Description

We recommend shorting Avis Budget Group (NASDAQ:CAR). We believe CAR, the second-largest rental car operator globally with roughly 20% US market share, is over-earning significantly due to (1) depressed operating expenses, which are in the process of normalizing, and (2) favorable revenue dynamics, which we believe are quite vulnerable to a reversal in supply and demand. While CAR has sustained earnings at elevated levels longer than many expected, we consider the thesis timely, especially given a +34% rise in the stock in the past month alone on very little company-specific news. 

We believe a balanced underwriting of trends already underway results in fair value 30% below current over the next 6 months, which is no more than a reversal of the recent factor-driven squeeze.

Our main purpose here is to provide an update on recent trends and share some additional thesis points relative to what has previously been written on this name (we encourage interested readers to revisit Siren81’s writeup of March 2022 for some quick background on market structure in the rental car industry and the drivers of CAR’s dramatic earnings growth since 2019). 

1. Americas Is the Key Driver of Value 

Although 27% of CAR’s global fleet resides in the International reporting segment (mostly Western Europe), this segment accounts for just 15% of group EBITDA before eliminations over the past 12 months, and we think the Americas business will continue to account for the vast majority of earnings going forward, just as it has historically. We believe Europe is a fragmented market, where local incumbents like Europcar and Sixt have meaningful market positions, low-cost disrupters have historically weighed on industry returns, and operators are structurally less nimble in adapting to changing market conditions. Some examples of this latter point are: (1) it is harder to right-size headcount given stricter labor laws across much of Europe, (2) different European markets have different regulatory regimes, so simply re-allocating fleet throughout the region to match demand is impossible. In addition, the European rental business has a higher percentage of program vs. risk vehicles, meaning vehicle disposition values are more fixed and outsized gains from canny fleet management are less achievable. All of this reinforces our view that as Americas goes, so goes Avis Budget, and we will focus on the Americas business in the remainder of our discussion.

2. Quantifying Americas Over-Earning

A quick review of what has driven Americas EBITDA up from $652mm in 2019 to $3,366mm over the last 12 months is a helpful starting point. The table below presents annualized snapshots of the business at various points in time: 2019, the EBITDA peak in 3Q22, the most recent period (1Q23), and a “run-rate” based on the following management guidance from the May 2, 2023 and February 14, 2023 earnings calls:

• “As we guided to our last earnings call, we expect our full year 2023 monthly vehicle interest per vehicle to be roughly $100, so there are additional headwinds to overcome on the vehicle interest line.”

• “Our straight-line depreciation in the fourth quarter was in the high 200s and we expect this figure to settle in the 300 to 320 range for 2023.”

While management has consistently sand-bagged the depreciation guidance up to this point by booking large gains on vehicle dispositions (they dispose of older vehicles into the used market, where post-pandemic shortages and elevated demand boosted residual values above CAR’s carrying values in many cases) this is an inherently transitory benefit well understood by the market; depreciation will eventually converge up to the $300/month straight-line rate, and any beats in the meantime are unlikely to be capitalized by the market. Thus we may say with a fair degree of certainty that run-rate earnings power in the Americas segment today is ~$2,000mm, not the inflated figure of $3,366mm reported by the company over the last 12 months.

Table I. CAR Americas Segment EBITDA Bridge since 2019

The key drivers of the nearly 500% increase in segment EBITDA from 2019 to 3Q22 are visualized in Chart I below. 

Chart I. CAR EBITDA Bridge: 2019 to Peak, Peak to Run-Rate

All of the $3bn increase is due to (1) $2.3bn of high incremental margin pricing gains (+37%), expressed by revenue per day or RPD, and (2) $1.1bn of declines in vehicle depreciation expense (-73%), while fleet growth net of opex has resulted in no net improvement in earnings. As discussed above, point (2) is clearly a transitory tailwind, and will recede naturally as CAR refreshes its fleet, provided only that the market values for used rental risk vehicles do not perpetually inflate relative to CAR’s internal depreciation schedules. On this point, we can see that the most widely-followed indicator of used vehicle values, the Manheim Index, has been correcting from highs. We believe that rental risk values have fared a bit better but are tracking directionally with Manheim. 

Chart II. Manheim Used Vehicle Wholesale Auction Index

Source: Cox Automotive

Point (1), the dramatic increase in RPD, is unprecedented. For many years leading up to the pandemic, despite the oligopolistic US market structure (Enterprise Holdings (“ERAC”), Hertz Global Holdings (“HTZ”) and CAR have jointly held at least 85% market share since 2007), the industry was fiercely competitive. For instance, a glance at CAR’s historical reporting (with a couple of our own adjustments to make 2014 and 2015 apples-to-apples) reveals a -0.7% CAGR in Americas RPD between 2014 and 2019, a period during which US core CPI averaged +2.0% per annum, for an annualized decline of ~3% in real terms. Worse still, CAR was seemingly unable to accurately forecast RPD, missing their annual guidance in 5 out of 5 years, by an average of about 2%:

Table II. CAR Historical Reported Revenue per Day (Adjusted to New Segment Reporting)

The company’s inability to drive pricing gains during this extended period, we believe, is characteristic of an industry where high fixed costs (namely, the depreciation and financing costs per unit that are incurred whether or not a vehicle is hired out, the large fixed network of retail outlets needed to operate an efficient fleet, and the heavy upfront costs required to win desk share at airports) disincentivize idling of assets that can be rented out at a lower price. This structural feature of the market is compounded by the existence of Enterprise, an operator nearly triple the size of CAR, which has a distinctive business model and is not subject to the same quarterly investor scrutiny that its smaller, publicly-traded peers endure. More on ERAC below.

We present all of the detail above to illustrate that this has been a challenging market for CAR for many years, and to emphasize the extraordinary nature of the +37% increase in RPD from 2019 to peak levels in 3Q22.

Once the anticipated normalization in vehicle depreciation and the projected uptick in vehicle interest expense materialize, we arrive at the $2bn run-rate level of Americas EBITDA shown in Chart I above. We believe this is a little ahead of the consensus view (Visible Alpha consensus for CAR’s Americas segment this year is $1.8bn, embedding $200mm or so of incremental RPD headwinds or -6% beyond the current run-rate). However, this is still an EBITDA number at least $1bn higher than 2019 levels, substantially all of which is pricing. How sustainable is that?

3. Signs of a Normalization in Process

We believe, in a base case, that CAR can retain about 50% of the Americas pricing taken during the past two years, at least for the time being. In short, of the $20 in RPD growth CAR achieved between 2019 and today, we believe they give back $10, landing 2024 RPD at ~$66. Below we walk through numerous datapoints suggesting pricing at current levels is unsustainable and needs to reset ~15% lower.  

First of all, CAR management is not denying that an RPD normalization is underway in their own quarterly guidance, as stated here on May 2: 

• “I think when you look at it on the Americas…we're getting back to normal seasonality. So I think the increase that you see sequentially in RPD from 1Q to 2Q, the last normal year we had was 2019, and you could look back going forward. It's kind of in the 2%, 3% sequential increase. And I think that's what you can expect out of us on a consolidated basis for 2Q.”

Simply plugging in the high end of this guidance, +3% Q-o-Q, would suggest that Americas RPD will be -9% Y-o-Y, more than half of the way back to our 2024 expectation. While the management team has consistently outperformed these quarterly RPD forecasts up to now, and could do so again in 2Q23, it is noteworthy that management is not trying to defend the sustainability of pricing.

We would note also the circumspection that has crept into the company’s messaging about long-term steady-state profitability. Back in 2021, CAR drew a line in the sand at $1bn of EBITDA:

• November 2, 2021: “Now that we've broken the $1 billion annual adjusted EBITDA barrier, we're never going back…this is a new chapter at Avis, where an annual plan below $1 billion in EBITDA is no longer acceptable.”

But by a year later, despite group EBITDA surpassing $4.1bn, management seemed less willing to commit:

• November 1, 2022: [Sell-side question]: “…I remember it was just last year…when you said that the Board of Directors wouldn't accept a plan for the year ahead that would include less than $1 billion of EBITDA. Here, you're doing $4 billion. Do you think the fact that you're doing $4 billion instead of the $2 billion that was expected by analysts at the start of the year…is that reason to believe that out-year earnings in 2025 or 2026, may be higher than was estimated a year ago or so? And how do you go about thinking what the normalized earnings potential of the business might be?...

• [CAR answer]: “We've made the decision of the management team to forego giving formal guidance for 1 year out. So we're definitely not giving long-term guidance on this call. But I'll reiterate what we said in our prepared remarks, we're seeing normalization in key variables that drive our business, like you said, RPD, rental days, used car values. But the question is, what does normalization mean? And what does the glide path towards normalization look like? Is it a hard reset back to 2019 levels? Or is it a gentle slope with a longer runway that settles at a level materially higher than 2019? You'll have to make your call…”

Second, we see a fair amount of support from checks and alternative data sources for the fact that Americas RPD is correcting. This makes intuitive sense to us from a top-down macro perspective, as Americans deplete the cash reserves accrued from government transfers over the previous 2 years and react to broad-based inflation and higher interest rates by curbing big-ticket discretionary consumption. 

Northcoast Research noted soft 2Q rate trends in their recent airport rental survey:

• “Based on our recent channel work on rental car pricing, we offer the following observations: (1) U.S. rate dynamics in April and May which were below year-ago levels at a low-teens rate, (2) forward pricing curves in the U.S. for the months of June and July which imply a slight firming in the pace of y/y declines, with rates down high-single digits, and (3) pricing trends in Europe which remain elevated by historical standards, though down double digits compared to a year ago as well.”

Chart III. Northcoast Research May 2023 US Airport Rate Survey

Source: Northcoast Research

Morgan Stanley’s most recent AlphaWise survey of 2,000 US consumers highlights sequential softening in travel spend intentions:

• “Summer travel plans softened further this wave of the survey with 19% of consumers expecting to travel more this summer compared to last year and 38% expecting to travel less, yielding a net of -19% (vs. -16% last month). Budgets are also smaller for summer travel this year, with 23% expecting to spend more and 38% less (netting to -14% vs. -11%).” 

In lodging, an industry closely tied to car rentals, we have seen checks turn incrementally cautious in 2Q, suggesting the pent-up demand that boosted travel and leisure during 2021 and 2022 may have run its course. Below are some excerpts from recent Cleveland Research checks with hotel operators:

• “May was the first month we did not exceed budget since Omicron last year. On a y/y basis, 1Q RevPAR was up 12%, April down 2%, and May down 1%. First 12 days of June are down 6% y/y vs. plan for up low-singles and now looks like the first month that will show y/y rate declines. What saved us in 21-22 was the beach markets, and now they are the ones making us look bad. All in, we are pegging 2023 up 3-5% for the year, down from our original 6%. 

• We took 5 points of y/y growth out of our 2Q RevPAR forecast over the past 60 days. In 1Q, our full serve portfolio was up 27% and select serve up 23%. For 2Q, both were planned up 12% on 4/1. We now expect full serve up 7-8% and select up 5-6%. Group is still strong.  BT is weaker than expectations, hovering at 80-85% of 2019. Leisure is soft vs. plan which we attribute to y/y strength in areas like Europe and cruises. For just the last 45 days…relative to 5/1 we have taken ~3pts of y/y growth out of our RevPAR forecast for May through August. 

• The average leisure guest is being more price conscious and shopping for value dates while last year they paid a premium for their first choice dates. But, our high end room types are still doing very well.”

• “May-June softened up a bit.  Budget coming into the year was up 6-7% for that period. By early May, our May/June budget had moved to +2-3% and we are looking to land around 1% as of today. Florida is feeling the biggest hit. 3Q is mixed bag – relative to prior forecast July is a bit worse, August similar, and September we are optimistic.  The gaps in May through July have brought our full year growth plan down by maybe 1-2 points. The majority of our shortfall in ADR has come from less demand for premium product/room types vs. last year.  Bright spot is group has remained strong. 

• There has been an adjustment in our rate strategy for 2H.  From an ops perspective we are not just dropping rate, rather holding the headline ADR and participating in more promos.”

While none of this tells us where exactly Americas RPD will land in the longer run, it is clear that the market is loosening up. A third, more structural argument, relates to ERAC:

• ERAC is privately-held by the Taylor family, although there is publicly-traded debt, so certain high-level financial and fleet information is available.

• The company operates a roughly 1.5mm unit global fleet, significantly larger than CAR’s 660k unit fleet. The strategy is anchored by ERAC’s dominant position in the insurance replacement market, where long-term agreements with the nation’s largest P&C insurers yield a much steadier business than the cyclical and seasonal airport business comprising 2/3 of CAR’s revenues.

• We believe that the insurance replacement market confers several utilization benefits upon ERAC, including longer average rental period (fewer turnarounds per vehicle per month, resulting in higher in-service rates and therefore higher fleet utilization), higher mid-week utilization (balancing out the relatively low leisure demand on Mondays, Tuesdays and Wednesdays), and complementary seasonality (insurance replacement demand is a function of collision frequency, collisions increase when the weather is bad, and the weather is worst, on average, during the 4Q and 1Q low season for leisure rentals, meaning ERAC gets better utilization than CAR during lulls in leisure demand).

• During the pandemic, however, we believe ERAC’s insurance replacement business was a liability. Given a shortage of new vehicles being sold by OEMs into the rental channel, ERAC had too little fleet to adequately meet both airport and off-airport demand, and they chose to honor their long-term commitments to insurers by pulling vehicles out of the airport channel and pushing them into insurance replacement. This resulted, we believe, in a loss of market share for ERAC within the lucrative airport channel, a boon mostly accruing to CAR given HTZ’s financial distress early in the pandemic.

• Based on our research, we believe that ERAC share came under pressure in early 2021 as vehicle supply tightened up, a 600-700bp tailwind to CAR’s market share; by 3Q22, ERAC share had begun to recover modestly as vehicle supply improved, and we do not think it is coincidental that 3Q22 marked the high point of CAR’s TTM Americas EBITDA. Going forward, we believe ERAC will aggressively seek to recover its lost airport presence, which remains about 400bps below its pre-pandemic levels today.

• In addition to ERAC’s superior scale and rigorous focus on customer service, the large insurance replacement business gives the company a structural advantage in fleet utilization; they can price lower and capture the same margin as CAR. Thus it seems CAR has few levers to defend share aside from RPD concessions in the face of a bullying action from ERAC. We would argue this is reflected in the 2Q23 data, but there is more to follow.

Taken together, we believe management messaging, checks, and the advantaged position of ERAC relative to CAR and HTZ all suggest ample evidence that a large portion of CAR’s pandemic-era RPD improvement will ultimately fail to stick.

4. Implications for the Stock

Incorporating our view of Americas RPD discussed above, we arrive at the following base case for CAR’s Americas segment:

• RPD corrects to just +16% vs. 2019 levels, compared to +36% at current run rate; this is the critical assumption, but we take some comfort in the Northcoast channel checks suggesting that 2Q23 trends are already -HSD, more than half of the way to our 2024 expectation, and management’s commentary also supports this view.

• We hold the current vehicle depreciation and interest per vehicle per month largely flat with current run rate.

• We assume some incremental SG&A and opex savings despite a slightly higher fleet size, and despite the fact that CAR’s per-unit SG&A and opex have consistently inflated since 2019.

• Given the demand pressures and share losses we anticipate for CAR from ERAC’s airport resurgence, we assume modest utilization rate pressure (-50bps from current).

• The result is a ~$900mm EBITDA level, roughly $200mm above pre-pandemic but less than half the current run rate.

Table III. Our Base Case 2024 Americas EBITDA Forecast

Using this estimate to build up to our view of CAR value per share, we further give credit for a 10% margin on $3.1bn of sales for International, and we apply the long-term pre-pandemic EBITDA multiple of 8.0x to our $1,095mm consolidated estimate.

The result, we believe, has some built-in conservatism (in a recession, after all, why would any of the extraordinary pricing gains remain?), yet still suggests an attractive risk-adjusted return for the short. 

Nor are we advocating for the stock to do something it has not already frequently done; CAR was trading below our price target as recently as last month, and has bottomed below this level fairly regularly over the past two years.

Table IV. CAR Price Target and Expected Return

Catalyst

Downward revisions in quarterly RPD outlook

Earnings misses