Significant agency problems and multiple directorships

Part of my analysis on multiple directorships considers corporate governance theories such as agency theory and applies this against the issue of multiple directorships.

Agency theory in a nutshell is where somebody parts with control over their assets to the control of another due to, inter alia, an inability or unwillingness to deal with them personally. Inherent in such a relationship is the ability of the person in control to use those assets and the access to them to serve their own self-interest. In a corporate context the owner then must incur agency costs to align the interest of those in control with his own.

This is a premise for "good corporate governance" and what are the best ways, effectively, to minimise agency costs. Agency theory has developed a number of solutions to curtailing self-interest through a number of governance factors (alongside any existing legal deterrents to self-interest such as fiduciary duties). The theory identifies factors such as splitting the role of CEO and chair to two different individuals so no one individual controls the board, holding regular board meetings (as well as directors not missing too many), regular committee meetings such as the remuneration committee, executives not holding too many external appointments, a high level of independent directors to executives, high equity ownership from executives, high percentage of pay for performance element in compensation package, and a good sized board all as features of good governance.

These are just a few of potentially numerous agency problems in a firm but gives a good account of what is considered good governance according to the UK Corporate Governance Code.

The amount of agency problems present in any given firm may vary. Executives may be more able than others to impose higher agency costs. Agency theory would predict this is due to insufficient monitoring from the independent non-executives. A reason why non-executives are not monitoring effectively may be from their own external appointments that they have viewed them as a perquisite consumption due to the increasing value of remuneration available in such positions. Therefore one may predict that where a firm has more agency problems then the firm will also have a higher level of non-executives with more external appointments.

To test for this, indicator variables were created to show whether a firm had an agency problem from those just mentioned above (9 in total) and were collated. This figure was put in to a multiple regression model with other predictors (non-executive remuneration and non-executive equity) to predict the outcome of non-executive external appointments.

The first regression model did not demonstrate agency problems had a significant relationship with multiple directorships. However, to further this analysis the regression model was adapted to include each individual agency problem to identify whether there were any significant individual agency problems. Therefore the regression model included non-exec remuneration and equity as well as the 9 individual agency problems.

These results demonstrated that three agency problems (CEO-chair duality; board meetings missed; executive equity ownership) had a significant positive relationship with non-executive external appointments. Thus where a firm has these significant agency problems increasing then the non-executives will hold more appointments. There was also one significant negative relationship with the ratio of independent directors to executives. Thus where the ratio of independent directors to executives is greater the non-executives will hold fewer appointments.

Using these significant agency problems the final stage was to create dummy variables for each firm to indicate whether a firm had either 0, 1, 2, 3 or 4 of these agency problems present in the firm. These dummy variables were then placed in to the regression model again with non-executive remuneration and equity to see if there were any significant results. The dummy variable for 0 and 4 factors were omitted since they were constant.

The results show that where the firm has more of these significant agency problems present the significance of the relationship to predicting non-executive multiple directorships also increases. As a result there is evidence available that agency problems increase the amount of external appointments for non-executives and executives are thus able to place more agency costs on the firm.

Model		Unstandardized Coefficients		Standardized Coefficients	t	Sig.
		B	Std. Error	Beta
1	(Constant)	13.048	2.095		6.227	.000
	Non-Exec Remuneration per NED	.000	.000	.223	2.850	.005
	Equity held per Non-Exec (percent)	-16.154	4.186	-.297	-3.859	.000
	One significant agency problem	1.912	1.576	.119	1.213	.227
	Two significant agency problems	2.453	1.808	.128	1.357	.177
	Three significant agency problems	7.050	2.592	.229	2.719	.007

Writing up!

As mentioned in my previous post I am in the process of writing up. For anyone who has overcome this hurdle I salute you. To all those yet to experience it I do not envy you.

It is a difficult process. There are a number of issues that I have come across that I thought I would highlight in a brief blogpost to prepare the next unsuspecting victims of this torment.

1) So close but yet so far: Yes that dreaded question of when will I ever finish? I feel so close to completing but the goal does not seem to be getting any closer.

2) Fear: Do my arguments make sense? Is my external examiner going to think this is some kind of joke? Trying to make a thesis fit together even when you have a general theme becomes a difficult task. Have I explained myself clearly enough? I think I wrote the same argument and point in five different ways today in what I first perceived as clarity, but then appeared to be repetitive.

3) Forgetfulness or even cockiness: OK so I wrote my first chapter 3 years ago and now it needs some revision. I know the basic premise of the arguments in this chapter I have explained the principles and justified the arguments in the chapter. But can I remember much beyond the basic arguments? Well not really, not without going back over some of the chapter and rereading some parts of articles. But your brain often refuses to comprehend the task. It almost shuts down trying to tell you, you have done this before why do you need to rewrite it. Is it not enough that I understand the premise and justifications of the argument and people should just accept it without me having the task of rewriting it.

4) Procrastination: It seems easier to procrastinate when you get closer to a goal or finishing. The perception that you do not have long to go often convinces you that it is OK to take a quick break more often. I have often found myself saying well I'll just watch one quick event at the olympic rowing regatta. By the end of the day you get that sinking feeling that another day has passed and that you did not actually accomplish that much.

Anyway, these are four torments that I have come across. To my other readers if you have any you would like to add please feel free to post and contribute to this "discussion".

For anyone thinking of doing a PhD do not let this put you off. Nothing in life is going to be simple. The vast majority of my life as a PhD student has been fruitful and worthwhile. But bringing together 3 years of work to achieve one goal is never going to be easy.

I discussed my PhD recently through a project ongoing at UEA's Centre of Competition Policy. I hope this to be available soon and will post a link when it is made available.