Outline:

• Why this paper deserves our attention.
• Thoughts on improvement.
  1. The setting as a laboratory.
  2. One research question.
  3. Clear predictions.

Why this paper deserves our attention:

• Most of the research in Accounting and Finance is about how incentives work.
• Pathological incentives are understudied.
• Quality of care is vital.
• Excellent work designing a survey to complement archival data.

Thoughts on improvement:

1. The setting as a laboratory.
2. One research question.
3. Clear predictions.

The setting as a laboratory: Germany

• Every national healthcare system is an experiment in delivery of effective care.
• Each system has advantages and disadvantages for researchers and patients.
• The German system deserves consideration per se.

The setting as a laboratory: Germany

• A clear discussion of how your setting lets you see things previous researchers could not.
  • This will help readers understand your approach and contribution.
  • A clear taxonomy of healthcare will help. (single payer/single provider/independent doctors/etc.)

The setting as a laboratory: Chief Physicians

• Why do we need a study about chief physicians?
  • The title raises this question, it should be clear in the first few paragraphs.

The setting as a laboratory: Chief Physicians

• If we are interested in quality of care, and incentives (we are), then clarifying the roll of the chief physician in determining quality of care is essential.
  • In the US context the attending physician has direct influence on the quality of care.
• A simple model of the hospital’s/physician’s production function may add clarity.
• If the production function is not separable (e.g. Cobb-Douglas) and some inputs are unobservable, then positive weights on observables may lead to high pay and zero output.

One research question:

The study lists several research questions:

1. Which performance measures are relevant for chief physicians’ compensation-based controls?
2. Which performance measures are relevant for chief physicians’ non-compensation based controls?
3. How do compensation-based controls and non-compensation based controls affect the quality of care?

Based on the title, and my interest, I recommend focusing on 3.

Clear predictions:

• Based on theory and institutional details
• It seems like you have strong priors about the interaction between compensation and non-compensation controls
• State them, highlight the tension
• Focus empirics on the research question

Minor issues:

• Reconsider the terminology: monetary and non-monetary incentives.
• Keep monetary/non-monetary separate from qualitative/quantitative performance measures.