2. Carvana Co. (NYSE: CVNA)
Number of hedge fund holders: 64
Number of Hedge Funds Having Stakes in the Company Out of Top 100 Hedge Funds: 12
Carvana Co. (NYSE: CVNA) is an Arizona-based online used car retailer founded in 2012. It is placed second on our list of 13 stocks that best performing hedge funds are piling into. The stock has returned more than 173% to investors over the past year. According to Insider Monkey calculations, twelve hedge funds out of the 100 best performing ones are bullish on Carvana. The firm offers users the ability to research, analyze through imaging technology, purchase, and schedule cars for delivery through their platform.
Carvana Co. (NYSE: CVNA) posted earnings results for the first quarter of 2021 on May 5, reporting earnings per share of -$0.46, beating market estimates by $0.21. The revenue over the period was $2.25 billion, up 104% year-on-year.
At the end of the first quarter of 2021, 64 hedge funds in the database of Insider Monkey held stakes worth $7.5 billion in Carvana Co. (NYSE: CVNA), up from 63 in the preceding quarter worth $7 billion.
In its Q1 2021 investor letter, Steel City Capital LP, an asset management firm, highlighted a few stocks and Carvana Co. (NYSE: CVNA) was one of them. Here is what the fund said:
“Carvana’s (CVNA) 4Q’20 results weren’t particularly great. EBITDA was negative ($70) million, a stark turnaround on a sequential basis from a first-ever EBITDA profit of $21 million in 3Q’20. The culprit was a steep drop off in retail unit GPU ($1,265 vs. $1,857) and wholesale unit GPU ($358 vs. $1,113) as some of the COVID-driven aberrations in the used car market began to abate.
The company’s presentation of EBITDA (calculated “bottom up”) is dubious, as it commingles non-operating items including mark-to-market changes in its retained securitization portfolio. With the exception of 1Q’20, when ABS markets were going haywire, this line item provided a tailwind throughout 2020, including a gain of $5 million in 4Q’20. Also on the non-operating self-help front, management released a reserve for vehicle service contract cancellations in 4Q’20, adding another $7 million to EBITDA, and boosting “Other” GPU by $96.
Putting it all together, I put operating EBITDA closer to negative ($82) million vs. the $70 million printed by the company. This is a larger loss than 4Q’19 (calculated on a similar operating basis) despite the company selling 43% more retail units y/y!
Management didn’t provide formal guidance for 2021, but did offer guardrails for how to think about the year. Retail unit growth is expected to accelerate from last year’s 37%, with total revenue tracking in-line with retail unit growth. Total revenue per retail unit was $22,885 last year, meaning the company thinks it can hold this metric relatively flat throughout the year. Management also noted it expects some softening in retail ASPs throughout the year (“I think the gains that we saw in ASP in the back half of the year, we expect to moderate a little bit in 20216 “), with the implication being “Other” revenue – including financing – will serve as an offset.
Why look at total revenue per retail unit? The company guides to total GPU, which itself is an apples-and-oranges mix of total gross profit divided by only retail units. As for total GPU, management called out expectations for “mid- $3,000s” in FY21. Putting the pieces together, $3,500 of total GPU divided by $22,885 of total revenue per retail unit implies gross margin of 15.3% for the year, roughly 100 bps of pickup vs. last year’s 14.2%.
On the EBITDA front, management guided to continuing cost leverage but still a “small EBITDA margin loss” in FY’21. Splitting the difference between last year’s negative 4.6% EBITDA margin and breakeven gives us something in the realm of a 2.5% EBITDA margin loss for FY’21. So, 200 bps of total improvement, 100 bps of which we know is coming from GPU margin. The other 100 bps, therefore, must come from SG&A.
Applying a negative 2.5% margin to $22,885 of total revenue per retail unit implies about $575 of negative EBITDA per unit sold. This also allows us to back into implied cash SG&A per unit of $4,075, which is 17.8%, and 100 bps better than last year’s 18.8%.
The unknown variable is what retail unit growth actually looks like in 2021. All we know is it’s going to “accelerate” vs. last year’s 37%. Doing some back of the envelope sensitivity implies negative EBITDA ranging from a ($210) million loss at 50% unit growth to a ($250) million loss at 80% growth (A classic case of “We lose money on every sale but make up for it in volume!”). For context, the street is currently forecasting a negative EBITDA margin of 1.0% and negative EBITDA of ($87.5) million.
I think one of the big risks to the company’s outlook isn’t necessarily on the volume front – I believe management when they say they can’t keep pace with demand – but instead on the GPU front. Before 2H’20, only once in the prior 14 quarters did total GPU exceed $3,000. 3Q’20 reached an all-time high due to strong vehicle pricing and strong finance gross profit, while 4Q’20 got its boost from finance gross profit and the abovementioned reserve release. Finance GPU is a function of both absolute interest rates and ABS spreads, and the trajectory of absolute rates since the beginning of 2021 calls into question the company’s ability to maintain finance GPU at $1,400, let alone grow it.”
