Page 1
MODELING OF PUBLIC TRUST IN BASIC SOCIAL AND POLITICAL INSTITUTIONS:
A COMPARATIVE ECONOMETRIC ANALYSIS
DEMIDOVA [email protected]
15th Annual Conference of the International Society for New Institutional Economics , Stanford University, 17 June 2011
Page 2
ISNIE 2011, Stanford
The problem
photo
photo
photo
To what degree does trust in basic social and political institutions depend on the individual characteristics of citizens of different countries?
World Value Survey, 5th Wave, 2007-2008, 46 countries
Three types of models have been estimated. • The first model is based on the assumption of a generalized relationship for all
countries.• The second takes into account the heterogeneity of countries (using an
introduction of the country-level variables).• The third applies a preliminary subdivision of countries into five clusters.
Data
Models
Page 3
ISNIE 2011, Stanford
Dependent variables
photo
photo
photo
Dependent variables How much confidence do you have in… Answers
Government The armed forces 1 - A great deal; 2 – Quite a lot;
3 – Not very much; 4 – Not at all
Parliament Police
Armed_forces Government
Police Parliament
Political_parties Political parties
Justice The judicial system,
Press The press
Television Television
Page 4
ISNIE 2011, Stanford
Average level of confidence
photo
photo
photo
0 1 2 3 4
serbiazambiaburkina fasounited statesgreat britainukraineturkeytrinidad and tobagothailandswitzerlandswedenspainsouth africasloveniaviet namrussian federationromaniapolandperunorwaynetherlandsmoroccomoldovamexicomalimalaysiasouth koreajapanitalyiranindonesiaindiaghanagermanygeorgiafrancefinlandethiopiacyprustaiwanchinachilecanadabulgariabrazilaustralia
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Law enforcement institutions are the most popular among the residents of most countries
Law enforcement institutions are the
most popular among the residents of most
countries
Citizens trust the political parties and
parliament least of all
Page 5
ISNIE 2011, Stanford
Results of ordered logit models estimation
photo
photo
photo
Independent variables
Armed Forces
Police Govern-ment
Parlia-ment
Political Parties
Justice Press TV
Sex (1-M, 2-F) (+) (-) (+) (+) (-) (+)
Age (-) (+) (+) (+) (+) (+) (+)
Educmid (+) (+) (+) (+) (+) (+) (+) (+)
Educhigh (+) (+) (+) (+) (+) (+) (+) (+)
Income (-) (-) (-) (-) (-) (-) (-) (-)
Marital (-) (-) (-) (-) (-) (-) (-) (-)
Unemployed (+) (+) (+) (+) (+)
Supervisor(?) (+) (+) (+) (+) (+)
Page 6
ISNIE 2011, Stanford
Results of mixed logit models estimation
photo
photo
photo
Variables Armed Forces
Police Gov Parlia-ment
PoliticalParties
Justice Press TV
Sex Age Educmid Educhig Income
Marital
Unempl
Supervis
GDP CPI
Page 7
ISNIE 2011, Stanford
Cluster Analysis
photo
photo
photo010
2030
40L2
squa
red
diss
imila
rity
mea
sure
1 15 44 5 11 19 13 26 4 17 18 25 46 35 36 40 20 8 10 42 27 38 37 2 23 31 3 30 6 12 29 9 32 33 41 39 24 43 28 7 14 34 21 22 16 45
Dendrogram for wardslnk cluster analysis
Page 8
ISNIE 2011, Stanford
First Cluster
photo
photo
photo
0 1 2 3
viet nam
south africa
mali
malaysia
india
ghana
china
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Countries belonging to the first cluster have the highest level of
confidence
The highest level of confidence is in
Vietnam
China is slightly different from the
other cluster countries
Page 9
ISNIE 2011, Stanford
Second Cluster
photo
photo
photo
The geography of the nine countries of the
second cluster is rather various
Practically all the points are located in
the interval 2-3
In the second cluster the judicial system is
highly respected
0 1 2 3
zambia
thailand
spain
south korea
morocco
japan
iran
indonesia
burkina faso
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Page 10
ISNIE 2011, Stanford
Third Cluster
photo
photo
photo
Third cluster includes countries of the former Soviet Union, Eastern European countries
The points are shifted more to the right,
which corresponds to lower levels of trust
The ordered system of preferences: the army,
the police, the government, the parliament, the political parties
0 1 2 3 4
ukrainetrinidad and tobago
taiwanslovenia
serbiarussian federation
romaniapoland
perumoldova
mexicogeorgiaethiopia
chilebulgaria
brazil
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Page 11
ISNIE 2011, Stanford
Fourth Cluster
photo
photo
photo
The fourth cluster includes the
Scandinavian countries Switzerland, Turkey
and Cyprus
Almost all points are located, as in the
second cluster, in the interval 2-3, but the order of the points is
different
The press and television are
unpopular ii the fourth cluster
0 1 2 3
turkey
switzerland
sweden
norway
finland
cyprus
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Page 12
ISNIE 2011, Stanford
Fifth Cluster
photo
photo
photo
The fifth cluster includes the economically
developed countries
The most popular are the force institutions
and the judicial system
The government, parliament, political
parties, press and television are very
unpopular
0 1 2 3
united states
netherlands
italy
great britain
germany
france
canada
australia
mean of armed_forces mean of policemean of government mean of parliamentmean of political_parties mean of justicemean of press mean of television
Page 13
ISNIE 2011, Stanford
Results of ordered logit models for clusters
photo
photo
photo
Armed Forces Police
Govern-ment
Parlia-ment
Political_Parties Justice Press Television
SexI+ II+ III+ IV+ IV- V- IV+ V+ II- V+ II- III- I+
I+ II- IV- V-
AgeI+ II- III- IV+ V- II- III- V- III- III- V- II- III-
I+ II- III+ IV+ V+
II- III- IV- V+
II+ III-IV-
EducmidI+ III- IV+ V-
I+ II- IV+ V- I+ IV+ V-
I+ II+ IV+ V- I+ IV+ V- I+ IV+ V-
I+ II- III- V+
I+ II- III- V+
Educhigh I+ II+ IV+ I+ II+ IV+I+ II+ III+ V-
I+ II+ III+ V-
I+ II+ III+ V- I+ II+ V-
I+ II- IV-V+
I+ III+ IV+ V+
IncomeI- III- IV+ V-
I- II- III- IV+ V-
I- II- III- IV+V-
I- II- III- V-
I- II- III- V- I- II- III- V- I- III- I- III- IV-
MaritalII- III- IV- V- I- III- IV- I- IV-
I- III-IV- I- IV- I- IV- I-
I- II- III-IV-
Un-employed I+ II+ V+
I+ II+ III+ IV+ V+ V+ V+ V+
II- III+ IV+ V+ V+ II- V+
Supervisor I+ II- V- I+ II+I+ II+ III- V-
I+ II+ III- I+ III- I+ II-
I+ III+ IV+ V+
I+ III+ IV+ V+
Page 14
ISNIE 2011, Stanford
Conclusions
photo
• In modeling the degree of public confidence in basic social and political institutions, it is necessary to take into account the specific characteristics of each country.
• Marital status contributes to greater confidence in the basic institutions; therefore, it makes sense to promote family values.
• The degree of confidence in major social and political institutions increases with income, thus creating favorable conditions for the growth of welfare (a decrease in the tax burden and help with the development of small and medium enterprises, providing affordable loans), could lead to an increase in public trust.
• Reducing corruption in the country would increase the credibility of law enforcement and the judiciary.
Page 15
20, Myasnitskaya str., Moscow, Russia, 101000Tel.: +7 (495) 628-8829, Fax: +7 (495) 628-7931
[email protected]