Three Ways PE Returns Capital: Dividend Recaps, Refis, and Exits

Ask an LP what "return of capital" means and you'll get a specific answer: cash back in the fund account, ready to be pushed on to the pension, endowment or family office that wrote the original commitment.

Ask a sponsor the same question and you get a menu. There are three ways PE actually gets money out of a deal — and only one of them involves selling the business.

The other two happen while the portfolio company is still very much on the books. They matter enormously to the sponsor's realized MOIC, they matter to the underlying company's survival, and they occasionally leave sponsors partly whole on deals that wipe LPs out entirely.

The three channels

Every dollar that flows from a PE portfolio company back to fund investors travels through one of three channels:

1. Dividend recapitalization. The company borrows new debt and distributes the proceeds as a special dividend. Ownership doesn't change.
2. Refinancing distributions. The company refinances existing debt on better terms — lower rate, longer tenor, laxer covenants — and any excess proceeds get distributed.
3. Exit. The sponsor sells the equity — via IPO, secondary sale to another sponsor, or strategic sale to a corporate. Ownership changes.

Two of the three are variations of the same maneuver: the balance sheet does the work, and the sponsor puts money in the fund's pocket without giving up the underlying asset. Only the third actually crystallizes a valuation with a real buyer on the other side of the table.

That distinction is where LPs and sponsors sometimes see the same deal very differently. A big enough recap can leave the sponsor's fee-and-carry economics whole even if the ultimate exit turns out to be a zero.

Channel 1 — dividend recaps

A dividend recapitalization is the same maneuver a homeowner runs when they cash-out refinance a mortgage. The portfolio company takes on new debt above and beyond what it needs to refinance existing obligations, and the excess proceeds are distributed to equity holders as a special dividend.

The mechanic 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.
• It front-loads cash flow to the fund. Because IRR weights early cash heavily, a mid-hold recap of $100M–$300M can pop reported IRR by hundreds of basis points on a moderately performing deal, even when the eventual MOIC comes in flat.
• It's funded by the portfolio company's balance sheet, not the fund's. The sponsor doesn't write a check — new lenders do.

The trade-off is written into the same paragraph. The portfolio company now carries higher leverage than it did going in, and it has less cushion to absorb an operating miss. Every recap is a redistribution of downside risk from the sponsor to the company's operators, lenders and employees.

Channel 2 — refinancings

Refinancings sit on the same continuum as recaps but with a different purpose. Sometimes the goal is a lower rate; sometimes it's a longer maturity that pushes the covenant reset out beyond the sponsor's expected hold; sometimes it's tripping the accordion feature and adding incremental capacity for a bolt-on acquisition.

Most refinancings don't distribute anything — they just re-tune the capital stack. But when spreads tighten between year-one and year-three of a hold, a refi can generate genuine excess proceeds that flow back to the fund. A textbook example: swap a term loan priced at SOFR + 500 for one at SOFR + 350 after two years of debt paydown, upsize slightly, distribute the delta.

The mechanic isn't glamorous but it's cumulative. Over a 5–7 year hold on a well-run platform, a series of refinancings and small distributions can meaningfully shift the IRR without ever touching the operating business.

Channel 3 — the exit

The exit is where the story finally clears the fog. Three flavors:

• IPO — the sponsor takes the company public, sells a portion of the equity in the primary/secondary offering, and sells the rest over subsequent quarters via block trades or forward sales. This is a multi-year process; the reported "exit date" is usually the initial pricing.
• Sponsor-to-sponsor sale — another PE firm buys the equity outright at a negotiated price. Clean, fast, no ongoing exposure.
• Strategic sale — a corporate acquirer buys the equity, often at a premium to what another sponsor would pay because of synergies. Rare in the middle market; common at scale.

Every one of these ends the sponsor's economic interest in the business. The difference between the three is mostly speed of full monetization and how long the sponsor stays exposed to post-close volatility.

Hilton — a textbook exit-driven return

Blackstone's Hilton buyout is the cleanest example of a fund making its money almost entirely through channel three. The deal closed in October 2007 at a $26.0B enterprise value on $1.2B of LTM EBITDA — a 21.7x entry multiple, financed at 8.4x leverage with about $5.6B of sponsor equity.

The hold immediately hit turbulence. The 2008 hotel-demand collapse forced Blackstone to restructure roughly $4B of debt in 2010 to avoid default. That restructuring wasn't a distribution — it was a survival move. No cash went back to the fund during the crisis years; the sponsor was buying time on the balance sheet.

Distributions only began after the December 2013 IPO, which priced Hilton at a $32B enterprise value. Even then, the exit was a multi-year sell-down: primary IPO shares, subsequent block sales, plus the spinoffs of Park Hotels & Resorts (real estate) and Hilton Grand Vacations (timeshare) before the final sell-down in May 2018.

Notice the shape: roughly $8B of fund profit sitting on ~$5.6B of invested capital, essentially all of it delivered at the exit. There is no mid-hold recap padding out the IRR here — Blackstone couldn't have recapped a levered hotel portfolio in 2009 even if they'd wanted to. The MOIC and IRR are honest; the number reported to LPs is very close to the number LPs actually received.

That's the model when the underlying asset survives. Patient exit, single clean channel of return, no interim daylight between what LPs see on paper and what they get.

Toys "R" Us — recaps and fees while LPs got zero

The Toys "R" Us story shows what happens when the balance sheet does all the work and the exit never materializes.

KKR, Bain Capital and Vornado closed the deal in July 2005 at a $6.6B enterprise value on $550M of LTM EBITDA — a 12.0x entry multiple at 5.7x leverage ($5.3B of debt). Sponsor equity at close: approximately $1.3B of fund LP money.

The debt load translated to roughly $400M per year of cash interest against $550M of EBITDA. That left the company almost nothing to reinvest against Amazon's toy warehouse, which had launched two years before the buyout. When EBITDA drifted down, there was no cash for an e-commerce transformation. The company filed for Chapter 11 in September 2017 and liquidated by June 2018. Roughly 33,000 employees lost their jobs.

For the fund LPs, the story ends there. Zero proceeds. 0.0x MOIC. Negative-100% IRR.

For the sponsors, it doesn't. Over the 12-year hold, KKR, Bain and Vornado collectively extracted approximately $470M in management fees and dividend recapitalizations. That isn't a fund-level return in any honest sense — it's an entity-level fee-and-recap stream that flowed to GPs regardless of what happened to LP equity.

This is the version of "return of capital" that most stings LP allocation committees. The sponsor was made partially whole by channel 1 (recaps) and by management fees running through the holding period. The LP equity was never made whole at all.

Why the three channels aren't interchangeable

Sponsors sometimes talk about the three channels as if they're substitutes — pick your route, get the money out. In practice they're very different animals for the people on the other end of the cash flow.

• Channels 1 and 2 are debt-funded. They increase the portfolio company's leverage in exchange for cash today. They don't validate an underlying valuation. If the exit later fails, LPs may still lose more than they got back from the recaps.
• Channel 3 is buyer-validated. A real transaction at a real price closes the loop. The sponsor's MOIC calculation stops being an opinion and becomes a receipt.

The order of return matters too. A recap in year 2 that returns half the fund's cost basis materially reduces LP downside — even if the eventual exit disappoints. But the same recap taken by a sponsor who then runs the company into bankruptcy is small comfort at the LP level. Toys "R" Us LPs, on balance, would have preferred a lower-leverage capital structure at close over the recaps they eventually saw.

What to look for in your own model

If you're building an LBO, the three channels of return should be explicit line items on the returns waterfall, not embedded assumptions buried inside the exit-multiple table. Try running the same base case through a lightweight tool like our free LBO returns calculator — it separates the interim distributions from the terminal exit so you can see which channel is doing the work.

Then ask three questions of any base case:

1. How much of the projected MOIC comes from mid-hold distributions vs the terminal exit? A model where 60% of MOIC is a year-3 recap is not the same deal as one where 100% is the year-6 exit. LPs will discount the former.
2. What's the IRR spike from any modeled recap? If IRR jumps 400 basis points because of a year-2 dividend, disclose it — don't let it lurk inside a headline number the IC will assume is operating-driven.
3. What does the IRR look like with zero recaps? Strip the interim distributions out and re-run. If the deal only clears the fund's IRR hurdle with recaps included, you're relying on channel 1 as a substitute for channel 3.

Our All-in-One PE Model breaks distributions into interim recap, refinancing proceeds, and terminal exit on separate lines of the returns block, so you can see which channel is doing the work in every sensitivity case. See the sheet preview if you want to verify the structure first.

The Hilton and Toys "R" Us cases sit at opposite ends of the spectrum: one deal returned $14B entirely through the exit, the other returned zero at the LP level while paying sponsors through channels 1 and 2. Both are labeled "PE deals" in industry summaries. Only one of them was a return of capital in the sense the LP originally meant it.

---

Sources:

1. Hilton S-1 Filing (Dec 2013), SEC EDGAR — entry multiple, sponsor equity, and IPO details
2. Bloomberg, "How Blackstone Made One of the Greatest Deals in PE History," May 2018 — total proceeds and exit sell-down history
3. Toys "R" Us Chapter 11 Petition (S.D. NY 2017), Prime Clerk docket — original capital structure and debt load
4. House Financial Services Committee Hearing (Feb 2019), "The Toys R Us Bankruptcy and the Aftermath" — interest expense, EBITDA, and layoff figures
5. Bloomberg, "How Vulture Capitalists Ate Toys R Us," March 2018 — sponsor fees and dividend recap history