Bitcoin - Potential Q2 2021 Impact

[The Accountant]

Here is my estimate for the Bitcoin Impairment Charge for Q2 2021:

Q2 Earnings Outlook with this Impairment:

 

[GeorgeSymonds]

Bitcoin was $58,700 on 31st March. They either didn't revalue the holding on that date in which case why would they need to now, or the holding back then using your other figures was $2,477M. The value today is circa $30k a coin as you say, the total holding is now $1.2B or so, and in total they've had a little over a $1B hit on the bitcoin investment in the last quarter (although a much smaller one compared to what they paid). I not a US accountant so I don't know if they have to revalue every quarter, but if they did, I think they're in for a dreadful number.

 

[MP3Mike]

Bitcoin was $58,700 on 31st March. They either didn't revalue the holding on that date in which case why would they need to now, or the holding back then using your other figures was $2,477M. The value today is circa $30k a coin as you say, so the total holding is $1.2B or so, so in total they've had a little overa $1B hit on the bitcoin investment in the last quarter (although a much smaller one compared to what they paid). I not a US accountant so I don't know if they have to revalueevery quarter, but of they did, I think they're in for a dreadful number.

They can only revalue in two cases:

1. They sell and re-buy the BTC.
2. The value of BTC goes below what they last valued it at, at any point during the quarter.

And in both cases they have to revalue it. (They can't revalue just because the price of BTC went up.)

So if BTC went to $0.01 even for split second in the quarter they would have to revalue to that price point. But then they wouldn't have to revalue it again unless it went below a penny.

 

[henchman24]

Bitcoin was $58,700 on 31st March. They either didn't revalue the holding on that date in which case why would they need to now, or the holding back then using your other figures was $2,477M. The value today is circa $30k a coin as you say, so the total holding is $1.2B or so, so in total they've had a little overa $1B hit on the bitcoin investment in the last quarter (although a much smaller one compared to what they paid). I not a US accountant so I don't know if they have to revalueevery quarter, but of they did, I think they're in for a dreadful number.

GAAP handling is that gains cannot be recorded (revalued) until the sale. Tesla took an impairment loss in Q1 that is estimated to be at the 31,490.33 number (GAAP forces the losses to be recorded as a non-cash event).

 

[MP3Mike]

Some discussion in another thread lead me to writing this post:

What makes you think that they purchased around Feb 8th? We know it was much earlier than that. They announced the purchase had already occurred in their January 27th 10k filing:



So it seems that they purchased sometime in January before the 27th.



So we know the price of BTC went down after they purchased it. The low price in Q1 for BTC was January 21st at about $28.9, let's assume that they purchased after that point. The next lowest point was on January 27th was about $29.3k. So they purchased between the 21st and 27th. (Which lines up with disclosure rules that they had to announce their BTC purchase within 4 days.) Making the 27th the low point in Q1 that they have already written the value down to.

So I have to revise my math now. 29.3k-28.8k= 0.5k loss per BTC. 0.8 * 42069 BTCs = $21M loss to recognize.

If they purchased before January 21st they would have already written it down to $28.9, meaning they need to recognize even less of a loss.

Can anyone, calling @The Accountant, find where I messed up that calculation?

If that is correct the impairment should be closer to $21M for Q2.

 

[heltok]

So silly. This should not matter, what should matter is the current value of it, not the min during the period. But great that it will fool all the silly bears who think that GAAP is the only valid metric who keep getting frustrated that the bulls who believe in phony non-GAAP keep getting richer.

 

[The Accountant]

Some discussion in another thread lead me to writing this post:



If that is correct the impairment should be closer to $21M for Q2.

I can't find fault with your computation of the $21m loss. I have calculated the Q2 loss from a different angle
(I have tweeked my numbers slightly from what I posted above in the first post)

The Q1 10Q states:
Note 3 – Digital Assets, Net
.................As of March 31, 2021, the carrying value of our bitcoin held was $1.33 billion, which reflects cumulative impairments of $27 million. The fair market value of bitcoin held as of March 31, 2021 was $2.48 billion.
-----------------------------------------------------------------------------------------------------------------
If Tesla states that Fair Value was $2.48 billion and we know the price of bitcoin at Mar 31 of $58,930 per unit, then it means they held 42,084 bitcoin units.
With this information we can compute the book value per unit at $31,603 (the $1.33B divided by 42,084 units)

---------------------------------------------------------------------------------------------------------------
If the low for Q2 is $28,815, then we compute a loss of $117m.

------------------------------------------------------------------------------------------------------------------

The problem with my calculation is that I cannot reconcile the March 31, 2021 book value per unit of $31,603 with the low in Q1 (the Q1 impairment).
I have seen others on Twitter, and I believe Rob Mauer as well, also point to $31,603 (or something similar) as the book value - so they are probably using my method of computation.
But I cannot argue with your logic. If the low in Q1 after Jan 27 was $29.3k per unit, then the $21m loss seems right.

So I guess we have a range of $21m to $117m. Thanks for pointing this out....big differnce.

Linking your $21m post here:

Near-future quarterly financial projections

If anyone has *a lot* of time on their hands I would appreciate some comments on my first draft income model out to 2035. I tried to make it reasonably comprehensive. I'm sure there are some dramatically bad estimations here and I'm gonna keep tinkering with presentability and justification. I...

teslamotorsclub.com

 

[MP3Mike]

The problem with my calculation is that I cannot reconcile the March 31, 2021 book value per unit of $31,603 with the low in Q1 (the Q1 impairment).

Yeah, that was my problem as well. The only thing I could come up with is if they had received ~3,300 Bitcoins for vehicle orders near the $58k top, and only held ~38,769 Bitcoins at the Q1 impaired level. Doing the math that way would come out with a Q2 impairment of ~$118M.

But could they really have gotten paid ~$194M worth of Bitcoins in order fees during the little time they had it available as an option in Q1? The only way I could see that being possible is if there were more than 700 Roadster orders placed and they paid the whole $250k purchase price in Bitcoins upfront. (Where they even taking Bitcoins for Roadster orders?)

 

[Chunky Jr.]

Does an impairment reset their cost basis? eg If BTC falls to 25K this quarter, then goes to 50K next quarter, would they then have a gain of $25K/coin if they sold?

I'm not an accountant, so sorry if this is a dumb question.

 

[The Accountant]

Does an impairment reset their cost basis? eg If BTC falls to 25K this quarter, then goes to 50K next quarter, would they then have a gain of $25K/coin if they sold?

I'm not an accountant, so sorry if this is a dumb question.

Yes, it resets there cost basis. So when they sell, it is off the impaired lower value and the gain is higher.

 

[The Accountant]

Yeah, that was my problem as well. The only thing I could come up with is if they had received ~3,300 Bitcoins for vehicle orders near the $58k top, and only held ~38,769 Bitcoins at the Q1 impaired level. Doing the math that way would come out with a Q2 impairment of ~$118M.

But could they really have gotten paid ~$194M worth of Bitcoins in order fees during the little time they had it available as an option in Q1? The only way I could see that being possible is if there were more than 700 Roadster orders placed and they paid the whole $250k purchase price in Bitcoins upfront. (Where they even taking Bitcoins for Roadster orders?)

I was playing around with that scenario (bitcoin obtained from vehicle orders) as well but saw it couldn't work.
Another scenario was they sold to test liquidity (booking the gain) and later repurchased more in Feb/Mar in the 50k range. However, I think if they had done this they would likely have mentioned it in the 10Q.

 

[MP3Mike]

Does an impairment reset their cost basis? eg If BTC falls to 25K this quarter, then goes to 50K next quarter, would they then have a gain of $25K/coin if they sold?

That is my understanding. The cost basis is continually marked down, and then gain is calculated from the lowest price it was marked down to, but only when it is sold.

 

[MP3Mike]

I was playing around with that scenario (bitcoin obtained from vehicle orders) as well but saw it couldn't work.
Another scenario was they sold to test liquidity (booking the gain) and later repurchased more in Feb/Mar in the 50k range. However, I think if they had done this they would likely have mentioned it in the 10Q.

Yeah, I guess they could have tested for large liquidity and then bought most of it back. But I would think they would have ended up with more than $127M of profit doing that.

In the end it will likely only be a mystery for this quarter, as going forward it sounds like there won't be any changes in their holdings for a while. (No orders accepted in Bitcoin and no Bitcoin sales/purchases.)

It will be interesting to see what they publish in about a month.

 

[The Accountant]

Reconstructing Tesla's Bitcoin purchases, sale and impairments:

Initial Purchase & Q1 Impairment

Tesla first reported their Bitcoin purchase with their 10K filing. The PwC Auditors Report is dated February 8, 2021 so we know that Tesla purchaed their Bitcoin between Jan 1 and Feb 8. I have computed that Tesla purchased their Bitcoin on Jan 28 for $32,070 per coin and had an Impairment that day as Bitcoin dropped to around $31,493. Bitcoin on January 28 was actually as low as $29,922 but Tesla purchased on that day AFTER that low.

Bitcoin Q1 Sale

To test liquidity, Tesla sold 4,541 Bitcoins at $59,683 (on March 20) generating a gain of $128m in Q1

March 31, 2021 (Q1) Bitcoin Book Value

After the sale of 4,541 coins, Telsa Bitcoin balance at March 31 was 42,231 units valued at the $31,493 per coin.

Estimated Q2 Bitcoin Impairment

The Bitcoin low for Q2 was $29,031 on June 22 and requires a $104m impairment charge for Q2.

Bitcoin Pricing Source: Coindesk

 

[The Accountant]

@MP3Mike @st_lopes

See my new post above on the Bitcoin Impairment.
I believe I solved the puzzle. I was never able to solve the puzzle in the past looking a the lows of each day in January. I finally figured out that Tesla purchased on Jan 28 after the low of the day and was only subject to the impairment on a slight dip in price that day after they had purchased their bitcoin.

 

[st_lopes]

@MP3Mike @st_lopes

See my new post above on the Bitcoin Impairment.
I believe I solved the puzzle. I was never able to solve the puzzle in the past looking a the lows of each day in January. I finally figured out that Tesla purchased on Jan 28 after the low of the day and was only subject to the impairment on a slight dip in price that day after they had purchased their bitcoin.

Hats off to you on this one!

 

[MP3Mike]

I do doubt that they bought, or sold, all the Bitcoins at one time/price. And we know they received some Bitcoins from people placing orders. But that is probably close enough.

 

[The Accountant]

I do doubt that they bought, or sold

Yes - That's possible. The Q2 10Q will give us another data point to nail this down better.

 

[MXWing]

Very timely POWER MOVE in Bitcoin. This is great support for TSLA going into earnings.

Hope action sustains. This will be of great help when impairment is discussed in both the letter, conference calls and analysts report.

... eh yeah we hit impairment when BTC was 31K. It's 38K right now so weird accounting rule.

 

[MP3Mike]

I guess in general we over estimated the impairment amount: https://tesla-cdn.thron.com/static/ZBOUYO_TSLA_Q2_2021_Update_DJCVNJ.pdf?xseo=&response-content-disposition=inline;filename="q2_2021.pdf"

Our operating income improved in Q2 compared to the same period last year to $1.3B, resulting in an 11.0% operating margin. This profit level was reached while incurring SBC expense attributable to the 2018 CEO award of $176M in Q2, driven by a new operational milestone becoming probable. Operating income increased YoY mainly due to volume growth and cost reduction. Positive impacts were partially offset by growth in operating expenses including increased SBC, Model S/X ramp (negative margin in Q2), additional supply chain costs, lower regulatory credit revenue, Bitcoin-related impairment of $23M and other items.

But I got pretty close:

So we know the price of BTC went down after they purchased it. The low price in Q1 for BTC was January 21st at about $28.9, let's assume that they purchased after that point. The next lowest point was on January 27th was about $29.3k. So they purchased between the 21st and 27th. (Which lines up with disclosure rules that they had to announce their BTC purchase within 4 days.) Making the 27th the low point in Q1 that they have already written the value down to.

So I have to revise my math now. 29.3k-28.8k= 0.5k loss per BTC. 0.5 * 42069 BTCs = $21M loss to recognize.