Abstract (eng)
This thesis provides an empirical analysis of the relationship between executive compensation structure and performance in the banking industry. The aim is to decompose remuneration as precisely as possible in several short- and long-term components and examine their association with performance for a couple of model specifications and in several settings. For this purpose, I employ a unique “hand collected” data set of 52 Eurozone banks in the period 2010 – 2014. The list of my sample banks emerges from the SNL Financial database, which is also the source for all financial and stock market data that I use. I obtain my compensation data from the banks’ annual reports, remuneration reports, corporate governance reports, and pillar 3 disclosure reports.
In my analysis, I first account for the term structure of remuneration by dividing executive pay in short- and long-term incentives. Subsequently, I fragment remuneration in cash- and share based short- and long-term incentives in order to reflect the form of granting compensation. In a further model specification, I differentiate between short-term incentives, and long-term incentives divided into deferred pay and other forms of long-term incentives in order to examine the use of debt-like remuneration mechanisms. I use continuously compounded stock returns as a measure of performance and provide additional robustness checks with two alternative performance measures - return on average risk-weighted assets and Return on Average Equity (ROAE). I control for bank size and leverage. For the main part, my analysis concentrates on the whole management board, but I also compare and contrast my results for CEO and non-CEO board members. I also examine the relationship between compensation structure and performance for large and small banks by using a threshold of €50 billion in total assets, and account for government ownership stakes.
I find that long-term incentives exhibit a positive and significant relationship with performance, much stronger than short-term incentives. Moreover, cash-based long-term pay components have a stronger relation to performance than long-term equity-based incentives. I do not find evidence for this effect in the case of short-term incentives. Deferred pay has the strongest relationship to performance as compared to all other compensation forms. My findings are valid for the whole management board, as well as for CEO and non-CEO board members, and are robust with respect to alternative measures. However, I find that long-term incentives exhibit the strongest relationship to performance when the performance measure is continuously compounded stock returns, and the weakest association with performance when performance is measured with ROAE. Exactly the opposite holds for short-term incentives. Furthermore, the relationship between compensation structure variables and performance is stronger in the case of non-CEO members than in my CEOs sample. Splitting the Management Board sample in large and small banks, confirms my general findings in the case of small banks. Among large banks only short-term incentives are significantly related to performance. Government ownership is significantly negative related to performance in the whole management board, insignificant in small banks and highly significant in the presence of large banks. In the presence of a government stake, compensation structure does is not significant in the large bank sample.
My thesis has several theoretical and practical contributions. It is the first study in the banking literature that examines the direct relationship between executive pay and performance in several countries outside the U.S. By using the period 2010 – 2014 I update even the most recent scholarly findings, majorly focusing on the pre-crisis or crisis period. My results have various implications for the design of executive compensation schemes in the banking sector, for regulators, and policy makers.