Pacific Crest’s Tesla analyst, Brad Erickson, downgraded Tesla from Overweight to Sector Weight and has a fair value of $293 for the shares. This is a day after Deutsche Bank’s Rod Lache downgraded Tesla from Buy to Hold while raising his target price from $245 to $280. The shares have dropped from $280 last Friday to $258 on Tuesday.

Erickson downgraded Tesla due to the risk/reward being more balanced due to the shares moving higher from $185 at the end of March to $268 yesterday (where it traded up to when positive previews for June quarter Model S sales were published). While various headwinds have subsided such as concern about low oil prices, competition and China demand he believes Model X optimism has risen meaningfully going into its September launch.

He is also lowering his 2015 and 2016 estimates for Tesla due to lower leasing revenue but that is partially offset by higher battery revenue even though June quarter Model S sales were better than expected. For 2015 Erickson’s EPS moves from $0.71 to $0.65 (vs. the Street expecting a loss of $0.03) and 2016 goes from $5.27 to $4.46 (Street is at $3.42).

Erickson is also expecting a much smaller loss in the just ended June quarter of $(0.21) vs. the Street’s $(0.58).

2020 estimates are shots in the dark

Erickson at Pacific Crest initiated 2020 revenue and EPS forecasts of $31.2 billion and $24.75, respectively. Lache at Deutsche Bank increased his 2020 EPS estimate from $20.00 to $22.20.

Erickson is projecting Tesla’s revenue to increase over 50% between 2019 and 2020 going from $19.8 billion to $31.2 billion. This is based on the Model 3 ramping from just under 36,000 units in 2018 to almost 113,000 in 2019 to 332,000 in 2020. While the Model 3 would not be a top 10 selling vehicle worldwide I suspect it won’t have global availability meaning it would probably have to be one of the top selling models in the geographies Tesla will be selling it.

Erickson also has the company’s gross margin increasing from 28.5% in 2018 to 30.5% in 2020 while average selling prices (ASPs) decrease from $86,700 to $63,100.

Maybe all the stars will align for this to occur but given the challenges to ramp production and increase margins as ASPs decline I believe these assumptions will meaningfully change multiple times (and mostly downward) over the next few years. Using these estimates in a discounted cash flow model along with $3 billion in battery revenue (possible but the plant site is still under construction) to me is very risky.