The theoritical literature makes ambiguous predication about the relationship between the extent of rent extraction by private parties and its impact on economic activities. One view argues that after putting ethical consideration aside, corruption may in fact improve efficiency, particularly in developing economies. In this model the size of bribes by different economci agents could reflect their different opportunity cost. Better firms are more willing to buy effective ered tape. The theory suggesting that bribery may lead to ower effective red tape is known as the efficient grease hypothesis. The crucial assumption of this model is that the red tape and regulatory burden can be taken as exogenous, independent of the incentive for officials to take bribes. The opposing view on the other hand asserts that because the bureaucrats have discretionaty power with given regulation, regulatory burden may be endogenously set by corrupt officials such that they customize the nature and the amount of harrasment on firms to extract masimum bribe possible. In this model firms that pay more bribes could still face higher, not lower, effective red tape. Consequently corruption could lower economic efficiecy insteaf of improving it. In this paper we estimate a model whereby bribe could be endogenous. The model stresses the role of firms' commitment ability's function of their characteristics. We use the data from the recently completed survey on the governance of the local governments in Indonesia. The data cover bribe that needs to be paid by firms to goverment officials at the district level. Other aspects of governance such as transparency, accountability, efficiency and the general attitude toward business sector are included in the data as well. The firm level regression suggests that the amount of bribe paid increases with the firms size. Thus rejects the efficient grease hypothesis.