CEO pay: the role of performance, governance and political connection

Abstract

CEO pay has attracted substantial public attention since its introduction as one of\nthe remedies to the agency conflict. This is driven by the extensive and sustained\nrise in executive pay in most of the developed countries. Two questions arise\nfrom increased executive pay. First, does rising executive pay reflect top\nmanagement contribution to firm performance or it is just following patterns\nfrom past years? Second, what is the role that is played by corporate governance\npractices to control this rising pattern in CEO pay? This thesis has two main\nobjectives, first to investigate whether CEO pay contributes to enhanced\ncorporate performance by differentiating between the impacts of short- and longterm\ncompensation on short- and long-term performance indicators respectively.\nSecond, the study further investigates the role of corporate governance and\npolitical connections on the CEO pay.\nThe thesis uses the System GMM method to analyse a sample of 777 nonfinancial\nUK firms during the period 2000-2012. This method allows counting\nfor endogeneity and unobserved heterogeneity that are likely to emerge in the\nmodels. The findings show that CEO bonuses are positively associated with\nshort-term performance measures, while they have a negative impact on the total\nshareholder return. Evidence also suggests that long-term compensation has a\npositive impact on long-term performance, supporting the main claim of the\nAgency Theory. The results also show that the required role of corporate\ngovernance practices is passive in UK companies, reflecting a source of\nmanagerial power for the executive officers. The thesis is the first to show that\npolitically connected CEOs are paid higher compensation compared to their\npeers. However, evidence suggests that those highly powerful managers are well\ngoverned as a way to hedge against future uncertainty comes from being\npolitically connected.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 20 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.