Article Thumbnail
The case for "Golden Chargers"

October 25th, 2023 by Ole Henrik Hannisdahl

In a recent poll conducted by the newsletter Electric Avenue, 55% of respondents said they were in favor of reservation on public charging station. The newsletter targets industry professionals, and the close call in the survey reflects the topic’s controversy. What if we could solve the reservation use-case with a “Golden Charger” instead?

First, some comments on reservation in general:

  • Reservation is most valuable for the customer when a station is saturated (to avoid queuing)
  • When a station is saturated, there are cars physically queuing. A fancy reservation system is useless unless you can also manage the physical queue on-site somehow.
  • Unless you have good physical queuing management, reservation is likely to reduce throughput relative to a first come first serve system. This is probably not what you want at a saturated station.
  • Most customers queuing at a fast charging station in a saturation scenario are likely (relatively) inexperienced and won’t know about the reservation system beforehand. In addition to creating a bit of chaos, this may also raise temperatures on-site when people who have been queuing for a while are being bypassed by smug pros who are paying to get ahead.

For the abovementioned reasons, plus a few more technical ones, most CPOs are very hesitant to allow reservation except in a few very defined usecases such as dedicated taxi sites etc.

However, I think there are a few alternative solutions that might be worth testing. Most notably, taking lead from the airlines who have mastered loyalty programs and perks, would be to designate a charger on popular sites as a “gold member” charger. The charger cannot be reserved per se, but can only be used by customers who have reached a certain membership tier. This charger could be physically branded / signed so that other users would know not to queue there (think “Fast track” lanes at airport security), and since gold members would still have to queue there in the order they arrive, most of the physical and digital queueing issues are avoided.

This would stimulate frequent high-value customers to choose your network also in off-peak hours, and could overall generate higher customer satisfaction and more revenue than a pay – to – get – ahead reservation model.

The original poll can be found here

Article Thumbnail
What does it take for a CPO to be profitable?

July 24th, 2023 by Ole Henrik Hannisdahl

Out of all the companies out there owning and operating charging networks, none are profitable, and most burn through cash at an alarming rate. Yet, we see investors and analysts pricing CPOs at up to a whopping 70 times book value. What does it take for a CPO to make money in the long run? What do we have to believe in to justify these valuations? In this article, we break down the main components of a generic CPO financial model to try to answer these questions.

Firstly, a little bit of background. For those unfamiliar with the EV charging domain, “CPO” is most of the time short for “Charge Point Operator”. This usually, but not always, is the same company that owns the chargers, and CPO can also be short for “Charge Point Owner”. However, owning a charge point and operating a charge point are two entirely different things. Spoiler alert: The Operator side of a CPO is a fairly straightforward game, whereas the Owner side is where things really get interesting. The Operator side essentially creates the company’s EBITDA, whereas the Owner side needs to worry about asset value, and hence, EBIT. We will return to this later in the article. But first, let us have a look at the main revenue and cost components of a typical Operator:

Charging revenue

This is usually the main revenue source, and in many cases the only revenue source, for Operators. Some will also make money providing services to third party Charge Point Owners, or even through certificates trading in certain markets. More on that later on.

Charging revenue for an Operator comes through two main channels:

  • Drop-in payment. This is when unregistered customers pay directly at the charger, with bank cards or mobile payment.
  • EMSP revenue.

For those unfamiliar with this term, an EMSP stands for Electric Mobility Service Provider, sometimes just denoted EMP or MSP. In essence, EMSPs own customers, while CPOs own chargers. For an analogy, think about Mobile network infrastructure such as towers and core network servers (CPOs) and Mobile Operators (EMSPs) which sell you subscriptions across various network operators. Typically, a CPO will negotiate bilateral deals with larger EMSPs, or just list their network on platforms such as DCS, Hubject or Girève, which aims to connect EMSPs and CPOs. Some Operators also run their own EMSP, either exclusively or in parallel with other EMSPs. We could do a whole series of articles just on this, and we just might in the future.

For now, let’s simplify it: Drop-in revenue is from unregistered customers who just show up at a charger and want to charge. EMSP revenue is from customers who have some sort of subscription or registration. How charging is priced across customer segments is a whole other ballgame, that we may touch on in another article. But for now, let’s consider that most charging sessions are priced per kWh, i.e., as a direct function of the amount of electricity delivered to the end customer. The industry has gradually converged on this price model in the later years.

Now, let’s turn to cost components:

Cost of electricity

Electricity cost
How CFOs in CPOs felt about the electricity bill during the second half of 2022

This is simply the CoGS (Cost of Goods Sold). When calculating the total cost of this, you need to account for both the cost of electricity in the period it was delivered, as well as grid fees. Grid fees usually consist of a fixed element and a variable element that can depend on peak load, total energy delivered, time of use, and other factors. This cost can vary wildly across seasons, time of day and geography. Some Operators pass the spot price plus a margin directly on to end customers, but for practical and regulatory reasons (and because many Operators have locked in fixed prices to EMSPs), most choose to just average it out over time, adjusting customer pricing periodically based on cost forecasting. Some will try to hedge this, usually through a direct purchase agreement outside of the spot marked (called a PPA).

Transaction cost

The cost of processing drop-in transactions. Usually, the Operator will team up with a payment provider such as Adyen or similar, which clears payments for a fee. For EMSP revenue, the transaction cost is on the EMSP side.

Maintenance cost

Maintenance cost
The repair guy has arrived!

This is the cost of scheduled and unscheduled maintenance and repairs. Typically, all fast chargers will require at least a yearly maintenance visit. In addition, there is all the somewhat unbelievable stuff that can happen to chargers in the field. In short, if you can imagine it, you can be sure it has happened. Is it possible to crash so that a Polestar lands upside down on top of a charger? You betcha. Do you think a CCS cable is strong enough to pull a 500kg charger off its concrete foundation and on to the hood of an adjacent Nissan LEAF if the cable gets entangled on the towing hitch of an Audi e-Tron, and the driver just floors it? Absolutely. The list goes on. Be it in the form of direct cost, warranty handling, insurance payments, revenue loss or otherwise, all this stuff ends up costing money that must be accounted for.

Customer service

Customer service
The inspiration board for legacy fast charger UX

Every month, thousands of people buy an EV for the first time. At some point, these people will find themselves in front of a fast charger for the first time, with no clue about what to do. Imagine pulling into a petrol station for the first time, in a car you have no prior knowledge of, trying to work out where the fuel filler cap is located, or what the difference between diesel and petrol is. Our parents usually taught us all this when we learned how to drive, but EV drivers mostly have not had this luxury (yet). Charging user experience is another topic we can write a whole other series of articles about, but for now, let’s just underline the importance of having good customer service.

Site rental / revenue sharing

Are you an analyst trying to make sense of a CPO earnings forecast? Are you enthusiastic about their EBITDA margins? Look again, and try to find out what, if any, the CPO is planning to pay property owners. Chances are that they are greatly underestimating this component going forward. Unless the CPO is the property owner (think about petrol stations owned by a fuel retailer such as Circle K or BP), getting to build and own chargers at a good location will likely come at a higher and higher price. In the early days, CPOs could gain access to good sites simply by being willing to invest. However, as property owners and real estate companies become more and more aware of the value of their parking lot, and as new CPOs with deep pockets are trying to pay their way to critical mass, expect the power balance to tip in favor of property owners. We should probably write an article about this trend as well, at some point.

SaaS licenses

In addition to all the generic business IT systems, an Operator needs a system called CPMS: “Charge Point Management System”. Large legacy Operators may have built this themselves, and for some (like the company ChargePoint), this is part of their core business. For these companies, the systems are activated in the balance and typically written off over 3-5 years. However, the industry is pivoting to dedicated platforms such as Driivz, Greenflux, Virta or similar, payable through license fees.

Allocated overhead

The CPO may be part of a larger company with multiple business areas, or it may be a standalone company having to support the full cost of things such as accounting, marketing, HR, offices, etc.)

Net profit from adjacent business areas

Alt revenue
Don't worry, we can always sell credits!

We mentioned two other forms of revenue earlier: Services, and certificates. Let’s briefly look at these.

Services will typically mean building or operating chargers for others and charging a fee for it. CPOs are set up with processes and people to convert investor cash into chargers on the ground, and occasionally choose to hire out these capacities to third parties. This will typically be part of a deal with property owners, where the property owner owns the infrastructure themselves and the CPO builds and operates it for a fee. While this can provide some welcome revenue in the short run, as well as open for an asset-light market entry strategy for newcomers, the downside is that you might be building your own competitor. Over time we expect Charge Point Operation to become more and more of a commodity, which has some interesting strategic implications we may touch on in another article.

Certificates trading is a niche revenue source in certain markets such as California, usually designed to accelerate the transition to renewable energy. In such markets, CPOs may earn money by selling certificates or credits to other players who are obliged to buy them for regulatory reasons. Tesla famously earned significant money on this in critical periods, both on credits from cars sold and on credits from energy delivered through their supercharging network.

EBITDA

When we add up all the components above, we get the CPO’s operational profit, or EBITDA. For a larger CPO with healthy margin on CoGS, as well as some economies of scale on other cost components, the EBITDA margin could be in the 25-35% range, which is very decent. So all is well, then? A CPO with a healthy EBITDA ticks all the boxes: An operationally profitable company, solidly in the ESG sector, riding a megatrend that more or less guarantees hefty growth figures for the next decade and beyond? Unfortunately, that is only half the story. Let’s turn to the Owner side of things and look at EBIT.

The tricky “D” in “EBITDA”

EBITDA
The universal excuse in every company report that's short of expectations

Building and owning a charging network is essentially about digging money into the ground. Charge Point Owners are asset-heavy companies constantly looking for sites to build on. Once you commit to a site, roughly half of your investment will be sunk cost, namely the cost of everything below ground. All the stuff above ground can in theory be picked up and redeployed to a new site, but in reality, this almost never happens.

For all practical purposes then, when you invest in a site, the money you deploy must be recuperated through charging revenues from that site. While revenue sorts under EBITDA, the actual investments in the charging network sorts under the “D”, for “Depreciation” (remember that EBITDA stands for “Earnings Before Interest, Taxes, Depreciation and Amortization”). Financially, your network will typically have to be written off in anywhere from 5 to 10 years, depending on which audit standard you follow, the specific site contracts, your auditor’s mood, and other factors. This static, yearly depreciation is what you have to account for in the company’s balance sheet, and consequently, in the EBIT.

Let’s look at what this actually means using a very simplified example. Let’s say you want to invest in one fast charging bay at the cost of 50.000€ (a charging bay is CPO speak for the ability to charge one car. A site with 5 charging bays can charge 5 cars simultaneously). Let’s say your auditor agrees to a 10 year depreciation period. This gives you a yearly depreciation of 5.000€. And let’s say that the CPO’s EBITDA margin is 30%. This means that in order to account for the depreciation euro-for-euro, the site needs an annual turnover of (5.000 / 30%) = 16.667€ to break even. This should form the basis for your investment decision. Simple, right?

Unfortunately not, for three reasons:

1) Discount rate (IRR)
2) Uneven distribution of revenue over the depreciation period
3) Market saturation

Discount rate: The reduced value of future cash

IRR
Idea: Check. Investment: Check. Future revenue: Uncertain

All money comes at a price. In addition, all future revenue is risky, i.e. less than certain. These two basic truths form the basis of Internal Rate of Return, or IRR for short. IRR is a percentage that is the sum of two components.

The first component, the price of money, is the company’s weighted average cost of capital (WACC). This is essentially what the company has to pay in interest on its debt financing, and the expected return on equity from its investors. Depending on the company’s financial structure, the WACC will increase when general interest rate increases, and vice versa.

The second component, the risk factor of future revenue, is trickier. Its actual value is usually one of the more closely guarded company secrets. Calculating it means looking at a bunch of factors such as the sector you are investing in, competition, fundamentals and more. In the end, you are left with what’s usually referred to as the company β (beta). If you look at historical data for a company, you can calculate the β using regression analysis. But if you are trying to calculate the future β of a company in a new sector, such as for CPOs, you need to get a little more creative.

When we have both the components, we have the IRR, which is used to discount the value of future cashflow. What this means is that as we look farther into the future, the value of our revenue estimates becomes more discounted, i.e. worth less. Another way of looking at it, is that the IRR allows investors, or a company, to compare different investments opportunities with different profiles, and invest in those alternatives which provides the highest return on capital employed. We will show you a practical example of this in a minute, after we have had a look at our next issue: Revenue variation.

Uneven distribution of revenue

Now that we know how to find the value of future cash, we can look at the cashflow for a charging station over time. Essentially, all CPOs bet on increased usage over the lifespan of a charging station, simply because there will be more and more EVs on the road. This is a fair assumption, however with a couple of important caveats that we will look at in the next section. But for now, let’s assume that a charging station will see more and more usage every year. Typically, this means that a charging station will not cover its own depreciation euro-for-euro in the first years of operation, but that this is offset by higher revenues in later years. On a cashflow basis, this means that the station be cashflow positive over its lifespan. However, when adjusting for IRR, the picture changes drastically.

Consider the example we used above, with a 50.000 € charging bay depreciated over 10 years. Let’s add a hypothetical linear ramp-up in operational profit, and let’s use a fictional IRR of 12%:

Year Balance Depreciation Operational profit IRR - adjusted operational profit
0 50 000
1 45 000 (5 000) 3 500 3 125
2 40 000 (5 000) 4 000 3 189
3 35 000 (5 000) 4 500 3 203
4 30 000 (5 000) 5 000 3 178
5 25 000 (5 000) 5 500 3 121
6 20 000 (5 000) 6 000 3 040
7 15 000 (5 000) 6 500 2 940
8 10 000 (5 000) 7 000 2 827
9 5 000 (5 000) 7 500 2 705
10 - (5 000) 8 000 2 576
SUM (50 000) 57 500 29 903

On a cashflow basis, the station will generate 57.500€ over its 10 – year investment lifespan, which equals a 15% return over 10 years. Not exactly great to begin with, but when we adjust for IRR, the case falls apart. As most of the station’s revenue is in the future, and therefore highly discounted, the Net Present Value (NPV) of the cashflow is a measly 29.903€ measured in today’s investor money. Compared to an investment of 50.000€, this obviously does not add up.

There are two straight-forward fixes to this problem: Reduce the cost of building the station, and / or increase the revenue from the station. We will get back to this in our conclusion, after we look at our final issue: Market saturation.

Market saturation: How much electricity does an EV actually need?

EV charging is essentially a zero-sum game. As a rule-of-thumb, the real-world, year-round average consumption across the BEV (Battery Electric Vehicle) fleet is roughly 0,2kWh per kilometer. If you have a market with 100.000 BEVs, each driving an annual average of 12.000 km, you get a total energy consumption of (100.000 * 12.000 * 0,2) = 240 000 000 kWh, or 240 GWh. Unless you somehow manage to get everyone to drive more, or if you manage to convince OEMs to make less energy-efficient cars, the only real variable left to determine the total charging market volume is the number of BEVs on the road. Then you just need to split this volume up in to charging segments and leave CPOs to fight for market share in each segment.

Segmentation and segment volumes are topics that we should probably add to the increasingly long list of future articles we need to write. There are many ways to segment the charging market, but for now, let’s just leave it at this: Regardless of how you segment it, you are going to share the relatively fixed volume in any given segment with your competitors. This means that, all other things equal, your utilization rate per station will be heavily influenced by how much your competitors are building. And, as we saw earlier: Once a station is built, as long as it has positive EBITDA, the station will continue to operate.

You would think that the over-establishment of chargers is automatically avoided by means of rational investors: If you do the math, you should be able to work out when the market is close to being saturated. It would generally make little sense to deploy more capital at that point, since the value of your future revenue would not cover your current investment. However, for many different reasons, this is not always the case. This dynamic is also found in other markets such as the VLCC shipping market, which is largely cyclical. Investors pour money in to building new ships, only to find that freight rates collapse when too much tonnage enter the market at the same time. We might do a deep-dive in to various CPO strategies in a future article, but for now, let’s highlight a single factor: Location.

The bottom line: Location, baby! (as well as cost and contracts)

So you have worked your way through a long article, and you are still reading. It is time to reward your effort with an answer to the original question: What does it take for a CPO to be profitable?

Let’s assume that you are a reasonably decent Operator, with a given EBITDA margin. As an Owner, let’s assume that your IRR is given. As we mentioned above, the two main levers you can pull to increase profitability will be charging revenue and construction cost.

Earlier, we gave an example of a cashflow profile of a charging bay. In a zero-sum market, the average revenue per charging bay is a direct function of the number of cars and the total number of charging bays. But what if all charging bays are not equal? What if some charging bays can recuperate an above-average revenue (at the expense of other charging bays, which will then have to earn below-average revenue)? Every CPO knows from its own data that customers prefer some locations over others. Every CPO has done extensive modelling on this. And most CPOs will believe that they have cracked the code to finding above-average locations.

This not only solves the revenue problem, as you can beef up your estimates using a multiplier reflecting the value of having premium sites, but it also solves the saturation problem. It allows you to argue that in a saturated market most of the chargers constructed by newcomers at later stages will be at sub-par locations, since all the premium locations are already taken by the incumbents. And, because you can show from data that customers prefer your premium sites, you can argue that the revenue stream from these sites will hold up even if the market gets saturated, while locations built by newcomers will suffer.

Newcomers, on the other hand, can pull the cost card. They can argue that the incumbents have a higher cost per bay due to legacy hardware and technology, while newcomers can take advantage of years of iterative improvements and know-how to be more cost efficient when building out their networks from scratch. They do not have a tail of earlier mistakes such as malfunctioning chargers lacking proper load balancing and good user interface, or just wrong site selection in the early days, to pay for. All this, and more, means they can break even at a lower revenue per bay, both due to lower construction cost and higher EBITDA margin.

Common to all players it that contract length can be considered a silver bullet. If you manage to increase your contract length for a site from 10 to 15 years, you might get your auditor to let you depreciate certain elements of the site, such as substations and cabling, over a longer period. You also get a longer revenue tail, albeit heavily discounted. And to top it off, you may have locked in your revenue sharing with the siter owner at a reasonable, and not the least, predictable, level for a long time.

To sum it up: If you have good locations, long contracts, cost-efficient construction, and lean operations, you are positioned well as a CPO. However, you are still vulnerable to market saturation and the moves of your competitors.

And at some point in the future, you will have to renegotiate all your site contracts, in competition with other CPOs eager to take over your premium sites, or even the property owner deciding to just take over the site themselves. We will get back to a few scenarios for how this could play out in a later article.

Epilogue: How to justify a price / book ratio of 73?

This is actually more of a question that an answer. The example is taken from Allego, which is one of the few publicly listed CPOs out there. That means they have to report all their numbers in a way which make them easy to parse, as opposed to unlisted companies which require you to parse through a bunch of reports, or even CPOs who are part of a company structure that allows them to not really report anything interesting at all.

On July 12, 2023, Allego’s market cap was USD 743 M, with a stock price of 2,8. Shareholder’s equity was USD 27 M, which means that the market priced the equity at 27,5 times its book value (hence “price / book ratio”, or P/B in short). While this in itself is a high valuation, the average price target between 4 analysts that track the company was 7,4. Using the analyst consensus as a basis, we get a target market cap of USD 1 964 M, giving us a P/B ratio of 73.

The company has consistently posted losses both at the EBITDA and EBIT level and is burning through cash. It has raised funding multiple times, and at its current burn rate it looks set to have to do it again soon. And yet, analysts believe the company should be priced at 73 times its book value. I personally find this very hard to understand. If I was to try to understand it, I would look at the following factors:

P/B value
P/B of 73? That's nothing, I can do 200!

Generalist fund managers needing to buy ESG

Allego sits squarely in the ESG segment, and is one of the very few publicly available options for EV infrastructure investment. A quick look at the shareholder statistics show that 15% of stock is owned by insiders, and 84% of stock is owned by institutions. This accounts for a total of 99% of the total stock volume, meaning that the stock is not exactly a people’s favourite. As more and more mutual funds, pension funds, sovereign wealth funds etc are getting ESG targets, Allego might find its way in to indexes, baskets and portfolios that means large funds will hold the stock on autopilot. This creates artificial demand for the stock that is not necessarily rooted in the actual company performance.

Acquisition premium

The door into the CPO space is severely crowded. In the later years, large multinationals with deep pockets have acquired charging – related companies at hefty premiums in order to get a foot in the door. Examples of this include Shell’s acquisition of New Motion, and VW’s acquisition of the CPMS provider has.to.be. As Allego is in constant need for cash, and is one of the larger CPOs out there with a significant site footprint, it is possible to argue that the company is ripe for a buy-out, which means payday for existing shareholders.

Expectation about future revenue

“Yes, surely, the company burns cash and loses money everywhere, but this is all normal. The company is still in an investment phase, maximizing the potential of a landgrab scenario that it is well poised to win given its experience and know-how from data. It has premium sites in volumes, something which is not currently reflected well enough in the company balance sheet. We are confident that the company will become a cash cow in the future”.

None of these explanations really do it for me, and I am still mystified as to how supposedly rational analysts can justify a P/B of 73 for Allego. If anyone can help me understand what I am missing, I would greatly appreciate it.

Kindly,

Ole Henrik Hannisdahl
Founder, Incremental AS.