"What happens to gasoline prices when EV vehicles become prevalent" was the question raised in one of the comments in response to my earlier article on volatile raw materials for electric vehicle manufacturers. The question came in a very categorical (notice "when" and not "if") and matter-of-fact manner, triggering thoughts that took many days to organize, quantify to the extent possible and present in a brief text.
The problem with this question is the inherent lack of appreciation of the myriad of complexities on the way to becoming a dominant technology. The typical line of thought is: "adoption of electric vehicles increases exponentially as Tesla (NASDAQ:TSLA) (others are seldom mentioned) scales production and costs fall, therefore the demand for gasoline, and by extension - oil, disappears." As it happens, the question coincided with an article on Bloomberg titled "Energy Giant Shell Says Oil Demand Could Peak in Just Five Years," lending credibility to the question. Upon a cursory look, of course, it transpires that a brief remark by Shell's (NYSE:RDS.A) RDS.B0 CFO that oil demand may peak in somewhere between 5 and 15 years (his main point being that supply will not peak before demand), is rephrased into something altogether different. Let's also keep in mind that Shell has long decided to concentrate on natural gas, which makes such a statement politically more palatable.
Accepting the notion that demand for oil has disappeared assumes that we have answers to at least three issues of epic proportions: mass market adoption of electric vehicles (EVs), the path to superior total cost of ownership of EVs, and the most difficult one of all: the supply chain consequences.
Mass market adoption of EVs
Let us first take stock of the current situation. Despite decades of research, development and production (EV1 1996?), mass market adoption of EVs is still to happen. Bringing hard facts to the table often does curb the enthusiasm. According to the International Energy Agency, in 2015 there were 1.15 million plug-in hybrid and fully electric vehicles on the road globally. In the same year 88 million new vehicles were sold. To be clear, 88 million is only the added number of vehicles and not the existing fleet. There are somewhere around 1.2 billion vehicles globally, which means currently EVs are no more than 0.1% of the global fleet. Talking about the end of oil is a little premature?
The number of EVs is of course growing faster than the overall fleet, with the best-in-class Tesla posting ~ 50% year-on-year growth numbers in unit sales. Not arguing the validity of these numbers, by extrapolating this type of growth and average GDP growth rate for the overall fleet we find the insignificant impact zone - at least ten years from now - so long as the share of EVs in the total fleet is below ~5%, which is until about 2026, the question on the impact on oil demand is largely theoretical.
To be sure, plotting the line above is a futile exercise in extrapolation: the industry will hit a number of hurdles way before the 5% level is reached. Chiefly among them is the cost. Tesla's Model 3 is supposed to retail at $35,000, which is incidentally a number with great significance: the average vehicle price sold in the US was $33,560 in 2015 - i.e. Tesla is only one model away from the mass market. Or is it?
Unfortunately, as any student of statistics quickly finds out, averages (means) are very dangerous numbers, particularly if there are many outliers. Very few high-priced vehicles sold in the luxury segment have a disproportionate impact on the average number. Taking out European luxury brands from sales numbers leaves 92.5% of all vehicles. Although there is no vehicle specific pricing data to arrive at the median value, the number is likely to be in the mid-twenty thousand range.
Source: WSJ Car Sales Data
As it happens, a median income household in the U.S. cannot possibly afford the average price tag. The typical conclusion is that most families overspend on cars. But if we use the prudent affordability rule of thumb (20% down payment, four-year financing and outlays of no more than 10% of annual income), we arrive at a vehicle budget for a family with two median earners of $28,500. Since a family owns 1.9 cars in the US and one vehicle is likely to have been purchased used, the real sustainable mass market price can be stretched only to $20-25,000 - a far cry from the idea of mass market for Tesla.
This means we are at least a generation or two (10-15 years) away from mass market, which reminds me Amara's law: We tend to overestimate the effect of a technology in the short run and underestimate the effect in the long run.
The path to superior TCO
Despite most people's claims that consumers are irrational and buy predominantly based on impulse, hard data once again suggests the opposite. Consumers prefer larger, more comfortable and safer vehicles, but not at any cost. In fact the vehicle preference is highly correlated with the price of gasoline. The recent fall in oil price was accompanied by a significant shift towards large SUVs/trucks, which suggests that consumers not only look at the sticker price but also take into account the long-term total costs of ownership (TCO).
What is more interesting, such a strong correlation exists only when oil prices are over $60 per barrel. When prices were below $60, fuel economy was largely irrelevant and consumers generally preferred large vehicles.
Key inputs for calculating the TCO are the gasoline price, price of electricity and the residual value(RV) of the vehicle.
Unfortunately, the more successful EVs are in approaching the critical 5%-mark, the more volatile the key inputs for TCO calculation become.
In order to grasp the complexity of this issue let's examine a scenario: an individual makes a decision to buy an EV based on oil price of $50 per barrel and current residual value of $15,000. As TCO is in favor of EVs, more people make the same decision. The market catches on and oil prices fall due to the expected lower demand. Lower oil prices also affect the residual values of electric vehicles, so that the recalculated TCO becomes negative for EVs. Such a dynamic "disequilibrium" is typical for adoption cycles and is a huge hurdle for the new drive train.
How many of these yoyo moves it will take to approach mass market for BEVs is anyone's guess. However, every zig in the wrong direction would mean that a technology in its infancy has an uphill struggle to survive and fund itself. Companies with both electric and ICE products (Ford (NYSE:F), GM (NYSE:GM) etc.) will have significantly more stable basis to compete.
Supply chain consequences
While I do believe that the first two hurdles are probably manageable in the long term (over 10 years), the third one leaves more open questions still to be answered. The use of plastics in vehicles has been increasing over time not least because of the need to save weight and make them more fuel efficient. In value terms, ~8% of a vehicle's value is plastic materials: polypropylene (PP), polyurethane (PUR), polyvinylchloride (PVC) etc. Notice the common theme? These are all oil (or natural gas) derivatives (see the table, a bit outdated now, actual use may be much higher).
Source: Institute of Production Technologies, Slovakia Click to enlarge
Some (minority) of these materials can be commercially feasibly replaced by alternatives. Majority, at least in the automotive market, do not have alternatives.
Now getting back to our scenario, let us examine step two, when oil prices go down to $25 per barrel. Good news for plastics, of course: as oil prices come down, major resin prices follow (nearly 0.9 correlation there), but up to a point. If gasoline prices follow suit and remain sustainably low, oil refinery becomes loss-making. This results in lower oil volumes and by default, lower naphtha volumes - a by-product of refining and the key feedstock for plastics. Prices for gasoline may remain low, but prices for naphtha will skyrocket. Doubling plastics costs could result in high single-digit cost increase for car manufacturers, which is the size of the profit margin. In essence, achieving a critical level in adoption will trigger additional costs for vehicle manufacturers and make it harder to progress.
There are two key messages to be clearly understood by investors in pure-play EVs. First, the replacement of the drive train is not just a technological change, it is a drastic departure from the existing ecosystem in multiple markets. It takes a long time for markets to settle and find a combination of clearing prices and technology that work for all participants. This is not to say it will not happen, but we do still underestimate how long it will take and by extension - how much it will cost to sustain a pure-play loss-making company.
The second point is directed to the early adopters who (rightly so) smile and recall the quick demise of CDs and analogue photography as examples of just how quickly things can change. There is an almost tragic difference between music players, cameras (and Apple products) on one hand and automobiles. The latter is a capital investment item for companies and households, with significantly different life cycles. In this sense, unlike other consumer goods, cars are similar to planes and trains. Their cost is just too high to allow a quick change. Given the required time and complexity as well as likely volatility in multiple market prices for raw materials, OEMs with broad mix of technologies are better placed to ride the wave.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.