Private Equity

1. Anatomy of a leveraged buyout

Learning goal: Describe who puts money in, who earns what, and which three levers a buyout sponsor pulls.

A leveraged buyout (LBO) is the purchase of a company using a relatively small portion of equity and a relatively large portion of outside debt. Kaplan and Strömberg (2009) describe the typical target as a mature firm where the private equity firm takes majority control. That separates buyouts from venture capital, which usually backs young companies and does not take control. Their paper uses the terms private equity and leveraged buyout interchangeably, and this module follows that usage.

Money comes from limited partners (LPs), who commit capital to a closed-end fund run by a general partner (GP). The GP is paid in three ways in the standard description: an annual management fee on committed capital, a profit share called carried interest that is almost always 20 percent, and sometimes deal or monitoring fees charged to portfolio companies. Metrick and Yasuda (2010) found a 20 percent carry level in all 144 buyout funds in their sample.

Kaplan and Strömberg sort the changes a sponsor makes into three groups: financial engineering (heavy debt), governance engineering (board control and active oversight) and operational engineering (industry and operating expertise). The groups overlap in practice. Treat them as a checklist for asking where a return came from, not as three separate businesses.

Scope. This module covers US-style buyout funds. Tax and regulation refer to US federal rules unless a lesson says otherwise. UK, EU, Chinese and other regimes differ and are not covered. The NVCA model documents are written for venture financings, so they are not a buyout fund template.

Worked example

Fund Alder has $500m of LP commitments. In one deal it buys 100 percent of a company for $400m of enterprise value with $250m of new debt. How much fund equity does the deal use, ignoring fees?

1. The purchase price is $400m.
2. Debt covers $250m of it.
3. Fund equity covers the rest: 400 − 250 = $150m.
4. That is 150/400 = 37.5 percent equity, so the deal is about 62.5 percent debt-financed.

Mental model: A buyout is a controlled purchase of a mature company with a large debt share, funded by LP capital and paid for through fees and carry.

Common trap: Calling every PE deal an LBO. Venture and growth deals usually lack control and heavy debt.

2. Sources and uses: pricing the deal

Learning goal: Build a sources-and-uses table and find the equity cheque from price, fees and debt.

Every LBO model starts with one identity: total uses of cash equal total sources. Uses are the purchase of the business (enterprise value) plus transaction fees. Sources are new debt plus sponsor equity. The equity cheque is the plug: uses minus debt. Everything the fund later earns is measured against that number.

Enterprise value is set by an entry multiple applied to a measure of earnings, usually EBITDA. This lesson uses a fictional company, Harbor Components, with $40m of EBITDA bought at 9.0x, so enterprise value is $360m. Fees of 3 percent of enterprise value are $10.8m. Debt of 5.5x EBITDA is $220m. Uses are $370.8m and the equity cheque is $150.8m. These are teaching numbers and are not drawn from any real deal.

Debt sizing rests on credit-market conditions as well as the deal. In the Gompers, Kaplan and Mukharlyamov (2016) survey of 79 PE investors, capital structure choice was based about equally on trade-off considerations and market-timing considerations. A model that treats leverage as a fixed ratio hides that choice.

Worked example

Harbor Components has EBITDA of $40m. Entry is 9.0x, fees are 3 percent of enterprise value and debt is 5.5x EBITDA. Find the equity cheque.

1. Enterprise value = 40 × 9.0 = $360m.
2. Fees = 0.03 × 360 = $10.8m, so total uses = $370.8m.
3. Debt = 5.5 × 40 = $220m.
4. Equity = 370.8 − 220 = $150.8m.

Mental model: Uses = price + fees. Equity = uses − debt. A fixed debt amount means every extra dollar of price comes out of equity.

Common trap: Forgetting fees. They raise the equity cheque and drag on returns from day one.

3. Debt: coverage, tax shield and the interest cap

Learning goal: Test whether a debt package can be serviced and estimate how much interest is deductible under the US 30 percent rule.

Debt has two sides. Interest is a fixed claim that must be paid regardless of how the business does, so too much of it raises the chance of costly financial distress. In many countries, including the US, interest is also deductible, which can raise firm value through a tax shield. Kaplan and Strömberg note that the shield is hard to value because it depends on assumptions about the tax advantage of debt and how long the debt stays.

Two quick credit tests are debt / EBITDA (here 220/40 = 5.5x) and EBITDA / interest. At a 9 percent rate, Harbor Components pays $19.8m of interest, so EBITDA covers interest 40/19.8 = 2.02 times before any tax, capex or working capital. Thin cover is why the cash left over for repayment in lesson 4 can shrink fast in a bad year.

US federal tax scope. The IRS Q&A on section 163(j) says deductible business interest cannot exceed the sum of business interest income, 30 percent of adjusted taxable income (ATI) and floor plan financing interest. Its list of ATI adjustments treats depreciation and amortization differently for taxable years beginning after December 31, 2024, so ATI is closer to an EBITDA-style base than in 2022 to 2024. For taxable years beginning after December 31, 2025, the IRS also says a U.S. shareholder's CFC income inclusions under sections 951(a), 951A(a) and 78 are excluded from ATI. The IRS Q&A is current as of August 19, 2026. The limit is applied at the partnership level for partnerships. A gross-receipts exemption exists for smaller businesses ($32 million for 2026 per the IRS). This lesson assumes ATI equals EBITDA, which is a teaching simplification (in the IRS description ATI starts from taxable income and adds back items such as business interest expense, NOL and section 199A deductions, and depreciation, amortization and depletion for the years noted above) and not tax advice. Disallowed interest is carried forward, not lost.

Worked example

Harbor Components has $220m of debt at 9 percent and ATI of $40m (assumed equal to EBITDA). How much interest is disallowed under a 30 percent cap?

1. Interest = 0.09 × 220 = $19.8m.
2. Cap = 0.30 × 40 = $12.0m.
3. Disallowed = 19.8 − 12.0 = $7.8m, assuming no interest income or floor plan interest.
4. The $7.8m is carried forward and can be used in later years if there is headroom.

Mental model: Debt helps through tax and discipline and hurts through fixed payments. In the US the deduction is capped by a share of ATI.

Common trap: Treating the cap as lost money. Disallowed interest is carried forward.

4. The operating model and debt paydown

Learning goal: Roll a debt balance forward with a cash sweep and see how EBITDA growth and interest interact.

The operating model turns an earnings forecast into debt repayment. Start with EBITDA, subtract cash leakage (taxes, capex and working capital in this toy model), then subtract interest on the opening debt balance. Whatever remains sweeps to repay debt. Interest uses the opening balance so there is no circular formula.

In Harbor Components, EBITDA grows from $40m to $60m over five years and leakage is 35 percent of EBITDA. Year 1 cash before interest is 44 × 0.65 = $28.6m. Interest at 9 percent on $220m is $19.8m, so only $8.8m repays debt. By year 5 the balance is $138.9m. Interest falls each year because the balance falls, so repayment speeds up even where EBITDA growth is steady.

Model limits. The 35 percent leakage rate, the 9 percent rate and the EBITDA path are teaching assumptions. A real model has tranches, amortization schedules, covenants, minimum cash and a tax computation that depends on the interest cap from lesson 3. The point here is the mechanics: debt paydown is a result of operations, not an input.

Worked example

Harbor Components starts with $220m of debt at 9 percent. Year 1 EBITDA is $44m and leakage is 35 percent. How much debt is left after year 1?

1. Cash before interest = 44 × (1 − 0.35) = $28.6m.
2. Interest = 0.09 × 220 = $19.8m.
3. Cash sweep = 28.6 − 19.8 = $8.8m.
4. Closing debt = 220 − 8.8 = $211.2m.

Mental model: Paydown = EBITDA after leakage minus interest. Lower debt lowers interest, which speeds paydown.

Common trap: Using the original debt for every year of interest. Interest follows the opening balance each year.

5. MOIC and IRR

Learning goal: Compute a multiple of money and an annualized return, and explain why they can disagree.

The multiple of invested capital (MOIC) is equity returned divided by equity invested. In Harbor Components, exit enterprise value is 60 × 9.0 = $540m. After repaying the remaining $138.9m of debt, equity is $401.1m, so MOIC is 401.1 / 150.8 = 2.66x. The internal rate of return (IRR) is the discount rate that sets the net present value of the cash flows to zero. With one investment and one payout n years later, it simplifies to MOIC^(1/n) − 1, which is 21.6 percent for a five-year hold.

Gompers, Kaplan and Mukharlyamov report that the PE investors they surveyed rely primarily on IRR and multiples to evaluate investments. The two tell different stories: MOIC ignores timing and IRR ignores scale. A 2.0x return in three years and a 2.0x return in ten years have the same MOIC and very different IRRs. The ILPA performance template definitions refer to IRR and MOIC computed both with and without the impact of fund-level subscription facilities (lesson 10), which shows that cash timing is treated as material.

Use the single-cash-flow shortcut only when there is one cash in and one cash out. With interim dividends, solve for IRR numerically. The practice problem in this lesson stays in the shortcut case.

Worked example

Harbor Components invests $150.8m of equity and returns $401.1m after five years. Find MOIC and IRR.

1. MOIC = 401.1 / 150.8 = 2.66x.
2. Raise to the power 1/5: 2.66^0.2 = 1.216.
3. IRR = 1.216 − 1 = 21.6 percent.
4. Check: 150.8 × 1.216^5 = about 401.1.

Mental model: MOIC measures scale of gain, IRR measures speed. With one in and one out, IRR = MOIC^(1/n) − 1.

Common trap: Quoting IRR with no hold period or MOIC with no timing.

6. Where the return comes from: a value bridge

Learning goal: Split equity gain into EBITDA growth, multiple change and debt paydown, net of fees.

A value bridge splits equity gain into pieces. Equity gain equals the change in enterprise value, plus debt repaid, minus fees paid at entry. The change in enterprise value is split into EBITDA growth valued at the entry multiple, and multiple change valued on exit EBITDA. This ordering is one convention. Another ordering would assign the cross term differently, so state the convention when you compare deals.

For Harbor Components, equity rises from $150.8m to $401.1m, a gain of $250.3m. EBITDA growth contributes (60 − 40) × 9.0 = $180.0m. The multiple is unchanged at 9.0x, so multiple change adds nothing. Debt paydown adds 220 − 138.9 = $81.1m. Fees subtract $10.8m. The sum is 250.3, matching the gain.

Kaplan and Strömberg describe empirical evidence that operating performance after buyouts is largely positive, though selection bias and the age of the sample limit conclusions. A bridge cannot tell you whether EBITDA growth was caused by the sponsor or by the market. It only describes arithmetic.

Worked example

Harbor Components: EBITDA 40 to 60, multiple 9.0x both ends, debt 220 to 138.9, entry fees 10.8. Split the equity gain.

1. EBITDA growth = (60 − 40) × 9.0 = 180.0.
2. Multiple change = (9.0 − 9.0) × 60 = 0.
3. Debt paydown = 220 − 138.9 = 81.1.
4. Total = 180.0 + 0 + 81.1 − 10.8 = 250.3, which matches 401.1 − 150.8.

Mental model: Equity gain = EV change + debt repaid − entry fees. Bridges describe arithmetic, not causes.

Common trap: Counting multiple expansion as skill. It depends on the market at exit.

7. Exits, hold periods and timing

Learning goal: Name the main exit routes and compute how hold period changes IRR for a fixed multiple.

Most buyout funds have a limited contractual life, so exits are part of the design. Kaplan and Strömberg analyse a CapitalIQ sample through 2007. Among exited deals, a sale to a strategic (non-financial) buyer was the most common route at 38 percent, a secondary buyout to another PE fund was 24 percent and an IPO was 14 percent. These shares describe that sample and period and should not be treated as current.

The same multiple gives very different IRRs depending on when it arrives. For Harbor Components at 2.66x, IRR is 38.6 percent after three years, 27.7 percent after four, 21.6 percent after five and 15.0 percent after seven. That is why a sponsor with a good multiple on offer may prefer an earlier exit, and why LPs watch hold periods.

Exit choice also depends on price and market windows. A strategic buyer may pay for synergies, a secondary buyer gives liquidity but sells to another leveraged owner, and an IPO needs the public market to be open. The fund's end date matters too: its contractual life and extension terms sit in the fund documents, and SEC examiners have flagged advisers that extended private equity fund terms without required approvals or without following the liquidation provisions.

Worked example

A deal earns 2.66x. What IRR does a three-year exit give, versus a seven-year exit?

1. Three years: 2.66^(1/3) = 1.386, so IRR is about 38.6 percent.
2. Seven years: 2.66^(1/7) = 1.149, so IRR is about 15.0 percent.
3. The multiple did not change, only the timing.
4. This is why IRR is not a substitute for MOIC.

Mental model: Exit routes differ in price and speed. For a fixed multiple, a longer hold means a lower IRR.

Common trap: Reading historical exit shares as current market structure.

8. Fund economics: fees, commitments and net returns

Learning goal: Compute lifetime management fees, investable capital and the gap between gross and net.

A fund is a pool of commitments, and not all of it reaches companies. Metrick and Yasuda note that a fund charging 2 percent annual fees on committed capital for ten years would spend 20 percent of commitments on fees, leaving 80 percent as investment capital. In their sample of 144 buyout funds, 84.0 percent changed the fee basis after the investment period, 59 funds had an initial fee of exactly 2 percent and 74 were below 2 percent.

Kaplan and Strömberg give an illustration with a $2 billion fund and a 2 percent fee: $40 million a year during a five-year investment period, declining as investments are sold, for management fees of $300 to $400 million over the fund life. In this lesson a smaller toy fund of $500m charges 2 percent of commitments for five years ($50m) and then 2 percent of the invested cost still held ($300m, $250m, $200m, $150m, $100m), which is $20m. Lifetime fees are $70m and $430m is available to invest.

The ILPA performance template defines fund-level net TVPI as the fee-paying investors' share of net asset value plus their distributions, divided by the capital they contributed. Net measures come after fees and expenses, so gross deal returns are higher. Always ask whether a quoted multiple is gross or net and whether it is deal-level or fund-level.

Worked example

A $500m fund charges 2 percent of commitments for five years, then 2 percent of invested cost of $300m, $250m, $200m, $150m and $100m in years 6 to 10. What are lifetime fees?

1. Years 1 to 5: 0.02 × 500 × 5 = $50m.
2. Years 6 to 10: 0.02 × (300 + 250 + 200 + 150 + 100) = 0.02 × 1,000 = $20m.
3. Lifetime fees = $70m, or 14 percent of commitments.
4. Capital available to invest = 500 − 70 = $430m.

Mental model: Investable capital = commitments − lifetime fees. Net returns are always below gross.

Common trap: Comparing a deal-level gross multiple with a fund-level net multiple.

9. The waterfall: hurdle, catch-up, carry and clawback

Learning goal: Run a whole-fund distribution waterfall with a hurdle and 100 percent catch-up, and spot a clawback.

The distribution waterfall sets who gets cash first. A common whole-fund structure has four steps: LPs get back contributed capital, LPs get a preferred return (hurdle), the GP receives a catch-up until it has 20 percent of total profit, and the rest is split 80 percent to LPs and 20 percent to the GP. Metrick and Yasuda report that 92.4 percent of buyout funds in their sample had a hurdle return, versus 44.7 percent of venture funds, and the modal hurdle was 8 percent. Their worked example shows that with a full catch-up and enough profit, the GP still ends up with its full carry share of profits.

Take a toy fund with $200m called on day one, an 8 percent hurdle and 100 percent catch-up. After three years the hurdle amount is 200 × 1.08^3 = $251.94m. If the fund distributes $330m at that point, LPs first get $251.94m. Profit so far is $51.94m, so the GP catch-up is 0.2 / 0.8 × 51.94 = $12.99m. The remaining $65.07m is split 80/20: $52.06m to LPs and $13.01m to the GP. The GP total is $26.0m, exactly 20 percent of $130m profit.

The order of tiers and the timing of distributions matter. In a deal-by-deal waterfall, early winners can pay carry before later losers arrive. Where carry has been over-distributed, a clawback requires the GP to return it. ILPA Principles 3.0 call an "all capital back" waterfall the best approach to minimise clawback liability and say that deal-by-deal distributions should reflect return of all realised cost, partial write-offs and fees and expenses to date. ILPA guidance also says clawback amounts should be gross of taxes and paid back no later than two years after recognition, and that accrued carry should be held in escrow. ILPA Principles are non-binding best-practice guidance, not law, and each fund's limited partnership agreement governs.

Worked example

A $200m fund (all called on day one), 8 percent hurdle, 100 percent catch-up, 20 percent carry, distributes $330m at year 3. How much carry does the GP receive?

1. Hurdle amount = 200 × 1.08^3 = $251.94m to LPs.
2. Preferred profit = 51.94, so the catch-up is 51.94 × 0.2/0.8 = $12.99m to the GP.
3. Remainder = 330 − 251.94 − 12.99 = $65.07m, split 80/20: LP 52.06, GP 13.01.
4. GP total = 12.99 + 13.01 = $26.0m, which is 20 percent of profit of 130.

Mental model: Capital back, preferred return, catch-up, then 80/20. Check whether the catch-up is complete.

Common trap: Applying 20 percent to every distribution and skipping the hurdle and catch-up.

10. Measuring performance and the rules around it

Learning goal: Compute a public market equivalent (PME), judge performance claims and state what US regulation currently requires.

IRR and MOIC do not say whether a fund beat the stock market. The Kaplan-Schoar PME uses an index (the S&P 500 in the original paper) to value each contribution and distribution on a common date and compares the totals. In the form used here, PME is the future value of distributions divided by the future value of contributions, both grown at the index return. A PME above 1 means the fund beat the index. Kaplan and Schoar (2005) found that average fund returns net of fees roughly equalled the S&P 500, with large dispersion and strong persistence across funds raised by the same firm.

Harris, Jenkinson and Kaplan (2014) studied nearly 1,400 US buyout and venture funds from Burgiss. They report that buyout funds exceeded public markets for most vintages since 1984, outperforming the S&P 500 by 20 to 27 percent over the life of the fund and more than 3 percent per year. Datasets, vintages and benchmark choices matter, so read these as findings of those papers, not guarantees for any fund. Gompers, Kaplan and Mukharlyamov report that LPs they surveyed focus mostly on absolute performance, even though equity-based benchmarks such as PME suit a risky equity investment.

Jurisdiction and standards. This section is US-specific. In August 2023 the SEC adopted new private fund adviser rules covering quarterly statements, audits, restricted activities and preferential treatment. On June 5, 2024 the Fifth Circuit vacated those rules in National Association of Private Fund Managers v. SEC, and an SEC announcement dated October 31, 2024 lists the newly adopted rules and related amendments as vacated. That SEC announcement page was last updated Nov. 1, 2024, and as of October 4, 2026 the sources checked (sec.gov, the Fifth Circuit opinion, the Federal Register) showed no SEC re-adoption of those rules. The SEC and CFTC have separately kept working on Form PF, the confidential reporting form for private fund advisers: on April 20, 2026 they proposed amendments that would raise the filing threshold from $150 million to $1 billion of private fund assets, and on September 3, 2026 they extended the compliance date for the 2024 Form PF amendments from October 1, 2026 to July 1, 2027. Form PF is a separate regime from the vacated 2023 rules. Rule status can change, so check the current position before relying on it. Earlier SEC examination findings still show what examiners look for: failures to follow fee calculation practices described in fund disclosures, incomplete description of recycling, and misleading track record presentation. ILPA reporting and performance templates are voluntary industry standards, not legal requirements. Check the current status of any rule before relying on it.

Worked example

A fund calls $100m at time 0 and distributes $180m at year 4. The index level is 1.00 at time 0 and 1.50 at year 4. What is the PME?

1. Contributions grown with the index to year 4: 100 × 1.50/1.00 = $150m.
2. Distributions at year 4 are already at year 4: $180m.
3. PME = 180 / 150 = 1.20.
4. Above 1 means the fund beat the index. Discounting both sides to time 0 gives the same ratio.

Mental model: PME above 1 beats the index. The SEC 2023 private fund rules were vacated in 2024 and, in sources checked on Oct. 4, 2026, not re-adopted. Check dataset, benchmark and gross vs net.

Common trap: Quoting an IRR without a benchmark, or assuming vacated rules still apply.