Executive compensation and superstar CEOs in public sector

Another one of the entrenched narratives is that performance payment is the way to reform public systems. But, as I have blogged on several occasions, this narrative stands on very flimsy presumptions. 

Adding empirical basis to questioning the narrative, a new paper examines whether CEOs can impact the performance of large and complex public sector organisations. 

It uses the case study of CEOs of English public hospitals under the NHS system, the fifth largest employer in the world with over 1.2 million employees, and where a pool of CEOs keep moving across comparable hospitals. On average these hospitals have 4500 employees, multi-million turnover, and labour costs accounting for 70% of the costs of production. Reforms of the late eighties replaced the administrative model with a highly decentralised managerial approach to hospital management, and hospital boards have autonomy to compensate CEOs accordingly. The authors compare perceived differences in managerial ability, as proxied by their compensation, to actual differences, as they emerge from the analysis of the objective production measures. 

They write,

We study the extent to which CEOs are differentiated in terms of their pay, as well as a wide range of hospital production measures including inputs, intermediate operational outcomes and clinical outcomes. Pay differentials suggest that the market perceives CEOs to be differentiated. However, we find little evidence of CEOs’ impact on hospital production. These results question the effectiveness of leadership changes to improve performance in the public sector...

we... examine whether CEOs have a “style” regardless of the hospital they manage by testing whether there are statistically significant and “portable” CEO fixed effects, i.e. systematic differences in hospital production that are associated with the movement of CEOs across different hospitals. We... “stack” our hospital production variables into distinct sets of input, throughput, output (clinical and financial) and staff job satisfaction measures to take advantage of the fact we have multiple measures of production, and to simplify the exposition of our results while maximizing the number of observations. Our results show little consistent evidence that CEOs are able to generate persistent performance differentials across the organizations they lead. While we find the estimated CEO fixed effects are jointly statistically significant, these CEO fixed effects are essentially period-hospital-specific shocks rather than true CEO effects, since large deviations in one production measure in one hospital are typically not replicated by the same CEO in another hospital... 

hospitals are large complex organizations, in which highly trained (and hard to monitor) individuals run separate but interconnected production processes. Management at the very top of such organizations may find it difficult to engage in coordination and getting a large number of actors, who traditionally have not worked together, to work cooperatively. Put another way, a possible interpretation of our finding is that the organizational inertia of a large hospital is too strong for a single manager – even if this person is the CEO – to be able to impact performance within the short time period in which they are in office, and consistently across organizations. 

And this definitive conclusion is important,

Our results indicate that the CEOs of large public hospitals do not necessarily impact hospital performance, a result that stands in stark contrast with earlier findings relating to the private sector and to smaller public sector organizations... the lack of a CEO effect may also be due, more broadly, to the complexity of hospital production, which transcends the fact that the NHS is publicly owned. From this perspective, our results cast doubts on the effectiveness of a “turnaround CEO” approach–the model in which top managers frequently rotate across hospitals to induce meaningful changes in performance–for large public sector organizations.

There is a related lesson. Market-differentiation, based on executive compensation, the proxy for CEO ability and therefore compensation in the private sector (also the basis of differentiation chosen for study here), is unlikely to be effective in the public sector or with complex systems. In fact, as I have just blogged here, the case for high executive compensations even in the private sector is questionable.