America’s problem with electric vehicle charging
In our last post, we examined whether the lack of fast charging infrastructure was slowing sales of electric vehicles (“EVs”). While our findings were inconclusive, we wanted to take this line of thinking one step forward: if, hypothetically, charging infrastructure is a headwind for EV sales, what are the major headwinds for installing new, fast charging units?
Two obvious reasons exist. For one, this is a classic “chicken and egg” situation: property managers don’t want to install chargers if there are no electric vehicles to use them, and drivers don’t want to buy electric vehicles if there are no chargers to fuel them. Secondly, fast charging technology is still in early days, thus it remains expensive and takes time (typically ~30 minutes) to completely fuel a vehicle.
While these two factors are significant, we envision them naturally working themselves out as technology matures and the market develops. However, beyond these factors, more nuanced issues persist that are discouraging investment in fast charging infrastructure and may require more proactive solutions. Of these, we have identified four key headwinds, and ranked them (criteria for which is discussed later in this article) in the following order, from strongest to weakest:
1. Installation Investment Risk
2. Electricity Pricing
3. Low Utilization
4. Variation of Technology
In the remainder of this post, we detail each headwind as well as their potential solutions. We also rate each headwind based upon 1) impact; 2) difficulty of solutions; and 3) timing of solutions.
Headwind 1: Installation Investment Risk
The Problem: Investment in charging infrastructure can be a frightening endeavor for landlords or small business owners. The fixed costs of fast chargers are prohibitive: research has indicated that the total cost of a four-unit fast charging system is ~$205,000, including $30,000 for each unit, $20,000 for installation, $17,500 for utility service extensions, and $40,000 for a transformer. Moreover, the benefits of chargers to businesses are difficult to measure. Retail stores are often told that installing electric vehicle chargers may lead to increased sales or businesses that have chargers will increase employee retention, but these impacts can’t truly be calculated. Lastly, the technology is still in its early days and is changing rapidly. All these factors combined equate to a very risky investment for business owners considering purchasing a charger.
Possible Solutions: Charging unit manufacturers are attempting to minimize this investment risk for potential business customers. For one, manufacturers have offered leasing programs for a while now, reducing the upfront cost. More recently, EVgo unveiled a mobile fast charger that effectively eliminates fixed costs, allows for property owners to test out the technology before fully committing, and prevents a property from being stuck with the asset in the case the owner or renter changes.
In addition, governments can provide a great deal of support to minimize investment risk. Clearly, financial support in the form of tax rebates and subsidies are a strong driver for charging infrastructure, and the list of incentives in the United States is lengthy. Governments are also increasingly committing to cutting greenhouse emissions and adopting electric vehicles. These commitments create more certainty around the growth of electric vehicles and the demand for fast charging, thus lessening the associated investment risk. In our minds, however, it is unlikely that these programs are expanded further, thus incentives may not be accelerate the pace of charger installation going forward.
Investment risk is a strong headwind for fast charging infrastructure given high costs, uncertain benefits, and changing landscape. A few solutions exist, such as pricing programs and government incentives, but the one that may be the biggest driver, technological innovation, will likely take some time to develop.
Headwind 2: Electricity Pricing
The problem: Utility bills for commercial customers have two primary components: energy and demand charges. Energy charges are based on the total amount of energy used during a billing period, typically billed on a $/kWh basis. Demand charges, on the other hand, are based on the peak electricity usage of a customer during a billing period over a 15-minute interval and measured in kW. The rationale for these demand charges is to maintain sufficient capacity: the grid is only as good as the amount of electricity it can transmit during peak periods (otherwise, blackouts occur).
If a customer draws a large amount of electricity at a constant rate, then this demand charge will be a small slice of its energy bill. On the other hand, if a customer draws a large amount of electricity in short bursts, then the demand charge can be significant. Unfortunately, EV chargers fall under the latter category: chargers are used infrequently, but, when they are, they transmit a large amount of power. The below chart, displaying the hourly utilization of an individual EVgo host site, illustrates the infrequency and variance in faster charger use:
As a result, demand charges make up a significant portion — up to 94% for some sites in San Diego — of charging units’ energy bills. The issue stemming from this phenomenon is a revenue-cost mismatch. Chargers earn revenue through energy sales (or use), but costs are primarily based on peak energy demand. Moreover, simply raising prices to alleviate this issue is unlikely as it would erode competitiveness to substitutes like home charging and gas refueling of internal combustion engine vehicles.
Possible solutions: Mitigating this headwind appears to be in utilities’ hands, since, as mentioned earlier, charging stations are hamstrung to increase prices in order to compete with substitutes. And utilities are starting to take action with rate reforms. Many are offering discounts to demand charges, while some are taking more innovative approaches. PG&E, for instance, has proposed to replace demand charges with monthly fixed rates for certain levels of energy use, similar to a phone subscription plan, which, the utility estimates, will have sizable savings for EV chargers.
While there has been progress on reforming demand chargers, it has mostly been with a few utilities in more progressive states. We believe these other regions will likely be slower to adopt new pricing structures, until electric vehicle adoption hits critical mass. And, given that the solution is contingent upon regulation change, it is natural that this will take time.
The downside of these reforms is that they may aggravate the grid capacity problem that demand charges were ultimately created to solve. This is why utilities are also exploring time-of-use rates, which would require charging stations to pay more for energy use during peak periods of demand for the entire grid (~4–6pm). As stations pass this charge through to end-customers, charging would shift to off-peak periods, lessening the load on grid capacity. These rates may also reduce demand for fast charging as drivers decide to charger later in the day at home; however, since charging demand is relatively distributed compared to grid demand (as seen by the utilization graph above), we don’t believe this to be a major factor.
In summary, demand charges are significantly hampering profitability (and, thus, installation) of fast chargers. It will be up to utilities to employ a combination of demand charge discounts and time-of-use rates to mitigate this headwind. While some are already doing so, it will likely take time for others to follow suit.
Headwind 3: Low Utilization
The Problem: A charger utilization rate is the percentage of time that a charging port is actively fueling a vehicle. There is little public data available on fast charging utilization, but the majority of the data points we have found point towards utilization ranging from 5–15%. On one end of the spectrum, EVgo shared hourly utilization for one of its chargers, implying a 15% aggregate utilization. Conversely, New York State published a report back in 2015 that found its fast charging stations to have an average utilization of 3–4%. In contrast, gas pumps have been found to have utilization of 34%.
Regardless of the exact rate, utilization is too low for financially-viable fast charging. As you can see below, a Level 3 charger would require ~20% utilization to achieve cost parity with fueling a 24mpg (about the current US average) ICE vehicle. Significantly higher increases in utilization are required to compete with residential charging and more fuel efficient ICE vehicles.
This issue is amplified by the nature of charging. Any time a car has completed a charge, but remains idly plugged into the equipment, it is using up space without generating revenue for the charger owner. Moreover, because charging takes time (~30 minutes for a full charge), many vehicle owners engage in other activities during a charge (e.g., grocery shopping). The end result being that owners do not unplug their vehicles and reopen chargers at the exact time a charge is completed, translating to lower utilization.
Possible solutions: Some of this issue may be alleviated by having more EVs on the road, but it is still an open question as to whether new vehicles will use fast charging given the current heavy reliance on home charging.
Beyond that, the largest opportunity we see to increase utilization is to target fleet customers. EVs have low operational costs compared to ICE vehicles, making them ideal for owners that drive frequently. One type of owner that fits this bill is the fleet owner: fleets, like delivery and taxi companies, run vehicles for much longer periods of time versus the average car owner, and, therefore, would benefit the most from the lower operational expense of EVs. They are also the perfect customers for fast charging operators. Fleets represent a large, centralized supply of vehicles that require constant fueling and do not have the option to use cheaper residential chargers. On top of that, fleets tend to charge during off-peak hours: for instance, EVgo found that rideshare customers charge most often from 12–2pm and 8–9pm. This has the added benefit of putting less stress on the grid, avoiding time-of-use rates, and complementing demand from retail EV owners who charge most frequently ~5pm.
In order to increase utilization, charging operators should secure long-term contracts with fleets as well as offer exclusive access to chargers for certain times of the day for a price premium. This transition to electric fleets is starting to happen — more than a third of EVgo’s business in 2018 went to fleets — and it should only accelerate moving forward as more fleets look to go electric as a result of financial and regulatory incentives.
Another potential solution for low utilization is through subscriptions or loyalty programs, which charger operators can employ to lock in demand upfront. Some are starting to explore this model already, but it is unclear how successful they have been to date. In fact, some research suggests that subscriptions aren’t a viable option: one specific report found that fast chargers would require a substantial subscription fee of over $100 / month to simply break even on average energy usage, and, even then, customers would have an incentive to charge more, given the marginal cost would be zero. As a reference, EVgo’s membership charges only $7.99 / month, but still charges for electricity, albeit at a $0.03–0.04 discount.
Overall, utilization is a major obstacle to installation of fast charging infrastructure, but we feel that emergence of electric fleets will be a large driver to overcome this issue over the next several years.
Headwind 4: Variation of Technology
The Problem: Largely attributed to being a nascent industry, a variety of different fast charging technologies currently exist on the market. One such variation is connector type:
The cause for this differentiation lies within geographic differences in electricity transmission standards. Thankfully, the major EV charging unit manufacturers have largely solved this issue by including multiple connector types on chargers, the notable exception being Tesla’s proprietary network.
Where variation of technology truly comes into play is the large number of charging networks. These can be thought of as equivalents to gas station brands (e.g., Shell, Exxon). The issue stemming from this variation is that networks often require memberships (typically free), physical membership cards, and mobile apps, effectively fragmenting the market. The result is that, while a good deal of fast chargers exist, the supply is constrained due to network silos. While we do believe this is a headwind for fast charging infrastructure, we don’t think its impact is significant, at least compared to the other headwinds identified.
Possible Solutions: We see two paths to fixing this issue. The first being an over-the-top solution that allows users to locate and access chargers from any network. For example, ChargeHub is an excellent resource that aggregates charging locations of all network providers. Secondly, it seems likely that charging networks will consolidate in order to capture economies of scale and lock consumers into larger networks.
All in all, while the landscape for fast charging contains many different technologies, we believe the impact of this headwind to be slight, and the likely solutions to be realizable in the near-term.
Conclusion
Our premise when we started this post was that the challenges of building fast charging infrastructure may be causing consumers who would otherwise want an EV to delay their purchase. After evaluating the four key headwinds to EV charging, we believe they rank in the following order, from strongest to weakest:
We are certainly driving towards an electric vehicle revolution, but these structural issues around charging infrastructure may block the road ahead. We believe that implementing some of the solutions we have presented in this article will be critical to realizing the full potential of the EV market.
