From: "James O'Toole" <otoole@LCS.MIT.EDU>
Date: Fri, 28 Mar 1997 22:15:46 -0800 (PST)
To: "'cypherpunks@toad.com>
Subject: RE: Hard to Tax Scenario
Message-ID: <01BC3BDE.8901B420@slip-james.lcs.mit.edu>
MIME-Version: 1.0
Content-Type: text/plain



I think that the mythical "Doctor Ann" operating in this supposedly hard to tax scenario,
may have operating costs that are higher than non-anonymous doctors, and when these
operating costs approach her taxation costs, this "hard to tax" scenario becomes
uneconomic.  Because the operating costs will be roughly linear as a function of Ann's
gross, whereas the taxes are roughly linear as a function of Ann's net, Ann will discover
that it is cheaper to operate non-anonymously once her volume of business is sufficient.

This is one of the reasons that black-market (non-reporting) organizations generally aren't
common at large scale.  There are plenty of other reasons: higher cost of capital, difficulty
of proving disputes, etc.  Crypto (and other anonymizing technology) may lower Ann's
operating costs, but for any large organization, the %-of-gross will dominate the %-of-net
fee that the tax authority charges for operating on the white market.

However, Ann has another alternative, which large organizations will also have.  Ann can
operate completely non-anonymously, but bleed profits out of her business through the use
of anonymized business partners who exist solely to assist Ann in hiding her profits.  Crypto
might make this scheme more foolproof for Ann, by enabling her to anonymously own and
operate a network of covert subsidiaries that she does business with openly, and which collect
most of her profits, but which themselves are for some reason not vulnerable to the tax
authority.

Traditionally, multi-national businesses do this using transfer pricing and offshore subsidiaries,
and their battle with the tax authority has several features:
	taxes based on GROSS for transactions crossing borders that inhibit tracking
		(customs duties, higher withholding taxes for more anonymous trading partners, etc.)
	legal requirements for consolidated reporting, with rewards paid to informers who expose
		conspiracies
	shifting the burden of proof to Ann in cases of suspected self-trading with transfer pricing.
		(this means that if Ann can't prove who her anonymized trading partner is, and it
		 therefore could be a covert subsidiary of herself, then she automatically loses)

The gross-basis taxation is the most likely cost-effective tool that tax authorities could use to
combat crypto-based tax evasion, in my opinion.

I'd like to see the cypherpunks-quality analysis of how crypto/ecash would help the anonymous
Doctor Ann function in a society whose tax system was more like Europe's VAT system.

----------
Date: Mon, 16 Dec 1996 16:47:14 -0800
From: Hal Finney <hal@rain.org>
To: cypherpunks@toad.com
Subject: Hard to Tax Scenario
Cc: hanson@hss.caltech.edu
Sender: owner-cypherpunks@toad.com
Precedence: bulk

Robin Hanson, inventor of the Idea Futures prediction market, is a very
creative and thoughtful writer who has posted to this list occasionally.
He says he sent the message below to the CP list over the weekend, but
I didn't see it.  I am including it (in bits and pieces) in its entirety
for the benefit of others who may also have missed it.

Robin Hanson, <hanson@hss.caltech.edu>, writes:

> Hi.  The volume here is too high for me to subscribe regularly, but I
> subscribed recently so I could ask the following question:
>
>   How well thought out is the notion that widespread crypto could
>   allow a large fraction (>30%?) of the economy to avoid taxation?
>
> I've heard this speculation many times, and just saw it in print in,
> in David Friedman's article in the summer '96 issue of Social
> Philosophy and Policy.  But I have trouble imagining how it could
> work.

For those who don't know, by the way, David Friedman is the son of
Nobel prize winning economist Milton Friedman.  Both father and son
have libertarian leanings, and David in particular has tackled some
of the hardest problems which would be faced by an anarchic society.

> Imagine Ann is a doctor who wants to ply her trade without taxation.
> Patients go a local high res medical net booth, which Ann runs from
> long distance using several real time digital mixes.  To do this, Ann
> spends most of each day in her expensive home VR room.

So we are imagining a future scenario in which medicine is commonly if
not universally practiced via these remote means?  Or do we have two
classes of doctor, the anonymous virtual ones and the identified ones
that you go and see in person?  I ask because at least some of the
difficulties Ann faces seem due to her virtual practice.

> Patients pay Ann in untraceable cash, which she uses to pay for
> groceries and other net services.  Her cover story about why she
> spends so much time in her home VR room, and how she pays for
> groceries, is that she is a receptionist for some sham company.

The need for a cover story raises the question of from whom Ann has
to keep her secrets.  In a society where (we will stipulate) 30% avoid
taxation, the moral significance of not paying taxes will be different
than it is today.  We had some interesting posts in an earlier discussion
on this list describing the situation in Italy, where apparently tax
avoidance is raised to a higher degree than in the U.S.  It sounded like
it has the approximate moral status that speeding does here, a minor
infraction which almost everyone does some if not all of the time.
In some sub-cultures no doubt the tax avoidance rate would be even higher.

In such a society Ann may not have to care that much about keeping her
secrets, as long as she doesn't have too high a profile at tax time.

> Ann has many collegues which she does business with regularly,
> including equipment suppliers, a pharmacist, a nurse practitioner,
> emergency substitutes, and various specialists.  Ann has never met any
> of these people in person, and they all show each other fake faces,
> voices, and even rythms of walking and speaking.  Ann's social life
> outside VR is entirely divorced from her work life.

Well, that last part is true for me already; I telecommute to a company
300 miles away and have no social life with my co-workers.  For that
matter my wife and I have practiced cocooning for several years,
and I haven't had a close friend from work since the early 1980's.
Being married makes this easier, of course.

The other part of this scenario, where Ann interacts with her co-workers
via fake faces, does seem disturbing.  I could imagine, though, that this
might be common in such a culture.  Maybe everyone pretties themselves
up when on the videophone.  If there is widespread understanding that
most faces are at least somewhat false, then perhaps going all the way
to a completely faked up face would seem more acceptable.  But to someone
from my generation it will be hard to accept.

> To convince patients to trust her, Ann gets bonded by a certification
> service.  To obtain this certification, Ann must be careful to not
> refer to any people who know her "true name", such as her teachers at
> the physical school she physically attended.  And Ann must somehow
> assure the certification service that she will not resell the
> certification, allowing others to pretend that they are her.

It is possible that we might see a more performance-based certification
rather than a recommendation based one.  My wife is a physical therapist,
and she had to pass a licensure exam given by the state which qualifies
her to practice.

In an earlier message to me Robin pointed out the crucial role played
by recommendations in hiring decisions.  Certainly I would be much more
likely to hire someone who listed his previous jobs and for whom I could
get good recommendations by his earlier supervisors than an applicant
who insisted that this information was confidential.

Robin also suggested that there could be a selection effect, so that the
doctors from good schools with good grades would use these advantages to
maximize their income, and so the only anonymous ones would be the ones
who didn't have these qualifications.  This could lead to a situation
where most anonymous service providers were assumed to be inferior to
regular ones, so they would get less money even if they were actually
just as good.  (I apologize to Robin if I missed the point of his
earlier discussion or am presenting it incorrectly.)

Even with such disadvantages, a doctor like Ann might accept a lower fee
at first while she builds up her reputation as an anonymous doctor with
talent and ability.  After a few years she could hope to have overcome
the stigma which (we will suppose) anonymous doctors face and display
some strong recommendations based on her successes.  In the long run
this could be a winning strategy due to the tax savings.

(I haven't given the problem of reselling certificates enough thought to
discuss it in any detail.  There have been some discussions of "is a
person" credentials which could apply, but that opens up a big can of
worms.)

> If Ann ever slips up, revealing her true name to a virtual associate,
> failing to convince a physical associate of her sham employment,
> or if anyone ever breaks through her realtime digital mixes, Ann
> is open to expensive blackmail, she may have to start over with a
> new virtual persona, and may have to go to jail for a long time.

This is an interesting problem which I haven't seen discussed before
in this form.  In Vinge's original "True Names" people were afraid of
harrassment and physical threats if their identity were discovered, but
Robin's example of the danger of being exposed as a tax evader could be
very bad as well.  If there is this much tax avoidance, we might assume
that tax rates are high, and penalties for tax evasion are high as well.
On the other hand, if tax evasion is nearly universally practiced,
perhaps there are strong cultural pressures against turning someone in.

There is also the question of how good the technology is for anonymous
communication.  At best it would appear to require a widespread infra-
structure, and if this is used largely for tax evasion it is hard to see
how it could survive.  So I think this would be a very significant issue
to be faced by the prospective 'nym.

> It seems to me that Ann is paying a high price for an ability to
> avoid taxation, and at current tax rates it is hard for me to believe
> that she wouldn't just rather pay the taxes.  What am I missing?
>
> Robin D. Hanson  hanson@hss.caltech.edu  http://hss.caltech.edu/~hanson/

It is hard to judge how high the various prices are that she pays.
The socialization aspect may not be important if she has friends outside
work.  The risks of being caught will depend on factors we don't know,
like the technology and legal system.  Rather than assuming that tax
rates are the same, it might be more plausible to assume they have gone
up in order to keep revenues stable.

Another problem which Robin didn't mention is the issue of insurance
payments.  It is hard to see how Ann's patients can get reimbursed for
their expenses if we assume that Ann is in effect a "black market" doctor.
This problem may be somewhat specific to the medical scenario, but I
suspect that many other professions are going to have trouble switching
to a cash basis.  Anyone whose customers are businesses, for example,
will face the problem that the businesses' books will need to show that
an expense is justified in order to deduct it.  This will be a major
problem for the "anonymous firm" we have discussed occasionally.

One difficulty I find with this scenario is its science fictional nature.
It is hard for me to consider details about the life of a doctor who works
via VR.  Also, if we are already in a situation where 30% of people are
avoiding taxes there will certainly have been major changes in society,
but I don't know what they will be.  This makes it hard for me to focus
on the issues specific to Ann's anonymity.  However I do like Robin's
choice of a concrete and vivid example like this.

Hal