Tomorrow’s car will represent a steep change in form and function. According to Goldman Sachs 2025 cars will be green, convenient, safe and affordable (Goldman-Sachs, 2016). The transition has already happened with the EPA push for a fuel economy standard set tentatively at 50 miles per gallon despite the reluctance of the automotive industry (Snavely, 2016). Safety and convenience are strong megatrends that have taken an interesting turn: driver is not wanted anymore. According to NHTSA, drivers are the reason for 94% of crashes (NHTSA, 2016). Table 1 show the statistics of the critical reasons of accidents. What if the humane element can be removed from the equation?
Table 1: Driver-, Vehicle-, and Environment-Related Critical reasons
Table 2 shows more specifically the reasons of the humane error and what should be the core safety strategy of the algorithm of a computer-driven vehicle: recognition and decision.
Table 2: Driver-Related Critical reasons
However, completely autonomous vehicles programmed not to crash are already on the horizon. According to McKinsey, combining autonomous and non-autonomous vehicles in a single traffic mix will be a significant challenge (McKinsey, 2014).
The most difficult time is likely to be the transition period, while both kinds of cars learn to share the road before self-driving ones predominate. (“Self-drive only” lanes and dedicated roadways might be the first step.).Passengers will be responsible only for choosing the destination, would have the freedom to do what they please in a vehicle. Connected cars—especially self-driving ones—could also change the way people use their drive time.
In a 2013 survey, more than 50% of respondents said they would prefer to listen to music, talk on the phone, watch videos or browse the Internet while traveling by car. Disabled, elderly, and visually impaired people would enjoy much greater mobility. Throughput on roads and highways would be continually optimized, easing congestion and shortening commuting times. As self-driving technologies are quickly developing the next logical question is the ownership of car. If a car can drive by itself then it can do a lot more during the day then packing parking lots around cities.
The majority of vehicles worldwide are used only to commute or for short trips during the day, leaving them idle 95% of the time. Would drivers decide to forgo ownership and access cars only when they need them? Several options could be imagined to replace ownership but one of them is already being tested by Uber as self-driving Uber rides already tested in Pittsburg (Huston, 2016). According to McKinsey, the transition to fully autonomous mobility is ineluctable with a transformation of at least 35% of vehicles to some form of autonomy by 2040 (McKinsey, 2014).
According to the Economist, all over the rich world, young people are getting their licenses later than they used to—in America and also in Britain, Canada, France, Norway, South Korea and Sweden. Even in Germany, car-culture-vulture of Europe, the share of young households without cars increased from 20% to 28% between 1998 and 2008 (The future of driving, 2012). Unsurprisingly, this goes along with driving less. American youngsters with jobs drive less far and less often than before the recession. 16- to 34-year-olds in American households with incomes over $70,000 increased their public-transport use by 100% from 2001 to 2009.
There are different measures of saturation: total distance driven, distance per driver and total trips made. The statistics are striking on each of these counts even in America, still the most car-mad country in the world. There, total vehicle-kilometres travelled began to plateau in 2004 and fall from 2007; measured per person, growth flat lined sooner, after 2000, and dropped after 2004 before recovering somewhat (see chart). The number of trips has fallen, mostly because of a decline in commuting and shopping (of the non-virtual variety).
Given the current megatrends and the on-going self-driving tests conducted by Uber in Pittsburgh, one can consider the self-driving service as a viable and attractive option becoming available in the very short term. Therefore, we run a Cost Benefit Analysis between purchasing a new car and using driverless Uber rides to commute.
Few Assumptions are as follows
- The Car: First key assumptions are the annual mileage. Four annual mileage were considered in our sensitivity analysis: 10k, 15k, 20k and 25k miles per year to reflect the potential variations. The total initial investment is $15,000 for a 2016 mid-size car typically used for commuting in a relatively dense and urbanized area. According to the Independent Insurers Agent, there are a number of used cars you can purchase for under $15K, particularly if you are looking at cars that are 5 years old or older. In the under $15,000 price range, you will only find new compacts and subcompacts such as the Chevy Sonic or the Hyundai Accent (Agents, 2014). Average car prices are currently in the 30k range in the US but it represents a wide range of vehicles purchased for more than cost efficient urban commuting, e.g. minivan or trucks.
Annual costs for operating a car were based on a report published by AAA in 2015 (AAA, 2015) . These estimates include licenses and tax ($665) and insurance ($1115). We kept these items constant in our study although one can argue that insurance could vary somewhat with the mileage driven. The general regular maintenance ($766.5), the annual tire replacements ($147) and the fuel costs were pro-rated with the mileage. The nominal value provided by AAA at 15,000 miles per year was ratioed accordingly. We estimate the lifetime of the car to be 200,000 miles. The regular maintenance budgeted in our operating costs will keep the car running. For instance a mileage of 10,000 miles per year will assume that the car lifetime will be 20 years. We assumed that the car will show an average fuel economy of 25 miles per gallon over its entire life and that the fuel cost will average $2.80. These assumptions match the total fuel cost proposed by AAA of $1681.50 per year. Note that electric cars would represent a different analysis.
2. The Uber Ride: Another key assumption is the cost to use a driverless Uber service. According to Ark Invest, “with an autonomous system of $2,000, the price of an autonomous taxi could drop to $0.35/mile (after factoring in the cost of remote human operators that could guide vehicles in emergencies) given the improvement in autonomous error rates and the drop in electric vehicle battery costs, while still providing a healthy return on capital to the service provider.” (Keeney, 2016). Uber will probably start by offering higher rates so our sensitivity analysis includes $0.40, $0.45 and $0.50/mile.
3. Financial Assumptions :Additionally, we applied a 2% annual inflation rate to the costs required to operate and maintain the car. When calculating the net present value, we assumed a discount rate of 5%
The total mileage was divided by the year mileage to obtain the lifetime of the car. After applying the ratio to the fuel, maintenance and tire expenses, the costs were added on an annual basis to obtain the car cost per year. This number is divided by 12 to obtain the car cost per month and also divided by the mileage to provide the cost per mile.
|Compact Car||$ 15,000|
|license, tax, etc.||$ 665.00|
|cost per mile||$ 0.29|
Table 3: Cost hypothesis
The mileage rate was multiplied by the car mileage and divided by 12 months to obtain the monthly cost of the Uber ride.
Net monthly cash flow is the difference between the monthly car cost and the Uber ride cost. 2% inflation was applied to the net cash flow using this formula
Inflation = NCF(1+2%/12)^n
Discount Cash Flow (DCF) was calculated as
DCF = Inflation (1 +5%/12)^-n
The Net Present Value was the recurring difference between the initial investment minored by each monthly DCF payment. The Internal Rate of Return was computed as the interest rate that would bring the NPV to zero at the lifetime set as the ratio of the total mileage (200k) over the year mileage. Following tables are showing the results of the computation of the Cost Benefit Analysis.
Table 4 shows the monthly gain versus the mileage and the Uber cost per mile. This metric show that there is always a monthly gain in owning a car except at the lowest mileage and the lowest Uber cost. Only with that combination that Uber rides are most cost effective solution. This situation would arise in a dense city with multiple commuting options such as New York where the mileage put on a car would be necessarily low.
Table 4: Monthly Gain ($)
Present Value at maturity, considered here as the car lifetime which varies from 8 years to 20 years, indicates at what condition the monthly benefit is worth the ownership of a car in the long term (Table 5). Even at the lowest cost of Uber rides (bottom raw) the long term benefit of riding with Uber is a better option than owning a car if the distance covered exceeds 20,000 miles year.
Table 5: PV of Benefits ($)
At the lowest level of mileage, riding Uber is a viable possibility as long as the cost per mile is lower than $0.40. Table 5 shows that despite some monthly benefit at owning a car a few options are not overcoming the initial investment of the purchase of a car. At low mileage or low Uber cost a few options are most sustainable.
The internal rate of return shown in Table 6 confirms that low mileage and low Uber cost mean a significantly low DCF in order to achieve break even at the car lifetime. By the same token, higher Uber cost and extensive mileage cause a relatively high IRR.
Table 6: IRR for Car lifetime (%)
Table 7 shows a response of the Profitability Index (PI) ratio consistent with the growth of Present Value with time which can overcome the initial investment in the ownership of a car. PI higher than one indicates a go decision in owning a car versus a PI lower than one suggesting that riding Uber is the best option. Once again the relatively small variation of the cost per mile of the Uber ride translates into a dramatic shift of the financial rationale to drive a car.
Table 7: PI for car lifetime
Table 8 shows the number of years before achieving the financial break-even point taking into account inflation and Discount Cash Flow. This duration needs to be put in perspective of the car lifetime for each annual mileage. Based on our hypothesis of maximum mileage equal to 200,000 miles, the lifetime of the car at the highest mileage is 8 years. At this annual mileage, the initial investment would paid off by the gain from Uber rides. At a rate of 20,000 miles per year, the car lifetime is only 10 years and would not be recovered if the Uber rides fair at $0.35. The same conclusion can be made at an annual rate of 15,000 miles because the car lifetime is 13.3 years. However at the lowest mileage rate of 10,000 miles, the Uber rides would be advantageous at $0.40 because the car would break down before reaching the break-even time.
Table 8: Break Even time (Years)
Intangible benefits must be addressed in this important decision-making process. The most significant benefit of using Uber ride is the possibility to work, sleep, phone in the car and use the commuting time more effectively. For example, the ride could be accounted as work hours by progressive employers. The gain from driving would be realized through a higher productivity during this time of day.
Uber rides would also represent the possibility for more people to move around town with less cars. It would also help reduce the growth of parking lots in crowded areas. Uber rides will ultimately communicate information on their destination opening the possibility to share a ride with more than one person.
Even though the maintenance costs are factored in our calculations, there is a definite amount of time taken by repairs or recalls when owning a vehicle. It also requires to own a garage and in the case of extended family to take over the neighbor lawn as well. One major concern would be the impediment of driverless cars in bad weather or in an accident. Uber indicated a central command would always be in touch with the fleet. Would the infrastructure enough on the road to support driverless cars. The US government just releases a proposal for an investment of 4 billion dollars in the preparation for driverless car accommodations as Michigan just adopted the most permissive regulations for this new technology. Lastly there is always a possibility of misuse of personal information or hacking of the software in the autonomous cars. Leading to unsafe and unpleasant situations.
Our analysis has shown that above 15,000 miles/year a commuter is recovering its investment when comparing with the cost of Uber rides especially above $0.40/mile. This critical price milestone has been announced as a realistic target but is based on decreasing radar production costs that need to be demonstrated. However, there are some significant intangible benefits at not taking care of a vehicle, not to have to buy a house with gigantic garages and increase our productivity during our commute. It would definitely help with a lower stress level.
PS: This paper was written as part of group assignment for Financial management
AAA. (2015). Retrieved from http://newsroom.aaa.com/2015/04/annual-cost-operate-vehicle-falls-8698-finds-aaa-archive/
Agents, I. I. (2014, 06 2). Retrieved from https://www.trustedchoice.com/insurance-articles/wheels-wings-motors/cars-under-15000/
Goldman-Sachs. (2016). Retrieved from http://www.goldmansachs.com/our-thinking/technology-driving-innovation/cars-2025/index.html?cid=PS_01_55_07_00_01_15_01&mkwid=
Huston, C. (2016, 09 15). Watch Uber’s self driving cars hit the road in Pittsburgh. MarketWatch. Retrieved from http://www.marketwatch.com/story/watch-ubers-self-driving-cars-hit-the-road-in-pittsburgh-2016-09-14
Keeney, T. (2016, 2 13). Low Cost LiDAR Will Save Lives. Retrieved from https://ark-invest.com/research/low-cost-lidar-system#fn-6025-5
McKinsey. (2014). Retrieved from http://www.mckinsey.com/industries/automotive-and-assembly/our-insights/a-road-map-to-the-future-for-the-auto-industry
NHTSA. (2016). Retrieved from https://crashstats.nhtsa.dot.gov/Api/Public/ViewPublication/812115
Snavely, B. (2016, 10 11). Automakers lobby to ease MPG standards despite improved fuel economy . Chicago Tribune. Retrieved from http://www.chicagotribune.com/classified/automotive/sc-fuel-economy-cafe-midterm-autocover-1013-20161010-story.html
The future of driving. (2012, 09 22). The Economist.