Image source: SolarCity.

One of the hardest things to reconcile with residential solar leader SolarCity (NASDAQ:SCTY.DL) is how a company with little revenue and losses quarter after quarter can be generating value for shareholders. What the company has tried to do is sell investors on a concept called retained value, which measures the value of solar systems they install over as long as 30 years.

Retained value has become a common metric in the solar industry, but it hides some important costs that go along with building solar systems. Below, I'll try to uncover what value SolarCity is really building in its business and if it's really as much as investors think.

What is retained value?
Before we dig into the numbers, let's take a step back and define what we're talking about. Retained value is a concept invented by SolarCity to put its value creation into one easy number to digest. It takes all estimated future cash flows and discounts them at 6% to today's dollars, giving investors an approximation of how much value the company is adding in the thousands of leases it signs each quarter.

Let's start with retained value under contract, because this is really the only part of the calculation with any reasonable probability of being accurate. The definitions on retained value below can be found in the company's latest earnings presentation. SolarCity puts retained value under contract this way:

Retained Value under Energy Contract represents our estimate of the forecasted net present value at a discount rate of 6% of Nominal Contracted Payments Remaining for all lease and PPA Energy Contracts (excluding consumer loan energy contracts) and estimated performance-based incentives and contracted solar renewable energy certificates allocated to us, net of amounts we are obligated to distribute to our fund investors, upfront rebates, depreciation, renewable energy certificates, solar renewable energy certificates and estimated operations and maintenance, insurance, administrative and inverter replacement costs, based on contractually agreed amounts as well as historic expenses.

I've put this explanation in a diagram below to try and explain it more clearly. After a contract is signed, SolarCity can assume certain cash inflows and outflows based on the contract and past experience with insurance, maintenance, and other costs. But it's key to understand that costs incurred before the contract is signed aren't included in retained value. These are sales, marketing, administration, and other costs that are paid for upfront to acquire the contract.

Image created by the author.

The renewal value assumes that customers will renew leases for 10 years at a 90% rate after the initial 20-year contract expires. Here's how SolarCity explains it:

Retained Value Renewal represents our estimate of the forecasted net present value at a discount rate of 6% of the payments we would receive upon renewal of all lease and PPA Energy Contracts (excluding consumer loan agreements) through a total term of 30 years at a rate equal to 90% of the contractual rate in effect at expiration of the initial term [net of estimated costs].

We'll get into why I question the renewal figure below, but that's the justification. It's these two value numbers that SolarCity adds together to get retained value, which is reported each quarter.

What SolarCity says it's worth
In the past year, SolarCity has added $1.4 billion in retained value. Below are the retained value balances at the end of 2013 and 2014 and the calculation of the value added during the year last year. These are a snapshot of the value remaining as of each date, so subtracting the value at the end of 2014 from the value at the end of 2013 gives us the value added in 2014. 

Retained Value Component 

RV as of Dec. 31, 2013

RV as of Dec. 31, 2014

Retained Value Added in 2014

Under Energy Contract

$660 million

$1,685 million

$1,025 million


$391 million

$738 million

$347 million

Lease/PPA Total Retained Value

$1,051 million

$2,423 million

$1,372 million

Source: SolarCity. Calculations by the author.

What's missing from retained value
What this calculation doesn't include is the costs that go into generating those sales and operating the business on a day-to-day basis or the debt used to finance projects. In the image above, those operating costs are along the blue part of the timeline.

In 2014, SolarCity spent $414.2 million on operating expenses, including $238.6 million on sales and marketing. These are costs that should be pulled out of retained value above because they take place before the contract is signed but are integral to the overall business. But they're not the only costs.

SolarCity also uses debt to fund some of its projects and working capital. In 2014, it added $806.8 million in net debt as a result of these costs. This should also be pulled out of retained value, which management stated directly in the announcement that they would be reporting "levered retained value," or retained value minus the cost of debt, in the future. In the fourth-quarter earnings release management made clear that debt pulls value directly from equity shareholder's portion of retained value saying:

Relative to our Retained Value forecast at the end of 2013, our total forecasted enterprise value creation increased by $1.4 billion in 2014. Net of an increase in net debt (including convertible debt) of approximately $0.8 billion, this suggests we increased total net present value creation for equity shareholders by approximately $600 million in 2014. 

The real value of SolarCity
Since I don't think there's any way we can reasonably assume a customer will renew their lease in 20 years for 90% of their rate in year 20, I give no value to the renewal portion of retained value. But the calculations are laid out so that you can add them back in if you disagree.

I think SolarCity's value creation in 2014 should be calculated as the retained value of energy contracts signed minus operating costs minus net debt.

$1,025 million-$414.2 million-$806.8 million = ($196.0 million)

In reality, I think SolarCity lost nearly $200 million in total value last year. Even using the quote above of $600 million in equity shareholder value (which includes renewal) if we subtract operating costs to obtain that value we get total value of around $200 million. That's not a lot for a company that's worth $5.1 billion. And remember that there are some rosy assumptions built into the retained value number SolarCity gives investors.

More problems with retained value
While retained value may now be an industry standard for measuring value -- it certainly has its shortcomings. Most come down to very rosy assumptions the industry is making about the future of solar leases based on very limited data. Here are the three biggest risks that could blow up retained value.

  • Will homebuyers be willing to take on legacy leases for aging solar panels? For example, would a homebuyer want to buy a home with 10 years left on a lease for 10 year old solar panels?
  • There's little data on leases older than 5 years, so we don't know if default rates will be significant. Retained value assumes they won't be, but this isn't a mortgage on a home that can be repossessed and if a customer is unhappy with their system they can stop paying for it. If they do, SolarCity would have little recourse but to ruin someone's credit and take the system down. Even that would be a costly process. 

As I've shown above, these risks are on top of the fact that total value added to shareholders is far lower than the $1.4 billion retained value number the company advertises. 

Look before you leap
On the surface, growth in retained value may give investors the idea that shareholder value is being created, however, a deeper dive could suggest that there are certain costs SolarCity isn't accounting for like operating costs that go into acquiring customers and debt that pays for building the system itself. 

Despite acquiring hundreds of thousands of customers over the past five years, I would argue that SolarCity has created little to no long-term value for shareholders when all appropriate costs are included. But it'll take many years for that thesis, or even a bullish thesis, to be proven correct. 

At the very least, I'd be cautious assuming value is as the company advertises. There's more to the story than the company would like to admit.