“Beware of geeks bearing formulas.” – Warren Buffett

Earnings before interest, taxes, depreciation, and amortization — discussed more commonly using the acronym EBITDA — has become a popular standard by which to measure business performance. Public companies use it on earnings calls to demonstrate achievement. Reporters use it interchangeably with cash flow to describe earning power. Banks look to it as a way to understand the likelihood of debt repayment. If EBITDA has become the gold standard, what’s the problem?

Buffett’s longtime business partner, Charlie Munger, expressed Berkshire Hathaway’s position on this particular formula best: “I think that, every time you see the word EBITDA, you should substitute the word ‘bullshit’ earnings.”

The Upside

Before diving into why EBITDA is ultimately lacking, let’s quickly review what it actually highlights. As a display of pre-tax, pre-interest, pre-investment earnings, EBITDA can give a good, short-term snapshot of the raw earnings potential of a business. That’s valuable in assessing the potential upside of the business prior to necessary evils like financing, required upkeep, and Uncle Sam.

I keep qualifying these statements with the word “potential” because I’ve never seen a business whose owners have been able to stash away anything close to EBITDA.

EBITDA’s Limited View

In breaking down EBITDA’s shortcomings, the easy place to start is taxes. Depending on the structure of the business and the taxing entities, the government takes between 22.5 percent (federal dividend/capital gains plus Obamacare tax, with no state income tax) and 52 percent of profits (highest bracket for ordinary income, plus California’s state income tax as well as San Francisco’s city income tax) before owners get to save or spend. How the company is structured and where it gets taxed matters tremendously to its real earning power. Again, I’ll leave it to Warren Buffett to explain:

“People who use EBITDA are either trying to con you or they’re conning themselves. Telecoms, for example, spend every dime that’s coming in. Interest and taxes are real costs.”

Like taxes, paying interest on borrowed money doesn’t affect business operations, but it certainly affects the magnitude of earnings. Howard Marks of Oaktree Capital Management explains that “leverage is the ultimate two-edged sword: it doesn’t alter the probability of being right or wrong; it just magnifies the consequences of both.”

If leverage is employed, it’s a crucial component of the business, especially for the equity holders who eat last after profits cascade down through the debt holders. In a debt-free company, a 20 percent downswing in profitability is nothing to worry about. There are still profits accruing for the owners, just at a lesser rate. Contrast that with a highly leveraged company, where a 20 percent decline can force bankruptcy or, at the very least, ensure that the owners don’t see profits for a very long time.

The bottom line is that although debt and taxes don’t directly affect operations, they certainly affect owners, creating opportunities for high-EBITDA companies to have their earning power eroded to nothing.

Emphasis on the ‘DA’

By far, the most dangerous use of EBITDA lies with the “DA.” While technically considered a non-cash expense for accounting purposes, the costs of depreciation and amortization are real. As a refresher, depreciation and amortization are accounting inventions that allow a business to spread the costs of various items over a period of time that (supposedly) mirror their usage. If I buy a piece of equipment (a capital expenditure) with a 10-year life cycle, depreciation allows me to slowly deduct its value from my taxable earnings over that period of time.

It sounds simple enough, and it is. Depreciation can rightfully show the actual earnings power of the business over time, assuming the depreciation schedule and the capital expenditure’s usefulness match up. The challenge lies in dismissing it outright because it’s a non-cash expense after the initial investment is made. That line of thinking makes no sense to anyone trying to estimate future earning power, as Charlie Munger explains:

”There are two kinds of businesses: The first earns 12 percent, and you can take it out at the end of the year. The second earns 12 percent, but all the excess cash must be reinvested — there’s never any cash. It reminds me of the guy who looks at all of his equipment and says, ‘There’s all of my profit.’ We hate that kind of business.”

When evaluating a business, I break depreciation down into three categories. The first is depreciation as an average proxy for maintenance CAPEX, or the amount of reinvestment necessary to keep the business level. The second is a proxy for growth CAPEX, or the amount of investment made to support projected future growth. Lastly, there is tax-planning depreciation, which can result in 100 percent of the anticipated depreciation being taken in the first year, depending on current tax law.

Like any other metric, the goal of looking at depreciation is to understand the risks, rewards, and likelihood for outcomes. If a company has an EBITDA of $2 million, yet spends $2 million per year on new equipment to merely stay open, that’s a terrible business. Conversely, a business with $2 million of EBITDA and very little capital expenditure certainly looks attractive from the owner’s chair, depending on the price you’d have to pay for that piece of furniture.

The point is that depreciation and amortization are real expenses that have a variety of implications for the business. To ignore them is to risk providing a deceiving picture of reality to yourself and others. A great example came in 2002, when WorldCom inflated its EBITDA by classifying $3.8 billion in normal expenses as CAPEX. As expected, that didn’t turn out well.

EBITDAM

When Groupon was getting ready to go public, it tried to amortize its user acquisition expense, claiming that its customers would continue to buy more stuff over time. While that may have been true, that’s always been the case with marketing expenses. In a perfect world, the expense pays off over a long period of time, generating a high return. Just like depreciating a fixed asset, why wouldn’t we want to evaluate companies without taking into account the marketing expense?

To explain the shortcomings of EBITDA, let’s imagine I invented a concept called EBITDAM to evaluate two companies, with the “M” standing for “marketing.” Company A has $100 million in EBITDAM and $50 million in marketing spend. Company B has $100 million in EBITDAM and $10 million in marketing spend. They both trade for $1 billion.

Now, let’s say I came to you and said, “Hey, I found these two companies trading for 10x EBITDAM!” You’d look at me funny and say, “Brent, why the heck do you care what multiple of EBITDAM they’re trading for? What matters is the multiple you’re paying versus what they’re actually earning.”

Then I’d say, “Oh, I know that B is cheaper, but I’m looking for companies without a lot of required marketing spending, so I use EBITDAM.” It’s fair to say you’d assume I was a nutcase.

If you agree with me there, I’ll ask, “Why treat real depreciation and amortization any differently? Why bother with throwing around EBITDA?” The answer you hear is that “D” and “A” are “non-cash,” and that the real annual cash expenditure may differ from the amount recorded in the statements. So they use EBITDA as a proxy for cash flow and make tables of EBITDA multiples.

But this makes very little sense. Let’s say, in our above example, that they did all of their marketing spend in year one, then amortized it over five years in the income account. Would you start paying attention to EBITDAM multiples? Would EBITDAM become a reasonable proxy for cash flow for you? Would you accept a spreadsheet full of EBITDAM multiples? I sincerely doubt it. You’d be deeply skeptical of what I was pushing. You’d subtract whatever real annual marketing spend was necessary and go on your merry way, not paying attention to EBITDAM.

So my problem is that, in most cases, EBITDA doesn’t mean anything more to me than EBITDAM would to you. I’m not looking for a company with a lot of EBITDA any more than I’m looking for a company with a lot of gross profits. What I want to know is what cash earnings — after accounting for some normal level of capital spending — am I getting versus price paid?

Still the Standard

Despite its shortcomings, EBITDA is still used as the standard for valuing businesses. Financial statements are dense and, especially among private businesses, open to many variations of accounting. EBITDA certainly allows for an easier comparison, but at what cost? I’ll let investor Jeff Annello have the final word:

“EBITDA has some use, I suppose, but I think it’s misleading and used to mislead, not to clarify. Often, I think companies fool themselves as well. It’s just a tool I don’t trust in most hands. I think Munger is right to call it ‘bullshit’ earnings.”

This post was originally published on Forbes.