Council paper
EA
2022-09-14
USE change as DV : change_pr_limit_meaningf
mean probability of change indicating LIMITATION by the Council” (using only meaningful provisions
##
## Table 1: Change as a DV
## ===========================================================================================================
## Dependent variable:
## ---------------------------------------
## change_pr_limit_meaningf
## (1) (2) (3) (4)
## -----------------------------------------------------------------------------------------------------------
## PO_oppose_change -0.296** -1.263*** -0.190 -1.295***
## (0.115) (0.234) (0.162) (0.443)
##
## wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.552*** 0.554*** 0.556*** 0.606***
## (0.090) (0.090) (0.090) (0.093)
##
## cntr_polarization_t_trilogue_imputed 0.245*** 0.311*** 0.246*** 0.231***
## (0.083) (0.084) (0.083) (0.082)
##
## weighted.mean_EU_position_lowe -0.075*** -0.073*** -0.073*** -0.073***
## (0.013) (0.013) (0.013) (0.013)
##
## n_comm_consulted -0.016** -0.016** -0.016** -0.015**
## (0.007) (0.007) (0.007) (0.007)
##
## NEW_LR_Proposal_Probability 0.676*** 0.668*** 0.674*** 0.687***
## (0.051) (0.051) (0.051) (0.051)
##
## factor(as.character(form))Directive 0.377*** 0.354*** 0.376*** 0.376***
## (0.062) (0.062) (0.062) (0.062)
##
## factor(as.character(form))Regulation 0.134** 0.102* 0.132** 0.135**
## (0.060) (0.061) (0.060) (0.060)
##
## PO_oppose_change:cntr_polarization_t_trilogue_imputed 1.890***
## (0.398)
##
## PO_oppose_change:weighted.mean_EU_position_lowe 0.061
## (0.065)
##
## PO_oppose_change:wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 1.202**
## (0.515)
##
## Constant -2.947*** -2.981*** -2.946*** -2.989***
## (0.135) (0.135) (0.135) (0.136)
##
## -----------------------------------------------------------------------------------------------------------
## Country FE? Yes Yes Yes Yes
## -----------------------------------------------------------------------------------------------------------
## Observations 5,427 5,427 5,427 5,427
## ===========================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01Using the probability of the Council limiting the proposal as the main DV (Mean probability of the CATEGORY 0 for meaningful provisions)
Model set up:
dv: cn_sq_ratio
IV: PO_oppose_change wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed cntr_polarization_t_trilogue_imputed weighted.mean_EU_position_lowe n_comm_consulted NEW_LR_Proposal_Probability +
FE: Form, country_name
Method: GLM quasibinom
plot the DV
dv<- dt %>%ggplot(aes(cn_sq_ratio) ) +
geom_histogram( alpha=0.4, binwidth = .008, color="#e9ecef", fill="#377EB8", position = 'identity') +
theme_minimal()+
labs(title="",
x="Mean Pr. of Council's position maintaining the SQ",
y="Count")
dv## Warning: Removed 250 rows containing non-finite values (stat_bin).
ggsave("DV.png", plot=dv, width = 12, height = 12, units = "cm" )## Warning: Removed 250 rows containing non-finite values (stat_bin).##
## Table 2: DV= Mean Probability of the Council maintaining the SQ in EU authoirty
## ===========================================================================================================
## Dependent variable:
## ---------------------------------------
## cn_sq_ratio
## (1) (2) (3) (4)
## -----------------------------------------------------------------------------------------------------------
## PO_oppose_change 0.367*** 1.092*** 0.260** 0.312
## (0.096) (0.203) (0.130) (0.381)
##
## wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.550*** 0.544*** 0.546*** 0.552***
## (0.073) (0.073) (0.073) (0.074)
##
## cntr_polarization_t_trilogue_imputed -0.138** -0.178*** -0.140** -0.138**
## (0.069) (0.069) (0.069) (0.069)
##
## weighted.mean_EU_position_lowe 0.021** 0.020* 0.020* 0.021**
## (0.010) (0.010) (0.010) (0.010)
##
## n_comm_consulted -0.013** -0.014** -0.013** -0.013**
## (0.006) (0.006) (0.006) (0.006)
##
## NEW_LR_Proposal_Probability 0.086** 0.091** 0.087** 0.087**
## (0.042) (0.042) (0.042) (0.042)
##
## factor(as.character(form))Directive -0.076 -0.055 -0.076 -0.076
## (0.050) (0.050) (0.050) (0.050)
##
## factor(as.character(form))Regulation -0.103** -0.077 -0.101** -0.102**
## (0.048) (0.048) (0.048) (0.048)
##
## PO_oppose_change:cntr_polarization_t_trilogue_imputed -1.434***
## (0.354)
##
## PO_oppose_change:weighted.mean_EU_position_lowe -0.065
## (0.053)
##
## PO_oppose_change:wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.066
## (0.441)
##
## Constant -1.617*** -1.595*** -1.616*** -1.618***
## (0.107) (0.107) (0.107) (0.107)
##
## -----------------------------------------------------------------------------------------------------------
## Country FE? Yes Yes Yes Yes
## -----------------------------------------------------------------------------------------------------------
## Observations 5,428 5,428 5,428 5,428
## ===========================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
## Additing plots for individual effects of polarization and
salience
predict_sal <- ggpredict(m1_c, terms = "wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed[all]")
figureH1_sal= ggplot(predict_sal, aes(x=x, y=predicted )) +
geom_line(stat = "identity" , color="#377EB8", linetype="solid") +
geom_ribbon(aes(ymin = conf.low, ymax = conf.high) , fill="#377EB8" , color="#377EB8" , linetype=1, alpha = .2)+
labs(title = NULL, x = "Salience of the EU action in policy area", y = "Predicted Pr. of CN keeping SQ",
linetype="Salience of the EU action in policy area") +
theme_minimal() + theme(legend.position="bottom",
legend.title=element_text(size=14),
legend.text=element_text(size=14)) +
theme( axis.text.y = element_text(size = 12),
axis.title.x = element_text(size = 15),
axis.title.y = element_text(size = 15),
axis.text.x = element_text(size = 12)) +
theme(legend.position="none")
#figureH1=figureH1+geom_rug( sides="b", color="#377EB8")
figureH1_sal
ggsave("H1_sal.png", plot = figureH1_sal, width = 12, height = 12, units = "cm") Individual effect of polarization
predict_polariz <- ggpredict(m1_c, terms = "cntr_polarization_t_trilogue_imputed[all]")
figureH1_polariz= ggplot(predict_polariz , aes(x=x, y=predicted )) +
geom_line(stat = "identity" , color="#4DAF4A", linetype="solid") +
geom_ribbon(aes(ymin = conf.low, ymax = conf.high) , fill="#4DAF4A" , color="#4DAF4A" , linetype=1, alpha = .2)+
labs(title = NULL, x = "Public polarization in the policy area", y = "Predicted Pr. of CN keeping SQ",
linetype="Public polarization in the policy area") +
theme_minimal() + theme(legend.position="bottom",
legend.title=element_text(size=14),
legend.text=element_text(size=14)) +
theme( axis.text.y = element_text(size = 12),
axis.title.x = element_text(size = 15),
axis.title.y = element_text(size = 15),
axis.text.x = element_text(size = 12)) +
theme(legend.position="none")
figureH1_polariz
ggsave("H1_polarizl.png", plot = figureH1_polariz, width = 12, height = 12, units = "cm") Interpretation:
When the public becomes highly critical of the EU engagment in the policy area by the time of negotitiations start, the council is less likely to adhere to the SQ if the public is divided; BUT They will listen and maintain the SQ when the public is NOT polarized and critical of the EU.
plotting interaction of polarization and public opinion at level different from extremes
# get the summary
summary(dt$cntr_polarization_t_trilogue_imputed)## Min. 1st Qu. Median Mean 3rd Qu. Max.
## 0.0000 0.3479 0.4955 0.4962 0.6463 1.0000mean(dt$cntr_polarization_t_trilogue_imputed) # 0.4961652## [1] 0.4961652mean_plus_sd=round(mean(dt$cntr_polarization_t_trilogue_imputed)+sd(dt$cntr_polarization_t_trilogue_imputed), 2)
mean_min_sd=round(mean(dt$cntr_polarization_t_trilogue_imputed)-sd(dt$cntr_polarization_t_trilogue_imputed), 2)
# predict
predict_m2_polar_SD= ggpredict(m2_c, terms = c("PO_oppose_change ", "cntr_polarization_t_trilogue_imputed[.28, 0.71]"))## Data were 'prettified'. Consider using `terms="PO_oppose_change [all]"` to get smooth plots.graph1_m2c_sd =ggplot(predict_m2_polar_SD , aes(x=x, y=predicted, group=group)) +
geom_line( aes(linetype=group, color=group))+
geom_ribbon(aes(ymin = conf.low, ymax = conf.high, fill=group , color=group, linetype=group) ,
alpha = .2)+
scale_linetype_manual(values=c("solid", "dashed"), name="mean +/- 1SD CNTR Polarization at t of trilogues") +
scale_color_manual(values=c("#E41A1C" ,"#377EB8" ), name="mean +/- 1SD CNTR Polarization at t of trilogues") +
scale_fill_manual(values=c("#E41A1C" ,"#377EB8" ), name="mean +/- 1SD CNTR Polarization at t of trilogues") +
#scale_y_continuous( breaks = c(0.1, 0.3, 0.5, 0.7, 0.9),
# labels = c("0.1","0.3","0.5", "0.7", "0.9"), limits = c(0.1, 0.98))+
labs( title = NULL,
x = "Change in Public opposition",
y = "mean +/- 1SD CNTR Polarization at t of trilogues",
linetype="mean +/- 1SD CNTR Polarization at t of trilogues") +
theme_minimal()+
theme(legend.position="bottom", legend.title=element_text(size=14),
legend.text=element_text(size=14))
graph1_m2c_sd 
ggsave("H2a_mean_SD.png", plot = graph1_m2c_sd , width = 12, height = 12, units = "cm") interpretation:
The Council position is more likely to respond to the changes in public preferences over EU policy action (between the time proposals was initiated by the Commission and the time when the negotiations started,) in the Member States governed by pro-EU governments than in the Member States governed by Euroskeptic governments
Interpretation: CHANGES IN EFFCET DUE TO CHANGE IN SCALING
OLD RESULT: Likelihood of the Council adhering to the SQ position increases for non-salient issues as the opposition to the EU action in policy domain increases from the time of proposals till the time of negotiations.
NEW RESULT (HOLDS WITH THE SCALING from 16092022):: The Council is more likly to adhere to the SQ position whent he issue is SALIENT and the opposition to the EU action in policy domain increases from the time of proposals till the time of negotiations (in comparison to the cases when the population becomes more supportive of the EU action on salient issues). There is no effect of public opinion for non salient issues
adding parliamentary control as a control variable
pr_cont=read_csv("C:/Users/nasta/Downloads/NPdata_version2020_v1.csv")## Rows: 915 Columns: 15
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (3): country, ctr, notes
## dbl (12): year, memo, eac, seccom, reserve, mandate, information, constraint...
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.table(pr_cont$year)##
## 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973
## 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 9
## 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989
## 9 9 9 9 9 9 9 10 10 10 10 10 12 12 12 12
## 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005
## 12 12 12 12 12 15 15 15 15 15 15 15 15 15 25 25
## 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020
## 25 27 27 27 27 27 27 28 28 28 28 28 28 28 27table(dt$country_name)##
## Austria Belgium Bulgaria Croatia Cyprus
## 208 208 208 159 208
## Czech Republic Denmark Estonia Finland France
## 208 208 208 208 208
## Germany Greece Hungary Ireland Italy
## 208 208 208 208 208
## Latvia Lithuania Luxembourg Malta Netherlands
## 208 208 208 208 208
## Poland Portugal Romania Slovakia Slovenia
## 208 208 208 208 208
## Spain Sweden United Kingdom
## 208 208 208table(dt$year_trilogue)##
## 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
## 108 189 594 1020 728 364 532 868 868 504pr_cont=pr_cont%>% select(country, year, oversight) %>% filter(year>=2004)
#### merge
dt=left_join(dt, pr_cont, by=c("year_trilogue"="year", "country_name"="country"))rerun the models with the praliemantary control included (Winzen, 2021. Avaiable at: https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/1BARB6 )
##
## Table 3: DV= controllling for parliamentary control
## =====================================================================================================================
## Dependent variable:
## -------------------------------------------------
## cn_sq_ratio
## (1) (2) (3) (4) (5)
## ---------------------------------------------------------------------------------------------------------------------
## PO_oppose_change 0.367*** 1.092*** 0.260** 0.311 0.334
## (0.096) (0.203) (0.130) (0.381) (0.277)
##
## oversight 0.076 0.101 0.070 0.079 0.076
## (0.881) (0.877) (0.880) (0.881) (0.881)
##
## wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.550*** 0.544*** 0.546*** 0.552*** 0.550***
## (0.073) (0.073) (0.073) (0.074) (0.073)
##
## cntr_polarization_t_trilogue_imputed -0.138** -0.178** -0.140** -0.138** -0.138**
## (0.069) (0.069) (0.069) (0.069) (0.069)
##
## weighted.mean_EU_position_lowe 0.021** 0.020* 0.020* 0.021** 0.021**
## (0.010) (0.010) (0.010) (0.010) (0.010)
##
## n_comm_consulted -0.013** -0.014** -0.013** -0.013** -0.013**
## (0.006) (0.006) (0.006) (0.006) (0.006)
##
## NEW_LR_Proposal_Probability 0.086** 0.091** 0.087** 0.087** 0.086**
## (0.042) (0.042) (0.042) (0.042) (0.042)
##
## factor(as.character(form))Directive -0.076 -0.055 -0.076 -0.076 -0.076
## (0.050) (0.050) (0.050) (0.050) (0.050)
##
## factor(as.character(form))Regulation -0.103** -0.077 -0.101** -0.103** -0.103**
## (0.048) (0.048) (0.048) (0.048) (0.048)
##
## PO_oppose_change:cntr_polarization_t_trilogue_imputed -1.435***
## (0.354)
##
## PO_oppose_change:weighted.mean_EU_position_lowe -0.064
## (0.053)
##
## PO_oppose_change:wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.067
## (0.441)
##
## PO_oppose_change:oversight 0.062
## (0.487)
##
## Constant -1.655*** -1.646*** -1.651*** -1.658*** -1.655***
## (0.455) (0.454) (0.455) (0.456) (0.455)
##
## ---------------------------------------------------------------------------------------------------------------------
## Country FE? Yes Yes Yes Yes Yes
## ---------------------------------------------------------------------------------------------------------------------
## Observations 5,428 5,428 5,428 5,428 5,428
## =====================================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01Parliamentary control has no effects
add indicator for elections: Checking the effects of trilogue_elections ( days between trilogue and the upcoming elections)
##
## Table 4: Additing elections into the model: Days from TRI to elections
## =====================================================================================================================
## Dependent variable:
## -------------------------------------------------
## cn_sq_ratio
## (1) (2) (3) (4) (5)
## ---------------------------------------------------------------------------------------------------------------------
## PO_oppose_change 0.367*** 1.085*** 0.259** 0.313 0.361**
## (0.096) (0.203) (0.130) (0.381) (0.172)
##
## tri_elect -0.00003 -0.00003 -0.00003 -0.00003 -0.00003
## (0.00003) (0.00003) (0.00003) (0.00003) (0.00003)
##
## wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.551*** 0.545*** 0.547*** 0.552*** 0.551***
## (0.073) (0.073) (0.073) (0.074) (0.073)
##
## cntr_polarization_t_trilogue_imputed -0.138** -0.178*** -0.140** -0.139** -0.138**
## (0.069) (0.069) (0.069) (0.069) (0.069)
##
## weighted.mean_EU_position_lowe 0.022** 0.020* 0.021** 0.022** 0.022**
## (0.010) (0.010) (0.010) (0.010) (0.010)
##
## n_comm_consulted -0.013** -0.014** -0.013** -0.013** -0.013**
## (0.006) (0.006) (0.006) (0.006) (0.006)
##
## NEW_LR_Proposal_Probability 0.087** 0.091** 0.088** 0.087** 0.087**
## (0.042) (0.042) (0.042) (0.042) (0.042)
##
## factor(as.character(form))Directive -0.076 -0.055 -0.076 -0.076 -0.076
## (0.050) (0.050) (0.050) (0.050) (0.050)
##
## factor(as.character(form))Regulation -0.102** -0.077 -0.101** -0.102** -0.102**
## (0.048) (0.048) (0.048) (0.048) (0.048)
##
## PO_oppose_change:cntr_polarization_t_trilogue_imputed -1.422***
## (0.354)
##
## PO_oppose_change:weighted.mean_EU_position_lowe -0.065
## (0.053)
##
## PO_oppose_change:wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.064
## (0.441)
##
## PO_oppose_change:tri_elect 0.00001
## (0.0002)
##
## Constant -1.599*** -1.580*** -1.599*** -1.600*** -1.599***
## (0.108) (0.108) (0.108) (0.109) (0.108)
##
## ---------------------------------------------------------------------------------------------------------------------
## Country FE? Yes Yes Yes Yes Yes
## ---------------------------------------------------------------------------------------------------------------------
## Observations 5,428 5,428 5,428 5,428 5,428
## =====================================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
checking the effect of proposal_election variable
Neither on its own nor in the interaction the election proximity matters
##
## Table 5: Additing elections into the model: days b/n proposal and elections
## =====================================================================================================================
## Dependent variable:
## -------------------------------------------------
## cn_sq_ratio
## (1) (2) (3) (4) (5)
## ---------------------------------------------------------------------------------------------------------------------
## PO_oppose_change 0.367*** 1.091*** 0.260** 0.312 0.347**
## (0.096) (0.203) (0.130) (0.381) (0.168)
##
## prop_elect -0.00001 -0.00001 -0.00001 -0.00001 -0.00001
## (0.00003) (0.00003) (0.00003) (0.00003) (0.00003)
##
## wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.550*** 0.544*** 0.546*** 0.551*** 0.550***
## (0.073) (0.073) (0.073) (0.074) (0.073)
##
## cntr_polarization_t_trilogue_imputed -0.139** -0.179*** -0.140** -0.139** -0.138**
## (0.069) (0.069) (0.069) (0.069) (0.069)
##
## weighted.mean_EU_position_lowe 0.021** 0.020* 0.020* 0.021** 0.021**
## (0.010) (0.010) (0.010) (0.010) (0.010)
##
## n_comm_consulted -0.013** -0.014** -0.013** -0.013** -0.013**
## (0.006) (0.006) (0.006) (0.006) (0.006)
##
## NEW_LR_Proposal_Probability 0.086** 0.091** 0.088** 0.087** 0.086**
## (0.042) (0.042) (0.042) (0.042) (0.042)
##
## factor(as.character(form))Directive -0.076 -0.055 -0.076 -0.076 -0.076
## (0.050) (0.050) (0.050) (0.050) (0.050)
##
## factor(as.character(form))Regulation -0.102** -0.077 -0.101** -0.102** -0.102**
## (0.048) (0.048) (0.048) (0.048) (0.048)
##
## PO_oppose_change:cntr_polarization_t_trilogue_imputed -1.433***
## (0.354)
##
## PO_oppose_change:weighted.mean_EU_position_lowe -0.064
## (0.053)
##
## PO_oppose_change:wmean_Salience_DK_cntr_cap_year_t_trilogue_imputed 0.066
## (0.441)
##
## PO_oppose_change:prop_elect 0.00003
## (0.0002)
##
## Constant -1.611*** -1.590*** -1.611*** -1.612*** -1.611***
## (0.108) (0.108) (0.108) (0.109) (0.108)
##
## ---------------------------------------------------------------------------------------------------------------------
## Country FE? Yes Yes Yes Yes Yes
## ---------------------------------------------------------------------------------------------------------------------
## Observations 5,428 5,428 5,428 5,428 5,428
## =====================================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01