DV3165 Development Management Chapter 2.2

Development Management
Chapter 2: Political Accountability and Public Action

Handout 2.2
 Hotelling Beach Game with 3 players (Equilibrium or disequilibrium?)
- For any initial distribution of vendors along the beach, the players will tend to converge on the
midpoint. (for the same reasons mentioned in the previous two games)
- But once all three players have reached the middle of the beach, they will not be able to find a
stable equilibrium.
- This is because the player that ends up in the middle loses his entire market share to the two
outer players.
- Hence he will relocate to a position just to the left or the right of one of the two players
surrounding him.
- But this will leave another player stuck in the middle, who will then behave likewise.
- In an abstract game such as this, with no relocation costs, this sort of competitive churning will
go on forever, until the end of time, and no stable solution will ever be found.
- Any stable solution for the three player game would entail one of the players accepting zero
market share, which none of the three will do.
 Hotelling Beach Game with 6 players
- This game has a solution that is unambiguous and stable at the positions depicted in your
figure.
- At these positions, the outermost players will capture the entire market between them and the
endpoints
- The four interior players will split the remaining market between them, where each interior
player catches the line segment between himself and the nearest dotted line (which are the
midpoints of the two interior line segments).
- This solution is somewhat counterintuitive but it provides stability.
 The Hotelling beach model has been applied frequently to the analysis of party positions in
political ‘space’
- Here the beach becomes an ideological continuum varying between left-wing and right-wing
positions. (Two player game)
- In particular, the two and three player games have been used to explain why third parties, and
third-party candidates, fare so badly in countries like the US and UK, with entrenched two-party
systems.
- In the US’s two party system: Democrats and Republicans
- In the UK’s two party system: Conservative party and the Labour party
- There is no ideological space for a third party to occupy such a system, the Hotelling model
implies.
- When a third party forms in such a system, the existing two parties will manoeuvre to take
away its market share, in this case potential voters, by relocating so as to surround it on both
sides.
- Such dynamics also explain why two-party systems always battle for the political centre.
- Neither the left nor the right-leaning political party will actually locate on the left or right of the
ideological spectrum.
- Both will locate firmly in the middle, because the middle ground is where elections are won.
- It is impressive that such a simple model can shed light on such complex real-world
phenomena as the number of parties that survive in different political systems, and the
ideological positions they take up.
- However, Hotelling’s model does not shed light on the two conditions for responsive
governments.
- Unfortunately, both uni-dimensionality and ‘single peaked-ness’, key assumptions of the
Hotelling model, are implausible if votes are to transmit useful information to politicians.
- With respect to voting, uni-dimensionality implies that voters can only care about a single
domain of their lives, such as education, transport, or defence.
- Given that in the real world, voters do care about more than a single domain, and that
elections must transmit information across a number of domains if they are to make politicians
accountable to voters, uni-dimensionality is simply unrealistic.
- And given multi-dimensional tastes, single peaked preferences are implausible.
More recent scholars, including notably Downs (1957), and Besley and Coate (1995), have
devised a variety of models of voting.
- But voting equilibria are problematic for deep reasons first recognized over two centuries ago by
the Marquis de Condorcet (French philosopher, mathematician and political scientist)
> He concluded that voting in a multi-dimensional space will lead to the possibility of cycling aka
Condorcet Cycling.
- To understand the logic behind this, consider the following experiment:

- After a long three students crave ice cream.
- They find themselves in the last ice-cream selling venue still open in London, and with only
enough money between them for one tub of ice cream.
- The vendor offers them Chocolate (C), Strawberry (S) or Vanilla (V)
- The students are named 1, 2, and 3
- They vote for which flavour to buy
- The order of their individual preferences over the three flavours are:
- There is a stable equilibrium in the situation

- It is revealed that chocolate is preferred to vanilla by a majority of two votes to one
- Now changing every student’s preferences such that they are:
- There is no stable equilibrium for this situation

- None of the three flavours is preferred by a majority and no stable equilibrium will form around
any particular flavour.
- Further explanation:
> Imagine the shopkeeper calls out ‘chocolate’ and student 1 raises her hand.
>Then the shopkeeper calls out ‘strawberry’ and student 2 raises his hand.
> When the shopkeeper calls out ‘vanilla’ student 3 raises her hand.
> The prospect of getting chocolate, which is student 2’s least favourite flavour, leads him to
propose a vanilla alliance with student 3, which student 3 accepts.
> But student 1 likes vanilla least, so she approaches student 2 and suggests a strawberry
alliance.
> At this point, student 2 breaks the vanilla alliance, and joins the strawberry alliance, which
flavour he prefers.
> But strawberry is the flavour student 3 likes least… and so on.
> In the absence of frictions and transaction costs, meaning in practical terms that the students
have nothing else to do, and do not die of hunger, these broken-coalition “elections” can go on and
on and on until the end of time without ever a stable equilibrium being reached.
- There are two distinct problems in this situation.

1. Impossibility of the existence of a stable equilibrium in multidimensional space.
2. Instrumentality, (i.e. of the voting mechanisms we commonly use as a means for bringing about
such an equilibrium.)
- A number of scholars have tried to solve one or the other of these problems, by proposing
increasingly restrictive assumptions about the sorts of preferences voters are allowed to have, or
by positing probabilistic voting mechanisms, which remove the mechanical link between voters’
beliefs and/or preferences, and their voting behavior.
- Although some progress has been made, none of these contributions has managed to solve
either problem in a way that is approximately realistic, and that allows our two conditions –
information and accountability – to be met.
- Thus democracy can produce stable equilibria – that is to say, clear results of some sort – only in
the simplest models, and under highly unrealistic assumptions, that make the transfer of
information from voters to politicians impossible.
- Relax the assumptions of these models even a little, to allow for – say – three dimensions, and
even the equilibria disappear, and the models have no solution.
- Do these results imply that democracy is a sham?

> Such conclusions, sometimes put forward by democracy’s critics, are too harsh on two counts. >
First of all, real world elections are generally one dimensional.
> We cast a vote for one party or another, or for one candidate or another, rather than expressing
detailed preferences across a menu of policy options.
> Expressing detailed preferences would much better inform our leaders as to what we want them
to do.
> But unfortunately, real-world voting mechanisms by and large do not work that way.
> Hence it is not necessarily our models of voting that are limited, but rather our real voting
systems, which the models in some fundamental respects try faithfully to represent.
> A second objection is that voting is not everything.

> What about interest groups, pressure groups, and civil society?
> What about campaigns, polling, lobbyists and individual encounters between voters and
politicians?

DV3165 Development Management Chapter 2.2

Uploaded by

Copyright:

Available Formats

DV3165 Development Management Chapter 2.2

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DV3165 Development Management Chapter 2.2

Uploaded by

Copyright:

Available Formats

Development Management

Chapter 2: Political Accountability and Public Action

 Hotelling Beach Game with 3 players (Equilibrium or disequilibrium?)

- This solution is somewhat counterintuitive but it provides stability.

- And given multi-dimensional tastes, single peaked preferences are implausible.

- To understand the logic behind this, consider the following experiment:

- There is a stable equilibrium in the situation

- There is no stable equilibrium for this situation

- There are two distinct problems in this situation.

- Do these results imply that democracy is a sham?

> A second objection is that voting is not everything.

You might also like