More Good News for EV Buyers
The broadening EV market means that government subsidies will flow to buyers – not sellers.
The electric vehicle market continues to grow. There are now more than 100 different EV models for sale in the United States, up from only 2 in 2011. Current offerings come from Audi, BMW, Cadillac, Chevrolet, Ford, Genesis, Hyundai, Jaguar, Jeep, Kia, Lexus, Lucid, Mercedes, Mini Cooper, Nissan, Polestar, Porsche, Rivian, Subaru, Tesla, Toyota, Volkswagen, and Volvo.
This broadening of the market is great news for EV buyers. There are more options than ever before, and the market is becoming more competitive, with lower prices for EVs and shrinking profit margins for EV manufacturers. A recent piece by Liam Denning, for example, points to shrinking markups at Tesla.
For today’s post, I want to talk about an additional related benefit for buyers. Throughout the last decade, supply constraints led EV sellers to capture a large part of government EV subsidies. But the broadening of the market is relieving these constraints and – in the language of economics – making the supply of EVs more elastic. This means that moving forward government subsidies will tend to flow to buyers – not sellers.
Incidence 101
As economists like to point out, subsidies can cause market prices to change, shifting subsidy dollars between buyers and sellers. This may seem like some obscure technical issue, but this question of subsidy incidence is central to thinking about the effectiveness and distributional impacts of policy.
A good example comes from January 1, 2019. On that day, the U.S. tax credit available for Tesla vehicles decreased from $7500 to $3750. And what did Tesla do? They decreased their prices by $2,000, effectively offsetting about half of the change in net price to consumers. More recently, Tesla did the same thing, lowering the price of vehicles that were no longer eligible for the full tax credit.
Pricing behavior like this suggests that Tesla – not Tesla buyers – has been capturing a large part of the tax credit. The key question to ask is, “How did prices change?”. When we see an increase or decrease in subsidies “pass-through” to prices, this means EV sellers are capturing part of the subsidy. These are, of course, just anecdotes, but this is also what we would expect based on economic theory. When supply is constrained and relatively inelastic to price, then sellers capture a large part of government subsidies. Throughout much of the 2010s, Tesla was up against supply constraints. Early production of the Model 3, for example, was described as “production hell” with Elon Musk sleeping at the Fremont factory as the company grappled with intense production bottlenecks. With supply constraints like this, the supply of EVs was relatively inelastic, with Tesla already selling every car it could make.
–
Here Come the EVs
Fast forward to today. There are 100 different EV models for sale in the United States and the market is more competitive than ever. Tesla still dominates the U.S. EV market, but its market share has fallen from 80% in 2020, to below 60% today, and there are large inventories of EVs available from multiple manufacturers.
All this points to buyers – not sellers – capturing the majority of U.S. government subsidies. No longer can Tesla simply increase their prices to capture the subsidy for themselves. To the contrary, Tesla and all the EV manufacturers are feeling distinct downward pressure on prices due to market competition.
Economic theory is again helpful here. When supply is not constrained – relatively price elastic – then subsidies do not drive up prices, and EV subsidies tend to flow to buyers – not sellers. As we continue to come out of the broader automotive chip shortage, the supply for EVs is going to be increasingly elastic, with a large number of sellers competing in most light-duty categories.
–
Conclusion
The key idea is that subsidies can change market prices, shifting who actually gets a subsidy. In the extreme, if a buyer receives a $7,500 subsidy, but EV prices are $7,500 higher than they would be otherwise, then the after-tax price for the buyer hasn’t actually changed at all.
This is not just about EVs. We are going to hear a lot more about tax incidence as the subsidies under the Inflation Reduction Act continue to ramp up. For example, there is also a $2,000 tax credit for heat pumps, so it will be interesting to see how these subsidies impact heat pump prices.
It also means that policymakers need to be careful, because subsidies may not have the effect they expect. You can implement a subsidy hoping to help buyers in some market, only to learn later that the subsidy actually flowed to sellers. It is important to think hard – beforehand – about the structure of the market and who will get the subsidy.
–
Keep up with Energy Institute blogs, research, and events on X @energyathaas.
Suggested citation: Davis, Lucas. “More Good News for EV Buyers” Energy Institute Blog, UC Berkeley, October 16, 2023, https://energyathaas.wordpress.com/2023/10/16/more-good-news-for-ev-buyers/
Categories
Lucas Davis View All
Lucas Davis is the Jeffrey A. Jacobs Distinguished Professor in Business and Technology at the Haas School of Business at the University of California, Berkeley. He is a Faculty Affiliate at the Energy Institute at Haas, a coeditor at the American Economic Journal: Economic Policy, and a Research Associate at the National Bureau of Economic Research. He received a BA from Amherst College and a PhD in Economics from the University of Wisconsin. His research focuses on energy and environmental markets, and in particular, on electricity and natural gas regulation, pricing in competitive and non-competitive markets, and the economic and business impacts of environmental policy.
Energy Institute Blog 
Tax rebates for EV’s discriminate against all of us who are too poor to pay a lot of taxes, but who would buy EV’s if the costs were lower for them.
RE: “EVs are stacking up on lots because nobody can afford them & about 20% of the buying public wants them.” That too is about right, in California per DMV transition prices data, 18% of all vehicle sales are the highest priced models of a calendar year (all non-subsidized models prior to EV sales). This would suggest the car buying population size that is less sensitive to vehicle prices, this is somewhat consistent the historic and current EV buyers to some extent. Now growing EV sales past 20% will be encountering more cost-conscious mainstream buyers. I suspect in the near future slower growth rates than the past.
Regarding “EVs do not reduce emissions, at least not unless the new EV purchase displaces a new ICE vehicle purchase.” Perfect timing, I just posted a new analysis article finding that all the EVs sold to date have not reduced any pollution greater than the new gasoline vehicles. EVs will not start to reduce criteria pollutants until they are greater than 6.0 million, today in California, they are 1.15 million. It may surprise you to know that 40% of current gasoline models have lower NOx on the City Drive Cycle, and greater than 90% of the gasoline models have lower NOx on the Highway drive cycles. (This was generally true in 1998 as well).
The EV assumption error stems from the governmental policy of using average deteriorated lifetime emissions of new vehicles, and average vehicle fleet emissions, instead of comparing the cleanest gasoline vehicle populations at their cleanest emission rates and their replacement with EVs.
First, we need to have less than 6 million EVs on the way to exceeding 6 million EVs. Are you proposing that we stop selling EVs right now because along the way the emissions might be higher now than in a much longer future?
Second, it is completely appropriate to use average lifetime emission rates for ICEs because EV lifetime emission rates will not deteriorate over time–those are completely dependent on the electric system emission rates which do not deteriorate. Why compare unrealistic and temporary new car emission rates in this situation?
Third, and most importantly, the marginal emission rates from EVs are not the emissions for the last unit dispatched in the CAISO loading order. Gas units are not ramped up to meet EV charging load, especially during peak hours when gas units are already fully loaded. California has not built a new gas unit since 2010–all of the new generation has been zero emission renewables (much of this on rooftops.) Gas generation has declined 40% over the last dozen years. The true marginal generation for EVs is solar or wind added at the bottom of CAISO’s loading stack.
1. The article illustrates the magnitude of the emission averaging error of the EV policy and shows the errored analysis used to justify EV mandates and ICE Ban. The article shows one instance where the justification for the EV mandate was oversold to the public, and this new way of examining EVs finds a very real analysis omission.
2. The point of the article was that we have been ignoring the “short-term low emissions ICEVs. I can assure you as former CARB new vehicle certification staff EPA certification emission values are not “unrealistic” and examining the “temporary new car emission rates” was the point. How long does the cleanest car maintain their super clean emission rates needs further study in addition to the 2005 CeCERT study which confirmed extremely low levels for in use vehicles. The atmosphere and us breathers will not see one bit of air quality benefit until the EV population is cleaner than the cleanest gasoline model population. That was the point of Table 1.
3. I totally understood marginal emission rate issues. If you believe CARBs developed CA GREET overestimates emissions, please provide an alternative source. Gasoline cars and trucks have reduced NOx and PM emissions 98-99.5% from 1970 levels and power plants also have reduced emissions similarly from existing plants. CARB’s EMFAC 2017 shows Heavy Duty Diesels are 99.8% lower on PM than pre-1987 trucks, and they are 100% lower due to the engine consuming and cleaning up the particulates when driven on the road in dirty air.
So how do you propose that we bridge from 1.5M to 6M EVs if we don’t sell any in between? You haven’t proposed a viable alternative path forward. We may or may not have higher emissions during that interim period, but we will certainly have lower emissions in the end.
As for the GREET model emission rates, I’ve often proposed alternative methods in this model and other models and analyses, but CARB, CEC and CPUC staff have almost always refused to accept the proposals despite often acknowledging the problems with their assumptions and methods. The difficulty in changing these models are political, not technical.
Richard McCann, RE: “So how do you propose that we bridge from 1.5M to 6M EVs” The EV is a False Choice for clean air. The EV is cleaner than PZEV, ULEV, and several nominally certified ICEVs for the first 10-15 seconds of the cold start. After 15 seconds the ICEVs PZEV and ULEV at a minimum are cleaner on-air violation days and roads. The Clean Air Act authorizes EPA and CARB to address air violations . . . -these occur 20-40 days of the years. Modern ICEVs operated in air violation roads have significantly lower emissions due to the engines air consumption and emission controls concurrently reducing on-road air pollution in addition to the ICEVs additional contribution. In many cases ICEVs have net-negative criteria pollutant emissions. I presented these findings at last year’s COORDINATING RESEARCH COUNCIL workshop. The cleanest 20-40% Light and Heavy-duty vehicle emissions are zero or less when operated in air violation areas – using CARB EMFAC vehicle surveillance data.  As an example- you may have heard that several cities including LA experienced higher ozone and other air pollution during the COVID lockdowns. I also presented the 5-years of increased ozone pollution in LA during the 2008-2014 recession-reduced driving (fuel sales). ICEVs both create and resolve their own emissions to a point. Biogenic (soils) NO, NO2 emissions dominate mobile sources in several locations. CARB is unwilling to bring this into their emission inventory.  RE: alternative methods for GREET. Any suggestion is welcome. As an alternative I could use a Natural Gas Fired Turbine Power Plants even those in SCAQMD as an alternative? Below is an example of the varied CO2 rates that I would assume 80% recharging at low rates and 20% at mid-day??? I would have to build out a reasonable NOx and PM rates from this CO2.Â
Gary Yowell916-524-3794 Cell
Here is the complete article if you want to understand the new approach.
Not sure if the link will be posted properly.
Do EVs reduce NOx or PM emissions more than combustion engine vehicles? The answer may surprise you. – Stillwater Associates
October 16, 2023 By Gary Yowell From the 1990s to the present day, the California Air Resources Board (CARB) and the U.S. Environmental Protection Agency (EPA) have maintained that EVs have lower nitrogen oxide (NOx) and particulate matter (PM) emissions than internal combustion engine vehicles (ICEVs). In fact, California’s State Implementation Plan (SIP) relies heavily… Read More
This issue will be relevant for heat pumps when the hefty IRA subsidies kick in next year.
Leaving aside the question of “who gets the money”, the more fundamental question is what effect are EV subsidies actually having on EV sales — and on GHG emissions?
A relative of mine recently bought a Model 3, for which he got both the $7500 tax credit and California’s $2000 CVRP rebate (Clean Vehicle Rebate Project). I asked him whether his purchase decision was at all influenced by the $2000 rebate and he said absolutely not; he would have purchased the Model 3 with or without the rebate. As for the $7500 tax credit, that was a much tougher call; he could not say whether it was determinative to his purchase choice. So maybe his purchase was influenced by the tax credit, but what would his alternative action have been without the credit? If the alternative would have been to purchase a new ICE vehicle, then the credit had a positive emissions benefit. If he would have instead bought a used car, EV or ICE, or just kept his old car, then the tax credit would have subsidized increased emissions because it would have added a new vehicle to the existing fleet (low-emission, but not zero-emission) without a compensating reduction in ICE vehicle sales. The marginal surplus in used vehicles resulting from new-vehicle subsidies would not necessarily have resulted in increased vehicle scrappage; there is plenty of demand in the export market to absorb the surplus.
EVs are expected to reach price parity with ICE vehicles by 2025, at which point EVs will be cheaper and will have substantially lower operating costs and about 50% longer lifespan than ICE vehicles. At that point nobody will need a subsidy to motivate them to favor a new EV over a new ICE vehicle, and EV subsidies will just be financing expansion of the vehicle fleet — more EVs in the U.S. and more old, polluting clunkers being shipped to Mexico, Nigeria, … .
EVs do not reduce emissions, at least not unless the new EV purchase displaces a new ICE vehicle purchase. The only way to reduce vehicle emissions is by either curtailing the sale of new ICE vehicles or accelerating the early retirement of old ICE vehicles.
California is discontinuing its CVRP program and shifting its resources to “Clean Cars 4 All” (CC4A), which pays low-income consumers $9500 or more to scrap their old ICE vehicle and replace it with a cleaner alternative. CC4A, on average, subsidizes the scrappage of a model-year 2001 vehicle and replacing it with a model-year 2020 used car. It is unclear what emissions benefit there is from scrapping a car that is already well beyond its expected lifespan, and the subsidized purchase of a used, low-emission vehicle merely makes that vehicle unavailable to other buyers.
A more cost-effective way of phasing out ICE vehicles would be to convert them to EVs. Right now, EV conversions in California typically cost more than used EVs, but that situation might be changing. A European company, EV Evolution EU, for example, sells EV conversion kits for less than $10,000, and as EV battery costs come down (a 2X reduction is expected within a few years), EV conversions will become an increasingly more affordable option for low-income buyers. Currently, EV conversions do not qualify for CC4A rebates (or for the federal $4000 tax credit for used EV purchases), but that situation might change as policymakers begin to recognize the economic potential of this option.
The key to EV sales and reduced pollution is Bi-directional Charging where the EV battery can also power a home and the home having 150% output from rooftop solar panels. Being able to re-direct Solar power from the grid during high curtailment to the EV battery, then, tap that during high demand pricing will also help pay for the batteries in the EV and the replacement batteries when purchased. The EV battery is equivalent to 4 Tesla Powerwall’s costing more than an EV when purchased for the home.
Having done a scrappage study for the California Air Resources Board (admittedly a while ago), the scrappage effect created by inducing replacement of a fairly new used car with a brand new car is quite significant. In addition, shipping a used car over seas to anywhere but China is an expensive endeavor that probably costs more than what the car is worth. As for Mexico, even those used U.S. cars induce scrappage of used cars there–nothing runs indefinitely.
Another important factor is that the scrappage programs are aimed just as much at reducing criteria air pollutants, not just GHGs. Pre 2004 cars in particular are much higher emitting and getting those off the road is an important strategy.
Are there any other notable examples in the RE space where the suppliers capture the incentive. I noticed something like this in Germany. When the FiT rates ratcheted down the installation cost ratcheted down in step such the ROI/IRR stayed in consistent band. Arguably, the FiT rates and installation costs came down in lockstep because of the falling costs of modules, inverters etc.
Have installation costs of PV in California dropped in reaction to the removal of NEM?
Have installation costs in any other areas dropped in reaction to the removal of incentives for wind/solar/batteries/etc?
I scrapped my end of life Civic for $6000. Up here in BC this was a privately funded program called Scrap-It. This was definitely a motivator for getting my PHEV.
When my home burned, I contacted Tesla for a new quote to replace my Tesla Solar Roof. Having paid less than $60,000.00 in 2020 for the roof I was not expecting the replacment cost for a new Tesla Solar Roof to be $121,500.00 to get back the same roof. Power walls went up from $6,000.00 each to $12,000.00 for the first replacment Power Wall then $8,000.00 each for the next 3. Replacements do not get the IRA tax credits yet they keep pricing these things like they do making the replacment un-affordable under NEM-3.0. ROOFTOP SOLAR will disappear under NEM-3.0 and even existing NEM-1.0 and NEM-2.0 will run out and when it becomes time to buy the replacment, the companies, that originally sold them, will all be out of business. Just look at Nevada when they ended NEM and all the companies left the state.
Betazeded
I hope your insurance plan included a replacement rider!
Back in 2021, a couple of months after the Caldor fire we found out that our farm policy didn’t cover trees falling onto our fencing. A very old cottonwood tree twisted itself apart during a storm that took out a power pole across the street too. Three transformers ended up hanging in the air when the PG&E pole snaped. Power was out for 3 days.
I thought our insurance plan might have covered some of our costs to get the tree off the road- no such luck. That was the last power outage we experienced in CA as we moved before another storm took the local grid out for close to a week. If we had remained in CA we would have installed a 3.2Kw ground mounted system to work in concert with our existing 6.1kW system. Two Tesla batteries would have been added to suck up the extra juice during the day for use during peak times and overnight. I don’t know about you, but NEM 3 would have had me finding a way to give excess generation from the new system to our neighbors’ rather than send the energy to the grid and have PG&E credit us pennies for the juice.
\Hope:
1) your rebuilding plans are going smoothly.
2) That your fire wasn’t associated with a wildfire, or grid failure fire, as the FAIR plan insurance rates are very\ pricey
Mark Miller
Richard, I haven’t seen your scrappage study, but that could be very useful in quantifying the cost-effectiveness of CVRP and CC4A. CARB has also supported research on the impact of consumer choice (“rebate essentiality”) on the cost-effectiveness of vehicle rebates, but they don’t actually use any of that research in their cost calculations. The CVRP cost is rated at $193/MTCO2e and CC4A is $1,097/MTCO2e [California Climate Investments 2023 report, Appendix A], but CARB’s cost quantification methodology lacks any coherent rationale and is basically meaningless. [AQIP FY 2023-24 Funding Plan, Appendix A] LAO has criticized CARB’s methodology for not considering full lifecycle vehicle emissions and consumer choice. Moreover, the studies on rebate essentiality are defective, in that they are based on the survey question, “Would you have purchased/leased your [clean vehicle] without the state rebate?” That tells you nothing about whether the consumer would have alternatively bought a new ICE vehicle – the default assumption – or would have bought a used car or just kept their old car.
The quantified health benefits of reduced criteria pollutants from scrapped vehicles might vastly outweigh any GHG benefit, assuming that the consumer wasn’t just collecting a $9500 rebate on a vehicle that was about to be scrapped with or without the incentive. I’ve not seen any studies or data on the impact of CC4A on either scrappage rates or criteria pollutants.
In a paper I wrote earlier this year I said this about CC4A: “Since 2015 the program has subsidized approximately 15,000 vehicle replacements (1% of the state’s 1.3 million potentially eligible replacement vehicles), and has reportedly reduced GHG emissions by 85,000 MTCO2e (0.06% of annual statewide transportation emissions) at a cost of $1090/MTCO2e. Without a guiding standard of cost-effectiveness, policies such as ‘Clean Cars 4 All’ can only provide benefits for very few.”
Two things: First, the inefficiency of tax incentives and/or rebates — that is, the likelihood that the incentive/rebate beneficiary would’ve made the qualifying purchase even with no incentives — has been a challenge ever since the inception of tax/rebate incentives. That’s the nature of the beast. The goal is to design a program that minimizes the “I would’ve bought it anyway” problem. Yes, that’s much easier said than done.
As for EV conversion kits for “less than $10,000,” I’m strongly inclined to think that the labor costs would be huge given that one would have to remove the ICE and drive train. I just don’t see that as being economical now, if ever.
Labor costs could probably be ~$2000 with an efficient operation, using standardized conversion kits configured for specific vehicle models.
EVs are stacking up on lots because nobody can afford them & about 20% of the buying public wants them. Look for Lucid & Rivian to go down by the end of the year, and by next summer Ford & GM crawling before Congress begging for yet another bailout. I was reading about GMs Hummer EV, 6 tons & $144,000 sticker price. During a single quarter last year GM sold exactly 2 EV Humvs. This after spending $4 billion on a new factory.
I don’t know what’s the worst scamm, EVs or the ‘Transition to Renewables’ grift? Marine ‘Windfarms’ are already ‘dead in the water’ (no pun intended) on the east coast, and Evs are rusting in dealer’s lots.