Long, complicated, boring, poorly written (but hopefully informative!) post incoming. Don't read it if you plan on complaining, or else skip down to the "advantages" section if you don't care about how the regulations work and just want my analysis (including current guesses for Q1 regulatory revenue).
WARNING: Any or all of the following information could be wrong. I am often wrong about many things. Sometimes I am wrong about the same thing more than once.
Reading 1,500 Pages of Regulations So You Don't Have To
Ok, I've been thinking for a few weeks now about compiling my research on regulatory credits into a blog entry to help clarify the issues and to correct some of the idiotic misinformation floating around out there (I'm looking at you John Peterson).
My original plan was to publish it in the first week of May, because the deadline for manufacturers to submit MY2012 information to California is May 1st, and the shareholders meeting is May 8th. However, I'm not certain California will release the initial reports immediately, and the final reports aren't due until October. So I'm only about 30% sure there will even be new data in the next few days. The fact that I've been waiting since March for the Feds to update (the Federal deadline is in March) and it hasn't happened yet doesn't give me a lot of confidence.
The only solid data I have is from MY 2010 and MY 2011, which kinda sucks because those were trial years and the regulations really didn't kick in until MY 2012. The lack of MY 2012 data really inhibits my analysis, so hopefully they put that out there sooner rather than later.
As to the credits themselves, I have identified 3 primary markets for regulatory credits. There is the market for California ZEV mandate credits, the EPA GHG market, and the NHTSA CAFE market.
Currently the EPA GHG market likely represents the most important revenue stream for Tesla over the next decade. It is possible that Tesla is not currently participating in the NHTSA CAFE market, and indeed appears not to have done so through 2010 -
That said, the joint EPA-NHTSA rule for 2012-2016 clearly envisions credits for EV's participating in the CAFE market, and Tesla IS participating in the GHG market established by that rule.
Here is a quick primer from C2ES that summarizes both of the adopted rules, 2012-2016 and 2017-2025 -
Federal Vehicle Standards | Center for Climate and Energy Solutions
Here are links to the two rules and supporting information -
Regulations and Standards: Light-Duty | Transportation and Climate | US EPA
WARNING: Both rules together run something like 1000 pages. By ignoring side issues you can cut that down to maybe 300-400 pages.
The California ZEV market is what really sold me on TSLA in August 2012. At the time only it and the joint EPA-NHTSA rule for 2012-2016 was in effect and the EPA was relatively weak in terms of promoting pure EV's.
In contrast, the ZEV mandate covered at least through 2025 and specifically favored Tesla's large battery strategy. The regulatory synergy was set up to potentially allow Tesla a large source of revenue from established players, along with market incentives that likely would have resulted in a large market share for Tesla in California and the associated Section 177 states.
However, late last year the EPA and CARB decided to collaborate and normalize their regulations for 2017-2025. CARB amended the ZEV Mandate, removing all references to model years following 2017 (which defaulted them back to federal control in those years) and the EPA published the joint rule, extending and enhancing the 2012-2016 regime.
As a result, my previous model, based heavily on the 2017+ ZEV mandate was DOA. Thanks a bunch, EPA. Thanks. A. Bunch.
Alrighty then, moving on to specifics, here is the (modified) California ZEV Mandate -
http://www.arb.ca.gov/msprog/zevprog...g/attachb1.pdf (actual regulation)
2009 Zero Emission Vehicle Program Review (supporting links)
Basic Summary of CARB is that they wanted to force automakers to sell ZEV's, with the requirements ramping up to 16% of the market in 2018. There was a grace period (which has been effectively active since 1990 when the law first went into effect, with early requirements being abandoned in the early 2,000's and leading to the famous repossessions and subsequent crushing of early EV's), but the fundamental ZEV requirements really began in MY 2012-2014 with 12% of all sales needing to satisfy ZEV requirements, and there is a hard floor of 0.79% of all sales needing to be pure ZEV's (meaning EV's right now). The rest of the 12% can be satisfied with a mix of other advanced technologies, but ZEV's are the gold standard and are worth far more than credits from a Volt, which itself is more valuable than a Prius.
Based on my research a ZEV credit is worth ~$5,000+, and lessor credits don't have a lot of value. Every Model S is worth 4 credits, and each credit is equal to 1 car. So if a manufacturer sold 10,000 cars/year in California, they would need to come up with 79 ZEV credits, and 1,121 credits of lessor value, which can be satisfied with ZEV credits, or filled with a complicated mix of credits from less advanced powertrains. Those requirements then ramp up in 2015-2017.
Here is a quick breakdown -
As you can see, by 2018 16% of the market needed to be 100% ZEV.Code:Model Years Total% ZEV Volt's etc. etc. 2012 – 2014 12 0.79 2.21 3.0 6.0 2015 – 2017 14 3.0 3.0 2.0 6.0 2018+ 16 16
Also, there are travel provisions that allow California credits to count towards the requirement for other Section 177 states, and vice versa, but there is a discount depending on the state. Fundamentally, this reduces the overall requirement because a sale in one state partially fulfills the requirements in the others, which is why you see some compliance cars being marketed only in California. If they sell enough in California they can satisfy the requirements for all Section 177 states. The travel provisions were to have expired in 2017, which would have dramatically increased the ZEV requirement, but all of that has now been superceeded by the new joint CARB-EPA-NHTSA rule for 2017-2025.
If not for the joint rule, Tesla would have been able to bank huge numbers of credits, and various quirks in the law would IMHO have put Tesla into a position of playing no limit Poker with a huge pile of chips, against a group of players with few or no chips and little prospect of catching up. Whatever. Thanks a bunch EPA. Thanks. A. Bunch.
The Federal rules, both the joint EPA-NHTSA 2012-2016 rule and the triple joint 2017+ rule have a rather different philosophy.
First, all vehicles are regulated on an "attributes" basis. Fundamentally, they look at the "footprint" of a vehicle. Basically, draw a box with a wheel at each corner and that's your footprint. I calculate the footprint of the Model S to be slightly more than 53 square feet for purposes of this regulation.
They then have target values for GHG (for EPA) and MPG(NHTSA) that are calculated with a mathematical function based on the footprint attribute of each car. If you scroll through the rule documents I provided you can easily find graphs with "curves" that are based on the regulatory target for that year. Here is the 2017+ rule and you can see an example starting on page 22 of the PDF
Just look to see where the 53 square foot mark intercepts the curve you are interested in and you will know the regulatory target for that year.
For comparison, 2012 BMW 750li is virtually the same regulatory size as a Model S and currently emits 523g/mi CO2. Looking at p88 of the 2012 rule you can see that a vehicle of this size is supposed to hit a target of ~345g/mi in MY 2012, which is a deficit of ~180g/mi.
In the future, you can see that in 2016 it drops to about ~300g/mi for a vehicle of this size (detailed explanation of the curve starts on p32 of the 2012-2016 rule).
As to the "credits" the g/mi CO2 value is converted into metric tonnes over the life of the vehicle (195,264 miles for cars, and 225,865 miles for trucks). The units are Mg which is megagrams. Every megagram is a metric tonne. Every Mg of CO2 is essentially a "credit" under this program.
For regulatory purposes, the Model S generates zero g/mi CO2 until well into the 2020's at a minimum.
Here is the current credit balances through MY 2011 -
The amount of the credit is based on the target curve IMO. When reading the curve, there is an upper limit and a lower limit and the slope. I have been modeling the Model S based on where its target is on the slope, but the actual credit is arbitrary because the Model S doesn't actually generate emissions at the tailpipe, while the curve is a regulatory tool that in practice has vehicles clustered around the line, both higher and lower.
So its not entirely clear to me what the actual value assigned to the Model S is. Here is the equivalent CAFE formula
Because the upper bound and lower bound can be picked by the equation (ie 1/a or 1/b), it certainly seems possible that the Model S would get credit for one of those instead of the point where it intercepts the slope. There is a similar equation for CO2, but the Model S doesn't have a CO2 equivalent to solve it, while it does have an MPGe assigned. When I solve the problem, it looks like the Model S gets the 1/a value, which translates into the high MPG regulatory limit (a favorable outcome). Regardless, I'm not completely confident in my interpretation, and if anyone bothered to read this far I'd be interested in what you think.FOOTPRINT is the product of the vehicle's wheelbase and average track width in square feet, a and b are high and low fuel economy targets that increase from 2012 to 2025 and are constant for all vehicles, and c and d are adjustment factors. Parameter c is measured in gallons per mile per foot-squared, and parameter d is measured in gallons per mile.
1. Has no fleetwide emissions to balance until well into the 2020's at least (if they build a lot of cars, EPA will start counting some "upstream" emissions)
2. For 2012-2016, if Tesla participates in the CAFE market they get a 0.15 divisor for their MPGe value. So 89 MPGe (for the 85kWh) gets turned into a credit of 593.33 MPGe. Automakers are capped in how much they can use this to shift their fleet average (the cap moves from 1.0 MPG to 1.2MPG depending on the MY). This cap somewhat limits the market during this period, and I haven't researched the carry forward provisions for Tesla to use these credits in future model years.
3. For 2017+ CAFE still offers the 0.15 divisor, but there is no cap. This raises the value of these credits to Tesla.
4. For 2012-2016 there is no multiplier for being an EV, but the size of the Model S footprint and 100% credit give it a big advantage compared to a Volt (which is smaller, only gets credit for electric miles, and must balance its own gas emissions). The advantage compared to a Leaf is much, much smaller.
5. For 2017+ the Model S gets a multiplier of 2 for its GHG credit, which ramps down to 1.7 by 2021 before being phased out. This is a small advantage compared to a Leaf. The Volt starts with a multiplier of 1.6 and ramps down in a similar fashion. This is a smaller credit than a Leaf, but still a nice advantage. Plugins of all types are strongly favored, and in fact manufacturers with ICE fleets will need to produce very large numbers of plugins (or purchase credits) to meet post 2021 requirements.
6. All Tesla GHG credits (not certain about CAFE) generated until 2016 are good through 2021, and there is a 5 year carry forward (phase out) of credits after that (use it or lose it), so a 2017 credit expires in 2022.
7. Tesla will be one of the few manufacturers to have a large surplus of any kinds of credits through 2016, after which the bigger changes start to kick in. Tesla might end up being the only major seller of credits in these markets over the next decade. Even Nissan will likely have to horde its credits, and GM is struggling a bit right now, even with the Volt. Toyota has huge numbers of credits, but also less room to improve their fleet in the future and no incentive to help competitors. Also, their "early" credits are not protected until 2021 (protection starts in 2011) so many, many early Prius credits will age out by mid decade. Still, among traditional automakers, only Toyota is really well positioned to get to 2025 without scrambling.
8. Direct competitors, BMW and Daimler are already experiencing significant deficits on the credit market. A special program called TLAAS was created for LVM's (low volume manufacturers, U.S. sales < 400k) which allows them to flag a TOTAL of 100,000 vehicles during the model years of 2012-2016 to have a GHG target multiplier of 1.2. This represents a fraction of their production during this time, but it should keep things from getting totally out of hand. Nevertheless, my assessment is that they will be unable to bank any significant numbers of credits before the 2017 ramp up.
If they go into 2017 without significant PHEV or BEV products they will have real problems. This is also true of LVM's like VW, but VW and the others have a much better profile in terms of having smaller more efficient vehicles to balance the scales. The most likely buyers of Tesla GHG credits are BMW and Daimler.
9. Successful introduction of GENIII in 2017 will allow Tesla to accumulate large numbers of GHG credits when the multipliers are highest. They get an additional bonus if they manage to sell a total of 300k plugins in MY 2019-2021. If they do so their production cap for MY 2021-2025 increases from 200k to 600k before they start getting assessed upstream CO2 costs (ie, the CO2 cost of electricity generation).
10. Based on the structure of the credit bank, it seems obvious that CO2 credits are likely to be able to be sold on any regulatory CO2 market. They can already be exchanged for ZEV credits in the CARB market, and the EPA is already trying to establish similar regulations for large point sources.
11. EPA has listed the "Social Cost" of CO2 at ~$23/Mg. By itself, that likely means something like $1,800 per Model S for 2012 GHG credits. But the cost on this small regulatory market is likely much higher, with plenty of potential buyers, and few sellers. The EPA has itself recognized an implied multiplier of 2 by 2017 (the most critical phase of the ramp up). The marginal cost for automakers who have yet to develop and deploy compliant fleets is likely higher yet. With a multiplier of just 2 and ~2,400 sales in Q4 you get about $8.6m GHG revenue, while assuming 4,750 Model S's in Q1 that represents ~$17M in GHG revenue.
For automakers who are not able to meet the CAFE standards Tesla should be able to leverage additional revenue if they participate in that market. I have not identified any market signals which might be helping to set that price, but with a high MPGe and a 0.15 divisor, it seems clear that Tesla has a potentially large number of credits in that market.
Tesla also has enough of the "gold standard" ZEV credits to be able to satisfy any demand on the CARB exchange. The base penalty for non-compliance on that market is $5,000/missed credit, and the market cost to avoid being forbidden to sell cars in California and the other Section 177 states is higher yet. With 4 credits/car sold in eligible states, and with no non-EV's to balance against, Tesla dominates this market.
With over 1.6 million 2012 California car sales, the minimum ZEV requirement (0.79%) for 2012 represents a base requirement of a maximum of 12,640 units.
1.75 million California sales are projected for 2013, which implies a base requirement of a maximum 13,825 credits, with a similar number expected in 2014. For 2015-17 the base requirement goes up to 3%, which would imply a requirement of 52,500/year at 2013 sales rates.
With ~2,400 sales in 2012 and at least 50% in California (with some unknown additional number in the Section 177 states), Tesla could in theory have generated a minimum of 4,800 ZEV credits representing in excess of $24 million in revenue. The balance between sales of ZEV credits and GHG credits (and possible CAFE credits?) for Q4 will be unclear until the California ZEV MY2012 data is released (initial deadline is May 1, final report WILL be release mid October... unclear if interim data is released).
With just GHG and California ZEV credits we have a reasonable estimate of ~$33m in Q4 revenue, and the additional revenue that was actually recorded is likely some combination of Section 177 ZEV and possibly CAFE. It's also possible that the GHG multiplier is smaller than 2 in 2012 and that the base price for the ZEV credit is substantially more than $5,000. For Q1, GHG assumptions give ~$17m, while ZEV is still possibly $40m+.
I get to the Q1 ZEV number with an estimate of ~2,000 California sales ("3,000 California Registrations by March 21" + ~200 more before the end of March - 1,200 for 2012). This gets ~8,000 credits. Automakers would likely still have had a fairly large deficit from 2012, along with 2013 requirements to fill.
Keep in mind that the base requirement is something like 26,000 credits for 2012-2013, and Tesla only supplied ~4,800 in 2012. Only Leaf had substantial sales in California which help satisfy this requirement, with just 2 credits generated per Leaf sold, and a need to potentially bank credits for 2015+ when the base requirement goes up to 3%. Also, 2012 was a terrible year for Leaf sales with fewer than 8,000 sold in the U.S. Assigning 50% of U.S. sales to ZEV states is a generous estimate and would result in 8,000 ZEV credits for Nissan, with virtually none for anyone else.
As a result, I show a minimum potential of $57m in Q1 credit revenue from a model that predicts $32m (with over $40m actual) Q4 revenue. Just as in the case of the Q4 model I am ignoring Section 177 ZEV credits and potentially CAFE credits. It's possible that I am overestimating the GHG multiplier, and there is no requirement that Tesla actually sell any credits anywhere. Also, some SVM manufacturers are somewhat insulated from the ZEV requirement, so 26,000 credits is a maximum requirement.
On balance, I think its likely that speculation about Q1 revenue from regulatory credits has substantially understated the potential, especially since $5,000/credit looks to be a minimum, and manufacturers had an incentive to get to a positive balance before the May 1st reporting deadline. This is very much a sellers market.
12 states besides California, representing a significant fraction of the U.S. automarket have fully adopted the ZEV mandate -
US: Section 177 States - Transportpolicy.net
As of the date of the linked chart, Florida had agreed to the mandate but not yet ratified it in the legislature. If they do so, they will be the 13th state, while New Mexico is scheduled to participate in 2016 and 2017.
Travel provisions allow some credits generated in one state to "travel" to satisfy the requirements in another state. I have not yet performed an analysis to determine the total market size, and it seems likely to be a moving target. The explanation of the travel rules is in the source data provided in the ZEV links above.
Nissan will have many fewer available credits, but will also be a seller in this market. All other companies are likely buyers. GM is fairly likely to be satisfying a fair portion of the requirements with Volt and can cascade down from there (I haven't analyzed this), but the total requirement is huge and requires large numbers of regular hybrids, Neighborhood EV's and other technologies (remember, the base requirement for ZEV's is only 0.79%, but the total requirement is 12% for 2012/13).
Tesla can sell their ZEV credits to satisfy any of these requirements, while a Prius credit (for example) has relatively limited utility. The value in this segment is less because Toyota has huge numbers of lessor credits and will be a seller, but again only a portion of the 12% can be satisfied with Prius credits. Many companies are at least partially fulfilling their requirements through their own product sales using Hybrids and NEV's, etc. It is difficult to drill down enough to really model this market without the MY 2012 data. The secondary market is even more difficult to analyze for the Section 177 states.
Nothing prevents an automaker from purchasing Tesla credits and banking them for future use. And there is no requirement that Tesla sell any of its credits right now. So the actual pace of sales is not necessarily driven by current needs, but rather by an attempt by both sides to maximize utility.
To wrap this up, the important takeaway is that regulatory credits are a lucrative part of Tesla's income stream, and I see little evidence of that changing any time soon. ZEV credits are incredibly lucrative, and will be for a few years. GHG credits represent a steady income stream going into the next decade and beyond, with revenue strongly correlated with the price of carbon and with a nice bump coming along just as ZEV expires. CAFE is a big unknown, but the maximum utility looks to be 2017+ when the caps limiting use are removed. If they are bankable using the same rules as GHG's (which is likely since its the same rule, and probably in a section I skipped) then it seems possible that Tesla would bank those and that they could be an important supplement to GHG's in the 2017+ time frame.