When KKR, Bain Capital and Vornado bought Toys "R" Us in July 2005, they put about $1.3 billion of fund LP equity into the deal. Twelve years later, that equity was worth zero — the company filed Chapter 11 in September 2017 and liquidated by June 2018.
Read the press from the time and you'd think the sponsors had been crushed too. They hadn't. Across the 12-year hold, the sponsor group collected roughly $470 million in management fees and dividend recapitalizations — money that flowed to GPs regardless of the eventual zero return to LPs.
That decoupling is what makes dividend recaps the most contentious mechanic in private equity. Let's unpack what they actually are, how the cash gets out, and why LP and sponsor outcomes can run in opposite directions on the same deal.
What a dividend recap actually is
A dividend recapitalization is the same maneuver a homeowner runs when they cash-out refinance a mortgage. The portfolio company adds new debt to its balance sheet, and the proceeds of that debt are distributed to equity holders as a special dividend — bypassing the need to actually sell the business or generate the cash from operations.
The structure has three properties that make it irresistible to sponsors when credit markets cooperate:
- It returns capital without exiting. Sponsors distribute cash to fund LPs without losing ownership of the business — cake and eat it.
- It spikes IRR dramatically. Because IRR weighs early cash flows heavily, a recap in year 2 or 3 can lift the IRR even when MOIC stays modest. (See the MOIC vs IRR tradeoff — recaps live exactly at the seam between the two metrics.)
- It's funded by the portfolio company's balance sheet, not the fund's. Sponsors don't write a check. The company's new lenders do.
The economics are clean when the underlying business can carry the additional debt service. When it can't, the recap merely accelerates the timeline to distress.
How the mechanic works in practice
Picture a portfolio company two years into the hold. Original capital structure: $300M of debt at 5x EBITDA of $60M. EBITDA grows to $80M; leverage drops to 3.75x organically. The credit market is open; spreads have tightened.
Sponsors raise a new $400M term loan, refinance the existing $300M, and distribute the $100M of incremental proceeds to themselves. Leverage resets to 5x — exactly where it started — but the sponsor has now banked $100M of return without selling a share.
If the original equity check was $200M, the sponsor has just realized a 0.5x distribution on cost in the first three years of the hold. The remaining equity is still worth whatever the eventual exit pays — but the worst-case downside has been partially derisked by cash already pulled out.
It's a perfectly legal, perfectly common transaction. It's also a precisely engineered transfer of risk from the sponsor to the portfolio company's lenders and employees.
The Toys "R" Us case — sponsors took $470M while LPs got zero
In July 2005, KKR, Bain Capital and Vornado closed Toys "R" Us at a $6.6 billion enterprise value on $550 million of LTM EBITDA — a 12.0x entry multiple at 5.7x leverage. Sponsor equity at close: roughly $1.3 billion of fund LP money.
| Metric | At close (2005) |
|---|---|
| Enterprise value | $6.6B |
| LTM EBITDA | $550M |
| Entry multiple | 12.0x |
| Leverage | 5.7x ($5.3B debt) |
| Annual cash interest | ~$400M |
| Sponsor equity | $1.3B |
Twelve years later, the equity was worth zero. The Chapter 11 filings cite the $5.3B debt load explicitly as the reason the company couldn't compete with Amazon. Roughly 33,000 employees lost their jobs in the 2018 liquidation.
| Metric | At exit (2018) |
|---|---|
| Total proceeds to LPs | $0 |
| MOIC to LPs | 0.0x |
| IRR to LPs | -100% |
| Hold period | 12 years |
| Sponsor fees + recaps collected | ~$470M |
That last row — ~$470M to sponsors against a $0 LP outcome — is the heart of the dividend recap controversy. Two sets of investors in the same deal saw outcomes that ran in opposite directions.
Where the $470M came from
The $470M aggregates two sponsor compensation streams that the grounding documents and Bloomberg's 2018 reporting describe flowing across the 12-year hold: annual management fees paid by the fund vehicles, and the proceeds of dividend recapitalizations executed when credit markets were supportive.
The annual management fee runs for the life of the hold whether the deal is going well or not. The fee is calculated against committed capital, accrues regardless of portfolio company performance, and compounds across the years before LPs see any net carry distribution.
The dividend recaps sit on top. The mechanic, common to large PE recaps of the era: the portfolio company raises new debt, refinances the existing facility, and distributes the incremental proceeds to the fund vehicles that own the deal — and, by extension, to the GPs that manage those vehicles.
Both streams are senior to LP returns in the cash waterfall. The portfolio company pays its lenders first, then its management fees, then its sponsors. Whatever's left at exit — zero, in the Toys case — is what LPs receive.
Why the capital structure broke
The 5.7x leverage at close translated to ~$400M of annual cash interest against $550M of EBITDA. When EBITDA fell — driven by exactly the e-commerce competition that Amazon's launch had telegraphed two years before the buyout — coverage collapsed.
The sponsors had ~$150M of EBITDA-minus-interest in the good years to fund every other obligation: working capital, capex, store renovations, e-commerce transformation. By the time the business actually needed to spend on digital, the cash simply wasn't there.
If you want to see exactly how this dynamic collapses on paper, the free debt capacity calculator shows the leverage cap and the DSCR floor on the same screen. Stress EBITDA down 20–30% and watch coverage break — that's the test the 2005 sponsors should have run against an Amazon downside case before adding any recap debt on top.
The underlying capital structure error is the one the sources & uses walkthrough covers in depth: the S&U was internally consistent, but the leverage was wrong for the business. Each dividend recap during the hold reset that structure to roughly its original (wrong) leverage just as EBITDA was about to crack.
Why LPs tolerate the structure
If sponsors can win even when LPs lose, why do LPs sign limited partner agreements that allow it?
Three reasons:
- Recaps generate distributions in good deals too. When the underlying business actually performs, recaps return capital faster than holding to exit — improving DPI and IRR for LPs alongside GPs. The same mechanic that produced $470M to sponsors in Toys produces meaningful early LP distributions in deals where EBITDA does grow to support the new debt.
- Management fees are how funds operate. A fund without management fees can't pay associates, fund diligence, or maintain a deal team between transactions. The fee is the price LPs pay for active management.
- Carried interest still ties sponsor wealth to LP returns. GPs make truly large dollars from carry on successful exits, not from management fees. Toys cost KKR, Bain and Vornado a meaningful carry pool that — if the deal had worked — would have dwarfed the $470M in fees collected along the way.
The grievance isn't that fees exist. It's that the structure permits situations where fees alone make sponsors meaningfully whole on deals that wiped out LP capital.
The stress test before any recap
The honest question a senior partner should ask before signing a dividend recap is the same question that should have been asked at the original Toys S&U: can this business carry the post-recap debt against a realistic downside EBITDA?
If the answer requires base-case EBITDA growth to maintain coverage, the recap isn't conservative balance sheet management — it's an option on continued favorable conditions. When conditions turn, the recap locks the company into a debt level it can't service.
Run the test in 30 seconds with the LBO returns calculator — set the original leverage, growth, and exit assumptions, then layer a recap event in year 3 and see what post-recap coverage looks like in a –20% EBITDA scenario. If coverage breaks under the downside, the recap is the wrong call regardless of what current credit markets are willing to fund.
What this looks like in a real model
Our All-in-One PE Model builds the recap mechanic directly into the LBO tab — configurable recap events with their own debt tranches, distribution timing, and the downstream impact on IRR, MOIC, and DPI. The sensitivity tables show what each recap does to LP returns under base, downside, and severe stress cases — exactly the test that the Toys recaps consistently failed against an honest Amazon downside.
See the LBO sheet preview before buying.
The dividend recap is one of the most powerful tools in the PE playbook. Used against a business with genuine excess debt capacity, it accelerates LP returns and compounds DPI. Used against a business that can't carry the additional debt, it transfers wealth from future LP outcomes — and from the employees and lenders left exposed — into present-day sponsor cash.
Toys "R" Us is the canonical cautionary case. It's worth knowing the mechanic cold before signing off on the next one.
Sources:
- Toys "R" Us Chapter 11 Petition (S.D. NY 2017), Prime Clerk docket — original capital structure and bankruptcy filing
- House Financial Services Committee Hearing (Feb 2019), "The Aftermath of the Toys R Us Bankruptcy" — interest expense, EBITDA figures, employee impact
- Bloomberg, "How Vulture Capitalists Ate Toys R Us," March 2018 — deal close details and sponsor fee + dividend recap history
- Center for Popular Democracy, "Pirate Equity: How Wall Street Firms Are Pillaging American Retail" — analysis of sponsor compensation versus LP outcomes