| |
|
Economic Impact of the Personal Nanofactory
Deflationary forces resulting from mass availability of desktop personal nanofactories can be opposed by inflationary forces competently initiated by governmental monetary authorities.
Originally published in Nanotechnology
Perceptions: A Review of Ultraprecision Engineering and Nanotechnology,
Volume 2, No. 2, May 8, 2006. Reprinted with permission on KurzweilAI.net,
May 8, 2006.
Is the advent of, and mass availability of, desktop personal nanofactories
(PNs) [1] likely to cause deflation (a persistent decline in the
general prices of goods and services), inflation (a persistent general
price increase), or neither?
A definitive analysis would have to address: (1) the technical
assumptions that are made, including as yet imprecisely defined
future technological advances and the pace and order of their introduction;
(2) the feedback-mediated dynamic responses of the macroeconomy
in situations where we don’t have a lot of historical data
to guide us; (3) the counter-leaning responses of existing power
centers (corporate entities, wealthy owners/investors, influential
political actors, antitechnology-driven activists, etc.) to the
potential dilution of their power, influence, or interests, including
their likely efforts to actively oppose or at least delay this potential
dilution; (4) legal restrictions that may be placed on the widespread
use of certain technological options, for reasons ranging from legitimate
public safety and environmental concerns to crass political or commercial
opportunism; (5) the possibility (having an as yet ill-defined probability)
that nanotechnology might actually “break the system”
and render conventional capitalism obsolete (much as solid state
electronics obsoleted vacuum tubes), in which case it is not clear
what new economic system might replace capitalism; and (6) the changes
in human economic behavior that may result when human nature itself
may have changed.
A definitive answer is beyond the scope of this essay. Here, we
take only a first look at the question.
Price/Performance Assumptions
Our preliminary analysis begins with an assumption that at the
end of a 20-year period of introduction, almost every household
in a given developed country has purchased a PN. The PN will be
capable of building any manner of consumer goods using simple molecular
feedstock such as acetylene or propane gas that will be piped into
the home via a utility connection, similar to present-day hookups
that deliver natural gas, water, and electricity. There are other
delivery scenarios such as bottled gas feedstock, and more self-sufficient
feedstock provisioning scenarios such as nanoblock premanufacturing,
solar-powered recycling, biomass/biowaste extraction, or even atmospheric
extraction, but these will be set aside here both in the interest
of simplicity and because they may be heavily regulated [2] or in
some cases even declared illegal [3].
If we further assume that (1) the price to acquire the PN is approximately
US$4400 (see below), (2) the PN has a mass of 10 kg and produces
consumer products at the rate of 1 kg/hr [1], and (3) the PN is
operated 50% of the time throughout a useful lifetime of 10 years,
then the PN during its useful life produces 44,000 kg of consumer
products which then have an amortized capital cost of $0.10/kg,
a cost that is built into every product manufactured by the PN.
The $4400 price point for the PN was taken as a plausible figure
that might reasonably be chosen by a U.S. manufacturer. This price
point cannot be too cheap or there is no profit, nor too expensive
or there are no buyers. For comparison, $4400 will purchase a good
quality large-screen TV, a high-end refrigerator/freezer, or a topnotch
laptop computer in the U.S. today. The PN will be a versatile appliance,
able to manufacture whatever is deemed legal in the 21st century
such as: (1) consumer goods including nondurables such as food and
durables heavily laden with nanosensors, nanomotors, nanopumps and
nanocomputers, (2) all the patterned sheets or chunks of diamond
that anyone might want, and (3) some kinds of medical nanorobots
for personal use, though these may be heavily regulated. We assume
that the PN will not be allowed to manufacture contraband, nor various
types of weapons systems including ecophages [4], or more PNs (which
would nullify the R&D funding and manufacturing business model).
If the public is not allowed to manufacture PNs using PNs [5], then
the production cost of a PN using a PN becomes almost irrelevant
to the retail price of a PN. The manufacturer may charge $4400 for
a PN even if it only costs them $10 to make one (see below), using
the difference to pay for showrooms and sales staff, marketing and
advertising, legal costs for defending the brand and a monopolistic
pricing regime, product liability insurance costs (which could be
substantial), warranty and servicing costs, online and print publications
including consumer how-to books and magazines, websites and online
help desks, executive overhead and corporate perks, and perhaps
modest dividends from profits for the shareholders.
Assuming the average person in the developed world consumes 2000
kcal/day of food, and taking the average energy density of food
(arbitrarily weighting protein:carbohydrate:fat in a 4:3:3 ratio)
as 24 million J/kg [6], then the person is consuming about 130 kg/yr
of food. Further assuming this average person consumes four grocery
bagfuls per week of nondurables with each bag containing 2 kg of
useful product, then the average person requires 400 kg/yr of consumer
nondurables (of which 130 kg/yr is food). The PN is assumed to produce
4400 kg/yr of consumer products. Given that the average person in
an industrialized economy needs 400 kg/yr of nondurables, or 1600
kg/yr for a household of four people, this leaves 2800 kg/yr either
for increased nondurables consumption or for the manufacture of
desired consumer durables. Durables might include clothing, appliances,
furniture, and cars. Large automobiles that weigh 2000 kg today
might weigh as little as 200 kg if made of much stronger diamondoid
materials [7], so the production budget would allow up to 14 diamondoid
cars per year to be built. Thus, a single PN per household with
the above parameters is probably sufficient to satisfy all reasonable
household needs for residents of industrialized countries.
Cost of PN-Manufactured Goods
The base operating cost of a working PN, and hence the price of
anything that can be manufactured by the nanofactory, should approximate
the cost of the material and energy inputs. Of course, “cost”
is not price—in a capitalist economy, prices of goods are set
by competitive markets. In a stable equilibrium economy, the price
of manufactured goods that are in demand cannot long persist below
the base cost of the material inputs. But added to those base input
costs, and hence indirectly added to the price of a good, will also
be various intangible costs, which for the PN might include some
or all of the following:
(1) acquisition financing or opportunity costs;
(2) PN licensing costs;
(3) the cost of product-description information not in the public
domain (e.g., IP fees, per-use fees, per-bit fees);
(4) communications toll charges for PN data links, with flat baseline
fee varying by transmission speed, fidelity, security, and availability
for product data downloads;
(5) fee for dangerous-product construction blocking (e.g., governmental,
parental);
(6) fee for contraband or socially-taboo construction blocking
(e.g., governmental, parental, religious, etc.);
(7) regulatory costs (e.g., environmental impact/mitigation, thermal
pollution credits, content policing, federal mandates enforcement);
(8) federal/state taxes (e.g., sales/use, value-added, property,
excise, luxury, inheritance, estate, gift, etc.), State Nanofactory
Commission tax, and surcharges for unrelated social purposes;
(9) local taxes (e.g., feedstock fire services, toxic gas mitigation,
metering services, low-income subsidies, universal access subsidies,
9-1-1 emergency phone number fees);
(10) federal protective trade and domestic subsidy-support tariffs;
(11) federal economic stabilization surcharges;
(12) fee for regular government-mandated safety inspections, analogous
to annual smog inspections commonly required for automobiles;
(13) international surcharge for disadvantaged nations, administrated
through United Nations auspices (perhaps imposed as single fee collected
at time of sale, or paid as annual international tax);
(14) registration fee for all PNs, analogous to automobile registration
fees;
(15) the cost of government mandated insurance for accidents or
injuries to others as a result of products manufactured by the PN,
similar to the automobile personal injury insurance required in
many states; and
(16) miscellaneous other fees and expenses. A precise quantification
or estimate of all these costs cannot easily be made, but prorated
intangible costs of at least $0.20/kg appear plausible [8].
This means that even if the cost of material and energy inputs
fell to essentially zero—say, through the use of recyclable
nanoblocks that could create the potential for cheap disassembly
and re-use of nanoblocks, thus recovering the majority of the energy
investment in a given amount of diamondoid mass—there would
still be an amortized capital cost of $0.10/kg plus a fixed intangible
cost of $0.20/kg built into all products manufactured by the PN,
giving an irreducible total minimum cost of $0.30/kg. (The fixed
intangible $0.20/kg component should prove relatively insensitive
to fluctuations in the general price level.)
Bulk petroleum-based hydrocarbon gases like methane (CH4)
or propane (C3H8) can cost as little as ~$0.17/kg
[9a] or ~$0.31/kg [9b], respectively, at wholesale, but the retail
price for home delivery—as would be required to supply a PN—was
$2/gallon for liquefied propane gas in December 2005, or $0.90/kg.
Nanofactories may need to be restricted to such feedstocks for reasons
of public safety (because development of more flexible feedstock
technologies might facilitate malicious development of self-replicating
systems [10], e.g., by using nanofactories to construct them). For
a complete PN feedstock gas, small amounts of other gases containing
additional non-carbon atoms needed for mechanosynthetic processes
still must be added to the mix, raising feedstock costs by an additional
but unknown amount—for example, $1.50/kg for hydrogen (H2)
gas, $110/kg for Si atoms in silane (SiH4), $354/kg for
P atoms in phosphine (PH3), and $1150/kg for Ge atoms
[11].
Theoretical studies of diamond mechanosynthesis [12] often assume
the use of a more chemically convenient carbon-rich hydrogen-poor
precursor carbon-source molecule such as acetylene (C2H2).
Note that the chemical energy of all the interatomic bonds in one
kilogram of acetylene is about 67 MJ, implying a relatively low
theoretical cost of $1.85/kg at the current electricity price of
~$0.10/kW-hr, even though large-volume commercial acetylene prices
are typically $18/kg [13] because of costly special tanks required
by law for the safe transport of this unstable high-energy gas.
If electricity could be obtained more cheaply, then in principle
the cost of acetylene could fall accordingly. For instance, if electricity
costs dropped 10-fold to $0.01/kW-hr, then acetylene costs could
fall to $0.185/kg, and ignoring the cost of ancillary non-hydrocarbon
gases the minimum total cost of PN manufactures would then decline
to $0.185/kg + $0.10/kg (capital cost) + $0.20/kg (intangibles)
= $0.485/kg. If electricity costs dropped 100-fold to $0.001/kW-hr,
then acetylene costs could fall to $0.0185/kg, implying a total
cost for PN manufactures (again ignoring the cost of ancillary gases)
of at least $0.3185/kg. The cost of electricity cannot as a practical
matter decline to zero [14] because energy is an inherently scarce
resource, electrical generators have input material costs and fixed
capital/site expenses, transmission lines require maintenance, waste
heat and other pollution byproducts must be properly disposed of,
natural environments should not be despoiled, and so forth. But
even if the cost of electricity did fall to zero, the total cost
of PN manufactures would still be at least ~$0.30/kg as noted earlier.
Supporting the plausibility of our assumption of a future ~$0.10/kW-hr
electricity price is the historical fact that U.S. prices of electricity
in nominal dollars have been remarkably constant over the last 100
years. This period has seen economic booms and depressions, war
and peace, and tremendous technological change and gains in industrial
productivity. Specifically, the average price in nominal dollars
for delivered U.S. residential electricity was $0.105/kW-hr in 1907
[15] and $0.099/kW-hr in August 2005 [16]—that is, no change.
(The U.S. Consumer Price Index (CPI) has risen 22-fold during this
same period which means that the real cost of residential electricity
has declined 22-fold over the last 100 years, but the nominal price
has stayed constant at 10 cents. Industrial electricity prices have
behaved similarly.) We can consider an idealized alternative scenario
for decentralized energy production in which solar energy is captured
by inexpensive thin solar panels placed as ground cover over cheaply-priced
vacant rural land, initially bringing the cost of solar-derived
electricity down to ~$0.001/kW-hr [17]. Since this cheap electricity
can be immediately resold at the prevailing $0.10/kW-hr price in
competition with existing utility suppliers, a likely rapid escalation
of free-market vacant land values or the imposition of regulatory/protectionist
tariffs by government should quickly close much if not most of the
price gap. Of course, environmentalists may prevent the project
from going forward at all for a variety of reasons—such as
(1) the possible death of covered vegetation due to sunlight deprivation;
(2) the alteration of local weather patterns and watershed hydrodynamics
including soil drainage, erosion, and water tables; (3) the destruction
of natural “view easements” with solar collectors that
are aesthetic eyesores; (4) the interference with insect and bird
navigation; and so forth. Ocean-based solar collection systems have
similar costs with different sets of environmental issues.
Optimistically assuming that feedstock gases can be purchased by
the consumer for a delivered price of $0.70/kg, and adding in the
amortized initial capital outlay of $0.10/kg to buy the PN, and
another $0.20/kg for prorated intangible costs, then the cost of
manufacturing consumer products using a personal nanofactory should
be about $1/kg. This is moderately cheaper than the current cost
of nondurables like bread and meat at the grocery store ($3-$5/kg)
[9b, 18], and 2-3 orders of magnitude cheaper than the current cost
of commonplace durables such as a good pair of shoes ($100/kg) or
a good laptop computer ($1000/kg) [9b]. Supplemental licensing fees,
intellectual property restrictions, taxes, regulatory tariffs and
so forth may drive up the cost of some luxury or specialty items
to well above the $1/kg price of manufactured goods, but the current
prices of expensive items such as beluga caviar ($796/kg), pearls
($27,000/kg) and pharmaceutical interferon ($190,000,000/kg) [9b]
(all made from intrinsically cheap atoms) also have room to fall
many orders of magnitude. Over time, most basic consumer goods should
become commoditized with prices asymptotically approaching production
+ capital + intangibles costs, though perhaps with some government
tariffs on personal manufacturing zeroed out on basic staples and
necessities to accommodate the neediest citizens.
A sudden lowering of manufacturing costs might lead to deflation.
An interesting historical precedent for technology-driven hyperdeflation
of a commodity price is the example of aluminum metal. During most
of the 1800s, aluminum could only be produced commercially using
an expensive chemical process involving highly reactive alkali metals.
As late as 1884, the pure element still was regarded as a precious
metal with a price of $16/pound, equivalent to $23/oz in current
2006 dollars. Aluminum was used for fine tableware (much like sterling
silver, then and today) and the metal was chosen to cast the small
pyramidal cap atop the Washington Monument [19]. In 1886 a cheap
new electrolytic technology called the Hall process was discovered
which by 1914 had sent the price of the pure metal plunging to just
$0.19/lb, a price floor determined by the cost of electricity. This
represents a nearly 100-fold single-commodity price hyperdeflation
in less than 3 decades—the technology-enabled price drop could
occur because aluminum atoms are plentiful and widespread in nature,
unlike, say, gold atoms [24]. Still, a bit of perspective is needed
here. Average U.S. consumer prices have only risen 50-fold during
the last 300 years and 10-fold in the last 60 years [20], a +4%/yr
long-term rate in modern times; the U.K. has a similar record. The
sharpest general price deflation in U.S. history occurred during
1780-89 when prices declined 100-fold from temporarily inflated
Revolutionary War profiteering highs [20]. But except in similar
cases involving bursting economic bubbles, deflations are generally
mild and rare—across the worldwide Great Depression during
1920-1933, U.S. consumer prices dropped by -35% (not even 2-fold)
[20].
In light of the above considerations, a conservative assumption
is that the introduction of personal nanofactories over a time period
lasting, say, two decades will result in the average prices of consumer
nondurables falling perhaps 5-fold from today’s prices, and
the average prices of consumer durables falling perhaps 100-fold.
How will this affect the overall inflation rate? Perhaps surprisingly,
not much.
The Deflationary Impulse
In the 2005 U.S. economy, consumer durables accounted for 12.8%
of all Personal Consumption Expenditures (PCE), nondurables were
29.0% of PCE, and services were 58.2% of PCE [21]. If the aforementioned
5-fold and 100-fold price declines occur over a 20-year period during
which PNs are introduced at a linear rate of market penetration
growing from 0% to 100%, and if the services component of PCE continues
to inflate at the long-term historical +4%/yr rate, then the net
inflation rate for all of PCE falls from +4%/yr in the last year
prior to PN introduction to +1.4%/yr during the 20 year period in
which PNs are introduced into the economy. In other words, there
is no actual deflation, just -2.6%/yr worth of disinflation. By
the end of the 20-year PN introduction period, services make up
95.5% of the private consumption economy, nondurables are 4.4% of
PCE, and durables have almost vanished at 0.1% of PCE. In year 21,
inflation returns to the long-term historical +4%/yr trendline.
It’s unclear what effect PNs might have on inflation in the
service sector [22]. On the one hand, it can be argued that services
may experience some disinflation during the PN introduction period
because PNs in the hands of service providers will make those businesses
more efficient, hence able to lower their prices for competitive
advantage. On the other hand, the principal determinant of the price
of services is the cost of human labor. With workers generally more
affluent and less desperate for necessities, labor wage rates might
actually need to rise, not fall, to induce PN-goods-satiated indolent
citizens to remain in, or return to, the workforce. Of course, if
most services no longer needed to be provided by humans—e.g.,
if most services could be provided by AI-enabled robotic slaves
cheaply manufactured by PNs, thus allowing humans to become pure
consumers with no need to produce anything of value or to serve
anyone—then the value of human labor would have effectively
fallen close to zero and the cost of services would then fall to
the same degree as durable goods, since robots and AI-embedded computers
are also durable goods. The degree to which humans will find robot-only
services to be socially and aesthetically acceptable is unclear,
as is the legal status of enslaved AI-enabled robotic “persons”
[52], and both issues are beyond the scope of this essay.
The Inflationary Impulse
However, this is not the end of the story. Much like the extreme
but narrow economic dislocations experienced by developed countries
when suddenly confronted with cheaper external sources of goods
(often causing companies to outsource), personal nanofactories are
the economic equivalent of below-cost trade competition for most
retail manufactured products. One likely result is that much of
the manufacturing, agriculture and wholesale/retail trade sectors
(about 40% of all employment in developed economies [20]) could
face continuing downsizing pressures throughout the 20-year introduction
period of the PNs. Up to 40% of all workers could be affected and
they may experience wage stagnation, job loss, and a need for additional
education or retraining for gainful employment in the service sector—much
as displaced agricultural workers fled to the manufacturing sector
a century ago.
Consider the likely governmental response to these displacements.
If a significant fraction of manufacturing-sector employees lose
their jobs and thus have no money to pay for necessities, then as
their unemployment or union benefits and savings become exhausted
these people will naturally turn to the government for further assistance
during their period of career realignment.
They will demand to receive transfer payments of subsistence cash
to buy the commodity goods (e.g., food, clothing) and services (especially
nanomedical care) that they need but can no longer afford. Politicians
will feel little restraint against printing money as fast as it
is needed to make voters whole, equivalent to Milton Friedman’s
famous proposed “helicopter drop” of cash [23].
This forced expansion of the money supply will be inflationary.
But by how much? In the scenario analyzed earlier, over a 20-year
PN introduction period the economy loses a combined average of $160B/yr
in consumer spending on durables and nondurables, rather than gaining
a combined average of $220B/yr that would have occurred in the absence
of the PNs. In the simplest case, the government (using the United
States as an example) would just issue the swing difference ($380B/yr)
as direct grants to displaced people in the affected industries.
This would increase the current growth rate of the 2005 U.S. M2
money supply from $300B/yr (+5%/yr) to $680B/yr (+10%/yr), perhaps
adding about +5%/yr to the annual CPI inflation rate (assuming for
simplicity no multiplier effect) during the 20-year PN introduction
period. Combining this new government-induced inflation with the
disinflation of -2.6%/yr caused by the rapid fall in the prices
of PN-manufactured goods (see above) gives a net inflation rate
increase of +2.4%/yr, i.e., from the current +3%/yr rate to a slightly
higher +5.4%/yr rate of U.S. consumer price inflation. The macroeconomic
effects of this change might be hardly noticeable to non-economists.
Even this small increment to the CPI inflation rate can be entirely
neutralized (at least on an economy-wide basis) if the government
increases the M2 money supply growth by only an average of $200B/yr
(rather than $380B/yr) of direct grants of assistance to displaced
workers. In this case, the +2.6%/yr inflationary impact of increased
government outlays would exactly counterbalance the -2.6%/yr deflationary
impact of the PNs on the general price level, resulting in a zero
net change in the overall inflation rate. The missing $180B/yr not
directly paid to the displaced workers would be offset in most cases
by the incremental appreciation of tangible assets owned by them,
that cannot readily be manufactured by nanofactories with any clear
cost advantage (e.g., houses (whose conventional manufacturing cost
is already $1/kg [9c]), land and other real estate, precious metals
[24], and so forth). The 2004 U.S. national home ownership rate
was 69.0% [25], and an extra +2.6%/yr of general inflation applied
to the current $19.1T worth of owned residential real estate [26]
amounts to an annual appreciation gain of $500B/yr.
Deflation and Inflation Can Be Balanced through Government Action
Note that because government has the ability to “lean against
the wind,” then, within broad limits, the inflationary effects
of increased government outlays can always absorb the deflationary
effects of PNs, even if cheaper electricity allows the cost of PN-made
goods to fall from $1/kg to the irreducible minimum of $0.30/kg.
The same result occurs if PNs are introduced into the economy over
a period substantially shorter than 20 years. For example, a faster
10-year introduction period produces a mild -0.8%/yr deflation over
that period [27]. In this case, the +4.8%/yr inflationary impact
of $365B/yr of increased government outlays would exactly counterbalance
the -4.8%/yr deflationary impact of the PNs on the general price
level, again resulting in a zero net change in the overall inflation
rate.
In a worst-case scenario, an unrealistically fast 1-year introduction
period of PNs would produce a 1-year -33.5%/yr deflation rate (in
the absence of any government response), almost exactly matching
the total peak-to-valley deflation that was recorded during the
Great Depression (see above). But rather than allowing money supply
to naturally decline as in the 1930s (which enhanced the contraction
in real economic growth), this time the governmental monetary authorities
would probably increase the M2 money supply to match the $3,320B
in lost final sales. That’s the equivalent of handing every
man, woman, and child in the U.S. a check for $11,600. The one-time
stimulative action would increase the M2 money supply by about +50.1%
in a single year.
Such stimuli are not unprecedented in U.S. economic history [28].
For instance, a two-year +50.1% M2 increase during 1942-44 at the
beginning of WWII produced only +4%/yr CPI inflation in those years,
in part due to the simultaneous implementation of price controls
[15, 20]. A one-year +60.6% increase in total U.S. currency during
1862-63 at the beginning of the Civil War produced +23%/yr CPI inflation
across those years, though by 1878 consumer prices had returned
back down to 1862 levels [15, 20].
In the above worst-case scenario, the one-time issuance of government
largesse would permanently balloon federal debt [29] from $7,900B
(2005) to $11,200B in a single year. Assuming interest rates remain
roughly unchanged along with the inflation rate (since interest
rates loosely track the inflation rate), federal interest payments
on the debt [30] would only rise from $350B/yr (2005) to $500B/yr,
a $150B/yr federal budget expenditure increase that seems modest
in comparison to the usual $3,900B/yr (2005) in U.S. federal outlays
[31]. Even a modest increase in long-term interest rates, due to
the increased debt level, is probably tolerable to the federal budget.
In the low-probability 1-year PN introduction scenario, federal
outlays would spike +85%, from $3,900B to $7,220B for one year,
but this is historically comparable to the +51% increase in federal
outlays during 1862-63 and the +172% increase in federal outlays
during 1942-44.
Conclusions
Deflationary forces driven by advances in molecular manufacturing
(MM) can be opposed by inflationary forces competently initiated
by governmental monetary authorities. This allows the two forces
to remain roughly in balance, with the incremental inflation at
the general price level remaining close to zero as PNs are introduced.
Since an MM-rich economy will be dominated by services and information,
not goods, our expectation is that the prices of services and information
might rise very slightly as the prices of PN-manufactured goods
falls significantly. For example, if services and information comprise
95% of the economy and goods are only 5% of all sales, then a deflationary
-20% decline in the prices of goods can be largely offset by an
inflationary rise of just +1% in the prices of services and information.
Manufactured goods will represent a decreasing portion of the total
economy as PNs are introduced, hence the influence of such goods
on the general price level in the overall economy will tend to decline
as the diffusion of MM throughout the economy approaches saturation.
Similarly, services and information will represent an increasing
fraction of the total economy as PNs are introduced, eventually
reaching near 100%. Hence the amount of incremental inflation in
the prices of services and information that is required to offset
the deflation caused by falling goods prices will also tend to decline
toward zero as the diffusion of MM throughout the economy approaches
saturation [32].
Both deflationary and inflationary levers exist in the economy.
Either may be pulled on harder, as needed, to maintain a stable
price regime (or at least to ensure “soft landings”),
up to the point where the whole economic system breaks down. That
is, a monetary expansion (possibly creating soaring public debt
politically rationalized by a Presidential proclamation of a “War
on Deflation”) can continue until the required inflationary
counterstimulus becomes so enormous that the financial stresses
induced by large and rapid monetary flows literally break the economic
system, possibly in an extended analog of the stock market crash
of 19 October 1987 [33]. Quantifying this threshold systemic breakage
point—in both local and global economies—should be an
urgent matter for future econometric and economics-history research.
Acknowledgments
The author thanks Steve Burgess, Steffen Christensen, Tom Craver,
Ray Kurzweil, Chris Phoenix, Mike Treder, Michael Vassar, and Brian
Wang for helpful comments on earlier versions of this manuscript.
References and Footnotes
1. K.Eric Drexler, Nanosystems: Molecular Machinery, Manufacturing,
and Computation, John Wiley & Sons, NY, 1992, Section 14.4.
2. A simple calculation (http://www.nanomedicine.com/NMI/6.5.7.htm#p4)
of the hypsithermal limit for Earth, when combined with a few other
reasonable assumptions, suggests that a per capita allocation of
about 10 kg of active nanomachinery will generate about the maximum
amount of extra thermal pollution that our planet can handle without
sustaining ecological damage. This limit or something like it may
someday be enshrined in international law.
3. For example, consider the case of a private individual who sets
up equipment to extract CO2 from the air and combine it with water
to synthesize his own acetylene gas. Authorities may not allow individuals
to set up unregistered and unregulated systems (i.e., acetylene
generator + PN) that are capable, in effect, of metabolizing air
into diamond. Such a system would amount to the construction of
an aerovore (http://www.rfreitas.com/Nano/Ecophagy.htm#Sec8-2_GrayDust)—a
free-range replicator that can copy itself by feeding on air—because
it would be fairly trivial to miniaturize the proposed acetylene
synthesis system down to a micron scale system, and because basic
scaling designs for submicron-size self-replicating molecular assemblers
that use acetylene as feedstock have already been published (http://www.MolecularAssembler.com/KSRM/4.11.3.htm).
Since the proposed system, once miniaturized, could be used to effectuate
a “gray dust” scenario thus precipitating a global ecological
catastrophe, it is likely that systems of this type (and their obvious
precursors) may be deemed illegal except when deployed in highly
regulated and extensively monitored (probably government) facilities,
probably with an international supervisory body such as the IAEA
continuously watching the entire proceedings via ubiquitous cameras.
Because of the ecophagic risks, the building of such a device by
an unauthorized and unsupervised person might even be viewed as
an act of terrorism.
4. Robert A. Freitas Jr., “Some Limits to Global Ecophagy
by Biovorous Nanoreplicators, with Public Policy Recommendations,”
Zyvex preprint, April 2000; http://www.rfreitas.com/Nano/Ecophagy.htm
5. It seems unreasonable to presume that the existing capitalist
power structure will knowingly fund the development, and then later
willingly allow the widespread sale and unrestricted use, of a product
which has no economic value whatsoever to capitalist manufacturers—and
even worse, will likely threaten the economic value of all other
product lines of all manufacturers. It may be hard to make money
selling a product to consumers that can build more copies of itself,
thus obviating future sales. This is amply demonstrated by the commoditization
of agriculture, a “(bio)manufacturing” sector of the economy
that relies heavily on self-replicating (bio) machinery. (Observe
that big agribusiness survives in part by selling farmers new seeds
each growing season.) It is unlikely that PNs will be sold that
can make other PNs, because these could be a serious threat to the
existing capitalist system and thus will be vigorously resisted
by all possible means, once the power structure fully appreciates
the threat and believes the threat to be both possible and imminent.
Hacker-proofing is likely to be strictly enforced; the technologies
that will make this possible are already being developed in the
private sector IP area today. And governments of all types will
be motivated not to let their citizens have too much freedom to
manufacture whatever is locally defined as “contraband”
or “taboo”, so a fair degree of international cooperation
in enforcing these restrictions can be expected. Consequently, restricted-use
(and monitored-use, perhaps even publicly-published use) technologies
in the personal manufacturing area cannot be ruled out and in fact
seem highly probable in view of the security and economic risks.
It is possible to imagine an alternative scenario in which PNs
are given away for free, but cannot build anything unless you pay
a subscription fee for software and product-build instructions for
all the products you want, similar to the situation today in some
companies where cell phones are given away free as a marketing ploy
to sell subscription services. The success of this business model
applied to PNs would strongly depend on the manufacturer having
absolutely secure control over the source, and the use, of the purchased
software and product-build information. We will not further evaluate
here the economic consequences of this or other alternative business
models for PNs.
6. “Table 6.1 Energy Storage Density for Various Chemical
Fuels,” in Robert A. Freitas Jr., Nanomedicine, Volume I:
Basic Capabilities, Landes Bioscience, Georgetown, TX, 1999,
p. 141; http://www.nanomedicine.com/NMI/Tables/6.1.htm
7. The mass of diamondoid products whose utility is primarily a
function of their strength may decrease by between two and three
orders of magnitude. (If we are manufacturing food, there will likely
be almost zero mass reduction in the products.) The only published
detailed design analysis for the comparative mass reduction of any
diamondoid products is for the respirocytes (http://www.foresight.org/Nanomedicine/Respirocytes/index.html),
which are calculated to achieve a 1620-fold mass reduction compared
to a volume of red blood cells of equal functional capacity. But
respirocytes are a microscale durable product. As of late 2005,
no similar detailed analysis had yet been undertaken for macroscale
consumer durables such as shoes, bicycles, automobiles, and the
like. So, while it is reasonable to assume a 10-fold mass reduction
in consumer durables, our ability to achieve 1000-fold mass reductions
in macroscale durables (e.g., a 10-gram bicycle) remains somewhat
speculative.
Another scaling issue is whether future consumer durables, fabricated
using molecular manufacturing (MM) techniques, will be given the
same functionality as today’s products, or somewhat greater
functionality, or will have dramatically augmented functionality
(superfunctionality) due to the incorporation of a significant mass
fraction of active nanomachinery. Over time, products will probably
emerge across the entire spectrum, ranging from current functionality
to superfunctionality.
In the case of MM-fabricated superfunctional nanomachinery-rich
consumer products, a possible further complication of reducing product
mass is the reduction in thermal capacity that this causes. Thermal
capacity can be recovered by incorporating high heat capacity bulk
materials such as water into the product or adding extra conductive
structures in contact with external heat sinks, but either solution
increases total product mass. Additionally, a superfunctional macroscale
product containing a large quantity of active simultaneously-actuated
nanoscale machinery can generate significant waste heat (http://www.nanomedicine.com/NMI/6.5.3.htm#p11)—mainly
from computational tasks by nanocomputers but also from some kinds
of mechanical tasks, both of which may employ power densities many
orders of magnitude larger than those commonly found in biological
systems. The fraction of active nanocomponent mass that is present
in a superfunctional nanomachinery-rich macroscale product is a
design decision. This fraction may range from 0% (e.g., the product
is purely passive nanomaterials with no active nanocomponentry)
to 100% (e.g., the product is purely active nanocomponentry as is
approached by “utility fog”), depending on the application
and the objectives of the designer and his customers. For example,
a respirocyte will have 8.37 x 10-17 kg of active nanocomponents
(e.g., computers, sensors, pumps, and engines) for every 3.56 x
10-16 kg of total dry mass of final product (which includes
inactive mass such as hull and other support structure), so 24%
of the respirocyte dry mass is active nanocomponent mass. The problem
of heat removal for a superfunctional product of size-independent
power density grows more restrictive as product size increases (the
familiar Square-Cube law), making this a more important issue for
superfunctional macroscale products than for superfunctional microscale
products such as medical nanorobots (e.g., respirocytes). Design
solutions for superfunctional macroscale products incorporating
large mass fractions of active nanomachinery may include: (1) reducing
overall power density of the product, (2) reducing the fraction
of active nanomachinery present in the product, (3) operating active
nanomachinery present in the product at less than a 100% duty cycle,
or (4) adding cooling systems such as onboard refrigerators coupled
to radiators, disposable cooling packs, or hookups to external sources
of cooling fluids or other heat removal systems.
In the case of MM-fabricated consumer products having similar functionality
to present-day durable goods, cooling will be no more difficult
for these products than it is today—and may, in fact, be easier
because of the increased efficiency of nanoscale machinery compared
with traditional macroscale machinery. Cooling only becomes an extra
consideration when the designer adds significant new functionality
that doesn’t exist in today’s consumer products.
8. What is the impact of government taxation and regulation on
the cost of goods? There are many possible estimates, none of them
very precise but all suggesting a consistent range of impact on
costs. Starting with the broadest view, in the U.S. National Accounts
for 2005QIII the “taxes on production and imports less subsidies”
were 6.8% of Gross National Product [21], implying an equivalent
load on a $1/kg PN product of about $0.068/kg. Alternatively, in
2005QIII the total of “personal current taxes” plus “contributions
for government social insurance” were 20.4% of Personal Income
[21], which would imply an equivalent load on a $1/kg PN product
of about $0.204/kg. The average effective state and local tax burden
in the U.S. is 10.0% of income [34], which equates to a prorated
$0.10/kg assuming the value of PN output is treated as income at
$1/kg. Academic and institutional sources [35] estimate the economic
cost of all U.S. federal regulations during the 1995-2005 period
as $0.865T-$1.27T/yr, equivalent to a $0.07-$0.10/kg regulatory
burden on a $1/kg product.
In the author’s home state of California (USA), vehicle registration
fees are based primarily on purchase price. According to the state’s
online vehicle registration fee calculator [36], the purchase of
a new gas-powered vehicle in El Dorado County having a purchase
price of $4400 (identical to the purchase price posited for the
PN) would require payment of a total first-year registration fee
of $409, which includes a one-time sales/use tax of $319 on the
initial purchase and $90 for the first annual registration fee.
Over the 10-year useful life of a PN requiring registration, this
would amount to a total of $1219 in fees, which implies a registration
fee of $0.028/kg when amortized over the lifetime PN product output
of 44,000 kg. A listing of the items comprising the annual automobile
registration fee (most of which are suggestive or have obvious analogs
to a PN purchase) include Current Registration ($31), Vehicle License
Fee ($29), California Highway Patrol ($9), Smog High Polluter Repair
Fee ($6), Smog Abatement ($6), Air Quality Management District ($4),
Abandoned Vehicle Fee ($1), Auto Theft and/or DUI Crime Deterrence
Program ($1), County Service Authority for Freeway Emergencies Fee
($1), Fingerprint ID Fee ($1), and Reflectorized License Plate Fee
($1).
The author’s review of personal residential landline telephone
and electric bills for 2005 reveal essentially fixed government
fees, taxes, and other surcharges averaging $6/mo and $45/mo, respectively,
regardless of quantity consumed. Applying these amounts to a PN
producing 367 kg/mo would imply a minimum fixed burden of $0.02-$0.12/kg,
or higher if the full 367 kg product mass budget for the month was
not manufactured. Property taxes are usually imposed on a per-parcel
or per-house (or per floor-area) basis, regardless of the level
of use (e.g., ignoring number of occupants, children, cars, etc.)
of the real estate. An additional 1%-3%/yr personal property tax
imposed on a PN costing $4400 producing 4400 kg of products annually
would equate to a $0.01-$0.03/kg tax. Property taxes based on property
market value range from 0.13%-2.3% in the U.S. [34], 0.1%-2.0% in
Russia [37], and 1.4%-2.1% in Japan [38]; or as a percentage of
national income (GDP), ~1% in Germany, ~3% in France, and ~4% in
U.K. [39]. The City of Baltimore, MD, also assesses a 5.77% personal
property tax and a 5.77% utility tax [40].
Copyright royalties for owned written intellectual property (IP)
are typically 10%, which might suggest an additional $0.10/kg of
IP-related costs on products initially priced at $1/kg, except that
user fees might here be calculated according to the level of use
(e.g., the number of users or “seats” authorized to operate
the PN, as is common in commercial software contracts) rather than
as a percentage of sales, hence might not be tied to unit price
of product like a sales tax would be.
Financing a $4400 capital expenditure on an interest-only basis
at a 5% interest rate for a device producing 4400 kg/yr equates
to a minimum $0.05/kg financial overhead if the device is always
operated at the 50% production rate. The cost of lost opportunity
is calculated similarly, with the interest rate taking the role
of a discount rate on foregone alternative uses of the capital.
Note that the additional amortized capital cost of $0.10/kg still
applies because interest-only financing means the borrower still
owes the $4400 after 20 years of paying only interest. (These are
only crude estimates; an explicit calculation of discounted future
value, etc. would be needed to establish more precise estimates.)
Based on the above considerations, an irreducible intangible cost
of at least $0.20/kg for government fees, IP rights, and financing
costs seems plausible.
9. (a) Sudheer Indala, “Appendix A: Cost Estimation Procedure
For Carbon Monoxide,” Development and Integration of New Processes
Consuming Carbon Dioxide in Multi-plant Chemical Production Complexes,
Master’s Thesis, 18 December 2003; http://etd.lsu.edu/docs/available/etd-01212004-125820/unrestricted/14Appendix.pdf.
(b) “Table 6.1 Current bulk prices of nanoscale-structured
products,” in Robert A. Freitas Jr., Ralph C. Merkle, Kinematic
Self-Replicating Machines, Landes Bioscience, Georgetown TX,
2004, p. 203; http://www.MolecularAssembler.com/KSRM/Tables/6.1.htm.
(c) Is the cost of housing really irreducible? It is true that personal
residences of current design, if constructed using diamondoid materials,
might be buildable using an order of magnitude less processed mass
(possibly incorporating water, dirt, or stone ballast to provide
wind-resistance bulk) than present-day houses, hence reducing homebuilding
costs tenfold assuming a $1/kg rate. However, opposing this deflationary
factor will be consumers’ inflationary desire for houses with
more square footage and more extensive features—such as pools
and saunas, interior waterfalls and decorative ponds, reconfigurable
rooms, escalators and elevators to upper floors, greenhouses and
extensive outbuildings for hobby purposes, home theaters, ubiquitous
fireplaces and wall-sized “view” windows, hurricane and
earthquake-proofing, and so forth. And it is likely that any remaining
savings in house construction costs may be spent to bid up the price
of scarce land to build upon in desirable locations, since lower
house prices would increase the demand for more houses.
10. Robert A. Freitas Jr., Ralph C. Merkle, Kinematic Self-Replicating
Machines, Landes Bioscience, Georgetown TX, 2004; http://www.MolecularAssembler.com/KSRM.htm
11. Hydrogen (http://www.solartoday.org/2004/may_june04/h2_afford_it.htm),
silane (http://microlab.berkeley.edu/text/2K4YER/bob.html),
phosphine (http://www.aciar.gov.au/web.nsf/att/JFRN-6BN95D/$file/wp16.pdf),
germanium (http://www.answers.com/topic/germanium).
12. Ralph C. Merkle, Robert A. Freitas Jr., “Theoretical analysis
of a carbon-carbon dimer placement tool for diamond mechanosynthesis,”
J. Nanosci. Nanotechnol. 3(August 2003):319-324; http://www.rfreitas.com/Nano/JNNDimerTool.pdf.
David J. Mann, Jingping Peng, Robert A. Freitas Jr., Ralph C. Merkle,
“Theoretical Analysis of Diamond Mechanosynthesis. Part II.
C2 Mediated Growth of Diamond C(110) Surface via Si/Ge-Triadamantane
Dimer Placement Tools,” J. Comput. Theor. Nanosci. 1(March
2004):71-80. Jingping Peng, Robert A. Freitas Jr., Ralph C. Merkle,
James R. Von Ehr, John N. Randall, George D. Skidmore, “Theoretical
Analysis of Diamond Mechanosynthesis. Part III. Positional C2 Deposition
on Diamond C(110) Surface using Si/Ge/Sn-based Dimer Placement Tools,”
J. Comput. Theor. Nanosci. 3(February 2006):28-41; http://www.MolecularAssembler.com/Papers/JCTNPengFeb06.pdf.
13. “Price List: Millennium Acetylene Cylinders,” Taylor-Wharton/Harsco,
10 November 2005; http://www.taylor-wharton.com/pricing/current_pricing_master/current_pricing_master.pdf
and http://www.taylor-wharton.com/Pages/HP&A/Acetylene.htm.
Price is $17.91/kg for acetylene in largest WM-420 tank, capacity
13.0 kg of commercially pure acetylene gas at 250 psig and 70 oF.
Calcium carbide may be purchased (http://www.chembargains.com/inactive-chemical-supplies-11359.php)
at $400 per 100 kg, which when reacted with water (assumed free)
and the acetylene thus generated is collected (also assumed free),
the cost for the 40.56 kg of liberated gas would be $9.86/kg, not
counting the additional disposal costs for the calcinated sludge
byproduct.
14. An uncritical optimist heedless of many technical, political,
public safety, and other practical obstacles might propose obtaining
almost zero-cost energy by employing a single well-shielded 2500
micron3 selenophage of mass ~10-11 kg that
is delivered to the Moon by hitchhiking for free on the next scheduled
lunar lander mission that someone launches. The minimum possible
energy cost for such a launch approximates the gravitational potential
energy of a selenophage resting on Earth’s surface, or -GMEarthmSelenophage/REarth
~ 1.7 x 10-10 kW-hr which would cost ~$1.7 x 10-11
assuming today’s energy price on Earth of $0.10/kW-hr. The
pure energy cost to manufacture the one selenophage unit in a PN,
assuming $1/kg, is also $10-11. Using lunar-available
visible-spectrum solar energy (25 W/m2 on cloudless Moon
[17]) absorbed on the device’s surface (184 micron2)
and local materials, the device replicates copies of itself (at
64 MJ/kg, 6.7 x 10-5 J needed per copy; 4.6 nW available to the
device via solar energy; assume only 1/6 is used for replication/mechanosynthetic
activities) at a leisurely rate of 1 day per copy cycle. Acceleration
of each selenophage to the required average diasporic speed of 1.5
m/sec across the lunar surface costs (1/2 mv2 =) 0.01
nJ, demanding only 0.002 sec of available onboard power, a negligible
drain from replicative activities. After 81.2 doubling times, say
82 days, the entire lunar surface is coated with solar-electricity-producing
selenophages. Conservatively assuming a solar-electric lifetime
of only 10 years before replacement is needed, the circumlunar selenophagic
layer now generates a total output of 500 terawatts (current human
energy use ~ 10 terawatts worldwide) at a net cost to the original
human instigator of (~$10-11) / [(500 terawatts) (10
years)] ~ $ 2x10-28 / kW-hr.
15. U.S. Bureau of the Census, Historical Statistics of the United
States, Colonial Times to 1970, Washington, D.C., 1976; “Average
Prices for Electric Energy,” Series S 116, p. 827; “Money
Stock, M2 Money Supply,” Series X 415, pp. 992-3; “Currency
Stock and Currency in Circulation, Total Currency in U.S.,”
Series X 420, p. 993.
16. “Table 9.9 Retail Prices of Electricity Sold by Electric
Utilities,” Energy Information Administration, Monthly Energy
Review, November 2005; http://tonto.eia.doe.gov/merquery/mer_data.asp?table=T09.09
17. Taking average ground-level cloudless noontime optical-band
solar intensity as 200 W/m2 [41] and correcting for day/night
cycles (0.5), cloud cover (0.5), solar angle (0.5), and 50% conversion
efficiency (0.5), the panels deliver ~50 kW/acre (12.5 W/m2)
or ~4.4 million kW-hr/acre over an assumed 10-year lifetime for
the panels. Taking the average cost of vacant rural land as $500/acre
[42], the carrying cost at 5% mortgage interest and a 1% property
tax rate is $30/yr, or $300/acre for 10 years. Thin diamondoid solar
panels with their support structures to cover 1 acre will have a
total mass of ~2900 kg [43], costing $2900 if fabricated for $1/kg
(see below), bringing total cost over 10 years to $3200/acre and
thus the cost of energy to ~$0.001/kW-hr. Cost is similar if the
ground is first cleared by a bulldozer at $50-$150/hr [44] plus
move-in/out fees, allowing laying solar collectors on the ground
without supports. This analysis also ignores any additional costs
of local energy storage or transmission to a distant site.
18. Even if the base cost of molecular manufacturing using PNs
was lowered to $0.30/kg, food prices would always have less far
to fall than durable goods prices because nondurables made using
current production methods such as agriculture start out from a
cheaper price base. Agriculture may be regarded as a form of biology-based
nanotechnology (the plant or animal cells do most of the work of
creating the molecular structure of the food, not the farmer) whose
prices already have much of the manufacturing cost advantages of
nanotechnology built into them.
19. George J. Binczewski, “The Point of a Monument: A History
of the Aluminum Cap of the Washington Monument,” JOM
47(1995):20-25; http://www.tms.org/pubs/journals/JOM/9511/Binczewski-9511.html
20. U.S. Bureau of the Census, Historical Statistics of the
United States, Colonial Times to 1970, Washington, D.C., 1976,
Series E 135 (CPI All Items, p. 211) and Series Z 557 (Wholesale
Price, Colonial and Pre-Federal Years, p. 1196); current CPI and
employment data from Statistical Abstract of the United States,
various years.
21. “Personal Income and Outlays: October 2005,” U.S.
Bureau of Economic Analysis, Dept. of Commerce, 1 December 2005;
http://www.bea.gov/bea/newsrel/pinewsrelease.htm.
See also: “Gross Domestic Product and Corporate Profits: Gross
Domestic Product: Third Quarter 2005 (Preliminary),” U.S. Bureau
of Economic Analysis, Dept. of Commerce, 30 November 2005; http://www.bea.gov/bea/newsrelarchive/2005/gdp305p.htm.
PCE represents ~70% of the U.S. Gross Domestic Product (GDP), with
the balance of GDP consisting of private domestic investment (~15%)
plus government consumption expenditures, government investment,
and the import/export trade deficit.
22. The impact of PNs on the food sector is illustrative. PNs may
bring down the prices of basic food and food products 5-fold, with
basic nutrition becoming public domain. But food with a strong service
component may not suffer and might even thrive. For example, visiting
a restaurant is mainly experiential service consumption, not nondurable
goods consumption. Served food derives much of its value from ambiance,
service, and variety. The restaurant chef may be fabricating all
the food using a PN in the back of the kitchen, but the PN will
be running specialized software that allows the chef to synthesize
each course using unique molecular recipes he’s written himself
incorporating the latest flavors, smells and textures, and to create
a dazzling variety of special effects.
23. Milton Friedman, “The Optimum Quantity of Money,”
in Milton Friedman, ed., The Optimum Quantity of Money, Aldine
Publishing Co., Chicago, 1969.
24. Robert A. Freitas Jr., “Tangible Nanomoney,” Nanotechnology
Industries Newsletter, Issue II, July 2000, pp. 2-11; http://www.rfreitas.com/Nano/TangibleNanomoney.htm
25. “Table 15. Homeownership Rates for the United States,
by Age of Householder and by Family Status: 1982 to 2004,”
U.S. Census Bureau 2004; http://www.census.gov/hhes/www/housing/hvs/annual04/ann04t15.html
26. “Table B.100 Balance Sheet of Households and Nonprofit
Organizations,” FRB Release Z.1, 8 December 2005; http://www.federalreserve.gov/releases/z1/current/z1r-5.pdf
27. For a PN introduction period lasting τ years and assuming
neutral fiscal and monetary policies and all else equal, a crude
approximation of the inflation rate is Iτ ~ 100%
[exp( ln{PCE (τ) / PCE(0)} / τ) -1], with PCE(t) = ND(t)
+ D(t) + S(t) and taking t = 0 as Year 2005 data: NonDurable goods
ND(0) = $2566.8B, ND(τ) = ND(0)/5, Durable goods D(0) = $1129.9B,
D(τ) τ = D(0)/100, and Services S(0) = $5159.8B, S(τ)
= S(0) (1.04)τ.
28. Keynes [45] had argued in 1936 that public spending was a perfect
substitute for private spending in terms of its effect on the overall
economy. During the 1940s, the positive impact of WWII on the American
economy was interpreted by many as a vindication of Keynesian economic
principles, suggesting that it was possible to manage the overall
economy through a judicious use of public spending.
29. “Historical Debt Outstanding – Annual 1950-2005,”
U.S. Department of Treasury, 2005; http://www.publicdebt.treas.gov/opd/opdhisto4.htm
30. “Interest Expense on the Debt Outstanding,” U.S.
Department of Treasury, 2005; http://www.publicdebt.treas.gov/opd/opdint.htm
31. “Government Current Receipts and Expenditures,” Bureau
of Economic Analysis, 2005; http://www.bea.gov/bea/dn/nipaweb/GovView.asp
32. Ray Kurzweil (personal communication, 21 December 2005) suggests
that once a technology becomes an information technology, it becomes
subject to the law of accelerating returns (LOAR) [46], which is
a strongly deflationary force. Some technologies, he says, are already
subject to LOAR such as computers and communications, while others
are becoming subject to it such as drugs and medical technologies
that are in the process of switching from the hit-and-miss paradigm
of drug discovery to the information-driven paradigm of rational
drug design. Information technology is now only about 10 percent
of the economy but will grow substantially over the next 20 years,
ultimately dominating the economy. However, Kurzweil adds: “There
is a natural counterbalance to the deflationary effect of the LOAR,
which is the tendency to increase consumption with greater price-performance.
The price-performance of electronics (in terms of the cost per transistor
cycle) has been coming down by half about every year. But our consumption
of electronics has more than kept pace: the industry has grown by
18% per year in constant dollars over the past fifty years despite
the fact that you can buy about twice as much capability each year
for the same number of dollars. People didn’t buy ten thousand
dollar iPods ten years ago. We’ve seen that at each stage of
automation. In the eighteenth century, few people could afford any
well-made clothing. With the automation of the textile industry
starting around 1800, the common man and woman could afford well-made
clothes for the first time, and people were no longer satisfied
with owning a single shirt. This led to the invention of fashion,
in which people did not want to keep wearing the same clothes for
very long. So this increase in consumption with increasing price-performance
acts as a natural counterbalance so that the deflationary effect
of the LOAR does not actually shrink the size of the economy as
measured in constant currency.”
33. Donald MacKenzie, “The Big, Bad Wolf and the Rational
Market: Portfolio Insurance, the 1987 Crash and the Performativity
of Economics,” Economy and Society 33 (February 2004): 303-334;
http://www.sociology.ed.ac.uk/Research/Staff/Mackpaper5.pdf.
See also: “Black Monday – the Stock Market Crash of 1987,”
http://www.stock-market-crash.net/1987.htm;
“Lope Markets – 1987 Stock Market Crash,” http://www.lope.ca/markets/1987.html.
34. “Effective State and Local Tax Burdens by State and Rank
for Calendar Year 2005,” Retirement Living Information Center,
http://www.retirementliving.com/RLtaxburdens.html;
“Taxes by State,” http://www.retirementliving.com/RLstate1.html
35. Clyde Wayne Crews Jr., Ten Thousand Commandments: An Annual
Snapshot of the Federal Regulatory State, 2002 Edition, Cato
Institute, Washington DC, 2002, http://www.cato.org/tech/pubs/10kc_2002.pdf;
“Cost of Health Services Regulation: Working Paper Series,”
Center for Health Policy, Law and Management, Duke University, 26
December 2005, http://www.hpolicy.duke.edu/cyberexchange/Regulate/CHSR/CHSR.html;
Richard K. Vedder, “Federal Regulation’s Impact on the
Productivity Slowdown: A Trillion-Dollar Drag,” Center for
the Study of American Business (CSAB), Washington University, St.
Louis, Policy Study Number 131, March 1996, http://heartland.org/Article.cfm?artId=4677;
Thomas D. Hopkins, “Regulatory Costs in Profile,” Center
for the Study of American Business (CSAB), Washington University,
St. Louis, Policy Study Number 132, August 1996, http://wc.wustl.edu/csab/CSAB%20pubs-pdf%20files/Policy%20Studies/ps132%20hopkins.pdf;
Peter Ferrara, “The Government Tab,” Cato Institute, Washington
DC, 26 June 1999, http://www.cato.org/dailys/06-26-99.html.
36. “California Vehicle Registration Fee Calculator,”
California Department of Motor Vehicles, https://vrir.dmv.ca.gov/FeeCalculatorWeb/index.jsp
37. Boris Vishnevsky, “Tax Reform: A Breakthrough?” The
Moscow News, 10 December 2004; http://english.mn.ru/english/issue.php?2004-12-10
38. “NNI Business Glossary, Tax: Property Tax,” NikkeiNet
Interactive; http://www.nni.nikkei.co.jp/FR/TNKS/TNKSHM/glossary/tax_03.html
39. “General News: September could be bad for investments
- papers 2nd Sept,” IFAOnline, 2 September 2002; http://www.trustnet.co.uk/general/news/printable.asp?db=general&id=35552
40. “2005-2006 County Tax Rates,” Maryland State Department
of Assessments and Taxation, 25 October 2005; http://www.dat.state.md.us/sdatweb/taxrate.html
41. “Section 6.3.6 Photonic Energy Conversion Processes,”
in Robert A. Freitas Jr., Nanomedicine, Volume I: Basic Capabilities,
Landes Bioscience, Georgetown, TX, 1999, p. 156; http://www.nanomedicine.com/NMI/6.3.6.htm#p5
42. U.S. rural farmland in 2004 ranged from $265/acre in NM to
$10,200/acre in CT and RI; U.S. average is $1360/acre [47]. Canadian
rural farmland in 2004 ranged from $350/acre in Saskatchewan to
$3337/acre in Ontario [48], with one writer reporting prices as
low as $188/acre in southeastern Saskatchewan [49]; Canadian average
is $927/acre [48]. In 2005, relatively inaccessible virgin Brazilian
timberlands were available for $18/acre in 10,000-acre parcels [50],
remote but water-accessible Brazilian grasslands could be bought
for $200/acre in 200-acre parcels [50], and cattle grazing property
was for sale in remote Arizona for $125/acre in 20,000-acre parcels
[50].
43. A solar-collector sheet of diamondoid material of dimension
w = 67.1 m (edge length of 1 square acre), working strength σw
= 1010 N/m2 (~0.2 times the failure strength of diamond;
http://www.nanomedicine.com/NMI/10.3.1.htm#p3)
and density ρdiam = 3510 kg/m3 frontally
impacted by a wind of density ρair = 1.29 kg/m3
at STP traveling at maximum gust velocity v = 44.7 m/sec (100 mph)
must have thickness t > ρairwv2/σw
~ 17 microns to avoid tearing, giving a sheet of mass Msheet
= ρdiamw2t = 270 kg. Cylindrical diamond
supporting columns (one at each corner) of length L = 10 m with
Young’s modulus Ediam = 1.05 x 1012 N/m2
must have radius R > (4L2 ρairw2v2
/ π3Ediam)1/4 = 7.7 cm to avoid buckling
(http://www.nanomedicine.com/NMI/9.3.1.2.htm#p2)
under a maximum whole-sheet 1200 metric ton wind load, giving total
column mass Mcolumns = 4 (πR2Lρdiam)
= 2630 kg, hence total installed diamond mass Mdiam =
Msheet + Mcolumns = 2900 kg. (Mdiam
is minimized using the largest possible sheet rather than multiple
smaller sheets because Mcolumns ~ w-1.) The
above estimate ignores possible damage from seasonal snow loads,
an additional 400 metric tons per foot of snow depth assuming a
snow pack density of ~300 kg/m3 [51].
44. “Machinery Rental, Wisconsin, 2004,” Wisconsin’s
2004 Custom Rate Guide, Wisconsin Agricultural Statistics Service,
p. 9; http://www.nass.usda.gov/wi/custom_rate_2004.pdf;
“Summary Of EERA Rates, Bulldozers,” California Interagency
Emergency Hire Of Equipment Rental Rates, New Agreement Period:
May 1, 2005 – April 30, 2008 , No. 29 April 2005, p. 3934-6;
http://www.fire.ca.gov/php/fire_er_content/downloads/3934rev0405.pdf.
See also:
http://www.cityofmiddletown.com/Purchasing/Bids/2005-038.pdf
45. John Maynard Keynes, The General Theory of Employment, Interest
and Money, Harcourt Brace, New York, 1936.
46. “Chapter 2. A Theory of Technology Evolution: The Law
of Accelerating Returns,” in Ray Kurzweil, The Singularity
is Near: When Humans Transcend Biology, Penguin Books, New York,
2005, pp. 35-110.
47. “Table 9-14. Farm real estate: Average value per acre
of land and buildings, by State, Mar. 1, 1970 and Jan. 1, 2000-2004,”
Agriculture Statistics 2005, U.S. Department of Agriculture,
p. IX9; http://www.usda.gov/nass/pubs/agr05/05_ch9.pdf
48. “Table 2. Value per acre of farm land and buildings,”
Value of Farm Capital, Agriculture Economic Statistics, November
2005, Statistics Canada; http://www.statcan.ca/english/freepub/21-013-XIE/2005002/t002_en.htm
49. “Farmland Prices on the Prairies,” Warren’s
Farm & Ranch Review, Spring 2001, pp. 14-15; http://www.warrensdirectory.com/pdf/2001/getting_intense.pdf
50. “Land for sale, Farms for sale, Rural property for sale,”
Rural Property Bulletin, 17 December 2005;
http://www.landandfarm.com
51. “Snow Cover. Properties of Snow Cover: Snow Density,”
The Atlas of Canada, 18 April 2005; http://atlas.gc.ca/site/english/maps/environment/climate/snowcover/1
52. Robert A. Freitas Jr., “The Legal Rights of Robots,”
Student Lawyer 13(January 1985):54-56; http://www.rfreitas.com/Astro/LegalRightsOfRobots.htm
© 2006 Robert A. Freitas, Jr.
| | |
