WEBVTT

00:00:09.940 --> 00:00:11.156
PROFESSOR: I

00:00:11.156 --> 00:00:17.670
LAWRENCE SUSSKIND: I think,
collectively, I did not

00:00:17.670 --> 00:00:21.870
hear the full-fledged
justification

00:00:21.870 --> 00:00:23.690
of the precautionary principle.

00:00:23.690 --> 00:00:25.980
And I think that
focusing on the fact

00:00:25.980 --> 00:00:30.480
that its application
would be the greatest

00:00:30.480 --> 00:00:32.280
good for the greatest
number of people

00:00:32.280 --> 00:00:35.850
is probably manifestly
wrong because I

00:00:35.850 --> 00:00:38.370
think there are plenty
of instances where

00:00:38.370 --> 00:00:42.180
being precautionary
means that you look out

00:00:42.180 --> 00:00:45.210
for subgroups or
sub effects which

00:00:45.210 --> 00:00:48.750
in fact, are very unlikely.

00:00:48.750 --> 00:00:51.460
In a world of
certainty, one can argue

00:00:51.460 --> 00:00:54.600
utilitarian ethics applies.

00:00:54.600 --> 00:00:57.210
If you ban a chemical, you're
going to have a certain cost

00:00:57.210 --> 00:00:59.925
and you going to have
a certain benefit.

00:00:59.925 --> 00:01:03.990
As soon as the cost
becomes uncertain,

00:01:03.990 --> 00:01:06.540
or the benefit
becomes uncertain,

00:01:06.540 --> 00:01:08.130
which is related
to risk assessment

00:01:08.130 --> 00:01:10.890
because the benefit of
controlling the chemical

00:01:10.890 --> 00:01:15.390
is derived from doing some
sort of risk assessment

00:01:15.390 --> 00:01:17.030
which has high
uncertainties in it.

00:01:17.030 --> 00:01:19.154
But it's not the only thing
that has uncertainties.

00:01:19.154 --> 00:01:21.030
Costs have uncertainties, too.

00:01:21.030 --> 00:01:24.390
So as soon as you depart
from a well-defined benefit

00:01:24.390 --> 00:01:27.210
and a well-defined cost.

00:01:27.210 --> 00:01:30.420
Utilitarian ethics is no
longer important because what's

00:01:30.420 --> 00:01:36.240
important is how you value the
mistakes that you could make.

00:01:36.240 --> 00:01:39.610
A type I error comes when you've
failed to regulate something,

00:01:39.610 --> 00:01:41.580
and it later comes
out to be worse

00:01:41.580 --> 00:01:45.474
than you thought or at the tails
of the distribution of what

00:01:45.474 --> 00:01:46.140
you're thinking.

00:01:46.140 --> 00:01:49.500
A type II error is
where you regulate,

00:01:49.500 --> 00:01:52.160
and things turn out to be
not as bad as you thought

00:01:52.160 --> 00:01:54.540
and then impose a
higher cost upon.

00:01:54.540 --> 00:01:58.480
But there's a type
III error which

00:01:58.480 --> 00:02:00.606
is you're working on
the wrong problem.

00:02:00.606 --> 00:02:02.230
And I think that's
what you all really.

00:02:02.230 --> 00:02:06.030
Missed the point is that
when you ban a chemical,

00:02:06.030 --> 00:02:07.960
that isn't all that happens.

00:02:07.960 --> 00:02:11.950
What happens is you
end up coming face

00:02:11.950 --> 00:02:14.860
to face with a substitute,
with an alternative technology,

00:02:14.860 --> 00:02:17.960
with alternative ways to
meet the public needs.

00:02:17.960 --> 00:02:21.520
And so a straight
cost-benefit analysis,

00:02:21.520 --> 00:02:27.700
even with uncertainty, unless
it includes the alternatives

00:02:27.700 --> 00:02:29.810
that you have is not
a complete analysis.

00:02:29.810 --> 00:02:31.820
And you can make
anything look good

00:02:31.820 --> 00:02:35.510
depending upon how you value
life or value the environment.

00:02:35.510 --> 00:02:39.310
And so the precautionary
principle, properly,

00:02:39.310 --> 00:02:42.270
which is which applies when
you have great uncertainties,

00:02:42.270 --> 00:02:44.200
has to ask the
question, what are

00:02:44.200 --> 00:02:49.042
the alternatives to the action
that you've decided to promote.

00:02:49.042 --> 00:02:50.000
There are alternatives.

00:02:50.000 --> 00:02:52.420
There are there substitutes

00:02:52.420 --> 00:02:55.090
I'll give you one more comment.

00:02:55.090 --> 00:02:57.460
I know the word of
Cass Sunstein well.

00:02:57.460 --> 00:03:00.340
I graduated from
[INAUDIBLE],, same law school.

00:03:00.340 --> 00:03:04.170
I think he is a poor scholar.

00:03:04.170 --> 00:03:05.230
He does not understand.

00:03:05.230 --> 00:03:06.659
SPEAKER 1: You heard it here.

00:03:06.659 --> 00:03:08.200
LAWRENCE SUSSKIND:
He does not really

00:03:08.200 --> 00:03:11.390
understand what this
whole issue was about,

00:03:11.390 --> 00:03:14.860
has never really understood it.

00:03:14.860 --> 00:03:20.320
For example, people say, suppose
you regulate vinyl chloride.

00:03:20.320 --> 00:03:23.230
You put worker controls
on vinyl chloride.

00:03:23.230 --> 00:03:26.290
And the workers now
earn a smaller salary

00:03:26.290 --> 00:03:27.750
because the vinyl
chloride producer

00:03:27.750 --> 00:03:29.229
can't afford to pay them.

00:03:29.229 --> 00:03:30.020
So what do they do?

00:03:30.020 --> 00:03:31.228
They quit their health clubs.

00:03:31.228 --> 00:03:33.890
They buy less lean hamburger.

00:03:33.890 --> 00:03:37.780
And the health impact,
is argued by Sunstein,

00:03:37.780 --> 00:03:39.760
is worse than if you
regulated the hazard

00:03:39.760 --> 00:03:41.710
because you made
the workers poor.

00:03:41.710 --> 00:03:44.720
And they eat fat hamburger
and don't exercise.

00:03:44.720 --> 00:03:46.150
But that doesn't
take into account

00:03:46.150 --> 00:03:49.760
the fact that when you regulate
vinyl chloride, what replaces

00:03:49.760 --> 00:03:52.220
it, let's say, is polyethylene.

00:03:52.220 --> 00:03:54.460
So those industries
increase production,

00:03:54.460 --> 00:03:57.510
increase the wages of their
workers, who now join health

00:03:57.510 --> 00:03:59.195
clubs and lean hamburgers.

00:03:59.195 --> 00:04:01.195
Now if you're really going
to take this calculus

00:04:01.195 --> 00:04:03.580
to the extreme, you
could see how ridiculous

00:04:03.580 --> 00:04:06.250
it is because you
can't really determine.

00:04:06.250 --> 00:04:08.900
But what you can determine
and what regulation

00:04:08.900 --> 00:04:13.060
shows, when it's stringent,
you don't make marginal changes

00:04:13.060 --> 00:04:14.220
in production.

00:04:14.220 --> 00:04:17.260
You may see serious
shifts in the nature

00:04:17.260 --> 00:04:18.820
of industrial policy.

00:04:18.820 --> 00:04:20.589
And you shift to
nonchemical alternatives.

00:04:20.589 --> 00:04:23.050
For example, the banning
of CFCs didn't just

00:04:23.050 --> 00:04:26.890
supply one inhalant which
was slightly better.

00:04:26.890 --> 00:04:29.220
People started to use pump cans.

00:04:29.220 --> 00:04:31.000
You sell less
product, then that's

00:04:31.000 --> 00:04:32.380
too bad for the producer.

00:04:32.380 --> 00:04:35.070
But you don't need to
inhalant to deliver

00:04:35.070 --> 00:04:37.190
deodorant or those things.

00:04:37.190 --> 00:04:40.460
So it is narrowly defined.

00:04:40.460 --> 00:04:45.040
You work on the wrong
problem, and you ignore how

00:04:45.040 --> 00:04:47.500
you feel about air avoidance.

00:04:47.500 --> 00:04:49.605
That's when the precautionary
principle comes in.

00:04:49.605 --> 00:04:50.480
SPEAKER 1: Beautiful.

00:04:50.480 --> 00:04:54.302
AUDIENCE: --go along with
it, slash bring up that point

00:04:54.302 --> 00:04:58.230
that I think there is a middle
ground where with some things

00:04:58.230 --> 00:05:00.864
I think we should
look into alternatives

00:05:00.864 --> 00:05:05.560
and have them as backseat
if we find that whatever

00:05:05.560 --> 00:05:07.948
precautionary didn't work out.

00:05:07.948 --> 00:05:13.198
I think it's really dangerous to
just ignore other alternatives.

00:05:13.198 --> 00:05:18.826
Like the Susskind article
said, going with that action

00:05:18.826 --> 00:05:25.840
leads to unintended things that
maybe could've been a benefit.

00:05:25.840 --> 00:05:29.620
LAWRENCE SUSSKIND: Take the
global warming situation.

00:05:29.620 --> 00:05:32.500
This middle ground, I
think, is a fiction.

00:05:32.500 --> 00:05:36.650
You either have to regard the
probability of serious global

00:05:36.650 --> 00:05:42.250
impacts as within the realm
of possibility, or you don't.

00:05:42.250 --> 00:05:45.070
Even people who aren't
certain about the science,

00:05:45.070 --> 00:05:47.290
you ask them the
question, should we

00:05:47.290 --> 00:05:49.810
plan for a rising sea level?

00:05:49.810 --> 00:05:52.240
Should we plan for
floods and droughts

00:05:52.240 --> 00:05:54.710
would be a bad thing to do.

00:05:54.710 --> 00:05:56.560
And the answer
invariably comes, no, it

00:05:56.560 --> 00:05:58.660
should be a good thing to do.

00:05:58.660 --> 00:06:01.080
And if you act in
a timely fashion,

00:06:01.080 --> 00:06:04.519
even if you turn
out to be wrong,

00:06:04.519 --> 00:06:06.310
imagine what would
happen if you turned out

00:06:06.310 --> 00:06:09.490
to be right and turn out
to have the coastal cities

00:06:09.490 --> 00:06:12.340
disappearing [INAUDIBLE].

00:06:12.340 --> 00:06:14.410
I think the
precautionary principle's

00:06:14.410 --> 00:06:16.210
been criticized
by people who have

00:06:16.210 --> 00:06:20.470
an economic dog in that
fight, that it's too expensive

00:06:20.470 --> 00:06:23.140
and their particular
industry will disappear.

00:06:23.140 --> 00:06:26.830
But the alternative is to
have, generally, other people

00:06:26.830 --> 00:06:28.340
bear the cost.

00:06:28.340 --> 00:06:31.360
So who bears the cost
and who reaps the benefit

00:06:31.360 --> 00:06:33.070
of the decided policy?

00:06:33.070 --> 00:06:34.360
It's very important.

00:06:34.360 --> 00:06:38.500
And the distributional effects
of going one way or the other

00:06:38.500 --> 00:06:39.490
are really serious.

00:06:39.490 --> 00:06:43.420
If you decide to put a carbon
tax on oil, let me tell you,

00:06:43.420 --> 00:06:46.990
the petroleum
industry will survive.

00:06:46.990 --> 00:06:49.120
If you don't put a
carbon tax on the floor,

00:06:49.120 --> 00:06:51.650
it's not clear that the
planet will survive.

00:06:51.650 --> 00:06:55.690
So if they ask whose
ox is being gored,

00:06:55.690 --> 00:07:00.490
how certain are we that you are
producing an irreducible effect

00:07:00.490 --> 00:07:01.420
of these people.

00:07:01.420 --> 00:07:03.700
And do you want to
keep your options open?

00:07:03.700 --> 00:07:05.950
So I reject the fact that
what you're trying to do

00:07:05.950 --> 00:07:07.730
is find a middle ground.

00:07:07.730 --> 00:07:09.688
LAWRENCE SUSSKIND: But
I want to turn the dial,

00:07:09.688 --> 00:07:14.530
as a last note, one more tick
beyond what Nick just said.

00:07:14.530 --> 00:07:18.460
I believe, take the
climate change example,

00:07:18.460 --> 00:07:21.940
that if you say, let's
take this seriously

00:07:21.940 --> 00:07:25.420
because it might happen
that a whole series

00:07:25.420 --> 00:07:27.370
of second and third
order decisions

00:07:27.370 --> 00:07:30.940
can get made that, in fact,
wouldn't get made otherwise.

00:07:30.940 --> 00:07:33.460
It's related to the point
you're making first.

00:07:33.460 --> 00:07:35.820
And that will create
all kinds of benefits

00:07:35.820 --> 00:07:38.170
because if people don't do
it to create these benefits,

00:07:38.170 --> 00:07:39.610
they'll do it to
create benefits.

00:07:39.610 --> 00:07:41.111
But they never
would have searched.

00:07:41.111 --> 00:07:43.610
They never would've tried, They
never would've looked there.

00:07:43.610 --> 00:07:47.350
So when we talk to communities
about taking various moves

00:07:47.350 --> 00:07:50.484
to manage potential
climate risks,

00:07:50.484 --> 00:07:52.150
and once you do it
in a way that creates

00:07:52.150 --> 00:07:56.860
all kinds of other second
and third order benefits now,

00:07:56.860 --> 00:07:59.080
instead of just looking
at what the costs are

00:07:59.080 --> 00:08:01.360
and arguing that the long
term benefits outweigh

00:08:01.360 --> 00:08:03.700
the short-term
costs, my argument

00:08:03.700 --> 00:08:06.700
is why don't you try to do
something that deals with those

00:08:06.700 --> 00:08:09.940
risks and creates all kinds
of short-term benefits.

00:08:09.940 --> 00:08:12.190
And that becomes a
design consideration

00:08:12.190 --> 00:08:13.744
in formulating the policy.

00:08:13.744 --> 00:08:15.410
And if you can't do
it, you can't do it.

00:08:15.410 --> 00:08:17.680
But I would argue
that the problem

00:08:17.680 --> 00:08:22.450
with the precautionary
principle is it doesn't link to,

00:08:22.450 --> 00:08:27.460
and let's look at ways of using
this theory that we're not

00:08:27.460 --> 00:08:28.840
going to do something.

00:08:28.840 --> 00:08:32.200
I told you the story about the
central Arctic and fisheries

00:08:32.200 --> 00:08:33.010
question.

00:08:33.010 --> 00:08:34.570
You can use a
period of time to do

00:08:34.570 --> 00:08:37.570
more scientific investigation
that might in fact lead

00:08:37.570 --> 00:08:40.419
us to understand
better ways of using

00:08:40.419 --> 00:08:44.290
those resources as opposed
to, oh my god, we're scared.

00:08:44.290 --> 00:08:45.540
A bad thing could happen.

00:08:45.540 --> 00:08:47.010
It might only be a 1% chance.

00:08:47.010 --> 00:08:49.160
But, oh my god, if
it happens horrible,

00:08:49.160 --> 00:08:51.860
horrible things will
affect us in the future.

00:08:51.860 --> 00:08:53.220
So let's not do it.

00:08:53.220 --> 00:08:54.040
LAWRENCE SUSSKIND: But the
precautionary principal

00:08:54.040 --> 00:08:56.070
has been linked to
alternatives analysis.

00:08:56.070 --> 00:08:59.210
Now nobody doesn't ask,
what are the alternatives.

00:08:59.210 --> 00:09:03.490
And by the way, the big failure,
Bush and Cheney decided,

00:09:03.490 --> 00:09:05.920
we're going to get
Saddam before he gets us.

00:09:05.920 --> 00:09:08.530
He was operating under
precautionary principle.

00:09:08.530 --> 00:09:11.710
Look what the application of
that precautionary principle

00:09:11.710 --> 00:09:12.910
brought us.

00:09:12.910 --> 00:09:15.640
SPEAKER 1: If you
only say, let's

00:09:15.640 --> 00:09:18.760
look at alternatives,
that's not the same as let's

00:09:18.760 --> 00:09:19.802
search for alternatives--

00:09:19.802 --> 00:09:20.843
LAWRENCE SUSSKIND: Right.

00:09:20.843 --> 00:09:22.450
SPEAKER 1: --that
both reduce the risk

00:09:22.450 --> 00:09:24.783
and create a variety
of alternatives.

00:09:24.783 --> 00:09:25.530
LAWRENCE SUSSKIND: And
develop those alternatives.

00:09:25.530 --> 00:09:27.280
SPEAKER 1: And let's
be clear about trying

00:09:27.280 --> 00:09:29.526
to distribute the benefits
of those alternatives

00:09:29.526 --> 00:09:30.400
in a progressive way.

00:09:30.400 --> 00:09:31.990
You could add that, too.

00:09:31.990 --> 00:09:32.560
Anyway.

00:09:32.560 --> 00:09:35.310
Thank you both for coming.