Why 8.4x Leverage on Hilton Worked but 5.7x on Toys R Us Didn't

Here's a piece of trivia that breaks most associates the first time they hear it.

Blackstone bought Hilton in 2007 with 8.4x leverage. KKR, Bain and Vornado bought Toys "R" Us in 2005 with 5.7x leverage. Hilton ran more debt against EBITDA, closed at the peak of the cycle, and rode straight into the worst hotel demand collapse since the 1930s. Toys ran less debt against EBITDA, in a category — kids' toys — that had been a slow-growth utility for forty years.

Hilton returned $14B in proceeds at a 2.5x MOIC. Toys returned zero.

If you've been taught that leverage is the lever you set carefully in cell B17 because more debt equals more risk, this comparison should bother you. The arithmetic was the other way around, and the outcome was the other way around.

What was actually different — and how senior partners actually size leverage — is the rest of this post.

What the leverage ratio actually measures

Debt-to-EBITDA is a snapshot. It tells you how many years of today's EBITDA it would take to retire the debt if you applied 100% of it. That's a useful sanity check, and it's the number lenders quote in their term sheets. It is not a forecast of whether the company can service the debt next year, or the year after, in the world that actually arrives.

What kills LBOs isn't the headline leverage ratio. It's three things underneath it:

1. The stability of the EBITDA that the ratio is built on. $550M of EBITDA that's structurally declining is not the same as $1.2B of EBITDA that's cyclically depressed.
2. Cash conversion — how much of EBITDA actually shows up as free cash flow after capex, taxes, and working capital. Two businesses with the same EBITDA can have wildly different cash-on-cash service capacity.
3. The interest coverage — EBITDA divided by cash interest. This is what determines whether covenants get tripped in a bad year, not the multiple of EBITDA you closed with.

Hold those three in mind. They're why the headline leverage ratios in this comparison lie.

Hilton at 8.4x — why it worked

Blackstone closed Hilton in October 2007 at a $26.0B enterprise value on $1.2B of LTM EBITDA. That's a 21.7x entry multiple — the kind of multiple that makes a healthy associate's stomach turn — financed at 8.4x net debt to EBITDA. Sponsor equity was $5.6B.

The deal looked terrible by mid-2009. Hotel demand collapsed, RevPAR cratered, and on paper Blackstone was sitting on a covenant breach waiting to happen. Then two things saved it.

First, Blackstone restructured roughly $4B of the debt in 2010 instead of letting it default. They bought back debt at discounts, extended maturities, and pushed the wall out far enough that the recovery in hotel demand could catch up to the capital structure. The cash interest burden came down materially in that exchange.

Second — and this is the part associates miss — Hilton's underlying EBITDA stream was cyclically depressed, not structurally declining. Hotels are franchise-heavy, brand-driven, and benefit from a steady secular tailwind in business and leisure travel. The 2009 EBITDA hole was real but temporary. Once travel recovered, EBITDA recovered with it, and the same 8.4x leverage that looked terrifying in 2009 looked normal again by 2012.

By the December 2013 IPO, enterprise value was $32B. Blackstone sold down over the next four and a half years, ultimately returning $14B in proceeds — a 2.5x MOIC, an 18.7% IRR over a 6.2-year weighted hold.

The 8.4x leverage didn't kill Hilton because Blackstone bought a cyclical business at the peak, restructured the capital structure proactively when the cycle turned, and held long enough for the cyclical EBITDA to recover. That's the patient-capital thesis. The leverage ratio looked dangerous; the cash flow profile underneath it was survivable.

Toys at 5.7x — why it didn't

KKR + Bain + Vornado closed Toys "R" Us in July 2005 at a $6.6B enterprise value on $550M of LTM EBITDA — a 12.0x entry multiple, financed at 5.7x leverage. Sponsor equity was $1.3B. The capital structure carried $5.3B of debt, which generated roughly $400M of cash interest per year.

That single ratio — $400M of cash interest against $550M of EBITDA — is the whole story. Interest coverage on day one was about 1.4x. There was almost no room for EBITDA to fall before cash flow went negative on a maintenance basis, and zero room left over to invest in the business.

Then Amazon happened. Amazon had actually launched its toy warehouse two years before the buyout, in 2003, but the sponsors didn't underwrite the e-commerce disruption seriously in the model. By the early 2010s the structural shift was unmistakable: toys were one of the easiest categories on earth for Amazon to win, because parents shop on price, the SKUs are standardized, and there's no try-before-you-buy dynamic.

EBITDA didn't crash overnight. It bled. And every dollar of EBITDA decline ate directly into the cash that would have funded an e-commerce transformation — exactly the investment Toys needed to make to survive.

Here's what the cash math looked like in a normal year versus a bad one:

At close there was $100M of reinvestment cash. In a "bad year" — and bad years became the new normal — there was nothing. You cannot rebuild a digital business with $0 of growth capital.

There's a separate ledger people forget. Sponsors collected roughly $470M in management fees and dividend recaps over the twelve-year hold. So the sponsor-level economics weren't a total wipeout. But fund LPs were wiped out: 0.0x MOIC, -100% IRR, $0 in distributions. The 5.7x leverage ratio looked moderate next to Hilton. It wasn't — because the underlying EBITDA couldn't support it once the world changed.

The two deals, side by side

The leverage column is the one that gets all the attention. It's also the one that's most misleading in isolation. Look across the row at "underlying EBITDA character" — that's where the deals actually parted ways.

How PE pros actually size leverage

Senior partners don't size leverage off a number-of-turns rule. They size it off a downside-EBITDA stress test. The questions they actually run are:

• What is EBITDA in a recession year? Not LTM. Not a budget. The realistic trough.
• Is the trough EBITDA cyclical or structural? Cyclical EBITDA comes back. Structural EBITDA doesn't.
• What is interest coverage at the trough? EBITDA divided by cash interest. Below ~1.5x in the trough year and you're a covenant breach away from losing control of the company.
• What is the FCF conversion rate? EBITDA isn't cash. After maintenance capex, working capital, and taxes, what's left to service the debt and reinvest?
• What does the maturity wall look like? A 5x leveraged business with all the debt maturing in year 6 is much more dangerous than the same business with the same leverage in tranches that mature staggered out to year 10.

Apply that stack to Hilton and the 8.4x leverage clears — the trough was bad but the recovery was real, and Blackstone bought themselves time with the 2010 restructuring. Apply it to Toys and the 5.7x leverage doesn't clear — the trough kept getting lower because the structural decline was permanent, and there was no operational pivot available without growth capital the capital structure had eaten.

For the mechanics of how a sources-and-uses table builds out — and how net debt, equity rollover and financing fees fit together when you're actually setting the leverage — the sources & uses walkthrough covers the construction step-by-step.

For a quick downside test on a specific deal, the free debt capacity calculator flips the question around: input EBITDA, interest rate and a coverage covenant, and it shows you the maximum debt the business can carry before coverage breaches. Stress-test it at a 25–30% EBITDA decline and you'll see which deals have margin and which ones don't — exactly the test the Toys R Us sponsors should have run against an Amazon-shaped downside in 2005.

If you want to see the returns side play out at different leverage levels for a specific deal thesis, the LBO returns calculator runs the debt-paydown loop and surfaces IRR and MOIC against your inputs in seconds.

The practical takeaway

Stop quoting leverage as "5x" or "8x" like it's a property of the deal. It's a property of the business. The same 6x leverage is conservative on a regulated utility with 95% cash conversion and terrifying on a fashion retailer with negative comps. The Hilton-vs-Toys comparison is the cleanest illustration of that in modern PE history: more debt on the better cash flow profile worked, less debt on the worse cash flow profile didn't.

When the IC asks "what's the leverage?" the right answer is never just a number. It's the number, the trough EBITDA, the trough coverage, and a one-sentence answer to whether the trough is cyclical or structural. Get those four right and the deal is either obviously safe or obviously not.

Our All-in-One PE Model builds the leverage and coverage stress tests directly into the LBO and Debt Schedule tabs — seven tranches, configurable covenants, and sensitivity tables that show coverage and equity returns across a 5×5 EBITDA-and-multiple grid. See the sheet preview before buying.

The deal that breaks isn't the one with the scary leverage ratio. It's the one with the scary EBITDA underneath it that nobody flagged.

---

Sources:

1. Hilton S-1 Filing (Dec 2013), SEC EDGAR — entry multiple, leverage, and original transaction terms
2. Bloomberg, "How Blackstone Made One of the Greatest Deals in PE History," May 2018 — total proceeds, 2010 debt restructuring detail
3. Blackstone Q4 2018 Earnings Release, IR.Blackstone.com — final exit IRR and MOIC
4. Toys "R" Us Chapter 11 Petition (S.D. NY 2017), Prime Clerk docket — capital structure and debt load detail
5. House Financial Services Committee Hearing (Feb 2019), "The Toys R Us Bankruptcy and the Aftermath" — interest expense, EBITDA, and sponsor fee history
6. Bloomberg, "How Vulture Capitalists Ate Toys R Us," March 2018 — deal close detail and dividend recap history