WEBVTT

00:00:00.000 --> 00:00:01.956 align:middle line:90%
[SQUEAKING]

00:00:01.956 --> 00:00:03.423 align:middle line:90%
[RUSTLING]

00:00:03.423 --> 00:00:06.357 align:middle line:90%
[CLICKING]

00:00:06.357 --> 00:00:22.118 align:middle line:90%


00:00:22.118 --> 00:00:23.160 align:middle line:90%
RICHARD DE NEUFVILLE: Hi.

00:00:23.160 --> 00:00:25.050 align:middle line:84%
It's me again--
Richard de Neufville.

00:00:25.050 --> 00:00:27.300 align:middle line:84%
And I'm happy to
talk with you again.

00:00:27.300 --> 00:00:29.820 align:middle line:84%
And I'm going to be talking
some more about this notion

00:00:29.820 --> 00:00:32.310 align:middle line:84%
of the forecast
being "always wrong,

00:00:32.310 --> 00:00:36.000 align:middle line:84%
where "always" is in
quote marks to give

00:00:36.000 --> 00:00:38.920 align:middle line:84%
the general idea that we
can't count on it being right.

00:00:38.920 --> 00:00:41.610 align:middle line:84%
But of course, sometimes,
we make a forecast.

00:00:41.610 --> 00:00:44.520 align:middle line:84%
And what actually happens
is exactly what's forecast,

00:00:44.520 --> 00:00:47.520 align:middle line:84%
although the general
rule is, it is not.

00:00:47.520 --> 00:00:50.940 align:middle line:84%
I showed you some
slides previously

00:00:50.940 --> 00:00:55.920 align:middle line:84%
about how our cost estimate
of creating projects

00:00:55.920 --> 00:00:57.150 align:middle line:90%
was uncertain.

00:00:57.150 --> 00:01:00.370 align:middle line:84%
But now, let's look at
predictions of project demand.

00:01:00.370 --> 00:01:02.850 align:middle line:84%
So this is another
class of possibilities,

00:01:02.850 --> 00:01:07.620 align:middle line:84%
where we can deviate from
what's actually happening.

00:01:07.620 --> 00:01:11.280 align:middle line:84%
This one is one of
the many predictions

00:01:11.280 --> 00:01:15.000 align:middle line:84%
of the price of oils in
a particular timeline

00:01:15.000 --> 00:01:16.470 align:middle line:90%
between here now--

00:01:16.470 --> 00:01:20.370 align:middle line:90%
2004 and 10 years later on.

00:01:20.370 --> 00:01:24.480 align:middle line:84%
And what we have here is an
official projection, which

00:01:24.480 --> 00:01:26.730 align:middle line:90%
is this line that's green.

00:01:26.730 --> 00:01:28.140 align:middle line:90%
And that's the low one.

00:01:28.140 --> 00:01:30.180 align:middle line:84%
And then, there's the
red one, which is a high.

00:01:30.180 --> 00:01:33.240 align:middle line:84%
And the best forecast
is that blue one,

00:01:33.240 --> 00:01:36.510 align:middle line:84%
where it's supposed to
be around $40 a barrel,

00:01:36.510 --> 00:01:40.610 align:middle line:90%
as predicted in 2004.

00:01:40.610 --> 00:01:47.160 align:middle line:84%
The big line up here at the top
is what was actually happening.

00:01:47.160 --> 00:01:50.990 align:middle line:84%
And you can see that the
line, the prediction,

00:01:50.990 --> 00:01:54.350 align:middle line:84%
was not only not within
the range of what

00:01:54.350 --> 00:01:58.130 align:middle line:84%
was originally predicted, but
it was about twice as much.

00:01:58.130 --> 00:02:01.130 align:middle line:84%
And it went up to
over $100 a barrel

00:02:01.130 --> 00:02:03.650 align:middle line:84%
in that period, which
is only 6 years ahead.

00:02:03.650 --> 00:02:07.920 align:middle line:84%
So the expert agency that had
been doing that was saying,

00:02:07.920 --> 00:02:09.810 align:middle line:84%
well, it's going to
be in this range.

00:02:09.810 --> 00:02:13.760 align:middle line:84%
It's a reasonably tight
range of $40 to $60, perhaps.

00:02:13.760 --> 00:02:15.680 align:middle line:84%
And in fact, what
actually happens

00:02:15.680 --> 00:02:20.430 align:middle line:84%
was terrifically different
from what occurred.

00:02:20.430 --> 00:02:23.210 align:middle line:84%
Now, I've just shown you
this one particular graph.

00:02:23.210 --> 00:02:27.410 align:middle line:84%
But the fact is that
oil price predictions

00:02:27.410 --> 00:02:32.030 align:middle line:84%
are notorious for being
different than what actually

00:02:32.030 --> 00:02:32.990 align:middle line:90%
happens.

00:02:32.990 --> 00:02:36.570 align:middle line:84%
And you can appreciate
why, because first of all,

00:02:36.570 --> 00:02:39.170 align:middle line:84%
there have been
wars in this area,

00:02:39.170 --> 00:02:42.230 align:middle line:84%
in the Middle East and
elsewhere, which change things.

00:02:42.230 --> 00:02:44.600 align:middle line:84%
There have been huge
economic developments--

00:02:44.600 --> 00:02:46.640 align:middle line:90%
the rise of China.

00:02:46.640 --> 00:02:50.180 align:middle line:84%
Its appetite for oil has
raised the prices enormously.

00:02:50.180 --> 00:02:53.270 align:middle line:84%
And then, not yet
shown on this graph,

00:02:53.270 --> 00:02:56.360 align:middle line:84%
but you will know it, which is
the development of fracking,

00:02:56.360 --> 00:02:59.880 align:middle line:84%
which has really reduced the
prices of oil considerably.

00:02:59.880 --> 00:03:04.820 align:middle line:84%
So in general, the
forecasts of what

00:03:04.820 --> 00:03:08.930 align:middle line:84%
will happen to the price of
oil have not been correct--

00:03:08.930 --> 00:03:11.840 align:middle line:84%
systematically lobbied
correct, not just

00:03:11.840 --> 00:03:13.850 align:middle line:84%
by the average
person on the street,

00:03:13.850 --> 00:03:17.240 align:middle line:84%
but by the International
agencies and national agencies

00:03:17.240 --> 00:03:19.550 align:middle line:84%
charged with projecting
the price of oil.

00:03:19.550 --> 00:03:23.060 align:middle line:84%
It doesn't happen the
way they say it will.

00:03:23.060 --> 00:03:24.560 align:middle line:90%
And I'm not blaming them.

00:03:24.560 --> 00:03:26.270 align:middle line:90%
They're doing the best they can.

00:03:26.270 --> 00:03:29.300 align:middle line:84%
But there are things
that happen around them

00:03:29.300 --> 00:03:31.980 align:middle line:90%
that change the possibilities.

00:03:31.980 --> 00:03:33.390 align:middle line:90%
Here's a different one.

00:03:33.390 --> 00:03:37.050 align:middle line:84%
Here is a projection for,
in the United States,

00:03:37.050 --> 00:03:41.670 align:middle line:84%
the production of wind
energy installations.

00:03:41.670 --> 00:03:46.410 align:middle line:90%
So the blue line was a forecast.

00:03:46.410 --> 00:03:48.480 align:middle line:84%
And it seemed
perfectly reasonable.

00:03:48.480 --> 00:03:51.420 align:middle line:90%
It showed an active growth.

00:03:51.420 --> 00:03:57.090 align:middle line:84%
And that was considered to be
fairly aggressive at that time.

00:03:57.090 --> 00:03:59.340 align:middle line:90%
And then, what happened?

00:03:59.340 --> 00:04:02.670 align:middle line:84%
Actually, the red bars
show the actual amount that

00:04:02.670 --> 00:04:04.300 align:middle line:90%
was installed in those years.

00:04:04.300 --> 00:04:07.470 align:middle line:84%
And it got to be
easily twice what

00:04:07.470 --> 00:04:10.710 align:middle line:84%
happened in most of the
years except the last one.

00:04:10.710 --> 00:04:12.160 align:middle line:90%
What happened here?

00:04:12.160 --> 00:04:14.550 align:middle line:84%
Well, what I
believe happened is,

00:04:14.550 --> 00:04:17.970 align:middle line:84%
the explanation was that
somewhere in the government,

00:04:17.970 --> 00:04:20.070 align:middle line:90%
they passed regulations--

00:04:20.070 --> 00:04:24.180 align:middle line:84%
or in state governments, they
passed regulations encouraging

00:04:24.180 --> 00:04:26.890 align:middle line:90%
the use of wind energy.

00:04:26.890 --> 00:04:28.500 align:middle line:90%
So people responded.

00:04:28.500 --> 00:04:32.070 align:middle line:84%
This was unexpected by the
people who made the forecast

00:04:32.070 --> 00:04:34.210 align:middle line:90%
originally, but it happened.

00:04:34.210 --> 00:04:36.120 align:middle line:84%
This is the kind of
thing that does occur.

00:04:36.120 --> 00:04:38.280 align:middle line:84%
Sometimes, it leads
to an increase.

00:04:38.280 --> 00:04:40.590 align:middle line:84%
Sometimes, it leads
to a decrease.

00:04:40.590 --> 00:04:45.210 align:middle line:84%
But the thing is that
there are changes.

00:04:45.210 --> 00:04:46.540 align:middle line:90%
Let me go to another one.

00:04:46.540 --> 00:04:50.760 align:middle line:84%
This time, it's for water
use in a sector of England

00:04:50.760 --> 00:04:54.510 align:middle line:84%
called Anglia done by
the Anglia water people.

00:04:54.510 --> 00:04:59.730 align:middle line:84%
And the line in the middle here,
the black line going straight

00:04:59.730 --> 00:05:01.530 align:middle line:90%
up, is the projected demand.

00:05:01.530 --> 00:05:05.430 align:middle line:84%
And they said, well,
we're not sure about this.

00:05:05.430 --> 00:05:08.820 align:middle line:84%
So let's go basically
plus or minus 10%.

00:05:08.820 --> 00:05:10.770 align:middle line:90%
Gives us a reasonable range.

00:05:10.770 --> 00:05:15.160 align:middle line:84%
Some other people earlier had a
more aggressive line going up.

00:05:15.160 --> 00:05:17.820 align:middle line:84%
But basically, it
was continued use

00:05:17.820 --> 00:05:22.200 align:middle line:84%
associated with more people
and more washing machines,

00:05:22.200 --> 00:05:23.310 align:middle line:90%
and whatever you have.

00:05:23.310 --> 00:05:25.530 align:middle line:90%
But it was a growth pattern.

00:05:25.530 --> 00:05:28.020 align:middle line:84%
Now, remarkably, what
actually happened

00:05:28.020 --> 00:05:30.840 align:middle line:90%
was, the water use decreased.

00:05:30.840 --> 00:05:33.720 align:middle line:84%
And then, at some point,
it increased a little bit--

00:05:33.720 --> 00:05:35.410 align:middle line:90%
or up to what was expected.

00:05:35.410 --> 00:05:36.840 align:middle line:90%
And then, it went down again.

00:05:36.840 --> 00:05:38.220 align:middle line:90%
It was different.

00:05:38.220 --> 00:05:42.060 align:middle line:84%
Now, it's just
another example of how

00:05:42.060 --> 00:05:44.880 align:middle line:84%
the forecast and the
actuality is different.

00:05:44.880 --> 00:05:46.590 align:middle line:90%
What happened here?

00:05:46.590 --> 00:05:51.070 align:middle line:84%
It wasn't they stopped
washing or things like that.

00:05:51.070 --> 00:05:55.500 align:middle line:84%
But what I believe happened
was that the water company

00:05:55.500 --> 00:05:59.820 align:middle line:84%
decided to raise the
prices to finance

00:05:59.820 --> 00:06:02.190 align:middle line:90%
new facilities-- good thing.

00:06:02.190 --> 00:06:04.950 align:middle line:84%
But people said, ah,
if I'm paying more,

00:06:04.950 --> 00:06:06.780 align:middle line:84%
I'm going to pay
attention to my leaks.

00:06:06.780 --> 00:06:11.070 align:middle line:84%
And in fact, in many water
supply systems, about 1/3

00:06:11.070 --> 00:06:15.360 align:middle line:84%
of the actual use of
water goes into leaks,

00:06:15.360 --> 00:06:22.890 align:middle line:84%
because faucets drip because
the mains were built 50,

00:06:22.890 --> 00:06:26.760 align:middle line:84%
100 years ago, and they leak,
and because people don't bother

00:06:26.760 --> 00:06:29.220 align:middle line:84%
to save water by installing
low-flush toilets

00:06:29.220 --> 00:06:30.270 align:middle line:90%
or what have you.

00:06:30.270 --> 00:06:33.900 align:middle line:84%
So it's in fact,
quite easy for them

00:06:33.900 --> 00:06:37.680 align:middle line:84%
to decrease the
water that they use,

00:06:37.680 --> 00:06:39.720 align:middle line:84%
or the water that goes
through their households,

00:06:39.720 --> 00:06:44.260 align:middle line:84%
or their industries if they have
the right incentive to do it.

00:06:44.260 --> 00:06:46.320 align:middle line:84%
Same thing happened
in Boston when

00:06:46.320 --> 00:06:50.610 align:middle line:84%
they built a large
sewage protection

00:06:50.610 --> 00:06:52.920 align:middle line:90%
unit in Boston Harbor.

00:06:52.920 --> 00:06:55.650 align:middle line:84%
They spent about
$6 billion for it.

00:06:55.650 --> 00:06:57.240 align:middle line:90%
They raised the prices.

00:06:57.240 --> 00:07:01.680 align:middle line:84%
My prices went up from about
$50 a year to about $500 a year.

00:07:01.680 --> 00:07:04.320 align:middle line:84%
And like other people,
I installed all kinds

00:07:04.320 --> 00:07:06.630 align:middle line:84%
of water savings
devices in my house,

00:07:06.630 --> 00:07:10.230 align:middle line:84%
and the water usage in
Boston went down also.

00:07:10.230 --> 00:07:15.000 align:middle line:84%
But that wasn't anticipated
by the engineering experts,

00:07:15.000 --> 00:07:18.000 align:middle line:84%
by the experts that
were thinking about it.

00:07:18.000 --> 00:07:22.530 align:middle line:84%
They're making the
projection as the water

00:07:22.530 --> 00:07:24.150 align:middle line:90%
system was being developed.

00:07:24.150 --> 00:07:29.880 align:middle line:84%
So again, you have a
situation where the forecast

00:07:29.880 --> 00:07:32.250 align:middle line:90%
and the actuality don't match.

00:07:32.250 --> 00:07:38.160 align:middle line:84%
You might ask, how do I
get these kinds of graphs?

00:07:38.160 --> 00:07:41.760 align:middle line:84%
And this gives you an exercise
you can do for yourself.

00:07:41.760 --> 00:07:45.660 align:middle line:84%
I ask my students
to choose any topic

00:07:45.660 --> 00:07:48.690 align:middle line:84%
they want to, to look
for a series of forecasts

00:07:48.690 --> 00:07:52.140 align:middle line:84%
for the future, and to compare
them with what happened.

00:07:52.140 --> 00:07:55.980 align:middle line:84%
And this particular one
was done by one student.

00:07:55.980 --> 00:07:57.880 align:middle line:84%
The others were done
by other students.

00:07:57.880 --> 00:08:02.490 align:middle line:84%
But you can test my
supposition, my argument

00:08:02.490 --> 00:08:05.640 align:middle line:84%
that the demands
are not as predicted

00:08:05.640 --> 00:08:08.670 align:middle line:84%
by choosing a topic
that you want yourself--

00:08:08.670 --> 00:08:14.190 align:middle line:84%
the sales of the Ford
pickups, the introduction

00:08:14.190 --> 00:08:17.520 align:middle line:84%
of electric vehicles, whatever
you might be interested in.

00:08:17.520 --> 00:08:22.110 align:middle line:84%
You could see what forecasts
were done 5, 10 years ago, what

00:08:22.110 --> 00:08:25.840 align:middle line:84%
has now been happening, and
make the comparison yourself.

00:08:25.840 --> 00:08:30.210 align:middle line:84%
Sometimes, the forecast
will be a good indicator

00:08:30.210 --> 00:08:31.380 align:middle line:90%
of what actually happened.

00:08:31.380 --> 00:08:33.659 align:middle line:84%
But in general,
these are the kind

00:08:33.659 --> 00:08:36.539 align:middle line:84%
of results that my students
bring up to me when I give them

00:08:36.539 --> 00:08:37.570 align:middle line:90%
that exercise.

00:08:37.570 --> 00:08:41.200 align:middle line:84%
These are the kind
of graphs, pictures,

00:08:41.200 --> 00:08:45.300 align:middle line:84%
diagrams that you can
have for yourselves,

00:08:45.300 --> 00:08:46.620 align:middle line:90%
create for yourselves.

00:08:46.620 --> 00:08:49.640 align:middle line:90%


00:08:49.640 --> 00:08:54.220 align:middle line:84%
So why is it that the expert
forecasts are often wrong?

00:08:54.220 --> 00:08:56.500 align:middle line:90%
What is happening?

00:08:56.500 --> 00:09:00.190 align:middle line:84%
The general answer is
because things happen.

00:09:00.190 --> 00:09:01.210 align:middle line:90%
What things?

00:09:01.210 --> 00:09:04.360 align:middle line:84%
Well, there might be
technological advances

00:09:04.360 --> 00:09:07.900 align:middle line:84%
that you don't need to do things
the way you did them before.

00:09:07.900 --> 00:09:10.660 align:middle line:84%
Or there's fracking, which
made the price of oil

00:09:10.660 --> 00:09:12.520 align:middle line:90%
cheaper, for example.

00:09:12.520 --> 00:09:15.760 align:middle line:84%
There are weather patterns
that account, maybe,

00:09:15.760 --> 00:09:18.970 align:middle line:84%
for some of the increase and
variability in the use of water

00:09:18.970 --> 00:09:21.590 align:middle line:84%
in England in that
particular example.

00:09:21.590 --> 00:09:23.110 align:middle line:90%
There are economic crises--

00:09:23.110 --> 00:09:27.880 align:middle line:84%
the economic situation
isn't so good,

00:09:27.880 --> 00:09:29.860 align:middle line:90%
and people don't buy as much.

00:09:29.860 --> 00:09:34.820 align:middle line:84%
And things are cheaper or
more difficult. Things happen.

00:09:34.820 --> 00:09:37.970 align:middle line:90%
That is the lesson from today.

00:09:37.970 --> 00:09:39.990 align:middle line:90%
Thank you very much.

00:09:39.990 --> 00:09:51.000 align:middle line:90%