Download PDFOpen PDF in browser

The underlying dynamics of the institutional crisis of the Dutch Tax and Customs Administration: An unsupervised machine learning approach.

EasyChair Preprint no. 1152

32 pagesDate: June 10, 2019

Abstract

An institutional crisis occurs when a policy sector or public organization is confronted with a sharp decline in legitimacy (Boin & ’t Hart, 2000), indicating that the viability of the policy sector or organization is at stake; its future can no longer be guaranteed (Ansell et al., 2016). Specifically, institutional crisis indicate a mismatch between the organization and its environment. Yet, while several institutional crises have been identified over the years, little attention has been paid to the specific causes that create institutional crises.

This article seeks to fill this gap by exploring the underlying dynamics that led to an institutional crisis in the context of the Dutch Tax and Customs Administration. The case is analyzed over a 24 year period (1995 – 2018) using a state of the art research method, i.e. topic modeling an unsupervised machine learning technique. This technique reveals which topics are being addressed over the years in the media in relation to the DTCA. In addition, a sentiment analysis identifies in which period in time the DTCA had received negative media attention. In the second phase, the data is studied more in-depth to understand what mismatches occurred. The preliminary results show that, contrary to the assumptions in research to date, institutional crises may also arise because of a mismatch between political ambitions, organizational capacity and operational activities. Secondly, the culture of an organization or government may act as a catalyst. 

Keyphrases: Causes and dynamics of institutional crisis, Perceived performance versus actual performance, Viability of policy sectors/ public organizations

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
@Booklet{EasyChair:1152,
  author = {Vera Leijtens},
  title = {The underlying dynamics of the institutional crisis of the Dutch Tax and Customs Administration: An unsupervised machine learning approach.},
  howpublished = {EasyChair Preprint no. 1152},

  year = {EasyChair, 2019}}
Download PDFOpen PDF in browser