Hilton vs Toys R Us: Why One LBO Returned $14B and the Other Returned Zero

If you set the two deals next to each other on a whiteboard, they look like variations of the same recipe. A megafund buys a household-name consumer business, loads it with debt, holds it for the better part of a decade, and hands the story to the trade press to write up.

Blackstone bought Hilton in October 2007 for a $26.0B enterprise value and returned $14.0B in proceeds by 2018 — a 2.5x MOIC and an 18.7% IRR over a 6.2-year weighted hold. KKR, Bain Capital and Vornado bought Toys "R" Us in July 2005 for a $6.6B enterprise value and returned zero. Total proceeds to LPs at the end of the 12-year hold: $0.0B. MOIC: 0.0x. IRR: -100%.

Same era, same playbook on the surface, opposite outcomes. This post is the side-by-side. Not a drill-down on leverage or on the retail vs. hospitality question — those are separate posts. This is the whole scorecard: what the two sponsor teams saw, chose, and did across the full lifecycle of each deal.

The one-page comparison

Everything downstream of this table is commentary on it.

Read across the two columns and the differences that jump out are entry multiple and leverage. Read down each column and the story changes. Hilton went in at what looked like the wrong price and the wrong leverage; Toys went in at what looked like the safer price and the safer leverage. Then the world happened.

Difference 1 — the underlying business

The single most consequential thing that separated the two deals was not visible on the S&U or in the returns waterfall. It was in the P&L that gets built underneath both.

Hilton's $1.2B of LTM EBITDA at close was cyclical — hotel demand collapses in a recession and comes back. By the December 2013 IPO the enterprise value had grown from $26.0B to $32.0B, reflecting the recovery in hotel demand from the 2009 bottom.

Toys R Us's $550M of LTM EBITDA at close was structurally declining. Amazon had launched its dedicated toy warehouse two years before the buyout — the diligence pack should have called it out and the model should have pressure-tested it, but the sponsors underwrote the deal as if kids' toys were a slow-growth utility. The EBITDA didn't come back after a downturn; it kept drifting down as e-commerce ate the category. By the mid-2010s the $400M/year of cash interest on the $5.3B debt load was chewing up nearly all of what remained, leaving nothing to invest in the digital transformation that might have saved the business.

Same 12x-21x range on entry multiple; wildly different EBITDA trajectories underneath. That's why the "high entry multiple" screen alone is misleading. The number that mattered wasn't the multiple paid — it was the durability of the earnings you were paying it against.

Difference 2 — margin for error in the capital structure

Hilton went in with 8.4x leverage, higher than Toys' 5.7x. But interest coverage isn't set by the leverage multiple — it's set by the cash interest bill against the EBITDA that shows up in a bad year.

Toys' math looked like this at close:

Any meaningful drop in EBITDA and the entire cash cushion evaporated. There was no room for capex on e-commerce, no room to fund a store-remodel program, no room for the working capital hit that comes when inventory sits longer. The capital structure was built for a world where EBITDA grew — and the world that arrived wasn't that world.

Hilton had the mirror-image profile: high headline leverage, but on an EBITDA base that could scale back up. When RevPAR cratered in 2009, Blackstone still had the balance sheet flexibility (and the sponsor willingness) to renegotiate rather than default. Toys, on a smaller absolute EBITDA base against $400M of annual interest, had no equivalent lever. The debt was smaller in dollar terms but harder to service in reality.

If you want to see the same dynamic play out on your own inputs, the free debt capacity calculator surfaces whether interest coverage or the leverage cap is your binding constraint at close — and stress-tests what breaks if EBITDA falls. Running Toys' numbers through it in 2005 would have raised red flags well before the diligence memo hit IC.

Difference 3 — active capital structure management during crisis

The deals also diverged sharply in what sponsors did when the world got bad.

By 2009 Hilton was staring down covenant issues as hotel demand collapsed. Blackstone didn't fold — in 2010 it restructured roughly $4B of Hilton's debt to avoid default. That single decision preserved the option value that eventually paid off at the 2013 IPO. Without the 2010 restructuring there is no $14B exit.

Toys' sponsors did the opposite in spirit: they extracted rather than invested. Over the 12-year hold, KKR, Bain and Vornado collected roughly $470M in management fees and dividend recaps combined. That's a real number — enough to make sponsors partially whole even after LPs lost 100%. But every dollar paid out to the sponsor was a dollar that wasn't retained to defend against the Amazon threat, and every dividend recap raised the debt load on the way in.

By the time bankruptcy was on the table in 2017, Toys' capital structure had ratcheted up rather than down. The real-estate spinoff strategy — using Vornado's expertise to unlock Propco value — got blocked by lender covenants once the company slid into distress. The lever that was supposed to save the deal was disabled by the deal's own capital structure.

Difference 4 — the exit path

Hilton exited through a staged IPO and sell-down. First tranche: December 2013 IPO at a $32.0B enterprise value. Then a multi-year sell-down through 2018, alongside the Park Hotels & Resorts and Hilton Grand Vacations spinoffs that separated the real-estate and timeshare assets from the operating brand. Total proceeds to LPs across the whole exit sequence: $14.0B.

Toys had no exit path. Chapter 11 filed September 2017; the sponsors tried and failed to restructure through the winter; liquidation announced March 2018; final store closings June 2018; ~33,000 employees laid off. Enterprise value at "exit" is not a number that gets computed — the estate paid down creditors in the waterfall order and equity was wiped.

Notice how those two exits also reshape how you'd frame the returns. If you had been shown Hilton's numbers in isolation — a 2.5x MOIC over 6.2 years for a 18.7% IRR — you might call it fine but not heroic. It sits comfortably in the middle of most funds' return distributions. The reason trade publications called it the greatest PE deal of all time isn't the multiple, it's the absolute dollar profit — roughly $8B of fund carry-eligible profit, on the largest single equity check the industry had ever written. For a deeper walk-through of why hold-period-adjusted returns matter more than headline multiples, see MOIC vs IRR.

Toys is the opposite side of the same lesson. A 12-year hold with a 0.0x MOIC is not a slow deal — it's a total loss with fees extracted on the way to zero. The hold-period-adjusted returns metric there isn't -100% IRR because that's not what LP paperwork actually shows; it's the vintage-year IRR drag against an entire fund's performance, plus the reputational drag on the sponsor's next fundraise. Both are real, and both track back to the same underlying failure: the underwriting missed the direction the world was moving.

Difference 5 — what the sponsors actually got paid for

The subtlest difference is what each sponsor team was earning across their hold.

Blackstone earned by being patient. The value the fund captured was realized on the exit side — through the recovery in hotel demand, the reset in trading multiples, and the strategic separation of Park Hotels and Hilton Grand Vacations. Compound patience through 2009, restructure the debt in 2010, hold through the recovery, take proceeds when the market is ready. If any of those decisions had gone the other way, the outcome moves toward zero.

The Toys sponsors earned by extracting. Roughly $470M in management fees and dividend recaps over 12 years made sponsors partially whole even as the underlying business faded. Every dollar paid out was a dollar not retained to defend against Amazon, and every recap raised the debt load. Sponsors who reinvest into a deal during distress tend to win more of the tail; sponsors who extract during distress tend to keep themselves whole and leave the LPs holding the bag.

What this comparison is (and isn't) evidence for

It's easy to over-generalize from two deals. Not every hotel LBO wins. Not every retail LBO loses. The Toys sponsors are not villains and the Blackstone team are not oracles — Toys made a defensible bet on a business that had been a steady utility for four decades, and Hilton made a defensible bet at what looked like a bubble-top multiple. Reasonable people running each deal at close would have signed the papers.

But the comparison does test whether the mechanical inputs on an LBO — entry multiple, leverage ratio, hold period — actually predict the outcome. They don't, not by themselves. What predicts the outcome is:

• Whether the EBITDA is cyclical or structurally declining
• Whether the interest bill leaves meaningful room to reinvest in the business in a bad year
• Whether the sponsor is willing and able to inject capital or restructure the debt when the crisis hits
• Whether the exit path stays open through the whole hold, not just the base case

Every one of those goes back to something you should be modeling before signing. Downside EBITDA cases, interest coverage stress tests, debt restructuring optionality, exit-multiple sensitivity. In every LBO worth building, they show up on the sensitivity tabs — and the number that appears on the returns page is the output of those inputs, not a standalone assumption.

Build it end-to-end

The All-in-One PE Model includes the full LBO returns build (base case + downside sensitivity), the debt schedule with covenant tracking, and a dividend recap module that shows what leverage looks like after each recap — so you can see what the Toys sponsors saw before the covenants tripped, and what Blackstone saw before the 2010 restructuring. See the sheet preview before buying.

Two deals. The same era, similar sponsor caliber, similar-looking sources & uses. One returned $14B and became the trade press's favorite illustration of what PE can do when it's done well. The other returned zero and became the case study for what happens when the model doesn't stress the right variables.

Everything that separated them was in the pages behind the returns page.

---

Sources:

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