Message from a Parallel Universe

[Phillip H. Zakas]

agreed with the liberal use of the phrase "perceived value" when referring
to transaction motivations...this is exactly the basis for equities markets.
A company is not "worth $8.1Bn"...the "perceived value" of a company may be
$8.1Bn.  A company is technically worth two things: a) the physical assets
at current fair market price (assuming liquidation of assets) and b)
projected value of intellectual property, market momentum, customer lists,
marketing program efficacy, team competency, projected sales, and other
intangible items.  The latter of the two is what defines the market cap of
companies.

Tim, you recommended we search out for your prior articles...I've done that
and found a couple of things (the '97 conference paper on this topic was
most relevant) but I couldn't find a whole lot...please provide an url if
you have one... after reading what I did find, I've come to the conclusion
that "reputation" is not necessarily an issue for certain kinds of
transactions...I believe it's useful to define when/where reputation means a
lot:

- If i'm hiring a painter or plumber, referrals mean a lot.  anyone can push
a pain roller.  few of us know how to do it well.
- If i'm hiring a contract programmer...past work and working relationships
matter.

Where reputation of a transactive party doesn't realy mean a lot (or means
little):

- napster.  I don't care who has the song I'm looking for...i'd download
from anyone (given the right bit rate, ping distance, connection type, etc.)
The product quality is judged by itself and my only potential 'downside' is
having spent a little time downloading a poorly encoded file...in which case
i delete the bad file and find someone else who has what i'm looking for.
- real life humint.  in this case, knowing the person is helpful, but not
important.  the key important factors are: a) is the intelligence valuable?
b) can it be verified via other sources? c) can i act on the intelligence
when i'm ready to act? d) can i control the information source?  in this
case you're necessarily dealing with people who you wouldn't trust by
definition.  the value of the relationship is maintained by how quickly the
contact is paid (unless ideologically motivated) and how consistently that
contact delivers.  but rating this information source is only useful to me
and my organization...i wouldn't share this information with a third party
'reputation credit system'.  I'm not a spy or a counter-spy so I'm only
guessing at these statements...but i think logically the above is true.
- when using an intermediary party to ensure the transaction terms were met.
in this case, the reputation of the transaction party is not important, it's
the reputation of the 3rd party intermediary.  and i don't believe this
means an "escrow" agent...it could be the financial institution guaranteeing
the funds (sort of like a letter of credit, but with verification of the
transaction details done by the bank.)  perhaps this means we would need to
have a new kind of 'bank' (or many banks) whose digital currency
incorporates triggers to ensure a transaction is completely smoothly and to
the favor of both parties (not sure how to deal with disputes though.)
- when the perceived value of the item is less than my level of acceptable
risk tolerance.  If i'm buying gum, I really don't care if i buy it from one
store or another.  i see the gum packaging looks right, the price is in the
right ballpark, then i buy it.

one other point:

the issues with "reputation credit reporting", etc. are being experienced in
some degree by ebay:

- ebay allows one to rate buyers/sellers.  if you receive negative feedback
for unfair reasons, you're stuck with that negative feedback...you can't
force ebay to change the feedback.
- ebay sellers will sometimes not sell to someone who is new or has no
feedback.  this encourages people to give each other positive feedback to
falsify their 'good reputation'.

finally, i think Tim's logic is right: don't try to come up with a complete
overhaul to the existing monetary system, just focus on the incremental
pieces.  in my opinion, infomarkets are the first best step...they're
digital, portable, somewhat protectable, and have value.

phillip

-----Original Message-----
From: owner-cypherpunks at Algebra.COM
[mailto:owner-cypherpunks at Algebra.COM]On Behalf Of Tim May
Sent: Monday, April 16, 2001 12:33 PM
To: cypherpunks at lne.com
Subject: Re: Message from a Parallel Universe



At 7:26 PM -0400 4/15/01, Declan McCullagh wrote:
>On Sun, Apr 15, 2001 at 07:26:24PM -0700, Tim May wrote:
>>  And my point is a very serious one: saying that "anarchy" cannot work
>>  in markets is not much different from saying anarchy (uncoerced
>>  transactions) cannot work in areas where in fact uncoerced
>>  transactions are the _norm_.
>
>Right. It's like telling a statist that certain government programs
>and regulatory systems aren't absolutely necessary. Their initial
>reaction is to disbelieve this notion: "But how would it work?"
>
>To phrase Tim's point another way, taking some liberties: If there is
>sufficient market demand for a product or service, it will become
>available, whether or not the government regulates it, encourages it,
>or prohibits it. If we knew tomorrow that copyright law would
>disappear in 10 years, the market would move to continue a system
>where new content would be available -- the incentives are too strong
>to ignore.
>
>Obviously ease of avoiding regulation and technological
>advances aiding reputational systems are big variables too.

Like I've said recently (and years ago, too), the "economics of black
markets" is an interesting topic. One could even take out the "black"
loaded term and just refer to economics of markets, except that then
sounds like ordinary Econ 101.

An agent typically undertakes/completes a transaction when:

Benefits > Costs

or, elaborating, when:

[Perceived benefits] > [Perceived costs] plus some epsilon for hassle/effort

(This can be puffed out, or term re-written, with various subterms
reflecting actual prices, risks, costs of money, criminal penalties,
chance of detection, etc. I did this some years back, but don't have
the time right now to track down my article or to rewrite it. I hope
most readers get the picture and can fill in the blanks themselves.
Probably more useful to do that.)

"Perceived benefits" has the usual meaning. The agent may be wrong in
his calculation or estimate, but it is for him to make the best
overall estimate of what he can get out of the transaction. Perhaps
by using the thing he buys, perhaps by selling it to another,
whatever.

"Perceived costs" are likewise his estimates of actual monetary cost,
risks to him (imprisonment, defection, bad deals) in the transaction,
etc. Into this side of the equation will go the possible law
enforcement issues. And maybe even "psychic costs" for buying
something illegal or unethical.

This is the sense in which black markets are simply markets. The
"costs" of being in the black market are simply more terms on the
cost side. Nothing new here, as black markets have been the norm in
most societies at most times, with the local thugs collecting some
percentage, etc.

For example, in the Napster case the actual monetary costs of
downloaded songs were zero (lacking MojoNation kinds of mechanisms).
The cost was partly bandwidth and sitting around waiting for
Metallica songs to download. There was also some slight cost to some
students at universities who were sanctioned for their Napster
habits. There was no "criminal cost," at least for the downloaders.

As another example, illegal drug users may put much heavier emphasis
on the criminal costs, i.e., the risk that they will be caught and
prosecuted, blah blah.

There are also the reputational issues such as we've been discussing.
These show up in the cost side as risks, actual costs to check credit
ratings, etc.

But still the Basic Equation stands:

Benefits > Costs

What Napster did was to alter several terms in the equation. What
cypherspace offers for, say, porn trading rings is the same thing.

Claims that transactions will simply "not happen" if some parts of
the equation get murkier than they might otherwise be are not
convincing.

Markets clear, and cypherspace will change the terms in the equation
but not the equation itself.


--Tim May
--
Timothy C. May         tcmay at got.net        Corralitos, California
Political: Co-founder Cypherpunks/crypto anarchy/Cyphernomicon
Technical: physics/soft errors/Smalltalk/Squeak/agents/games/Go
Personal: b.1951/UCSB/Intel '74-'86/retired/investor/motorcycles/guns