Image source: Getty Images.
Sunrun (RUN +3.83%) impressed investors on Thursday with quarterly revenue of $122.5 million and net income of $32.6 million, or $0.31 per share. And on the surface everything looks great: $51 million of NPV was added via leases and power purchase agreements (PPAs), and system & product sales more than doubled from the first quarter to $77.1 million, yileding the aforementioned profit.
But behind the numbers you'll see a business that's making extremely generous assumptions about its financials in the future. And that should worry investors.
Pulling the wool over investors' eyes
One thing hasn't quite made sense to me in Sunrun's financial reporting. The company has higher costs than competitors -- $3.67 per watt in Q2 compared to $3.05 per watt at SolarCity (SCTY +0.00%) and $2.94 at Vivint Solar (VSLR +0.00%) -- yet it says it's creating an incredible amount of value.
According to Sunrun, it is generating $4.61 in value for every watt it installed, while SolarCity is creating just $3.62 ($3.26 in contracted value). How does this make sense given the fact that they compete in a lot of the same markets for the same customers?
One answer hinges in how Sunrun defines value. Project value, according to Sunrun, is retained value plus upfront payments or benefits and tax equity. This retained value estimate includes an incredible amount of value that isn't contracted and would come 20+ years down the road. Here's a quote from the definition of retained value:
Estimated Retained Value represents the estimated nominal contracted payments remaining (discounted at 6%), less substantially all value from solar renewable energy credits ("SRECs") prior to July 1, 2015. It also includes a discounted (at 6%) estimate of the value of the solar system purchase or contract renewal at the end of the initial term.
That renewal value number is what's questionable. I'll cover why it's a questionable assumption below, but for now let's look at how it impacts financials.
To determine how much of that $4.61 in value comes from the renewal, you have to compare total retained value from Q1 2016 to Q2 2016 and divide by the 54.9 MW deployed under leases and PPAs. I've outlined the calculations under Sunrun's model in the left column below. In the right column, I've pulled out the value of any renewal or solar of systems at year 20, which I'll cover below.
|
Metric |
As Reported |
Without Renewal Value |
|---|---|---|
|
Estimated Contracted Retained Value |
$0.87 |
$0.87 |
|
+ Estimated Retained Value of Renewal |
$0.82 |
--- |
|
+ Utility or Upfront State Incentives + Upfront Payments from Customers + Tax Equity |
$2.92 |
$2.92 |
|
= Total Project Value |
$4.61 per watt |
$3.79 |
|
- Creation Cost |
$3.67 per watt |
$3.67 |
|
= NPV (on MW Deployed) |
$0.94 per watt |
$0.12 per watt |
Data source: Sunrun Q1 and Q2 2016 earnings reports. Calculations by the author.
You can see that when you pull out the renewal value there was very little value created during the quarter. And considering that the cash flows over the contracted 20 years are discounted at just 6%, you could argue that a higher discount rate would lead to lost value in the quarter.
Why is renewal so important?
The renewal number is important because it's an assumption based on no evidence. The lease and PPA was created only about 10 years ago and only became extremely popular in the last five years, so we have no idea if customers will renew after 20 years and I argue customers would be silly to do so.
Sunrun doesn't disclose exact renewal or buyout assumptions built into the renewal value, but most companies assume all customers will renew at 90% at prices of the last year of their contract, or 90% of customers will renew at the full value of the last year of the contract. That's much higher than their initial contracted rate, because Sunrun includes annual escalators in the price customers pay for solar electricity. For example, if the initial contract a customer signs is for $0.13 per kWh and there's a 3% escalator the 20th year, energy will cost $0.235 per kWh. 90% of that would be $0.211 per kWh.
If customers are signing contracts for brand new rooftop solar systems in 2016 for $0.13 per kWh, why would they renew a 20 year old system in 2036 for nearly double that price? And keep in mind that solar costs are coming down rapidly. In 20 years solar will likely be far cheaper than $0.13 per kWh.
A more appropriate assumption would be $0 renewal value because solar companies have no idea if customers will renew, buyout systems, or simply ask installers to remove a 20 year old solar system. The removal option is probably in a homeowner's best interest, and that would cost Sunrun money, so maybe the value at year 20 should be negative?
Numbers in solar matter
Solar companies that hold projects on the balance sheet often try to find ways to convince investors they're creating value. But sometimes those assumptions go too far. In the case of Sunrun, I would argue that aggressive assumptions are masking the fact that Sunrun is creating little to no value each quarter because it has much higher costs than competitors. And that's not something any investor wants in their solar company.
