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. These days I'm retired, and I can't by law give you advice. I do make mistakes, but I try hard to do my analysis thoroughly, and to make sure my data are correct. Remember: the unexpected sometimes happens. The expected does too, but all too often it takes longer than you thought it would.

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.

Tuesday, December 12, 2017

Update on costs of SpaceX's BFR

Slide from Elon Musk's BFR presentation at the IAC conference in Adelaide 2017
Costs increase from left to right -- should prolly be a log scale.  Click to enlarge. 

In his presentation to the International Aeronautical Conference in Adelaide, Musk said something significant which I didn't get, first time round.  He said that the cost per launch of the BFR/BFS combo, because the BFR and BFS will be fully reusable, would be cheaper than the cost per launch of the Falcon 1.  The published cost of a Falcon 1 launch was $8.5 million.   

Let's assume $1 million for fuel and another $1 mill for refurbishment every launch, which leaves $6.5 million to cover capital costs/depreciation and profit.  The booster (BFR) will be used again and again, perhaps 100 times (Musk thinks it could be 1000 times), but it's more complicated with the ship (the BFS) because it will be used for different purposes.  As a spaceship used only to reach Mars, its likely total number of uses will be limited, because Mars and Earth are only in opposition every 26 months.  On the other hand, it could be used far more often on the moon run (only 3 days away) and for point-to-point fights on Earth.  SpaceX would prolly use a Mars spaceship for moon missions and point-to-point flights in the 20 months between Mars missions, provided it could be unloaded it quickly enough on Mars for it to return to Earth before Mars and Earth moved too far apart.  So it's feasible that you could use the BFS almost as often as the BFR.

Let's use Sam Dinkin's capital cost from my earlier stab at estimating costings, $185 million for the BFR and $150 million for the BFS.  Let's assume just 50 launches for the BFS to allow for time taken to and from Mars (6 months) and some period spent on Mars loading and refuelling.  That means the capital cost per launch of the combo would be ($185/100)+($150/50) or roughly $5 million per launch.  Add our assumptions for fuel and maintenance and you get $7 million.

So how much would it cost to send one mission to Mars?  $7 million for the BFR/BFS combo, plus 5 refuelling launches which would each cost about $4  million ($7 million just takes the BFS to LEO -- low Earth orbit)  That's a total cost of $27 million.  Add in additional costs for the first mission, such as the two cargo ships containing equipment the first settlers will need, plus a cargo ship to accompany the spaceship, and that totals $100 million or so, still 3 orders of magnitude less than any competing proposal.  NASA's estimated cost for a mere handful of astronauts is $100 billion

It all hinges on re-usability.  If the BFR and BFS are re-usable even just 100 times, space travel becomes cheap.  Satellites launches, trips to the ISS, trips to the moon to start a moon base, colonising Mars, mining the asteroid belt: all become economically feasible.  When Musk first said that he would make his rockets re-usable everybody fell about laughing.  No one else has managed it, they chortled.  Now SpaceX routinely lands the Falcon 9 first stage and re-uses it.  He now says he will be able to re-use his rockets 1000 times, and we all doubt him, even me.  Perhaps we shouldn't. 

US solar output up 47% in '17

In the year to September, the electricity generated from solar panels was 47% more than the same period of the previous year.  New investment in solar was prolly pushed up by the likelihood of a big rise in tariffs on imported panels next year.

It's tempting to say, "so what?", because solar produced just 1.9% of total electricity generated in that period.  But consider.   Let's suppose the rate at which the percentage of solar rises as new solar farms are built is just 23% per annum rather than 47%, slowing next year (see chart) because the prices of solar panels go up after tariffs are imposed.   This takes the percentage from solar to 5.5% by 2022, which is what   GreenTech Media and SEIA (Solar Energy Industries Association) forecast.

Source: GTM/SEIA
Note: the left axis shows new capacity not cumulative capacity
 Assuming that growth rate continues--and it might accelerate, because the cost of solar will resume its falls after next year's increase, and meanwhile, storage will more than halve its cost over just the next 5 years--that implies 15.5% of all electricity generated will be from solar  by 2027, and 44% by 2032.  In practice I think solar will be a bigger part of electricity generated by then because its cost will have declined so much it will easily edge out gas and what is left of coal.  But even 44% implies that soon after, more than 50% of the USA's electricity will come from solar.   Since wind will also be growing, though more slowly, its current 5.6% will have risen to 20% or more, and hydropower will be 5 or 6%, its current level.  By 2040, easily 100% of US electricity generation will be from renewables.

Record highs far outpacing record lows

In a time series which has no trend, the number of record highs should be equal to the number of record lows.  Let's say you start measuring temperatures in 1950.  In 1950, the record high is, say, 25 degrees C, the record low 0 degrees C.  For a new record high to be registered, it's got to be higher than 25, and for a new low, it's got to be lower than 0.  Let's say, the next year, the high is 25.5, and the low -0.5.  These become the new record highs and lows.  But each new record makes the next record harder to reach.  If the underlying trend in the data is sideways, new record highs and record lows will become rarer, because they will be more likely to have been achieved before.  (Reminder: popular use of "record high" or "record low" as meaning "very high" or "very low" is incorrect.  The terms mean the highest high or lowest low since records began.)

So if you do see a pattern of ever higher highs, or ever lower lows, whether it's in market indices, or share prices, or climate records, it means there is a trend.  A trend doesn't mean it's higher/lower every year, it means that over time new highs/lows keep on being attained.  One year may be lower than the previous year without meaning that an upward trend has changed.  On the other hand, if successive observations do show a new pattern of falling highs and lower lows, then it becomes more and more likely that a new trend is in place.

Source: Climate Central

Daily record highs are vastly outpacing daily record lows in the U.S. We will always have warm years and cold years, but in a world without global warming, those warm and cold years would balance over time. However, that’s not what we are seeing. According to the 2017 U.S. Climate Science Special Report, after a rigorous reanalysis of GHCN stations back to 1930, 15 of the last 20 years had more daily record highs than daily record lows. The number of daily record highs outpaced daily record lows more than 4 to 1 in 1998, 2012, and 2016.

A first look at the data from NOAA/NCEI indicates that 2017 continues the warming trend, as daily record highs are beating daily record lows by a 3.5-to-1 margin so far. Below are some preliminary 2017 stats through the end of November. Visit the NOAA Daily Weather Records tool to get the daily updates on these numbers:
  • Monthly record highs have outnumbered monthly record lows at a rate of 9.7 to 1
  • All-time record highs have outnumbered all-time record lows 8.7 to 1
  • Record high minimum temperatures have outnumbered record low minimums 4.6 to 1

As the concentration of greenhouse gases increases in the atmosphere, the ratio of record highs to record lows will likely increase even further. With no change in current emissions trends, model projections indicate that record highs could outpace record lows by 15 to 1 by the end of the century.

[Read more here]

The evidence that climate change is happening right now, and that we don't have to wait for decades to see its effects, is mounting.  Records aren't just falling for temperatures, but also for severity and length of droughts, rainfall events, and hurricanes.  Almost everybody, even if they are concerned about global warming, thinks it's something which will only affect our grandchildren.  But it's happening right now.  By the time our grandchildren are adult, it will be incomparably worse, unless we move aggressively and rapidly to de-carbonise our economies.  Fortunately,  technology and the costs of renewables are helping facilitate this transformation, but we need to speed the process up, and also address areas where such technological progress is not being made: air travel, cement, iron & steel, land clearing, shipping.

Monday, December 11, 2017

Plunging battery costs

Storage is key to our transition to a carbon-free economy.  Even if we have a grid with a good mixture of wind and solar, there will still be periods when the two together don't provide enough electricity and other periods when they provide too much.  And though we can provide electricity to trains or trams via an overhead cable or an additional rail, we can't do that for cars or lorries.  For them, we will need stored energy.  Now, that's provided by petrol or diesel, but in future it will have to come from electricity stored in batteries.  So it's really important that batteries get cheap.

The chart below comes from BNEF via Climate Denial Crock of the Week.


The cost of storing 1 kWh has fallen from  $1000 in 2010 to just over $200 in 2017, an 80% decline, which works out to a cumulative/compound annual rate of decline of 20%.  The rate of decline of the last 3 years, though, has been much faster.  From 2014 to 2017, the rate of decline has been 29% per annum.  This faster rate of decline has coincided with the development of Tesla's first battery gigafactory in Nevada.  Right now, Tesla needs all the batteries it can produce for the ramp up in Model 3 production. So it's unlikely Tesla will be cutting the price of its batteries for public sale yet.  But in 6 months' time, as the gigafactory itself moves towards completion, and Model 3 production beds down, I have no doubt Tesla will be cutting the prices of its Powerwall and Powerpack products. 

Let's be conservative and project a continued rate of decline in battery prices of 20% per annum, though it's far more likely that with 12 battery gigafactories opening round the world, prices will decline faster than that.  At 20% per annum continued rate of decline, by end 2020, the cost per kWh will be down to $100/kWh of storage, half what it is now.

Currently the cost of the Powerpack, Tesla's utility-scale storage solution, is $387/kWh, but that includes the inverter/transformer.  Inverter costs are also falling, so it seems entirely plausible that by the end of 2020, Powerpack costs will have been at least halved.  Currently, the cost of power from the Powerpack is 10.5 cents/kWh delivered ($387/365 days/10 years).  A 50% price cut will take it down to 5 cents/kWh delivered, or $50/MWh.  Currently, gas peaking costs more than $150/MWh.  So it's probable that by 2020, no new gas baseload plants will be built.  Instead, utilities will start using battery storage to firm and balance electricity output.  Also, the cost of wind/solar with 10 hours of storage will be $70 or below, comparable to gas baseload in the US (where gas is cheap), and cheaper than gas baseload in the rest of the world.  Just as coal demand has peaked, so will gas demand peak not long after 2020.

The fall in battery costs will also drive down the sticker prices of EVs to the same as ICEVs.  $100/kWh for storage is where this will happen, but battery costs won't stop falling when they've reached $100/kWh.  They will halve again, making EVs and grid/household battery storage irresistibly cheap.

This is only 3 years away, the end of 2020.  In 3 years' time, EVs will be as cheap as ICEVs, and sales will be exploding.  In 3 years' time, renewables with storage will be even cheaper than they are now relative to coal, and starting to compete head-to-head with gas.  Frankly, those corporations and countries (legacy car manufacturers, oil companies, Saudi Arabia/Kuwait, utilities, etc.) which do not start to plan now for the deluge of profound change that is coming will be wiped out.  The good news for everybody else is that carbon emissions will start to fall really fast in the 2020s, giving us a fighting chance that emissions will be close to zero by 2050, which is what we have to achieve to limit global warming to 2 degrees C.

Transportation now biggest US emissions source

Source: Climate Central

Emissions from electricity generation have fallen sharply because gas and renewables have replaced coal.

The rapid growth in EVs is coming just in time.

Sunday, December 10, 2017

Donald will fix it

by Cathy Wilcox

Model 3 deliveries

It's no secret that Model 2 production has been way behind target.  In the chart below, which compares GM's Bolt sales to Tesla's Model 3 sales, you can see how behind schedule Tesla is--Bolt sales in Novemnber are where Model 3 sales were planned to be by now.  Tesla had been hoping for 1000 deliveries of the Model 3 per week in December.  The Model 3 is a far superior car to the Bolt, but sells at roughly the same price.  If Model 3s were properly available, their sales would surely greatly exceed the Bolt's.

Source of data: InsideEVs.com

So, should we give up hope that Model 3 production will never pick up and that Tesla is finished?    Well, no.  The chart below shows Model 3  VINs as seen by the public (source Teslarati).  The first Model 3 built had a VIN of 1 (or close to; the relationship isn't exact)  Note that recent VINs are running above 2000.  Note also that the curve is exponential, as Tesla said it would be.

I also use Musk's behaviour as a guide.  At the International Aeronautical Congress in September, he was puzzingly distracted, even depressed.  Then there was the tweet of him camping on the roof of the gigafactory while he fixed the problems delaying production of the batteries used in the Model 3, camping because it would take too long to drive to the nearest town to spend the night at a motel.  What other chief exec would do that?  Then came the launch of the Tesla semi and the new Roadster, where he was very upbeat, even ebullient.

From the chart above, I estimate that production of the Model 3 has already reached 300 per week.   Their new target of 1000 per week by the end of March seems achievable.  But production actually has to be much higher than that.  Over 400,000 people have reserved a Model 3.  Production of 1000 a week equals just 52,000 a year. It would take 8 years to satisfy the pent up demand.   From 1000 a week, production will have to rise to 2000 and then to 4000 by the end of 2018.  The exponential curve will have to continue to steepen.  Can Tesla do it?  Yes, I think so.  The first few months were the hard bit. 

What happens when Tesla "uses up" all the reservations?  Nothing.  Tesla Model 3s will be ubiquitous.  Everybody will know someone who has one, someone who will give their family or friends test drives in their new car, someone who will extol the pleasures of owning and driving a Tesla.  And if that isn't enough, Tesla can start advertising.

This production schedule will double US EV/PHEV  sales by end 2018.  And that's without the new longer-range Nissan Leaf or the Bolt., and there could easily be a 50%+ increase from those heights in 2019.  Which is vital.  We must cut carbon emissions as rapidly as possible.