DeCSS, crypto, law, and economics

[Karsten M. Self]

on Wed, Jan 08, 2003 at 08:17:41AM -0800, Eric Rescorla (ekr at rtfm.com) wrote:
> "Karsten M. Self" <kmself at ix.netcom.com> writes:
> > on Tue, Jan 07, 2003 at 04:10:27PM -0800, Eric Rescorla (ekr at rtfm.com) wrote:
> > >         However, if he can price discriminate, he can sell two copies,
> > >         one at 3 and one at 6. This makes it profitable for him to
> > >         produce the book.
> > 
> > ...and the usual mechanism is to produce various versions of the book:
> > 
> >   - A premium hardcover.
> >   - A "trade paperback".
> >   - A pulp paperback.
> >   - A premeium, leather-bound, acid-free archival quality, hand-signed,
> >     and specially illustrated, collectors edition.

> Well, that's certainly one option. However, there are certainly
> other examples, such as senior citizens discounts. 

Or matinees.  Or generic branding.  Or outlet stores.  Or rush-delivery
premiums.  Or personal shoppers.

The point being that traditional price discrimination mechanimsms _work
by appealing to preference differentials among buyers, not by
arbitrarially, and with force of law, *imposing* such differentials_.

> I think part of the point here is that legal measures to enforce price
> discrimination might well be Pareto-dominant in some cases. When
> there is a conflict between liberty and Pareto dominance, economists
> get a headache. [1]

"Pareto-dominant" isn't a term I'm familiar with (though Google returns
some hits).  Pareto efficient or optimal is:

    the "best that could be achieved without disadvantaging at least one
    group." (Allan Schick, in Louis C. Gawthrop, l970, p.32)

    http://pespmc1.vub.ac.be/ASC/PARETO_OPTIM.html

...and I can think of any number of reasons why legally mandated
discrimination is *not* Pareto optimal ipso facto, starting with the
fact that it violates several assumptions of the free market model.  The
classic test given by Varian is:

    [S]uppose that [a given allocation] were not [optimal].  Then there
    would be some other feasible allocation that gave everyone at lesat
    as large a utility, and someone strictly greater utility.  But the
    welfare function is an increasing function of each agent's utiluty.
    Thus the new allocation would hav eto have higher welfare, which
    contradicts the assumption that we originally had a welfare maximum.

    Hal R. Varian, _Intermediate Microeconomics_, WH Norton & company,
    1987. pp 534-535.

Since CSS, ipso facto, prevents certain allocations, it is not Pareto
efficient.

Peace.

-- 
Karsten M. Self <kmself at ix.netcom.com>        http://kmself.home.netcom.com/
 What Part of "Gestalt" don't you understand?
   A guide to GNU/Linux partitioning:
     http://kmself.home.netcom.com/Linux/FAQs/partition.html

---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo at wasabisystems.com