Lobbying And Vote Buying (Social Planner, Partial FF, Lobbying And Buying Supermajority Model Flashcards
Last time we studied voting behaviour.
How else can voting process be influenced (2) (topics of this)
Lobbying
Vote buying
Lobby:
A group attempting to influence voting by influencing parties or voters
Thurber estimations on US lobbying:
How many people does it involve
How much is lobbying spending annually
100,000+ lobbyists
9bn annual spend
UK’s view on lobbying
They find lobbying legitimate and necessary part of democratic process, however needs regulation
UK lobbying annual spend and employment
14,000 people employed
1.7bn annually by 2007
(100k+, 9bn US)
Revolving door
People previously employed in government are now in lobbying
Statistic on revolving door
In US 34 of top 50 lobbyists have previously worked in gov
2 views on revolving door lobbyist importance (those who worked in gov previously)
Expertise/experience
Connections
Expertise/experience vs connections view on why those who already worked in gov are influential
B) empirical work is small, but which view is more supported
Expertise: lobbyists use experience to influence
Connections: lobbyists use who they know and their networks
B) connections view i.e networks are more valuable
Revolving door study
Looked at lobbyist revenue after their connection leaves office
If lose revenue: supports connection view, since value was tied to their connection, not necessarily their skill!
If revenue constant after connection leaves; supports expertise
Finding
24% fall in revenue after connection leaves!
And even larger effects if connection was more senior!
Supports connection view - not about skill
Caveat
It doesn’t mean expertise doesn’t matter - since variation only comes from changes in connections
Bertrand further study finds
Evidence for both expertise and connections - but connections more important
Theory of Special interests politics -
shows how lobbying influences policy
What do we normally get from special interest policies?
Policy that provides benefits to small well-defined groups (hence the name special interest), costs to large unorganised groups
E.g local public good funded by central taxation
3 cases to understand impact of a lobby
Social optimum of a utilitarian government
Partial fiscal federalism (centrally financed public good)
Lobbying
Lobbying model set up:
3 groups of citizens with equal income
Total population a=1 and groups a1 a2 a3
a1+a2+a3=1 and y1=y2=y3 = y >1
All citizens consume private good Ci and a local public good. 3 local public goods for each of the groups gi
What is utility function
Ui = Ci + ln(gi)
First case:
Social optimum of utilitarian government; recall what they do
B) maximisation problem
C) find indirect utility function, then differentiate to find intuitive result
Maximise utility of whole society
Max U= a1U1 + a2U2 + a3U3
Subject to budget constraint
a1 (g1+C1) + a2 (g2+C2) + a3 (g3+C3)
<=
y(a1+a2+a3) which = y (since recall a1+a2+a3=1)
C) sub utility function (Ui = Ci + ln(gi) ) into maximisation problem
Then FOC to get g1=g2=g3=1
Result: g1=g2=g3=1 meaning
Under social utilitarian gov,
Each citizen gets same amount of public good, independent of the size of group they’re in
2nd case: partial fiscal federalism
What happens to gi (public good) now
Public good gi quantity decided by each group separately, but funded centrally through tax rate t
I.e decided locally but financed publicly by common tax)
Budget constraint for partial fiscal federalism
B) utility max problem pg 22
C) use both to make final budget constraint
D) then get indirect utility function by using Ci
t(a1+a2+a3) = a1g1 + a2g2 + a3g3 = ty
(ty is revenue gov receive)
B) Max u = Ci + ln(gi)
C) Ci = y - ty
So sub ty from above in
Ci = y - (a1g1 + a2g2 + a3g3)
D) essentially the same, just add the ln(gi)
Max u = y - (a1g1 + a2g2 + a3g3) + ln(g)
Then FIC to find g’1 i.e the quantity of public good chosen - final expression
B) Is it > or less than case 1 utilitarian social plannerj
g’1 = 1/a₁ > 1
B) Partial fiscal federalism gets More quantity than the utilitarian social planner (g’1 > g)
So g’1>g i.e when decided locally but financed publicly (by common tax) , we spend more than optimal on the public good
Why?
Individuals do not fully internalise cost of pubic goods, and smaller groups spend more and internalise even less.
Proof by aggregate pooling comparison (aggregate pooling is total
g* = a1g1 + a2g2 + a3g3)
Social optimum g1=g2=g3 =1 so aggregate provision g* is
g* = a1g1 + a2g2 + a3g3 = 1
Partial fiscal federalism g’1 = 1/a₁ so aggregate provision g’* is
g’* = a1(1/a₁) + a2(1/a₂) + a3(1/a₃) = 3
3>1 !!! Classic common pool problem; groups overspend when burden is fully on themselves i.e financed publicly
Final case: lobbying
Now assume group 1 decides to become a lobby to influence politicians, with tax rate determined centrally in.e this group chooses contributiom, dependent on policy outcome
How do we express contribution, and
Utility of group 1 from policy (public good)
B) Difference between the 2 is known as
C1(g1)
U1(g1)
Difference between the 2 is known as the reservation utility of group 1, and given by
b1 = U1(g1) - C1(g1)
Politician have an objective function. What does it include (2)
Give expressions for them too
Social welfare α (weighted 0<a<1)
Lobby contributions (weighted 1-a)
I.e care about society, but also can be influenced by contributions from lobby…
So what is the objective function of the politician
B) utility functions for groups 2 and 3
C) important: sub utility functions into objective function and then find FOC to get….
D) important: intuition;
W(g1,g2,g3) = αU + (1-α)a1C1(g1)
Social utiltiy + utility from lobbyist contributions
C)
g’’1>1
G”2 <1
G”3 <1
D) All <1 except the lobbying group, who receive more than optimal while other groups receive less
So lobbying group receives more while others less:
When will misallocation be even worse
If lobby contributions are weighted more heavily in the politicians objective function *(i.e if a smaller and closer to 0)\
Since lobbying is (1-a) so
If a closer to 0 stronger misallocation
If a closer to 1, approach social optimum (politician cares little about lobbyist contributions)
Flaw of model
We saw partial fiscal federalism also has overspending, but we cannot compare to find which has more.
Do not know if there is aggregate overprovision as depends on individual parameters
So social utiltarian gov g1=g2=g3=1 (equal provision)
Partial fiscal federalism g’1 = 1/α₁ which is >1
Lobbying is
g’’1>1
G”2 <1
G”3 <1 i.e only lobbying group receives more (esp if α is low i.e politicians weight lobbyist contributions into their objective function more i.e can be bought…)
So lobbying distorbs allocoation of public resoiurces
Limitations of our results (3)
ignores free-rider problem
only assumed 1 group can lobby. didnt look into opposing lobbies
We also said lobbying can be legitimate and necessary (UK view), or undue influence; our models dont differentiate between
Lobbying ex-post vs ex-ante
ex-post is once in office
ex-ante is during campaign
Looking at ex-ante (campaign) contributions modelling
Contributions enter as a shock. assume more contributions means candidate is better.
some groups can contriubte a lot, some can’t. candidates assign greater weight to larger groups and groups that can contribute more
What is the equilibrium
In equilibrium; no campaign contributions from any group
But possible contributions TO opponent prevents deviation (threatens candidates money given to other candidate); thus possibility of lobbying can influence policy without need for actual contributions!
Does contributions signal quality?
Yes, more comtributions correlate with ex-post effectiveness (once in office)
Why is voter’s knowledge on contributions important
since if they know who receives the most contributions, it can signal candidate quality and vote accordingly
Previously vote buying was common
explain situation in UK pre Ballot act 1872
elite could observe voting, and try influence with bribes
Now we have secret balloting so elites can no longer observe voting.
but what is the puzzle
With the secret ballot; a voter can accept bribe and vote as they wish anyways. but we still see vote buying! WHY?
Alternate explanation for this
turnout buyiing - reward people they believe would vote a certain way, for actually turning up
(so target people who think would vote in your interest, and then reward turnout - this is more common than vote buying)
example of turnout buying (Nichter)
Democrats offered cigarettes beer and medicne to encourage turnout of the poor (who they assumed would vote for them)
i.e mobilises traditionally unmobilised groups, and create habit forming, which is why still a form of election fraud
roll call votes
b) what is important to find about this
Public, votes are known. Vote yes or no.
b) these are often passed with a supermajority. e.g landslide vote we’ll see why..
buying a supermajority form
Imagine 7 person legislature (voters); all indifferent between new proposal and status quo.
2 vote buyers, A and B
A prefers proposal (values it at Wa)
B prefers status quo (values it at Wb)
A makes offer of bribes to a set of legislators. What does B do?
How should A respond
B attacks weakest part of A’s coalition
A should thus offer same offer to all legislators. Now how should many of 7 should they bribe…
How many should A offer the same offer to? or perhaps what amount should they offer?
4 enough for majority, but B could take majority by offering an arbitrarily smaller amount
So thus A should set their offer (a) > Wb
B’s value of status quo
So A has to pay 4Wb in total to ensure loyalty of all 4
now we can prove why buying a supermajority occurs
Suppose A bribes 5 legislators. What does B do,
How does A respond
B needs to buy 2 back to make it 4-3. offer a+ε to each.
To prevent, A must set a >= 5Wb/2
so pays 5Wb/2 to ensure loyalty: KEY RESULT - PAYS LESS IF THEY BRIBE MORE LEGISLATORS
(logic carries on up till 7, B will have to buy 4 back, so A should set offer a>Wb/4 so 7Wb/4 total cost)
