Leadership, CEOs and Bosses Flashcards
Betrand and Schoar 2003 = what is the key idea of the identification strategy?
As a manager joins the firm; their change in performance is as a result of the manager. This manager effect can be used to intepret strategy and performance of firms.
Betrand and Schoar 2003 = What are the strategy areas look at by the authors?
Investment level, CapEx, Acquisitions, Cash position, Debt Leverage, R&D, Advertising
Betrand and Schoar 2003 = what are the performance measures?
OROA and ROA (financial based performance)
Betrand and Schoar 2003 = What are the findings?
-Managers influence strategy and performance a great deal.
-Different MMT styles are captured through comparing different fixed effect combinations.
-Higher Acquisition FEs and Higher Cash position FEs give a lower performance.
-Older managers are more conservative.
Bennedsen 2020 = What issue does this paper solve in Betrand and Schoar?
the problem of CEO FEs capturing confounders. This could be that when managers change firms they move as a result of economic shocks/high or low performance of incumbent firm. Bennedsen uses CEO staying in the same firm as a solution.
Bennedsen 2020 = what is the first regression used? What is the magnitude of effect
-OROA = number of days in hospital + firm FE and year FEs.
- 10 days of hospitalisation results in around a 7% drop in OROA.
Bennedsen 2020 = what are the issues with using the primary regression? What is done to account for these issues? Does the new regression yield similar results?
- a hospitalised CEO maybe replaced with a worse one which would affect performance not throught the CEOs hospitalisation
-A good CEO maybe hired in good years.
To account for these concerns a Firm-CEO match FE is included.
-the results are very similar to before.
Bennedsen 2020 = what extensions are made to the regression? what does this extension show?
capturing different amounts of time hospitalised (1-4, 5-9, 10+) this extension shows the longer time spent in hospital the greater the negative effect on companies financial performance.
Bennedsen 2020 = what are the threats to causality and how are alleviated?
(1) Bad year causes CEOs stress => more hospitalisations. Using a control for stress related-illnesses shows no difference
(2) Bad year causes CEO to be more willing to go to hospital => more hospital. Controlled same results as before.
(3) Good year causes CEO to spend less time in hospital as leaves earlier. Using an IV of average disease hospitalisation finds no difference in estimates.
(4) Firm reorganisation as a result of CEO Hospitalisation. no effects of this on performance.
Bennedsen 2020 = what other conclusions does Bennedsen confirm?
-CEOs are unique. other top levels dont show the same issues when absent.
-Young CEOs absence has a larger impact on performance than older CEOs.
Lazear et al 2015 = describe the papers setting.
Evidence from a large data service company where employees are all conducting the same task. supervisors are tasked with managing ~9 workers where they train, teach and motivate workers. workers productivity is measured in output per hour. The idea of this setting is to try and figure out the effects that bosses have on their workers.
Lazear et al 2015 = what is the main regression used. and how do results be intepreted.
-output of each worker = controls, worker FEs, time FEs, and Boss FEs.
-using SD you can determine bosses impact.
If SD is near zero, no impact.
If SD is large, big impact of bosses.
Lazear et al 2015 = what are the initial results?
-Bosses matter alot for workers performance
-Bosses effects are dispersed.
-Replacing a boss from the bottom 10% with a boss from the top 90% will be the equivalent to adding another extra worker to a team of 9 members.
Lazear et al 2015 = what extensions are made to the primary regression? What are the results of these?
Determining the bosses persistence, FE for bosses of 6 and 12 months prior is included.
These results show bosses have a long lasting effect on worker performance. Past bosses effect workers today, implying some sort of training involved.
Lazear et al 2015 = What other conclusions can be found in the paper.
-The Boss-Worker wage gap is fair due to the productivity difference between them. Assuming the top 10% of workers are similar to bottom 10% of bosses (as good workers become bosses). A boss is 1.75x more productive than a worker.
-Good bosses are expected to improve output levels of all workers but improves good workers more. Paired together gives a positive net effect.
