Discover more from Overlooked Alpha
The Benefits And Perils Of EBITDA
Bullsh*t earnings or useful valuation metric?
Editor’s Note: This is the fourth installment of our new Fundamentals series. If you enjoy this series, let us know in the comments.
EBITDA, or earnings before interest, taxes, depreciation and amortization, has become one of the most widely-used metrics in the investment community — and one of the most controversial.
EBITDA is controversial to some investors because it excludes very real costs. Interest expense represents cash out the door. The same is usually true for taxes1. Depreciation and amortization are both non-cash expenses, but each represents the multi-year accounting treatment of assets paid for in cash.
Making matters worse, modern companies now use an “Adjusted EBITDA” metric, which backs out even more expense. Most notably, nearly all companies that report Adjusted EBITDA exclude stock-based compensation to employees. That compensation is not necessarily a cash expense, but it’s certainly a real expense. The additional shares paid to employees who exercise their stock options mean existing shareholders are diluted: the size of their ownership stake in the company declines2.
As a result, some investors derisively refer to Adjusted EBITDA as “earnings before everything”. Charlie Munger, Warren Buffett’s partner at Berkshire Hathaway, famously said, “I think that every time you see the word EBITDA, you should substitute the words ‘bullsh— earnings’”.
The criticisms of EBITDA are valid — in some cases. EBITDA, by definition, does make profits look better. Companies can and do take liberty with the metric. Certainly, valuing a stock — or many stocks — on EBITDA alone can get investors into trouble.
But at the same time, there is value to understanding, and to using, EBITDA — and even Adjusted EBITDA. Neither metric is perfect, but neither should be dismissed out of hand, either.
According to multiple sources, EBITDA was first used by John Malone in the late 1970s. Malone, a billionaire who still heads the Liberty Global complex, at the time was rolling up a series of small cable television systems across the U.S. He would eventually sell the group to AT&T in 1999 for a deal worth, in total, a whopping $48 billion (about $90 billion in 2023 dollars).
To this day, Malone is exceptionally adept at avoiding corporate income tax, a key reason why Liberty’s companies have enormously complicated corporate structures. One way to minimize income tax is to minimize reported earnings under existing accounting frameworks.
And so, at a time when investors focused intently on earnings per share, Malone developed EBITDA to provide a framework for the market to better understand his businesses. The attractiveness of EBITDA to Malone, perhaps counterintuitively, was that it could allow investors to get a better understanding not of net earnings, but of cash flow.
Overlooked Alpha is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.
Again, the ‘D’ (depreciation) and ‘A’ (amortization) are non-cash figures. But from EBITDA, investors can get to a reasonable estimate of free cash flow by subtracting cash interest expense3, cash taxes, and capital expenditures. The latter figure, essentially, is the cash flow statement version of depreciation: capital expenditures are the actual cash spent in a given year, while depreciation is the accounting-based, non-cash, cost of the asset spread out over a given period4.
Certainly, Malone was trying to make his business look better by using EBITDA. But it’s worth noting that, in this original example, he was also providing more accurate information to investors. In the case of a business that was growing quickly, booking upfront losses to acquire customers, and using debt to fund acquisitions, net earnings per share simply wasn’t that useful a metric.
And indeed Malone was right. Shareholders in his cable company, Tele-Communications Inc., made massive profits. In this case, the use of EBITDA wasn’t an attempt to create the perception of value that wasn’t there. Rather, Malone was trying to help investors understand the value that was there.
Where EBITDA Is Useful
To be sure, not every business is going to be what Tele-Communications was in the 1970s: fast-growing, acquisitive, and a bit messy. But, broadly speaking, there are industries and profiles in which EBITDA is much more likely to be useful.
In our earlier piece on price-to-earnings multiples, we noted that real estate investment trusts (REITs) almost always use a metric known as funds from operations (FFO). FFO is generally seen as more accurate because REITs have exceptionally high depreciation expense, as the cost of constructing, say, an office building is depreciated over 39 years. But, in fact, that depreciation expense isn’t actually ‘real’ at all: assuming proper maintenance spending, the building is not worth zero after 39 years. Usually, it’s worth more than it was nearly four decades earlier.
A similar dynamic plays out in the resorts and casinos sector. Casinos are inordinately expensive to build: the Wynn Resorts WYNN 0.00%↑ location outside Boston, for instance, cost $2.6 billion. But once that money is spent, the depreciation expense going forward isn’t really of interest to investors. And so for a casino operator, what really matters is not net income — which is depressed by that depreciation expense — but cash flow. And EBITDA provides a better example of understanding what cash flow looks like: again, investors can deduct cash interest expense, the capital expenditures required to keep the property updated, and whatever cash income taxes are required. Indeed, casino executives and Wall Street analysts covering the industry will tend to use EBITDA almost exclusively.
Not coincidentally, casinos are also usually heavily leveraged companies, with substantial amounts of borrowings. And for indebted companies, EBITDA is useful in two ways.
First, EBITDA is commonly used in judging a company’s credit. Because the metric excludes non-cash expense and interest costs, it basically includes every dollar the business generates that could, in theory, go toward interest repayments. Indebted companies will usually report a leverage ratio, which is the total amount of debt (or debt net of cash on the balance sheet, in the case of a net leverage ratio) divided by EBITDA over the past four quarters.
Leverage ratios provide a quick look at the health of a company’s balance sheet. Very broadly speaking, a net leverage ratio over six times is exceptionally dangerous, and 3x-5x can be dicey for the wrong business or one in the wrong industry.
Similarly, EBITDA can provide a framework for understanding the valuation the market is assigning to the entire business, not just the equity. Analysts will usually use enterprise value to EBITDA multiples; enterprise value (market capitalization plus net debt) is the proper numerator because it represents the value of the entire business.
Two companies can have similar price to earnings multiples, but different EV to EBITDA multiples because of dissimilar balance sheets. For instance, building products companies Masco MAS 0.00%↑ and Gibraltar Industries ROCK 0.00%↑ both trade at about 13x this year’s forecasted net earnings per share. But Masco has nearly $2.5 billion in net debt on its balance sheet; Gibraltar basically has zero borrowings.
And so, on an EV/EBITDA basis, MAS trades at about 10.1x, and ROCK 9.1x. What that tells us is that the market is valuing Masco’s business more dearly, even if it is valuing the two companies’ equity essentially identically.
Adjusted EBITDA Takes Hold
While Malone’s original use of EBITDA might have had some value, EBITDA admittedly has proliferated massively in the nearly five decades since. The metric became more popular in the late 1990s, when amid the ‘dot-com bubble’ more unprofitable companies went public.
Importantly, it wasn’t just a ‘dot-com bubble’; the concurrent telecom bubble was even larger. Perhaps ironically, many companies were looking to take Malone’s strategy in cable and apply it to communications assets. One of those companies, Global Crossing, in the third quarter of 2020 reported a net loss of $602 million, but positive Recurring Adjusted EBITDA of $355 million. Unlike TCI, however, Global Crossing’s use of EBITDA was not highlighting hidden value: the company filed for bankruptcy just 15 months later.
As tech stocks recovered, and as they led the market in the 2010s, the use of EBITDA, and in particular Adjusted EBITDA, spread. The sheer number of unprofitable companies, and particularly unprofitable companies with huge valuations, was a key factor. For companies with negative net income, EBITDA provided a metric that could be positive, or close. And for those companies, stock-based compensation tended to be much higher than in mature industries. The variance in the accounting for SBC made adjusted figures smoother and (in theory) easier to understand5.
By 2017, even Buffett himself was criticizing adjusted numbers, and implicitly Adjusted EBITDA:
Too many managements – and the number seems to grow every year – are looking for any means to report, and indeed feature, “adjusted earnings” that are higher than their company’s GAAP earnings. There are many ways for practitioners to perform this legerdemain. Two of their favorites are the omission of “restructuring costs” and “stock-based compensation” as expenses.
Many investors felt the same. Again, stock-based compensation is a real expense, and for many tech companies a very real expense. In the first three quarters of 2023, stock-based compensation at Snap SNAP 0.00%↑ has been over 30% of revenue. The company reported $2.4 million in Adjusted EBITDA over that period — a figure that excludes a stunning $991 million in stock-based compensation.
Snap is a somewhat extreme example, but most big tech companies are excluding SBC from their Adjusted EBITDA figures. And that fact gets to the idea that the expansion of EBITDA usage, and of ‘adjusted’ figures more broadly, has become a tool for companies to inflate their performance, rather than to give investors a more accurate picture of the business.
WeWork Takes It Too Far
That idea was strengthened in 2019, when WeWork WE 0.00%↑ tried to go public. At the time, the office sublessor was still flying high, with investments from Softbank valuing the business at $47 billion. As part of the initial public offering process, WeWork filed a Form S-1 with the U.S. Securities and Exchange Commission, and included a novel term: “Community Adjusted EBITDA”.
Unsurprisingly, by the Community Adjusted EBITDA standard, WeWork’s business was performing quite well. In 2017, it had posted Community Adjusted EBITDA of $233 million on revenue of $866 million. CAE had increased 143% year-over-year.
The problem was that Community Adjusted EBITDA almost literally was “earnings before everything”. The metric excluded interest, taxes, depreciation and amortization, but also:
revenue from sponsorships and ticket sales to WeWork-branded events;
expenses from running ancillary businesses (including a coding academy);
sales and marketing expense (!);
growth and new market development spending, which incredibly consisted of things like developing WeWork designs;
pre-opening expenses; and
general and administrative expenses — in other words, corporate overhead.
If you squint, you can see WeWork’s point as an evolution of Malone’s. What WeWork was trying to communicate with the figure was, essentially, the spread between what it charged tenants and what it paid landlords in existing developments. But even in that context, the exclusion of overhead, in particular, was laughable, and the Community Adjusted EBITDA metric spawned no shortage of jokes.
There was a serious side to the metric as well, however. Between CAE and a claim in the initial prospectus that WeWork’s mission was to “elevate the world’s consciousness”, the initial public offering process itself punctured the aura surrounding WeWork and its founder Adam Neumann. The IPO failed, Neumann was removed, and WeWork wound up going public via a merger with a special purpose acquisition company in October 2021. In its second quarter 2023 earnings report in August, the company warned that it was possibly headed for bankruptcy.
Keep It Flexible
The irony of the WeWork IPO, however, is that the company’s use of Community Adjusted EBITDA actually disproved some of the claims made by critics of EBITDA and/or adjusted profit figures. The market’s response was so intense that a long-awaited IPO of one of the world’s most valuable companies was canceled. That response alone shows that “bullsh— earnings” can only go so far, and that investors indeed did, and do, understand what adjusted figures really mean.
That in turn gets to a key point for new investors: to remain flexible. There are no hard and fast rules in investing. Yes, Adjusted EBITDA figures make the business look better — but they also provide more information. Even the nature of that information can be useful. Is the disclosure of EBITDA figures even useful, or is it being released (as Buffett alludes to above) simply to provide a bigger figure than net income? Is management working to make the business look as good as possible — or removing figures (such as a non-cash gain on sale of an asset) that would actually make performance look better?
There is no single investing metric that tells the entire story. A net earnings figure is useful in most situations — but constrained by accounting rules and decisions, and not necessarily applicable to every business during every reporting period. EBITDA does make a business look better, but it’s also valuable when performance doesn’t reach the standard of positive net income and/or non-cash figures are impacting the bottom line.
Simply put, there are reasons why EBITDA has become such a common metric. One of those reasons is that, often, it makes executives looks good. But that reason alone isn’t enough. As WeWork’s Community Adjusted EBITDA shows, if investors didn’t believe there was some value in the metric, it wouldn’t exist at all.
As of this writing, Vince Martin is long shares of Gibraltar Industries.
Disclaimer: The information in this newsletter is not and should not be construed as investment advice. Overlooked Alpha is for information, entertainment purposes only. Contributors are not registered financial advisors and do not purport to tell or recommend which securities customers should buy or sell for themselves. We strive to provide accurate analysis but mistakes and errors do occur. No warranty is made to the accuracy, completeness or correctness of the information provided. The information in the publication may become outdated and there is no obligation to update any such information. Past performance is not a guide to future performance, future returns are not guaranteed, and a loss of original capital may occur. Contributors may hold or acquire securities covered in this publication, and may purchase or sell such securities at any time, including security positions that are inconsistent or contrary to positions mentioned in this publication, all without prior notice to any of the subscribers to this publication. Investors should make their own decisions regarding the prospects of any company discussed herein based on such investors’ own review of publicly available information and should not rely on the information contained herein.
Owing to deferred tax assets, on occasion the amount of taxes paid in cash is much lower than the amount of taxes recorded in the profit and loss statement — but the broad point holds.
To use round, overly simplistic numbers: an investor owns 1 million shares of a company with 100 million shares outstanding. The CEO exercises options for 10 million shares, meaning there are now 110 million shares outstanding. The investor’s stake has shrunk from 1% of the company (1 million /100 million) to 0.909% (1 million/110 million).
As with taxes, interest on the earnings statement and cash paid for interest are not necessarily the same thing. For instance, if bonds are issued at a discount to their face value, that discount is amortized over time, and the amortized figure is added to reported interest expense even though the company is not actually paying that cash out.
The exact length of the period depends on the type of asset.