Price variation is acceptable in most markets because: 1) higher prices reflect better quality or convenience; 2) the price, and usually the quality, of those goods is generally known to the consumer before that good is purchased, and at the very least after it is purchased; and 3) the "search costs" of discovering the price or quality is reflected in the final price. But these attributes are generally inapplicable to health care, where prices, though widely variable, are non-transparent and do not reflect the quality of medical services. This study investigates the determinants of price and drivers of price variation using adjudicated fee-for-service claims from a large commercial insurer with nearly three quarters market share. The scope of the study was narrowed to diagnostic and therapeutic colonoscopy, a well-defined procedure with substantial price variation. Consistent with both the empirical and theoretical work in the bargaining and price concentration literature, I found a substantial positive relationship between market share and a facility's colonoscopy price relative to the price in the market. I estimated that for every percentage point increase in a system or individual facility's bed share, relative price increases by three to four percentage points; this result was stable across a number of specifications and included controls for facility, system, and county characteristics. Further, I found that an increase in the Hirschman-Herfindahl Index (HHI) by 1,000 points was associated with a decrease in the coefficient of variation, a measure of spread, by only 1.6 percentage points. While this effect size was quite small in magnitude, (for comparison, the standard deviation of the coefficient of variation was 0.14), it was robust to the addition of several county-level controls. Colonoscopy is a well-defined procedure whose negotiated price within a given CPT code is highly variable and strongly dependent upon market share. Knowing which "policy lever" to pull to address price variation in the commercial market will require data on more markets and procedures to understand whether price variation is justified; if market share is associated with better quality, then price variation is warranted, but if market share represents only bargaining power, then variation may be unwarranted. The positive relationship between market share and price is of particular policy significance in the current political climate with merger incentives created by the 2010 Patient Protection and Affordable Care Act.