Subscribe
Share
Share
Embed
Works in Progress is an online magazine devoted to new and underrated ideas about economic growth, scientific progress, and technology. Subscribe to listen to the Works in Progress podcast, plus Hard Drugs by Saloni Dattani and Jacob Trefethen.
I went to Sizewell B recently
Sizewell B is Britain's only
pressurized water nuclear reactor
and it's absolutely spectacular.
When you go into the turbine hall,
it's like a cathedral to energy.
It has two gigantic turbines, which are
like giant jet engines encased in steel,
but you can go up and touch the steel
and you can feel the turbine spinning
at 3000 revolutions per minute or 50
hertz, which I learned while I was there.
Very interesting.
I thought 50 revolutions per second.
But the thing is Sizewell B was finished
30 years ago and Britain hasn't completed
a new nuclear reactor since then.
The same story goes in lots of other
countries, which built often dozens
of nuclear reactors in the post-war
era, and then for some reason stopped
in the eighties and the nineties.
Now though, thanks to demand for data
centers, thanks to concerns about
decarbonization and concerns about
energy security, nuclear is back.
People are interested in whether
we can make nuclear work again,
somehow make it cheaper, somehow make
it faster, and restart the golden
age of nuclear power that we had.
So I'm joined today by my colleagues
Ben Southwood and Alex Chalmers, both
of whom have written quite a bit on
nuclear, to talk about why nuclear
was so promising to begin with, why
it became a flop, and whether it's
possible to bring nuclear back.
Then maybe you could just start
us off by talking about like,
what was the promise of nuclear?
Why did anybody care about
it in the first place?
So nuclear power is very
interesting for one reason.
Basically it's like fossil fuels,
but just much more energy dense.
So a lot of the cost, about half of the
cost at least, of producing power with
the coal power plant is digging up all
the coal, transporting it, burning
it, then getting rid of the enormous
amounts of waste it generates, and
completely put aside the fact that, you
know, it causes lung cancer and makes it
impossible to see in London the 1950s.
They didn't care about these
things that much at the time
when nuclear power was coming in.
But because you need to use 300,000
times less fuel if you could get hold
of it easily, which turned out we could,
if you can enrich it easily, which
turned out we could then it could be
not 300,000 times, cheaper, but you
could just get rid of the fuel cost.
The fuel cost is basically trivial.
It's like 2% of the cost of
running a nuclear power plant.
So we just got rid of that fuel cost.
So what we should have is just a half
the cost of a coal power plant, coal
power plant, because you've got the same
turbines, you've got the same boiling, you
are instead of burning, coal to boil it,
you're creating nuclear reaction by having
particular metals in a particular place.
But essentially you are boiling
water that turns a turbine.
So it's the same thing.
You should have stuff that's just half the
cost of a nuclear, of a coal power plant.
That's why the American utilities
loved it when in the 1950s when
they started building them.
That's why the British government,
through the central Electricity generating
board was interested in the 1950s.
And it also came out to be true.
So, they didn't work it out straight
away, but over the first 10 years of
nuclear power, it got cheaper, in
both Britain and the US and they were
building nuclear power plants for, in
modern money about $500 per kilowatt
of resultant, of capacity that it has.
Which, what they call overnight
costs, technical jargon.
But if you put a recent nuclear
power plant in that, those terms, I
dunno, I dunno how to pronounce
it, Vogtle (/ˈvoʊɡəl/ VOH-gəl).
In Georgia then I think it's about $16,000
per kilowatt in modern money versus $500.
So about 30 times more expensive.
So, and by the way, they were building
these about four years instead of 27 years
as Vogtle took to finally get plugged in.
So.
You can see why people
are excited about it.
Right?
Like there, there's a new form of
electricity that's gonna make it so
abundant that we're gonna be able to have
electricity that's like half the price.
And by the way, in the 1960s, the
US Army Corps of Engineers, I
think was making this calculation,
but I could be wrong there.
They judge that in modern money, the
electricity was 3 cents per kilowatt hour.
So let's say like 2.5 p And I can
imagine that's to the end consumer.
Yeah, I can imagine, Sam,
that your electricity does not
cost 2.5 p per kilowatt hour.
No, mine is like 21 pence.
23 pence per kilowatt hour.
There we are 10 times worse at
producing electricity than, than
they did in the 1950s and 1960s.
So that's the promise.
It started off as potentially the
cheapest way that, the key thing is that
it was the cheapest way to present it.
They didn't care about,
about the fact it was greed.
No-one cared about carbon then.
Yeah, they did care a bit
about, local pollution.
But not that much.
They maybe acid rain when it finally
came, became obvious, but that
just wasn't what was on their mind.
And everything went perfectly.
And now we have incredibly cheap energy,
but it's too cheap to meter, right, Alex?
Pretty much, I don't actually
know why we're sat here having
a conversation about it, so, no.
Yes so I think...
So What went wrong?
What went wrong?
Yeah, so I think Ben, gets
exactly right when he talks about
why people were enthusiastic.
And that narrative held through mid to
late 1950s through most of the 1960s.
But the exact point begins to go wrong,
varies by country, but a few sort
of common patterns begin to emerge.
So one costs start to rise from
the late 1960s, early 1970s.
And this happens for a
couple of different reasons.
One, I think in both the UK and
the US, the industry essentially
over promises and under delivers.
It begins moving to bigger and bigger and
more and more complicated reactor designs.
It assumes that costs will scale
linearly, companies underbid to
win contracts in commercially
competitive situations.
And this essentially doesn't play out.
And this begins to erode popular and
bureaucratic trusts in the industry.
In what way?
So, what do you mean it doesn't play out?
And how does that erode trust?
So, for example, if you look at the,
the US it's the like, essentially.
Companies promise regional utilities
that we will build a reactor in
this time to this budget and they go
over time and they go over budget.
In the UK it goes much
more spectacularly wrong.
So the UK originally had this
design called the Magnox for the
first wave of its nuclear program.
It built 26 of these in about 15
years, which is pretty spectacular.
And then in the mid 1960s, it decides
to move to a larger and more efficient
design, known as the advanced gas reactor.
And this unleashes Dungeness B, which is
probably the worst infrastructure project.
In post-War Britain, where
this, where a company that is,
That's a very competitive field,
Which is a very competitive field, and
it's a really weird story so that they're
tendering for this, nuclear power station.
And Britain had the system because there
was no one company that was capable
of building all the components and
all the bits and employing enough
people to build a power station.
So it would tender these out
to these competing consortia.
And the bid to build Dungeness B was
won by a consortium that was actually
on the brink of going bankrupt that
had won it by insanely underbidding
on time and cost to build essentially
a first of a kind reactor design.
They claim they could build
this thing in four years.
No one had built one at full scale before.
The company goes bust a few years
in, huge amounts of the engineering
work end up having to be redone.
Nuclear inspectors refused
to sign off on the reactor.
It finally, I think having started
in the mid sixties gets switched
on in the 1980s and I think is
running about 12% of the time.
So I mean, these kind of white
elephant projects begin to
damage confidence in the industry.
But there's a sort of
parallel story playing out.
So you have a degree of hubris and
bluster on the engineering front, but
at the same time, you have a changing,
popular and regulatory context.
So one of the big changes you see
in the 1960s and the 1970s across
much of the west is the growth
of the environmental movement.
So, for example, in 1971, the US nuclear
regulator loses a landmark Supreme
Court case known as the Calvert Cliffs'
decision, which means they start having to
do proper environmental assessments before
they can build nuclear power stations.
In the 1970s, one of the big drivers
of the early environmental movement
is a slightly melodramatic backlash
about nuclear waste and the industry
because it believes that nuclear
waste is a sort of a fake problem.
Like you can just put it in some
concrete and forget about it.
They don't take this very seriously
and suddenly find themselves
on the receiving end of a lot
of popular hurt and popular upset.
And then finally, you begin to see
a changing regulatory context for
licensing decisions and around nuclear
safety, which becomes progressively
more conservative as the 1970s press on.
Some of this is driven by
evolutions in the science.
Over the course of the fifties and
sixties, scientists become more and
more concerned about the effects that
radiation can have on the body, this
is embodied in something called the
Linear No-Threshold (LNT) hypothesis,
which states that damage and harm from
risk from radiation is both cumulative,
so it builds up in the body over time.
The body doesn't heal from it,
and there is no safe threshold.
This is as opposed to a model that
says that your body experiences small
stresses, but it repairs itself.
Yes.
Like, you know, the difference
between drinking 365 pints
over a year or in a night.
You know, clearly your body does do
some recovery or some processing of
the alcohol when it comes to alcohol.
The question in nuclear
is, does the body do that?
Yeah.
It's exactly that.
So in, in their defense in the 1950s,
we didn't really know as much about
things like DNA and how the body heals.
And most of the evidence we, and our
kind of theories about this developed
over the course of, but we basically
knew most of this, including on some
of the kind of more extreme, what are
known as double strand breaks in DNA.
We basically knew this by the 1980s.
We didn't know this in the 1950s,
but the model we used to regulate
radiation has not been updated in line
with these scientific developments.
In fact, we probably got this right
in the thirties, so our operating
assumption like historically was there
was a safe level of radiation and you
could be hit with that about a hundred
millisieverts/s, you'd probably be fine.
Then above that, the risk starts to
build up, which was our old assumption.
We sort of junks this in the fifties
and sixties and now it's quite funny.
I talk to people who work on this
and they're basically saying, if you
look at our first, international set
of radiation guidelines from 1934, it
was actually probably closer to the
truth than the model we used today.
It's a rare case of us going backwards.
So my question, or I'm curious
about the environmentalist movement.
Yeah.
How much are they worried
about nuclear waste?
How much are they worried about radiation?
Or like, are there other things?
What is it exactly in the
1960s, for example, that the
environmentalist movement is
worried about with nuclear power?
So it's an interesting one and
it's a case where I think a bunch
of relatively weak arguments get
lumped together, which sort of bulk
them out and make them stronger.
So some of it was essentially concern or
panic about the prospect of nuclear war.
Mm.
And this was sort of an
anti-imperialist, anti NATO thing.
And this kind of spilled over
into civilian nuclear power.
Because to be fair the
reactors were being used to
produce plutonium in some cases.
Right?
By the 1960s, everyone
had worked out chemicals, so
no one was really using them.
It turns out that we were doing that.
Because we're using the
old kind of reactor.
The original wave of nuclear reactors
built in Britain anyway, were done...
Yes.
The US nuclear power plants by and
large were not being used to enrich
plutonium to make bombs from.
And we didn't use most of
the Magnoxes to do that.
So the Magnox design emerged
from the nuclear program, but I
think after Calder Hall, the
ones we built were single use.
We had way too much plutonium by
like now it's still there in big pile.
Yeah.
One of the things that you highlighted
that I'd never heard about before
is, some of the nuclear bomb tests.
So firstly, the organization running
regulating nuclear power was in fact
in the us the same organization that
was doing bomb tests and stuff.
So the associating, the-
Atomic Energy Commission
Yes, associating them
together was reasonable.
And secondly, some of the bomb tests
they were doing were pretty risky.
Yeah.
There were a lot of mysterious sheep
deaths in Colorado in the 1950s, and there
was a whole panic about like strontium
ending up in milk in the late fifties.
But I think the most notorious
one they did was Castle Bravo in
1954, which was done in the Bikini
Atoll, near the Marshall Islands.
And the wind speed changed unexpectedly.
The bomb generated much, much
more fallout than expected.
And the US essentially rendered an
entire island perpetually uninhabitable.
So this island called like Rongelap
where they had to evacuate the entire
population because ash rained down
on it and children went out to go and
play with it, thinking it was snow.
Wow.
And then got quite ill.
And there were elevated cancer
levels in the population perpetually.
And some of the fallout hit
a Japanese fishing ship, the
ironically named Lucky Dragon.
And the crew found like
pus leaking from their eyes.
And by the time people had
realized what had happened, their
fish was being sold in Japan.
And this triggered like
an - Understandably.
Also, considering what happened in
Japan a decade earlier, triggered
a wave of like national panic
and fish being bulk destroyed.
Mm-hmm.
So yeah, some of these tests did go
wrong in, in a quite like spectacular way.
So nuclear has for maybe
understandable reasons, like
quite bad vibes at this point.
Are people are environmentalists, like
at what point do they start to worry
about meltdowns and things like that.
And actually if either of you could
talk me through or talk us through
what a actually is a meltdown or like
what's happening when a meltdown is
happening, that would be quite helpful.
I will do environmentalism
and Ben has strong opinions
on how real meltdowns are.
So we'll leave that one to him.
I think Alex is better on both.
But so on the fear of meltdowns, I
think like nuclear safety as a concern
I think really became heightened in
the 1970s, partly because you had a
series of like smaller accidents.
So for example, there was, I think
one at Brown Ferry in the United
States, and this is also when the waste
conversation begins to get going, like
one of the unfortunate instance you get
is like for the UK nuclear industry
is Greenpeace in one of its first ever
bits of activism, essentially catches
UK ships dumping nuclear waste at sea,
which as you can imagine, goes down
really, really well with the public
and it actually caused such an outcry
over it that sort of transport work
workers like threatens, like boycott the
industry if they don't stop doing it.
And the other one with the UK in the
mid seventies is, I think it's 1975,
although someone will write in to
correct me that the UK plans to build
a reprocessing facility at Sellafield.
So countries can ship their
nuclear waste to the uk, get it
reprocessed, and then they can
theoretically use it against fuel.
And there's this daily mirror headline
that says Secret Planned Turn
Britain into World's nuclear dustbin.
Which again, then leads to huge protests.
I think Friends of the Earth chartered
a train from London to Sellafield, and
they held a debate with representatives
from the site on a football
pitch next to the power station.
A) I didn't know you could actually
charter trains to do that, so this
actually begins to generate like
quite a lot of public concern and the
industry thinks it's a fake issue.
They're like, "Well, we'll
be able to use this to like
fuel advanced future reactors.
It's, basically safe in tanks."
Mm.
And then plans to build Nuclear fuel
disposal facilities get protested in
every community they're proposed in.
Sellafield.
Possibly related to what
you're talking about.
When I was growing up, Sellafield
was like absolutely one of
the top stories in Ireland.
Because it's on the west coast, on the
north, you know, up towards Scotland.
At the time the concern was that
there would be some kind of accident
and the wind would blow it
Yes.
To Ireland rather than, so it
would be it basically, it was
a sort of a moral hazard problem.
It's much like this concern you got
after like Chernobyl in 86' of, "Oh,
the wind will blow the radiation line."
Hence the name of the-
There was still like fake
German evidence, so showing that
Chernobyl led to effect in Europe
And he hence for name of the film when
the wind blows, which was very scary.
And very good.
Great film.
Yeah.
And speaking of trouble, that
was, I think the last pillar
really helped the environmental movement
was the meltdown at Three Mile Island in
1979 in Pennsylvania, though that unit
is now being apparently revived, which
it didn't actually harm anyone based
on like the best scientific evidence
that we have now, but it really
fueled the anti-nuclear movement.
They had to evacuate the area, did they?
Or..
I mean, it turns out that they would've
been fine not to evacuate the area.
They did evacuate the area.
But there was also a massive
degree of coverup, like, right?
This seems, from what Alex has told me, it
seems like this is a pretty much standard
- these guys are covering everything up.
And then hope and in the short
run oh, great, no one's been
polarized more against nuclear.
But in the long run, everyone seems
to come, it seems to come around in...
Yeah, I think this is exactly right
and I think part of this is about the
kind of culture of the nuclear industry
in the early nuclear industry is that
lots of the people involved in it
had previously worked in like wartime
nuclear programs, where the idea that
public opinion or public concern was a
factor that you had to take seriously,
just like didn't really occur to them.
So I think this is a consistent theme
that, you know, you see through the
sixties and seventies is the kind of
idea that the public put their trust in
the enlightened engineer or enlightened
expert or enlightened technocrat just
sort of dies a little bit of a death,
but the technocrats don't really
realize that it's happening to them.
So let's get onto ALARA.
Because a lot of what you've talked
about, I think reasonably focuses is
quite like country specific, but the story
that we're telling about nuclear is
worldwide or at least kind of a cross
the west, like most western countries
with one or two really interesting
exceptions, which we'll get onto.
Most western countries, experience
the same kind of decline in Yeah.
Support for nuclear power and kinda
rising concern about nuclear power.
So why is that?
So I think to be fair, I don't think
there's actually that much of a
decline in support for nuclear power.
In terms of if you poll people, the
decline comes quite a lot later.
Mm-hmm.
I think the decline comes after nuclear
power becomes slow and expensive.
And they're like, "Oh, it's a bad source
of power." I don't actually think
that the disasters cause a massive
fall off in popular opinion for it.
And most of the costs come
in before the disasters.
Which is in fact a lot of the costs
come in before ALARA comes in, right like
a significant that comes in afterwards.
So let's talk about Ara though.
Okay.
Explain what that, what
does that stand for?
What is that?
Yeah, so ALARA stands for 'as low as
reasonably achievable', and it's a risk
management philosophy that the nuclear
industry begins to embrace in the 1970s.
It's called ALARP or 'as low as
reasonably practicable' in the UK.
But it essentially amounts to
a version of the same thing.
And this is the idea, that risks
should be sort of pushed down to, a
level beyond which pushing them down
becomes, very disproportionate.
I think, that is obviously I agree
with that, but I think another way of
putting it that it's quite useful is.
If you can afford IE if there
are any profits left over after,
currently in the project plan.
Yeah.
And then you can afford a mitigation
that will reduce nuclear- will reduce
emissions or the risk of emission, of
radiation emissions it to any degree,
then you're obliged to do it so long
as there is any profit left over
and you can see why bringing this in
and then having two oil crises led
to, a massive explosion of cost.
Right.
So like.
Just at the time where wholesale
prices are going up massively.
And so therefore profits
for nuclear power plants in
principle are going up massively.
They're like, okay, just
do all these things.
And the nuclear operators end up saying,
well, okay, we can take it, we can take
it on the chin because prices are so high.
But then obviously afterwards when
the regulations have already kicked
in, you can't ratchet them back down.
But the key thing I think is it makes
profits illegal or like it makes
anything above the minimum profit
for it to go ahead illegal, and that
means that prices, any potential - you
invent any new mitigation, like
come up with a new kind of filter?
Yes.
You have to implement it.
Yes.
It means, it essentially means that
you get punished for good behavior.
So any innovation that anyone
else makes in safety suddenly
becomes like the new floor.
Or well, they have
an extra backup system.
Why don't you have an
extra backup system?
And you actually see this in
how regulation gets rewritten.
So the Office for Nuclear regulation,
the UK's nuclear regulator issues,
you know, these guidelines and one
of them is something called the
baseline safety objective, which sort
of, I won't go into the full UK risk
management philosophy, but it's once you
-
This is today we're talking about
Yes.
In the present day.
This in the present day, and it's,
if you get like below this level
of risk, you are probably okay,
although not guaranteed to be.
And they actually said that on radiation
they've lowered some of the baseline
safety objectives further, not because
of any change in their risk assessment or
the science, like it explicitly says this
in their safety assessment principles.
It's, we just think that the
industry can do this easily now.
So we've made it, we essentially
just moved the baseline
.
And I do want somebody to explain
what a meltdown is, because the term
meltdown is, everybody knows what the
term is, but I think almost nobody
can actually describe to you what is
happening when a meltdown is happening.
So explain that, but also as
part of that, what is it that
ALARA is actually trying to reduce?
Is it background radiation?
Is it meltdown risk?
Is it something else?
Is it everything?
So in the UK - this happens in every
different country, but I've looked
at the UK things a lot of the time.
It happens differently in every country.
Yeah.
It's basically, it's
similar in every country.
They have like 14 different
things they're trying to avoid.
And their risks are all different parts
of the chain, so background radiation
is not - they do have rules for
background radiation, but it's so low.
It's just irrelevant, in practice.
And to be clear, that doesn't mean
they've never done stupid stuff to
avoid trivial amounts, nugatory levels
of background radiation, but there are
also questions of like the percentage
chance of a bad outcome happening and
multiple bad outcomes at the same time.
So the EPR 1, has two
containment domes, right?
Yeah.
And the outer one can take a direct
hit from a plane flying into it,
but like at the same time that all
of their other systems are failing.
So it's more and more rare
potential circumstance.
They're not impossible, but they're
extremely unlikely and probably
pointless in many different cases.
Yeah.
So we're talking about is it like
10 to the minus six likelihood?
Events is sort of like over
-
That would be one in, that would
be like one in a hundred thousand.
No, there's like one in like a million
or 10 million reactor operating
years are the kind of level of
risks that we're trying to mitigate.
But in practicewhen you're trying
to evaluate that level of risk, it's
so sensitive to your inputs that it
becomes a little bit of a nonsense.
Which is why nuclear companies often don't
actually try and challenge the kind of
cost benefit analysis because it all feels
like a little bit fake and they tend to
just sort of suck up what the regulator
asks them to do out of convenience.
And let's be clear.
So let's assume, so Alex and I, we
talked about this before, so I know
what Alex thinks about all this stuff.
Alex and I have a view that the
standard way they measure the health
impacts of radiation is incorrect.
But let's just take for a second
the assumption that it's true.
Even if that's true, the baseline
standards before we kept escalating
them were pretty damn high relative
to all other types of power, right?
Maybe that's different now
that we have particularly safe
ones like solar, Photovoltaics
(PV) solar is very, very safe.
Like occasionally people die when
they're in the factory when they're
making it, or fall off their
roof when they're installing it.
But it's extremely safe.
So maybe there's a safer form of power
now, but compared to basically all of the
other ones, nuclear power is an extreme-
Even if you assume we're gonna
get a Chernobyl every 40 years or
something like that, which by the
way, there's no need for us to have a
Chernobyl every 40 years because their
management was absolutely disastrous.
Nothing like that has
ever happened in the West.
But let's say even if we're assuming
at Fukushima, every 40 years or
something, Fukushima, I mean, according
to the UN assessmen didn't harm
anyone through radiological effects.
So, so-
So talk me through what's going- so if
something goes wrong What, what happens?
So do you wanna explain a meltdown?
No, I think you do it better.
The basic thing that's happening
here, right, is that you have rods.
I mean, it might be pebbles in a kind
of reactor, there are all kinds of
crazy reactor designs, but mostly rods.
You have rods, there are moderators with
them that are slowing down the reaction.
You lift them up and down to speed it up
or slow it down based on trying to hit
what you're - keep it at a steady level.
If your coolant, which is usually
light water, sometimes what's
called heavy water, which we don't
need to go into that, and sometimes
a gas, like in the UK's power plants,
we were using gas cooled reactors.
If that escapes out or something so
like that, that's not being held in
there or it can't get through the
system where it gets cooled down,
then it might start overheating.
Or you get a loss of coolant incident
'cause it's broken into or something.
And this is, you've had like three
big reactor meltdowns over time.
Or if like in, Chernobyl you do
like a crazy test saying "Why not?
Let's do an unregistered test of
what would happen if we made all the
parameters to their extreme levels
and would something go wrong?"
You get one of these things then it keeps
heating and the metal has a melting point.
And then it melts.
And it's very hard to control
after that point because it's like
extremely hot metal just melting
through everything around it.
And then it starts escaping out.
And now we can probably just tank
that because we've got good enough
containment domes, so the act
actually a meltdown happening without
something else really bad happening,
probably won't cause any harm.
In good western modern reactors.
But if you were doing it in the first
ones we had, which didn't even have a
containment domethen that could lead to
the UK's worst nuclear disaster in 1957.
Something like this is happening.
It's not quite that actually, because
there isn't a full meltdown, but there's
a gas plume and like iodine, radioactive
iodine is getting into the gas plume and
then spreading all around like a huge
area, and then if it could get, if it
gets into the milk, then you don't want
drink the milk 'cause it concentrates
in the thyroid and that causes a much
higher dose -all those sorts of effects.
So meltdown is basically just
the core, which is made of
metal melting, gets really hot.
And because it's a nuclear reaction
that's melting it, you can't cool it.
Like, it gets past a certain
- There's nothing you can do.
What are you gonna like, I dunno, get
something with really high specific
heat capacity, force it through
it really - basically you have to
wait until it just goes for ages.
And that's like in the cases we've had,
they've just gone for a really long time
Because so many people think of a
nuclear meltdown as a nuclear bomb going.
And they imagine there'd be a huge
explosion, and obviously you don't
want that and you wanna keep those
very far from the population.
But really what's happening is you
have a release, an uncontrolled
release of radioactive material
into the surrounding area.
Like really, it isn't the melting,
per se, the melting is bad because
it breaches the containment area and
it releases radioactive material.
Is that right?
Yes.
Although I would hope that if we had a
meltdown in one of our modern ones, they
wouldn't breach the containment area.
But maybe it would.
I think my understanding is that
if you stand on the boundary of a
UK nuclear site and there's a full
meltdown, you shouldn't get this
bigger than 10 millisieverts which is
-
-which is one full body CT scan.
So it's not a big deal.
Like you can have one of
those and it seems to have
no effect on people's health.
Okay.
So bring us up to present day.
So we've had, we've gotten up
to around Three Mile Island.
And we've gotten up to ALARA.
So here's the interesting thing, right?
We're still building nuclear
reactors at this point, right?
Ish.
Ish.
So in the US they're still ordering
loads of nuclear power plants 'cause
they're thinking like gas prices,
oil prices are gonna be high forever.
We're heading towards
peak US oil at this point.
You know, they don't know what's
gonna happen in the future.
We need something, even if it's
really expensive, they don't
know about renewables getting
good in the future either.
So they're still ordering power plants
then as things cool off and it doesn't
turn out to be as bad as they thought
with importing fossil fuels, and
these plants keep taking ten, five
years longer than they expected.
And coming in more expensive, a
lot of them get canceled, but they
do finish off a bunch of them.
Yeah.
So I think 120 get canceled
after Three Mile Island.
So like you do see like a spike on
like the charts of international
orders in the US you see in 1979,
they go off a bit of a cliff.
The UK is actually already
failing by this point.
It's failing for the whole seventies.
Actually, it is gone and
France hasn't really started.
That's the interesting thing.
We'll get on to France in a minute.
France is, France is so interesting.
Yeah, just on the UK I think it's
an interesting one because 1979
Thatcher comes in and she's really
into nuclear power and she wants
to build like a load more of it.
So the government's kind of mulling
building maybe as many as like,
and in the UK by this point
under Thatcher decides we're gonna
junk these gas cooled reactors.
We don't have any of the kind
of like residual weirdness about
buying American technology that
our kind of predecessors had.
So they're gonna build, they're
gonna finally switch from like
the bad expensive design to the
cheap, efficient international one.
And they, it begins to all kind
of go off the rails in the 1980s.
So a couple of things happen.
So the eighties is when they decide
they're going to build Sizewell
B, which, you mentioned that you
visited, but kind of cool public fears
about building a pressurized water
reactor just after one has like melted
down in America, they decide to hold
this full public inquiry and the
Thatcher government are like, "Great.
This is gonna be our opportunity to just
defeat all of our foes and all of the
environmentalists and all of the haters.
We are gonna humiliate them with the
sheer strength of our case." They
allow the inquiry to be defined along
these like really, really broad terms.
And so the technocrats, right, the central
electricity generating board -to my point
about being, you know, technocratic,
complacency don't prepare for the inquiry
very well when it gets going in public.
It turns out they haven't completed
reams of the safety documentation.
They have to kind of repeatedly file
all these addenda, the kind of cost
data they provide on how affordable
nuclear is, is kind of nonsense.
And the chair says he doesn't believe it.
And because the inquiry is really
loosely defined, it spirals off
into discussions about the ethics
of like importing uranium from
like Namibia, which is occupied by
apartheid South Africa at this time.
Aborigine land rights and the impact
of the uranium mining industry on them.
Other countries react to designs.
The whole thing goes off the rails
and takes, takes over a year.
And the government just doesn't-
It took over a year to
plan a nuclear power plant?
How could they survive this?
Sadly it was at that point the longest
ever planning inquiry in like UK history,
then to be eclipsed by Heathrow Terminal
five, you know, not that long later.
And so by the time they get round to
building it like that has dented faith.
And then when they privatize energy
in the UK in the late 1980s, investors
begin like picking through the cost data
and they realize that the enlightened
technocrats hadn't been entirely honest
about how they were cross-subsidizing
nuclear from like coal and oil revenue.
And the government has to pull
nuclear out of the privatization at
the last minute, which is this huge
embarrassment for the industry.
And then plans to then the
rest of the PWR program is
effectively shelved at that point.
And the US is just finishing the reactors
it ordered, but most of them not.
Finishing the reactors
ordered but no more.
Just like it turns out that
nuclear is very expensive.
Everyone now discovers.
It turns out that it was a bad idea.
It turns out that like, we're
just not gonna do this anymore.
And then, I think it is the right time
to pivot onto talking about France.
Because this is just the time when
they decide, you know what, after
this experience of- Because at
the time they were importing like
all of the oil they used, right?
They didn't discover North Sea
oil like we did at the time.
They didn't have a massive supply of
their own like the US, and they thought
we just cannot be in this situation again.
It's so interesting, the exact
opposite; Basically the US and the
UK, they have high electricity prices.
They kind of think, well, we've got a
lot of slack in the system, so we're
going to regulate this to make it safer.
We'll spend some of
those profits on safety.
France does the exact opposite thing
and so, take up the story, yeah.
So yeah, with France, France is an
interesting one because initially
they start dabbling in civilian
nuclear power in the sixties and it's
largely a car crash; they have these
expensive like domestic French designs
that use natural uranium that are
incredibly inefficient and there were
technical problems with all of them.
In the early 1970s they then switched
to the light water reactor, and yet
after the 1973 oil crash, they go.
Yeah, we can't handle this, we
need energy security and the
French program goes really well.
I mean, over the course of the seventies
and eighties they build over 50 reactors.
And these are big reactors and these are
like double the size of the, or five
times the size of the smallest UK...
Yeah they are like quite chunky for sure.
And I think the French have like a few
different things going in their favor
is 1) French Fifth Republic, thanks to
Charles Degal, peace be upon him has
like a very centralized political system.
So essentially all decisions about
nuclear regulation and licensing,
are like taken away from Parliament.
The public essentially has no say,
you don't get the kind of hearings
and inquiries you get in the UK
in the seventies and eighties.
So they, that's
I wanna push on that because-
-I, I'm, don't worry-
-because in the UK we had that
system in the 1960s, right?
Like we had that system, we took it away
from ourselves in response to the public.
So the interesting bit is what you're
about to say next, I think, which is
how they sustain that system through
a massive build out, despite the fact
there were already, we had Three
Mile Island in 1979, and then they
had Chernobyl, which is an actually
really bad nuclear disaster in 1986.
And they just keep going
through all of that.
Yeah, they-
-and don't get too expensive
and don't get slowed down.
They slowed, they're
slowing down by Chernobyl.
But, so I think France for one is that
it's centralized, but it also means
that nuclear licensing decisions are
taken in like relative secrecy -until
1979 France doesn't actually have a
set of codified nuclear safety rules.
And there is like no obligation for EDF
France, a state energy company to run any
kind of environmental impact assessment.
They just pick sites and build reactors
on them that are deemed appropriate.
But there isn't really much, there is
a bit of a public backlash in a couple
of places, but not a sustained one.
And the reason is essentially
how French taxes work.
So unlike if you build a nuclear
power station in the UK, any sort of
revenues that you might get from it
through business taxes, immediately
go back to central government
to distribute as it wishes.
In France, these are retained locally.
So if you have a nuclear power station
built in your area, your town council
becomes like absolutely minted.
It's crazy how locally
this is in France as well.
So France has three main
types of local government.
They have the commune, the
department, which is roughly
parallel to a UK district council.
There's nothing really that in the
US - maybe a county in the US- and
then the region and the commune
could be, I mean, there are some
communes with zero people in them, or
occasionally one or two people in them.
Or it can be Paris inside the periphery of
2 million people, and they range in - But
the point is they can be like 1000 people.
So Chooz which is a funny nuclear power
plant 'cause it's on, what do you call a
long bit of a country going into another
country without being like a salient.
If it was in the first World
war, battle line, it'd be called a
salient, it's a salient into Belgium.
They built Chooz on the very
end of the salient into Belgium.
So it's basically in the middle of
Belgium, just happens to be connected.
But there are a thousand people
in the commune, including Chooz.
So they've basically not paid
any property taxes for 40 years
and it's extremely popular there.
They get free satellite television paid
for by the local authorities as well.
My first knowledge or understanding of
this came from a conversation with one of
our colleagues at Stripe who was a French,
- he's French and I was talking to him about
this EDF build out and he was like, "Oh
yeah, the playground I used to play in
when I was a kid was an EDF playground."
Yeah.
They built that playground or you
know, I don't know if they physically
built it, but presumably there was
a plaque somewhere saying, you know,
paid for by the good people at EDF.
So it's funny how much of French
society maybe just suffused with this
background support, background knowledge.
So my little theory, and I'm not sure if
this is true, but this is the tidbits of
a starting point of a theory is like, why
didn't the French environmental movement
take off and stop them from doing this?
'cause like the UK started
in the same position.
The US it was decentralized, but
it started with the ability to
mark its own homework to basically
just put power plants out, ask
no one, but there were political
competitors who came in, took it away.
And I think one possible reason is that
if you make it so that the people who
would be most likely to give support to
those, um, moves for maybe real reasons or
maybe pre pretextual reasons to use a word
that we like to use in Works In Progress
pod, but if there are people who might
find themselves being thinking, yeah, I
do really care about environmentalism.
I really care about environmentalism.
The thing I care about
most is environmentalism.
Which by the way will stop this hulking
great concrete mass that maybe puts
out invisible emissions that kills me.
If environmentalism will stop that,
turns out like I have a really strong
gut feeling that environmentalism is
amazing and I believe in environmentalism.
If you don't have an incentive to believe
that, possibly, that's the reason why
it doesn't get off the ground in France.
And so they don't ever, impose any of
these constraints on themselves, which
like the public could vote for that.
Like, they wouldn't have had to
do a revolution to to impose that.
And actually Mitterrand was the big
nuclear hater in French politics
because I think he essentially
just opposed it because Degal was
into it as far as I could tell.
I could never actually work
out a principled reason why
he didn't want nuclear power.
And he originally, I think in the
seventies campaigned against any further
kind of nuclear build even then he ended
up having to like moderate this position
because it wasn't like supremely popular.
I think he was kind of assuming there
was a big environmental constituency for
this and found that there wasn't one.
Do you know much about how EDF and
some of our listeners will be very
frustrated that we haven't talked about
the fact that this was all state led?
So I think we should talk about that.
And I'm curious about how
important that was in your view.
There are five great, really
great nuclear buildouts, right?
Um, South Korea, France,
Britain, the US and
...
China?
Japan.
We can count China, but I mean in
developed western countries and three
of them were run by the state, two
of them are private, so Japan and
the US are basically all private.
And it's like regulated utilities buying
from private companies, producing them.
And then in the UK, France and
South Korea, South Korea may
even build them themselves.
I dunno if you know that, I dunno if
you know the answer to that question,
but in the UK and France, it's like
EDF contracting from, there's a
state utility contracting from them.
I mean clearly there's something going
on here that, well, I dunno, clearly.
It's definitely interesting
that it was successful multiple
times with a state led build out.
Doesn't seem to have been
the only way you could do it.
And it seems more like that
having the regulatory enviro- like
EDF has got, had more and more
trouble over time building out.
The UK continued to be a state
led thing while it failed.
It wasn't taking away the state that
led to the failure of UK nuclear.
We took away the state in what,
1998 or whatever it was, or 1990.
1990. I think you, maybe it is, but the
central electricity generating board was
procuring the nuclear power plant between
1952, whenever it started, and 1990.
And most of the failure happened
during that period, not after 1990.
Yes I don't really buy that if
Hinkley Point C was being built
by the British nuclear company
instead of EDF, that essentially.
I don't think, I, you know, theoretically
like, but EDF was then paying for
Hinkley Point C off its entire balance
sheet anyway, so you don't really have,
like, the whole cost of capital thing.
EDF is building nuclear power
plants much more cheaply in
France than it is in the UK.
So you would say it's a test case of
the same state run company what the
difference a regulatory regime might
make between France and Britain.
Well, but also, I mean, different,
um, purchasing regime, right?
Or like, doesn't that kind of support
the idea that maybe if we had a domestic
EDF it would do something as well
as France's domestic EDF does it.
I'm open to this idea.
I think that the other factors like
the fact that we haven't tried to build
a nuclear power plant in 30 years...
Let me put the case forward then, so
what's really, the thing that really
seems likely to me is that scale and
certainty and doing like lots of nuclear
reactors, like having a pipeline of lots
and lots of projects doing the same thing
again and again, rather than doing this
in stops and starts and knowing that
you've got a customer that will pay you.
All of those things seem to be
very, very important factors.
Even in the story that you've told
about where things went wrong in
Britain, one of the big elements
seems to be like basically kind of
really dysfunctional competition.
It's kind of companies, underbidding
companies sort of doing things that
are not really wise in their long term
interests, but maybe make sense in some
sort of short termist way because they
have to keep the company afloat and
they need cash flow or something like
that and they think they can figure
out the problem later down the road.
The argument for a state led approach
to this is that it gives you certainty,
it gives you scale, you know that the
customer will still be there, that
the government, the one thing about
a state is that it almost certainly
will be able to pay its debts and
will be able to fulfill its contracts.
And if the supplier is also the state,
then it has much less of an incentive
to undercut those incentives.
Now it may have broken
incentives in other ways.
And I think your point Ben is right that,
I think it's a good point that basically
there are successful models of each and
there are unsuccessful models of each.
So it's not at all clear that the reason
that one like France was successful is
this factor, but looking at the specifics,
it definitely seems like if scale is
really, really important and certainty
is really, really important, then...
By the way, I don't think
scale is that important at all.
Okay.
I also don't think certainty
is that important at all.
So like, if you look at, the other
kinds of things we build, like gas
power plants, coal power plants, they
get better every year, and they get
cheap - and they're cheap to build.
Right?
And these are built by a completely
wide diverse array of suppliers.
And the point is that like everything
that goes into a nuclear power plant
is except for a tiny, only a small
number of the components need to be
like specific to a nuclear power plant.
They're all things that you could
supply chain up reasonably well
if you were able to do that.
If you look at the cheapest
power plants of all time, they
were not copy and paste jobs.
They were ones where they just
took decisions, so these are ones
built in the late 1960s in the US.
These are ones where they took decisions
on the fly, they did different designs in
every case, they basically, they had the
option of just doing what the previous
guy did, but if there was a cheaper
way of doing it, they would do that.
So they got better every time.
Now I am not saying that if we had
a state run nuclear body that I
would be fighting to disempower it.
I think that it would be a mistake
to focus so heavily on that.
Right now we have the UK government now
is basically building, Sizewell C like
it's taken, it's got the largest stake
in the project, it's funding it now.
Sizewell C being the successor
next door to which I saw as well.
So next door to Sizewell B.
Yeah, so the government is doing that.
We'll see if it makes a difference.
I basically do think it could
be done really well through the
states and I think we have lots of
examples of that around the world.
I just think.
Just as a, like an intellectual
point, I think it'd be wrong
to focus on it above as the only
thing, only factor of that matters.
And I suspect if the UK
government did it right now, it
would help in certain respects.
Because if it was taking on the cost
directly on its balance sheet, they'll
become this really powerful stakeholder,
the treasury trying to keep your
taxes down because then they we're
gonna get the money from if 'cause
they don't wanna ask you for more.
That would be like pushing the price down.
They might just say, let's not build it.
I think there are some helpful,
there will be some helps and forces,
but I think a lot of the forces
would just act on the government.
Like we have a lot of stuff that is being
built by the government in the UK that
ends up being really expensive anyway,
like the lower thames crossing or there
are loads of projects that the government
builds and they become really expensive.
And so it seems it'll be you know,
HS2 or across, like all of these
ones are built by the government,
they end up really expensive.
So I suspect it can't
just be government run.
No, I, I agree.
Like one bit of the government is
really, really good at tying the hands
of another bit of the government.
And we also actually
see this historically.
So for example, British nuclear power
stations through the seventies and
eighties were run with significantly
lower capacity factors than their US
peers because, well, power individual
power stations weren't profit centers.
They were part of a kind of
big like, you know, sort of
nationalized kind of energy system.
So the people running them had
no incentive to like, invest
or learn how to do better.
Preventative maintenance didn't
really kind of have much of an
incentive to figure out like what
are the gaps between like maximum
potential output and actual output.
So like when they were forced to do
this in the 1990s because they were
taken out of the nationalized energy
system and put in their own company,
even though it was state run and everyone
was watching them, it turned out that
they could like eek out all these
performance improvements they simply
sort of hadn't bothered doing before.
And they, it was the equivalent
of adding two new reactors.
In the late nineties when
they professionalized the
management of the existing stock.
But as you say, that was done under
the, under a different state body.
It was done under a different state
body, but when the industry's
feet were being like held to the
fire and there was an intention
to eventually privatize it again,
I mean what I would take from that
though is, and I think you agree,
is like you can have well-run state
organizations and you can have badly
run ones and there's a massive, the
bigger difference is well and badly
run, and I think that's what we see in-
Yeah.
I think it's about like incentives
and with both Hinkley Point C and
Sizewell C, we got the incentive
sort of catastrophically badly wrong.
Alright, well then let's bring us up to
Hinkley Point C and Sizewell C. Because
then we can move on to the present day
and what we do about it and things like
that, but let's finish the story of
global nuclear with a kind of special
twist or special focus on Britain.
So we do Sizewell B-
Yeah.
It's great.
It's amazing.
Pretty expensive.
I think by historical standards.
That's 6 billion in modern money.
So it was, the prices were
beginning to climb for sure.
Then we don't do anything
for 10 something 15 years.
And then Hinkley Point
C comes onto the agenda.
Talk to me about that.
What happens with Hinkley Point C?
So Hinkley Point C kind of begins
and comes up in essentially
the late two thousands.
So 2007, 2008 we've committed to
these ambitious climate change targets
and so we suddenly decide, you know.
actually, maybe nuclear power would
be, some of our nuclear power stations
beginning to be decommissioned by
this point, some of the old ones.
And we're like, well, maybe we should
bring some of this capacity back.
And so we government publishes a
white paper saying nuclear's back,
but we don't wanna subsidize it.
But if anyone else would like to pay for
nuclear power station, here are some,
sites that will be nice to plonk them
on, one of which is like Hinkley Point C
Because it's already got grid
connections and things like that.
Yeah, it's because it'd been the
site of two, Hinckley Point A and
Hinkley Point B already, in fact.
Yeah.
I think pretty much all the sites they
outline as potential ones, like former
nuclear sites of one sort or another.
And then EDF proposes this design, the
European pressurized reactor, which it
was already trying to build in France.
And the EPR was essentially
a product of Chernobyl.
So France stops building new nuclear power
after 1991, partly because public mood
shifted a bit, but primarily because they
built so much nuclear capacity that they
didn't need, like they were actually, most
of their, a lot of their plants were like
running below their theoretical capacity
because they had like too much energy.
So they froze the nuclear
build in 1991, 2004.
They decided to try and bring back nuclear
power, but with a much warier public.
So they team up with the Germans
to design the safest reactor ever.
Which is built to Germany's very, very,
very strict nuclear safety standards.
So it has like insane amounts
of redundancy built into it.
Some of the most advanced
safety systems ever.
But it's an incredibly
complicated, convoluted design.
A lot of nuclear engineers sort of
wince when you talk about it because
they consider it such an aberration.
And EDF proposes to build one of these
and they won't build it without subsidy.
So like we're not taking
on that kind of risk.
So it goes through the UK's
nuclear approvals process, which
has become incredibly conservative
in the intervening period.
They demand several sort of crucial
changes to the design versus the
French reference design that's
already being built at Flamanville,
which itself runs over budget
because the design's so complicated.
And then this essentially they
significantly increase the
amount of concrete required,
the amount of steel required.
They desire, they demand a backup
analog control system, which means
they have to re- they have to actually
design one of these analog control
systems, which takes 12 years.
So they build in all this additional
complexity and then EDF goes,
yep, yep, we'll do all of this.
And then we agreed a sort of fixed
price at which we'll like buy
energy from them in the future.
Same way we buy, or we pay for wind
power and solar power in the UK.
But because I think they reach the
financial assessment with us after we've
asked them to make all of these changes,
they essentially have no incentive to
push back on any of them because they
can, they end up making a loss on it,
but they can push a lot of the cost back
onto the British taxpayer and the CFD, we
agree on Hinkley Point C, I think is more
expensive than the UK- So as high as the
UK retail energy price has ever been apart
from briefly during the Ukraine shock.
So we agreed to buy energy from them,
very expensively, and they then build a
very, very expensive power station that
is running several years behind schedule.
Mm-hmm.
Already, and is set to be the most
expensive power station of all time built
anywhere in the world, ever, and should
appear at some point in the early 2030s.
So essentially we concluded that
like nuclear would always be slow and
expensive and essentially with the
complicity of the industry of turn
this into a self-fulfilling prophecy.
And to be fair, while some countries
have, you know, done better in that no
one's done as badly as as Hinkley Point C
in cost or as Sizewell C will be I gather
that the Finnish, who are pretty good at
building nuclear power plants are still
building their most expensive ones ever.
And the French obviously
are doing that as well.
And then famously, the US took 27
years to build their most recent
nuclear power plant, Vol - Vogtle I
never know how to pronounce it Vogtle?
And that's also extremely expensive.
Well, no, not quite as bad as ours.
But extreme, extremely expensive.
This was happening everywhere.
But we were the worst.
It's also not been happening everywhere.
So if you go to South Korea
...
Fair point.
Or the UAE, like they managed to
build, I think that South Korea can
build nuclear power to sixth of a
cost per kilowatt hour versus us.
Per kilowatt, per kilowatt of output.
Yeah.
Yeah.
Yeah.
And China's reported, I mean, I don't
know how much their numbers that they
put out are exactly accurate, but
they're reporting roughly similar costs.
And to be fair, several developing
countries or countries that
are, have quite low wage rates
are reporting low cost as well.
But I don't know how comparable
they would be to the UK.
Yeah.
I mean, China actually did build one EPR.
I think they managed to
build it for $3 billion.
Who did they build it with?
How did they get the design?
I think they just, I think
they just licensed it.
No, they did a joint venture.
It was a nuclear technology
transfer venture, but by a plant.
So it was still China's
worse nuclear build.
Basically, some countries survive, like
Japan was actually doing, still doing
pretty well in both speed and cost until
Fukushima when, in my opinion, they
significantly overreacted since then.
But that's a problem since then.
China, South Korea, a few other
cases have managed to keep costs low.
Russia hasn't been as bad as the
west in it, in its cost explosion,
but it is a roughly general story.
And the US and UK is a very similar
story, albeit, it's so funny how
parallel it is given that we had a
Soviet style, planning system for and
they had a private competition system,
albeit with regulated private utilities.
But roughly the same story,
roughly the same causes.
Although it's complicated,
obviously different between
countries, that gets us to now.
I guess it's now a world where everyone
believes the nuclear power is destined to
be costly and slow, and no one is like,
very few people are optimistic about it.
Okay, so we're up to the present day.
There's another version of this
conversation that we could have done,
or we could have framed this in a
completely different way, which is,
solar power is really promising.
It seems to be dropping in price
by an extraordinary amount.
It's already proving to be really, really
cost effective in some parts of the world.
Sub-Saharan Africa, it allows places that
are not really connected to a decent
national grid to leapfrog that need.
And you're now getting solar plus
batteries providing kinda increasing
amounts of electricity in some,
in some developing countries.
And it's becoming more and more cost
effective in places like the United
States, especially in the Southwest.
So like, why should you care about nuclear
power if you are optimistic about solar?
And Alex, I would love to hear, you
know, why should you not necessarily
be completely bush on, on new, on solar.
Before, before we get to that, I'll
give the solar optimist case for
nuclear, which is, I think there are
kind of three or four elements to this.
The first is you can never be certain
about anything like learning curves.
A paper came out recently that
showed that learning curves in terms
of cost reductions, they're not
predictive of future cost reductions.
So you can't be sure that solar will
continue or batteries will continue
the cost reduction trajectory,
like they might, and it'll be
absolutely brilliant if they do,
and I really, really hope they do.
But it would be a really bad idea to
just assume that they will, because
if they don't, you're in trouble.
The next is, there are just some
parts of the world where solar is very
unlikely to be useful in the near future.
Britain is clearly one of them, some
of the Nordic countries are now,
obviously Norway has a lot of oil
and gas and it has a lot of hydro.
Sweden gets a lot of hydro from
both itself and from Norway.
But there are parts of the world that
just will not be able to rely on solar
barring, like a much, much bigger
breakthrough than anything anybody is sort
of saying is imminent in the near future.
Another is that nuclear can be used
for things that solar can't be.
So nuclear reactors generate a lot of heat
, fossil fuels are good at generating heat.
Solar is not that good at generating heat.
It's, I'm not saying it's not possible,
but a nuclear reactor is directly, as
Ben said at the beginning, a nuclear
reactor is like a fossil, it's like a
fossil fuel way of producing heat to
begin with that you then convert into
electricity rather than solar, which
would be converting electricity into heat.
So.
Those are the kind of three
sort of objective reasons.
I think the most promising reason,
or I think the most kind of
compelling reason for me is that
we've already done this before.
There is no technological breakthrough
that is needed to get cheap nuclear.
There is a political or policy
breakthrough that's needed.
Now, you might very well think that
that's a silly waste of people's time.
Now, people in San Francisco, in
California basically kind of take
political constraints as fixed and
that's a really smart way to act, right?
Like you, you make a, you can make
a lot of progress in the world by
taking policy constraints as fixed
and building around those things.
But there are people, and I think of
us as being among them, and there are
lots of other people out there who
can in fact, change policy or who
can at least try to change policy.
And if those people who are probably
not gonna set up a really great energy
startup, or a really great startup, or
maybe, we'll, maybe we will, but if those
people can try to fix the regulatory
side of nuclear and return nuclear to
what it was say 50 years ago, then that's
a really, really great way of reducing
electricity costs without a massive
opportunity cost, at least in energy.
Maybe it has some
opportunity cost elsewhere.
I don't know.
I think that those are to me, those
are the reasons that, you know,
even if you think solar and, and
you know, lots of listeners I think
with good reason will feel really,
really optimistic about solar.
But dismissing nuclear and putting
all your eggs into the solar basket,
I think has a lot of drawbacks.
But at the same time, I wouldn't
say that you should put all your
basket eggs into the nuclear basket.
Yeah.
I think, I think that
would be a mistake as well.
No, I completely agree with all of that.
And my view is as probably the residents,
so skeptic at works in progress, if
other people want to like, raise money
and build companies trying to like crack
long duration battery storage, they have
my wholehearted support in doing so.
Where I get nervous is where, as you see
in places like California, public policy
begins to run ahead of the technology.
And even somewhere like California, like
solar capacity factors drop below 10%,
like extended stretches of the winter.
It's like how big a battery storage
system are we going to build?
Compare that to their
summer capacity factors?
About 30 is it?
So it, it depends where you
are, but can be into like the
twenties I think so, like-
So it's a variation of three times.
Yeah.
It's, so how big a battery storage system
are we gonna have to build, or are we
gonna have to build it all in one part
of the state and build, you know, sort
of huge sort of high voltage cables?
Like is the, these, you know, these
things are theoretically soluble,
but they're, they're not cheap.
Or you go to the northeast of the
us where capacity, solar capacity
factors can be four or 5% in the winter.
You're seeing gas plants being taken
offline and replaced with solar.
And as a, like this to me seems like a
little bit of a disaster in the making.
I will say one thing in favor
of our American friends.
When we discuss this question is that
so as Sam said one of the things that.
Is relevant in this is that the UK
is probably a place where solar is
gonna be, even if nearly everywhere
in the world has a good time, our
latitude is not very good for it.
So we're intrinsically, we find it less
exciting just because although it may help
everyone else, it's gonna help us less.
Maybe we should migrate to the sun as we
should, perhaps always should have done,
but another feature that makes it, solar
better for some countries than others,
especially the US, is that if you don't
care that you're gonna burn fossil fuels
at the rare times when they're off.
Like if you're very relaxed about
that, then the cost of renewables is
much just, is just much lower, right?
Like if you have gas plants, but gas
plants can be boot up quite quickly.
At relatively low cost.
You can just kick them in when you
have any shortfall of solar, even if
it's like a cloud that's unexpected.
If you're willing to keep that capacity
up then, and you're willing to use
it, then solar can be much cheaper.
If you are committing to a world where
you are not willing to do that and you
are closing down gas power plants and
you are not using them for the backup,
then it's gonna be much more expensive.
And so in a world where.
You can only do, if you are trying to
reduce carbon emissions towards zero.
Yes.
Which feel like seems most people who,
you know, even in the UK the the goal is
95% renewable production of, or carbon..,
Is that what Clean power 2030 is now?
Clean power 2030 is 95%.
I think basically, I don't think anybody
is seriously arguing it should go to zero.
Which really strengthens the
case for renewables and solar in
particular, that said Brian Potter
which I'm sure I'm sure many people
listening read his excellent blog.
But Brian Potter did a good analysis
of like how much of the US's energy
could solar supply and it started
getting quite expensive, above
60% on like today's technology.
Obviously a lot of this depends on
what, what happens in the future.
If in 10 years time we've.
The curves continue the way they've
been continuing today, then that might
just, that might not be the case.
You might also be able to get, even, you
can imagine a world where if things get
sufficiently cheap that you can overbuild
the UK enough that it, I mean, it's
very hard to imagine given how little
sun there is in winter, but you can
imagine a world where that happened.
I think we would need like 20
batteries per person for that.
I think we sat down and tried to do
the maths on this once, and I think
it was like
trillion
pounds or something.
I think we had I think we came up
with something like a 4 trillion pound
battery storage system that we...
I mean, 4 trillion pound under
current prices though, right?
Like, the premise here and the premise
here is that prices fall exponentially,
or prices fall incredibly, yes.
Significantly, as they have already, like
battery, battery prices have fallen a lot.
Not as fast as solar, but
still very, very fast.
And if that continues, then it
might not be as crazy to do this.
Like I don't think anybody's
really proposing yes.
To have battery storage to cover a
country's needs over like three or four
months of the year or something like that.
I think thats a bit of a strawman,
but I think it is reasonable to
imagine a world where this moving
target is moving in the direction
that it is, has been moving Yeah.
And continues to move in that way.
There are really big economic incentives.
I don't know enough about battery
technology to say whether there are
like really unsolvable problems ahead.
But I don't think that, that, it's crazy.
But yes, you do need a lot
of gas backup in the system.
Sometimes you come up against
technological hard walls.
Other times you come up with just new
ways that are totally way better than
the previous things that you did.
And you don't even need to keep improving,
you know, lithium ion battery batteries
or those kind ... Or invent a new kind..
New kind.
Right.
Like Austin Vernon is trying to build
you heat up some dirt and then that
stores your energy over the year.
That could be the answer.
I don't know.
I'm completely open to that stuff.
I think that we'd be, like you said, it's
probably foolish to bet completely on it
and if we made it so that we could build
nuclear power as quickly and as affordably
as South Korea, then there wouldn't
really be this either or question anyway.
Like China builds loads of nuclear power
plants and also puts loads of solar in...
There isn't actually an
opportunity cost, that's the point.
Right?
Like if there was an opportunity
cost, then this would be really hard.
And I, and a very, very like,
interesting, tricky question.
I would probably end up falling on
the nuclear side, I have to admit.
. But there isn't an
opportunity cost here, there,
Unless it turns out you'd be an amazing
employee of a renewable energy company
and then there's a huge opportunity cost.
Or, okay.
Or, or if I'm actually thinking about
it, like maybe permitting would be,
you know, the opportunity cost is
for people who are working on cheaper
nuclear is like, can you make it
cheaper to build or deploy like solar
grid connections or like general,
general electric grid connections.
So like maybe that's, maybe that's
where there's an opportunity cost.
To be honest though, those are
really, they work well together.
Mm-hmm.
Like if you fix the energy market
for... You could kind of see it
all as a fixing the energy market
thing, which would benefit solar
and also nuclear at the same time.
So what's what's your model of
what a successful next five years
looks like or next 10 years looks
like for getting cheaper nuclear?
So the one thing I think may just
happen, and it may just be that we
are a sideshow to this and it's just
going on, is that it used to be that the
costs of expensive nuclear power were
spread very finely across the country.
And or the benefits of cheaper nuclear
power would've been spread very finely.
Everyone gets a few pennies off their
bill each, you know, hour or whatever,
and it takes a long time to come about and
you don't know exactly, you can't trust
that it's actually going to come about.
Something else might happen in the way.
And so it was, it's been hard
to coalesce a movement in favor.
You can get movements to come
in favor for really dispersed
benefits, but it takes a while.
You have to really build them up, and we
haven't really been winning the people
who have cared about these things.
But suddenly with AI data centers,
like this week meta signed a 6.6
gigawatt deal with a bunch of
nuclear providers to over the next
10 years build like a 6.6 gigawatts.
So like the equivalent of five normal
size nuclear reactors or a bit more than
that, whatever that suggests that the
benefits are now becoming of cheaper
power are becoming very concentrated,
especially in the US where there's
this huge investment, in build out.
If you have very concentrated benefits,
you can imagine that if you are
meta you and you are paying say five
times more than what you could be
paying for this power, you suddenly
become fairly powerful advocate for.
Showing the benefits of this change, and
your, you can, you could, by capturing
those benefits, you can become this try
and work out who are the blockers to this?
How can I spread the
benefits with these blockers?
How can I like, make them a
broad coalition for change?
Who do I need to fund?
All those sorts of things.
They might mess it up.
They might not necessarily succeed,
but if all of the five biggest
companies in the US are advocating
for cheaper nuclear power, that
might be enough to make it going.
Now a bit more worried about the UK.
Yeah, that's where it gets trickier
is that we don't quite have that same
like potential lobby emerging, but
I think we have like other, we have
other reasons probably to be hopeful.
So for example.
Towards the end of last year, a review
was that the UK government, commissioned,
was published and it came out with some,
a relatively punchy set of recommendations
about bringing the cost of nuclear down.
So, for example, streamlining
some of the very time consuming
approvals processes that exist.
So, you know, a developer doesn't have
to go to like four different regulators
to like, determine the same issue,
removing some of the incentives that
make people who make it really, really
easy and really cheap for people trying
to sabotage development to sort of bring
spurious legal claims and also then
remove like... Essentially directing
regulators not to kind of mitigate and
demand goal plating for these trivially
small radiation doses that have, forced
the installation of, extra filters.
So like some of these measures
can, you can imagine bring down
some of the costs of these like
big UK nuclear mega projects.
And I think the other encouraging thing
we're seeing is that a lot of European
countries are now beginning to look
at reviving their nuclear programs.
So for example, Sweden is
exploring building SMRs Hungary
and, Czechia Poland are all have
nuclear programs are going ahead.
Romania is weighing up competing options.
Denmark is looking at lifting its
50 year ban on new nuclear power.
So like, you know, there are these
kind of, You know, they are these
kind of like green roots of hope.
I mean, in my view, ideally to keep
the statists who are tuning in happy
is that it would make much more sense
just to have European wide coordination
, on SMR, so you don't have like nine
parallel supply chains being built
by all of these different countries.
But small module reactors are interesting
because my understanding is that
yeah, there is a significant trade
off in efficiency the smaller you go.
But you may have, um, basically
two, two countervailing forces.
One being you might be able to get kind
of one and done regulatory approval
for a small modular reactor Yes.
That you can't get for a
large scale nuclear reactor.
So you get like sort of regulatory
economies of scale from that, that's
uncertain, you know, most countries,
I don't think any country has
like finalized how it's regulating
these things, but that would be a
way that it could be really nice.
The other is if they can be mass
produced in factories and they can be
assembled kind of relatively easily or
relatively kind of, straightforwardly.
Then you might get the kind of
learning curves that you get in
a factory that, that solar has
obviously benefited from and like most
manufactured goods get to benefit from.
That does seem somewhat promising.
Yes.
But it's definitely a kind of second
best to getting big scale reactors.
Yeah.
Cheap and, and like, is that right?
Or do you disagree?
I think SMRs are inefficient
versus the platonic ideal of
the gigawatt scale reactor.
But we never build the platonic
ideal of the gigawatt scale reactor.
So if like we were building like,
you know, big nuclear projects
efficiently, I think that the sort
of the trade off would be much bigger
and much more obvious, but we aren't
currently operating in that world.
There are some things where having
small reactors would be really cool.
Like there are a lot of big ships.
Yeah.
They basically want like
a steady amount of power.
They emit a lot of carbon as well.
Yeah.
Yeah.
They emit, they're like, they're
burning loads of kerosene or whatever.
It's, they burn diesel.
We could just like 20 megawatts
that would do 'em forever.
Yeah.
Like a hundred years of power
.
They could stay at sea all the time.
Yes.
That'd be nice.
Yeah.
Well, like nuclear subs, right?
Nearly all the time.
Yeah.
I, there there are a
lot of things like that.
I think also you were saying
about process heat earlier.
There are some things
we just want heat for.
Yeah.
Rather than we want high temperatures now.
Most of the nuclear power plants we have.
Work with pressurized steam.
Or a pressurized water or pressurized,
sometimes boiling water or steam.
That's sometimes good, but it's not
gonna be, it's not gonna like give
you the a thousand degrees or like
1500 degrees or 2000 degrees you
wanna use in lots of, but there are
some designs that all these advanced
SMR, I mean, they don't exist yet.
So, a lot of times something that really
annoys me is when you read copy about
SMRs, like copy is the right word.
'Cause it's like kind of ad copy
like that people are writing about
trying to promote, but they often say
like, this SMR does x. Well, no, no.
You mean the design is
designed to do this?
Like it doesn't exist anywhere.
You can't tell me that it does this.
In principle, especially if
we're doing, we're starting
from square one on regulation.
There are a lot of design
types that might be cool.
That we could run them with like much
hotter coolant much hotter coolant means
that we can have, higher process heat.
We can use it for like lots of
cool, so you could have this in
principle 20 megawatt reactor that
powers all the electricity that a
plant needs, but also does its heat.
Yeah.
Because converged electricity of is often
better, but it just has its downsides.
Right?
And that's why have the
downsides if you don't need them.
That's what I find really interesting and
I am like genuinely unclear as to how much
of the kind of the industry question or
the industry problem can nuclear solve.
'cause obviously, , but not
obviously, but like electricity does
top off at a certain heat level.
It's like 300 degrees
or something like that.
I think the arc furnaces can get
significantly hotter than that can they?
But we definitely can't get
as hot as a blast furnace.
Yeah.
And maybe it's just a case
that we just have to keep using
fossil foods for those things.
That maybe that's fine, but.
Perhaps nuclear does have this like
extra thing about it that other, for
me, it sources of energies don't have.
I suspect that what we should be doing
if we take climate change seriously is
mitigating all the stuff that's cheap to
mitigate and electrifying it, and then
building loads of clean energy so we have
abundant clean energy, and then we'll find
like the 5% or the 10% of things where
it's like, it's quite costly to do this.
Like is it worth doing it?
Could we not come up with some way of
sucking carbon outta the atmosphere or
like, have we gone far enough and the
natural feedback mechanisms are gonna
solve this problem, or whatever it is.
But , you can imagine a world
where, okay, it's expensive.
So currently to extract carbon outta
the atmosphere, but you can imagine a
world where that was less expensive than
just getting rid of blast furnaces and
switching them to, I mean, actually I
gather that arc furnaces aren't that
bad, but there are other ca cases which
I cracking plants or whatever that I
don't understand this stuff very well.
There's definitely, there's
definitely potential there.
I think SMRs are probably better than
just regulatory get around because of
the scale thing, like scale is just good.
Mass production scale you mean?
Yeah, exactly.
I think, but, but unproven, there
are a few plugged in, in the world
that count as small modular reactors.
Mm-hmm.
Um, I'm always doubtful on to what
extent they're actually modular and Yes.
Are they really that small?
The ones I like are, like, last energy
design is really actually quite small.
It's like a one ship, like
a few shipping containers.
Whereas some of them are like a
giant, a couple of warehouses.
It's like, yes, it's smaller than
Sizewell B that you mentioned.
Rolls-Royce SMR is as...
It's not actually an SMR.
Yeah.
It's like two football fields, right?
Yeah, it's huge.
I mean, the SMR definition was
traditionally like 300 megawatts or under,
and it's over 400 and it's not modular.
Yeah, yeah.
It's neither small nor modular, but it is
a reactor, so we can't do the real joke.
Yes.
I'm reasonably, I basically, I think
that it will be - it's foolish when
people say, look, you're working on
a problem that is, it would make the
world better if you solve that problem.
But there is another and more important
problem because like, there are a
lot of problems and we shouldn't all
work on the most important problem.
We still need someone to make our shoes.
Like, that's not the most important
problem in the world, but it'll
suck if we didn't have shoes.
And there are loads of things like that.
So I think it's foolish to say like, you
shouldn't try and solve nuclear costs.
I'm hopeful that it will that it
will solve itself now that there
are strong stakeholders in the game.
But that doesn't, that's not
a reason for complacency.
Great.
Before we wrap up, is there anything
that we haven't talked about
that you'd like to talk about?
So I don't wanna go off on one
about this because it's definitely
a topic that people who talk to me
have heard too much about already.
But I think that is the fear of
radiation is very exaggerated.
And I think there's a really interesting
question as to why people have a
fear that's so out of line with what
I think the evidence clearly shows.
And I'm coming up with some
theories right now, but that's
not what I wanna talk about.
Just the evidence against there being
big harms from the normal levels of
radiation that you might be expected
to experience if you were near a
melting down reactor, for example.
So like, take Chernobyl, according
to, I think it's the UN um, radiation
committee's report, they reckon
killed about 200 people total.
Of those most of the deaths are estimated,
there's no proven link, but they
think probably 200 people ish died of.
Raised cancer rates due to
infected milk and things like that.
There are about what, 20
first responders who died?
28?
Yes, something like that.
Um, now those guys had doses
of, we, we are, we often talk
in millisieverts, that's jargon.
Let's think of it in
terms of CT scans, right?
Because a full body CT scan is the most
radiation that norm someone would normally
experience in their life unless they
had like radio radiotherapy for cancer.
But that would be focused in one place
so it wouldn't have, um, or body harms.
Um, they're 10 millis verts.
So these guys were getting like
within hours.. Hundreds of seavers,
so like hundreds of thousands.
So like, you know, thousands
of CT scanners and they
died like pretty quickly.
Within hours.
Um, or days.
Uh, then no one else.
Um, the people who are, anyone who is
nearby, people who are like, no one else
has any proven harm except for higher
cancer rates like decades later, and
this is the worst nuclear disaster ever.
You can look at other examples
that are pretty cr like
these are the extreme cases.
Yeah.
I think it's worth looking at
the extreme cases rather than the
cases where we try and detect an
effect from really tiny, amounts.
'cause I think that the statistical
techniques are all like, kind of
making something out of nothing there.
So other crazy cases people, so
Hiroshima and Nagasaki, right.
Obviously their main effect is they
just directly killed loads of people.
But also if you weren't in the killed
by the direct blast and you're within
1200 meters, you got enough dose
to have a detectable health effect.
Am I right?
1200 meters?
Yes, roughly.
Um, and then beyond that point, they
have load, they, they collected data
on loads of people through their
whole life beyond that fo point
noted that no detectable effects.
And these guys got quite a high dose
in quite a short period of time.
And they found like, no, like
heritable effects either.
So like the descent, you know, you have
this idea, so you often people about like
talk people about Hiroshima and Nagasaki
go, well, you hear about all these babies
who were born like with like deformed...
I think there were deformed babies,
but those were people within 1200
meters who were pregnant at the time.
They were hit by the blast.
But there were no like delayed effects.
Yeah.
You have people with this imagination that
were like waves and waves of them, which,
So these are really extreme cases.
Yes.
So in the Hiroshima one, they
got a hundred CT scans in a second.
Right?
And they had no effect,
no detectable effects.
That's like the, um, the edge case.
Um, then, but we, we can go,
like, there are loads of different
examples looking at stuff like this.
So there's this really horrible case
of, in the 1920s, people used to like
to have glow in the dark watches.
Yeah.
And the only way they knew of doing it at
that time was to get radium, radioactive
radium paint, paint it on the dials.
And there were women, it was all
women who licked brushes with radium.
And they, they were told to lick
them for the first eight years.
And then they realized that it was
making these women's jaw fall off from
like in turn, basically inside taking
the radium inside led to much higher
doses than being near the radium.
The radium wasn't necessarily a
problem when your skin could block it.
But when it was inside, it
was causing, alpha emitters.
Yes.
Yeah.
Cause lots of damage when they're inside.
They stopped doing that anyway.
They tracked up, they
tracked all of these women.
Um, and you can see a straightforward
curve of when they got to, is it
a hundred millisieverts a day?
Yes, that sounds right.
Right.
Um, so 10 CT scans a day.
The, the girls who were getting a
were the women who were getting a
dose of a 10 CT scans a day, had like
long-term, very noticeable cancer
effects, but when they were like
five CT scans a day, they weren't
getting, any noticeable cancer effects.
Yeah.
Um, and there are many, many, many
studies like this on extreme doses.
They were really crazy ones
like the CIAs, is it, is it CIA
did the nasty one where they-
Is where they injected plutonium
in like terminal cancer
patients to see what happened?
Yeah.
So they found a bunch of, um,
I think they're all men Yeah.
Who were dying of cancer and they
were like, okay, let's see if we
can kill them quicker by injecting
them with loads of plutonium
because they're gonna die anyway.
They didn't kill any of them.
One of them didn't have cancer.
He was misdiagnosed,
also didn't affect him.
He lived till 84 and they gave him
like hundreds of doses of a dose.
They were deliberately trying to
see can we create an effect here?
And they weren't able to do it.
Now, I'm not saying radio like radiation,
clearly it kills you instantly.
If you go into Chernobyl
there are loads of doses.
Some of these doses, like the Hiroshima
guys, I am very surprised by that.
And I would've thought that
a hundred CT scans in a day
would be enough to affect you.
Long run.
But the point, and, but the
point is actually, I think the
evidence is that the harms of
radiation are massively overstated.
Um, and in fact, even a meltdown usually
doesn't cause significant health harms.
Like we should work to prevent
meltdowns, but we should also live
in a world where a certain number of
meltdowns is an acceptable cost of
doing business, so long as we have
reasonable protocols to deal with them,
which we do and like, which we do.
So anyway, that's just like a bugbear of
mine is that people massively exaggerate
the potential harms from radiation.
Very interesting.
Thanks very much, Alex, Ben.
Anybody if you're listening or watching
and you found that interesting, I
really, really recommend, Ben and Alex's
writing, especially Alex has written
tons for works in progress on nuclear.
The French nuclear build out to me is
the most interesting and absolutely
mind boggling and makes me a real
kinda Franco file or more than, more
than I already was, a Franco file.
Uh, and you've got more to come.
Yeah.
So, yeah, so, um, lots to
be excited about there.
Thank you very much for
listening and for watching.
If you've enjoyed this, check out works
in progress.co/print to order the,
and subscribe to the print magazine.
Next issue is going out very shortly
and, thanks very much for listening
and see you again next time.
Bye-bye.