Re: Protocols for Insurance to Maintain Privacy

[Adam Back <aba@dcs.ex.ac.uk>]

Adam Shostack <adam@homeport.org> writes:
> Many insurance companies with Mutual in their name (Liberty Mutual is
> large in the Northeastern US) get that from being founded as mutual
> insurance companies, where you pay to be part of the mutual insurance
> group, and when you get sick, injured, etc, the group pays money
> towords your treatment.  I think it broke down with increases in
> mobility.  They were implicitly based on reputation capital, and were
> not highly fraud resistant.
> 
> Adam's suggestion of a charity which only pays for the treatment of
> those who donate thus recreates an old system.

Saying that it would be a charity is just a statement of reality.
There exist no methods to insure when there is a near certainty of a
payout being required, as I think we all agree (Tim, Adam, Jon).

To call it anything other than charity is misleading, so I am really
just arguing for honesty in advertising.  (Something socialists are
not big on :-)

An insurance company would probably find customers and hope to make a
profit by offering a mutual fund, or charitable donation service.
There would be an agreement as to who could claim, and what types of
illness were covered, maximum payouts etc.  Then the business would
use the funds (minus it's profit margin) to pay for those treatments
affordable within the current funds.

This means that as funds may be limited that some people will go
untreated, if their treatments are too expensive and funds are short.
Efficient usage of funds is expected -- if it costs $1m to fix up some
old guy to live for a couple more years as opposed to giving cheaper
treatments to several younger people, well the old guy gets to die.

This might seem heartless to socialist types, but this kind of
decision gets made all the time by medics, who don't have unlimited
funds even in socialist countries.  In the UK some practices are funds
holding, which means that they get to manage their own money, they get
to decide which drugs to use, and who to give them to, and they get to
use profits (or more correctly surplus government grants) in certain
ways (equipment, etc).  (Or they did, Labour were threatening to
confiscate these surpluses).  I talk to medics a bit, as I am into
medical messaging security, and some of the people are bilingual
medics and medical informatics people.  Examples are drugs which are
3% more effective but cost 5x the price.  These kinds of optimisation
of available resources problems are all over the place in medicine,
and doctors balance these within budgets.

Another example is that doctors are able to tell smokers with lung
cancer who refuse to stop smoking that they will not treat them.  Or
switch off life support machines if it looks pretty hopeless, and is
costing a fortune to keep a half-alive person ticking over.

The failure with mutual funds is that people will attempt to start
donating once they realise they have some big medical bills coming (if
only donators get payouts).  To combat this the company might perhaps
have a policy of not paying out within 5 years of the first donation.
Then you'll have the counter attack of people re-selling proof of
donation receipts.  And then the company can counter that with a hash
of DNA code mixed in to the donation receipt such that only with the
patients consent can the receipt be used to link to him. 

The key point is the choice.  Many companies, many choices, freedom to
donate or not, or to stipulate terms of donation (eg., not to fund
lung cancer treatment for smokers).

It seems to me that people are on average more generous and charitable
than socialists would have us believe also.  For example a charity to
support people who had no insurance, or hit the maximum payout on
their insurance might get donations.  Perhaps even enough donations
that it would cover most cases.

Adam
-- 
Now officially an EAR violation...
Have *you* exported RSA today? --> http://www.dcs.ex.ac.uk/~aba/rsa/

print pack"C*",split/\D+/,`echo "16iII*o\U@{$/=$z;[(pop,pop,unpack"H*",<>
)]}\EsMsKsN0[lN*1lK[d2%Sa2/d0<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<J]dsJxp"|dc`