Fuzzy Math That Fueled Junk Debt Boom Is Sparking Jitters

Getting creative with earnings adjustments helps sell deals. 

Fuzzy Math That Fueled Junk Debt Boom Is Sparking Jitters
Markings and formulas are displayed on a whiteboard. (Photographer: Gilles Sabri/Bloomberg)  

(Bloomberg Businessweek) -- Imagine walking into a bank to borrow money. The loan officer might ask for your pay stubs and tax returns to prove your income, as well as for information about your debts and monthly expenses to determine whether you’re a worthy borrower. If the numbers don’t add up, you’d be out of luck.

But what if you could convince your bankers that your income is higher than your stack of documents indicates? What if you promised some belt-tightening over the next couple of years that included moving to a cheaper neighborhood, getting rid of your gym membership, and cutting back on travel? Would they believe you?

In the mid-2000s—the heyday of “liar loan” mortgages—maybe they would have. But banks learned their lesson about taking too much on faith from consumers. For corporations, though, credit is easy again. Some of the riskiest borrowers, and the private equity firms often behind them, have been massaging a measure known as Ebitda—earnings before interest, taxes, depreciation, and amortization—to appear more creditworthy and take out bigger loans.

Money managers, regulators, and credit rating companies say they’re being vigilant. Even so, the fuzzy numbers may be creating a false perception of safety in the $2.5 trillion market for low-rated corporate debt. That market is a key source of funds for companies with less-than-stellar credit and for private equity firms, which typically load the businesses they acquire with debt to boost returns. Institutional investors have increased their purchases of high-risk loans and bonds over the past decade, as near-zero interest rates made other fixed-income assets less attractive. That demand helped make bigger and riskier loans possible. Matthew Mish, head of credit strategy at UBS Group AG, says even a garden-variety recession in the next year or two could cause earnings for the weaker borrowers to drop by as much as 40% from peak to trough, rivaling the declines seen during the global financial crisis. Such a meltdown of corporate balance sheets could fuel a cascade of defaults and bankruptcies.

Fuzzy Math That Fueled Junk Debt Boom Is Sparking Jitters

Ebitda, which became popular in the 1980s as a tool for corporate raiders to better evaluate potential targets, has become a mainstay of corporate finance. It’s seen as a relatively direct measurement of a company’s ability to generate cash, because it strips out the effects of management’s decisions on capital investments and indebtedness. In theory it provides evidence of how much money the company could have available to service its debt.

Over the past several years, Wall Street lawyers and advisers have worked to squeeze generous adjustments into the Ebitda calculation, helping make purchase prices look smaller and debt loads more manageable. Exela Technologies Inc., a document processing company formed through the merger of SourceHOV and Novitex, relied on several rounds of adjustments to boost Ebitda when it borrowed money to finance the deal in 2017.

After removing interest, tax, depreciation, and amortization, plus some one-time costs, the company arrived at Ebitda of $247 million. In another round of adjustments, Exela counted expected benefits from shutting offices, cutting workers, and renegotiating vendor contracts. The resulting “further adjusted Ebitda” showed it pulling in $353 million for the 12 months ended on March 31, 2017. Based on generally accepted accounting principles, the combined company would have lost almost $63 million over that period. In a statement, Exela Chief Executive Officer Ronald Cogburn said it was important for investors to consider both a company’s current financial reporting and the potential beneficial impact of strategic planning on future plans.

In perhaps the most flagrant example of creative Ebitda, office-sharing company WeWork turned a $933 million loss into $233 million of what it called “community adjusted Ebitda” last year, when it issued its debut bond. The widely ridiculed—and since discontinued—metric excluded even basic general and administrative expenses. “We have never seen a net negative adjustment to Ebitda—it only goes up,” says Jason Dillow, CEO of Bardin Hill, a credit investment firm. “It is basically what can you get away with while keeping a semi-straight face.”

Professors Adam Badawi of the University of California at Berkeley and Elisabeth de Fontenay of Duke used a machine learning algorithm similar to those used to identify spam in email traffic to screen more than 4,000 credit agreements filed with the Securities and Exchange Commission from 2011 to 2018. Their study showed wide variation in how companies define Ebitda.

Take Del Frisco’s Restaurant Group Inc., the New York steakhouse chain. The company sought a $390 million loan last year to finance its acquisition of Spanish tapas chain Barteca Restaurant Group. It took Del Frisco 2,723 words to explain what Ebitda means, the most of any borrower included in the study. Its definition included 22 types of adjustments it could use to boost its reported earnings. By contrast, blue-chip companies such as chemical giant 3M Inc. and advertising agency Omnicom Group Inc. used less than 40 words for their Ebitda definitions. The most concise borrower used 10, barely enough to spell out the acronym. The simpler the definition, the less “creative” the Ebitda.

An overly complex definition can weaken protections for buyers of the company’s debt. That’s because key tests to determine if the company can take on additional debt, make investments, or pay dividends are typically calculated with Ebitda. “In instances where you have extraordinarily permissive language, it is much more like a self-certification than anything else,” de Fontenay says.

Ebitda inflation has picked up. In the first quarter of this year, companies issuing leveraged loans for mergers and acquisitions inflated their Ebitda by an average 43%, according to Covenant Review, an independent research firm that analyzes debt documents for investors. That’s the highest of any quarter in the data, which dates to 2015.

Evidence suggests that when companies make optimistic adjustments to Ebitda, disappointments are the norm soon after a deal is done. According to a recent study by S&P Global Ratings, companies involved in a merger or a leveraged buyout in 2015 missed their own earnings projections by an average 29% in the first year following the deal. For deals originating in 2016, the projections were off 35%.

Investors have already had a taste of what happens when things don’t work out as expected. Since 2017, Exela has been unable to improve earnings significantly with cost cuts and other moves, according to Moody’s Investors Service, which in November lowered Exela’s credit rating to Caa3, nine notches below investment grade. Exela’s profit margins have dwindled amid stiff competition, while charges for restructuring costs and goodwill impairment contributed to a net loss of $282 million in the 12 months ended on Sept. 30. Its bonds, meanwhile, dropped to just above 30¢ on the dollar. No reason to panic, though. The company says its further adjusted Ebitda is still growing.

To contact the editor responsible for this story: Eric Gelman at

©2019 Bloomberg L.P.