I just read a nifty little Cato Institute article by Glen Whitman (CSU-Northridge) titled "Against the New Paternalism: Internalities and the Economics of Self Control". Here's the ex. sum:

Economists have long argued that government intervention makes most sense in situations that involve externalities. Externalities are costs or benefits that spill over onto third parties. When individuals bear the full costs and receive the full benefits of their own actions, the justification for government involvement is much weaker. But a new generation of economists contends that paternalistic intervention can be justified to correct problems of self-control. If people don't fully consider the costs their choices impose on their own future selves, the theory goes, those choices impose within-person externalities dubbed "internalities." The internalities approach provides a novel argument in favor of paternalistic government policies such as sin taxes (including fat taxes), marketing restrictions, mandatory savings plans, and so on.

The theory of internalities is explicitly modeled on the theory of externalities. However, the former stands about where the latter stood in 1960, just prior to Ronald Coase's seminal work on the subject. Exposing internality theory to Coasean insights reveals serious flaws. Specifically, internality theory in its current form unjustifiably "takes sides" when it chooses to favor some personal interests over others. Furthermore, it ignores the possibility of within-person bargaining and other private solutions to self-control problems. Finally, it gives insufficient attention to the possibility of government failure. Taking those objections into account severely damages the case for paternalistic government intervention to address problems of self-control.

ATSRTWT