Together with Cognitive Credit:

Welcome back!

Today we’re going to talk about Carvana - its rise, its fall, what everyone got wrong, and why the stock has traded so strongly. There’s a lot of information about Carvana online but this is one of the more important credit stories an investor can learn about. Why? Because so many people got it wrong (including myself) and left money on the table. 

I posted this the other day on Instagram - if you didn’t buy something, and it outperformed, then you were wrong. Every time you pass on a Company that outperforms, you are wrong, and you need to think about why you were wrong so you can improve going forward. If you get the timing wrong on a catalyst or event, then you were wrong. There’s only so many companies someone can buy, and some would rather just invest passively, but if you’re comparing your returns to a passive index and underperforming because you didn’t buy a name or two then you need to fundamentally examine why this is the case. This is hard for people who carry an overly bearish tone, or who think inside a box, to swallow but do you want to make money or do you want to win useless hypothetical brownie points? 

The biggest rule in Credit is that everything has a price. Carvana eventually hit a “screaming buy” price that not many investors hit because it didn’t meet their typical criteria of what constitutes a good investment. So let’s remove that type of blockage from your mind and get into this deep-dive of Carvana.

What was initially an ugly duckling, hasn’t looked back since its Sept. 2023 restructuring - the company quickly saw significant margin expansion, improved unit economics and solid FCF.

Now obviously this isn’t financial advice, and not to say Carvana is a “Buy” “Sell” or anything else today - it’s to say, similar to Tesla, “Hey look there was a massive inflection point and the stock shot higher - you can be a Bear today, but if you were a cranky bear on the way up, you need to evaluate your process because you missed something.”

History:

Based in Tempe, Arizona, Carvana was started in 2012 by Ernest Garcia III, Ryan Keeton, and Ben Huston. The idea came to Garcia III at a wholesale car auction, where he noticed people were deciding on cars after looking at them for just 30 seconds. That’s when he realized the whole process could be done online. But the company’s roots start all the way back in 1990 with his father, Ernest Garcia II. In 1990, Garcia II was found guilty of bank fraud regarding his role in the Lincoln Savings Bankruptcy, having hidden ~$30mm in risky investments; he was sentenced to 3 years of probation and forced to cooperate with the prosecuting attorney in the landmark national case of the failure of Lincoln Savings (chairman Keating was charged with 73 counts of fraud, racketeering and conspiracy and sentenced to 12 and half years). However, Garcia II also built an automotive empire in Drivetime, which started with the purchase of Ugly Duckling’s assets in 1991. At that time, he merged his own financing company with the newly acquired used car retailer to create a company targeting subprime borrowers with poor credit histories. In 1996, the business flourished and ultimately went public. However, shortly after, in 2002, the company’s CEO and Ernest Garcia decided to take the company private once more and operate it independently. Today, Drivetime (newly named) is known for its proprietary credit scoring model and large operations, with annual revenues of ~$750mm.

Garcia III graduated from Stanford (BS in Management + Engineering), and after various roles (including consumer based investing at RBS Greenwich Capital; now NatWest), he joined the family business in 2007. He held various positions there (including Treasurer, and Director of Quant Analytics) and got his feet wet in the industry. Eventually, in 2012, he made the bold leap of creating his own company, but with Daddy’s help. At this point, CVNA’s operations were largely being supported by DriveTime’s balance sheet, as all of the inventory and loans were provided by the parent company. However, as the company kept picking up steam, with some even calling it the “Amazon of Cars”, Ernest Garcia III and Co spun off from Drivetime in Nov 2014 and began their independent journey.

This decision freed the proverbial shackles and allowed the company to scale and reach new territories. The business grew aggressively, and in the Spring of ‘17, the company filed to go public. 

Why defining a Credit Data Strategy is now existential for Credit Investors

As AI rapidly advances, credit investors who fail to leverage its potential will find it impossible to compete with more agile peers in terms of both performance and AUM. The best way for firms to position themselves for the market’s continued digital evolution is to define a “Credit Data Strategy”.

Recorded in partnership with the European Leveraged Finance Association, Cognitive Credit Founder and CEO, Robert Slater, explains what a best-in-class, data driven approach to credit research looks like in 2025 in this on-demand webinar.

Topics include:

-          How real-time fundamentals can drive faster decision-making.

-          Leveraging structured data & automation to unlock operational efficiencies

-          Using advanced analytics to enhance credit portfolio management

-          Best in class-examples for 2025

Cognitive Credit is the market's premier provider of fundamental credit data and analytics for credit investors. Want to see how our data & analytics could make 1Q25 earnings easier? Request your demo today.

Business Description:

Carvana is a national online used car retailer that allows customers to (1) buy, (2) sell, and (3) finance vehicles entirely through digital channels. The company operates a fully integrated e-commerce platform that delivers cars directly to customers' homes, eliminating the need for traditional car dealerships. 

Revenue Mix:

The company splits up its revenues into 3 different segments.

1. Retail Vehicle Sales ($9.7B; 70.8% of FY2024 Total Revenues): Aka online care sales. Customers can either (1) pick up the vehicle at a nearby IRC (Inspection and Reconditioning Center), (2) have it delivered to their home (80% of customers choose this) or (3) pick it up at a nearby vending machine (mostly for show but the company does report higher “market penetration” after the installment of one). This is the Company’s highest margin sales channel with FY2024 gross margin at 14.2%. To also note, all sales within this channel are subject to a 7-day guarantee, this is especially relevant because its a main sticking point for online purchases.

2. Wholesale Sales ($2.8B; 20.7% of FY2024 Total Revenues): The company also generates revenues by selling vehicles to other wholesalers. Specifically, this category includes sales to independent dealers, franchise dealers, and wholesale buyers. The company even mediates auctions on its platform. The main source of inventory for these sales is customer trade-ins that don’t meet Carvana’s retail marketplace status. As a result of the lower quality units + customer needs (dealers), gross margins are tighter vs. the retail channel (see FY2024; 10.3%)

3. Other Sales ($1.2B; 8.5% of FY2024 Total Revenues): This channel is mostly miscellaneous. It covers 4 activities. ((1) Carvana originates auto loans and sells them via ABS, with no recourse after the sale; (2) it sells VSCs, which function like warranties covering vehicle repairs; (3) it offers GAP Waiver Coverage that forgives remaining loan balances in cases of theft or total loss; and (4) it earns affiliate fees when customers buy auto insurance through Root.

Cost Structure:

Within CVNA’s business there are 5 main cost drivers - notably, the company prides itself on being vertically integrated.

1. Vehicle Acquisition Costs – The cost of purchasing used vehicles (excluded from the slide's breakdown). The Company only expenses this if the sale has been made, otherwise it’s registered under a cash outflow in the CFO section. This will be relevant as we enter the discussion of CVNA’s 2022 shortcomings.
   

2. Non-Vehicle Cost of Sales – Mostly variable in nature; reconditioning (parts, labor, overhead), inbound transport, inspection and IRC and auction facility expenses. Also includes more fixed components such as management payroll and facility under-utilization.
   

3. Operations Expenses – Mostly variable in nature; operations payroll (dealing with outbound transport for its last mile deliver services but excl. IRC payroll), non payroll operations exp (fuel, repairs and maintenance of outbound fleet, third party transportation services) and finally, transaction expenses (including fees for title and registration, finance platform expenses, and warranties). 

4. Advertising Expenses – Primarily variable costs related to performance marketing, media spend, and brand advertising to drive customer acquisition

5. Overhead Expenses – Mostly fixed; facilities expenses (corporate, customer service, last mile delivery hubs, storage facilities), corporate expenses (miscellaneous incl. HR, legal, tax etc.) and technology expenses (engineering, product, data science and tech services).

A Look into Carvana’s Business Model:

Pre-Restructuring Business Performance:

Carvana’s late 2010s financials were pretty ugly, which is why a lot of Credit Investors were very skeptical of the name.

Although the company continued growing its revenue (4.58x from 2017-2019), the unit economics were unfavorable due to elevated COGS mostly relating to non acquisition cost of sales. Within this category the overwhelming rhetoric is that the company was operationally inefficient. Although sales were scaling rapidly, the logistics were not ready to handle the demand. As a result, any gains won with expanding locations and IRCs were offset by underutilization and high inbound transport costs. Essentially, these ballooned cost of sales, squeezing the Company’s gross margins to low single figures. Add to that attempted scaling via advertising and its gross margins were turned into negative operating profits every year until FY2023 (as we’ll delve into later). 

A complex and intense purchasing model: Adding to already difficult unit economics was its purchasing model. Creating and scaling a used vehicle retailer requires constant updating of its vehicle fleet. Regardless of digital accessibility, customers are less likely to use your dealership if your vehicle selection choice is poor (to note: CVNA currently has 53k vehicles on its platform vs. ~200 for your local car dealership). Again, this puts even more pressure on cash flows as dealerships are forced to be constantly purchasing new vehicles. 

Capital Intensity: The second dynamic to note is the capital intensity of CVNA’s scaling. During these years, the company was set on expanding to new markets (cohorts). They achieved this in a few different ways. 

1. IRCs: Remember, these are basically warehouses where CVNA inspects the cars for quality before shipping them to customers. These centers also have big lots adjacent to them that act as vehicle storage for the company. 

2. Vending machines: These are see-through buildings that store the company’s cars and also act as pickup locations for their vehicles. From CVNA’s perspective, this achieves multiple purposes: (1) Stores the vehicles efficiently - like crowded NYC parking lots where employees pick up your car and use an elevator to park it on another floor in the building. (2) Awareness. The company has been smart with the placement of these hubs as they are often hard to miss. Many of them are along trafficked highways or otherwise plots of land with few high rises near it, meaning it’s very likely it will draw your attention if you pass by it. 

However, with both of these initiatives, the company somewhat negated the positive effect of a digital-only model. This is because by making in-person locations such a pivotal part of their operations, CVNA invested significantly ($230mm in ‘19, $144mm in ‘18 and $78mm in ‘17) in physical locations.

Leverage Profile

To accelerate growth, the company heavily leveraged the debt and equity markets. At the start of 2020, leverage was not an issue as they had ~$1.5B in total debt vs. ~$4.5B in market cap, running the business at 25% LTV, all while EBITDA was still a negative $230mm at this point. CVNA was able to raise a total of ~$775mm in equity financing across 3 years (‘17-19; excluding IPO proceeds, $469mm) from investors. At this stage, strong future growth prospects still gave the company wide access in both the debt and equity markets.

All of the above factors lead to a pretty soft financial picture by the end of FY2019 (Feb of 2020). The company’s inefficient operations were leading to pressure on its gross margins, which with an elevated SG&A base led to negative EBITDA. Additionally, its purchasing model focused on early vehicle acquisition and its capital intensive growth further pressured FCF generation.

COVID Speeds up Carvana’s Growth 

As COVID came around, the used car industry stood to gain from multiple tailwinds. Carvana specifically was one of the few companies during this time that (to some extent via luck) was well positioned.

For the Company, there were 4 main drivers of improved business activity:

1) Supply chain lockups. As supply chains worldwide came to a halt, manufacturing of vehicles also stopped → supply for cars was driven by the secondary market → prices skyrocketed. Axios reported that “every major car manufacturer”, during Mid-march, had stopped manufacturing vehicles. From January to December of 2020, this led to an average price increase of 14%, almost 10x the rate of inflation during that time.

2) Digital Tailwinds: CVNA was the main digital-first player in the used car market at the time of COVID → positioned as the main beneficiary of online car purchases. To note, other used car online retailers also performed very well during this time. Vroom, for example, saw 43% YoY revenue growth in 2020 (vs. 41% for CVNA).

3) Financing: As financing became a lot cheaper due to expansionary fiscal policy and low interest rates, more consumers were able to enter the market for used cars. Consequently, this meant general demand in the market increased and, along with other factors, prices skyrocketed.

4) Stimulus checks: Finally, consumers also had more capital as a result of government programs. Management discussed this in earnings calls (Q1 2021) as they saw “distributions of stimulus payments coinciding with a meaningful uptick in activity.”

From the start of the pandemic in 2020 to the end of 2021 the stock gained $149 (from $82 to $231), implying a 180% return. To also note, during this time the stock reached a high of $370 in the summer of ‘21. However, towards the end of 2021, and the beginning of 2022, as the frenzy (and car prices) slowed down, the stock began to correct to more rational levels..

Post COVID

Although the company greatly benefited from the industry tailwinds of COVID from 2020-2021, the reality of the state of its operations set in toward the beginning of 2022, as interest rates were set to start rising. After the pandemic, there were 3 main negative catalysts that propelled the Wall St. darling into a Ch. 11 candidate by the summer of 2023.

Aggressive Investments

Adesa Acquisition  – In May 2022, Carvana acquired ADESA U.S., a wholesale vehicle auction platform, for $2.2 billion in cash, funded through $3.275 billion of privately placed unsecured high yield notes (10.25% interest rate; half was financed by Apollo). The acquisition was a strategic move to expand CVNA’s reconditioning and pickup locations, with Adesa’s 56 in-person centers. Additionally, it was an attempt to further diversify revenues with the addition of Adesa’s wholesale dealer network. Although strategically there was a case to be made for Adesa’s acquisition, the company, as we’ll see later on, completely mistimed things.

As a slight side note, the financing for the deal was especially interesting. It was rumored that initially the acquisition was supposed to be financed with a mix of bonds and preferred shares. However, after not reaching enough market demand on a $1.2B share issuance (backstopped by $600mm of preferred shares to Apollo), Apollo offered to commit $1.6B (~50%) of a $3.275B bond issuance. Additionally, the new issuance barred the company from repaying the debt early. Compared to an equity share issuance, Apollo’s proposed alternative made the firm 1.6x as much money, stemming from comparatively elevated interest payments (as a result of unpaid principal). Ultimately the deal was executed at an annual interest rate of 10.25%, well above average HY debt pricing at the time. The company used the additional $1.2B ($3.275 from issuance less $2.2B in Adesa Acquisition) in liquidity for general corporate purposes.

There were further aggressive capital investments – Throughout 2021 and 2022 the company focused on expanding its operations via physical location expansion. This took two main forms. First is by (1) installing new IRCs to recondition vehicles and secondly, (2) with the construction of new vending machines. For 2021, this figure amounted to $557mm, representing a y/y increase of $197mm (2020).The story was somewhat similar the following year (2022), which is slightly surprising due to the significant expansion of locations as a result of the Adesa acquisition. The new acquisition came along with 56 new sites which, although not fitted to function as IRCs at the time of the acquisition, still provided useful real estate. Regardless, the company opened 5 new IRCs and 2 new vending machines amounting to total expansion costs of $512mm. Seeing as vehicle units and unit pricing were contracting, this once again proved to be a poorly timed bet.

Premium Inventory Purchases

Across 2021 and 2022, the company, with the goal of unlocking its growth, continued to invest in new inventory. To fuel the massive top-line expansion, the company had to ramp up its vehicle sourcing programs and prepare for demand fluctuations. It was estimated that between March and June of 2021 (the market peak), the company had amassed a catalog of 90,000 vehicles. To contextualize this figure, currently, Carvana holds 58,000 vehicles on its platform. On a $ basis, 2021 inventory purchases amounted to ~$2B. For further context, the previous year (FY2020), total inventory purchases totaled $263mm. Clearly, the company overshot demand.

Worsening Unit Economics

GPU (Gross Profit per Unit) – On the unit economics front, the company also suffered during this period. Specifically, within its Gross Profit mechanics, there were two main drivers of worsening outlook. (1) First, the company continued its extensive purchasing of inventory throughout 2021. Essentially, this meant that as prices normalized from 2020 and early 2021 levels, CVNA was left with a fleet of cars that were very expensive. In Q4 FY2021, the company recorded a $52mm of inventory allowance adjustments as historical levels didn’t match market conditions.

High SG&A

In 2022, Carvana’s financial downfall was exacerbated by persistently high SG&A expenses, driven in large part by overinvestment in advertising and general overhead increases. The company had ramped up marketing spend aggressively (both in absolute $ and on a per-unit basis) anticipating growth that ultimately didn’t materialize. As sales volumes declined sharply, Carvana found itself “overbuilt for the sales volume we ultimately realized,” with an infrastructure and cost base that was too large to sustain. While SG&A expenses did decline in absolute terms during Q3 and Q4, the elevated pace of unit sales decline meant that SG&A per unit remained stubbornly high. This disconnect between cost structure and actual sales performance severely pressured operating leverage and accelerated the company's downturn.

In the Fall of ‘22, the company found itself in a delicate position with low ammo to combat its problems. Its aggressive investments hadn’t paid off, the market’s turn devalued its inventory, and its unit economics were further suffering. The question on everyone’s mind was whether the company was going to have to restructure? Was there a way back for management? Lack of confidence from the debt markets (with some maturities event trading at 50 cents on the dollar), elevated yearly interest expense ($486mm in 2022, up from $176mm in 2021) and the 2025 $500mm maturity wall all looked to be worthy challenges, was the company going to survive?

Restructuring

Pre-Transaction Leverage (Q3 2022; Nov ‘22)

Pre-Transaction Liquidity (Q3 2022; Nov ‘22)

The Restructuring Timeline: We’ll navigate the restructuring transaction of July 19th, 2023 via a timeline. We’ll talk through meaningful and logistical milestones in order to paint the most accurate picture of the process. 

Fall 2022 – Carvana is rumored to have engaged restructuring advisers

• Kirkland & Ellis and Moelis & Co are rumored to have consulted with the company to consider possibilities to deal with the heavy debt burden

Dec 7, 2022 – Co-op Group Formation 

• Apollo Global Management Inc. and Pacific Investment Management Co. are the main forces in assembling a co-op (BlackRock Inc., Ares Management Corp. and Knighthead Capital Management also part of the group)

  • The Group held $4bn of unsecured debt (70% of the total outstanding)

• Advised by White & Case LLP and PJT Partners Inc.

• Includes fewer than 10 lenders in total

• Lenders were barred from independently negotiating with the company for at least 3 months - this was a smart move by creditors to present an unified front and mitigate the risk of a “deal away” - aka a third party deal that would disadvantage a large constituency of the lender group

Mar 22, 2023 – Carvana’s First Debt Exchange Offer: This was Carvana’s attempt to extend their runway on their terms. The Company had cleverly moved Adesa assets into an unrestricted sub, and wanted this to remain the case, but this was unacceptable to the credit group:

• Offer size: Up to $1.0B of new notes

• Minimum tender: $500M required to execute

• Issuance of new notes:

• Senior secured 2nd-lien notes due 2028

• 9% cash interest or 12% PIK (first 6 periods)

• Collateral: ADESA assets excluded - prior to the commencement of this offer, CVNA moved Adesa’s assets (~$2bn and 45% of CVNA’s liquidity at the time) into an unrestricted sub

• Co-op group rejects this offer as they see meaningful value left on the table with exclusion of Adesa’s assets

• The offer was set to expire on April 19th, 2023

Apr 19, 2023 – Company extends offer deadline:

• The Company extended the deadline for bondholders to accept the offer to May 3, 2023 + improved their offer given lack of interest. This is a classic sign that the debt exchange wasn’t going to happen on their terms

• The only difference between Mar 22nd and Apr 19th offer is increased Sr. Unsecured Notes due 2025 consideration (from $808.75 to $858.75, including $20 early exchange premium in both scenarios), everything else remained the same

Apr 2023 – The bondholder group stays united and extends the co-op agreement:

• Cooperation amongst bondholders extended to a minimum of Nov. 2023, with rumors that it contained the ability to extend the pact to May 2024.

May 3, 2023 – Company extends offer deadline once more (2nd time):

• This offer only included an extension (from May 3rd to May 17th), with no changes in the terms

May 17, 2023 – Company extends deadline once more (3rd time):

• The offer did two things: (1) Pushed back the deadline to June 1st (Previously, May 17th) and (2) firmed up withdrawal deadline. This meant that bondholders who had offered to exchange their notes under the previous proposed offer, had less than one full day to decide whether they wanted to still exchange their notes, under the new revised offer deadline.

June 1, 2023 – Offer expires:

• Since no minimum participation was reached ($500mm), the offer expired without being exercised. During this period the lenders and Company worked on a more suitable deal.

• For anyone on “the buyside of a restructuring” - you guys know this period includes a lot of confidential discussions and back & forth where lenders huddle with their advisors, and the advisors go back and forth with the company’s advisors on how to work out a deal. Well after enough zoom calls and revised transaction updates - a deal finally got worked out 👇️ 

Jul 19, 2023 – The Company reaches an agreement to restructure its debt with the bondholder group: We’ll get into the details below, but here’s the initial rundown.

• Lenders agreed to reduce Carvana’s total debt outstanding by over $1.2 billion

  • This included plans to eliminate more than 83% of Carvana’s 2025 and 2027 unsecured note maturities 

• New notes offered in the exchange are secured by Carvana and ADESA assets 

  • This is what you see in a classic LME - an important asset, like ADESA, being included in the collateral package as a “stick and carrot” way to get a deal done.

• Lowered required cash interest expense by over $430 million per year for the next two years

• This time, the exchange launched with support from Noteholders representing over 90% (~$5.2 billion) of Carvana’s senior unsecured notes.

Details of the Transaction:

• There were two other terms to the transaction. (1) First, the company would exchange variously dated unsecured maturities for newly issued secured notes. The notes that would be retired totaled $5.2bn of unsecured commitments while the new exchange would be offering $4bn of notes, highlighting a $1.2bn debt reduction. Most notably, the new notes would be secured, classifying the transaction as an “Uptier” where subordinated noteholders take a haircut on the principal of their previous notes to achieve stronger collateral in newer notes. (2) Secondly, this transaction also detailed that the company would be purchasing unsecured 2025 notes in the open market. To note, this was prompted in conjunction with an equity raise

• The exchange was conditional on the company’s ability to raise $350mm via a common stock issuance. This was broken up into two tranches, (1) An at-the-market transaction and (2) a privately placed offering by the “Garcia Parties” and had to reach a $126mm sum.

Jul 27, 2023 – CVNA announces satisfaction of ATM public equity requirement

• The Company successfully issued $225mm (~4.9mm shares) in the equity market. This was one of the requirements under the TSA

• Citigroup and Moelis acted as joint sales agents under the program

• Tbh - once this was announced, with the borrower and lenders locking hand, getting the rest of the agreement done was very procedural

Aug 2, 2023 – CVNA commences execution of exchange program

• The following are the terms the company used in placing offers

Aug 21, 2023 – Garcia Parties satisfy TSA condition with stock issuance

• The Garcia Parties raise $126mm. The purchase price per share was $46.3, representing an 18% premium to close on Aug. 18th.

Aug 31, 2023 – Company announces closing of exchange program and final results

• 96.4% of noteholders agreed to exchange $5.52bn of Sr. Unsecured notes (2025-2030) for new Sr. Secured notes (2028-2031)

• Met equity offering requirement (public ATM + Garcia Parties)

• Open market purchases of Sr. Unsecured Notes due 2025 of $341mm

• Results

  • This lowered Carvanaʼs required cash interest expense by $456 million per year over the next two years. 

  • Extended the runway to reduce 2025 and 2027 maturities by 88%

  • Lowered total debt outstanding by $1.326B

  • Left room for further debt issuances of $1.15B (can grow to $1.5B)

  • Most importantly, the Garcia family maintained control of Carvana

Post-Transaction Leverage (Q3 2023; Nov ‘23)

Post-Transaction Liquidity (Q3 2022; Nov ‘23)

There was roughly a year or so to buy the dip in Carvana, before results went vertical again in 2024

Post restructuring:

With the restructuring completed, interest costs slashed, and an extended runway, there was a limited opportunity to get bulled up.  

Carvana was a heavily shorted stock and suddenly people were offside very quickly. This is a nightmare scenario and people didn’t cover fast enough. The initial short squeeze led to a massive rally and started the foundation of the rebound. 

A few better than expected results set into motion Carvana’s return to greatness - starting in 1Q23. 

Carvana’s “core” loss (adjusted EBITDA loss) in 1Q23 was projected at only $50mm-100 mm, a dramatic improvement from the $348mm loss a year earlier.

The company generated a positive Adj. EBITDA of over $50mm in Q2 (a milestone in Carvana’s history) and nearly broke even on the bottom line. Net loss in Q2 2023 narrowed to just $58mm, a vast improvement from the $238mm loss in Q2 2022. Carvana had executed over $1.1 billion in SG&A cost reductions, driven largely by strategic headcount cuts and disciplined expense management. In 2022, the company fired a total of 4,000 employees, with a further 3,000 (not directly disclosed, but judging by reduced employee headcount) in 2023.

The company reduced non-vehicle cost of sales per unit by ~$900 from Q3 2022 to Q3 2022, achieved through insourcing, staffing efficiency, process improvements, and logistics optimization. Additionally, operations expenses per unit dropped by ~$1,400 from peak levels in Q1 2022, including a $900 reduction since Q4 2022, aided by tighter control over payroll, delivery logistics, and warranty costs. Advertising and overhead expenses also saw marked progress. Advertising spend per retail unit fell by ~$400 since 2021, hitting a record low in Q3 2023, as the company optimized channels and improved conversion metrics. Meanwhile, overhead costs—especially those embedded in SG&A—remained elevated on a per-unit basis, but showed room for operating leverage.

To showcase this jump in productivity, gross profit per unit climbed to $5,952 in Q3 2023, up 70% y/y (nearly $2,500 higher per car than a year earlier). The company deliberately sold fewer cars in 2023 than in 2022, but made significantly more profit on each sale – demonstrating a strategic shift to healthier growth, prioritizing unit economics over volume expansion.

In 2024, Carvana’s recovery solidified. The company returned to growth mode profitably. In 3Q24 Carvana’s revenues grew 32% y/y and it delivered $429 mm in Adjusted EBITDA (11.7% margin) and record GPU of over $7,400 per vehicle – all-time highs driven by improved efficiency in reconditioning, logistics, and inventory management.

It was clear to the market - after years of heavy investment, increased scale, and a better refinement of unit economics, Carvana was out of the woods and was once again the high-flying darling people thought it originally was. 

Bulls were rewarded:

Hindenburg’s short report was for naught: The well-known short-seller, Hindenburg, came out in January 2025 with a report accusing Carvana of accounting manipulation and lax underwriting, showing $800 million in suspicious loan sales and massive insider stock sales. Hindenburg also called out the somewhat mis-aligned incentives structure, calling Carvana - “A Father-Son Accounting Grift For The Ages”. The market largely ignored this - and shortly after, the founder of Hindenburg felt the urge to call it quits, with the pressures of short-selling starting to be too much for him. 

What you need to learn: When it comes to the Carvana situation and how you can apply it going forward, you need to think through the following: 1) Have a multiple where you love the company and want to buy it hand over fist, where you hate it, and where you’re just okay with it. 2) Understand where expectations are for earnings and FCF and if something materially happens that bumps those expectations up - get bullish. Or - pre-emptively understand what levers can be pulled & what problems can be solved and then get ahead of it. For Carvana, the equity skyrocketed once they got unit economics right - once Ebitda was about to inflict positively that was the time to load up. So that's a massive change in what the market's perception was - they were hypothetically going to get to that phase one day so in a way, post the reorg, and you were waiting for that event to play out.

Concluding thoughts - so as things stand right now, Carvana sits at a resounding Enterprise Value of nearly $50B (a 30x EV/EBITDA multiple). The world has continued to move more digitally, and after improving the cost structure and increasing operating leverage, Carvana has blown out the Bears who counted them out, while enriching those who were able to remove their negative bias. We have no recommendation and are not publishers of financial advice - as who knows, Hindenburg could always get the last laugh from the grave - but we are here to help you 1) understand restructuring processes and 2) identify changing situations where assets may temporarily mispriced.

Until next time 

-Harry

Want to gain access to the full research library? Upgrade below. Eventually all of our research reports will be part of the HYH Premium library.

Looking for a job on the Buyside? Access our compensation analytics and job board platform, Buyside Hub: