Document History

Original Publish Date: 18 May, 2020

Updated on: 03:08 PM – 06 August, 2021

This is a refined version of the old models page. In this page we have only included regression on the UP data specifically focusing on the formation of political family. We explore the question of who gets to form a political family on using a few different data sets for different years. This has been done for variety of reasons.

  • To ensure that our significant variables stays significant in different time settings.

  • One of the caveats of our dynasty data is that, many of the variables are static and collected as recent as 2017 whereas our dynasty data goes back to 1970s. So, technically we are using their latest data point even to asses the possibility of family formation that might have happened long before that. We provide a solution to this by a few models that are limited to the post delimitation years using dynamic data provided by ADR.

Regressions

UP

AE - GE all

In this model we consider winners and runner-ups in the elections held between 197x to 2019 in Uttar Pradesh.

\[{Political \ family} = \alpha +\beta_{1}Minister + \beta_{2} Economic \ diversification + \beta_{3}Land \ ownership + \\ \beta_{3} Election \ won \ by \ first\ member + \beta_{3}Female \ first member + Z + \epsilon \]

Where the outcome of interest \({Political \ family}\) is a dummy variable that is 1 if the family have been successful in forming a political family and 0 otherwise; \(beta_{1}Minister\) is another dummy variable that is 1 if there was at least one minister from that family in the state cabinet; \(\beta_{2} Economic \ diversification\) represents the number of industries and educational institutions owned by that particular family; \(\beta_{3}Land \ ownership\) is continuous variable that denotes the land owned by a family; \(\beta_{3} Election \ won \ by \ first\ member\) is continuous variable that is the number of elections won by the first member of the family; \(\beta_{3} Election \ won \ by \ first\ member\) is dummy variable that is one if the first member of the family was a female 0 otherwise; \({Z}\) is the constituency and candidate level controls; \({\epsilon }\) is the error term clustered at the constituency level

Summary of the AE GE all contestants unique data
Measure Family Minister(Dummy) Economic diversification Land ownership (quartiles) Elections won by first family member Female first member
Length 4867.00 4867.0 4867.00 4867.00 4867.00 4867.00
Unique-N 2.00 2.0 5.00 5.00 11.00 2.00
Mean 0.07 0.1 0.77 2.11 1.04 0.05
SD 0.25 0.3 1.02 1.20 1.29 0.23
Min 0.00 0.0 0.00 1.00 0.00 0.00
Max 1.00 1.0 4.00 5.00 10.00 1.00
Regression Results - AE-GE All
Dependent variable:
Political Family
(1) (2) (3) (4)
Political Diversification
(only cabinet) 		0.126*** 0.126*** 0.127*** 0.127***
(0.019) (0.018) (0.019) (0.018)
Economic diversification 0.045*** 0.036*** 0.044*** 0.035***
(0.007) (0.006) (0.007) (0.006)
Land ownership 0.023*** 0.015*** 0.024*** 0.016***
(0.005) (0.005) (0.005) (0.005)
Elections won by first member 0.003 0.0003 0.003 0.0001
(0.005) (0.005) (0.005) (0.005)
First member - female -0.022* -0.036*** -0.023* -0.038***
(0.012) (0.012) (0.012) (0.012)
Fixed effects Decade Sub-region Deacde and Sub-region
Clustered SE PC PC PC PC
Caste control Yes Yes Yes Yes
Observations 4,867 4,867 4,867 4,867
R2 0.121 0.148 0.123 0.150
Adjusted R2 0.119 0.146 0.120 0.147
Residual Std. Error 0.233 (df = 4856) 0.230 (df = 4853) 0.233 (df = 4849) 0.230 (df = 4846)
Note: p<0.1; p<0.05; p<0.01

AE GE All - Won elections

This model is run only among entities who have at least one an election.

Regression Results - AE-GE All
Dependent variable:
Political Family
(1) (2) (3) (4)
Political Diversification
(only cabinet) 		0.124*** 0.123*** 0.125*** 0.125***
(0.020) (0.018) (0.020) (0.019)
Economic diversification 0.056*** 0.043*** 0.054*** 0.042***
(0.008) (0.008) (0.008) (0.008)
Land ownership -0.015** -0.021*** -0.015** -0.021***
(0.007) (0.007) (0.007) (0.007)
Elections won by first member -0.034* -0.049** -0.035* -0.051***
(0.019) (0.020) (0.019) (0.020)
First member - female 0.031*** 0.022*** 0.033*** 0.023***
(0.007) (0.007) (0.006) (0.006)
Fixed effects Decade Sub-region Deacde and Sub-region
Clustered SE PC PC PC PC
Caste control Yes Yes Yes Yes
Observations 2,943 2,943 2,943 2,943
R2 0.119 0.164 0.123 0.168
Adjusted R2 0.116 0.160 0.118 0.162
Residual Std. Error 0.288 (df = 2932) 0.281 (df = 2929) 0.288 (df = 2925) 0.280 (df = 2922)
Note: p<0.1; p<0.05; p<0.01

AE-GE 2009:19

This is run among the elections that happened post delimitation, limited to winner and the runner-up

  • the dataframe in the

This regression run amongst all the contestants in the general elections held during 2009:19 period.

Summary of the GE 2009:19 contestants unique data
Measure Family Minister dummy Economic diversification Land ownership(quartiles) Elections won by first member First member female Total assets (in millions) Serious crime Turncoat Incumbent
Length 2092.00 2092.0 2092.00 2092.00 2092.00 2092.00 2092.00 2092.00 2092.00 2092.00
Unique-N 2.00 2.0 5.00 5.00 8.00 2.00 2017.00 2.00 2.00 2.00
Mean 0.23 0.2 1.49 2.94 0.17 0.07 50.61 0.26 0.12 0.23
SD 0.42 0.4 1.13 1.11 0.70 0.26 87.37 0.44 0.33 0.42
Min 0.00 0.0 0.00 1.00 0.00 0.00 0.50 0.00 0.00 0.00
Max 1.00 1.0 4.00 5.00 8.00 1.00 561.97 1.00 1.00 1.00
Regression Results - AE-GE 09:19
Dependent variable:
Political Family
(1) (2) (3) (4)
Total assets (log) 0.045*** 0.043*** 0.045*** 0.045***
(0.008) (0.008) (0.008) (0.008)
Serious crime -0.034 -0.044* -0.035 -0.044*
(0.024) (0.023) (0.024) (0.023)
Turncoat 0.033 0.036 0.030 0.031
(0.031) (0.031) (0.030) (0.030)
Fixed effects Year Sub-region Year and Sub-region
Clustered SE PC PC PC PC
Caste control Yes Yes Yes Yes
Observations 2,059 2,059 2,059 2,059
R2 0.052 0.062 0.066 0.074
Adjusted R2 0.048 0.057 0.059 0.065
Residual Std. Error 0.412 (df = 2050) 0.410 (df = 2046) 0.409 (df = 2043) 0.408 (df = 2039)
Note: p<0.1; p<0.05; p<0.01

AE-GE 2009 :19 elections won

This model is run among the contestants who have won at lease one election in their political career

Regression Results - AE-GE 09:19
Dependent variable:
Political Family
(1) (2) (3) (4)
Minister 0.047*** 0.047*** 0.047*** 0.047***
(0.010) (0.010) (0.010) (0.010)
Economic diversification -0.034 -0.034 -0.034 -0.034
(0.028) (0.028) (0.028) (0.028)
Land ownership 0.016 0.016 0.016 0.016
(0.034) (0.034) (0.034) (0.034)
Fixed effects Year Sub-region Year and Sub-region
Clustered SE PC PC PC PC
Caste control Yes Yes Yes Yes
Observations 1,641 1,641 1,641 1,641
R2 0.071 0.071 0.071 0.071
Adjusted R2 0.063 0.063 0.063 0.063
Residual Std. Error (df = 1625) 0.438 0.438 0.438 0.438
Note: p<0.1; p<0.05; p<0.01

5 PC

Var1 Freq
2009 163
2012 1613
2014 279
2017 1296
2019 190
dynast turncoat recontest incumbent
0.25 0.12 0.35 0.17
Regression Results - AE-GE 09:19
Dependent variable:
Political Family
(1) (2) (3) (4)
log(assets) 0.043*** 0.046*** 0.044*** 0.047***
(0.006) (0.007) (0.006) (0.007)
serious_crime 0.019 0.015 0.019 0.015
(0.020) (0.020) (0.019) (0.019)
turncoat 0.022 0.023 0.021 0.022
(0.025) (0.025) (0.025) (0.025)
Fixed effects Year Sub-region Year and Sub-region
Clustered SE PC PC PC PC
Caste control Yes Yes Yes Yes
Observations 2,909 2,909 2,909 2,909
R2 0.031 0.033 0.038 0.039
Adjusted R2 0.029 0.030 0.034 0.035
Residual Std. Error 0.405 (df = 2901) 0.405 (df = 2900) 0.404 (df = 2895) 0.404 (df = 2894)
Note: p<0.1; p<0.05; p<0.01