navguy12
Active Member
Welcome back.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Near-future quarterly financial projections
- Thread starter luvb2b
- Start date
No, the majority of that decline was due to Panasonic IRA production credits being partially passed on to Tesla. That’s like $1500 per 2170 vehicle, and while it will persist, will not create any further step change reductions in future quarters.
The Accountant
Active Member
Yes - the IRA credits helped COGS; I should have mentioned this. It was a significant impact to COGS but not a majority of the decline.
The $1k drop per vehicle equates to about $400m reduction in COGS and the IRA credit was likely no more than $160m of this amount.
Thank you for sharing the data! Do we have any reason to believe that COGS reduction will exceed $1k per unit between last quarter and this? If ASP falls by $3.5k and COGS improves by $1k, the net hit to margin will be $2.5k, or approximately 5 percentage points. Isn't a cut from 18% to 13% over just one quarter pretty concerning?
Income before taxes last quarter was $2.8B. A $1B hit will reduce this by 35%, leaving net income per share at 47 cents. Additional IRA incentives may amount to perhaps $100M? Additional deliveries may bring in another $100M, partially offset by inventory discounts. How do analysts fill the $800M gap to come up with expected EPS of 80 cents?
The Accountant
Active Member
I am not forecasting quarters at this time but here is a forecast by "James Cat" coming in at $0.81 eps using an ASP of $45k.
Perhaps his COGS per car of $36.9 and his deliveries of 450k may be too bullish.
I see. Pretty optimistic estimates. No way ASP comes in near $45k. He said ASP was impacted by Europe cuts but the US cuts were on par if not higher due to inventory discounts, even though I believe there hasn’t been any price cut since April in China. Since China accounts for around 30% of sales, let’s say the global ASP went down by $2.8k ($4k in the West and $0 in China). $44k would be much more realistic than $45. COGS is also too low. Deliveries can get to 450k though; we do know that Tesla sells all cars they produce and Musk did say that 2023 deliveries can hit 2 million units.
I'm going to assume that James Cat's numbers are accurate, except for ASP and COGS. I think ASP is going to come $1k lower, and COGS $700 higher, per unit. So gross profit should be $1,700 * 450k = $765M lower. Take off 10% for taxes and divide by number of outstanding shares and we have 20 cents lower adjusted EPS. My prediction is 61 cents per share for Q2. I have no idea what the impact of disappointing EPS would be on stock price though. Guidance may be more impactful than EPS.
Remember the S/X quanities "in transit" last quarter though. That alone should hopefully push ASP up by ~$2k, likely offsetting the price cuts to some degree.
The Accountant
Active Member
I see 3 items that impact the overall ASP that are sometimes overlooked:
Product Mix (what you mentioned) - There will be more Ys and S&X's sold in Q2 as a percentage of the total deliveries than in Q1. Model 3 was about 30% of deliveries in Q1 and will come in at about 27% in Q2 based on data I am seeing.
Geography Mix - A higher mix of sales outside of China increases overall ASP (as China has the lowest prices). China was 33% of sales in Q1 and it should drop to 32%.
Foreign Exchange - This may be a wash as the Euro, GBP & CAD$ strengthened vs the US$ but the Yuan weakened (Q2 vs Q1).
The product mix dynamics I mention above would also be a benefit to margins as Y, S & X have better margins than the 3 and margins outside of China are better than the margins in China (from what I understand).
EDIT: Many of the forecasters who publish quarterly estimates do account for the items above. I know that Troy, Stephenson and others have detailed models that factor these items into their estimates.
Last edited:
Has anyone compared vehicle inventory levels in Days on Hand for prior quarters/years with the present ones?
Multiple sites do have the historical and current data by model and options, including both new and used. I have access to one of them, but am not eager to spend the time required to compile the histories if someone else already does it.
EV-CPO is the one I have, which also tracks by option codes and locale so it is entirely feasible to directly access the information.
I suspect this information could be useful to enhance short term forecasts, although because it relies on published Tesla data, it will always be less than perfectly accurate. In all probability the historical comparisons should be informative on option makeup data, which will drive improved ASP estimates.
Please opine on this, hopefully with an idea of who already does this, or if not, how best to organize the data.
Multiple sites do have the historical and current data by model and options, including both new and used. I have access to one of them, but am not eager to spend the time required to compile the histories if someone else already does it.
EV-CPO is the one I have, which also tracks by option codes and locale so it is entirely feasible to directly access the information.
I suspect this information could be useful to enhance short term forecasts, although because it relies on published Tesla data, it will always be less than perfectly accurate. In all probability the historical comparisons should be informative on option makeup data, which will drive improved ASP estimates.
Please opine on this, hopefully with an idea of who already does this, or if not, how best to organize the data.
Knightshade
Well-Known Member
The issue seems to be a lot of debate about what the actual inventory levels are--- with some claiming dupe configs don't show up individually (for example if one location has 5 2023 RWD Model 3s that are all blue with 18" wheels and black interior and no FSD, they only show as one vehicle in the system).
Matt Jungs tracker for example only shows roughly 2500ish cars in the whole US inventory, which is 1-2 days of production and it's been roughly that since end of Q1... Whereas we have Tesla reporting quarterly on days of inventory, IIRC it was 15 most recently, MUCH higher... so unless Inventory in ROW is massively worse Matt might be running into that exact "dupes only show as one" issue.
Matt Jungs tracker for example only shows roughly 2500ish cars in the whole US inventory, which is 1-2 days of production and it's been roughly that since end of Q1... Whereas we have Tesla reporting quarterly on days of inventory, IIRC it was 15 most recently, MUCH higher... so unless Inventory in ROW is massively worse Matt might be running into that exact "dupes only show as one" issue.
Let’s put some numbers on these.
Let’s say on average a Model S/X sells for $45k higher than ASP, and this quarter Tesla is selling an additional 10k units of these models - this boosts revenue by $450M, or $1k per car delivered. And let’s say on average a Tesla sells for $5k lower in China than ASP, and this quarter Tesla is selling 5k cars more in other markets - this boosts revenue by $25M which is negligible. If we assume that Tesla is delivering 20k Model S/X units this quarter, ASP can be propped up by $1k.
The question is of course whether Tesla is actually selling this many of these high-end cars. From what I've seen in the US, Model S/X inventory has moved very slowly despite deep discounts and other incentives. Remember that the delta between production and deliveries was over 3k in Q4 2022 and widened to 9k in Q1 2023? So there was already a build-up of 12k cars by the end of Q1. I don't remember what Tesla's justification was, but 12k 'in transit' or whatever is a very high number given that the peak quarterly deliveries was 19k units (in Q3 2022). My best guess is that Tesla delivers no more than 15k Model S/X units in Q2 2023. Let's assume best case scenario of 15k units - this would still boost ASP by $500. Assuming that gross profit on each Model S/X is twice that of each Model 3/Y or around $15k per vehicle, this would boost gross profit by $65M, and EPS by 2 cents per share. My revised EPS estimate is 63 cents per share. So in the grand scheme of things, the product and geo mixes don't have much of an impact.
The only inventory we can actually count is used inventory. New inventory lists configurations available, not units available - units available is completely hidden from users, from all the research I've done. The only numbers we have for days on hand that we can rely on at all are the official numbers from Tesla.
Doggydogworld
Active Member
The "Inventory Days" Tesla publishes in the quarterly updates include the unsold cars EV-CPO and others try to track plus the unsold duplicate cars tesla doesn't show (as mentioned above) plus cars in transit to fulfill existing orders. There's no way to compare these numbers with each other.
If you want to track Inventory Days over time the Q1 Update has the last 5 quarters (3, 4, 8, 13 and 15). You can easily access quarters for prior years by opening the Q4 Updates from 2021 and earlier at Tesla IR.
The factory stalkers and Tesla's pricing actions probably give a better feel for supply/demand than sites like EV-CPO. Tesla experiments with inventory listings too much for any trend analysis, IMHO.
The Accountant
Active Member
In the scenario I use below, ASP would increase by $578 due to mix. There is clearly going to be a decline in ASP due to price cuts but we should see some small offset from product and geography mix.
Currency could swing things too. The EURO exchange rate in Q1 was 1.073 and now in Q2 it is 1.087.
So an asp of 50k Euros in Q1 translated to US$53.7.
Now a 50k Euro vehicle translates to 54.4k. That is a $700 increase in ASP from the Euro currency strength.
However, the Yuan is weaker . . . meaning sales in China denominated in the Yuan will translate to lower US$.
I have not run the numbers to see if currency is an overall plus or minus to ASP.
I have no clue if EPS is going to be $0.40, $0.60 or $0.80. My point is that when trying to assess ASP for the Quarter, there is more than just price cuts and price increases to consider. Mix and currency have an impact.
Oh I completely agree. My initial calculation was simplistic by design - I focused on the big ticket items to get a sense of the ballpark. I'm having a hard time bridging the gap between my initial EPS and the analysts', while the impact from the mix and currency change, which I'm glad you brought up, of maybe 2-3 cents, is a drop in this 20 cents bucket. Perhaps they are just a lot more optimistic than I am about COGS which is a black box to me. I just don't think anything has happened over the last few months that would drop COGS per unit significantly, except for the introduction of the Model Y AWD which also likewise drops ASP. Bullish as I am on TSLA long term, I'm quite bearish on auto margins and EPS this quarter.
The Accountant
Active Member
Some will be surprised about what I am about to state: I think @Troy (TroyTeslike on Twitter) may have the best P&L model (and data) amongst the Tesla forecasters. Once the deliveries are known, he has very good data on pricing by region by model and usually nails the revenue number. COGS is difficult for all forecasters as there is less data available but he does well here too. I believe his current ASP estimates are available now to his paid subscribers.
Edit: I too don't have a warm and fuzzy feeling on the Q2 earnings but how many times has Tesla done the opposite of what I thought? Plenty.
Thank you this is helpful to know. I wonder how Troy takes into account the inventory discounts this quarter. Inventory discounts (and inventory too) are a new phenomenon this quarter, so there isn't a whole lot of data on which to build a reliable estimate methinks.
