Thursday, October 20, 2011

Investing in Virgin Galactic

Abu Dhabi’s Aabar made a second investment in Virgin Galactic in July, increasing its stake in the company to roughly one-third ownership.  This marks the second investment by the Abu Dhabi based fund in the last three years. 

Bottom line:
  • 2009 investment for $280M equaled a 31.8% stake.  Post Money valuation of $900M.
  • 2011 investment for $110M increased its ownership stake to 37.8%.

Based on this data I estimate one of two scenarios:

Scenario #1: The 2009 investment was a down round with prices per share less than what Virgin had previously valued the company, and the 2011 was an up round.  An example of this scenario is provided below.  Note the percentage change between share price is valid but the share price itself is not publicly known, so I am using a simplified $1 per share for example purposes.

Scenario #2: The share price has not changed since the company's founding.  In addition to the two Aabar investments Virgin has brought in $21.5M from other outside investors.  An example of this math is below.

Both of these scenarios match the data provided by Aabar for the last three years. 

Reasons for Virgin taking additional investment range from:
1.       Preparations for new growth (Nanosat Launch vehicles or other new products)
2.       Paying for the delays in reaching commercial operations for its suborbital product
3.       Building up a war chest for a rainy day (when money is available sometimes you just take it)

Monday, October 17, 2011

Risk Pricing for the Reusable Falcon 9

In my last post I explored how the SpaceX's reusable Falcon 9 (rF9) could threaten those companies offering suborbital launch services.  

Discussion has focused on the risks preventing the rF9 from reaching the optimistic breakeven price point of $130 per kg (discussed in my last post).  Here are a few of the risk categories you have raised:

  • Increased variable costs: Elon may claim only $200K of propellant per flight, but the variable costs of an rF9 flight will surely be higher
  • Reduced number of flights: When you factor in the complexities of reusing a launch vehicle and the potential for a crash or loss of vehicle
  • Reduced payload capacity: Adding reusability will increase the mass of non-payload components – reducing the payload mass

What rF9 breakeven price points might we expect if we take these concerns into account?  In the table below, I explore these risks and their impact on breakeven price per flight and breakeven price per kg.  

In the second column are the breakeven prices today for an expendable Falcon 9.  This is the upper end of cost.  Weight any of these risks to the point you get price points beyond $5K per KG and customers will prefer the current Falcon 9 over the reusable version.  The third column shows the optimistic assumptions from my last post.  Columns four through six:

  • Increase variable cost per mission from $200K to $2M
  • Reduce reusability from 47 missions per vehicle down to only 10 missions per launch vehicle
  • Reduce payload capacity by 50% from 10,450KG to 5,225KG

These risk values give us a range we can talk about.  Even variable costs of $2M per flight, only 10 flights per vehicle, and half of the payload mass consumed with reusability hardware, SpaceX should be able to reach breakeven price points of about $1500 per KG and $7.5M per flight. 

And the great thing about risks…they get retired.  Be as pessimistic as you want to be about the capabilities of the initial versions of the rF9.  Variable costs will drop over time.  Flight rates per vehicle will rise, and payload mass will creep back up.  The key has been (and will always be) flight rates.  I wouldn't be surprised to see SpaceX subsidize their first generation rF9, offering first generation customers prices SpaceX won’t be able to satisfy profitability until the second generation rF9 – all in the name of increased flight rates.

How optimistic or pessimistic are you about rF9 capabilities?  Here is an interactive spreadsheet for you to explore your own risks and their effects on breakeven prices.    

Saturday, October 1, 2011

Will the Reusable Falcon 9 Kill the Suborbital Launch Industry?

With SpaceX’s announcement this week that the company would not only develop a reusable first stage for its Falcon 9 family of rockets but would make a completely reusable rocket system (I will use Clark Lindsey’s nomenclature: "rF9" for reusable Falcon 9), I have been wondering about the future of the young NewSpace companies developing reusable suborbital rockets.  Will companies like Masten, Armadillo and to a lesser extent XCOR and Virgin Galactic, survive this incursion from a well-funded NewSpace Cousin?

(the youtube video via Clark Lindsey's youtube channel.)  
SpaceX has announced the company is developing the “Grasshopper,” a 100 foot-tall suborbital Falcon 9 first stage that SpaceX’s cadre of young, talented engineers will use to test this initial piece of the rF9.  SpaceX has NOT announced any intention to commercialize the Grasshopper.  But if Masten, XCOR, and Armadillo continue to delay bringing a product to market that can reach 100KM, and SpaceX continues to develop products in its typical rapid fashion, might customers ask to buy payload space on an upcoming Grasshopper test?  

Or would SpaceX be willing to sell Grasshoppers to operators who then provide a suborbital launch service to users using the Grasshopper all before Masten has reached 100KM?  Could the unmanned Grasshopper be modified to carry passengers and compete with Virgin and XCOR?  If an operator came with funding, wouldn’t SpaceX take their money to make the modifications to "manrate" Grasshopper?

But the big money is the orbital market.  Most of the suborbital companies have expressed interest in using their suborbital experience and even their suborbital vehicles to expand current offerings to include an orbital system.  XCOR has published this image of an orbital capability.  

Virgin Galactic even took investment money from the Middle East to jump start their orbital program.  Could an rF9 meet all market demand for both suborbital and ultimately orbital launches as well?  And if they do, are the current suborbital companies doomed? 

It all comes down to money.

How cheaply could SpaceX really launch their new rF9?  We don’t know.  SpaceX does not even know yet.  But we can make some interesting estimates.   The heart of these projected orbital price reductions stems from reusing the rF9 like Southwest reuses its 747’s (which can fly commercially for 30 years with proper maintenance).  How many reuses is SpaceX planning on? 

At this point, the best data I have is a nugget SpaceX's CEO, Elon Musk, said this last week that he is targeting $500K trips to Mars as a market for his reusable craft.  

Let’s make some assumptions so we can approximate SpaceX’s reusability assumptions:
  1. A price for a Dragon/Falcon 9 trip to Mars will be equal to the price SpaceX is currently charging NASA for ISS visits ($130M per trip) - optimistic assumption
  2. 5 paying passengers per Mars Trip - optimistic assumption
  3. 10% profit per launch
  4. All maintenance and between-flight costs are included in the launch price - optimistic assumption

SpaceX breaks even after 47 flights (but that is a lot of assumptions).  here is a table to help visualize the math:

 Assuming a 47-flight amortization, what could be SpaceX’s breakeven price per KG to LEO?  Or to say it another way, how low would the suborbital company’s prices have to be to beat SpaceX?  

Again, let’s make some assumptions:
  1. A price for an rF9 to LEO is the same as current LEO Falcon 9
  2. Falcon 9 payload to LEO is unchanged
  3. 10% profit per launch
  4. All maintenance and between flight costs are included in the launch price.
  5. Propellant Cost per Launch = $200K
  6. rF9 breaks even after 47 flights

Based on these assumptions, SpaceX's breakeven Price to LEO for rF9 is $130 per KG or ~$1.4M per flight.  Again, here is a table to summarize how I came to this conclusion.  At the end of this post is a link to an interactive spreadsheet where you can modify these assumptions to create your own analysis.

These SpaceX prices are surely the most optimistic for the near term:
  1. What if the rF9 doesn’t get 47 flights per vehicle?
  2. What if between-flight maintenance costs for the rF9 are significant?
  3. What if payload capacity has to be significantly reduced to accommodate rF9’s reusability elements?
  4. What if near term launch demand is not high enough to fly as often as they need?
Even with the identified risks, this analysis would indicate:
  • Yes, rF9 could compete against suborbital companies for suborbital market share (especially if SpaceX sells the Grasshoppers to entrepreneur operators)
  • Yes, rF9 could compete against suborbital companies for orbital market share through extraordinarily low prices

So how can XCOR and Masten compete?  

I continue to be bullish regarding the utility of Nanosat-class launch vehicles.  When suborbital companies start offering orbital services (a second generation service), their initial orbital offerings would probably be within this Nanosat class - broadly speaking, payload space significantly under 100kg.  Is there still a market for suborbital companies to offer this type of orbital service?  Even if SpaceX may be able to now match (or beat) them on price?  

Yes.  Here is why:

Sometimes smaller is better.  The smaller vehicles these suborbital companies will eventually offer on orbit should:
  • Be easier to "fly full"– to get the $130/KG price on an rF9, you have to wait for the manifest to fill.  Not so with a smaller vehicle.  XCOR was talking about a payload of 12-20KG initially.
  • Be easier (and cost less) to maintain.
  • Be launched with less integration or preparation – this advantage is the BIG one.  XCOR talks about multiple flights on the same day, taking off and landing from existing airports.  Even if the rF9 could launch that often, it will be some time before regulations allow SpaceX to fly that often - especially if they are still flying from the Cape or Vandenberg where ops tempo is measured in "launches per month" not "launches per day".

Nanosat launchers are the future, but only if their ops tempo is fast enough to justify paying a premium for preferential launch windows.  

This advantage of the small won’t last forever.  SpaceX will keep improving its initial RLV offerings.  Spaceport operations will grow to allow for more airline-like ops tempos.  So Nanosat launch operators (today’s suborbital companies) will have to keep improving too.

But there is a market for Nanosats and it hinges now on ops tempo.  There is hope.

The bigger worry…

…is in the near term.  I mentioned earlier, I doubt SpaceX will pursue commercializing their Grasshopper suborbital vehicle.  But they may be open to selling this suborbital vehicle for others to operate.  Such a suborbital operator flying the Grasshopper would have tremendous suborbital market advantages and could be a major competitor to those suborbital companies focusing on suborbital research (Masten, Armadillo, etc.).

Suborbital companies should be worried, but not panicking.  If the reusable Falcon 9 hastens the development of viable Nanosat launchers, the industry will be doubly blessed – low launch costs from the rF9 and high ops tempo from Nanosat launchers.

Here is the interactive spreadsheet so you can build your own rF9 assumptions.