My Tesla Story - and why a Model S is less expensive than a Honda Odyssey

[ecarfan]

This topic should be merged with the pre-existing topic that was started by the guy who put that spreadsheet together.

@Objective1, in the San Francisco Bay Area with PG&E on their EV rate plan, off peak rates are $0.11/kWh.

 

[Yuri_G]

A couple questions:

The spreadsheet seems to be assuming electricity cost of recharging at $0.12 per kWh. That is very low for California, New York, and many other places Tesla buyers live, isn't it?

And it assumes gas retail prices will rise over 8 years. Why is that? Gas prices are not on an upward curve the last couple years, and a production revolution is underway thanks to fracking (and not, in the long term, just in the US).

Thanks for any insight you can provide.

Fracking doesn't affect gas prices, however oil prices do. Oil prices rise and fall based on global demand, which has been rising steadily the past 15 years.

Motor Gasoline Sales Through Retail Outlets Prices

 

[Objective1]

In my state, the price of gas isn't much higher now than it was in 2006, and it's been floating up and down around that point. Oil prices are like all prices, they are affected by both supply and demand. Fracking in the US hasn't driven down oil prices so far, but that's because it's world-wide impact on supply has hardly begun to be felt.

 

[RobStark]

A couple questions:

The spreadsheet seems to be assuming electricity cost of recharging at $0.12 per kWh. That is very low for California, New York, and many other places Tesla buyers live, isn't it?

And it assumes gas retail prices will rise over 8 years. Why is that? Gas prices are not on an upward curve the last couple years, and a production revolution is underway thanks to fracking (and not, in the long term, just in the US).

Thanks for any insight you can provide.

Southern California Edison Tier 1 is $0.13 per kWh and there are various EV rate programs available to lower the cost. This to incentivize overnight charging while the grid is usually idle. I would say $0.12 is a high estimate.

Gasoline price fluctuations are strongly linked to global demand and supply for oil. Spikes in prices not related to global demand and supply are generally short time price fluctuations in the US.

While fracking will increase domestic supply and put downward price pressure on global prices in will likely be overwhelmed by increases in global demand.

Any fall in prices the next two years is likely to be short term.

 

[AndreyATC]

One thing to add, which affects electricity pricing, is supply charge
My kw cost is only $0.06 (delivery charge), but with additional $0.10 per kw supply charge it comes to about 0.16 on the bill

 

[GDH]

But Batman drives the Odyssey so did anyone take that into consideration?

CLICK ME

 

[russman]

I'm in CA and have solar. I'm on a Time of Use plan and my low cost usage is about $.085 for the lowest rate. So $.12 is actually high, but doesn't hurt.

 

[bluedevil30]

Thinking about buying a Tesla and enjoyed your spreadsheet. A few comments after a very quick review:

-Did you include the cost of installing the home charging station in your initial costs?
-I think total cost of ownership - total and per mile (3,5,8 yrs) is a more relevant comparison tool than NPV
-This model would be way more accurate if Tesla offered a true lease option - then residual value can be eliminated.


Since the Company is now very stable the biggest risk in my mind is residual value. In five years, I don't want to have a new twice as efficient model come out for 50% of the price and crush my resale value. Nobody really knows the resale market for a used Tesla yet.

 

[JonathanD]

The 3-year buyback guarantee is good for piece of mind. By then we should have preliminary estimates on Model E pricing and can see how the resale market has been doing.

 

[VolkerP]

the biggest risk in my mind is residual value. In five years, I don't want to have a new twice as efficient model come out for 50% of the price and crush my resale value. Nobody really knows the resale market for a used Tesla yet.

There is a bottom for efficiency established by physics. Don't worry. But we can expect Tesla to improve EVs dramatically in the years to come.

The only way to justify a Model S on a financial basis is, if you plan to keep it for a long time - 8 ears or more. Then the biggest risk will be failure rate and cost of repair when out of warranty.

 

[tnagpal]

The model really bothered me for a while since the CPO Mercedes came out horribly. A used car should definitely come out better in just about every case.

The problem is the 10,20,30% depreciation assumption that is applied across any car. A better assumption should be ~30% in the first year and roughly increase 7.5% every year. For a used car, you depreciate at 7.5% the first year and then incrementally at the same rate.

I took depreciation numbers from cars.com for a Mercedes Benz E350, and then for the used MB E350 car, I started with the incremental depreciation in year 3 since the first year hit is usually the largest. What you get then is that buying a used car saves you a massive amount of money.

You can take a look at the revised numbers here - would appreciate anyone catching any errors or critiques.
https://dl.dropboxusercontent.com/u/712730/Tesla%20Model.xlsx

 

[Larry93428]

I have both a Model S and an Odyssey.
Service on that Od is expensive.
Last service on the Od they wanted to replace the serpentine belt at $1200. They said this is routine.
One of the sliding doors is starting to fail. That will cost a lot to repair!
Are such maintenance costs included in this thread?

 

[vevans]

I have these vehicles also. I find it amazing how it is difficult for people to grasp how much they spend in fuel. Someone I know said it best they are just used to buying fuel. This is called a paradigm something we are used to doing even if it costs more. Anyone with basic business skills should be able to understand why a Tesla makes sense with gas prices hovering around the four dollar mark. I really think the Tesla will hold its value alot more than what is shown.

 

[brianman]

This is called a paradigm something we are used to doing even if it costs more.

"Pair of dimes grift"

 

[alexv]

I'm sorry to burst everyone's bubble, but there are two glaring problems with this spreadsheet.

TL; DR: MS actually costs ~$25K more than the Ody over 8 years, given OPs assumptions.

First, and most importantly, the spreadsheet misinterprets depreciation-based TCO model. I'll try to explain the mechanics of the model the way I understand it (not taking discounting into account yet):

1. At the end of every year of ownership, you perform an imaginary sale of your car at its then-current resale value
2. The difference between the price of the car at the beginning of that year and the price at which you "sold" plus all other expenses (gas, service, taxes, what have you) is the total cost of ownership of the car for that year.
3. If you intend to keep the car for another year, you immediately perform an imaginary buy-back of the car (at the same resale value you just "sold" it for)
4. If you don't intend to keep the car, you just turn that imaginary sale into a real one.

The beauty of the model is that it works for any number of years, and whether you intend to keep the car for another year (i.e. whether the sale was imaginary or not) has NO effect whatsoever on the TCO calculation for the previous year(s).

The problem with OP's spreadsheet is that on year 8 (cell J46), it deducts the resale value of the car from that year's cost of ownership, supposedly because that's when the car is actually sold. But you can't do that! The model already took the sale into account!

This is (apparently) not easy to grasp, so let me provide a couple of examples to show the flaws in the spreadsheet.

First, imagine that our beautiful MS has no depreciation whatsoever (for example, it's mister Musk's own car, or whatever). Delete all the values from the "Depreciation factor" row. See the NPV? It's MINUS $53K now. That means that you paid $92K for your car (plus taxes, minus tax incentives), then paid each year for tires and electricity, then sold it for the same $92K (so no profit), and came out with a $53K discounted(!) profit COMPARED TO INVESTING SAID $92K AT 5% RATE (sorry for caps)!

For a more realistic example, imagine you sold the car after the first year. It's pretty clear from the table, that your total cost of ownership for that year should be around 20K (again, without discounting). However, if you remove all the years after first and perform that resale value subtraction in C46, the NPV becomes minus $50K! That's more than 50% profit, and you get to ride an awesome car for a year! How is that possible, you ask? Well, it isn't.

Anyway, after getting rid of that deduction (but leaving everything else untouched), discounted MS TCO for 8 years comes up to $62K. However, I argue that, since we're doing NPV discounting, even that has to be adjusted. To see why that is true, again imagine we only own the car for one year. For simplicity, also imagine the depreciation for that year is 0, there were no taxes or fees to be paid and no tax incentives, and also you charged the battery for free. Remember that we're comparing to investing the money at 5%, right? So, you buy the car for $92K, spend nothing else, and at the end of the year sell it for $92K. Does it mean that you suffered no losses compared to investing? Of course not! You have to prorate the initial investment ($92K) and only THEN subtract resale price from it to get the year-end cost of ownership, in this case $4.6K (or, after discounting, $4.3K NPV). So for the real calculation, for each year we have to prorate the resale price at the beginning of the year and subtract year-end resale price from THAT - to get adjusted incremental year-end depreciation.

I've updated the formulas for all the cars in the original spreadsheet and put it here: https://docs.google.com/spreadsheet/ccc?key=0Aoyh3mqbEkS4dDNXQ1pGOF8tdEdUSUlGWUZkZ05tU2c
With OP's original assumptions, MS TCO comes up to $83K - that's $2K more than even the MB E350'14 , and $25K more than the Ody.

 

[alexv]

Edit: reverted Miles/yr back to 15,000 and made another small fix - now MS is $1200 less than MB E350 '14 (yay!). Still $25K more than the Ody though.

 

[llavalle]

I'm sorry to burst everyone's bubble, but there are two glaring problems with this spreadsheet.

I always found the calculations a bit over-optimistic. Before buying my S85D (last month), I did a TCO comparison between the S and my current car, a 2013 Volvo S60 T6. (bought for $52k CAD + taxes)

With the current situation in Quebec, Canada, I was getting a break-even point at around 80months, and that's with a base model S85D with dual charger and tech package.

Current prices in QC :
Electricity : 8.26c /kWh.. in CAD, that's $0.0731 USD at the current exchange rate. We also get the first 30kWh per day at 5.57c!
Gaz : We're paying around $1.40 cad per LITER (regular unleaded), that's around $4.70 USD a gallon

As you can see, electricity is cheaper and gas costs more...I found it very weird that with the prices you guys have on both of these you could beat a Honda Odyssey (worth 2/3 of my Volvo) after 8 years... it just did not add up.

Thanks for noticing this flaw in the calculation!

 

[brucet999]

Solar Panel TCO Model

PL804,
Nice write up. Now, if you do the numbers on owner/builder solar and then plug those into your cost of energy

I did just that for my location in Huntington Beach, CA.
4.5KW self-installed array with 96% inverter efficiency, azimuth 270 inclination 22º would cost about $12,000 less 30% tax credit and projects to produce 6,500 to 7,000 KWh per year. Enough to offset all charging costs for 12,000 mi per year plus reduce home usage by 2,500 to 3,000 KWh per year; at SoCal Edison tiered rates, would reduce my electric bill about $500 per year.

To get a $0.12/KWh rate here, you have to install a second dedicated meter for EV charging at a cost of about $2700 (You need an electrician to cut and splice Edison's incoming lines) and charge during off-peak hours - 10pm to 8am. 5-year TCO for my proposed $90,000 list price MS using Grid power comes out to $47,800. Solar panel scenario comes out to 5yr TCO of $48,300. Sale of house in 5-yr scenario (likely for us) would increase home value by at least as much as Solar array cost, (likely $5k more using commercial installation price) and makes fuel cost a negative $500 per year; 5yr TCO becomes $36,300

For roughly comparable cars in size and luxury I chose Lexus ES300h and Lexus ES350, both pretty loaded, at $43,900 and $42,500 respectively. 5yr TCOs for those come out to $35,000 and $38,300 respectively. Even my best scenario, with zero cost for solar, MS costs $1300 more than ES300h. Still, the premium to own a Tesla works out to less than I had expected.

BTW, I include tires at around 30k and 60k miles for all three cars (according to Lexus and Tesla service departments) and brakes according to Lexus. Depreciation for Lexus came from Edmunds.com TCO and for Tesla from used prices in the area discounted for typical used car dealer trade-in price. For 2014 and 2013 MS those prices were unexpectedly high, giving depreciation rates of 18% first year and 23% 2nd year, using today's new sales prices comparably equipped as basis. (I did not want to try to figure out prices from a year and two years ago, even though I know that they were somewhat lower then.) For depreciation in years 3,4,5, I used Edmunds' rate progression for Lexus as they had been close in the first two years. Given that Tesla will be busy rolling out the Mega battery facility, Model X and Model 3 through at least 2018, I doubt that they would lower Model S pricing even if their battery costs decrease, so I wouldn't expect a collapse in resale prices.

 

[GreenPenguin]

I did just that for my location in Huntington Beach, CA.

To get a $0.12/KWh rate here, you have to install a second dedicated meter for EV charging at a cost of about $2700 (You need an electrician to cut and splice Edison's incoming lines) and charge during off-peak hours - 10pm to 8am.

You don't need to install a second meter you just need the TOU-D-A/TOU-D-B rate plan. Super-off peak 10pm-8pm is actually $0.11.

So a little cheaper.

 

[30seconds]

I did just that for my location in Huntington Beach, CA.
4.5KW self-installed array with 96% inverter efficiency, azimuth 270 inclination 22º would cost about $12,000 less 30% tax credit and projects to produce 6,500 to 7,000 KWh per year. Enough to offset all charging costs for 12,000 mi per year plus reduce home usage by 2,500 to 3,000 KWh per year; at SoCal Edison tiered rates, would reduce my electric bill about $500 per year.

To get a $0.12/KWh rate here, you have to install a second dedicated meter for EV charging at a cost of about $2700 (You need an electrician to cut and splice Edison's incoming lines) and charge during off-peak hours - 10pm to 8am. 5-year TCO for my proposed $90,000 list price MS using Grid power comes out to $47,800. Solar panel scenario comes out to 5yr TCO of $48,300. Sale of house in 5-yr scenario (likely for us) would increase home value by at least as much as Solar array cost, (likely $5k more using commercial installation price) and makes fuel cost a negative $500 per year; 5yr TCO becomes $36,300

For roughly comparable cars in size and luxury I chose Lexus ES300h and Lexus ES350, both pretty loaded, at $43,900 and $42,500 respectively. 5yr TCOs for those come out to $35,000 and $38,300 respectively. Even my best scenario, with zero cost for solar, MS costs $1300 more than ES300h. Still, the premium to own a Tesla works out to less than I had expected.

BTW, I include tires at around 30k and 60k miles for all three cars (according to Lexus and Tesla service departments) and brakes according to Lexus. Depreciation for Lexus came from Edmunds.com TCO and for Tesla from used prices in the area discounted for typical used car dealer trade-in price. For 2014 and 2013 MS those prices were unexpectedly high, giving depreciation rates of 18% first year and 23% 2nd year, using today's new sales prices comparably equipped as basis. (I did not want to try to figure out prices from a year and two years ago, even though I know that they were somewhat lower then.) For depreciation in years 3,4,5, I used Edmunds' rate progression for Lexus as they had been close in the first two years. Given that Tesla will be busy rolling out the Mega battery facility, Model X and Model 3 through at least 2018, I doubt that they would lower Model S pricing even if their battery costs decrease, so I wouldn't expect a collapse in resale prices.

why would you use the ES and not the LS or GS?