« ATOM Award of the Month, July 2017 | Main

Comments

fatcat

One obvious way to repurpose [semi-]deserted malls in the bay area is to create a 24/7 “Spa/Gym” with private bed/cabin similar to some Japanese airports. Technically not residential yet in direct competition with the rental properties in the aria.

Geoman

This is a pretty bold prediction.

I don't think you are necessarily wrong, but the more complex the prediction string of if/then, the more easily that chain can be disrupted by a unknown and unplanned repercussions, such as changes to human wants, needs, and behavior.

For example, if the cost of driving, in unproductive time, gasoline, and car ownership, drastically plummets, one effect might be that cars get bigger and more luxurious. Instead of an electric automated car, why not an electric automated motor home? I mean, I can sleep, use the bathroom, even sit at a desk and do work while I commute. In such a case, might not larger parking spaces be needed? Might I not be willing to drive even more? Perhaps the motor home becomes a part of my house - plugs in on return and becomes my home office? maybe the suburban mall becomes the "plug in" point for a vast array of motor homes, with the retail space converted to homes, and the open interior simply a common area.

Maybe it is the office buildings that are torn down to make way for more parking lots? And "office" is just a collected of hallways attached to mobile homes?

I sounds weird only because it is so different, but my point is things can change even more than you are imagining, and in even stranger ways. I can envision futures with no parking, or even more parking. With malls, or no malls.

Geoman

On your graph above, I would imagine retail and manufacturing to have continued their rapid growth since 2010 when the graph was created.

But I wonder about construction - how is it measured exactly? Gross value added is the measure of the value of goods and services produced. The graph suggests our construction is "less" valuable (less profitable) than it once was. But it occurs to me this may be a function of low interest rates - it is a well know phenomena that housing prices rise as interest rates fall - the homeowner can afford "more" home for the same monthly payment. So the productivity loss may reflect this more than anything. I guess just noting that construction is a different sector than the others on the graph, and may respond quite differently.

In addition land prices might have an outsized impact - productivity was much "higher" when land prices were cheaper. My construction cost in Los Angeles may be less, but the land cost is much more....so I show a negative growth in productivity?

Kartik Gada

Geoman,

Perhaps. But I am not so sure about bigger cars at that point. If most automobile transportation becomes autonomous, then there is a good chance that two-seater pods emerge, in order to not have as much unused capacity. Individual legroom may be more, but the total vehicle size may be less.

Existing lanes, turns, driveways, etc. are to embedded, so I don't think vehicles too large to use those will be widespread. Not to mention that most countries don't have that much room, so the volume of smaller vehicles will just be more (and hence have inherently lower price).

Overall, retail malls are being vacated, and the parking lot land represents dead capital. The huge amount of space devoted to retail in the US represents massive misallocation that will have a dislocating correction much more drastic than in countries that did not devote this much land to retail to begin with.

Kartik Gada

Geoman,

The interest rate point is interesting, and does seem to have a correlation with the timeline of construction productivity declines.

In that case, moving to ATOM-DUES is even more urgent, for construction productivity cannot rise until then.

The land price factor may be less when so much construction is redevelopment of land with old structures on them (i.e. demolished). For example, NYC is undergoing a huge skyscraper boom right now, the likes of which NYC has never seen. But it is still far more expensive to build a skyscraper in NYC vs. Shanghai, even after accounting for different labor wages.

The linked article from The Economist has more information on construction productivity declines.

Geoman

The one thing the ATOM does not do is create more land. Hence construction lags.

But the ATOM goes around the problem. The internet creates more land by making everything else need less (telecommuting, Amazon.com). The ATOM creates more land by removing cars from the road via automation, and doing away with car ownership. Or it could create more land by making it attractive for people to occupy the land with automated motor homes.

Or all of it, and a couple more things we haven't even dreamed of yet. Maybe automated drones flying us to and fro. Maybe Musk like tubes underground. Maybe gondolas like La Paz. The ATOM tries to do everything, and surges into every crack and crevice.

Thinking like the ATOM means accepting that everything is possible, anything can be disrupted, and nothing lasts, even our dreams. Last week's sure thing is discarded for next weeks promises.

Kartik Gada

Geoman,

People will have the option of calling for a pricier full-size vehicle that is large enough to lie down in, while others may want an economical two-seater pod. But overall, this disruption is continent on self-driving + electrical technologies making on-demand more economical than car ownership even for the lower-income car owners of today.

But yes, it is important to note that the ATOM relentlessly probes every crack and crevice, as you point out. That is why this construction productivity loss is huge. The Economist's article says that if it had merely kept up with the broader economy's productivity gain, annual world GDP would be $1.6T/year higher. Hundreds of millions more would have clawed out of poverty, vs. now.

One other area where the ATOM creates more is VR. People who live in a cramped little apartment can nonetheless put on the goggles and roam around 'their' custom mansion. They don't really own anything and can't show it to anyone out in the real world, but the dopamine rush to the brain is probably the same.

Sunny

Kartik - I am surprised you still have not included Blockchain in any of your posts as a disrupter. Matter of fact, I am curious to read your thoughts/opinion on this subject. I was dismissive of the technology until recently but have come around to its disruption, reason being its adoption (atleast the ledger part) by reputable and large multinational companies and banks. So the question is, first of all do you buy the disruption aspect of it and if so How pervasive and disruptive will be the impact of the associated crypto currencies that leverage Block chain, such Bitcoin and Ethereum.

Kartik Gada

Sunny,

I wrote about Blockchain here :
http://www.singularity2050.com/2016/09/deep-learnings-atom-effects.html

computer_guy524

https://hardware.slashdot.org/story/17/08/30/2152209/new-t-shirt-sewing-robot-can-make-as-many-shirts-per-hour-as-17-factory-workers

There ya go. Now, robots will take the jobs of even Chinese citizens, not just Americans :-) .

Computer vision is amazing.

Drew

I agree with Geoman in thinking that the construction productivity issue is a pricing issue more than _decreasing_ productivity. I can understand the Economist's belief that fragmentation and government contracts cause issues, but that is also the issue with eds and meds and even with those we don't see _decreasing_ productivity.

I did a quick Google search but did not find much useful information on the issue.

Kartik Gada

Drew,

Well, in a place where housing prices have risen greatly, but construction wages have not (like the Bay Area), isn't the high price a much higher profit margin for new residential and office construction?

In the Bay Area, builders are fighting tooth and nail to get projects approved, and the city councils only approve a small trickle out of the pipeline. Under a free market, the level of construction in the Bay Area would be five times what it is now, despite the low productivity of construction.

We keep hearing about how America can no longer build the same bridges and highways that it could build 50 years ago, and the productivity decline would seem to be the root cause of that.

It does appear that the same productivity metrics used for manufacturing should apply to construction, despite the interest-rate component (which again could be favorable given the higher prices inherent to lower interest rates).

But I am willing to be convinced otherwise.. Especially since that makes the ATOM-DUES even more necessary than before (as I said above).

Drew

Ah, that may be the issue. I was thinking productivity in terms of simply the physical construction itself. If they are measuring the entire process including design, permits, environmental studies, re-design, etc. and delays, then I can see a decrease in productivity. Simply comparing the original construction time and costs for the Twin Towers vs. their replacement for an example. I have my doubts that they are including the design time and legal costs for the productivity measurements in other industries, but those industries probably also have a lesser proportion of total costs coming from those costs. However, that isn't clear from the article. The comments on the article do call out the bespoke nature of construction projects and greater safety tradeoffs as well.

Geoman

I wonder if they added government productivity to the chart what we might see!

Also, you could argue that the replacements for the Twin Towers, despite taking longer, are in fact "better" in many ways. Safer, longer lasting materials, taller, etc.

Kartik Gada

Geoman and Drew,

There are certainly some improvements in skyscraper construction technology, which is why NYC is seeing an unprecedented boom in skyscrapers (many of which are much thinner for their height than one would expect).

But this still could be amidst an overall decline in productivity. The costs in a non-democracy like China or the UAE are much lower, beyond just labor. The legal and government costs in the US are prohibitively high, and I think the chart above takes the apples-to-apples productivity stats across industries. Construction, like education and healthcare, is very entangled in government.

Regarding government productivity itself, that is certainly terrible. It is one of the core disruptions AI is going to deliver, IMO. AI, among other things, makes tax avoidance more accessible beyond just the biggest corps and ultra-wealthy individuals. The tax regimes of high-tax countries have no chance of keeping up with that.

A.M.

"It will be cheaper for almost everyone to take the on-demand service all the time, than to own a car outright or even take the bus."

If the buses are autonomous too, then buses will retain a cost advantage, due to sharing a vehicle. There will be a spectrum of vehicle sharing, with accompanying prices. Perhaps mostly shared rides during rush hour, and more solo rides off-peak, depending on the ratio of road capacity to demand - denser cities would have a higher balance of high capacity transit. Road capacity itself may be diminished (e.g. narrowing streets due to increases in spatial efficiency of vehicles).

Regarding an unmatched skyscraper boom in NYC, that's a fiction. People just lose their shit nowadays any time something is built. NYC was adding 100k units per year in the 1920s (according to Ed Glaeser), as Houston recently was. One reason why: "40 Percent of the Buildings in Manhattan Could Not Be Built Today." *

Personally, I think autonomous mobility is closer, based on the rapidly increasing investment flows. There's already small rollouts of autonomous buses. I'm predicting initial consumer availability some time in 2019, on distinct routes. Having autonomous cars operate on the top 1% of roads by usage alone would be massively beneficial.

Geoman, the mobile homes thing is interesting. I could see that happening - but the aggregate amount of parking is so vast, ~6-8 spaces per existing car, due to government mandates, that mobile homes wouldn't swamp that. Plus, the politics don't align - when people cease to own cars, they won't have the incentive to demand that government requires all buildings to have parking, like they do now. So our existing parking will more than meet mobile home growth.

* https://www.nytimes.com/interactive/2016/05/19/upshot/forty-percent-of-manhattans-buildings-could-not-be-built-today.html?mcubz=0

Kartik Gada

A.M.,

Regarding an unmatched skyscraper boom in NYC, that's a fiction. People just lose their shit nowadays any time something is built.

Those are two separate things that are not mutually exclusive. The NIMBYism is extreme, of course, but there is nonetheless a wave of NYC skyscraper construction that is more than it ever has been. Note that this is not quite the same thing as building raw square footage, which is usually met with mid-rise buildings that are probably being obstructed by NIMBYism.

As per this link, 18 of the 25 eventual tallest in NYC are 2014 or later completions :

http://skyscraperpage.com/diagrams/?cityID=8

35 of the tallest 50 are 2011 or later completions. The Empire State will soon be all the way down to #8 in NYC, despite being #1 until 2014.

If anything, the ATOM has adapted to NIMBYism by making unusually slender buildings possible, so that tiny footprints get more value.

Similarly, note that 16 of the 25 tallest currently under construction across the US, including all 5 of the top 5, are in NYC :

http://skyscraperpage.com/diagrams/?searchID=78438204

This, of course, reflects far more poorly on the rest of the country than it reflects favorably on NYC. San Francisco only has 3 underway, even though it is even more expensive than NYC.

If the buses are autonomous too, then buses will retain a cost advantage, due to sharing a vehicle.

Well, the busdriver's salary is a very small part of the bus operating cost. The proportion for an Uber driver is far more (Uber takes only 20-30% of the fare as its cut).

But yes, there will be a spectrum. Carpools of 2-3 riders will form dynamically, even if some deride that as a 'microbus'. Most rush hour traffic could easily form into dynamic, on-demand 2-3 person carpools. Niches like an entire camper/trailer as Geoman describes, are also possible.

A.M.

Ah, I was wrong about the skyscraper boom! Thanks for the correction.

I wonder why it's occurring - maybe between high land prices, mandatory highly skilled union labor and difficult permitting, that you might as well build as much as possible once you get everything together.

Well, the busdriver's salary is a very small part of the bus operating cost.

Actually, driver costs are a very high portion of bus costs:

"Driver labor, and related time-based costs, are the dominant element – often 70% or more — of transit operating budgets in the developed world. The only exception is a fully automated driverless service, such as Vancouver’s SkyTrain." - Jarrett Walker, a leading transit planner.

A manned mini-bus (say 10-15 seats) doesn't make much sense, because it can't carry many people at peak times while you still have to pay for a driver. With autonomous buses, having multiple smaller vehicles doesn't cost much more, and it allows you to run service more frequently (say 15 buses per hour instead of 5) which increases ridership.

It also means you can tailor your stops along a fixed route to the riders aboard, or at least have closer stop spacing, with less of a time loss to other riders.

http://humantransit.org/2011/07/02box.html

Kartik Gada

I wonder why it's occurring - maybe between high land prices, mandatory highly skilled union labor and difficult permitting, that you might as well build as much as possible once you get everything together.

That seems to be the case, since many of those buildings are far thinner than even appeared possible before. They are making the most of what little land has become available, using new technologies that enable buildings 12 times taller than their width despite heights above 1000 feet.

That is still better than San Francisco and Los Angeles, where very little is being built at all.

Actually, driver costs are a very high portion of bus costs:

That is surprising. Maybe busdrivers are paid far above market due to it being a unionized job. Otherwise, I can't see how that can be the case, given that the bus itself is just 3 mpg and over $300,000 in purchase price.

Autonomous driving will make most contemporary buses obsolete, even for poorer people, due to the point-to-point advantages. Although ATOM-DUES is needed as well to ensure that almost all can afford on-demand cars.

Kartik Gada

Here is another visual of how much is being added to NYC, 2014-2021 :

https://static.dezeen.com/uploads/2016/06/new-york-visualisation-skyline-2020-skyscrapers_dezeen_936_0.jpg

Anthony Scalzi

The under-construction 111w57th is even skinnier than that. At 65 feet wide and 1438 feet tall, it has a ratio of 22. It happens to be being built with nonunion labor.
Another proposed tower at 262 5th ave is 1050 feet tall and 50 feet wide for a ratio of 21. Both towers sit on lots that were formerly just a couple tenements. For zoning purposes though, developers still needed to buy up the air rights many neighoring buildings. These airrights tend to cost anywhere $200-300 per buildable square foot. A change in zoning could eliminate those costs overnight.

Fun fact-the Empire State building would be pretty much impossible to build today under current zoning since it has a floor/site area ratio of 25 and the current limit is 15 in much of midtown Manhattan. This is a big issue for developers that want to rebuild older office stock, since if they tear down an older overbuilt building, the new building will have to be the smaller size stipulated by the zoning.

A common justification given for these zoning restrictions is that transportation infrastructure couldn't handle the demand from larger buildings. That goes back to poor productivity for infrastructure projects.

Bob Reisner

I strongly agree with the forecasted strength of the disruptions but I do think there is some additional points that might make for a faster arrival.

Batteries

The biggest missed point is the residual value calculation on ICE vehicles that businesses and consumer financing entities will make once it becomes certain that battery cost will achieve a $50 KWH market clearing level. Let’s assume that, consistent with your chart, the $50/KWH battery cost will be achieved by 2024 with this certainty seen somewhere around 2020-2022.

One the $50/KWH level becomes certain, there will be a rush to build dozens of battery factories the size of the Tesla gigafactory and automobile manufacturers will rush to move the vast majority of production to an electric platform.

The auto companies will understand that beginning 2024, ICE vehicles will be hard to sell, requiring lower prices and ever higher marketing costs. The reason is that ICE cars sold from 2022 on will not have the normal expected financial life. And it will get progressively worse as battery costs drop to the $25/KWH level in the late 2020s.

EVs will have half the cost of an equivalent ICE vehicle when fuel, repair and maintenance costs are computed. The simpler EV has a longer service life which makes it even more attractive.

So businesses will defer ICE purchases to get the less costly EVs. Banks and finance entities will shorten the projected financial life of ICE vehicles which will substantially raise consumer monthly costs. And the consumer will only choose an ICE vehicle as a last resort because they will know that the ICE vehicle will have little value when paid off.

And EVs will have great residual values because of their operating cost structure and longer projected life. Which means lower monthly costs for initial purchases which puts even more downward pressure on ICE vehicle prices.

ICE will be in a sharp downward production spiral in the mid 2020s and almost gone by 2030 in first world countries.

Autonomous Vehicles

Autonomous vehicles are going to be here way before 2032.

Google /Waymo has millions of autonomous miles on real roads today. And billions via simulation. At the beginning of 2017, a driver had to touch a wheel only once per 5,000 miles. And moving quickly to once in a 100,000 miles in early 2020. We will see consumer sales of fully autonomous Self Drive Vehicles (“faSDVs”) in the early 2020s.

And they will also have a very fast adoption.

Once the true availability date is known (say 2022), at least two years earlier purchase behavior will change. Car rental, Taxi, Rideshare, sales, and service companies will defer human driven vehicle (“HDV”) purchases around 2020 or earlier. The productivity impact on businesses of the faSDV is huge enough that if a company has too many HDVs in use when faSDVs become widespread, they might be so uncompetitive as to go out of business. It’s a binary event, the companies without faSDVs die.

Selected consumer categories will also defer purchases of HDVs because the value of the faSDV is so great. Examples include commuters, families with two cars, elderly, and the physically impaired. And of course, the coming availability and then real availability of faSDVs will crush the residual value of HDVs.

And the story becomes like the battery one above. Some buyers will not buy a HDV at any price. HDV prices fall and marketing costs rise for everyone else. Declining residual values means higher monthly HDV costs and initially faSDVs will have lower costs because of longer life spans.

Auto manufacturers get crushed again and faSDV adoption accelerates at a high level.

============

Retail is going to be far worse than projected here. And construction is already moving quickly.
Watch this 30 story building go up in a bit over 15 days including interior finish: https://www.youtube.com/watch?v=ajlUVSiUvWg
Watch this 57 story building go up in 19 days: https://youtu.be/AhLk7L1B_fE

I’d write more on these two but this comment is waaay too long!

Kartik Gada

Anthony Scalzi,

Wow. A skyscraper over 20x taller than its width will be a revelation. It gets around many of the horrid zoning and permitting problems.

I am surprised that an alliance between renters and construction companies cannot sue for removal of some regulations.

The 'transportation cannot handle demand' excuse used by the obstructionists is a poor one. More supply, as long as the ratio of office to residential is maintained, does not cause more traffic, since more people can live closer to their work. Even weaker is the 'it changes the character' excuse, even though whenever something new is built, no one even remembers what was there before. It goes to human hardwiring, where weaker people are afraid of change, as their survival is dependent on others. More self-sufficient people are more enthusiastic about disruptive progress.

Kartik Gada

Bob Reisner,

Remember that for autonomous vehicles, the real benefits are when over 80% of all cars are autonomous. The first 10% doesn't do much. Hence, I don't think this level of diffusion will happen before 2032. Remember, this is a technology that has been hyped in the past, only to miss deadlines. Progress is happening, but average people are slow to adjust, and regulators even slower.

Plus, the BSkyB construction videos have been discredited as just demos. The various videos of the time-elapsed 30-story building are almost 5 years old, yet they have not demonstrated any volume production, otherwise they would have built hundreds across the world by now and there would be wholesale deflation of existing multi-story building prices. Plus, they announced a 200-story 'Sky City' back in 2013, only to deliver nothing. Broad Group is a Chinese company that has no clue about marketing, PR, or branding.

Things are moving quickly, but not *that* quickly.

Bob Reisner

Kartik Gada,

When faSDVs become available at HDV prices will anyone want the 'obsolete' HDV? The HDV will lose all its resale value when faSDVs become available in volume. And Taxis/Rideshare/Rentals will not buy HDVs at any price. faSDVs will be a production capacity issue, not demand. And manufacturers have the same problem ... no one will buy the HDV once the faSDV is available.

The transition will be very quick.

Kartik Gada

Bob Reisner,

Note that of the 100M+ HDVs in the US that are under 6 years old, many were bought with debt. If those lose resale value quickly, most people won't even have the money to buy a new faSDV. They will have to keep using the HDV for a few more years, whether they like it or not. The average household can't just take a $20,000 hit to upgrade.

The three stages are :
1) When does a faSDV achieve complete cost parity with an HDV?
2) When do faSDVs achieve 80% or more penetration, enabling VtV communication?
3) When do faSDVs (along with EV capabilities) make on-demand more efficient than ownership?

I say it takes until 2032 for Step 3 to complete, partly because hundreds of millions of people worldwide will be left holding the bag on their HDV that they can't afford to ditch. Sooner would be better, but I wouldn't bet on it. Remember that faSDV progress 2007-present was much slower than they claimed it would be in 2007 (when they acted like it was just around the corner).

A.M.

"I am surprised that an alliance between renters and construction companies cannot sue for removal of some regulations."

Renters act like NIMBY homeowners when they're under rent control, because they're (mostly) insulated from rent increases. And in California, the construction unions seem a lot more powerful than developers and landlords. Our Democrat governor, Jerry Brown, tried to pass a law allowing more housing without special zoning approvals, but the bill went nowhere because the construction unions didn't like more building being allowed if high wages weren't mandated [1].

" They will have to keep using the HDV for a few more years, whether they like it or not. The average household can't just take a $20,000 hit to upgrade."

They can't afford not to use them, because the marginal cost of driving your ICE will be higher than the total cost of using an SDV, which will be fleet owned and operated.

I recommend this white paper on SDVs: https://static1.squarespace.com/static/585c3439be65942f022bbf9b/t/591a2e4be6f2e1c13df930c5/1494888038959/RethinkX+Report_051517.pdf

[1] http://www.spur.org/publications/urbanist-article/2017-01-11/bye-bye-right-housing

Kartik Gada

A.M.

They can't afford not to use them, because the marginal cost of driving your ICE will be higher than the total cost of using an SDV, which will be fleet owned and operated.

Yes, but that takes a while, since the sunk cost of the HDV has to factor into the equation of what is truly cost-parity. The average new-car transaction price in the US is $34K, or over 60% of the annual household income. If 100M US cars are under 6 years old, that is a lot of loss-swallowing that has to be done, particularly if debt-financed. 80% penetration of faSDVs takes until 2032.

If it really does happen more quickly, all the more reason why the ATOM-DUES is urgently needed, as the sudden devaluation of 400M HDVs worldwide, many of which are debt financed, can cause a recession.

Renters act like NIMBY homeowners when they're under rent control, because they're (mostly) insulated from rent increases.

In California, very few are under rent control. It is somewhat common in SF and Berkeley, but not in Silicon Valley.

The construction unions are villains, as you point out, and they are part of why construction productivity has gone backwards over time. One more reason to cheer greater automation and factory-built modular construction, as their artificial union wages experience the same disruption that OPEC and taxi medallions saw.

Fatcat

Yes, but that takes a while, since the sunk cost of the HDV has to factor into the equation of what is truly cost-parity

Sunk cost will weight on the mental inertia. However, sucking tov the status quo because of the money spent is a sync cost fallacy

https://youarenotsosmart.com/2011/03/25/the-sunk-cost-fallacy/amp

https://www.google.ca/url?sa=t&source=web&rct=j&url=https://en.m.wikipedia.org/wiki/Sunk_cost%23Loss_aversion_and_the_sunk_cost_fallacy&ved=0ahUKEwinvqLX4JbWAhVs7IMKHXjtC1IQygQIcDAN&usg=AFQjCNGG_q1sH1VcODTPGsvjwDMGCmXLJQ">https://en.m.wikipedia.org/wiki/Sunk_cost%23Loss_aversion_and_the_sunk_cost_fallacy&ved=0ahUKEwinvqLX4JbWAhVs7IMKHXjtC1IQygQIcDAN&usg=AFQjCNGG_q1sH1VcODTPGsvjwDMGCmXLJQ">https://www.google.ca/url?sa=t&source=web&rct=j&url=https://en.m.wikipedia.org/wiki/Sunk_cost%23Loss_aversion_and_the_sunk_cost_fallacy&ved=0ahUKEwinvqLX4JbWAhVs7IMKHXjtC1IQygQIcDAN&usg=AFQjCNGG_q1sH1VcODTPGsvjwDMGCmXLJQ

Kartik Gada

fatcat,

Sunk cost fallacy is the reason why users will not ditch their relatively new HDVs for faSDVs. If anything, this delays the migration and ensures that this takes longer, not shorter.

Why would you think sunk cost fallacy makes it easier for the user base to take the $20,000 average hit and move to faSDVs?

Remember that the full vision requires multiple disruptions, so even if one is slightly faster than the others, it cannot do that much until the others catch up.

I still say 2032 for the above vision.

Bob Reisner

To: Kartik Gada

(Gada in normal type and my comments between /***/ marks)

Note that of the 100M+ HDVs in the US that are under 6 years old, many were bought with debt. If those lose resale value quickly, most people won't even have the money to buy a new faSDV. They will have to keep using the HDV for a few more years, whether they like it or not. The average household can't just take a $20,000 hit to upgrade.

/***/Not all HDV will be replaced immediately because USA production & sales capability is less than 20MM units per year. The faSDV move will compress a 10 to 12 year cycle to 5 to 7 years. Everyone will not move immediately but the vast majority will move much sooner than the normal replacement cycle. Some buyers will not move to a new vehicle but will end up with short term faSDV rentals or a 'timeshare' version of a rideshare faSDV. They might even save money even with a HDV disposal loss./***/

The three stages are :
1) When does a faSDV achieve complete cost parity with an HDV?

/***/faSDVs do not need to get to cost parity. They provide unique capabilities to taxi companies, companies with field personnel, elderly, disabled, those that shrink from 2 HDVs to 1 faSDV, long distance commuters, etc. Cost parity might be as early as 2022 but PRICE parity will not be for 5 to 7 years because high demand and shortage of production will allow for higher manufacturer price./***/

2) When do faSDVs achieve 80% or more penetration, enabling VtV communication?

/***/ 80% penetration might be late 2020s but no later than mid 2030s. But V2V communication is not that important. V2V helps with some edge conditions and traffic flow but not really a driver in vehicle automation. /***/

3) When do faSDVs (along with EV capabilities) make on-demand more efficient than ownership?

/***/For a substantial part of the population, on demand is viable on day 1 (think urban dwellers). Just a bit later as a way to eliminate suburban second & third vehicles. But there will always be consumer owned faSDVs. Demand peaks will always be an issue (by definition) as will the desire for some level of vehicle personalization and customization. A lot less ownership than now but still very substantial. Maybe even still a majority./***/

I say it takes until 2032 for Step 3 to complete, partly because hundreds of millions of people worldwide will be left holding the bag on their HDV that they can't afford to ditch. Sooner would be better, but I wouldn't bet on it. Remember that faSDV progress 2007-present was much slower than they claimed it would be in 2007 (when they acted like it was just around the corner).

/***/I've stayed away from world wide adoption predictions because faSDV adoption in some second world countries and almost all third world countries might be problematical. 2007 era predictions were all over the spectrum ... from never to tomorrow. My guide has been Google pronouncements and they have been off by only a year or two. The faSDV is an extraordinary product and the development speed has been astounding./***/

Kartik Gada

Bob Reisner,

80% penetration might be late 2020s but no later than mid 2030s.

This is what I am saying. 2032 is my prediction for the whole thing, as there are many components (including the e-commerce and real-estate component). Remember, the goal is to force real-estate repurposement.

It is not merely about V2V, but getting the majority of drivers towards on-demand, which requires both faSDV and electric capabilities to truly cause the migration of almost all current car-owners.

We aren't disagreeing by much. I am focused on when there is truly mass adoption of both faSDV and electric, leading to on-demand being 80-90% of all rides, leading to huge pressure to repurpose all the land wasted on parking lots and retail strip malls.

Even the first 50% won't do it. 80%+ penetration is needed, which is why it takes a while. The first 50% might take the $20K hit in stride, but the next 30% will hold out for a long time, even if the marginal driving cost per mile of an EV is lower. They stuck are in a trap for a few years due to the sunk cost and inability to just get out of it unless on-demand is really very cheap.

Geoman

When people own cars, they use them only about 4% of the time on average. The rest of the time we're in our houses and offices, and the cars are sitting idle. So in theory there are 25 times as many cars as necessary. 25 times more parking spots.

But that is not actually true.

If everybody's 4% at different times of the day, a dozen people could share a single car between them with no problem. The problem is that most of them want to use their cars at exactly the same times of day, before and after work.

So we happily pay for the idle time of the car, in order to get the reliability of the car always being available to us, and us alone, and not having to share.

Taxis have been around for 100 years, and they seem to have not reduced car ownership one tiny bit. I think it is bizarre to expect, just because we replace the cabdriver with a robot, or the petrol engine with a battery, suddenly taxis are going to rule the world. Why do you expect human nature to suddenly change?

So, no I don't think this version of the future is necessarily going to happen. It makes perfect sense mind you, and is totally logical, but human beings don't always act rationally. I think it will take a lot longer for this future to come about than most people think.

I think it is just as likely there will be even more cars on the road - automated driving means loads of elderly can get out and do more. It means children can cruise around town. It mean driving long distances will be even easier. if electric cars are cheaper, why not have 2 or 3 in the driveway? heck, one for every member of the family, including the dog, who can take himself to the park!

Kartik Gada

Geoman,

Taxis have been around for 100 years, and they seem to have not reduced car ownership one tiny bit. I think it is bizarre to expect, just because we replace the cabdriver with a robot, or the petrol engine with a battery, suddenly taxis are going to rule the world. Why do you expect human nature to suddenly change?

Because cost will be much lower. Uber did a pilot where a few people were allowed unlimited Uber usage for $120/month, and they did not use their own cars anymore at all.

Not to mention that the real savings of an EV happen from 150,001 to 300,000 miles of the car's life, which is hard to get to under private ownership (especially if the owner wants the electronics of the car to not be too obsolete by that time).

Plus, many parts of the world don't have nearly as much private car ownership as the US. This includes developed countries. The pressure to get more value out of urban and suburban land is latent, but real. Note the chart in the article about retail space per shopper. The US really overshot and will correct this error (at least in expensive big cities).

It mean driving long distances will be even easier. if electric cars are cheaper, why not have 2 or 3 in the driveway?

I think this is the very reason that on-demand will be the primary usage, rather than ownership. You yourself pointed out the advantages of Uber just recently. Now imagine Uber at 1/5th the cost via faSDV, plus tons more new land available from demolition of many strip malls and their parking lots.

I agree that a few hobbyists will own several cars, but the same goes for horse ownership in modern times.

We shall see.

Geoman

No comment on the electric automated doggie Uber? I was quite proud of that ridiculous leap of logic! It could be a cute little tiny car. It could drive up and whistle for the dog to come down and get in. Dogs love to drive around. They also love sniffing where other dogs have been. Win-win if you ask me.

Joking aside, I always think the future will be much weirder than we can imagine. So far it is certainly not what I expected as a kid - some things I thought would never change have changed, and others I expected to be long gone are still hanging about.

Kartik Gada

Geoman,

Joking aside, I always think the future will be much weirder than we can imagine. So far it is certainly not what I expected as a kid - some things I thought would never change have changed, and others I expected to be long gone are still hanging about.

Perhaps. But I think it is more about finding the true trendlines, rather than the 'decoy' trendlines. This blog has a very high success rate of prediction (over 80%), going back almost 12 years now.

It is also possible to quantify and to some extent rank how overdue a given disruption is. This does not enable a timing prediction, but does enable quantification of the speed and degree of eventual disruption.

Plus, there are many assumptions people make, that ought not be made. For example, I don't think humans will ever colonize space en masse. It will be AI, for reasons described elsewhere. Another example is the assumption that income tax is mandatory and unavoidable. I contend that it is not necessary after 2025 or so.

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment

Your Information

(Name and email address are required. Email address will not be displayed with the comment.)