Soldiers' and CEOs' Pay: A Coloured Ribbon
Apparently, Napoleon is not the only one arguing that "A soldier will fight long and hard for a bit of coloured ribbon". In Orders of Merit and CEO Compensation: Evidence from a Natural Experiment, a working paper funded by Bocconi's Centre for Applied Research in Finance (CAREFIN), Linus Siming (Department of Finance) not only shares the same view but he actually took a step further, collecting convincing evidence to prove Napoleon's statement. Through a quasi-natural experiment, the author found that CEOs demanded higher compensations after a change in Swedish regulation prevented them from obtaining government honours (specifically, "orders of merit"), suggesting that those honours acted as substitutes for –at least a piece of– their compensation.
Whilst most of the existing literature deals with financial incentives (e.g. bonuses or stock options) or firm's specific benefits (e.g. private cars or airplanes), the role played by positional goods –i.e. goods which are important not for their commercial value but for their desirability– is still rather unexplored in academic research. Notably, there is just one ther recent working paper, pointing in the same direction of the author's work, which suggests that CEOs of Fortune 100 most admired companies earn on average 9% less than their comparables. This may be one of the reasons why the work by Siming received a lot of attention even by international press (see The Economist, A bit of coloured ribbon).
The assumption behind the author's (as well as Napoleon's) idea is simple: CEOs (soldiers) will be effectively motivated in their actions by a public recognition of their efforts achieved through the awarding of orders of merit. Such a reputational incentive may substitute the monetary ones to the point that, ceteris paribus, CEOs who have been awarded will accept lower compensations compared to CEOs who will not receive such honours. Even though such expectation is easy to formulate, collecting supporting evidence presents many empirical challenges since it is not easy to say ex ante whether a CEO will receive an award or not. To solve this impasse, the author exploited a change in Swedish regulation that prevented CEOs from receiving the Order of Vasa (the Swedish merit awarded for excellence in business) after 1974 but left untouched the ones already awarded. This natural setting allowed for an effective identification strategy and, by implementing a difference-in-difference design, it was found that, after the reform, CEOs who had not been awarded the Order received on average almost 7% higher compensations compared to their pre-reform colleagues.
Even if this setting solved the most serious identification problem, other doubts concerning alternative possible explanations may arise. To accommodate for that, the author performs additional tests making sure that the reached conclusion is not due to personal eligibility characteristics (e.g. age) or to the awarding of merits different to the Order of Vasa, together with some robustness tests. The results hold regardless the chosen specification and, additionally, it was found that older CEOs required larger monetary compensation compared to younger ones. The only case where the results do not hold is when the author considers longer pre- and post-measurement periods; however, the explanation for the loss in significance most likely lies in the fact that, by enlarging the time horizon, there are other confounding effects captured which are not meaningful for the research question. Finally, the author highlights how the rise in the post-reform CEO compensation does not correspond with a rise in firm's performance, suggesting that the orders of merit serve as a shareholders' rather than society's incentive mechanism.
Governments are informed: Next time they will face the decision whether to cut on honours, this comes at a price, a price that shareholders will have to pay to make up for their CEO loss.