Eric Schmidt at Corporate Eco Forum

Uploaded by Google on 09.09.2008


M. R. RANGASWAMI: We are very privileged to have Eric
Schmidt join us tonight.
He's actually spoken at my conferences when he was at a
prior company.
But I met with him about six months ago when we were at a
dinner together.
And I really presented him the Corporate Eco Forum program
and asked that he be a keynote.
And he said, look, M. R., if I have the time on my schedule,
I'll do it.
And fortunately, he had time on his calendar.
So Eric is a very passionate individual, as an individual
and a person, about the green initiatives
and the green movement.
And also, Google as a company has undertaken numerous
initiatives in this space, done a lot of experimentation,
and really helping make a big, big impact on the Eco agenda.
So as you know, this is a forum
for Global 500 companies.
And who better to have be our keynote than a Global 500 CEO?
So please help me in welcoming Dr. Eric Schmidt.

ERIC SCHMIDT: Hi, M. R. Thank you very much.

This is a fun group, and it's been a fun
conference for the day.
Shall we get the screen up?
And what I thought we would begin with his do--
What's a dinner without a few presentations?
We don't use PowerPoint anymore.
We use some better technology.
And I have with me Michael Jones.
Raise your hand, Michael.
And Michael happens to be the inventor of Google Earth and
is a good personal friend.
And I asked him, if he were going to talk about the earth
in the way that he loves it and that all of us love it,
how would he talk about it?
And curiously, Michael, you have us starting at our hotel.
So where are we going next?
MICHAEL JONES: Well, let's find out where we are.
ERIC SCHMIDT: We're at the Fairmont, Michael.

MICHAEL JONES: A small piece of the earth.
ERIC SCHMIDT: So one of the first things we looked at was,
what are people doing with Google Earth with expected
climate change?
Are they, in fact, using this as a tool to try to understand
what happens?
And there's this interesting group called the Hadley Met
Office, which is the UK official weather and climate
information research center.
And the British Antarctic Centre created
this sort of layer.
And what's interesting about this layer is that, as the
temperature goes up, it starts off kind of--
blue is the normal temperature-- and then, as the
temperature goes up, you see it changing and changing.
And what's interesting is you take the
innovation, say it's 2020.
And I'm using midpoint projections
here for climate change.
North America, two degrees Celsius.
So what is that, three, four degrees?
No big deal.
Warmer is better probably, anyway, right?
But in the North Pole, it's 15 degrees Celsius.
So you see the little ice cap up there?
It's going to be really nice to all those
open shipping lanes.
That's the way I think about it.
Unfortunately, there's a lot of other things that happened
up there, as everybody knows, including basically all of the
wildlife and all the trees are gone.
But we're not done yet.
They then basically take us out to 2050, which is five
degree Celsius, which again is kind of toasty here in our
part of the world.
It's 40 degrees in the North Pole.
So if you think there's any ice up there, you're not
paying attention to freezing point.
And of course, it's all open.
It's all gone, along with probably a sixth of the
world's plant life.
Oh yeah, well, we had enough-- we have the other 5/6, right?
You get the idea.
This is a very, very, very tragic scenario.
Keep going, Michael.
So what happens, of course, is as it gets higher, it gets
even worse.
So what we thought we'd do is--
can government policies make a difference?
We thought, well, are there some examples?
Everyone's always saying, nobody can do
anything about this.
So what are some examples of government policies?
So here's an example.
Mexico and Guatemala.
And show us the border, Michael.
Actually, I'm sorry, this is over '74
through basically today.
And there's the border, the difference in policies of
Mexico versus Guatemala.
Kind of interesting.
You can just see it as a straight line, or in this
case, as a corner.
Go ahead.
Let go to [INAUDIBLE].
Anybody been to Uruguay, a well run country?

Which side is Brazil, and which side is Uruguay?
This is a trick question.
Uruguay is on the lower, and Brazil's on the upper.
Anybody care to know what the border looks like?
There it is.
Let's zoom in.
By the way, you see the trees there on the
left, on the right?
Those are little trees that they planted.
Government policy does make a difference.
Another question, can any of you make a difference?
Of course we can.
There's a fellow in Texas whose name is Luecke, who
decided to make a difference.

And Michael, you promise me this is a real picture.
You didn't just add this.
OK, you promise?
OK, these are half mile long letters.
And by the way, that is a power line,
in case you're wondering.
Those are the tall power lines, because you can't plant
trees next to the power lines.
It gives you a sense of the scale of--
a lot of people say lucky, but the correct
pronunciation is loo-key.
This is a guy who had a vision that he could make a
And he did, with a little bit of marketing.
So when we do these policies, we have to have some policies
about where do we want to spend our money.
and I'll talk about this a bit more-- has invested $10
million in enhanced geothermal, which I want to
talk about a little bit.
And what you do is, enhanced geothermal--
I think people here know-- is there's a lot of
heat inside the earth.
Remember this as children?
It's really hot.
It's why you can't dig to China.
That doesn't work.
There's a reason.
But basically, if you go down a little bit, there's heat.
But the heat varies.
So here's a chart of the suitability of states for
enhanced geothermal.
And what's interesting, of course, is the Western states
have higher suitability.
Again, let's make decisions with facts.
It's a big country.
We've got a lot of choices.
Let's put those things where they make a difference.
Keep going.

Showing the circulation pattern.
So here's an example of the temperature of depth.
Now, we used 5.5 kilometers, because that's sort of a
reasonable depth to use oil drilling equipment, to get
down there and start playing with this.
And again, you can see there's a fair amount of temperature
there that's available.
And we're talking about temperature of 150 to 200
degrees Celsius.
And by the way, the temperature doesn't stop.
You sort of don't run out of it.
There's a lot underneath there.
So we calculated that if you dug down 5.5 kilometers in a
two percent [UNINTELLIGIBLE], we'd have 15,000 megawatts of
energy annually in California, for example.
Anybody here been to the San Gorgonio Pass? it's the pass
when you're driving from LA to Palm Springs.
Anybody ever see all those wind--
everybody sees that, right?
So here we are.
And Michael's now driving along.
And the wind farms there generate 893 gigawatt hours of
electricity a year.
It's probably one of the best places in America.
And basically, the average wind speed there is about 20
miles per hour.
So again, you can see them here.
You can see them on Google Earth.
And there's plenty of things that you can do with this.
Now, let's take a look at Google.
And as an aside, One of the first meetings I went to, they
said, did you know--
I always start those things, did you know--
that 40 percent of climate contributions, CO2
contributions, are coming from buildings?
And I don't really know much about this.
And they said, and it's a big problem.
I'm going to talk about that a little bit.
And I thought, I could actually do
something about this.
I actually have a lot of buildings.
Big deal, right?
Google's a big place.
So I called the facilities people.
And I said, what are we doing about energy efficiency?
And they said, oh, well, we lease the buildings, we bought
these ones, the usual sort of answers.
And I said, well, anybody know how much it would cost to make
these buildings very efficient?
And here you see the Google campus.
And the initial estimate was $5 million a year.
So it's going to be $5 million in capital.
And I said, well, that's a lot of money.
I try to look like a CEO.
And everybody asked you $5 million, say it's a lot of
money, because you want it to be cheaper.
So they said $5 million.
And I said, well, how long is the payback?
And they said, two and a half years.
I said, the payback on a $5 million capital investment is
two and a half years, and you haven't done it already?
And they said, no, we didn't have permission.
I said, you don't have to ask permission for
that kind of payback.
In other words, what's wrong with you?

But I learned something.
And I learned something, that in normal corporations-- and I
suspect all of you have this problem--
nobody asks the questions in an integrated way.
$5 million, immediate payback, what's not to like about this?
Furthermore, this is when gas was half the price and
electricity was cheaper.
And one of the things that people-- and again, business
people will understand this-- is savings compound infinitely
when prices go up.
This is an easy thing to understand, very, very easy to
When prices go up, you've already got the savings.
The savings get bigger, not less.
It's called increasing savings.
I've just invented it, OK?
Why this is missing in corporate
accounting is beyond me.
Somehow they don't get this.

So then I said, what else could we do?
And they said, well, we actually have some ideas about
using solar.
And we'll build carports for all the cars.
And we'll put solar on top.
And in fact, here, we have a meter, which shows you our
we have dynamic systems inside our campus--
which shows you by building how we're doing against power
usage-- and we use lots of electricity--
and of course, the contribution from solar.
And my point here is not that these are hard to do.
These are simple to do.
But somebody had to do them.
So the question is, have you done them in your company?
If you haven't, why not?
It's just a scorecard.
And it's just capital, with a very, very quick payback.
In our case, we had a lot of fun with this.
And about a year and a half ago, when it was completed, it
was the largest private installation of solar on top
of the roofs of the buildings that we own.
And by the way, the savings continue.
The payback is incredible.
Payback was less than 10 years.
And that's ignoring all the various subsidies, which, of
course, we were happy to pocket.
So eight buildings, two carports, the total utility
meter, the building 40 dashboard, which we're
showing here now.
And here, again, you can show the contribution from solar.
You can show the net contribution
and the total demand.
We can monitor this.
We can understand it.
And we can go forward.
Michael, take us back to where we started here.
Here we are at Google.

Come back up to San Francisco, I think.

ERIC SCHMIDT: What's interesting about this kind of
a tool is you didn't realize what life was like after you
start using it.
How did you operate without it?
So here we are again.
But of course, we're really there.
We're actually up on the top part of that
building right there.
And it's not really daytime.
It's kind of dusk.
And by the way, anybody look out the window, do you see the
Golden Gate Bridge?
Because it's obscured by fog.
But in Google Earth, you can see it 100 percent.
So with that, thank you very much, Michael.

So I wanted to show you that, because I think it helps when
people can see it.
You can see how you can make a difference and so forth.
Unfortunately, what I've concluded is that people in
this room get it.
A lot of people I know get it.
But we have a total failure of political leadership, at least
in the United States, and perhaps in the
world, on these issues.
It is a fundamental failure of leadership.
And it comes from--
Let's start talking about the things that really threaten
Well, obviously, nuclear proliferation.
That's something which many people worked on when we were
born and when we were young.
And the other one is climate change.
Certainly, life as we know it is now clearly threatened by
this thing over the next 50 to 75 years.
And what's interesting to me is that, as a political
dialogue, there is a lack of understanding of the value of
global alliances.
There's a lack of understanding of the benefits
of technology and how technology can actually change
the dialogue.
There's a lack of understanding that there's an
opportunity before us that is a massive opportunity that
solves a lot of these problems at once.
And so it seems to me that this is time
to seize the moment.
And I think what I would like to lay down tonight is it's
time for everybody to get off their rear ends and start
taking this seriously.
So here's my argument.
It's very simple.
I'm not a politician.
The government is about to do a massive stimulus package.
They're going to reward all their usual friends, depending
on who their friends are.
Because the country is in this, are we in a recession or
are we not, kind of a state.
Everybody's talking about it all the time in Washington.
Read about it in the press.
So instead of just randomly handing out all the money to
all the normal suspects, why don't we take an opportunity?
A crisis is a terrible thing to waste.
Why don't we take the crisis, and use this as an opportunity
to retool the energy infrastructure
of the United States?
It's very straightforward, not a very hard idea.

And I've got a proposal.
Unfortunately, I don't have everyone in agreement.
So help me with this.
It's very simple.
What's the number one reason we're spending all this money
on the military?
Fundamentally, because of oil security, right?
So if we could make the price of oil be lower,
that would be positive--

positive for us, not for everybody.
If we can solve that problem by reducing demand, we've got
another problem.
We have an employment problem.
We have a job creation problem.
And by the way, the jobs that are created in energy policy--
solar, geothermal, wind--
are, in fact, high paying blue collar, white collar jobs.
And by the way, they're jobs in the US, in case anybody
didn't pay attention.
Furthermore, we have the best scientists and the best
technology, the best research community in the world, people
beginning to work on this.
Maybe we could build export industries and get some of the
stuff out, and build some more revenue for the country.
So it seems to me that you've solved-- oh, and by the way, I
forgot about climate change.
If you actually did this, if you actually managed to figure
out a way to replace--
get this--
replace all of the electricity generation in the country by
renewable sources over a 20 year period, and you got rid
of half of the cars' use of gas on a per car basis--
which, again, is achievable based on our analysis--
you would cut our climate emissions by half.
And by the way, this is a replicable strategy for all
the other countries, too.
So we could show them how to do it.
So I sat down and, with a whole team, did this
And it's actually pretty interesting.
If you can achieve 100% US electricity generated by
renewable electricity sources, 50% percent of the vehicles
being plug-in hybrids-- and I want to talk about that--
and assuming that per capita energy consumption remains at
the same levels as it has this year-- which is, by the way,
an easy target, because they can clearly be made lower.
California's done this for years.
We all know this--
our calculation was that the total cost of
this was $2.7 trillion.
How are we going to come up with that money?
That's a lot of money, even for the US.
Well the reason you can do it is because there's an
offsetting $2.1 trillion in savings.
And the reason the savings occurs is because of
Remember that whole thing of that
increasing efficiency curve?
And what's interesting is our calculations say that it turns
out to be about using an 8% discount rate.
It ends up being about $200 billion in net present value,
which is roughly a fraction of what we spend
on oil in this country.
So even if you disagree with my methodology, there is a
real opportunity right in front of us to actually solve
this problem if people would stop arguing about all of the
minor issues and start talking about the fundamental
opportunity, which is to retool the energy
infrastructure of the United States.
And by the way, I forgot--
because we're politicians here in the room, right?--
How many jobs does it create?
500,000 jobs in this calculation in wind alone.
And by the way, we're not running out of wind.
There's a lot of wind.
The other things to think about-- and big country, it's
useful to have a big country with lots of wind--
the other most interesting thing about it is that wind
occurs all the time.
There are no cases where the US has been becalmed.

That would be historic.
The wind stopped today.
Sorry we don't have any electric power because the
wind stopped.
You see my point.

So we focused on solar, wind, and enhanced geothermal simply
because they're the ones that are most likely.
We explicitly ruled out nuclear power for a number of
reasons, including the fact that it has infinitely high
capital cost and has lots of energy security issues.
So if you imagine that the root problems are: the energy
mix is too carbon intensive-- which has to do with history,
and so forth and so on--
that the way we move the power is inefficient, we have the
wrong price signals.
A regulatory system is basically targeted to the
existing incumbents.
We all know this.
The model that you want is a distributed renewable power
structure with cogeneration and other sources that
actually make the thing more efficient.
We know how to do all of this now.
It's just a matter of sitting in a room and saying, the
payback is how long?
Why are we not doing this now?
It's actually right in front of us.
Now, we all know the carbon intensity issue is basically a
coal issue.
And these plants that are being built are responsible
for about 80% of the greenhouse gas emissions from
Electricity is a large component of this.
We all know the issues of climate change.
The aging electric power infrastructure that exists was
designed without the kind of smart grid ideas that would be
obvious to computer scientists, obvious to people
who sit in this--
and there are reasons for it.
They didn't have the switching technology.
They didn't have the signalling technology.
These things were built 20 or 30 years ago, when the only
kind of computer they could get was a mainframe.
And they didn't want to put that next the power station.
There's a lot of leakage in this network, and so forth.
The regulatory system--
with the exception of a few states, including California--
are incentivized around revenue rather than ROI and
cost savings.
That's a straightforward regulatory change that can be
made on a per PUC basis, or at a national level if they won't
move fast enough.
The cost of fossil fuels is not internalized into these
systems. So all of these
externalities are not in there.
So how do you do this?
Well, my ideal scenario would be to say, let's take
everybody and let them all compete fairly without
subsidies, and let's focus on efficiency first. The problem
with that is that there are so many subtle subsidies for the
existing fossil fuels that it's not
really a fair strategy.
Though, you'd like to do that.
Maybe you could wean them off.
And again, you all have heard a lot about these subsidies.
But there are many, many, many in terms of special treatment
around an oil leases, various forms of tax
credits, and so forth.
Why don't we just take all those and shift them over to
this other thing to help out?
But my plan doesn't even require that.
So why don't we figure out a way to work on energy
We're good at that.
It requires good technology, good invention, and so forth.
And when utilities pay people to make their buildings more
efficient, they do it at about $0.01 to
$0.03 a kilowatt hour.
So one way to think about that on an economic basis is that
investing in energy efficiency is cheaper than the cost of
coal, or the cost of wind or what have you, which is $0.05
and higher per kilowatt hour.
So my important point here is that, if we change the
incentives so people start thinking about efficiency
first, businesses will do this.
They will go and insulate their plants, make their
businesses more efficient, and so forth and so on.
And then, the other thing that you have to do is, you have to
look at total cost of ownership.
What happens is, the way the accounting works and the way
executives work is they always think about a purchase rather
than think of the life cycle cost. The life cycle cost,
especially with these increasing costs, becomes
And by the way, the savings--
a typical example is, you don't really need to replace
the insulation in your house, but you have
to replace the heater.
The passive systems are much less likely to wear out than
the active systems, which have motors and maintenance, and so
forth and so on.
By the way, all that labor costs go up and so forth.
So if you take capital plus operating budgets, you
discover that you need to solve this problem.
In Google's case, the capital and operating budgets for our
data center is owned by one person, who happens
to be in this room.
So he's very, very good at making these long term
And of course, the other trick is to create a real time
information loop.
What happens is that there's a lot of evidence that things
which are measured, consumption goes down.
There have been a lot of studies about this.
Smart meters with homes, for example, there's been a 10%
reduction in use just by being aware of it.
Now, 10%, you go, no big deal.
10% over the United States in every home?
Think about it.
And we have the underlying computer technology--
monitoring and so forth-- to make these kinds
of monitoring trivial.
These are very small, simple devices, easily applied,
again, and retrofitted into the systems. So consumers
basically can understand all of this.
The same thing has to do with PC equipment.
One of the other interesting things about office buildings
is that the computers--
my world--
generate a lot of the electricity use.
And a lot of those computers are not doing very much.
In fact, a lot of the heat load and a lot of the ultimate
electric load is driven by the fact that these computers are
on all the time.
So small changes in power management settings of your
computers and so forth can actually make a big
There's a group called Climate Savers Computing Initiative a
number of computer companies are working on, which looks
like it can, basically by raising these efficiency
standards-- this is roughly like the car companies
actually welcoming raising their own efficiency
standards, which is a separate discussion.
If we could basically meet these standards, we think we
could take 11 million cars off the road.
That's how big the electricity use and power use of personal
computers and mainframes, and data centers and so
forth, really is.
Another example is building codes.
Most commercial buildings are built to be leased.
So again, the incentives aren't in alignment.
Now, you can fix this.
Other countries, Sweden being the classic example, require
when these buildings are built to have huge amounts of
Again, you get this enormous, enormous payback.
Fuel efficiency, everybody talks about it.
Why don't we just fix that?
Again, it's a call to action, a call to action to better
power, better performance, and of course,
product efficiency standards.
But the interesting thing is if you do all of that in
efficiency, you can then start talking about winds, solar
thermal, and geothermal.
And what's interesting about it is, well, the problem is
the wind doesn't blow all the time.

Geothermal, you have to dig really deep.
Solar thermal, the sun doesn't shine all the time.
Even in California, it rains occasionally.
This obviously doesn't add.
But if you do the math, and we actually mathematically model
the likelihoods and probabilities of when things
are on and off, they basically balance out, so that they're
roughly a permanently reliable sum.
And by the way, the shocking discovery is that they're more
reliable than power plants, because power plants break.
And the wind continues to blow, the sun
continues to shine--
thank goodness--
and of course, the earth continues to be very hot.
So in the wind case, we're very close on wind.
I think people here know how successful the wind
programs have been.
North Dakota is now known as the Saudi Arabia of wind.
That's what they're claiming about.
Texas is known as the Kuwait of wind.
I'm not sure why one took one, one took the other.
You can guess.
The important point is that wind is really becoming
And there are issues of getting, of course, wind power
into the grid, and so forth and so on.
But all of that is going to happen.
And it looks like it's happening well.
Solar thermal is interesting because you know that there's
a shortage of photovoltaic panels, and so
forth and so on.
Solar thermal is different.
Basically, you take the sun, you heat up a fluid.
So you're using, essentially, a mirror.
And by the way, it looks really, really neat.
It's like a space age, James Bondian type device.
And those things are now being built all over the United
States and all over Europe.
And it's particularly good in deserts, for
all the obvious reasons.
So there are more than a dozen of these plants being built.
And with the subsidies that are currently on the plan,
they get economically roughly close to where
they need to be.
And this is roughly a million homes, so these
are big, big numbers.
By the way, if you took a 100 mile by 100 mile piece of the
which, by the way, there's a lot of 100 by 100 mile pieces
of the Southwest that look pretty empty to me--
that single square could power all of the United States if
you didn't have transmission issues.
So there's a lot of solar, and there's a lot of wind.
And I forgot the underground.
Enhanced geothermal, if it works, go in, fracture the
rock, ship the water down or other
surfaces, cycle it through.
We understand this pretty well.
Good jobs in the drilling and oil industry.
It may be a major game changer, because, of course,
it's 24 hours a day.
We at Google have decided, with our renewable electricity
less than coal initiative-- and
remember, that's a formula.
We like formulas.
You can raise the price of coal, or you can lower the
price of renewable electricity--
we think we can do this.
We ourselves are investing.
We've invested more than $10 million dollars in wind and
wind power related activities.
We recently invested $10 plus million in
enhanced geothermal activity.
Governments have to have it, too.
This is where this failure of leadership and failure of
understanding what the opportunity is.
Obviously, increased federal R&D support for all of this,
basically targeting renewable portfolios by state
Some are getting it, some are not.
Some, about half the states, have portfolio standards
around renewables, and so forth and so on.
You've got to solve the transportation problem.

A rough analysis for you is that it's cheaper to operate a
car on electricity than gasoline.
If you look at the current hybrid situation, if you
simply take the battery and make the battery bigger, and
you put a normal plug-in, and then you plug it into your
wall at night-- and this is very important--
you'll get roughly double the miles per gallon of the car.
These are called plug-in hybrids.
These plug-in hybrids look like they can
make a material dent.
The most extreme calculation I found was one by Avery Lovins,
which roughly goes like this: if you move to hybrids, you
get a factor of 2/fuel use per mile--
which is the metric you want to care about.
If you move to plug in hybrids, you get another
factor of 2.
If you move to E85, which everybody knows about, is 85%
ethanol, you get another factor of 4.
If you drive it the way of optimal efficiency, you get
another factor of 2.
And that turns out that you've reduced the gas use in the
automobiles from 100% to 3% of what it was.
So even if he's off by a factor of 2 or 4, imagine what
that would do to the price of oil, given that that's sort of
the largest consumer now of petroleum products.
So we, of course, are doing a small fleet of hybrids to
prove this point.
There are now initiatives among many manufacturers that
are trying to prove this model.
And it looks like it's going to work.
If you put a price on carbon, this would
become a lot easier.
Of course, there's a big debate as to whether you
should have a carbon tax or a cap and trade.
You can imagine which one is going to have more political
momentum, the one that doesn't have the word tax in it.
But anything that actually causes the externality of the
price of carbon to be properly into these systems
would help a lot.
If we do that, and we do one more thing, which is we've got
to solve the grid problem.
And this grid that we depend on-- and we love our grid,
because without it, we wouldn't have any power--
is creaking for all the ways that you know.
There's about a 9% efficiency loss in current architecture.
It's very inflexible.
So it's very difficult to transmit things depending on
where the sources are.
So let's imagine a different structure.
Let's apply a different model.
And let's imagine, for example, that those cars that
were charged at night had batteries in them that, when
they could just--
the batteries are just sitting there, right?
Let's see if they could send that power back into the grid
during the day.
And you go, that's kind of a cheap trick.
Because there's efficiency loss.
You lose efficiency when you store the battery--
It turns out that math is very interesting, because most of
the cost of power generation is the peak power cost. Very,
very interesting.
It's that top 10%, when it's really hot, and everybody's
air conditioners are on.
That's when all of the costs are, because that's when, on
the margin, you have to go to spot fuel and
more expensive fuels.
Because obviously, when there's low demand, you'll use
your most efficient sources.
So all of a sudden, you discover that if you can
figure out a way to use batteries in cars, other ways
of shifting load off of peak, and also, reconfiguring the
grid so that people can make money, and doing it in a
material way--
in off peak hours, and in peak hours, in particular--
giving power back to utilities, all of a sudden, it
works really, really well.
Why does this not occur today?
The grid isn't really owned by anybody.
It isn't really managed right.
People don't really understand how to architect these things
at the regulator level.
But the technology's right there.
So when I look at this, I look at this sort of smart grid and
what is called the vehicle-to-grid--
and I imagine the smart garage, where I just sort of
plug my car in.
And the computers handle it, and the right thing happens.
And it always works, and it's always incredibly efficient.
And by the way, I make money by cost shifting.
So I love my car.
I especially love my car parked in my garage.
Sure sounds to me like a distributed computing problem.
Sure sounds to me like the internet.
And sure sounds to me like personal computers.
Now, speaking as a computer scientist, I'm sure that there
are a lot of complexities that we're talking about.
But what I see foremost before us is an opportunity that is
the largest opportunity that I could possibly imagine.
It's an opportunity that solves energy security--
which is a lot having to do with war, and war is bad--
energy price, an opportunity to create jobs and investment,
and an opportunity before us, oh, and by the way, solve the
biggest problem facing the planet aside from nuclear
proliferation, which is climate change.
With that, thank you very much.

M. R. RANGASWAMI: We can take a few questions from Eric.
So anybody want to ask a question?
M. R. RANGASWAMI: Go ahead.
AUDIENCE: I'm curious, you didn't really talk about the
transmission problem.

ERIC SCHMIDT: In the first place, there are people
working on better architectures.
There is a transmission problem.
And I had fun with my 100 by 100 square mile example.
But that's not how you'd do it.
The model of power generation is going to be distributed.
You're going to have a combination of a relatively
large number of gas turbines, which are quite midsized in
this technology, and then, a pretty good mixture of smaller
sources of generation, of which solar
thermal would be one.
But I think a reasonable assumption, since we're not
going to solve the grid problem in the next year or
two, we're going to be forced to having, perhaps, more of
these than we would like, that is, a
less efficient structure.
But there's still lots and lots of solar sources.
By the way, it's sunny very near southern California, it's
sunny very near northern California, it's sunny very
near most of the places, certainly, in the West. The
East Coast turns out to have a lot of enhanced geothermal and
a fair amount of wind.
So again, depending on the situation, one size does not
fit all, and you have to look at it.
There are enough people now doing California desert-based
solar solutions that do, in fact, pencil out, even with
the existing assumptions about the grid--
which is a problem--
that I think that, with a little bit more focus on the
grid, we can get that one solved.

AUDIENCE: A great friend, senior VP of the Rocky
Mountain Institute.
AUDIENCE: Thank you, thank you.
I'm curious, is all of this sounds as good as we'd like to
think, why isn't it happening more?
And we talked about politicians.
But in the business community and [INAUDIBLE].
ERIC SCHMIDT: Well, it's interesting.
The Rocky Mountain Institute wrote a report in 2004, which
talked a lot about this.
And so there been a lot of people in my research--
I'm new to this--
a lot of people have been talking about
this for a long time.
And I think, frankly, cheap oil made it easy to ignore the
good work and the good work on efficiency.
The question is, do you think the era of cheap
oil will come back?
And I'm assuming it's not.
I think it's a reasonable assumption.
And even if it did come back, it would be
bad for climate change.
So I think it's fundamentally because political systems and
democracies tend to work on the urgent
rather than the important.
I do think that it's a failure of leadership.
What I like about this opportunity is, if we can tie
the short-term problems we have, which are jobs, taxes,
all the usual problems, on high current energy prices,
maybe we can break through this.
Whereas having an intellectual conversation a few years ago,
you just couldn't get anybody's attention.

ERIC SCHMIDT: Well, in the first place, on Google
ourselves side, most of our energy use is, in fact, in our
data centers.
And Bill Weihl talked about it earlier today, so I did not
want to cover over his material.
We have a very, very sophisticated program about
really understanding the amount of power that we use.
And the answer on our data centers is very simple.
If we can get our computers to be more used,
we have fewer computers.
Therefore, we generate less power needs.
And therefore, everything else works.
So the simplest answer about computer efficiency turns out
to be keeping them at 100% rather than 20% or 30% or 40%.
One of the great tragedies about all these computers that
are sitting in you guys' offices running is, they're
not doing anything.
What are they doing, running Windows?
It's not that much computation required.
And there's not much going on, right?
So they're just sitting there.
And think about it.
If they were really heavily used, how much smaller and how
much less power they could use.
Now, on the measurement side, we've got a number of
initiatives where we're looking at end user ideas
about measurement.
But virtually all of the end user measurement things
require collaborations with a utility.
So the utilities have to decide to expose on a per
person, per household measurable data.
And then companies like Google and people here in the room
can write software that will basically take that stuff and
do lots of interesting stuff with it.
But you have to make the data available.
M. R. RANGASWAMI: Eric, a question over here.
ERIC SCHMIDT: Oh, I'm sorry.
Go ahead.
Yes, ma'am.
AMY: You all are doing a lot of R&D and onvestment
I was wondering if you could talk about the policy he's
given for the importance of decoupling from the other
policy initiatives that you talked about.

ERIC SCHMIDT: Well, there's a lot going on in policy.
And it's basically a lot of conversations with
politicians, trying to get them to understand these
Everybody would like to hear these messages, but they often
don't understand the subtle trade-offs.
The single best thing to do is to try to come up with state
or national standards around efficiency, and then remove
the barriers to renewables.
And obviously, the fact that the government cannot even
extend the renewables tax credit is another tragedy,
another failure of leadership.
M. R. RANGASWAMI: There's a question over here.
ERIC SCHMIDT: Well, because we care so much about it,
Google's been doing the math.
We're trying to figure out a way to get this kind of
calculation done.
What bothered me is I would go to these meetings, and I'd
say, OK, I'm tired of everyone grousing about it.
Show me a plan.
Even if it doesn't foot, show me a plan.
So I've produced a plan.
If you don't like my plan, produce your own plan.
What's your alternative, death?
So this plan is a Google plan based on the data that we
have. There are other people.
Al Gore, for example, announced something which was
sort of a similar approach with a different set of
shorter time frame, did not cover cars, covered
which is another example of a plan.
And there's a few other people working on them.
AUDIENCE: Did you come out with this publicly?
ERIC SCHMIDT: I just did today.

AUDIENCE: I have a hypothetical question.
Say next week, you wake up, and you find out--
ERIC SCHMIDT: Well, again, it's a hypothetical question.
It's always a mistake to answer hypothetical
questions on stage.

Well run businesses have people who can do
spreadsheets, and they can do these calculations correctly.
And I think most companies and most CEOs that I've talked
with are going to do these pretty much regardless of the
cost of oil.
And the reason is that, under any set of assumptions, the
various passive investments-- for example, investments in
reducing your use, efficiency uses--
pencil out under all assumptions.
And those are just good.
And should have everybody doing them.
And that's a big business, many, many billions of dollars
of building retrofits, business energy process
retrofits, those kinds of things.
How would consumers behave?
I think we all have seen this complex relationship between
oil price--
Hummers and that sort of stuff.
And that's a case where the government really has to
decide how much regulation they want to put.
But the fact of the matter is that there's a good
It's very difficult to raise the gas tax when
gas prices are high.
It's very easy to raise the gas tax when
gas prices are low.
So an obvious thing to do in that scenario would be to try
to increase the gas tax, and then as gas prices then go
back up-- because, of course, they're going back up, because
we're eventually running out, so prices will eventually go
to infinity--
is you lower that tax.
As to whether that's really going to happen in the US
political system, I'll let you decide.

AUDIENCE: So, you as a CEO, obviously, [INAUDIBLE]
ERIC SCHMIDT: Tell them it increases their earnings and
decreases their expenses.

Corporate America--
AUDIENCE: Do you have that experience when you're sitting
at dinner with them?
The problem is, people want to solve the problem.
And by the way, American corporations are accumulating
cash at a huge rate.
Anybody notice this?
The cash awards are getting bigger, for various reasons
involving accounting and the way these companies are run,
and the way that you could do cash and
acquisitions, and so forth.
So companies have cash.
I could understand if you said, well, I really want to
go invest in efficiency, but I can't get the cash.
And I'm not going to go to the debt market to borrow it,
because interest rates are too high.
Most companies, in fact, have very good capital lines.
So I disagree with that line of reasoning.
And I think that opportunity's before--
It's just a math problem.
And by the way, the numbers are big.
Because you look at it over a one or two year period, it's
pretty interesting.
You look at it over a five or ten year period, under any set
of assumptions, it's huge contribution.
So again, CEOs are supposed to look at the long term value of
shareholders, not the short term, long term.
This is one of the best things that you can do for the long
term value of shareholders.
M. R. RANGASWAMI: Eric, I had a question, which is, you
showed Google Earth and how the borders of countries were
easily identified by policies--
ERIC SCHMIDT: --Wasn't that fun?--
M. R. RANGASWAMI: How can you give that to the whole world,
so people in different countries--
ERIC SCHMIDT: I can just imagine you planting an M and
an R in trees.
You're such an ecological person.
M. R. RANGASWAMI: So you know what I'm saying?
How can the people in Mexico have access to that data, so
they can go their government and politicians--
ERIC SCHMIDT: It's called, and they can
download it.
M. R. RANGASWAMI: OK, but I think it would be
a phenomenal asset.
ERIC SCHMIDT: No, again, what we're trying to do with Google
Earth is we're trying to use this as a platform, obviously,
to communicate.
You can have fun with it and so forth.
But one of the most important things people are doing with
Google Earth is they're building overlays.
My favorite overlay is the overlay that shows the
flooding of the US capitol as sea levels rise.
So eventually, what happens is, the only building left is
the US capitol.
And all the other buildings, all those K Street buildings,
the White House, they're all gone.
And just because the water comes up out
of the tidal basin.
And if that doesn't scare you, you're not paying attention.
Yes, sir.
AUDIENCE: I had this question.
I think that--
ERIC SCHMIDT: We have. One of the issues is that these sort
of generic calculators don't really get interesting until
they have real data.
So we need to have data sources.
And I think one of our messages to people who have
the data is, make it available on the Web.
And companies like Google and specialized companies, and so
forth, get advantage of it.
Yes, go ahead.

If you use the part per million calculations, we're at
300 and some-odd parts per million right now.
Current projections have us, in various cases under
business as usual, going well to 450, 550.
550 was in one of the major reports.
450 was in another.
450 corresponds to five degrees Fahrenheit temperature
rise across a lot of the populated parts of the earth.
And that does include some sea level rise and a fair amount
of ecological destruction.
The problem that I have is that the modeling makes some
assumptions that I think are unlikely unless politicians
and our leadership act.
And so we're on a path to exceeding the 550.
And those are situations where we would essentially have a
repeat of the kind of global calamities that have occurred
every so many millions of years in the world.
It's worth noting that oil and oil shale originally came from
global warming some random number of 200 million years
ago, when the oceans became acidic, all the plants died.
They eventually went to the bottom of the ocean, and then,
over the period of that many years, turned into oil.
So we're in a situation where, if we don't act, we could
eventually cause that to occur.
Of course, the problem is, all the people would be dead
before that cycle was over.
They'd be replaced by something else, I guess.
It's a terrible, terrible image.
So we have to take it very, very seriously.
So the climate change issue is very, very serious.
And the things that I'm talking with
are part of the solution.
It will ultimately require not just a US action, which is
what I'm advocating, but also, all of the other actions that
people here know about.
M. R. RANGASWAMI: One last question.
Go ahead.
AUDIENCE: What are your thoughts about using wave
energy, as well as the renewable sources?
I see you had an announcement a couple days ago about your
data barges.
ERIC SCHMIDT: Yeah, the data barges is a clever idea.
I'm not aware of us building any at the moment, but at
Google, you never know.
And by the way, we patented the idea of a data center on a
barge in ocean, using wave power to
power the data center.
You never know.
So the studies on waves indicate that there's a lot of
wave power.
There's issues of citing, there's issues silting,
there's issues of the depth at which you put the paddles and
other structures.
There's a couple of initiatives where people are
trying these.
There have been a number of pretty successful experiments.
The reason i focused on the three that I did is that we
have proof points of them working pretty well.
So I would view the wave one as an example of maybe a
fourth or a fifth one that may come.
And I think we want to encourage all of that.

M. R. RANGASWAMI: One last one.
ERIC SCHMIDT: The lady over there had a question.

I have not publicly endorsed any candidate.
And I would love all of the candidates to stop focusing on
some of the short-term things and take this as an
opportunity to go and actually solve all these problems in
parallel, which I think be can be done.
Yes, ma'am.
AUDIENCE: Mine is not really a question.
It's more of a plea to you, as the chairman of a very public,
high-profile company, to use your position to give your
plan to the leadership, whatever it may be, and to
share it with other CEOs, because we really need that
conversation to happen in the C-level suite.
And I'm not assuming that you are not doing this already.
But I would really hope that you don't underestimate the
opportunity that you have to make huge steps--
ERIC SCHMIDT: Well, thank you.
Well, thank you very much.