A report funded by the Urban Institute and Medical Group Management Association examines the hourly compensation paid under Medicare Fee Schedule to the actual compensation received by providers, by specialty.  This analysis confirms substantial differences in actual hourly and annual compensation across specialties, and that, under simulated Medicare compensation, the compensation ratios across specialties are narrowed very little.

From the report:

A primary goal of the 1992 Medicare physician payment reforms based on a resourcebased relative value scale (RBRVS) was to create an economically neutral fee schedule -- one that rewards all physician work equally. To develop such a fee schedule, the Centers for Medicare and Medicaid Services (CMS) refined and expanded the estimates developed by William Hsiao and colleagues of the work required to perform physician services. The resulting Medicare Fee Schedule increased payment for evaluation and management services (E&M) and decreased payment for procedures and tests relative to historical payment levels. The expected effect of this shift was to raise Medicare payments per service to primary care specialties and to lower payment per service for most other specialties. [...]

Since those initial attempts, no one has simulated how the Medicare Fee Schedule would change compensation per hour worked, annual compensation by specialty, or how such simulated compensation compares with actual compensation as reported from various physician surveys. Although there is extensive survey data displaying physician compensation, all of the surveys understandably provide aggregate compensation, that is, compensation from all payers, thereby obscuring the specific impact of Medicare’s payments on physician compensation levels.

Full report: http://www.urban.org/health_policy/url.cfm?ID=412051

Urban Institute. (2010). What if all physician services were paid under the Medicare fee schedule? An analysis using Medical Group Management Association data. Berenson, R.A., Zuckerman, R.S., and Stockley, K.