Introduction
Democratization has suffered more reversals in 2021, with the percentage of people living in a democracy falling to well below 50% and authoritarian regimes gaining ground. The project focuses on the impact of crime and democracy using the specific variable “Rule of Law (RL)” as a measure. The data goes back from 1996 to 2021 and has various countries. However, there will be a specific obersvational study on Latin America and the Caribbean and a possible cross comparison to different regions of the world and their relationship to crime and democracy. The data contains over 5700 observations for countries as it relates to their democracy and crime index, and about 900 of those data points are focused on Latin America and the Caribbean.
In Latin America and the Caribbean countries, does crime have an influence democratic institutions? This study conducts a bivariate analysis of the relationship between democratic institutions and crime within Latin America and the Caribbean. I hypothesize that crime impacts their democratic institutions and that crime has increased over the years therefore as crime increase, democracy in the Caribbean decreases. Although I am using this data for the purposes of my final project it is imperative to point out that the data collected are subject to several biases, including under representation of the developing world, asymmetries in reporting countries’ definitions of homicide, and those imposed by difficulties in collecting and compiling reliable data in nations with limited statistical and organizational capacity. The explanatory variable of interest is the democratic institutions or overall democratic index of each country and the outcome variable is the rate of crime rates in Latin America and the Caribbean.
Data and Research Design
The EIU Democracy Index provides a snapshot of the state of world democracy for 165 independent states and two territories. The Democracy Index is based on five categories: electoral process and pluralism, civil liberties, the functioning of government, political participation, and political culture. To analyze the relationship between crime and democracy, I used data complied over the years from the Economic Intelligence Unit (EIU):
Our research design is observational analysis on the variables of crime (RL) on the democratic index (VA) within Caribbean countries. VA stands for “Voice and Accountability” within our data and is calculated based on a country’s democracy index, vested interests, accountability of public officials, human rights, and freedom of association. RL stands for “Rule of Law” and is calculated based on a country’s violent crime, organized crime rate, fairness of judicial process, enforceability of contracts, speediness of judicial process, confiscation/expropriation, intellectual property rights protection, and private property protection.
The RL variable is the outcome variable. A countries rule of law index was calculated on a scale of 0 -4 and it took into consideration questions such as “Are violent demonstrations or violent civil/labor unrest likely to pose a threat to property or the conduct of business? Is violent crime likely to pose a significant problem for government and/or business?” Where a score of 0 = rule of law/crime is not a problem, and 4 = rule of law/crime is a very significant problem. The outcome ranges from 0 to 4, with a median of .45 - .50. RL appears to be relatively right skewed in distribution, ranging from 0 to 1 and a center slightly greater than 0.5.
The VA variable is the explanatory variable. The VA variable was calculated on a scale of 0 -4 and it took into consideration questions such as “How accountable are public officials? Is recourse possible in the case of unfair treatment do safeguards/sanctions exist to ensure to ensure officials perform competently? Is there a risk that this country could be accused of serious human rights abuses?” Where a score of 0 = very low likelihood of this country being accused of human rights abuses, and 4 = very high likelihood of this country being accused of human rights abuses. The indicators range from 0 through 4 (bad), with the median being approximately .56. VA appears to be relatively right skewed in distribution, ranging from 0 to 1 and a center slightly greater than 0.5.
Results
Before regressing the “Rule of Law” variable outcome on the “Voice of Accountability” variable, the bivariate relationship between these variables can be visualized with a scatter plot.
The bivariate plot suggests that there exists a positive relationship between democratic index and rule of law/crime index. The higher the rule of law, the stronger enforcement against crime in the country.We see that countries that are considered more democratic generally are associated with stricter laws on crime. Regression analyses will offer more detailed information about this seemingly positive association.
Regression Results
Results from the regression model is below. The bivariate model, which regresses rule of law on the democratic index.
Regression results show that a statistically significant, positive correlation exists between a country’s rule of law and their democratic index.
# multiple regression
lm_multiple <- lm(RL ~ VA + PV + GE + RQ + RL + CC, data=caribbean_eiu)
summary(lm_multiple)$coefficients |>
knitr::kable(digits=4)| Estimate | Std. Error | t value | Pr(>|t|) | |
|---|---|---|---|---|
| (Intercept) | -0.0217 | 0.0164 | -1.3230 | 0.1864 |
| VA | 0.3381 | 0.0361 | 9.3760 | 0.0000 |
| PV | 0.1137 | 0.0344 | 3.3074 | 0.0010 |
| GE | 0.1234 | 0.0302 | 4.0834 | 0.0001 |
| RQ | 0.2668 | 0.0260 | 10.2657 | 0.0000 |
| CC | 0.1473 | 0.0236 | 6.2507 | 0.0000 |
Even though this is a multivariate model, I want to look specifically at the VA variable, which is the democratic index on the rule of law and which is statistically significant because because their p-values equal 0.000. When we keep all other factors constant, for every 1 unit increase in the VA variable we get a .338 increase in the RL intercept.
Statistical Significance
In the multivariate model, the association between RL and VA is statistically significant at the test level α=0.05. Specifically, the p-value on the coefficient is 0. If hypothesizing that the association between VA and RL was non-zero, then the coefficient’s p-value of 0 would suggest there is a near 0% probability that the estimated association observed here is as extreme and meaningless as any estimated association from the null distribution. On the other hand, there is a near 100% probability that the estimated association is not due to random chance, because it is so dissimilar to what the estimated association would look like in the null distribution.
Causality and Confounders
There are many factors that are additionally taken into consideration when looking at the bivariate relationship between a nation’s democracy and crime. However, I specifically chose to isolate otuside variables to have a deeper look at how crime, exactly, alter with a nation’s democracy. When you take about crime within a country, tangentially the conversation of economics should come into play. Therefore, environmental confounders like poverty level within a country, unemployment rates, and the impact of education should be taken into consideration.
Conclusion
The analysis only takes into crime which does not account for various confounding variables like poverty, education, and the unemployment rate of the country. Although the data contains wide data points on all nations’ in the world’s rule of law rate and democracy index, I thought it be more pertinent to look into Latin America and the Caribbean because these are countries that have stable democracies but, with high levels of crime embedded within their society. Typically, when you analyze democracy within a nation, predominantly when speaking about Western democracies, they have high enforcements on crime which shows in their democratic index and overall democratic score. However, with nations like Jamaica, Trinidad and Tobago, Barbados, and etc, these nations have seen high levels of crime but have remained stable and strong democracies therefore, I wanted to see the results for myself – both as someone interested in political science and a native Caribbean immigrant. Potential next steps could be to compare Latin America and the Caribbean to the different regions of the world. This way we could see how the Caribbean positions themselves to other areas of the world.
The regression results demonstrate that as the rule of law increases, as countries are more authoritative on crime it positively impacts their democratic index and significantly associated with increases in their democracy index (or specifically the VA variable). A multivariate model that controls for possibly influential covariates and confounders produces an even larger estimate of this association.