Bryan Caplan  

TANSTAAFL vs. Futurism

PRINT
The Long-term Debt Issue... The GMU Difference...
While writing my target essay for May's Cato Unbound, I decided I ought to take another look at Edwin Dolan's TANSTAAFL: The Economic Strategy for Environmental Crisis.*  First published in 1971, it is one of the earliest works of free-market environmentalism

The book's full of good material, and was very original at the time.  But by modern standards, it's amazingly Malthusian.  For Dolan, the last two centuries of rapid growth of population and prosperity appear to be a happy coincidence.  He's skeptical that rapid growth can continue.  Unless I'm missing something, never mentions the Simon/Kremer/Romer point that larger populations create more new ideas.

Most striking, though, is how Dolan reacts to the following scenario:
It has been estimated that the earth alone could accommodate twenty million times its present population, living at 120 per square meter in a 2000-story building covering the entire earth.  It would take us 890 years, at our present rate of growth, to get to that point, and by then we may have solved the formidable technical and economic problems of interstellar travel, and be able to export our surplus to the stars...
 Dolan responds to this scenario with derision, even contempt:
If people want to believe in this sort of thing there is no rational argument which can be presented to convince them otherwise.  If one were to "prove" that it is impossible to travel faster than the speed of light and reach the stars, they could legitimately counter by citing "proofs" by eminent 19th-century scientists that heavier-than-air machines could never fly.  For the present, however, [this] lies in the realm of pure speculation, not of reasoned discussion, so we will give it no further consideration...
Perhaps I'm out of touch, but my guess is that modern intellectuals across the political spectrum would take the prospect of space colonization by 2860 A.D. more seriously than Dolan did.  Environmentalist values are almost certainly much more prevalent today than in 1971.  But techno-optimism - especially about the far future - seems to have gained even more ground.  Am I wrong?

* TANSTAAFL stands for "There ain't no such thing as a free lunch."


Comments and Sharing





COMMENTS (25 to date)
Pandaemoni writes:

While space colonization is a possibility, I don't think it will ever solve many population problems. Let's assume we had the technology to do it right now and A) that the world's population is 6 billion people, and (B) that it grows at a rate of 1% per year.

Under that assumption we'd need about 60 million people to emigrate to space colonies *every year* just to keep the population on Earth level. So, every year, we have to launch, basically, the entire population of the United Kingdom into space. Assuming the Earth remains in its current shape, I'm not sure we'll ever be "exporting" that many human beings away from the comforts and personal connections of Earth. The way it would be most likely to happen is if we turn the Earth into a hellhole rather than it's usually hospitable self.

Meng Bomin writes:

I must say that I agree wholeheartedly with Pandaemoni in that space colonization is not a viable option to save the Earth from population growth, if indeed that is what one is trying to "solve".

Launching people into space is expensive and will always be so as simple result of the amount of energy required to enter, much less go beyond, Earth orbit. Colonists would be expensive to transport and in reality, the easiest place for a human to live in the universe is Earth, as we're the product of billions of years of evolution and adaptation to the unique properties of our home planet. If one is worried about resources, sending humans into space exacerbates rather than ameliorates those problems.

That's not to say that we shouldn't attempt space colonization once we've developed the technology and gathered the resources...it's just that we should admit that it will be a difficult, costly endeavor and that it will likely not result in any significant benefit to those of us back on Earth beyond the knowledge that we've broadened the horizon of terrestrial life.

Eelco Hoogendoorn writes:

We can be a tad more precise as to why emigration of any form is not a long term answer to competition strong enough to curb growth rates.

The part of the universe colonized by humans will grow at best with time cubed; which is fundamentally slower than exponential growth in time.

Not that I think its something anyone should worry about today, but as a philosophical point, not being willing to grapple with the question of what the world looks like when there are more people than food, feels evasive.

nazgulnarsil writes:

currently people are, in general, net producers of wealth. more people = more wealth. but what happens as the ability of low skill workers to create wealth slowly fades away?

mike shupp writes:

I think your thesis is being refuted by the commenters. Too bad! But I have to note that my recollection also is that people were much more expectant of future technological progress in the 1950s and 1960's than they are today.

Let me suggest an experiment. Ignore any discussion of computer hardware, computer software, or computer-related communications for say a week. How many conversations will you overhear or articles (or even books!) will you read discussing future technolgical progress?

Will you find discussions of how many people might be expected to dwell in lunar and Martian colonies by the end of the century? Will you have some notion of how many nuclear reactors will be running in this country in 2050? Will you have some handle on how nanotechnology is affecting the home improvement business? Will your understanding of genetic links between Eskimos and North American Indians be deeper? Etc, etc, etc. My suspicion is that unless you deliberately search such items out, you're not going to find them. Instead you'll be regaled by statistics about how many copies of the 11th edition of the Encyclopedia Brittanica can be written in the average school child's laptop hard drive.

I.e., our notion of technology has become "what computers do" and our idea of progress is perilously close to "what computers will do."

Floccina writes:

I have an idea on space colonization, you push Mars into an orbit close to earth then you push commits and asteroids into it to get water on it and the size right. You are attempting to build a near by planet similar to earth. Pushing planets around would take a huge amount of power but nuclear reactions make a lot of power possible.

Chris Koresko writes:

Aside from a handful of physics-based arguments like the ones presented by Pandaemoni and Eelco Hoogendoorn, it's not possible to draw many firm conclusions about the impact of technological progress that far into the future. Still, it's fun to speculate.

Contrary to Meng Bomin's comment, the cost of the energy (propellant) for conventional rocket-powered space launches is not the dominant factor; it's the hardware and support personnel, and the need to amortize large engineering costs across a small number of vehicles, that make spaceflight expensive. Consider that even for passenger airliners the fuel cost is still what, 20% of the operating cost, even though there are thousands of airliners in service and they're reused many times each. Also, in terms of deep space travel, getting to Earth orbit is half the battle: it's "halfway to anywhere". So if we launch at all, it's not much harder to head for Mars et al.

So there is no fundamental reason that space launch can't be much cheaper than it is. High flight rates are probably key for getting rocket launch costs down the short term, and fundamentally new approaches such as the "launch loop" or the more ambitious "space elevator" may enable orders-of-magnitude reductions if they can be made to work.

Supposing we find a way to launch Pandaemoni's 60 million people per year, where would they go? One possibility would be artificial habitats constructed by automated, self-replicating mechanisms: "Santa Claus machines". The tech for that isn't ready yet, but it looks like it may be within a few decades. Such habitats could be made quite Earth-like if desired. There's no shortage of raw materials or energy out there, at least if the goal is to accommodate a steady stream of emigrants from Earth rather than an exponentially-growing population explosion.

An extensive industrial base in space would likely render a lot of our current resource and environmental problems moot. It's already almost practical to do space solar power (orbiting power stations beaming energy to the surface in the form of microwaves). That gets around the "cover the state of Connecticut with solar panels" problem, and enables 24/7 production (with only the occasional interruption when the station passes through Earth's shadow).

My guess is that the far future -- a few millenia from now -- will see most of humanity living among the asteroids, with a population many times what it is now. Earth will likely still be inhabited, but it'll be a backwater, maybe kept as a nature preserve or some such.

Finch writes:

> How many conversations will you overhear or
> articles (or even books!) will you read
> discussing future technolgical progress?

The last non computer-oriented future-tech article I read was depressing. It's a clear indication of why the space program has foundered, and it has nothing to do with the technology being difficult and costly and everything to do with it being run by the same folks who bring you the department of motor vehicles:

http://www.politico.com/news/stories/0411/53495.html

> [Space colonization] will likely not result in
> any significant benefit to those of us back on
> Earth beyond the knowledge that we've broadened
> the horizon of terrestrial life.

Well, it will result in there being a lot more people living in higher living standards, and the resulting additional invention, profit, art, and culture, but hey, who's counting?

The benefit is that we can have more people, and that those people will do productive things. Even if an individual stays home, he can still trade with the emigrants. Europeans benefited from American colonization whether or not they emigrated themselves.

8 writes:

If in 700 AD, the Catholic monks are discussing these issues, and one of them says the Virgin Mary showed him a vision of 2010 AD, how does that help them deal with the next 1000 years of development?

In Tainter's Collapse of Complex Societies, he showed that the marginal return on R&D was declining, along with healthcare and education. Has this changed? It seems the gains from the industrial revolution are finally winding down, with the information revolution as the final way to maximize the return on current technology.

Ed Dolan writes:

Bryan-- Thanks for your comments. As it happens, I was looking at TANSTAAFL myself lately, with the idea of publishing an annotated 40th anniversary edition. I had not reread it for decades, and had forgotten many details of what was in it.

You are right that I bought into the then prevailing pessimism on population growth. There were two parts to the pessimism. One part was a set of forecasts for future population that look way on the high side today. As it turns out, the demographic transition has proceeded more rapidly than many people at that time thought. It is now considered realistic that people alive today will see peak population, and that global population will be declining by the end of the century, rendering the space travel issue moot as far as demographics go. (I don't see travel to the stars as a mass venture, ever, but I do hope a few people figure out a way at least to go out and have a look.)

The other part of the pessimism was the bacteria-in-a-jar kind of Malthusianism, where beyond a point, more people would necessarily mean less income per capita. Simon's Ultimate Resource would not appear for 10 years yet, and you are right, I did not anticipate his arguments. When they came out, I found them persuasive, and will mention them in the notes to the new edition.

Given the pessimism about growth of population, per capita income, and space flight, the only argument I was left to fall back on was plain old libertarianism: If people want to have lots of kids, they should be allowed to do so, even if some people find it offensive. Basically, I put having a big family in the same category as gay sex or snorting coke: They are things that I don't do myself, but I don't think I have the right to stop other people from doing them. I thought, and still think, a large global population was something we could learn to live with if we had to.

Look for the appearance of TANSTAAFL: The 40th Anniversary Annotated Edition sometime this fall.

Nick Bradley writes:

To Meng Bomin,

Space travel will not be that expensive in the year 2860 -- we will be far, far wealthier.

The long-run per capita GDP growth rate is about 2% per year. At that rate, per capita GDP in 2850 in the US will be a Trillion dollars per capita, while the real cost of space travel will fall due to technological advance.

Here are the per capita GDP numbers in 2860 based on different growth rates:

2% - $960B
1.75% - $120B
1.50% - $14B
1.25 - $1.8B
1% - $220M
0.75% - 27M
0.50% - 3.25M
0.25% - 400K

So if it costs a billion dollars per person, at a growth rate of 1.75% that's only 0.83% of that person's GDP...the equivalent of $390 per person today. That's the same as airfare today.

If the real cost of space flight declined only 0.5% a year from now to then, that $390 flight would be about $5.50....bus fare.

To be honest, I think that economic expansion will necessitate space colonization much sooner than that. I could see us harvesting asteroids for minerals in a couple hundred years or so.

Nick Bradley writes:

Pandaemoni,

you're not taking into account economic growth and wealth accumulation. Space travel will be the equivalent of a plane ticket or even bus fare in hundreds of years.

Evan writes:
you're not taking into account economic growth and wealth accumulation. Space travel will be the equivalent of a plane ticket or even bus fare in hundreds of years.
We can't also forget the distinct possibility that space travel will be less expensive because humans will be easier to transport in the future. Imagine if most people in the future are cyborgs with superhuman resistance to acceleration, and the ability to run most of their body's functions off electricity. You could pack them into spaceships like sardines and launch them into space with a big gun. Virtual reality would also make the separation problems that space travel currently generates less problematic, allowing civilization to remain closer together.
Roger Sweeny writes:

So there is no fundamental reason that space launch can't be much cheaper than it is.

There is a fundamental reason that space launch will never be cheap. We live at the bottom of a pretty deep gravity well and it requires a lot of energy to get out of that well. That will never change. Space launch will always be energetically expensive.

Will technologies that we can't imagine today make it possible to obtain this much propulsion energy cheaply? Maybe. But I would be willing to bet a lot of money against it.

Finch writes:

> There is a fundamental reason that space launch
> will never be cheap. We live at the bottom of a
> pretty deep gravity well and it requires a lot of
> energy to get out of that well. That will never
> change. Space launch will always be energetically
> expensive.

Sure, but that justifies it being expensive like flying to Tokyo is expensive, not $10,000 a pound expensive. Energy cost to orbit is maybe $10 per pound of payload, using chemical fuels and depending on what you burn.

SpaceX, using basically the same non-reusable technology as the usual government contractors, appears to have reduced costs by about a factor of ten just by viewing additional employees and complexity as costs rather than benefits. There seems to be a lot of room for improvement.

fructose writes:

There is a fundamental reason that space launch will never be cheap. We live at the bottom of a pretty deep gravity well and it requires a lot of energy to get out of that well. That will never change. Space launch will always be energetically expensive.

One can imagine a 13th century philosopher saying "Men will never fly like birds. We are too heavy and dense. We would have to flap far too hard to rise off the ground." The philosopher couldn't have predicted the vast amounts of energy fossil fuels put at our disposal. (Or more towards the point, he wouldn't believe we could effectively harness that power.) This kind of myopia is even less excusable, since we know of a power source (fusion) which would give us cheap enough energy to go to the stars. We can't effectively harness that power yet (though people are trying).

In any case, the point is moot since it is likely that most "people" will be Hansonian uploads, who can simply be transmitted by laser or radio waves to the stars (after remote drones are sent first).

Douglass Holmes writes:

As little as 20 years ago I could have explained to you that both radio bandwidth, tower real estate, and land-line connections to those towers would preclude cellular service from ever being cheap enough for cell phones to replace land line phones. Funny thing about progress; it has a way of making things cheaper.
However, travelling to the stars will take a fundamental change in the nature of humans or the universe. Humans don't like to take eight years to get somewhere. And I consider people who believe in faster-than-light travel to be no more informed than young earth creationists.

Pandaemoni writes:

@ Nick Bradley

I was not making an argument based on a monetary cost, I was simply pointing out that we would need to move many a lot of people, permanently, away from the Earth every single year. While the economic costs could be low (I assume they will be), I do not believe that would be a realistic solution to a population crisis. For every net new person on the planet, you'd need to find someone willing to leave the Earth and take up permanent residence elsewhere.

I am not certain we could find populations the size of entire nations who would want to relocate to an alien environment each and every year, away from meaningful contact with family and friends left behind. That is on top of the problem that each space colony will likely grow of its own accord, and may not want an influx of large numbers of new colonists. (Any more than Americans in general want mass immigration from foreign nations, but with the added issue that resource strain may be more of an issue in an alien environment).

I do not believe that the solution for the population "problem" (though I am skeptical that there is a real problem) is not to shoot people into space.

I do think space colonization is likely, and that the size of the colonies will grow large over time. I'm not averse to saying there could well be, some day, many more people living on other worlds than there are on Earth itself. I also think that if population size were a real problem, we'd need to find a better solution than extraterrestrial colonization.

Finch writes:

Historical emigration rates have been very small, even when the costs are quite low and the apparent benefits high.

That doesn't mean there won't eventually be vastly more people living off earth than on earth, it just means that emigration won't do much for population on earth and that most of the space population will be in-space descendants.

Robb writes:

Why don't we have flying cars? Not because we can't but because of practical considerations.

Same thing for space colonization. Why would you want to colonize a hostile miserable place where nothing lives when friendly death valley awaits with cheap land, easy access, and best of all, you can breathe the air. Also consider Antarctica and the Gobi desert.

I don't know where technology will be in a thousand years, but I bet most people will still not be flocking to live in places where it is unpleasant or deadly to walk outside your house/apartment in the springtime.

Finch writes:

New York City is unpleasant and deadly without technological aid. Save oxygen, everything you need - food, water, energy - is imported, sometimes from vast distances. Three months out of the year you'd stand a good chance of dying if you spent the night outside. We don't all live in the savanna anymore and almost all mankind is critically dependent on technology for their day-to-day survival. Space differs only in small degree, not in kind.

To your other point, nobody pays $10,000 a pound to send stuff to Antarctica or the Gobi desert. It's a red herring argument. LEO is the highest price destination mankind has with a robust market. What would the world look like if it cost $100 a pound to get there? We don't know, and confident expressions that it would look just the same as it does now don't carry a lot of weight without data.

Robb writes:

Finch says,

"To your other point, nobody pays $10,000 a pound to send stuff to Antarctica or the Gobi desert. It's a red herring argument"

People spend a lot of money to get to the top of mount everest, but no one is talking about colonization.

What I am saying is why would anyone condemn their children to living in the equivilent of a submarine? We may all desire to visit these places (like leo or the asteroids) , but why would anyone want to live there?

Roger Sweeny writes:

Energy cost to orbit is maybe $10 per pound of payload

I'm not sure where you get that number from. It seems very low to me. But it is also irrelevant. Payload doesn't lift itself into orbit. It requires a rocket and fuel. All that mass also has to be lifted--and there is an awful lot of it.

A number of commenters have pointed to excessively pessimistic predictions from the past (cell phones, heavier than air travel). But wrong predictions go in both directions. In the 50s, lots of science people confidently asserted that by now electricity generated by nuclear power would be so cheap that it would cost more to keep track of people's usage and send them a bill than it would cost to generate ("too cheap to meter"). Many kids were given dreams of personal jet packs and flying cars.

Progress makes some things cheaper but makes some things more expensive.

Something may come along that will make lift capacity relatively cheap (as in "forty hours of work at the median wage will make me enough money to put a pound of something into orbit.") But I would be very, very surprised.

Finch writes:

>> Energy cost to orbit is maybe $10 per pound of
>> payload

> I'm not sure where you get that number from. It
> seems very low to me. But it is also irrelevant.
> Payload doesn't lift itself into orbit. It
> requires a rocket and fuel. All that mass also
> has to be lifted--and there is an awful lot of it.

Falcon 9, to pick an example, is about 3 percent payload by mass. Pretend the rest is propellant, to exaggerate the problem. 3/4 of that propellant is liquid oxygen, which is essentially free. It's maybe 7 cents a pound. The rest is kerosene, and runs around $1 a pound, or $8 a gallon since it's RP-1, a very clean cut of kerosene. You have about 32 pounds of propellant per pound of payload. 24 pounds of that runs 7 cents a pound, 8 pounds runs $1 a pound. Total cost is $9.68 per pound of payload to LEO.

You can pick propellants that makes this look worse. The worst case I know of is the Titan series of rockets, which used exotic and highly toxic fuels. In that case propellant cost around $250 per pound of payload to orbit. LOx/LH2 is somewhere in between. You can also pick propellants, like LOx/Methane, that make it look better.

Finch writes:

> What I am saying is why would anyone condemn
> their children to living in the equivilent of a
> submarine? We may all desire to visit these
> places (like leo or the asteroids) , but why
> would anyone want to live there?

Usually people emigrate for better long term opportunities for themselves and their children. I think "room to grow" would be a possible reason, but there are many. "Dad works there and the pay is good" is another. "Niches aren't already filled" is another.

Anyway, it's like living in a submarine _today_, albeit a submarine with the best view in the world. But it's unlikely to remain so. In a hundred years, it will probably be nicer up there than down here. Your statement is like saying colonization of North America wasn't viable because the Mayflower was uncomfortable.

Comments for this entry have been closed
Return to top