Adam Przeworski

some notes

* classical and modern democratic theorists disagree strongly on political parties: now hard to see how we thought we could do without them. Kelsen: "the ideal of a general interest superior to and transcending interests of groups, thus parties, the ideal of solidarity of interests of all members of the collectivity without distinction of religion, of national­ity, of class, etc. is a metaphysical, more exactly, a metapolitical illusion, habitually expressed by speaking, in an extremely obscure terminology,
of an ‘organic’ collective or ‘organic’ structure."

* fun tidbit from classical small-population solution of rotation: "the ceremonial office of the epistates, who held the seal of Athens and the keys of the treasuries and represented Athens in relation to other states, could be held for only one night and one day in one’s life"

* why majority rule? one justification is May’s theorem: "simple-majority rule is the only rule of collective decision making that satisfies four axioms: equality, neutrality, responsiveness, and decisiveness." equality just means anonymity in the sense that any two individuals can interchange preferences and collective decision is unchanged. neutrality means decisions don't depend on the names attached to alternatives. decisiveness means it's clear what we must do when collective decision is made. responsiveness means at least one person changing preference could flip the decision.

* why majority rule? another view: it's a proxy we use to gauge and minimize potential of successful violent opposition to collective decision.

* why majority rule? third answer is Rae's theorem: "simple-majority rule maximizes the number of people who live under the laws they like."

* problems with majority rule: Condorcet's paradox and its generalization, Arrow’s theorem.

* deep equivalence between equality before the law and anonymity: "The law has to treat all citizens equally because as citizens they are indistinguishable"

* summary of palmer's Age of the Democratic Revolution: "democracy was not a revolution against an existing system but a reaction against the increas­ing power of aristocracy. It was aristocracy that undermined monarchy; democracy followed in its footsteps." also check out John Dunn's Democracy: A History.

* why historically equality just meant anonymity: "Abolishing all distinctions [is] a logical outcome of the struggle against aristocracy. The fact is that democrats turned against all distinctions. The only attribute of democratic subjects is that they have none as such. The democratic citizen is simply without qualities. Not equal, not homo­geneous, just anonymous." comes across strongly in Rousseau quotations.

* suffrage eligibility rules tricky in principle: kids, the insane. unclear to me why these are not negligible rounding errors, random noise, and low cost to include.

* levelers, babeuf, marx, ricardo and many other socialists and conservatives all agree: material inequality not compatible with democracy -- and yet!

* why not more economic redistribution with democracy? initially suffrage restriction replacing aristocracy with what was effectively competitive oligarchy. but universal suffrage leads to puzzle. possible explanations include (1) deluded populace, (2) power elite, (3) marx's explanation (see below), (4) przeworski's explanation (see below).

* marx on limits of bourgeois rights: "The state abolishes, in its own way, distinctions of birth, social rank, education, occupation, when it declares that birth, social rank, education, occupation, are nonpolitical distinctions, when it proclaims, without regard to these distinctions, that every member of the nation is an equal participant in national sovereignty... Nevertheless the state allows private property, education, occupation to act in their way – i.e., as private property, as education, as occupation, and to exert the influence of their special nature."

* przeworski's explanation: "governments of all partisan stripes must anticipate the trade-off between redistribution and growth. Redistributing productive property or even incomes is costly to the poor. Confronting the perspective of losing their property or not being able to enjoy its fruits, property owners save and invest less, thus reducing the future wealth and future income of everyone. This 'structural dependence on capital' imposes a limit on redistribution, even for those governments that want to equalize incomes" but he admits empirically there is reason to doubt this trade-off.

* american institutional setup pretty weird, madison based it on hunches and reading of philosophers. the checks and balances stuff in particular has problems: empirically it doesn't seem more stable, and even in theory doesn't make much sense

* true democracy likely has to be representative setup with universal suffrage, unicameral parliament, strict majoritarian rules. author makes simple but subtle point that non-unicameral setup is mathematically equivalent to supermajoritarian rules.

* there is nothing very natural about the government schematized into branches like executive, legislative, etc. after all, their powers supposedly overlap to intentional check ambition with ambition (in american setup), which implies the powers do not themselves delineate the branches. also, they are not exhaustive of components necessary to govern: eg, education system pretty different and arguably more fundamental for self-rule. important not to forget we inherited some stupid categories scribbled out by philosophers who had no experience with actual democratic government.

* there is some accomplishment in having replaced aristocracy with competitive oligarchy: elites imposing themselves different from elites proposing themselves. even autocrats pay respect to need for democratic mandate by bothering to hold sham elections.

* still, we must wonder if this is best we can do? america in particular had a lot of obviously cobwebbed and busted institutions designed by people who knew little about democracy, and the checks and balances and bicameral legislature should be first things we scrap.… (plus d'informations)

przewowski one of my favorite thinkers, and this is his opinionated introduction to the methods used by modern political economists. notes and quotes below:

- markets are exchanges between individuals with some initial endowments; they can only take place only under the auspices of the enforcing authority
- market equilibrium reached when trades cease, no one can be made better off without someone being made worse off (pareto optimal)
- pareto optimality is always relative to initial endowments, which sidesteps some of the deeper justice questions; pareto optimality reach only if transaction costs are zero and everyone has perfect information, but under those conditions centralized planning gets same result
- now turn analysis toward enforcing authority (eg, a cop) and ask why does he not just grab everyone's initial endowment for himself? this question widens analysis to include full political-economic equilibrium
- we focus on democratic state with markets, and ask how its enforcement authorities (eg, regulators) will behave
- going back to market, there are a variety of potential market failures: increasing returns to scale lead to monopoly and deadweight loss to society in static analysis; in dynamic analysis entrepreneurial rents only accrue if there is some monopoly; nonrival consumption of goods leads to undersupply by market; positive/negative externalities lead to under/over supply as well
- market failures much discussed in 1950s, and by 1970s corresponding failures of governance as well
- no market encompasses all possible goods, so all markets are said to be incomplete; while they are incomplete prices are not sufficient statistics summarizing forsaken opportunities in a global sense, only locally given the fact of missing markets (eg, in externalities or futures)
- imperfect information means that some individuals learn from seeing others trade, generating keynes-type beauty contest phenomena: multiple rather than unique equilibria result
- asymmetric information an issue, else many trades don't make sense: eg, why consult a doctor? but with asymmetric information comes principal-agent problems
- different contract types between landlord and peasant illustrate how principal-agent problems under imperfect symmetric information (nature moves and both observe) allocate both consumption and risk differently; wage/tenancy/sharecropping make landlord/peasant/both bear risk; optimal choice of contract depends on risk posture of principal and agent
- asymmetric information can be due to hidden action of agent (leading to moral hazard) or hidden characteristics of agent (leading to adverse selection); principal must tradeoff efficiency and incentives, paying rents (income above marginal product) to induce effort from efficient agent and separate out inefficient types
- to say anything interesting about what the public officials will do we need to assume something about their objectives: are they public-spirited, self-interested, or what?
- under predatory state model, public official extracts rents from society via theft or usufruct--ie, either appropriating fiscal residuum (patrimonial state) or merely enjoying perks of public office in excess of their next best use (bureaucratic state)
- patrimonial state will undersupply public inputs to production (shitty roads); bureaucratic state will oversupply public inputs to production (grand boulevards)
- accountability mechanisms like democratic elections can constrain both if with model official as maximizing rents/perquisites over time
- elaborating the model, we can have elected politicos (principals) who set policy and then delegate the implementation to unelected bureaucrats (agents), and in turn the bureaucrats regulate market actors; "One of the most puzzling, and poorly understood questions, is 'Why do people who have guns obey people who do not have them?'"
- "To solve for such equilibria, we must work backward. We must first ask about the effects of regulation on individual actions, then about the effects of decisions on the actions of those who implement them, and finally about the collective decisions. For every implementation, there exists a course of action that is best for private agents given their constraints. The bureaucrats know this (perhaps imperfectly or incompletely), so that they can choose an implementation that is best for them given private agents' reactions. The decision makers know this, so that they can choose the decision that gets them the best implementation. Such asymmetric Nash equilibria are called Stackelberg equilibria..."
- again, it turns out bureaucrats will face the typical tradeoff between principals and agents: either pay the agents rents or accept socially worse outcome -- eg, cost-plus contract or fixed priced contract. "It turns out that the regulator will want the efficient firm to exert the efficient level of effort and will allow the inefficient firm to shirk. This result, sometimes called, 'efficiency at the top,' is generic for this kind of problem"
- paradox of regulation: political systems that can easily pass legislation cannot credibly commit to any course of action, undermining their ability to regulate
- one endogenous regulation model is regulatory capture: powerful interests extract rents from powerless. but unclear why voters would permit it: needs to postulate rational ignorance or active obfuscation by state
- another endogenous regulation mode is virginia school, where government distorts markets and awards rents to regulated firms; author casts doubt that these are true rents or would necessarily distort economy, as government could play vying factions off each other and choose non-distortionary policy
- empirically hard to tell whether money buys policy or merely follows it
- oversight of bureaucracy is helped by fact that regulatory losers tattle to lawmakers
- oversight of bureaucracy can rely on incentives (ex post, managerial control) or conformity with rules (ex ante, bureaucratic control), also "fire alarm" of information provided by affected parties
- political control of bureaucracy of ambiguous value: if guided by partisan interests devolves into clientelism
- partisan motivation does not imply socially suboptimal
- checks and balances another institutional fix: multipartisan system (doesnt work well, as bureaucracy can play partisans off of each other to their benefit via "implementation coalitions") or contramajoritarian (oversight of the elected by non-elected, courts and independent agencies -- dubious democratic legitimacy)
- kind of remarkable that insulated bureaucracies work to the extent they do, and unclear what alternatives for oversight would be
- "Voters have only one instrument - the vote - to reach two goals: retrospectively sanction the incumbent and prospectively choose a better candidate. If they care about both, they must trade these goals against one another."
- on politics of economic growth: "Whereas everyone wants to consume, each wants all others to invest. Because an individual who consumes at one moment also wants to consume in the future, when the possibilities of one's future consumption depend on the stock of productive inputs accumulated by others, at each time each individual wants everyone else to invest as much as possible."
- median voter model founders if multiple policy dimensions not easily collapsed into unidimensional ideology: "The median voter model is simple and analytically tractable but wildly unrealistic and limited in scope. In turn, multidimensional models suffer from multiple equilibria and are ... analytically intractable"
- one reason egalitarian redistribution of income may not follow from median voter model: government does little pure income redistribution, and variety of public services lead to multiple orderings of voters, so no single median voter
- still, remarkable that taxes on rich are not higher. various theories: aspirant rich etc.
- social insurance emerged in sweden for militaristic reasons: "healthy boys to fight in wars"
- private insurance faces asymmetric information problem, risk pooling results; easy state fix here--compel insurance
- an endogenous models of public insurance: as smaller fraction of polity in dangerous jobs, voter demand for public insurance falls… (plus d'informations)