Tech bubbles are worth it

Via FTAlphaville comes a vision of the near future by Morgan Stanley filled with promise:


Autonomous capability: This is an “ideal” world akin to common science fiction in which all cars on the road have at least a Phase 3 level of autonomous capability (including retrofitting older cars), full V2V/V2X capability and the ability to drive from Point A to Point B with zero human intervention.

Scenario: A family of four wants to travel from New York to Chicago. They have dinner at home, climb into the vehicle at 9pm, watch a movie projected on the windscreen, and then go to sleep in their fold-flat seats, waking up at their destination the next morning.

Functions: Fully autonomous driving with no human intervention, with the focus likely to be on lifestyle/entertainment of occupants and manual car control as a back-up/supporting function (or disallowed). Cars will look very different from cars of today. Cars can also travel with no occupants. Remote control/disable functionality necessary.

The Luddite’s and rev-heads among us can take comfort from the fact that any revolution of this kind will be slow to arrive in Australia and we will, by definition, not participate in the construction of these vehicles.

The quotes come from a report into Tesla, the electric car company pioneered by futurist Elon Musk:

While acknowledging its ultimate potential, up to this point we have observed Tesla becoming more of an annoyance to the established order of automotive power rather than a true disruptor. But what if Tesla’s ambitions extended further than giving high-end OEMs a run for their money? What if a period of transformational technological change in the auto industry coincided with Tesla’s application of its capabilities in hardware, software, infrastructure, and manufacturing.


We argue Tesla cannot be valued on near-term multiple metrics like traditional auto companies given that we expect Tesla to multiply revenues by more than 10x from 2013 to 2016 by nearly 30x by 2020 and around 60x by 2028. We have thus chosen a 15-year time horizon for our DCF which captures the full maturation of the Model S, Model X (and top-hat derivatives) and also the ramp up of its mass market electric vehicle (the Gen 3). We have applied a 11% WACC with a range of 9% to 13%. The terminal value, calculated on a midpoint of 10x EV/EBITDA accounts for roughly 50% of the total DCF value across the range of methodologies we have applied to arrive at our PT.

Alphaville rightly gives MS a serve for another round of “this time it’s different” thinking. After all, with TESLA’s share price trading at three-quarters of the value of General Motors with only one hundredth the revenues, it’s got tech bubble 2.0 written all over it:


But there’s something else going on here that is more virtuous. While the capital underpinning TESLA maybe foot loose, it is also visionary. The ludicrous valuation of the firm provides it with abundant opportunity to pursue an accelerated vision and in a way that America owns it.

What I am arguing is that bubbles are not always all bad. In technology and stock prices they can leave spectacular legacies of social progress, just as they did in the pre-1929 crash railways boom and again in the internet revolution of the nineties.

Stock market bubbles based on technology can be productive, and although they implode like all others, because they are based largely upon equity the bust is far less destructive than in other asset classes where debt plays the major role.

So I raise a glass to Elon Musk this morning and say bring on utopia!

Houses and Holes


  1. Hear, hear….!

    Systems analysts like Marchetti have been analysing Kondratieff cycles for transport modes for years, and it is obvious that a new one is overdue.

    There are a whole lot of Luddites around who think that the next mode has to be “post automobile” and some great leap backwards to something less flexible and more restricted in its coverage of the urban area. DUH!!

    Driverless cars simply have everything going for them; potential utilisation rates; the parking externality problem; the congestion externality problem (time no longer need be “wasted by a driver”); the road space utilisation problem; they solve all of the problems of automobility and have none of the disadvantages of mass transit.

    BTW there is potential for long distance travel in driverless cars to be VERY high-speed and efficient via platooning on special guideways. The cost of building “guideways” for driverless cars would be a fraction of the current costs of building highways (per vehicle km travelled).

    Systems like these are extremely logical because firstly, roads are now said to be “too expensive” to build. But a large proportion of the cost is the result of engineering all roads for the heaviest vehicles; whereas “congestion” is a problem resulting from large numbers of lighter weight vehicles. “Obvious solution number one”, then, is to provide a lower-cost “alternative system” of roads or lanes, mostly elevated, in congested areas, for vehicles of lighter weight, as well as bridges, bypasses, and overpasses in obvious places (where cost has been a long standing obstacle to construction of typical heavy duty structures).

    Secondly, congestion is worsened by the fact that vehicles have to observe safe traveling distances, which at higher speeds is actually counter-productive of road lane throughput. “Obvious solution number two”, then, is automated vehicle control that allows very high speeds with very close spacing between vehicles. A certain “drafting” effect (as in NASCAR) under these conditions would also capture gains in efficiency via lower air resistance.

    Thirdly, fossil fuel importation and use, and CO2 emissions; involve “externalities” that policy makers seem to be insisting that we must ameliorate at any price. “Obvious solution number three”, then, is to move rapidly to a system based on electric powered automobiles, with gas and nuclear being the logical sources for generation. But electric powered automobiles have too low a range for most automobile users liking. “Obvious solution number four”, then, is to build electric power supply and vehicular take-up, into the specially provided lighter-duty road network. “Obvious solution number five”, is to place plentiful solar panels along the roads.

    Also logical: the vehicles compatible with the guideways to be also compatible with ordinary surface roads and capable of seamless transition from one to the other. Also; trucks and heavy vehicles will remain limited to the use of the existing road system that has been engineered to sufficient strength for them.

    • There are also a lot of people out there who think the next mode of automobile transportation will be post ICE.

      I think they are going to be disappointed. The ICE is going to remain the cheapest way of powering an automobile (on whole of life cost) for the foreseeable future and I would suggest it will do so not by a small margin.

      • Essentially this argument is the same as the anti-solar argument. It’s “older but more polluting technologies are cheaper”, which right now is true. The second old technologies are no longer cheaper, demand for them will utterly collapse. Seen any cathode ray tubes lately?

        I see your ‘nothing ever changes’ argument and raise my ‘exponential technology curve’ argument against it. Currently, fully electric cars in the USA (Nissan Leaf in this case) cost about 30% more than ICE cars. Electric cars will continue to drop in price. ICE cars won’t as the technology is fully mature. The list of EV and hybrids being released by pretty much every auto maker in the US in the next few years is rather large, so there’s obviously money in the technology, which will result in more R&D.

        Make no mistake that we are going through a technological revolution similar than or larger to the 1930s. We’re in for an absolute transformation of manufacturing and transport, and I suspect ICEs ain’t gonna cut it for shuffling humans about (I suspect heavy transport will remain ICE for quite some time).

        Exception: if someone works out how to economically make biodiesel from algae, all bets are off :).

  2. Lets not kid ourselves that this will necessarily be a social utopia. Consistent with most aspects of the new feudal society, there will be two networks the one for the wealthier participants in the economy and the one for the rest that was left behind.

    Big walls or tunnels will prevent those too poor to access the system disrupting it?

    • Sour grapes, mate. Driverless cars/vans will be a far cheaper means of providing “public transport” than the status quo.

      Like poor people getting mobile phones, this has scope for a new wave of democratisation of the benefits of advanced civilisation.

      People from the left of the political spectrum get all sour grapes about those “left behind” by automobility and auto-based urban development, when in fact a huge leap forward was secured by those things, relative to the preceding status quo, where only the very wealthiest few percent kept horses, a coach, drivers and stablemen; and where only the very wealthiest few percent lived in separate homes.

      The next wave of mobility technology – driverless cars – has the potential to FURTHER democratise mobility like the mobile phone has democratised communications. What person in their right mind today would say that “we need to build more telephone booths to provide poor people with communications”? It is just as stupid to say we need more public transport investment to provide mobility for them.

      • It wasn’t a problem for all the displaced farm labourers and domestic servants in the past, was it?

        They had somewhere to go where their (lack of) skills and knowledge were still relevant (factories, etc).

        Within a generation, two at the most, just about every manual or low-cognition-skill job you can imagine will be able to be performed more efficiently by a robot.

        A generation or two after that, I expect even most “knowledge worker” jobs will be doable by robots.

        Most low- and mid-level managerial jobs will be collateral damage – simply made irrelevant rather than replaced.

        This is not just a repeat of the industrial revolution. There’s going to be nowhere for most people to go.

        We already have a situation where the top percent or so suck up most of the productivity of the bottom 99% and many (especially on the Right) feel they are morally justified in doing so (using a circular “they wouldn’t be earning it if they didn’t deserve it” argument). Imagine what it’s going to be like when there’s simply nothing productive the bottom 90-95% _can_ do because of robots. Today’s rhetoric about “makers and takers”, and the like, will be amplified orders of magnitude.

        I’ve posted this before, but it’s worth doing it again:

        I try to be optimistic in life, but I struggle to see how our future holds a scenario better than a return to fuedalism (albeit under the guise of “democracy”).

      • DrSmithy,

        I’m sure you are aware that the future you talk about, is about now.

        And tech bubbles now are largely about enabling the elites to dumb down the outcomes of technological development all the while keeping them firmly in control of that outcome. The added benefit of this process is it also gives the elites the money to buy their mates useless technology at ridiculously high prices.

      • Fair point PB – but I’m seeing technology move more towards entrenching feudal status not unwind it theses days. In the end while we still have democ it’s up to us…

  3. Indeed. A Rand vision of society.

    Post hoc justification are not valid. It’s a politicians trick

  4. Without the need to pay for the labour of a driver – a taxi ride in a driver less cab would be extremely cost effective.

    Bring on the electric driverless cab company.

    Or a Go Get car without driver.

    Cashless as well.

    • But wages will be so low, due to mass unemployment caused by the innovations, that the cabride may still be comparatively expensive! Displaced manual workers will find alternative jobs, sure, but at what reward and how many displaced cab drivers will be in that competition?

      • It wasn’t a problem for all the displaced farm labourers and domestic servants in the past, was it?

      • @Phil – didn’t most of them move to the cities to then take place in the industrial revolution. The point being that automation represents a problem when there is nowhere else (market or location) for people to find work once it is completed.

      • Domestic servants are an odd example: in Edwardian England many employers only cut back their staff when the supply dried up, in part because other jobs had already become available,

        For example, despite the technology having been available affordably for decades, at at least one of their homes, the family of Virginia Woolf only installed toilets because chambermaids were too hard to find and retain.

        Many other labour saving devices already existed, but were only adopted because employing servants was too difficult – few servants were actually displaced.

      • The problem Phil is that the globalised financial and corporate oligarchies have the perfect systems in place for the extraction of the majority of gains in prosperity from technological advancement. Look at GMO foods and the supposed end of global famine. Look at the famous prediction of a (20?) hour working week thanks to information and communications technologies. The windfall gains from technology over the ages are mostly captured by the rich and powerful, as is human nature.

        Not that I’m anti-technology or diasgree with the “democratisation” of technology, and I agree with you on those points on balance, but I don’t subscribe to the utopian vision – because the empirical evidence so far is that the utopia never arrives, despite the perennial predictions of it’s arrival.

        If you listen carefully to the fringe views of people like Chris Martinson and Paul Craig Roberts, it is cheap energy that has exclusively led to the global rise in living standards. Technology is obviously a cause of that ability to exploit energy, but it is also to a large extent the pleasant side-effect of cheaply available energy, and as long as there is a threat to our ability to cheaply exploit resources and energy, there is ultimately a barrier to the world’s poor accessing the benefits of technological advancement.

        In summary, global financial orligarchy and dwindling net energy returns are two massive blocks to an otherwise logical and inevitable progression of living standards based on technological advancement. But that’s just the pessimist’s view 🙂

    • The big problem for Tesla is that some of the major car manufacturers are now taking EVs seriously, and they’ve got very deep pockets. BMW’s i3 is probably closer to an affordable mainstream EV that the Model S, while Tesla’s affordable model (aka Gen 3) is still some years away.

      From an article in The Oz:
      “Pricing for the BMW i3 is expected to start at about $60,000, whereas the Tesla Model S is tipped to be priced at about $200,000.”

      • I would assume not, but they have very, very deep pockets and also depend heavily on battery technologies.

      • Australian prices aren’t worth quoting; they’re not representative of anything except Australia’s small market and bat-shit insane parallel import restrictions. In the US, the BMW i3 is $41,350. Model S is $69,900. This makes sense considering they’re not the same class of vehicle; the i3 is a practical hatchback, the Model S is a sports coupe.

      • Fair point, but in the US most of the Model S’s sold are the higher end models that are pushing $100K USD.

        My point remains, BMW are closer to having an affordable EV with wide appeal than Tesla.

  5. I have a question about driverless cars. Who’s liable in a collision? The owner, the passenger, the manufacturer?

    • When it comes to liabilities, lawyers always go for the one with the deepest pocket.

      The bigger issue lies with how ‘safe’ the system is against tampering, and whether human will be allowed to drive afterward. The tech ‘utopia’ can become dystopia, A cyberattack have the potential to kill everyone on the road simultaneously, a government can use the system for targetted assassination. The potential for mischief is endless.

    • All liabilities are waived ofcourse, its part of the honey trap AB. In exchange for not charging you a licence and rego you’ll give up any right to seek damages.

      Its a dystopia not utopia in the hands of the corporate lawyers. We need more engineers and less politicians/lawyers.

      • @AB no they won’t but you won’t get squat. It’ll be an all mighty lawyer circle jerk bankrolled and backstopped by government collecting your taxes and paying them out to the corporate re-insurers that will effectively own the governments via technocracy…

      • There is no substantive case law on this – I looked at aircraft auto-pilot liability in the past, and there is no Australian precedent for product liability or tort. Perhaps the analogy will be with animals (a type of independent intelligence) and whether the owner of the animal is responsible for damage caused by it (See Trigwell v SGIC). Ultimately though, parliament will intervene through regulatory control, like CASA in relation to auto-pilots.

      • @casewithscience – thanks for the info. I can’t help but think that manufacturers are going to be setting themselves up for a lot of trouble unless they are granted very strong legal protection.

      • Or they could use the defence so beloved of the manufacturers of small arms:

        “Driverless cars don’t kill people….people kill people”.

        Frankly, I think I’ll stick with Star Wars and Bladerunner…they seem so much more real than this pie-in-the-sky bullshit.

        Just as an aside…can you visualise Beijing, Kuala Lumpur, Bangkok, or New Delhi with driverless cars?

        Comedy central.

      • Sigh… some Australians need to get out more.

        If self driving cars are ‘pie in the sky bullshit’, I suggest you have a pretty low bar for what qualifies as bullshit. The last time I saw a fully autonomous self driving car on a public road was… 2 weeks ago. Created by a company with over $50 billion in cash and a $400 billion market cap.

        And they already have a better safety record than humans, while on current public roads (not test tracks), with standard California traffic present.

        Anyone who thinks autonomous vehicles aren’t enivtable is akin to a farmer who just saw a Model T Ford and swears up and down that it could never replace horses. Driverless trucks are already being used in the Australian mining sector. My bet is that once the safety record is fully established, insurance companies will pretty much be forcing companies to go driverless.

        The next decade is going to shock the bejeesus out of a lot of people.

      • Onya, Duds….you saw a driverless car.

        Wow….I saw a Concorde once …..shot right past me while my plane was sitting on the runway apron waiting for the annoying little thing to get out of the way.
        Remember when we were all going to be flitting from London to New York in an hour or so in supersonic aircraft?
        That worked out well, didn’t it?
        By the way…those driverless trucks in the mining industry aren’t really driverless ….they are remotely controlled by….wait for it….humans.
        They are about as driverless as a US drone.
        Not that a rigidly controlled mining site bears any similarity to the chaos of everyday traffic, mind you.
        Maybe you need to get out to some mining sites a bit more often.
        Nice red herring, though.
        The only people who are going to get the bejeesus shocked out of them in the next decade are those who think that all the answers to our problems will be solved by increasing amounts of technology.
        And you can take that from one who is up to his ears in it.
        But, mate, keep on dreamin’.

  6. So having trillions of dollars evaporate in failed enterprises like is worth it? For Larry Ellison maybe

    • Yeah, tech bubble 1.0 really didn’t give us much. Not much has survived from that era except maybe Amazon.

      • Yup. Amazon/Yahoo and yahoo is arguably earlier.

        All the .crash brought us was a massive over-capacity in server infrastructure which nobbled the earning of HP/DEC(RIP)/SUN(RIP)/IBM/Dell for half a decade before cloud storage took off.

        No bubbles are worth they are ipso facto misalocation of resources, and just because it has been the fad for the last 30 years to get in on the new bubble doesn’t mean we’ve figured out a better way to invest, it just means we’ve become more stupid with our money — which completely inflationary hence the constant misalocation of it…

        EDIT: @Lorax LOL I’m with you there man yahoo sux, altough I think they’ve been picking up quite a bit of the search market last 2 years.

      • It gave us the world’s most disastrous merger.

        Unfortunately, the growth and profitability of the AOL division stalled due to advertising and subscriber slowdowns in part caused by the burst of the dot-com bubble and the economic recession after September 2001. The value of the America Online division dropped significantly, not unlike the market valuation of similar independent internet companies that drastically fell, and forced a goodwill write-off, causing AOL Time Warner to report a loss of $99 billion in 2002 — at the time, the largest loss ever reported by a company. The total value of AOL stock subsequently went from $226 billion to about $20 billion.[57]

  7. “A family of four wants to travel from New York to Chicago. They have dinner at home, climb into the vehicle at 9pm, watch a movie projected on the windscreen, and then go to sleep in their fold-flat seats, waking up at their destination the next morning.”

    this is ridiculous, why not utilise the existing and much more energy efficient rail-lines + car rental services at destination? They will be able to sleep in more comfort and it would cost them less.

    • this is ridiculous, why not utilise the existing and much more energy efficient rail-lines + car rental services at destination? They will be able to sleep in more comfort and it would cost them less.

      I LOL’d at the idea of a family of four sleeping comfortably in (presumably) the same car that dad drives to work every day.

      • Why can’t we already have fold-flat seats for everyone except the driver?

        Because there’s not enough room in the typical car to do it ?

        That’s before getting into the other issues that might arise like road noise, snoring, farting and a need for toilet stops (will anyone be allowed to make the car pull over ?).

    • Rail lines and car rentals WON’T be more efficient, or cheaper, as a SYSTEM. There is nothing to match “door to door” as a system element.

    • In a world with ubiquitous automated vehicles, private ownership will be impractical and rare. Rather, the family in question would check-out a “sleeper” vehicle for their overnight trip, and then switch to a touring or 4wd or SUV model at their destination, depending on the nature of their trip.

      Automated vehicles are coming, and their complete takeover is going to happen sooner and faster than most are expecting. A lot of the criticisms levelled at them currently are grounded in blinkered views of the way things are now.

      • In a world with ubiquitous automated vehicles, private ownership will be impractical and rare. Rather, the family in question would check-out a “sleeper” vehicle for their overnight trip, and then switch to a touring or 4wd or SUV model at their destination, depending on the nature of their trip.

        You have been able to get 99% of this scenario already for years with the combination of taxis, planes & trains, and vehicle rental.

        Yet most still own cars, because a HUGE aspects of car ownership is the “ownership” part.

        There is no reason I can see to expect autonomous vehicles to change this.

        I expect we will see autonomous vehicle control on highways and other limited-access roads as reasonably common in about ten years, and probably mandatory another ten to fifteen after that. I will be surprised if autonomous vehicle control is allowed in remotely pedestrian-heavy locations without specific legislation around liability.

    • “this is ridiculous, why not utilise the existing and much more energy efficient rail-lines + car rental services at destination? They will be able to sleep in more comfort and it would cost them less.”

      Why not utilise the much, much more cost effective airlines? Even with $100 a barrel oil airlines are still significantly cheaper than rail over such a distance and I strongly suspect will remain so even as the price of oil rises in the future due to increasing aeroplane efficiency.