Disclaimer. After nearly 40 years managing money for some of the largest life offices and investment managers in the world, I think I have something to offer. But I can't by law give you advice, and I do make mistakes. Remember: the unexpected sometimes happens. Oddly enough, the expected does too, but all too often it takes longer than you thought it would, or on the other hand happens more quickly than you expected. The Goddess of Markets punishes (eventually) greed, folly, laziness and arrogance. No matter how many years you've served Her. Take care. Be humble. And don't blame me.

BTW, clicking on most charts will produce the original-sized, i.e., bigger version.

Sunday, October 22, 2017


Source: NextBigFuture

If you followed Musk's presentation to the International aeronautical Congress in Adelaide last month closely, you would have seen that out of roughly 60 rocket launches next year, 30 will be by SpaceX.  Which sounds fantastic: in 2012 it did just two launches.  But SpaceX's new BFR*  will be so big it the will be able to launch up to 10 satellites at a time.  And to get people to Mars by 2026, they will need at least 6 BFRs.  Where is the funding for that coming from?  Musk said that launches between now and 2026 will generate enough cash to fund them.  But .... there's a lot of cash needed.   What sort of demand for launch services is SpaceX counting on?


Elon Musk is also planning a world-wide high speed internet, using his own satellites.  He set up a satellite division in SpaceX arguing that he could produce satellites more cheaply than the satellite industry was doing, just like he did with rockets.  And then he decided to build out a world wide ultra high speed internet, using these satellites:

Elon Musk's SpaceX wants to launch thousands of satellites into space with the aim of providing super-fast global internet coverage, according to a regulatory filing.

SpaceX – the company on a mission to colonize Mars – outlined plans to put 4,425 satellites into space in a Federal Communications Commission (FCC) filing from earlier this week.

That's three times the 1,419 satellites that are currently in space, according to the Union of Concerned Scientists, a not-for-profit group made up of scientists across the world.

Billionaire Musk – who is also the chief executive of electric car company Tesla – first announced plans for the project in 2015, with an estimated cost of around $10 billion. The FCC filing did not outline the financials of the project.

The plan is to launch 800 satellites initially to expand internet in the U.S. And then the rest of the satellites would follow, although there was no timeline for the launch.

"The system is designed to provide a wide range of broadband and communications services for residential, commercial, institutional, government and professional users worldwide," SpaceX said in the FCC filing.

SpaceX's satellites will orbit at altitudes between 1,150 kilometers and 1,275 kilometers, allowing each one to cover a space of around 2,120 kilometers wide. According to the official filing, each satellite will weigh around 850 pounds and be the size of a small car.

Once "fully optimized", the system will be able to provide bandwith of 1 gigabytes per second for users globally. That's over 180 times faster than the current global internet speed average of 5.6 megabytes per second which was recorded in the Akamai State of the Internet report at the end of last year.

Reports earlier this year suggested Google and Fidelity had invested $1 billion into SpaceX to support the satellite project.

[Read more here and here and here]

The Wikipedia article on what it calls the SpaceX Satellite Constellation says:

In March 2017, SpaceX filed with the FCC plans to field a constellation of more than 7500 "V-band satellites in non-geosynchronous orbits to provide communications services" in an electromagnetic spectrum that has not previously been "heavily employed for commercial communications services."

Called the "V-band low-Earth orbit (VLEO) constellation," it would consist of "7,518 satellites to follow the [earlier] proposed 4,425 satellites that would function in Ka- and Ku-band. The March 2017 plan calls for SpaceX to launch test satellites of the type in both 2017 and 2018, and as of May 2017, begin launching the operational constellation sats in 2019.
Full build-out of the constellation is not expected to be completed until 2024, at which time there are expected to be "4,425 satellites into orbit around the Earth, operating in 83 planes, at fairly low altitudes of between 1,110 kilometers and 1,325 kilometers."

By September 2017, the planned number of sats in each constellation had not changed, but the altitude of each constellation became explicit: the larger group—7,518 sats—would operate at 340 kilometres (210 mi) altitude, while the smaller group—4,425 sats—would orbit at 1,200 kilometres (750 mi) altitude.

[Read more here]

SpaceX will be calling this system Starlink.

At 10 satellites per launch of the BFR, that's over 1,200 launches.  Sheer genius: SpaceX will be creating its own market for the BFR.  That's why Musk feels so confident about the finances of building enough BFRs to start the first Mars colony. While just oh so casually also giving the world, the whole world, everywhere in the world--deserts, seas, forests as well as urban and built-up areas--an ultra high speed internet. 

The achievements of this bloke are just extraordinary: electric cars; electric trucks; cheap batteries to help green the grid; solar roofs; high speed suborbital rockets giving us 30 minute travel times between continents; ultra high speed internet, everywhere, for everyone; a moon colony; human settlement on Mars; high-speed travel via tubes (The Boring Company); and an intracranial mesh to allow the human brain to connect with the grid and AI (Neuralink).  You'll notice I've included a few things in that list which haven't yet happened yet.  But who can doubt that he will achieve his goals?  Even if it takes him longer than he says.

*I might as well let you into the secret: BFR stands for Big Fucking Rocket.  The name was coined by Tim Urban at WaitButWhy

Litter, or less than 2%

There are many who argue that since Australia produces less than 2% of global emissions, we shouldn't bother to try and reduce our own emissions.  Our output is too small to make a difference.  This is akin to saying that you can throw your rubbish out of the car window because you contribute only 0.0001% of total litter.  Countries each individually emitting less than 2% of the global total for CO2 together emit nearly 40% of the global total, which is more even than China. (Chart from Simon Holmes à Court )

Yes, China is very important.  But China is actually taking serious steps to reduce its emissions.  Yes, the USA is important and despite the orange clown, it too is cutting its CO2 emissions.  India is a big emitter, but it's started down the road towards a low carbon economy.  (We will avert our eyes from Russia.)  And if every country responsible for less than 2% of global CO2 emissions also did their duty, we would be able to limit global warming to 1.6 or 1.8 C by 2050.

Source: Simon Holmes à Court

Consider, however, that not throwing your litter out of the window costs you very little.  Compliance with that social norm is cheap.  Perhaps complying with the global need to reduce carbon emissions is expensive?  In which case, you might be justified "throwing your litter out of the window", if everybody else also did that.  Why bother to be moral when nobody else does?  An age-old question.

20 or 15 years ago, switching to renewables was very expensive.  Renewables were 10 or 20 times as costly as fossil fuels.  Only countries with strong social consciences started down the road towards green energy.  They dutifully disposed of their litter, or at least some of it, while the rest of us continued to toss it out of the window.  It was too costly, too hard, to switch to renewables.

Even 5 years ago, though their costs had declined, renewables were still more expensive than fossil fuels.  However, by then the cost gap had narrowed dramatically, and it went on narrowing.  Now, in most places in the world, wind is cheaper than coal, and large-scale solar either is or soon will be.  We now have a powerful economic incentive to stop CO2 pollution.  The moral imperative now coincides with the economic.  Within a few years (3? 4? 5?) new renewables globally, even with some storage, will undercut the cost of old, fully depreciated coal power stations--the total cost of renewables will be less than the fuel cost of coal power stations. The only factor restraining the shuttering of old coal power stations will be how quickly we can roll out renewables to replace them, not how much renewables cost.

The same rapid cost declines are taking place with electric cars (EVs)  In five years' time, electric cars will have sticker prices which match those of ICEVs (petrol/gasoline/diesel cars.)  EVs are already far cheaper to run than ICEVs.  Even in Australia with its high electricity prices a 100 kWh Tesla  can be "filled" for just A$26. This is about one quarter the cost of the petrol needed for a similar range in a luxury (=heavy) ICEV.  Maintenance is cheaper, too: basically rotate tyres and refill windscreen washer reservoirs.  EVs are quieter, smoother, funner to drive, less smelly.  And they will park themselves.  When sticker prices of EVs match those of ICEVs, EVs will be extremely attractive, and car sales will switch very rapidly.  Yet this will require that the source of the electricity which "fills" EVs is renewable, or the benefits of switching (for society) will be a lot smaller.

If renewables and electric cars are cheaper than fossil fuels, the last excuse of the soft denialists will be swept away.  No country will be allowed by the world community to avoid doing their moral duty to cut emissions.  They will have no excuse not to go green.  In fact, small countries will be the least likely to be allowed to get away with subsidising coal or petrol and resisting the switch to a carbon-free economy.  Those small countries together producing 40% of world emissions will be under intense, irresistible, political pressure from the rest of the world to do their bit.  And that includes Australia.

Once, littering was socially acceptable.  It isn't any more.  In some quarters, emitting CO2 is still preferable to going green. It's somehow a mark of your right-wing credentials to support coal (WTF?)   But the social, moral and political pressure against carbon will just keep on intensifying as the costs of renewables plummet and global temperatures rise.  The Right's crusade against renewables will inevitably fail. And it will leave them looking like the really bad guys.

Saturday, October 21, 2017

The ALP's 50% renewables target

The Right, as ever, have been working themselves up into a lather about Federal Labor's 50% renewables target by 2030.  "Unworkable!"  "Risky!"  "Utter madness."

It's actually not a particularly brave or extreme target.

AGL's new wind farm at Silverton, near Broken Hill, will deliver electricity at A$65/MWhAGL said that the cost of solar is around A$70/MWh.  The company estimates that the cost of new (black) coal power stations is A$110/MWh.  According to Lazard, the total cost of wind in the USA has been falling by 14% per annum since 2009, and solar by 24% per annum.  According to Origin Electricity, the cost of renewables in Australia (solar and wind) has halved over the last 5 years, a compound rate of decline of 13% per annum (the difference with Lazard's data is partly due to movements in the A$.)

Meanwhile the cost of fuel in Oz coal generators is rising as old contracts expire. AGL disclosed that its cost of generation was $37/MWh in 2016, but this is based on old long-term contracts.  As these expire, the cost of fuel alone will rise towards $60/MWh.  And the cost of maintenance of the older generators is rising too. I don't know what maintenance cost for aging coal power stations are, but  AGL mentioned $900 million over ten years for Liddell, which is about $7/MWh on generous output assumptions.  So right now, the cost of existing coal-powered electricity is close to the costs of new renewables.  But renewables will go on getting cheaper, while coal won't.  In a couple of years, the economic pressure to shutter older coal generators, even without a carbon tax, will be intense.

What about the variability of wind and solar, you say? Well, battery costs are falling rapidly too. They have halved over the last 3 years, which is a cumulative rate of decline of 30% per annum.   Let's assume that this rate of decline slows down to 20% p.a., though with 12 battery gigafactories being built round the world, it probably won't.

Tesla's South Australia's battery bank cost US$50 million, for 129 MWh of storage.  Assuming (conservatively) a 10 year life, that works out at US$106/MWh delivered, or A$141/MWh.  The CSIRO estimates that we would need less than 5 hours storage at 80% penetration.  So, being cautious, let's do the costs for 6 hours of battery storage at 50% renewables penetration.  6 hours of li-ion battery storage added to wind or solar would add A$35/MWh to the underlying cost of the energy--which is $65/MWh for wind, & $70/MWh for solar--taking total cost (power plus storage) to A$100/MWh and A$105/MWh respectively.

But .... the costs of wind, solar and batteries will go on declining.  Look at the table below:

(Notes: The total renewables cost assumes 50% wind and 50% solar with a conservative 6 hours of li-ion battery storage.  The fuel cost of coal is assumed to rise to $60/MWh as cheap coal supply contracts expire.  The coal price is assumed to remain flat, and the A$ is assumed to remain unchanged.  The cost of new coal is assumed to remain at A$110/MWh.  I've ignored the high probability that some form of carbon tax/cap and trade is introduced.  A $20/tonne carbon tax adds about $20 to the cost of coal-generated electricity.  I've ignored the profits that battery storage could make from arbitraging fluctuations in the wholesale electricity price over the day, which would reduce their cost.)


  • New coal power stations are already more expensive than a portfolio of new renewables with 6 hours of storage.
  • By 2022, new coal will be almost twice as expensive as the renewables portfolio
  • By 2020, new renewables will be as cheap as the fuel cost of coal, before maintenance.  Let me repeat that: the cost of a brand new portfolio of wind, solar and batteries will be cheaper than the running cost of coal power stations by 2020.  And the differential will only widen thereafter.
  • Coal power stations are 25 or 30 year investments.  They are already more expensive than renewables.  Within 3 years their fuel cost on its own will be more expensive than renewables. Why would anyone rational build a new coal power station?  And after 2020 or 2021, why would anyone keep an existing coal power station going when renewables are so much cheaper? 
  • I've used the example of li-ion batteries which are currently the most expensive commercially available storage options.  However, there are already cheaper options.  Pumped hydro storage costs vary a lot depending on the site it's located at, but I've seen estimates of A$30 per hour.  The CSP plant at Port Augusta has a PPA of A$75/MWh, which includes the cost of power and storage, little more than solar without storage, but that prolly assumes some profits from arbitrage.  
  • I haven't even included household/small business solar with behind-the meter batteries.  These by themselves could contribute the equivalent of a couple of hours of grid-level storage.

Labor won't have to push very hard to get 50% renewables by 2030. Market forces are already rapidly driving electricity generation that way.  The people who are "utterly mad" in this whole imbroglio are the right-wing coal spruikers in the "Liberal" Party, and the ill-informed hacks of the Murdochcracy.  Their passion for 19th century technology when cheaper, cleaner, and better replacements exist is, frankly, demented. 

My numbers will certainly be wrong, but the trend is crystal clear.

Friday, October 20, 2017

The Netherlands is to close all coal plants by 2030

Source: Lonely Planet

This week, The Netherlands introduced legislation to close all coal-fired power stations in Holland by 2030.  From PV Magazine:

The news will come as a blow to three utilities that as recently as 2015 commissioned three coal plants in the Netherlands – the newest in Europe – but will be welcomed by clean energy advocates across Europe.

Engie, RWE and Uniper each invested heavily in Dutch coal plants completed in 2015, but now all three utilities will likely suffer massive losses on the write downs made on these coal plants, said Gerard Wynn, analyst at the Institute for Energy Economics and Financial Analysis (IEEFA).

“The impairments reflect the impact of massive growth in renewable power in neighboring Germany, which has depressed wholesale power prices, and the utilities having failed to foresee flat or falling electricity demand,” Wynn said.

Wynn added that the announcement by the Dutch government highlights the risks involved with investing in new or existing coal-fired power generation facilities in Europe, with the Netherlands now joining a growing band of EU nations that are pursuing accelerated coal phase-out plans.

“This, combined with the rise of renewables and the impact on demand of improved efficiency, puts old electricity production models at risk,” Wynn said.
[Read more here]

Thursday, October 19, 2017

The success of the Energiewende

The Energiewende is Germany's "energy transition" to renewable energy, and out of nuclear and fossil fuels.  It is pronounced with a hard g, as in English "get" and w is pronounced as a v.

It is often criticised as a failure, because

  1. It's pushed up the price of electricity in Germany. Germany has among the highest electricity price in the world 
  2. Germany still uses lots of coal, so what's the point?
  3. They've been doing it for nearly 2 decades and they're still haven't reached 100% renewables.  Surely they should have done more?
When the Energiewende started in 2000, wind was very expensive.  Wind turbines were much smaller than today, and the electricity they generated was much more expensive than it is now.  But as the industry moved down the learning curve, costs fell--since 2009, wind has fallen 66% in cost.  Today, wind is the cheapest generation source.  Solar was even more expensive than wind in 2000.  Its costs have fallen 85% since 2009.  The feed-in tariff contracts signed were for a duration of 20 years, so it will take a while before the newer low-cost wind and solar start to influence electricity costs.  But electricity costs in Germany have probably peaked and will start falling from now on.

We must thank Germany (and Denmark) for their decision to install expensive wind turbines 17 years ago.  They started the world down the learning curve, making it cheaper and easier for us to do the same.  Die Welt sagt Danke, Deutschland.  Verden siger tak, Danmark

As part of the Energiewende, Germany had long-term plans to gradually scale back nuclear.  After Fukushima, the German government promised to close down all nuclear plants by 2022.  You can see what happened (in the chart below): nuclear halved from 10% of total supply to 5%.  Coal picked up the slack.  But through it all, renewables kept on rising, from 7% of total supply (mostly hydro) to 34%, a rise of 27 percentage points in 16 years.  I suspect coal has peaked, and will steadily decline from now on, even as nuclear goes to zero.  There's politics involved: lots of German coal miners.  But Germany is good at transitions, and will find a way.

Source Global Green Shift

The third criticism is plain silly.  Germany is a huge industrial economy, the largest to ever make this transition to a green economy (China will be next.) Although Denmark started before Germany, it was Germany which really got the ball rolling.  It's worth remembering that when they started in 2000, they didn't know how they would do the transition.  Biomass? Nuclear? Wind? Hydro?  Energy saving? Solar wasn't even in contention, it was considered too expensive, and it was thought that Germany was too far north for solar to be workable.  The Germans were pioneers, paving a road that the rest of the world could follow.  And despite the size of the German economy and its population, despite the huge learning curve they traversed, despite growing their heavily industrial economy and raising living standards, they still managed to lift renewables by 27% over 16 years.  If they repeat that achievement, renewables will provide  61% of their power by 2032 and 88% by 2048, very close to the necessary 0% by 2050.  

So far we've just been talking about electricity generation.  But that produces only part of global emissions.  Germany still needs to grasp the opportunity afforded by electric vehicles.  Germany is the auto powerhouse of Europe.  Even if the world gets to 100% green generation, it's still a long way from 100% green transport.  If Germany embraces EV targets as China and California have done, its car manufacturers, who have been dithering about EVs, will be forced into the technological future, a place Germany has always been comfortable in.  

[Read more here]

Don't mess with Texas

From Quartz:

Don’t mess with Texas. Although the White House is leading a campaign to burn more coal, states and utilities are largely ignoring the call. In April, West Virginia rebuffed efforts by Democratic governor Jim Justice to revive its moribund coal industry. On Oct. 13, Texas announced it, too, was turning to renewables.

The retirement of three coal-powered plants owned by Texas utility Luminant early next year means wind capacity in Texas will surge ahead of coal by the end of 2018. The Energy Institute at the University of Texas at Austin reports that the lost coal power capacity will be more than replaced by about 4,000 MW of wind power coming online.

[Read more here]

Source: Quartz

Note that that's capacity, not output.  Assuming reasonable capacity factors (60% for coal, 40% for wind) output should be around 8,800 MW from the coal and 9,800 from the wind, on average.  So over the course of 2019, wind will exceed coal in ERCOT.

The key point is this: renewables are cheap and getting cheaper.  So cheap that even in Republican, conservative Texas wind is replacing fossil fuels.  Coal doesn't stand a chance.  And in the USA, the transition is easy because gas is cheap, so "firming" supply (i.e., filling in the gaps caused by variability of renewables output) by using gas power stations is relatively cheap.  Outside the US, because gas is more costly, the transition requires storage, which is still a bit expensive.  But inside or outside the US, storage costs are plunging--gas is just an interim fuel until storage becomes cheap enough to replace it.

Wednesday, October 18, 2017

Why China and coal matter

China is by far the world's largest emitter of CO2.   This chart, from Global Carbon Project, shows the massive jump in emissions since 2000, as Chinese economic growth exploded, and the recent peak as (a) economic growth slowed and (b) the government started to seriously address air pollution.  US emissions have fallen, because coal-powered generation has been replaced by renewables and gas.  European emissions have fallen because of EU climate change policies including the Energiewende in Germany (more on that in a separate post)  India's CO2 emissions continue to grow rapidly, but the shift away from coal in India does seem to have begun.  Remember: to cut the level of CO2 emissions by 80% over the next 30 years, we'll have to reduce them by about 5% per year.

You get a different picture if you look at emissions per capita.  The US is among the highest CO2 emitters per capita (Australia is higher).  But it has also fallen the most.  In the US economic growth has decoupled from emissions.  Emissions per capita are back where they were in the 60s.  But remember, what matters for the climate is total emissions.  We need to get total emissions as close to zero as we can.

The biggest source of emissions is  burning coal. The next biggest is burning oil, mostly for transport.  Gas is way behind, because it produces half the emissions of coal for the same output.  Even better, gas generation can be scaled up or down quickly in response to demand fluctuations.  So we should concentrate of removing coal from power generation and on electrifying transport.  That doesn't mean we should ignore other sources of carbon.  Coal is used to make iron and steel, so we'll need to find a way to do that efficiently using renewables.  Cement production involves baking calcium carbonate to drive off the CO2, so we'll need to find different ways of making cement.

Around the world, coal is being replaced in electricity generation by renewables, gas and nuclear.   Electric car sales will explode over the next 10 years.  And China, the world's largest CO2 emitter is central to both these trends.  The USA is very important too, and the good news is that despite the orange clown in charge, her emissions will go on falling because states, municipalities and companies want clean power, and because the economics now favour renewables.

This year will probably show the peak in global CO2 emissions.   But the initial rate of decline in emissions will be slow.  Too slow.  World temperatures are rising by 0.2 C every decade.  We need to stabilise the level of CO2 in the atmosphere, which means we have to reduce CO2 emissions to below the rate which natural processes take CO2 out of the atmosphere.  And we need to do that fast.

[All charts from Global Carbon Project]