From: Daniel Fabulich (daniel.fabulich@yale.edu)
Date: Thu May 28 1998 - 22:38:19 MDT
On Thu, 28 May 1998, Michael Lorrey wrote:
> In any event you DID sign a contract when you bought the illegal information from an anonymous seller: you waived
> the seller, recognised that the seller made absolutely no guarrantees as it its accuracy, validity, or origin.
> I.E. you bought into the implied caveat emptor contract that goes with any commercial trade as a base of
> understanding.
There is a somewhat vocal group of libertarians who insist that an
implicit contract is a void contract; parties must agree to the terms of a
contract before they may be ethically enforced. Can you show that this is
wrong?
> > In addition, you get even worse problems when the merchant is also
> > anonymous. Suppose I post through anonymous remailers and a mail2news
> > gateway that I have a copy of your manuscript available for anyone who
> > sends me the price of a copy in anonymous cash. I don't know to whom I'm
> > selling, and they don't know from whom they are buying. How are you going
> > to prosecute me THEN?
>
> a) Your customers, if caught with the illegal copies in their posession, will be arrested and prosecuted. Their
> arrest and prosecution will be publicized in the media, thus generating a deterrent effect in the public that
> will impinge upon your ability to do business.
>
> b) nothing is totally anonymous. Anything on the net can be traced. As for your arguments about the cost of
> enforcement, I can also say that the internet also gives IP enforcement agencies the ability to greatly reduce
> the cost of IP enforcement, at least in regards to web commerce. I think that these costs will be reduced enough
> that it will totally offset, and then some, your imagined 'efficiency' of having no IP.
Since I find this scenario interesting, I'm going to run it down step by
step. You tell me where and how I get caught.
1) I license the use of a copy of information you created. 2) I encrypt
an advertisement and a pseudonymous public key to Dave's public key,
stating that I have a copy of your manuscript available for $0.05.
Only Dave can read this message; Dave has agreed to broadcast e-mails
which are sent to him via USENET. After enrypting it to Dave's key, I then
encrypt the ciphertext to Carol's key, adding a message requesting that
she send the e-mail to Dave; I encrypt that ciphertext to Bob's key with a
message requesting that he send the e-mail to Carol and send the message
to Bob. 3) Bob decrypts the e-mail and sends his decrypted text to Carol
as requested. 4) Carol, ignorant of who originated the e-mail beyond the
fact that they sent it to Bob, decrypts her ciphertext and sends the
remaining ciphertext to Dave. 5) Dave decrypts the e-mail, ignorant of
its author, and broadcasts it to USENET. 6) Zed reads this message and
decides to post $0.05 in anonymous money encrypted to my public key along
with his own public key; he does so through a similar chain of anonymous
remailers; no one but the first in the e-mail chain even knows who Zed is,
no one but me knows what the message says, because Zed encrypted the
e-mail to me before sending it along the chain. 7) I encrypt a copy of
your manuscript to Zed's public key and broadcast it anonymously.
In this way, Zed and I can conduct a trade perfectly anonymously. If the
remailers do not keep a copy of the e-mail which they forward (they
don't), then it is impossible to trace either myself or Zed.
What are you going to do about this?
> But you are talking about a market. Markets are not markets of brand names, but of product types. The cola market
> is not the Coca-Cola market, or the Pepsi market, but the cola market. Both have different recipes for cola,
> which are protected by patent, both compete to satisfy the public demand for colas with different products. You
> cannot say that Coca-Cola has a monopoly on the cola market. Thus my assertion that a patent is not a monopoly
> stands.
Well, let me put it to you this way: any economist who knew anything at
all about microeconomics would agree that Coca Cola charges more for Coke
than it would if anybody could make Coke, and that this is because it is
the only company that makes Coke. They would also agree that increased
prices lead to decreased quantity. They would agree that this decreased
quantity coupled with increased prices would represent deadweight loss, or
economic inefficiency, and that all else being equal this will drag down
the macroeconomy.
If you don't want to call this a monopoly, that's OK with me, but please
excuse me if I agree with the textbooks on this point.
> So you are for treating IP as a commodity. Any commodity market is going to trend toward an oligopoly situation,
> as economies of scale push smaller producers out or up in the market. Why is this? Because in the case of
> internet commerce, bandwidth is discounted for large users, and hardware like servers, routers, drive systems, CD
> writers, etc. are all cheaper on a per kilobyte basis the larger capacity you go.
Are you saying that it would be a natural monopoly? If so, since when are
libertarians opposed to natural monopolies? Also, I disagree with your
assertion that all commodity markets tend towards monopoly thanks to
economies of scale. There are lots and lots of commodities on the market
that aren't monopolistic, never have been and don't look like they're
about to become monopolistic in the near future. The existence of such
markets disproves your assertion that all commodity markets have this
behavior, though it may nonetheless be true for IP.
> In this scenario, the small home grown inventor is discouraged from becomeing a player in the market, unless they
> are a rich bachelor who expects to get ripped off by the big boys, playing it off as if they are great
> philanthropists, humantarians, etc..., or ulness the home grown inventor sells out to the established players in
> the market. You are not encouraging a free market, but encouraging oligopoly/mercantilism.
If you are correct, and it is more efficient for a monopoly or oligopoly
to produce the goods than it is for small producers to compete (or sell to
them), then by encouraging this process we are encouraging a free market;
this is because in the case of a natural monopoly, the free market is not
competitive. I am not convinced that this is actually the case, but it
does not help nor hinder either of our arguments if you are right on this
point.
> Yes it is weird, but your numbers for the present market are not worth extrapolating from as it still is less
> than 0.1 % of the market, yet even if LINUX catches on in the market beyond the hobbyist or iconoclast, companies
> like Red Hat will not make any decent level of revinue, until one of the established companies, like Microsoft,
> comes along and modifies the OS into something which it can protect, LinOS II. They then will promote this, and
> soak up the entire market, putting Red Hat out of business, or back in the hat closet.
Most Linux users would disagree; they argue that the reason Linux is good
and growing (some say that it is already better than Microsoft's product)
is because anyone is free to modify the source in any way they want, but
they are also prevented from protecting their invention under the GPL.
This means that it is easy for anyone with a good idea to improve the OS,
rather than wait for Microsoft to release the next version of its
bloatware.
Also, if there were no IP laws, information like Linux would make up 100%
of the market, not 0.1%. Remember, I'm not trying to say that Linux is
going to beat Microsoft in today's market; I'm just saying that Linux
continues to improve despite the fact that nobody can protect their new
Linux code. This flies in the face of your assertion that if inventors do
not receive royalties on their inventions, they will not invent. They
will invent, only slower.
> It would be a good thing. However I think that your expectation is wrong, and even if it does gain efficiency, it
> will be because the commodity market for IP pushes invention supply into an oligopoly situation, which any self
> proclaimed libertarian free marketer should abhor.
I don't abhor natural monopolies, nor should any other free marketer. It's
the COERCED monopolies that we have to worry about. Also, I am not at all
certain that the market for IP is naturally monopolistic, and I definitely
disagree with your assertion that all commodity markets monopolize.
> Considering that the economy is still pretty planned, with most industry still in the hands of Communist party
> apparatchiks, I think that a) most of its economic growth is from externalization of western IP costs via
> pirating of foreign owned IP properties, and the rest is b) from internal IP that is owned by state industries
> and enforced by party apparatchiks with connections with police or army authorities, or c) small consumer
> businesses, mostly in the service industries, that are privately owned.
I don't think you can get much more qualitative than this analysis, which
was all that I was trying to show. This is not to say that it is wrong,
necessarily, but simply that you can't do this quantitatively; you MUST be
qualitative.
> yet you claim with a wave of your hand that nicroeconomic efficiency is beneficial for the macro economy without
> showing HOW it benefits the macro economy, WHO in the macroeconomy truly benefits, and refuse to measure the
> effect in the macro economy of the disincentive to invent on the rate of invention, and reduction in productivity
> that results from stagnated technological development.
Well, figure it out yourself. The portion of surplus which is above the
price line and below the demand curve is consumer surplus; that below the
price line and above the supply curve is producer surplus. If IP were to
be eliminated, consumers would gain more than producers would in the copy
market (see http://pantheon.yale.edu/~dgf4/incent3.gif ); also, the
increase in consumer surplus when demand increases is greater under
perfect competition than under a monopoly, while producer surplus is
greater under monopoly than under perfect competition, as noted.
As for how it benefits the macroeconomy, you should know this perfectly
well: all else being equal, increased total microeconomic surplus results
in increased nominal GDP. However, this is similar to the argument for
kids drinking milk: all else being equal, a child will grow taller if
they drink milk than if they don't drink milk. How much taller? I cannot
say. All I claim to know is that microeconomic efficiency is better for
the economy than microeconomic inefficiency, but I cannot tell by how
much.
> I would love to see links or book refs to anything that shows that you can determine the relative size of two
> shapes without knowing the quantitative size of either shape. Sounds like magic to me.
Look, you want a REAL REAL simple example? Suppose I have a triangle
formed by the points (0,0), (x,0) and (y,0) and a triangle formed by
(0,0), (z*x,0) and (y,0) where z is less than 1. The latter triangle will
have less area than the former triangle. I have no idea what their areas
will actually be, but I know their relative sizes on a qualitative level.
No magic, just math. If I know enough about the supply and demand curves,
then I can use the same "magic" on the area of the deadweight loss, the
increase in revenues as demand increases, and the quantity produced by
monoplists as opposed to that produced under competition.
> Yet you show no quantified relative size.
I never claimed to! Did you read my conclusion? All I said was: "To
conclude, if we let A be the area of the blue trapezoid under perfect
competition, M be the area of the blue shape in the monopolistic diagram,
D be the area of the grey triangle and E be the cost to society of
enforcing IP laws, then we should eliminate IP laws if D + E > M - A, or
if A + D + E > M."
> Also, since you only count marginal costs, not fixed costs, your chart
> is absolutely meaningless, especially in an IP marketplace, where almost all costs are fixed costs (i.e. R&D),
> while the marginal costs of reproduction are nil.
Almost nil. That was the low red line in the third graph. As for fixed
costs, you're right, they don't show up on the graph, but they were the
central point of my use of those graphs: if the increase in revenues is
greater than the fixed cost of invention, then invention will take place,
if not, then it won't. This was the principle I set out in bold; I
believe it is the most robust part of my analysis because it is so
trivial. If you disagree with this part of the analysis, (and there's no
reason why you should: a priori, you could still be right even if this
were true) you'd better provide your OWN graphs showing why mine are
wrong.
> Additionally, your measure of efficiency, i.e. the quantity
> made available in the marketplace, is wrong. The efficiency differential is measured as the change in price
> times the change in quantity.
Gee. Funny how the area on the supply/demand graph is given in precisely
those units: Price times quantity.... SAY! Wait a minute! You don't
suppose it's a microeconomic PRINCIPLE that the change in efficiency is
given by the change in area on the supply/demand chart, do you? Naw. It
couldn't be...
Why do you think I spent all that time harping on about area, anyway???
Geez.
> In addition, you also only use straight lines in your graph, which any microeconomist can easily tell you are
> simplistic and inaccurate of a real market. The demand, cost and monopolist market line all will tend to be
> S-curves of one degree or another, depending on the product, the state of the art at that time, consumer
> confidence and utility, etc. In a market with such S-curves, the efficiency gained can be reduced or amplified
> over your straight lined graph, depending on how the lines intersect.
As you surely realize, I used lines because they're shitloads easier to
draw. If you want to draw out your own graphs using S-curves, go right
ahead. I'm anxious to see them. So long as your curves don't violate any
principles, they will show the same qualitative results as mine do.
> Because it is thanks to piracy in other countries, it is unprofitable. Since a piracy enforcement agency, whether
> a government or a PPL, cannot operate in a monopoly state jurisdiction without the expressed approval of that
> state monopoly, and must operate within the legal guidlines set by that jurisdiciton, any state that wishes to
> shelter piracy can make it unprofitable to enforce IP rights. Just as most terrorism and guerrilla warfare is not
> successfull without the shelter and support from states outside the zone of conflict, IP piracy is only sucessful
> with the direct participation, shelter and support of state monopolies.of power. Dont' make me label you a
> pro-statist.
I don't understand. It would be unprofitable to produce tamper-proof
software BECAUSE it is more tampered with in other countries? Care to
back this up a little more?
> And it is just as easy to have programs written on a CD which automatically run when the CD is accessed. This is
> how viruses get away with their mischeif. While AV software may be able to wipe a virus on a floppy, a) on a CD
> there is plenty of room for multiple copies to watch over each other, and b) a CD cannot be overwritten that
> easily.
Again, all I have to do is build a computer emulator that copies instead
of executes (especially easy when we're talking about CDs) and all the AV
software in the world wont help you as I burn myself an identical CD with
all of the AV software intact, and pop it into my real computer. The new
CD is no different from the old CD as far as the software can tell; what
will the AV programs detect if they are never executed?
> > Beg your pardon? By precisely how much will the incentive to invent
> > change?
>
> It can be measured by the percentage of patents issued which are brought to market with the help of measureable
> capital investment. Since capital invested in such a risk investment will demand a given return on that
> investment, if the total on investment is reduced by the lack of IP protection to the point that the given rate
> of return on investment demanded by the capital market will dictate that the amount of money an inventor can
> raise from investors for a given invention will decrease below the point required to bring the given invention to
> market, then the inventor will not bring that particular invention to market. I don't have any data right now, so
> I am not making any blind guesses (unlike you).
Yes, this is all true. However, how do you propose to measure the
demanded return on investment on IP in America without IP laws, all else
being equal? China won't cut it; planned economies don't let this
mechanism work at all. You've got to use America and all else must be
equal in order for this mechanism to apply; in other words, you can never
make this measurement.
> > My argument has been qualitative, arguing that the incentive to
> > invent would decrease, but not specifying absolutely by how much.
>
> Which is simply a cop out.
Well, go get your data and we'll see how YOU do.
> And I say that since the supply of inventions will contract markedly, your pricing will not change much, until
> consumer demand outstrips supply, due to population growth and limited resources, and no change in resource
> utilization efficiencies do to stagnated tefhnological growth. At which point you will have inflation and
> economic stagnation.
Actually, interestingly enough, NOT if you are right about that natural
monopoly thing. In that case, the big monopoly on invention-commodities
will have an incentive to pay a premium for inventions, up to but not
exceeding the area of the blue shape in my incentive analysis. Inventors
will therefore have an incentive to invent given by this shape, just as it
is today. That is, in that case, the incentive to invest is EQUAL in both
systems, as well as the deadweight loss. The only difference, then, is
the cost of enforcement; again, the vote in my favor, not yours.
Meanwhile, even if natural monopoly does not take place, the blue
trapezoid under competition expands when the demand for inventions
increses, so there IS a demand effect on the incentive to invest; again,
smaller, but not zero.
> Yet any web server/router/backbone owner will have a contract with an insurance company to protect their assets,
> especially if they have capital leins on the equipment. In a libertarian world, any sane insurance company will
> require that the network asset owner subject himself/herself to the authority of their contracted PPL. Most or
> all PPLs will have interlocking contracts of enforcement. Thus you will be able to track anyone on the net that
> vilates your IP contract, and you are covered by your IP protection clause in your insurance policy. Anyone you
> license your property to will have to subject themselves in your license agreement to ajudication and enforcement
> of your self licensed IP rights, ergo your PPL can do a network audit on your customers networks, on their ISP
> routers, etc. In an internet world, the cost of IP enforcement will go down remarkably.
I think I see. So your PPA will go to the PPA(s) of all of those who
reside on my anonymous remailer chain and get them to tell who I am. The
problem with this policy: this is effectively key escrow, which has been
proven many times over to be inefficient. If you have to hand over your
keys when the government/PPA says so, you can't do electronic commerce
securely. Period. If e-commerce is more efficient than physical commerce
(And there is a great deal of evidence that this is the case,) then
enforcing key escrow results in economic inefficiency. So again I fall
back on Friedman's rule of thumb: if it is economically inefficient, it
will not be enforced under anarcho-capitalism.
> > The cost of invention is sunk, which means that it doesn't change the
> > shape of the supply curve. Competitive disadvantage only takes place when
> > your supply curve is higher than the other guy's supply curve, so this is
> > clearly not the case. Instead, we see from the graphs that invention is
> > only profitable if the increased revenues from invention outweigh the sunk
> > cost of invention; this incentive is smaller than that under a monopoly,
> > but not zero.
>
> I think that this is a totally bogus claim. You are completely off your rocker here.
What? That sunk costs do not result in competitive disadvantage?
Competitive disadvantage is a technical term which refers only to the
supply curve, which is really not affected by sunk costs, honest to
goodness; you can go look it up. Perhaps you meant something else? The
average cost curve?
> Simplistic, basic microeconomic charts, any skilled economist will tell you, have ABSOLUTELY NO RELATION to any
> real world microeconomic scenarios. To make generalizations from a generalization like you are doing is the
> height of ignorant arrogance.
Yet they will agree that microeconomic efficienccy does lead to increased
GDP, ceteris paribus. It's the ceteris paribus which most economists will
catch you on, but since all I was trying to prove is that IP laws are
economically bad for us compared to not having them, microeconomic
efficiency is the perfect choice.
This archive was generated by hypermail 2.1.5 : Fri Nov 01 2002 - 14:49:08 MST
