The seventieth generation
Michael Tobis
http://www.dailykos.com/story/2007/7/28/214525/965
I especially think the comment "even if it is the seventieth
generation" is worth considering. I am inclined to say "especially if
it is the seventieth generation". The longer your horizon, the more
absurd our behavior looks, and the more obviously morality gets into a
terrtible tangle with the idea of conventional economic thought as a
sensible guide.
Consider David Archer's argument that our current behaviort is likely
to commit ourselves to a clathrate release a couple of millenia hence.
Admittedly this is somewhat speculative, but consider for the purposes
of argument that it were a certainty. Economic arguments would
discount the hunderds of generations in the future replay of the
Paleocene/Eocene cataclysm to maybe fortyseven dollars and eighteen
cents. Do we have a moral right to weight the distant future against a
more or less arbitrary contemporary measure of value?
mt
Don Libby
From: "Michael Tobis" <mto...@gmail.com>
Newsgroups: gmane.science.general.global-change
To: <global...@googlegroups.com>
Sent: Sunday, July 29, 2007 1:12 PM
Subject: [Global Change: 1981] The seventieth generation
> Economic arguments would
> discount the hunderds of generations in the future replay of the
> Paleocene/Eocene cataclysm to maybe fortyseven dollars and eighteen
> cents. Do we have a moral right to weight the distant future against a
> more or less arbitrary contemporary measure of value?
>
> mt
>
You seem to be suggesting there is some measure of value, or "moral right"
that is not more or less arbitrary. In reality we have a difference of
opinion about the discount rate: should we discount it to $47.18 or to forty
seven kazillion dollars and 18 cents? This leaves us with a political
struggle between those who prefer a high discount rate and those who prefer
a low one. Should we tax coal at $200 per ton or $200 kazillion per ton as
Hansen suggests (i.e. "ban coal")?
-dl
Michael Tobis
commensurable with dollars is dubious, but that's another topic. Let
me stipulate that we ahve some measure of value, and discuss whether
any discount rate at all is morally defensible.
While we must make some discounting of uncertainty, I do not think a
given probability of making the seventieth generation live in a world
subjected to some specific amount of damage should be weighed any
differently in our calculus than the same probability applied to the
current generation.
In the case of the clathrate release, we are discussing a high
likelihood of affecting hundreds of generations. I admit that the
likelihood is a real issue, but let's presume some mutually agreed
upon risk weighting, and presuming all else equal, why should we care
whether the hundreds of generations affected start in 2080 or 20800?
You may argue that no discount is mathematically pathological. I think
that's a separate question. (It seems moot, because risk weighting
will avoid the pathology, but that's not the point.)
My question is what the moral basis is for valuing *equally likely*
disaster in 20800 so much less than a comparable disaster in 2080.
mt
Robert A. Rohde
that they will be better equipped to deal with any concievable
"damage" than we are today. So yes, I think a harm applied in 20800
should be weighted less than one applied in 2080.
A related, but indepedent, concern is the cost to us now of avoiding a
harm in 20800 versus the cost to a future generation of doing so.
Once we start talking about these fantastically long timescales there
is essentially no way to predict future cost. Let's suppose it would
cost us 1% of GDP now to prevent a disaster in 20800. But perhaps by
2625, even if we do nothing now, the people with their flying cars,
fusion reactors, and interstellar gateways can prevent the same
potential disaster for 1/10000 the cost. If there were no
intermediate harm by waiting 600 years, then obviously the right thing
to do would be to wait.
Even if science can predict the long-term consequences of the physical
changes we make to the environment, it is implausible that we can
predict the future of human society to such a degree that we could
make any kind of cost-benefit analysis that would be sensible for the
distant future. Neither the relative significance of the harm nor the
cost of counter-acting it are things that are foreseeable on the
timescale you are talking about.
So yes, I do think discounting the very distant future makes sense.
There are plenty of reasons to be concerned with the next hundred
years without needing to get lost in open-ended speculation the next
hundred thousand.
-Robert A. Rohde
http://www.globalwarmignart.com/
> > -dl- Hide quoted text -
>
> - Show quoted text -
Don Libby
From: "Robert A. Rohde" <rar...@gmail.com>
Newsgroups: gmane.science.general.global-change
To: "globalchange" <global...@googlegroups.com>
Sent: Friday, August 03, 2007 10:41 PM
Subject: [Global Change: 2005] Re: The seventieth generation
> Once we start talking about these fantastically long timescales there
> is essentially no way to predict future cost.
<...>
>
>
> On Aug 3, 3:06 pm, "Michael Tobis" <mto...@gmail.com> wrote:
>> I agree that we need a metric. I think the idea that it is even
>> commensurable with dollars is dubious, but that's another topic. Let
>> me stipulate that we ahve some measure of value, and discuss whether
>> any discount rate at all is morally defensible.
>>
>> My question is what the moral basis is for valuing *equally likely*
>> disaster in 20800 so much less than a comparable disaster in 2080.
>>
>> mt
>>
>> On 7/30/07, Don Libby <dli...@tds.net> wrote:
>> > Sent: Sunday, July 29, 2007 1:12 PM
>> > Subject: [Global Change: 1981] The seventieth generation
>>
>> > > Do we have a moral right to weight the distant future against a
>> > > more or less arbitrary contemporary measure of value?
>>
>> > > mt
>>
>> > This leaves us with a political
>> > struggle between those who prefer a high discount rate and those who
>> > prefer
>> > a low one.
Well, you're looking in the right place - moral values underpin time
preferences. You will no doubt find a plurality of opinions, each rooted in
moral values. Which one is "right"? Scratching head, shrugging shoulders,
there's not much case-law precedent to draw on for Geocide. Reaching for
the bookshelf:
"Time and money: Discounting's problematic allure"
http://www.rff.org/Documents/RFF-Resources-136-pres.pdf
_Discounting and Intergenerational Equity_
http://www.rff.org/rff/rff_press/bookdetail.cfm?outputID=3115
The folks at Resources for the Future have given this some serious thought.
-dl
Michael Tobis
On 8/3/07, Robert A. Rohde <rar...@gmail.com> wrote:
>
> If there are people to worry about in 20800, then it is very likely
> that they will be better equipped to deal with any concievable
> "damage" than we are today. So yes, I think a harm applied in 20800
> should be weighted less than one applied in 2080.
I do not think that is obvious, and that is yet another topic worthy of note.
It appears to me that our capacity to alter our environment in a
premeditated and consistent way peaked about two generations ago, the
problems not being technical in a coinventional sense. Rather, the
technologies for manipulating opinion have advanced dramatically under
tyhe influence of broadcasting and information gathering technologies.
Accordingly, the relative strength of private interests over public
interests has increased dramatically, and sports stadiums get built
while bridges fall down.
The rest of your points are well taken, but I carefully factored them,
as well as the first point, out of my question as originally posed.
The discount rate infers that the value of an absolutely certain given
outcome be it positive or negative in the distant future is far less
than the value of the identical event in the near future. I understand
more or less how this emerges from economic logic, and I question it
on moral grounds. My question pertains exactly to equally certain
events at different times in the future.
This is not an altogether frivolous thought experiment. It appears
likely that we are in fact buying enormous and unavoidable climate
changes thousands of years into the future, when the contemporary heat
pulse propagates down to the mid-ocean and cause a carbon release
comparable to the direct anthropogenic one.
http://www.agu.org/pubs/crossref/2005/2004GC000854.shtml
mt
Don Libby
From: "Michael Tobis" <mto...@gmail.com>
Newsgroups: gmane.science.general.global-change
To: <global...@googlegroups.com>
Subject: [Global Change: 2012] Re: The seventieth generation
>
> Thanks for your thoughtful reply, Robert.
>
> On 8/3/07, Robert A. Rohde <rar...@gmail.com> wrote:
>>
>> If there are people to worry about in 20800, then it is very likely
>> that they will be better equipped to deal with any concievable
>> "damage" than we are today. So yes, I think a harm applied in 20800
>> should be weighted less than one applied in 2080.
>
>
> This is not an altogether frivolous thought experiment. It appears
> likely that we are in fact buying enormous and unavoidable climate
> changes thousands of years into the future, when the contemporary heat
> pulse propagates down to the mid-ocean and cause a carbon release
> comparable to the direct anthropogenic one.
>
> http://www.agu.org/pubs/crossref/2005/2004GC000854.shtml
>
> mt
I am reminded of the case of Permian Shield Volcano vs. All Life on Earth.
According to the prosecution's theory, the enormous outgassing kicked off a
warming that turned the oceans into a stagnant cyanobacterial pudding, with
the resultant silent but deadly hydrogen sulfide flux killing off 90% of
species on land and sea.
Species, being what they are, bounced back to fill the empty niches, only to
suffer yet another mass extinction, and another bounce-back yet again to
where we are today, in the age of mammals, contemplating the distinctly
mammalian morality of the next mass extinction. Viewed from this
perspective, the rhythmic pulse of Earth's biotic tides admit little room
for moral intervention; it's just earth pouring earth into earth.
Morality plays out in how humans treat each other, and there are plenty of
humans here and now to exercise our better judgment in shaping our behavior
towards one another. I agree with Rhode that our near-term predicament
offers enough opportunity for rational action and moral reasoning without
stepping into the issue of whether future civilizations should or should not
deploy orbiting parasols or submarine carbon bunkers.
Those who adopt the moral ethos of sustainability will attempt to exert
influence so as to not deprive future civilizations of the option to
geoengineer the global energy budget, and they will make this attempt in
competition or cooperation with groups aligned on other moral values and
principles; freedom, equality, happiness, survival, and so on. That is the
political struggle that ultimately determines what value or discount is
placed on long term consequences versus near term beneficial achievements.
-dl
Tom Adams
A few notes on long-term predictions:
1. A few years back, Verner Vinge noted that he and other sci-fi
writers had noticed that everything becomes plausible around 2030 due
the possibility of an explosion of self-designing AI technnology. This
has started a movement centered around something called the
"technological singularity".
2. If you ask a technologist in IT or biology to make a prediction
about a concrete advance, they will evidence a ~5-year prediciton
horizon. That's my observation.
Of course, we need to plan beyond 5 years, city planning/
infrastructure tends to have a multi-decade threshold.
Anyway there is bit of a problem here...
Michael Tobis
Our ability to affect the environment extends ever further in time but
our ability to predict those effects gets shorter and shorter.
Our own behavior is increasingly a dominant part of the system, and we
ourselves as a species are notoriously beyond the scope of our
predictive abilities. It seems that conventional wisdom leaves it to
our successors to have increasing ability to solve the increasingly
complex burdens we place on them.
This seems to me more like wishful thinking than like morally
well-founded behavior.
While technical progress is arguably monotonic, social progress seems
to have periods of substantial retreat. Coping effectively with
ever-increasing global environmental issues requires a maturity of
global decision making that seems beyond reach, and though there are
some bright spots on the whole I think the trends aren't good.
I believe that the more power we have over the environment, the
greater the amount of attention we should pay to far-distant
generations. As our physical power becomes ever greater, the
seventieth generation increasingly deserves to have their interests
accounted for.
Likely such a moral precept won't be taken seriously before a
generation arises that is severely impacted by a distant predecessor.
That distant predecessor will most likely be us. I don't take great
pride in that.
mt
gerh...@aston.ac.uk
> generation" is worth considering. I am inclined to say "especially if
> it is the seventieth generation".
It is an interesting philosophical question. Suppose our current
generation had a unique power to save a generation a billion years
into the future from the sun going supernova? Placed into that
hyopthetical quandary by a God like being, we might accept a lot of
sacrifice today. We might be willing to die ourselves to ensure they
would live, or alternatively, we might feel completely detached from
them and say it's their problem, not ours, or if we felt the future
generation deserved punished for their wickedness, we might actually
wish to contribute to their demise.
Interesting thought experiment though this is, in practise it
fundamentally runs across the problem of uncertainty (eloquently
pointed out by others here already), and across the related issue
whether we are in a unique position to prevent damage 70 generations
hence, or whether the damage can be reversed (CO2 removal from the air
between 2100 and 2200) or made innocuous (over 70 generations maybe
cities can be moved without too much dislocation and so sea level rise
mightn't matter too much to a very rich 70th generation).
On a related note, if it doesn't matter whether we prevent damage in
the year 2100 or the year 21000, prevention of another ice age becomes
a serious benefit of global warming. Usually that benefit is dismissed
after all by arguing that it is reasonably likely that the next ice
age would be thousands of years away absent human intervention.
Also our resources are limited, how do we add up the benefits/costs to
millions of generations into the far future and compare with what we
should spend today, unless we put in some cut-off via discounting?
Michael Tobis
In my opinion, it is in fact the uncertainty that partially saves us
from a mathematically pathological calculation.
Only partially, though. Back when sci.environment was functional
Steinn Siegurdsson discussed this matter with regard to large asteroid
impacts. It is reasonable to consider the cost of an impact sufficient
to obliterate all animal life on earth as infinite. The risk of being
imacted by an asteroid, though tiny, is not strictly speaking
infinitesimal.
Similarly, once the jackpot in a lottery exceeds the point where the
expected payoff is larger than the expected cost, is it rational for
me to sell or mortgage everything I can to buy lottery tickets?
There is indeed something missing here. I don't claim to know how to
handle this class of problem in detail. I'd welcome any substantive
quantitative ideas.
Getting back to grinding my axe, though...
My complaint is that I can see that the present method is increasingly
ill-suited to the actual policy decisions we must make, as our powers
of modification become larger.
The present method is based on a medium of exchange which operates
unidirectionally in time: we can affect the economics of the future,
but the future cannot affect our economics. There is a sort of free
trade assumption in which the trade partners are absolutely incapable
of stating their preferences.
The rational consequence (to the individual self-motivated investor)
works out to be a specific discount rate. This discount rate is based
primarily on what will happen to the medium of exchange, presuming
nothing terribly unfortunate happens. This is where the handwaving
becomes most frantic. It is, I believe, assumed without proof that
this discount rate is the same as that which needs to be applied in
policy decisions; i.e., that the rational collective action is the
same as the rational individual action.
I am not even convinced that the rational metric of collective action
is commensurable with currency at all. Even if it were, the question
of how to discount future events is fraught.
Not everything we do is about money, though it's hard to tell
sometimes. For instance, we no longer care about whether slavery is
economical. We have reached a moral consensus that the economic value
of slavery is moot. We just won't do that.
How much should we expend to avoid a non-economic damage; say
premature mortality of a million people of random age. I agree we must
measure this and I agree we must discount it; in this I am as
hard-nosed and unromantic as any right-winger. I am arguing that the
discounting must be entirely based on the uncertainty. A 50% chance of
killing a million people in 2008 or a 50% chance of killing a million
people in 20080 strike me as equally moral choices. Similarly for
losing a million species, or submerging a million acres of
historically important sites. These are things we must avoid, to the
extent we can, forever. There is no moral justification for ignoring
the distant future.
Again, I agree that uncertainty must be accountedf for in the
calculation, and I acknowledge that its effect will have some
similarities to standard discounting. I simply point out that they are
not at all the same. Our present-day commitment to sea level rise and
possibly to clathrate releases is not in any sense morally justified
by the likelihood that these problems will peak long after our great
grandchildren have passed away. In fact, we need to pay special
attention to these matters as a moral concern precisely because the
marketplace can be relied upon to discount them utterly.
Finally, as for the prevention of another ice age, I agree with Heiko
that this also becomes our responsibility henceforth. Nature has been
lucky so far, but now we need to be smart. There is no guarantee that
our planet will remain habitable indefinitely. (A mystical viewpoint
would be that Gaia noticed her long-term depletion of atmospheric CO2
and created us to fix it! Perhaps this view will win over the more
committed Luddites eventually...)
Which brings up the question of why we are so hasty to use up our
ammunition when that ice age obviously isn't imminent, even if it
originally would have been. (Opinions differ on whether another ice
age is naturally due about now; it's generally accepted that if so, it
has been cancelled, and the next candidate onset is some tens of
thousands of years into the future.)
mt
William M Connolley
On Sat, 11 Aug 2007, Michael Tobis wrote:
> Only partially, though. Back when sci.environment was functional
> Steinn Siegurdsson discussed this matter with regard to large asteroid
> impacts. It is reasonable to consider the cost of an impact sufficient
> to obliterate all animal life on earth as infinite. The risk of being
> imacted by an asteroid, though tiny, is not strictly speaking
> infinitesimal.
Not wishing to start off old discussions again, but I suspect that from an econ
viewpoint it can't be considered infinite cost to lose all life on earth. I
presume that econs would consider the sum total to be of finite value (even if
you assume infinite lifetime of the earth, due to discounting)
-W.
--
This message (and any attachments) is for the recipient only. NERC is subject
to the Freedom of Information Act 2000 and the contents of this email and any
reply you make may be disclosed by NERC unless it is exempt from release under
the Act. Any material supplied to NERC may be stored in an electronic
records management system.
James Annan
> Let me being by acknowledging Heiko's points.
>
> In my opinion, it is in fact the uncertainty that partially saves us
> from a mathematically pathological calculation.
>
> Only partially, though. Back when sci.environment was functional
> Steinn Siegurdsson discussed this matter with regard to large asteroid
> impacts. It is reasonable to consider the cost of an impact sufficient
> to obliterate all animal life on earth as infinite. The risk of being
> imacted by an asteroid, though tiny, is not strictly speaking
> infinitesimal.
>
> Similarly, once the jackpot in a lottery exceeds the point where the
> expected payoff is larger than the expected cost, is it rational for
> me to sell or mortgage everything I can to buy lottery tickets?
>
> There is indeed something missing here. I don't claim to know how to
> handle this class of problem in detail. I'd welcome any substantive
> quantitative ideas.
Well the easy answer here is that utility is not generally linear in
money, so although the first $1 might bring a positive expected return,
the last will not.
> Getting back to grinding my axe, though...
I'm not going to debate what is morally or ethically justifiable, but if
you personally are prepared to live your life according to a zero
discount rate, can I borrow your house and savings for a little while?
I'll give you them back in 100 years (index-linked of course), I'll even
add the overly generous return of 20% for your trouble.
Thought not :-)
James
Michael Tobis
> money, so although the first $1 might bring a positive expected return,
> the last will not.
A good answer that raises a lot of questions. Is this accounted for in
economists' thinking?
A lot of possibly related issues about the regressive nature of carbon
taxes seem to be trying to pop into my mind...
> can I borrow your house and savings for a little while?
> I'll give you them back in 100 years (index-linked of course), I'll even
> add the overly generous return of 20% for your trouble.
I think microeconomics is a straightforward if possibly complex matter
but macroeconomics is not.
I don't particularly want collective decision making to have the
identical structure as individual decision making, and I don't see any
reasoning why it should even be remotely similar.
Indeed, I think the point of collective decision making is to increase
the likelihood of beneficial results from individual decision making.
The macro presents a very different concern than the micro.
mt
James Annan
> > Well the easy answer here is that utility is not generally linear in
>> money, so although the first $1 might bring a positive expected return,
>> the last will not.
>
> A good answer that raises a lot of questions. Is this accounted for in
> economists' thinking?
Of course it is!
Stern uses a logarithmic utility, where 10% of a poor person's salary is
worth the same (in terms of overall global utility) as 10% off a rich
person's. There are arguments about how plausible this curve is,
especially in combination with the negligible discount rate. I don't
think anyone claims that the world works like this, but Stern asserts
that the world _ought_ to behave like this.
>
> A lot of possibly related issues about the regressive nature of carbon
> taxes seem to be trying to pop into my mind...
>
>> can I borrow your house and savings for a little while?
>> I'll give you them back in 100 years (index-linked of course), I'll even
>> add the overly generous return of 20% for your trouble.
>
> I think microeconomics is a straightforward if possibly complex matter
> but macroeconomics is not.
>
> I don't particularly want collective decision making to have the
> identical structure as individual decision making, and I don't see any
> reasoning why it should even be remotely similar.
>
> Indeed, I think the point of collective decision making is to increase
> the likelihood of beneficial results from individual decision making.
> The macro presents a very different concern than the micro.
Society is just the sum of all the individuals. I don't see how you can
propose society forgoing current wealth for the promise of the same
wealth in the future, and somehow believe that individuals that make up
society can avoid the same. Are you are saying that this sort of bargain
would be acceptable to you, so long as everyone else did it?
James
gerh...@aston.ac.uk
> worth the same (in terms of overall global utility) as 10% off a rich
> person's. There are arguments about how plausible this curve is,
> especially in combination with the negligible discount rate. I don't
> think anyone claims that the world works like this, but Stern asserts
> that the world _ought_ to behave like this.
http://heikoheiko.blogspot.com/2006/07/economic-benefits-of-minimum-wage-laws.html
The beauty of this economic modelling technique is that it shows the
value of redistribution. While taxing the rich may reduce GDP, it may
simultaneously raise total welfare, at least when adjusted for
logarithmic utility.
The reason this type of model gets applied to climate change and not
to the more obvious redistribution is I think the blame component and
nationalism, roughly speaking it seems that it's a lot more ok to
spend $10 to avoid killing Bangladeshis through reduced climate change
than it is to just hand them $10, even if that'll save ten times as
many people.
Economic models might say that taking $10000 from each American
earning more than $30000 per year and giving it to Indians making less
than a Dollar a day would be hugely beneficial for total global
welfare, but it's an awfully tough sell, well at least as far as
making the case to American voters is concerned ...