Professor of Political Science · Department of Government, London School of Economics
I am a political scientist at the London School of Economics. I model how people act together – parties and their factions, governments and the coalitions that make them, cabinets and their ministers, leaders and the led. One idea runs through the work: what holds a collective together is, past a point, what breaks it. I study where that point lies: coalition termination, party splits, ministerial resignations, and candidate exits. I have an interest in British political history and the development of its core institutions and voting behaviour.
A recent strand of my work reads classical rabbinic law as formal political economy. The Talmud, beneath its legal surface, argues about institutions – how to seat a court, how to divide a contested asset, how to render judgment when the bench itself may be captured – and its procedures often encode the very trade-offs a modern theorist would isolate. I take these sugyot as design mechanisms (in the spirit of Aumann & Maschler, 1985) and ask what, in the strategic setting each presupposes, the design achieves.
I teach courses on Social Choice, Public Choice and Game Theory (undergraduate and postgraduate), a graduate seminar on Leadership, and one on Institutions (Parties and Coalitions, Elections and Government).
As well as studying and teaching institutions, I help build them in my own field. I founded the Political Science and Political Economy (PSPE) research group at LSE in 2010, with Simon Hix, and the LSE/NYU annual conference, held since 2012, with Dimitri Landa. I am also a co-founder of the PolEconUK network. I currently run the PSPE research seminar, with Tim Besley, Tak-Huen Chau and Noam Yuchtman. In 2015 I hosted the inaugural Comparative Politics and Formal Theory conference at LSE. I am co-editor, with John Patty, of the Journal of Theoretical Politics.
My private passions are first and foremost Chelsea Football Club, where I have long stood on its famous “Shed End”, and now proudly stand alongside my two children; combat sports (mainly BJJ, Brown Belt since 2025 under Manxhina at London Fight Factory); and music (second generation punk, now classical). I enjoy food and wine and am currently writing a book (with Abhinay Muthoo), Breaking Bread: The Politics of Food, that connects models of leadership, bargaining, and conflict to chefs, restaurants and markets.
When a leader's clarity coordinates a movement so well it overshoots a catastrophe, the instruments of agreement – clarity, authority, cohesion – change sign.
When coordinated action can carry a collective over a catastrophic threshold, the institutions that manufacture agreement – unified leadership, cohesive parties, centralised authority – turn into hazards, and restraint takes the form of deliberate fragmentation: factions, federations, separated powers, staggered authority.
A party of two factions sustains a common manifesto only by tolerating occasional splits – the political analogue of Green–Porter price wars, on the equilibrium path and on factions that have not defected.
Coalitions as repeated disciplinary structures rather than one-shot divisions of spoils: who can govern together depends on whom they can monitor.
Career concerns set inside a cabinet: judging ministers against one another nets out the fortune they share, so a scarce promotion becomes a contest won at the margin – and the leader's instrument is no longer a wage but the composition of the bench she gathers for comparison.
Talent is drawn to a winning team, making recruitment self-reinforcing; a team's survival turns on stepping-stone types – some never observed to act – that carry it between prospering and dissolving.
The random draw of the FA Cup as a natural experiment in thousands of income shocks: a small club drawn away to a much larger one is handed a financial windfall and climbs the league pyramid over the following decade – an effect present even among clubs that lose the tie, and, for upward movement but not protection from relegation, conditional on the organisational capacity to spend it well.
The two-team sequel to Playing for the Winning Team: when organisations compete for the same talent, loyalty is what lets a trailing rival survive rather than collapse into a single dominant winner, and noise can sustain, even revive, the contest.
A two-sided variation on Playing for the Winning Team: winning outright ends the contest that motivated the effort, so the upper barrier becomes a cliff rather than a crown – the same force that builds a winning team is the one that can carry it over the edge.
The time-series companion to The Cabinet as Yardstick: a leader's scarce talent is worth one thing fielded and another withheld, so the best are husbanded while a government is strong and deployed as it weakens – a succession schedule, with some never fielded at all.
The buy–sell ('Texas shootout') clause and its Talmudic ancestor gud o agud (Bava Batra 13a): no priced clause can dissolve efficiently, protect each owner, and allocate to the higher-valuer at once – and where the contested object has a welfare of its own, a child, a holy site, the half-slave, the clause gives way to a court.
A decision-maker can be better served by bias than by neutrality: the Talmud's zabla panel (Sanhedrin 3:1) as the verifiable-disclosure micro-foundation of the party-appointed arbitrator – each side's judge certifies what it can prove, and between two opposed custodians the state is revealed.
The rule that a unanimous conviction acquits (Sanhedrin 17a) as the optimal response to a capturable court: because capture manufactures consensus, unanimity is its signature, and a court that refuses to convict on it deters the manipulation it cannot see.
The Mishnah's cap on ransoming captives (Gittin 45a) as a community binding itself against its own mercy: a cap, not a ban, denies the captor his rent, and the dispute over whether a man may redeem himself for whatever he wishes identifies which externality – the common purse or the next abduction – the rule was built to correct.
A test of Hirschman's exit–voice–loyalty logic on 134,000 individual votes across 28 English constituencies, 1832–1868: the exit of a favoured candidate drives party switching, an effect that fades as durable partisan loyalty consolidates.
Linking individual voting records to social class and mobility in nineteenth-century England, the paper asks how a voter's class origin, and movement between classes, shaped partisan choice as the franchise widened and industry remade the electorate.
Department of Government, London School of Economics, Houghton Street, London WC2A 2AE.
Email: t.dewan@lse.ac.uk