The beginning of the end. Remember all of the old “B” sci-fi movies from the 50s that either started or ended that way? Usually with a shot of an atomic bomb exploding in the background. Sir Winston Churchill had a much better way of saying it. “Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning.” UBER has reached the end of the beginning. The world’s most heavily valued taxi service is about to become the world’s most heavily valued robotics firm.

UBER has been clear for years that Robots are part of their future. The question was WHEN will UBER will join the robot revolution. The answer is, “Real soon!” UBER’s wants to immediately replace drivers, and if they can they will start in New York. For UBER, soon is just not soon enough! Unexpected events have dramatically changed their schedule, and they’ve got to get robot cars on the street in 1-2 years.

UBER is faced with some very unique problems. We all know what UBER is, and almost everyone is aware that UBER is valued at over $60 billion. Far fewer are aware that UBER loses money hand over fist. UBER expects losses of $250 million in 2015. Any company that has never made a profit and is losing nearly a million dollars a day has to start identifying the issues and developing projects to cut these losses. You can be sure that this is exactly what has been happening at UBER. So, before we get to the end of the beginning, let’s go back to…

THE VERY BEGINNING: UBER launched in 2009, and the UBER App arrived in July of 2010. UBER was just a taxi service (named UBER taxi) and dealt with the same issues as every other taxi service. All taxis have three basic costs: cars, fuel and employees. There are always other costs, such as a garage and fines, perhaps even some advertising, but these are usually far less significant. When the UBER App launched, gasoline was $2.78 a gallon. In less than a year, it rose to over $4 a gallon and then bounced between $3 and $4 over the next few years.

In order to break even, UBER needed to control fuel costs. UBER started to buy new cars for their new drivers, partially to ensure a supply of drivers and partially to ensure they were driving fuel efficient cars. UBER even experimented with electric cars in China. But then, in late November of 2014, the price of gasoline unexpectedly fell below $3 for the first time in more than six and a half years. The price fell then fell, then rose (to $2.92), and then fell again and hit a lot of $1.84. Now, Iran is re-entering the international oil market and will add a million barrels a day of oil production. This could be a very rough ride, but overall gas prices are likely to continue their fall throughout 2016.

With the price of oil cut in half, three things happen immediately. First, the cost structure for UBER and every other transportation service falls. UBER and their competition (including the famous Yellow Taxis) will start lower their rates, perhaps starting a price war. That’s good for UBER, which can easily change fares, and bad for Yellow Taxis that require a very complex process that requires government approval. Second, if UBER can retain some portion of fuel savings, they can close the profitability gap. Third, and most importantly, cutting one of the primary operating costs in half makes the remaining two appear even larger. Especially, the cost of employees.

DRIVERS DRIVE COST: Drivers are now the primary cost of an UBER ride, and UBER rides are the most expensive in New York City. By one estimate, New York City UBER drivers make $27 per hour. For our calculations, we’ll round down to $25 per hour. Humans can only drive for so long before they need to eat or sleep or get to a bathroom. In New York City, a taxi is often owned by a garage, rather than the driver. When one driver completes their shift, they return to the garage and another driver starts their shift. Currently, most UBER cars are run by a single human. They might choose a longer shift, or even a 7 day work week, but they are not on the road 24×7. A robot car would only stop for fuel or for maintenance.  A robot car could stay on the road indefinitely, replacing not one, but several drivers.

While UBER has 14,000 cars vs. slightly less than 14,000 Yellow Taxis, many of the Yellow Taxi’s are run for 3 or 4 shifts a week. UBER plus Yellow Taxi’s are probably equal to 20,000 “24×7” cars. For the purpose of calculation, allow 2 hours a day for refueling and maintenance, so 23 hours a day and 365 days a year yields 8,030 working hours for an autonomous car. At $25 per hour, if that had a driver, it would need to pay an additional $200,750 annually. We don’t need to go much further to understand how UBER can become profitable, and take over the world’s car service market.

In order to replace these drivers, you need robot cars. If UBER buys 14,000 new cars, that allows them to replace all of their current drivers and take over the work of many Yellow Cab drivers. Let’s be very conservative and assume that the first autonomous cars that UBER buys will be $75,000… the price of a Tesla or premium gasoline car. UBER would pay about a billion dollars for new cars, but save $2.8 billion EVERY YEAR. Let’s toss in UBER’s $250 million annual operational loss, and a year after adopting a 100% robot fleet, UBER’s profit rockets to $1.5 billion. Since we loaded up all of the car costs into year 1, in year 2 profit would rise again to $2.5 billion. That’s the result of merely converting the 14,000 NYC UBER cars, and not touching the 145,000 other UBER cars around the world.

Other locations won’t have the same combination of taxi density and high wages of NYC, but many urban environments will have similar numbers. Assume that the next 14,000 cars UBER automates only yield half of the year 1 benefit ($880 million), and the next 14,000 contribute half of that ($440 million). UBER’s 1st-year  profits rise to $2.8 billion. That’s without touching two-thirds of drivers. Still, once UBER converts these cars they will be swimming in cash and be able to dramatically reduce the cost of a ride.

This is where UBER can take over private shuttle buses (Uber-X ride sharing), city express buses and perhaps realize their dream to replace consumer cars. With robot cars, it will make economic sense to pre-park cars nearby customers who need cars on-time, every day. Second generation robot cars could be designed without a steering wheel or driver’s seat, allowing a “living room” configuration that is more comfortable for a working commute to work when it is more than an hour away.  For the even longer commutes of “remote” workers, who take a 2 hour or more drive to work once or twice a week, a configuration with a bed or sleeper seat might be very popular!

DRIVERLESS CARS MOVE FORWARD: Driers are more than a cost issue for UBER. Drivers have been a problem since their early days. UBER has tried to classify  drivers as software users, and not employees, to avoid paying benefits. UBER wants to experiment with pricing methods and special offers, to improve ride volume or profitability. Drivers dislike these unannounced changes and have found that most experiments cut their pay. In early 2016, the fee for UBER-X (shared Taxi’s) was cut 15%, causing another round of driver protests. It also launched a new competing app, Swyft, launched by disgruntled drivers.  That’s the sort of news that will make UBER’s investors… unhappy. Partially because it is bad publicity, and partially because it could one day be real competition. And mostly because 2016 is looking a lot like the year that UBER will launch its IPO.

At this very moment, robot cars drive as well or better than humans. In a short time, robot cars will improve beyond the ability of humans to compete. Which is exactly what has happened with every other task robots have performed. As soon as 2017, Tesla expects to introduce a fully autonomous car. Audi has similar expectations for 2017. Google is targeting 2018 for their autonomous car. Other manufacturers may hold off until 2020. Those who are scheduling a later release date are not waiting for technology. Instead, they are waiting for new regulations and opinion statements from insurance companies.

Early pilot programs show that autonomous cars are ready to take to the road today, and can drive at least as well as human drivers. But does that mean that they are legal to drive? Nevada, California, Michigan, Florida and other states approved laws that allow at least limited use of autonomous cars. When a consumer has an autonomous car, it will almost always have a human driver in the car, they will just allow the car to drive (and be ready to take over if necessary). A taxi will always drive itself, and the passenger may not know how to drive, even if an emergency occurs. The insurance industry is just beginning to say that they will accept the car as the driver, but that doesn’t tell us who gets sued when an accident occurs (UBER, the car manufacturer, the software developer, etc.). Do you need to wait until the laws are in place before you start using a robot car? America didn’t wait for new regulations when drones, electric bikes,

Alternatively, a taxi will always drive itself and passengers may not know how to drive, even if an emergency occurs. The insurance industry is just beginning to say that they will accept the car as the driver, but that doesn’t tell us who gets sued when an accident occurs (UBER, the car manufacturer, the software developer, etc.). Do you need to wait until the laws are in place before you start using a robot car? America didn’t wait for new regulations when drones, electric bikes, smartphones, and the Internet arrived. Will driverless cars be any different?

Today, UBER has gone as far as buying cars for its drivers to ensure that it can expand its program in New York City. It seems likely that that are negotiating an insurance policy that will allow them to start a driverless car pilot program in New York. As to regulations, UBER has proven to be quite ingenious in navigating regulatory issues. New York City’s Taxi Commission controls all car services, and strictly limits the number of Taxis, yet UBER has adroitly sidestepped these regulations. That’s how UBER grew larger than the entire NYC Yellow cab fleet.

TIME FOR A PILOT: The next step is to run a pilot program. Google, Tesla, and other autonomous cars have been doing road tests for years. However, none of these tests have been for Taxis, nor have they allows members of the public into autonomous cars. We already know that it makes great financial sense for UBER to launch driverless cars in NYC. NYC also makes a great location for a pilot. Unlike a driverless truck, that might have a route that goes through multiple states and many cities, UBER’s driverless Taxi pilot only needs to go through one City, and perhaps far less than that. Remember, with UBER, you call the car and tell it where you are going. UBER could start with just one high-volume route, say between Wall Street and the Upper East Side (where many Wall Street employees live). Run it for free. Let UBER become a true expert on this one route. If no accidents occur for 30 days, expand beyond this route. When all of Manhattan is running to satisfaction, expand to the boroughs (Brooklyn, Queens, Staten Island and the Bronx).

There will be resistance from City Hall and from drivers. Perhaps even the Insurance Industry. Resistance is less likely from Taxi riders, who have always had complaints about the quality of Taxi’s in NYC. NYC has a massive tourist industry. Some tourists take home great stories about NYC and it’s Yellow Taxis. Others tell the world that NYC’s Taxis are like something out of an underdeveloped country. Robot cars might improve NYC’s image in the tourist industry. What about traffic violations and accidents?  On May 20, 1899, America issued the first speeding ticket, to Jacob German.

Do I even need to tell you that Jacob was a NYC Taxi driver?  Just a few months later, on September 13^th of 1899, another NYC Taxi driver gets the credit for the first Taxi fatality. The NYC Taxi accident report was released in 1999, showing 13,134 accidents with 3,041 causing personal injury and 10 resulting in deaths. Robot cars would virtually eliminate accidents. Also, 14,000 or more robot drivers could significantly reduce NYC’s legendary traffic jams. Given the social benefits, could City Hall credibly argue against driverless Taxi’s?

Well, they might… if they looked at the employment numbers. There are 50,000 drivers behind the 14,000 Yellow Taxis and another 50,000 drivers in other forms of cars for hire. A successful driverless Taxi will accelerate driverless trucks and other vehicles. There are at least 300,000 full-time truck drivers in New York state and a similar number in New Jersey. So, across the NY Metro area, the relentless drive towards driverless vehicles will result in a million lost jobs. Not even including  a massive reduction in fender benders (and work for garages), more serious accidents (reducing medical work) and, of course, the beginning of a very rapid decline in the auto insurance industry (robot cars that don’t crash will have very, very low premiums). The stakes are very high, so there will be resistance. But resistance can only slow, not stop, the inevitable.

SUMMARY: The numbers don’t lie. UBER has such an overwhelming financial interest in replacing its drivers with driverless cars, that it is indeed inevitable. Perhaps UBER will do their IPO in 2016, perhaps in 2017, but once they go driverless they will go from losing money to multi-billion dollar profitability. And it will only take a pilot in Manhattan to achieve more than a billion dollars in first-year profits. Resistance to robot cars is FUTILE… but there will be resistance…. from drivers, politicians and insurance companies. But in the end, UBER, or something that looks just like them, is going to launch driverless cars, and then  it’s just a matter of time (and not much time) before all drivers of all road vehicles are replaced. That’s my Niccolls worth for today, and I’m sticking with it! Don’t agree? Then comment and tell me why!