Selective contracting is the practice where a health insurer contracts with some (but not all) providers. This is often seen as beneficial as it helps the insurer to lower prices for treatments. This then again helps to reduce total health care expenditures which tend to rise rapidly in most developed countries. In the US this is seen as a big success and the Dutch government tries to stimulate selective contracting as well in the Netherlands. We analyze selective contracting using the Bernheim and Whinston (1998) framework and focus on the externalities on prices for the non-insured. We find the following results. If two part tariffs are used in contracting between insurers and providers, selective contracting increases the rents for insurers but this is not passed on to consumers. Selective contracting raises prices for the non-insured although point of service options may remedy this partly.