Which poses a greater threat to democracy: economic inequality or political polarization?
Inequality is the greater threat. Polarization is what inequality produces when citizens notice the gap.
Rousseau, Luxemburg, and Hirschman each trace a separate mechanism from inequality to democratic collapse. Concentrated wealth distorts civic voice, funds elite exit from shared institutions, and lets demagogues fill the space left behind. Hirschman's Colombia case makes this concrete: when wealthy citizens exit public schools and hospitals, those institutions decay, and polarizing rhetoric finds ready audiences in the wreckage.
Arendt's counter-evidence is real and must be named. Weimar Germany in 1932 showed millions of non-starving citizens voting to dissolve a republic, driven by atomization rather than hunger. Polarization can accelerate democratic collapse even when material conditions are not desperate.
But the council's weight falls on the structural side: Palme's Sweden, built on the Rehn-Meidner wage compression model from the 1950s onward, reduced the grievances that polarization feeds on. Arendt names a genuine accelerant; Rousseau, Luxemburg, Palme, and Hirschman name the fuel.
Confidence summary: Strong aggregate confidence in the causal primacy of inequality, with a genuine and unresolved minority position from Arendt on polarization's autonomous destructive power.
1. The core argument
The most important finding from this council is not the verdict itself but the mechanism underneath it. Inequality does not simply make citizens poorer. It enables the wealthy to exit the shared institutions, schools, hospitals, public squares, where political life is actually conducted. Once elites leave, those institutions decay. Decayed institutions produce grievances. Grievances produce audiences for demagogues. Polarization fills the institutional space that inequality has already emptied.
This sequence matters because it changes the policy target. If polarization were the primary threat, the correct intervention would be at the level of discourse: restoring shared facts, repairing civic norms, reducing algorithmic amplification of conflict. If inequality is primary, discourse interventions operate downstream of the real problem. Palme's Sweden is the council's strongest evidence: compress wages, universalize social insurance, keep elites inside shared institutions, and the demagogue's wedge finds no gap to enter. The verdict rests on this structural logic, not on dismissing Arendt's counter-evidence, which remains real and requires a separate reckoning.
2. How each member frames it
Jean-Jacques Rousseau insists the question is about the conditions under which a general will can form at all. His deeper point, which the card could not fully carry, is that material inequality does not merely disadvantage the poor; it corrupts the rich. Citizens who possess what others need cannot reason toward genuine common interest because their private advantage is structurally opposed to the public good. Rousseau would accept significant constraints on property rights as the price of genuine civic equality. What he rejects entirely is the reformist position that you can have radical material disparity and functional democracy simultaneously. The two are, for him, logically incompatible.
Hannah Arendt does not deny inequality's damage, and the card understated how carefully she qualifies her position. Her specific and irreducible claim is that polarization can become independently lethal once a shared factual world collapses. In November 1932, many Germans who voted for dissolution were not destitute; they were disoriented. Arendt would push back sharply on Hirschman's institutional decay model: the Weimar case suggests that even functioning institutions cannot survive when citizens lose the capacity to agree on what is real. Her candid limit is that she offers no mechanism for restoring shared reality once it has fragmented, only a theory of why losing it is fatal.
John Rawls plays the role of structural integrator, and the card did not fully surface why that position is more radical than it appears. Rawls is not simply saying both threats matter equally. He is saying the basic structure of society, the rules governing property, opportunity, and political rights, determines whether public reason is even possible. On his reading, a society that permits extreme inequality has already corrupted the conditions for legitimate political deliberation, making Arendt's shared world structurally impossible to maintain. He would reject Luxemburg's revolutionary conclusion while accepting much of her diagnosis about what concentrated wealth does to legislative processes.
Rosa Luxemburg sharpens the critique of parliamentary incrementalism in ways the card could not develop. Her challenge to Palme is pointed: Sweden's Rehn-Meidner model succeeded in a postwar environment of organized labor, export-dependent capital that could not easily exit, and a political party with genuine working-class roots. Strip any one of those conditions and the reform path closes. She would argue that what Palme built was historically contingent rather than a reproducible model, and that the current period, with mobile capital, weakened unions, and captured media, more closely resembles her 1899 diagnosis than his 1970s achievement.
Olof Palme accepts Luxemburg's historical conditions argument more than the card suggested, but draws the opposite operational conclusion. He does not claim the Swedish model is infinitely exportable. He claims it proves the causal chain runs from material stakes to political solidarity, and that breaking the chain at the material level is both possible and democratic. His candid limit is Hirschman's challenge: the Rehn-Meidner model required elites to remain inside shared institutions. Once elite exit becomes available and affordable, redistribution alone may not close the gap fast enough to prevent institutional decay.
Albert O. Hirschman brings the most empirically grounded mechanism, and the card understated its pessimistic edge. His Colombia observations suggest that elite exit and institutional decay are self-reinforcing: the worse public institutions become, the more elites invest in private alternatives, which accelerates the decay. This dynamic makes Palme's redistribution model harder to initiate the longer it is delayed. Hirschman's implicit policy implication is that the window for redistribution-led recovery narrows over time, and that institutional restoration may need to precede, not follow, material redistribution in severely unequal societies.
3. Where the council agrees
The most surprising point of agreement is that polarization cannot be treated as a communication or media problem alone. Every member, including Arendt, locates the roots of democratic fragility in structural conditions rather than rhetorical or informational ones. Rousseau and Luxemburg attribute those conditions to property relations; Arendt attributes them to the collapse of shared political space; Hirschman attributes them to institutional hollowing. None treats better journalism, fact-checking, or civility norms as sufficient responses. The council also agrees that the inequality-polarization relationship is causal rather than merely correlational: inequality generates the material and institutional conditions that polarizing rhetoric then exploits. Finally, all six members agree that elite behavior is the critical variable. Whether elites stay inside shared institutions, whether they capture legislative processes, whether they fund exit from public goods, determines the trajectory more than mass citizen attitudes do.
4. Where the council splits
The genuine split is between Arendt and the remaining five members on a specific and consequential question: can polarization destroy democracy without inequality as its proximate cause?
Arendt says yes. Weimar's collapse was driven by atomization and the destruction of shared facts, not by the poverty of voters who chose dissolution. On her reading, a democracy can be brought down by an epistemic crisis even when material conditions are tolerable.
Rousseau, Luxemburg, Palme, and Hirschman say the Weimar case does not refute the structural argument; it illustrates it. The atomization Arendt identifies was itself produced by the structural dislocations of the interwar economy. The shared world had already been dismantled by inequality before the epistemic crisis arrived. Rawls sits closest to Arendt among the majority, but still holds that structural conditions determine whether epistemic repair is even available as a political option. Neither side is wrong on the evidence; they disagree on which mechanism is doing the decisive work.
5. For a policymaker to decide on
The council cannot resolve whether to target material inequality first, through redistribution, wage compression, and universal public services, or to invest first in restoring the shared civic institutions through which citizens develop common stakes. Hirschman's framework suggests elite exit makes both strategies harder simultaneously, and that sequencing matters enormously. A government with limited political capital must choose: does it open with redistribution and hope institutional recovery follows, or repair institutions first and risk losing the political window for redistribution entirely?