reappeared in advisory and governance roles. Reform agendas proliferated, implementation capacity fragmented, and several initiatives lost traction amid political conflict and administrative overload, a pattern Jónsson and Sigurgeirsson describe as post-crisis reform processes that “went nowhere” despite the opening created by the collapse.Footnote 134
Iceland’s rapid return to macroeconomic stability helped mask the extent of institutional continuity. By the late 2010s, the post-crisis banks were materially smaller, more cautious, conservatively funded, and less dependent on short-term foreign wholesale markets than their predecessors.Footnote 135 The post-crisis decade was the phase in which formal change was greatest and governing logic most durable. Legislators granted new supervisory powers, the government imposed capital controls, and criminal prosecutions created an unprecedented appearance of accountability. Even so, the reformed institutions continued to draw on the same small pool of domestically embedded actors. That asymmetry, necessity in 2009 and restoration by 2020, is the key finding. The period is best understood as layering and drift: new bodies and new rules were added, but the practical authority to use them remained concentrated within the field they were meant to transform. Reform therefore changed the appearance of banking governance in Iceland more than it changed the inherent social organization of power.
Reprivatization and the Return of Insider Capitalism (2021–2025)
More than a decade after the 2008 collapse, the Icelandic state still held large equity stakes in the commercial banking sector. This situation, resulting from crisis-led nationalization, provided a unique opportunity to assess whether post-crisis reforms changed the link between finance and political power. Could the government divest its ownership through clear, market-based methods rather than through discretionary deals via elite connections?
The reprivatization program began with a broadly marketed initial public offering (IPO) of Íslandsbanki in 2021. The ISFI sold 35 percent of the bank at ISK 79 per share (ISK 55.3 billion), attracting roughly 24,000 shareholders, leaving the state with 65 percent.Footnote 136 The offering was designed to signal openness and broad participation.
The second phase in 2022 took a markedly different form. ISFI executed an accelerated bookbuild offering of 22.5 percent (450 million shares) to a screened group of 207 investors at ISK 117 per share (ISK 52.7 billion), reducing the state’s stake to 42.5 percent. The shift from an open IPO to a restricted offer arrangement raised concerns about preferential access and the opacity of the allocation process.Footnote 137 When the buyer list was published on April 6, 2022, over the objections of the ISFI itself, it read as a reunion of the pre-crisis elite, specifically of the old Glitnir circle repurchasing its successor. Þorsteinn Már Baldvinsson, Glitnir’s chairman at its collapse, bought just over ISK 296 million through the holding company Eignarhaldsfélagið Steinn; FL Group, Glitnir’s largest pre-crisis shareholder, returned under its new name Stoðir hf., with its former chief executive Jón Sigurðsson among its principals; and Jón Ásgeir Jóhannesson, the dominant figure of that ownership bloc, participated indirectly through associated vehicles. The 118th-largest allocation, just under ISK 55 million for 0.1042 percent of the bank, went to Hafsilfur ehf., the investment vehicle of Benedikt Sveinsson.Footnote 138
Benedikt Sveinsson was not simply another qualified investor. He is the father of Bjarni Benediktsson, the finance minister who had approved the sale on March 22, and the family line runs back to this article’s opening cast: his own father, Sveinn, was one of the Benediktsson brothers who succeeded the Thors family at the Independence Party’s core in the 1960s, the brother who ran the family’s business arm while Bjarni the elder held the premiership and the third brother, Pétur, directed Landsbanki itself from 1956 to 1969.Footnote 139 Before 2008, Benedikt and his brother Einar had been principal shareholders of Íslandsbanki-Glitnir,Footnote 140 Einar serving as its chairman.Footnote 141 In October 2023, the Parliamentary Ombudsman found that the minister had lacked competence to approve a sale in which his father’s company bought,Footnote 142 while accepting that he had not known of the participation; Bjarni Benediktsson resigned as finance minister on October 10, 2023, and within a year was prime minister. The judgment was procedural, the sums trivial, the minister’s good faith undisputed. That is precisely the point: in a financial field this small and this dynastic, even a scrupulous minister could not stand outside the network his office was selling to.
In its subsequent audit, the National Audit Office concluded that the preparation and execution of the March 2022 sale did not meet expected standards of administrative practice: the sale’s objectives and operational criteria were insufficiently specified; the documentation provided to parliament did not give a clear picture of the arrangement; and ISFI itself recognized that the structure was not fully consistent with the spirit of the governing legislation concerning openness, transparency, and equality of bidders.Footnote 143 The audit also emphasized how reliance on external advisers and weaknesses in documentation made it difficult to test decisions after the fact, turning a method intended to secure market pricing into a legitimacy problem about traceability and equal treatment. Further sales were paused, and the episode became a public legitimacy crisis for the reprivatization program, but the transaction itself was not reversed.Footnote 144
Regulatory follow-up was more consequential. In June 2023, the Financial Supervisory Authority of the Central Bank of Iceland concluded the case through a settlement with Íslandsbanki, imposing an administrative fine of ISK 1.16 billion and requiring remedial measures. The settlement described multiple breaches in the bank’s conduct as a sales agent, including failures in investor classification, conflicts-of-interest controls, disclosure and communication practices, and shortcomings in governance and internal controls. It found that the bank had not satisfied its obligation to act honestly, fairly, and professionally and to promote market integrity, underlining that the regulatory concern was not political influence as such but failures of market-process safeguards.Footnote 145
The political system also responded by rewriting the legal and organizational architecture around bank divestment. The Alþingi adopted a special statute authorizing the disposal of the state’s remaining holding in Íslandsbanki (Act No. 80/2024)Footnote 146 through a marketed public offering open to both retail and professional investors. To address the legitimacy deficit exposed in 2022, the statute hardwired retail priority by splitting the offering into two books, with Book A reserved for individuals at a fixed price linked to recent market prices, and required public disclosure of transaction-level information, including the identities of ultimate buyers, as well as an independent post-offer review of compliance.Footnote 147 Alþingi abolished Icelandic State Financial Investments from January 1, 2025, and moved responsibility for the state’s financial-sector holdings to the Ministry of Finance and Economic Affairs.Footnote 148 Under this framework, the state completed its divestment in May 2025: a fully marketed offering settled on May 23, 2025 and raised gross proceeds of about ISK 90.5 billion, leaving the state with no equity stake in Íslandsbanki.Footnote 149
The reprivatization cycle is the first episode in this history that encountered a genuine external constraint. The 2022 sale initially reproduced the older pattern with striking clarity: a screened allocation process, a narrow buyer base, and a set of investors closely linked to the pre-crisis financial elite. The transaction therefore looked, at first, like another displacement of formal ownership that left elite access intact. What altered the story was the response. Public scrutiny, audit, supervisory enforcement, and political fallout forced the reprivatization process to become more procedural, more transparent, and less dependent on discretion. The 2025 divestment was therefore conducted under a different framework, one designed to reduce the scope for insider allocation. The important point is not that political capitalism disappeared. It did not. Rather, it shows that absorptive capacity can be constrained when external accountability is sustained and institutionalized. That is the boundary condition absent from the earlier phases of Icelandic banking history.
Conclusion
This article traces the evolution of Iceland’s banking system in six phases over more than a century of institutional transformation: from the founding of Landsbanki in 1885, liberalization and privatization, the rise and collapse of the privatized banks, re-nationalization and crisis management, and post-crisis reform, to the reprivatization of the 2020s. Across that sequence, the finding is consistent: elite influence was repeatedly reproduced and reconfigured through changing institutional forms rather than displaced by them. Only in the 2020s did that capacity to absorb reform meet a limit. Banks were merged, nationalized, privatized, restructured, and sold again, and regulators were empowered at each turn, but the channels of influence, political appointments, insider lending, revolving-door careers, and closed networks proved more resilient than the reforms directed at them.
Analytically, Beckert’s triadic lens helps explain why repeated reforms can yield persistent fragility: formal changes to rules and ownership reshape institutional openings, but actors exploit networks and cognitive frames to legitimize new practices and reconstitute influence. Mahoney and Thelen’s account of gradual change specifies how this occurs within the institutional dimension. Each episode’s mode, whether layering, drift, or conversion, or, at privatization, a displacement confined to ownership form, turned on how far the defenders of existing arrangements could block change and how much the rules could be reinterpreted; and across the century those conditions favored absorption over rupture.
This account also speaks to a broader debate. Calomiris and Haber argue that stable banking systems are those in which a concentrated financial elite operates at arm’s length from populist political interference, and that crises originate in the distortions populism introduces. Iceland fits the first half of their account and contradicts the second. Its banking system was elite-run and substantially insulated from populist pressure, the very configuration their framework associates with stability, yet it produced, in 2008, the largest banking collapse relative to national income in recorded history.Footnote 150 The Icelandic case suggests that elite insulation is not a guarantor of stability but can itself be a source of fragility when the insulated elite faces no external check and its networks span the institutions meant to constrain it.
Several comparisons suggest themselves. Ireland and Spain also experienced banking collapse and substantial institutional reform after 2008: did their elites persist, and did change there proceed through layering, drift, and conversion? A second line of comparison runs to other small states, and to those that built indigenous financial systems in the late nineteenth and early twentieth centuries, where the relation between elite formation and banking development may have followed different paths.
Iceland is an extreme case that renders the social organization of financial governance unusually visible.Footnote 151 It suggests a caution for both scholars and policymakers: institutional change cannot be assessed only by statutes enacted or agencies redesigned. It must also be evaluated in terms of who (re)gains access, how authority circulates, and which frames confer legitimacy. The Icelandic sequence suggests one further lesson. For more than a century, elite coordination absorbed every reform directed at it, metabolizing new rules into old logics. Only in the 2020s, and only under sustained external accountability from a free press, an independent audit office, and public pressure capable of unseating a minister, was that absorptive capacity visibly checked. Political capitalism, on this evidence, is not immune to reform, but the constraints that bind it tend to come from outside the institutions it has captured, and whether those constraints endure remains untested.
Open access
