Thesis icon

Thesis

Critical events and ministerial turnover in Latin American presidential democracies

Abstract:

What are the causes and consequences of ministerial turnover in presidential systems? This thesis presents three articles on different theoretical and methodological aspects of this question and integrates elements of the agency theory to explain the causes of ministerial terminations and their effects on presidential approval dynamics.

The first article refers to the creation of a novel dataset on turnover in 124 governments in 12 presidential cabinets in Latin America from the mid-1970s to the early 2020s. Its construction required extensive digitalisation of archives and the use of machine learning models, producing a dataset of unprecedented indicators, such as calls for the resignation of ministers and their reallocation within the cabinet. By contrast, most previously existing data focused on reshuffles or individual terminations. The article also includes measurement validity checks to cross-validate the dataset and information on potential applications.

The second article analyses the effect of resignation calls on reallocations and dismissals in presidential cabinets. A president can limit agency loss and incentivise ministerial performance by combining protection and dismissals. The article’s empirical strategy relies on competing risks models and propensity score matching to estimate the causal effect on ministers who received questioning. The risk increases eightfold from the second call for their resignation, but is mitigated in the case of nonpartisan and more senior ministers, with both conditions denoting strategies to limit agency problems and optimise presidential evaluation.

The third article asks how ministerial terminations affect presidential approval. It proposes that dismissing ministers who have received resignation calls presents an opportunity for the president to send signals to the electorate and, through a blame-shifting dynamic, achieve a corrective effect on popularity. This empirical expectation is tested using instrumental variables (IV) regressions. A corrective effect of about 10 points was found in coalition governments where the greater ease with which responsibility can be attributed to different actors facilitates blame-shifting.

Actions


Authors


More by this author
Institution:
University of Oxford
Division:
SSD
Department:
Politics & Int Relations
Oxford college:
St Hilda's College
Role:
Author
ORCID:
https://orcid.org/0000-0003-1510-6820

Contributors

Institution:
University of Oxford
Role:
Supervisor
ORCID:
0000-0002-1150-0954


DOI:
Type of award:
DPhil
Level of award:
Doctoral
Awarding institution:
University of Oxford

Terms of use



Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP