Goldman and the Tesla Hyper-Feedback-Loop

The electric car maker's outlook keeps getting more uncertain.

Here's the latest installment from the corporate feedback loop known as the Tesla-SolarCity merger.

On Thursday morning, Goldman Sachs cut its 6-month price target on Tesla stock by 23 percent, from $240 to $185. Goldman did this even as it raised its forecast for Tesla's 2017 revenue by 17 percent and boosted its estimates for Tesla's non-GAAP earnings per share in 2017 and 2018 significantly (for example, Goldman now forecasts a profit next year rather than a loss).

Goldman's rationale for this apparent dichotomy is that the impending acquisition of SolarCity, among other things, makes Tesla a riskier proposition.

The twist here is that Tesla's bankers, Evercore ISI, used earlier Goldman forecasts from July for their valuation of the electric-vehicle maker to justify the transaction's terms. The numbers in Goldman's model have generally moved in Tesla's favor since then, with forecast revenue and Ebitda for the years 2017 through 2020 rising by 11 percent and 5 percent, respectively. Yet they now translate to a substantially lower value partly because of that same merger.

Evercore's valuation analysis was already comprehensively dismembered in this video by NYU finance professor Aswath Damodaran. The bigger point for Tesla's shareholders is what Goldman's new report says about the SolarCity deal, on which they will presumably get to vote soon.

First of all, Goldman's new numbers offer yet another reminder of the fickleness of forecasts and price targets for Tesla. That revenue forecast for 2017, for example, was $10.8 billion in the July model that Evercore used, was cut to $10.1 billion weeks before the prospectus even came out, and has now jumped to $11.8 billion.

As I wrote here, sell-side analysts have a history of being wildly overoptimistic about Tesla's earnings, then cutting them as results approach, only to have the company miss those reduced expectations anyway.

So Tesla's future cash flow, and therefore valuation, are highly uncertain. That gets to the second, more important observation about Goldman's latest report: Buying SolarCity pushes investors even further into murky waters.

Consider that Goldman just raised its 2018 EPS forecast for Tesla as a standalone company to $3.50 from $2.63, a jump of one-third. Yet its target price has dropped by nearly a quarter. The implied multiple of 2018 earnings has gone from 91 times to 53 times.

Granted, on one level, all this means is that Tesla's valuation multiple has gone from ludicrous to merely crazy. But the implication is that the risk premium embedded in Goldman's valuation has risen significantly. In other words, the chances of Tesla actually earning anything like $3.50 in 2018 are looking very dicey.

The addition of SolarCity's own cash-burning acumen to Tesla's existing abilities on this front are a big reason for this. Here are Goldman's EPS forecasts for Tesla as a standalone company and the mid-point of the annual ranges pro-forma for the acquisition going ahead:

Sachs And The City

SolarCity's high costs wreak havoc on Goldman's EPS forecasts for Tesla

Source: Goldman Sachs

Note: Forecasts are for adjusted, non-GAAP EPS. Pro-forma figures are the mid-points of annual ranges.

Given the aforementioned inaccuracy of analysts' earnings estimates for Tesla, you might just shrug at this: So what if a bunch of forecasts that never work out anyway turn out to be wildly wrong, deal or no? The real damage, though, is to the balance sheet.

Coming Soon

Taking on SolarCity pushes Tesla's net debt to very high levels using Goldman's forecasts

Source: Goldman Sachs

Note: Pro-forma data are mid-points of annual ranges.

Heading off that looming leverage ratio is one reason why Tesla needs to come back to the market soon to sell more equity. All else equal, getting that pro-forma number for 2017 down to 3 times -- still high relative to Tesla's history -- by selling new shares alone would require raising almost $3.3 billion of new money, equivalent to 11 percent of the current market cap.

And that would come on top of the new shares being handed to SolarCity's shareholders as part of the deal, equivalent to another 7 percent. It's worth pointing out here that, even on a standalone basis, Goldman models Tesla's share count hitting 208 million in 2020, 40 percent higher than the current level.

The paradox at the heart of this deal is that SolarCity's shareholders will be paid with a currency, Tesla's stock, which will be debased by the very act of handing it to them. That's one reason why SolarCity's discount to the implied value of Tesla's offer has swung around wildly.

Solar Coaster

SolarCity trades at a wide discount to Tesla's offer, though it has narrowed recently

Source: Bloomberg, the companies.

Note: Data through July 29th use the mid-point of Tesla's preliminary exchange ratio for SolarCity.

Lately, the market's faith in the deal actually happening seems to have strengthened somewhat. Assuming the risk of the deal's failure has gone to zero, SolarCity's own stock price implies a value for Tesla of $179 a share -- uncannily close to Goldman's new target and 14 percent below where Tesla closed on Wednesday.

One last thing to consider: Goldman's new target for a riskier Tesla takes its valuation for the company down by $8.2 billion -- roughly 4 times the market cap of SolarCity. You might almost call that a hyper-feedback loop.

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

    To contact the author of this story:
    Liam Denning in New York at

    To contact the editor responsible for this story:
    Mark Gongloff at

    Before it's here, it's on the Bloomberg Terminal.