Hedge-Fund Strategies & Fund Structures

Ring: Finance-Plus (institutional/CFA breadth) · Branch: Alternatives (AL) · Node AL1.01 · ~16 hours · Mastery gate ≥ 85%

A hedge fund is not a product; it is a legal and fee wrapper around a strategy that ordinary long-only funds are not built to run. Take away the mystique and what is left is simple: a manager who can borrow (leverage), sell what they do not own (shorting), and use derivatives, organized inside a fund structure designed to charge for skill rather than for merely holding the market, and wrapped in redemption terms built to match how liquid the underlying bets actually are. Everything in this module (the five strategy families, the 2-and-20 fee, the master-feeder plumbing, the gates and lock-ups) is the machinery that makes that one sentence operational. This Ring is additive institutional/CFA breadth, it presents the toolkit and the plumbing alongside, not in place of, the program's process-over-P&L, concentration-over-diversification philosophy: nothing here argues you should own twenty uncorrelated sleeves instead of a handful of businesses you understand deeply; it argues that when you evaluate an allocator, a fund-of-funds pitch, or your own eventual capital-allocation decisions, you should be able to read a term sheet and a strategy description the way the other side of the table does.

By the end you will be able to place any hedge fund into one of five strategy buckets and state, from first principles, where its return comes from and what specifically can break it; decompose a 2-and-20 fee (management fee, hurdle, high-water mark) year by year through a real drawdown, in Python, so that no fee statement can surprise you; and read a fund's structural terms (master-feeder, side pockets, gates, lock-ups) as risk-management engineering rather than fine print. You will also carry, explicitly, the reason this corner of finance is the single hardest place in this program to verify anything an AI tool tells you, and the discipline that fixes it.

Learning objectives

By the end you can:

1. Classify any hedge fund strategy into one of five families (equity long/short, event-driven, global macro, relative value, managed futures) and state each family's return driver and its characteristic risk in one sentence apiece.
2. Compute and interpret gross and net exposure for a long/short book, and explain, with a worked pair trade, why a market-neutral position's P&L depends on a return differential, not on market direction.
3. Price a merger-arbitrage deal spread (the gross annualized spread and the expected-value return once deal-break risk is priced in) and explain the strategy's asymmetric payoff shape.
4. Build a full 2-and-20 fee waterfall in Python: management fee, a hard hurdle (no catch-up) and a soft hurdle with 100% catch-up, and a high-water mark, carried through a multi-year path that includes a drawdown, and state precisely why the high-water mark can block an incentive fee even in a year whose return clears the hurdle.
5. Explain master-feeder structures, side pockets, gates, and lock-ups as engineered answers to specific problems (tax/regulatory segregation, illiquid-asset ring-fencing, run risk, and funding genuinely illiquid strategies), and compute the numeric consequence of each for an investor.
6. State why hedge-fund data is this program's highest AI-over-trust risk (R10) (no public financial statements, self-reported and survivorship-biased performance databases) and name the actual primary sources (Form ADV, SEBI AIF disclosures) that substitute for them.
7. Reference, not re-derive, the mechanics this module borrows: covered interest parity and the carry trade (M7.06) for global macro, the credit-spread and priority-of-claims framework (E11.01) for relative-value credit trades, and the option Greeks (delta/gamma: DV1.02) for convertible arbitrage, so you spend your time on what is actually new here: strategy classification, fees, and fund structure.

Prerequisites & connections

Builds on. This is the Alternatives branch's first node, and it opens once you pass M3.02. The module leans on habits built earlier in the program: Phase 1's financial-statement fluency (a fund's NAV statement is just a very short balance sheet); Phase 3's M3.02 (beta and the CAPM, "net exposure" is nothing but a book's realized beta to the market, expressed in rupee or dollar terms instead of a coefficient) and M3.03 (the cost-of-capital intuition that makes a "hurdle rate" (a minimum return before a manager is paid for skill) feel familiar rather than arbitrary); and Phase 2's discipline of never taking a reported number at face value, which you will need the moment you read a fund's marketed "net return" and ask which investor's high-water mark it was computed against.

Three modules elsewhere in this Ring are referenced here rather than re-taught, exactly as this branch's authoring brief requires: M7.06 (International: FX, Flows & the Dollar System) already builds covered interest parity, the carry trade, and commodities-as-macro-signals from the ground up, this module borrows those mechanics wholesale for the global-macro section and adds only the fund-level lens (sizing, P&L attribution, fee structure). E11.01 (Credit Analysis & the Debt Investor's Lens) already builds the credit-spread decomposition (spread ≈ PD × LGD) and the priority-of-claims waterfall, this module borrows that framework for relative-value credit trades (capital-structure arbitrage, convertible arbitrage's credit component) without re-deriving it. M5.09 (Commodities, Real Estate & REITs) owns the cost-curve and cyclical-earnings lens that a managed-futures/CTA fund's commodity trend signals ultimately trade against; this module uses commodities only as one instrument in a systematic basket, and defers the full asset-class treatment to AL1.02 (Commodities, Real Assets & Infrastructure as Asset Classes), which follows this node. A fourth reference is forward-looking: DV1.02 (Options. Parity, Binomial & BSM + the Greeks) builds delta and gamma properly; this module takes delta as a given hedge ratio and shows only how a relative-value fund monetizes it.

Feeds into. AL1.02 (commodities/real assets, where a CTA's trend signals get their full macro treatment) and AL1.03 (digital assets, where several of this module's fund-structure issues (gates, redemption terms, and unverifiable self-reported performance) resurface in sharper form). It also feeds the portfolio-construction thread in AA1.01/AA1.02 (how much of a portfolio, if any, to allocate to an illiquid, fee-heavy sleeve is a risk-budgeting question, not a strategy question) and PW1.02/PW1.03 (side pockets are hedge funds' answer to the same illiquidity problem that closed-end private-market vehicles solve structurally by simply not offering redemption at all).

Two further nodes own what this one deliberately stops short of. AL1.04 owns operational due diligence on an alternative manager, the non-normality measures (skewness, excess kurtosis and the downside statistics) that a hedge-fund return series demands and a Sharpe ratio hides, and the allocation decision itself: how much of a portfolio should sit here, on what pacing, and against which benchmark. Every fee and structure figure computed above is an input to that decision rather than the decision. FI1.04 owns the convertible bond as an instrument, decomposing it into a straight bond plus an equity option and pricing the embedded optionality on a calibrated tree; Worked Example 7 here hedges a convertible position and takes that decomposition as given rather than deriving it.

The one-sentence version of this module. A hedge fund sells you a strategy, a fee schedule, and a set of promises about when you can get your money back, and the discipline this module installs is to price and verify all three separately, because a fund can be right about the market and still cost you money on the fee schedule, or be profitable and still trap your capital on the redemption terms.

Core teaching content

Event-driven: merger arbitrage and the deal-spread trade

When a company announces it is acquiring another, the target's stock typically jumps toward, but not fully to, the offer price. The gap is the deal spread, and merger arbitrage (risk arbitrage) is the strategy of buying the target at that discount and holding until the deal closes at the full offer price (in a stock-for-stock deal, the arbitrageur also shorts the acquirer in the deal's exchange ratio, hedging out the acquirer's own market risk and isolating the spread). The spread exists because the deal might not close on schedule, or at all: financing can fall through, a regulator can block it on antitrust grounds, a competing bidder can emerge, or shareholders can vote it down. Annualizing the gross spread ((offer − current price) ÷ current price × (12 ÷ months to close)) tells you the return if the deal closes as announced; it says nothing about the probability that it does. The honest number is an expected value across the deal-completing and deal-breaking outcomes, and the gap between the two is usually large, because a broken deal typically sends the target's price not just back to its pre-announcement level but below it (the market now prices in whatever went wrong). Worked Example 4, in US dollars, computes both numbers on a single deal and shows exactly how much smaller the expected-value return is than the naive annualized spread, merger arbitrage's return profile is small and steady when deals close, and occasionally sharply negative when they do not, the same general shape shared by many convergence and carry strategies across this program (M7.06's carry-trade worked example shows the identical asymmetry in a currency context).

Global macro: trading the world's largest, most liquid markets

Global macro funds trade top-down views on growth, inflation, monetary policy, and capital flows using the world's deepest and most liquid instruments: government bonds and interest-rate futures, currencies (spot, forwards, and FX options), commodities, and equity indices. Discretionary macro managers form a view, a central bank is behind the curve, a currency peg is unsustainable, a yield curve is mispricing a coming cut, and size a position with judgment; systematic macro managers encode similar logic (often with a heavier statistical/quantitative bent) into rules that generate and size positions with minimal daily human intervention, blurring into the managed-futures family below. This module does not re-derive the instrument mechanics: M7.06 already builds covered interest parity (the no-arbitrage link between spot, forward, and the two currencies' interest rates), the carry trade's mechanics and its "steady gains, then a sudden large loss" risk shape, and how oil, gold, and copper function as macro signals, read that module for the derivation. What a fund-structure lens adds on top is sizing and attribution: a macro fund does not simply "believe the rupee will hold"; it sizes a specific notional overlay against its NAV and reports the P&L contribution of that one position separately from everything else in the book, exactly as Worked Example 5 (India ₹/US $) demonstrates for an INR/USD carry position.

Managed futures / CTAs: systematic trend-following and volatility targeting

Commodity trading advisors run largely systematic strategies (trend-following above all, though also carry and mean-reversion signals) across a deliberately diversified basket of liquid futures spanning rates, currencies, equity indices, and commodities. The technique that most distinguishes this family from a simple "buy what's going up" heuristic is volatility targeting: rather than sizing every position at the same notional dollar amount, a CTA sizes each position so that it contributes roughly the same amount of risk to the book, using the rule position weight = target portfolio volatility ÷ the asset's own realized volatility. A low-volatility instrument (a short-tenor interest-rate future, say) needs a large notional position to contribute a meaningful risk budget; a high-volatility instrument (crude oil) needs a much smaller one. Worked Example 6 makes this concrete across a three-asset basket and shows why a CTA's summed gross notional exposure routinely runs well above 100% of NAV, not because any single bet is aggressive, but because volatility-scaling low-volatility instruments up to a normal risk contribution mechanically produces large headline notional. The characteristic failure mode follows directly from the technique: trend-following, by construction, loses money in a choppy, range-bound, directionless market, because every signal in the basket is simultaneously whipsawed by the same lack of a sustained move.

Why this corner of the curriculum is the corpus's highest AI-over-trust risk

Say this plainly, because R10 requires it and because the honest reason is structural, not a matter of degree. A listed company files a 10-K or an annual report that an auditor has signed and a regulator has received; you can trace almost any claim about it to a named, dated, public document. A hedge fund files almost nothing of the kind. There is no public income statement, no audited-and-filed balance sheet an outsider can pull, and (critically) performance is self-reported, voluntarily, into commercial databases (HFR, Eurekahedge, and similar), which is why every academic study of those databases finds structural survivorship bias (funds that blow up simply stop reporting and quietly disappear from the sample) and backfill bias (a fund only starts reporting once it has a good track record to show, then backfills its earlier history, so the visible universe is tilted toward winners by construction). Layer on top of that the fact that a fund's actual holdings are confidential, its short positions are never disclosed, and its own investor letters are marketing documents written by the people being evaluated, and you have an asset class where a fluent AI answer to "what did Fund X return last year, and how did it do it?" is, more than almost anywhere else in this program, impossible to distinguish from a fabrication without an independent primary source, because for most of the underlying facts, no independent primary source open to the public exists at all. The Madoff fraud (uncovered in 2008) is the extreme, real-world proof of the stakes: fabricated NAV statements were accepted by feeder funds, auditors, and thousands of sophisticated investors for well over a decade, precisely because the underlying "performance" had no independent primary source that anyone actually checked. The two genuine primary sources that do exist and are worth knowing cold: a US SEC-registered adviser's Form ADV Part 2A brochure (searchable free through the SEC's Investment Adviser Public Disclosure system), which discloses real fee terms, conflicts, and disciplinary history, though rarely net performance, and, for India, a Category III AIF's placement memorandum and SEBI disclosures. Treat every other hedge-fund "fact" an AI tool produces exactly as fluent and exactly as unverified as any other confident model output, and route your verification to one of those two named sources or do not use the figure.

Common mistakes & how experts think differently

1. Treating "hedge fund" as a return profile instead of a legal wrapper. Novices assume all hedge funds behave similarly (leveraged, aggressive, high-return). Experts know the label describes a fee-and-structure wrapper, not a strategy, a market-neutral relative-value fund and a directional global-macro fund can sit inside identical legal terms and have opposite risk profiles; the strategy classification (the five buckets above) is the only thing that tells you what actually drives the return.
2. Reading the annualized merger-arb spread as the expected return. The naive (offer − price)/price × (12/months) calculation assumes the deal closes with certainty. Experts always build the expected-value version (probability-weighted across completion and break scenarios) because deal-break outcomes are typically far worse than the spread being captured is good, and Worked Example 4 shows how large that gap can be.
3. Confusing gross exposure with net exposure, or either with risk. A 200% gross / 0% net book can still be extremely risky (concentrated pair bets, crowded shorts) while a 60% gross / 60% net book carries mostly plain market beta. Neither number alone tells you the risk; experts read them together and ask what specific bets make up the gross.
4. Assuming a management fee only applies "when the fund does well." It does not, it is an asset-based fee, charged in every year including loss years (Practice P16 and Worked Example 1's Year 2 make this explicit), which is exactly why investors scrutinize the management-fee rate on large, low-turnover AUM bases where it can dwarf any realistic performance fee.
5. Believing a return above the hurdle automatically means an incentive fee is owed. It does not, if a high-water mark is also in force and NAV has not yet recovered above it. Worked Example 1's Year 3 (a 10% net-of-fee return, comfortably above a 6% hurdle, and still a zero incentive fee) is built to break this exact assumption.
6. Treating a soft hurdle and a hard hurdle as roughly the same protection. They are not, once the fund's return moves into the catch-up zone: a soft hurdle with 100% catch-up hands the manager the entire marginal gain until the manager has caught up to a full incentive-rate cut of everything, after which the hurdle stops mattering at all, the comparison table in the fee section shows this converges to materially less net return for the investor than a hard hurdle at the same gross performance.
7. Assuming every investor in the same fund earned the same net return. Equalization means an investor's incentive-fee liability tracks their own subscription-date high-water mark, not the fund's. Two investors in the identical book can, correctly and non-fraudulently, receive different net returns for the identical period.
8. Citing a fund's marketed track record as if it were an audited fact. Hedge-fund performance databases are self-reported, with structural survivorship and backfill bias; a track record pulled from a factsheet or, worse, recited fluently by an AI tool with no attached source is a claim, not a verified figure, until traced to Form ADV, a placement memorandum, or (rarely) an audited investor statement you hold directly.
9. Confusing a delta-hedged position with a riskless one. A convertible-arbitrage position hedged at today's delta is protected against small moves; delta itself changes as the stock moves (gamma, DV1.02's territory), and financing costs, borrow costs on the short stock, and credit deterioration in the issuer are all real risks the equity hedge does nothing to touch.
10. Reading a gate or lock-up as evidence the fund is in trouble. Both are disclosed, pre-agreed terms present in the vast majority of funds trading anything less than fully liquid instruments, the actual red flag is a manager who changes redemption terms after the fact, or who imposes a gate that was never in the original offering documents.

Worked examples

WE1: India ₹: the full 2/20 + hurdle + high-water-mark waterfall through a drawdown

Godavari Absolute Return Fund (synthetic), a Category III AIF, launches at NAV ₹1,000.00 per unit. Terms: 2% management fee on beginning-of-year NAV; 20% incentive fee; a hard hurdle of 6% (no catch-up); a high-water mark. Gross annual returns over four years: +20%, −15%, +12%, +15%, a strong year, a drawdown, and a two-year recovery. Order of operations each year: apply the gross return → deduct the management fee → test the hurdle and the high-water mark → charge the incentive fee only if both pass → roll the new high-water mark forward.

Year 1 (+20%). Gross NAV = 1,000 × 1.20 = 1,200.00. Management fee = 2% × 1,000 = 20.00 → NAV after mgmt fee = 1,180.00 (an 18.00% return net of the management fee alone). 18.00% clears the 6% hurdle, and 1,180.00 exceeds the ₹1,000 high-water mark, so the incentive fee applies to the excess above the hurdle: 20% × (1,180.00 − 1,000×1.06) = 20% × 120.00 = 24.00. NAV after incentive fee = 1,156.00 → net return 15.60%. New high-water mark = 1,156.00.

Year 2 (−15%). Gross NAV = 1,156.00 × 0.85 = 982.60. Management fee = 2% × 1,156.00 = 23.12 (charged in full, a loss year changes nothing about the management fee) → NAV after mgmt fee = 959.48 (a −17.00% year). Below the hurdle and below the high-water mark of 1,156.00: incentive fee = 0.00. NAV after incentive fee = 959.48 → net return −17.00%. High-water mark stays at 1,156.00.

Year 3 (+12%). Gross NAV = 959.48 × 1.12 = 1,074.62. Management fee = 2% × 959.48 = 19.19 → NAV after mgmt fee = 1,055.43 (a 10.00% return, comfortably above the 6% hurdle). But 1,055.43 is still below the 1,156.00 high-water mark (the fund has not yet earned back Year 2's loss) so despite clearing the hurdle, the incentive fee is 0.00. NAV after incentive fee = 1,055.43 → net return 10.00%. This is the step to sit with: a double-digit return, comfortably above the stated hurdle, and the manager is paid nothing beyond the management fee, because the high-water mark is a separate, independent gate.

Year 4 (+15%). Gross NAV = 1,055.43 × 1.15 = 1,213.74. Management fee = 2% × 1,055.43 = 21.11 → NAV after mgmt fee = 1,192.63 (a 13.00% return). This finally exceeds the 1,156.00 high-water mark, so the incentive fee resumes, computed on that year's own hurdle-adjusted base: 20% × (1,192.63 − 1,055.43×1.06) = 20% × 73.88 = 14.78. NAV after incentive fee = 1,177.86 → net return 11.60%. New high-water mark = 1,177.86.

Read the whole path. Cumulative gross return over the four years is 31.38% (compounding 1.20 × 0.85 × 1.12 × 1.15); cumulative net-of-all-fees return is 17.79%, a cumulative fee drag of roughly 13.6 percentage points, decomposed as ₹83.42 of management fees and ₹38.78 of incentive fees per ₹1,000 of original capital, or about 2.88 percentage points of annualized drag (gross CAGR 7.06% versus net CAGR 4.18%). Notice where the drag concentrates: two full years (2 and 3) paid a management fee and zero incentive fee, precisely because the high-water mark was doing its job.

Extend it: a new investor subscribing at the trough. Suppose a new investor bought into the identical fund at the start of Year 3, at that day's NAV of ₹959.48, which becomes their own personal high-water mark under standard equalization, regardless of the fund's fund-wide mark of ₹1,156.00. Running that investor's own Years 3–4 through the identical waterfall: in Year 3, their NAV after management fee (₹1,055.43, the same fund, the same trade) is above their own personal high-water mark of 959.48, so they owe an incentive fee of ₹7.68, ending Year 3 at ₹1,047.75, while the original, Year-1 investor owed zero incentive fee that same year on the identical portfolio. In Year 4, the new investor owes a further ₹14.67, ending at ₹1,169.29. Two investors, one fund, one trading book, the same two years, and different fees, entirely because of when each one bought in relative to the fund's peak. This is exactly what equalization is built to do: charge each investor only on gains that are genuinely new to them.

WE2: US $: a soft hurdle with 100% catch-up, and the master-feeder it sits inside

Meridian Global Fund (synthetic) is structured as a Cayman master fund with a Delaware feeder (US taxable investors) and a Cayman feeder (non-US and US tax-exempt investors), both subscribing at NAV parity into the master. Terms: 2% management fee, 20% incentive fee, a soft hurdle of 5% with 100% catch-up. The catch-up threshold: T = hurdle ÷ (1 − incentive rate) = 5% ÷ 0.80 = 6.25%.

Year 1 (+15%). NAV $100.00 → gross NAV $115.00. Management fee $2.00 → NAV after mgmt fee $113.00 (a 13.00% return: well above T = 6.25%, meaning full catch-up is achieved this year). Incentive fee = 20% of the entire post-management-fee gain: 20% × ($113.00 − $100.00) = $2.60. NAV after incentive fee = $110.40 → net return 10.40%. New high-water mark = $110.40. Check: the manager's total take (management + incentive) is $4.60 on a $15.00 gross gain, and the incentive fee alone is exactly 20.00% of the $13.00 post-management-fee gain, confirming full catch-up leaves the manager with the same share as if there had never been a hurdle at all.

Year 2 (+7.5%). NAV $110.40 → gross NAV $118.68. Management fee $2.208 → NAV after mgmt fee $116.472 (a 5.50% return, above the 5% hurdle but below the 6.25% catch-up threshold: the fund is inside the partial catch-up zone). Here the manager takes 100% of the entire increment above the hurdle: $116.472 − ($110.40 × 1.05) = $116.472 − $115.92 = $0.552. NAV after incentive fee = $115.92 → net return exactly 5.00%, precisely the hurdle rate, to the decimal. That is the catch-up zone's defining, slightly startling feature: inside it, the investor's return is pinned at the hurdle no matter how far above the hurdle the fund's own performance runs, because the manager is entitled to the entire marginal dollar until their catch-up is complete.

The side-by-side that makes the difference concrete. Running the identical hard-hurdle (WE1's rule, 5% hurdle here for comparability) against this soft-hurdle-with-catch-up rule, from par, at several gross-return levels:

Below the hurdle, and even at a gross return whose post-management-fee return still sits under the hurdle (6.5% gross nets to 4.5% after the 2% management fee, under the 5% hurdle either way), the two structures are identical: zero incentive fee, because a hurdle test only ever begins once the return clears it. The structures diverge only once performance runs past the hurdle: at 9% and 13% gross, the hard hurdle consistently leaves the investor with materially more net return, because the soft/catch-up structure's 100%-to-the-manager zone extracts a bigger bite before settling into the same 20% split both structures share at the top end. Neither structure is "wrong", they are different, fully disclosed bargains, but an investor who cannot tell the two apart from a term sheet is agreeing to materially different economics without knowing it.

WE3: Equity long/short: exposure and a market-neutral pair (India ₹)

Fully computed in the Core Content section above: Sahyadri Long/Short Fund (synthetic), AUM ₹500 cr, long ₹350 cr, short ₹150 cr → gross exposure 100.00%, net exposure 40.00%. The ₹10 cr / ₹10 cr Konkan Bank–Malabar Bank (synthetic) pair trade nets +₹0.50 cr (+5.00% of the pair's capital) in both the market-up scenario (Konkan +8%, Malabar +3%) and the market-down scenario (Konkan −2%, Malabar −7%), the arithmetic (long P&L = capital × long return; short P&L = −capital × short return; net = sum of both) produces the identical result in each case because what actually varies is the five-percentage-point gap between the two names, not the market's direction, which is precisely the point a "market-neutral" label is claiming.

WE4: Merger arbitrage: gross spread versus the deal-break-adjusted expected value (US $)

Atlas Industrial Corp (synthetic) announces a cash acquisition of Meridian Robotics Inc. (synthetic) at $45.00/share, with the deal expected to close in 6 months. Meridian's stock, pre-announcement in the high $30s, now trades at $43.50, the deal spread.

Step 1, the naive gross spread. Spread = $45.00 − $43.50 = $1.50. Simple 6-month return = $1.50 ÷ $43.50 = 3.4483%. Annualized = 3.4483% × (12 ÷ 6) = 6.8966% (≈ 6.90%), the return if the deal closes exactly as announced, with certainty.

Step 2: price the deal-break risk. Assume the market is pricing a 90% probability of completion; if it breaks, assume Meridian's stock reverts to $34.00 (below its pre-announcement level, since a break usually carries bad news about why). Expected value: EV = 0.90 × $45.00 + 0.10 × $34.00 = $40.50 + $3.40 = $43.90. EV gain over current price = $43.90 − $43.50 = $0.40. EV period return = $0.40 ÷ $43.50 = 0.9195%; EV annualized ≈ 1.8391% (≈ 1.84%), roughly a quarter of the naive gross-spread figure.

Step 3, see the asymmetry. If the deal completes: +$1.50 (+3.45%) over six months. If it breaks: $34.00 − $43.50 = −$9.50 (−21.84%). A small, likely gain against a materially larger, less-likely loss is merger arbitrage's whole risk signature, and it is exactly why the naive annualized spread, quoted alone, understates the strategy's true risk by ignoring precisely the scenario that defines it.

WE5: Global macro: an INR/USD carry trade, sized and attributed at the fund level (India ₹ / US $)

Nilgiri Global Macro Fund (synthetic), India-domiciled, expresses a carry view: borrow in US dollars at 5.00%, hold rupee-denominated short-tenor paper at 7.00%, this module borrows the covered-interest-parity/no-arbitrage-forward mechanics behind this trade entirely from M7.06 and adds only the fund-level sizing and attribution below. Carry = 7.00% − 5.00% = 2.00% per annum, before any currency move.

Three FX scenarios, unhedged: if INR is flat against USD over the holding period, the net return is the full carry, +2.00%. If INR depreciates a mild 1%, net return is 2.00% − 1.00% = +1.00%. If INR depreciates a sharper 3% (larger than the rate differential), net return flips to 2.00% − 3.00% = −1.00%, the trade loses money despite a positive, unchanged carry, because an adverse currency move larger than the carry simply overwhelms it.

Fund-level P&L attribution. Nilgiri sizes this position at 25% of NAV notional via an FX forward overlay (rather than allocating the fund's entire book to one currency view). Contribution to fund NAV in each scenario: flat FX → 25% × 2.00% = +0.500%; mild depreciation → 25% × 1.00% = +0.250%; sharp depreciation → 25% × (−1.00%) = −0.250%. This is the fund-structure lens this module adds on top of M7.06's mechanics: a macro fund does not report "we were long the carry trade"; it reports a specific, sized contribution to NAV from that one position, alongside every other position in the book, exactly like the pair trade in WE3 or the futures basket in WE6.

WE6: Managed futures: volatility-targeted position sizing across a basket (US $)

Cascade Trend Partners (synthetic), a systematic CTA, targets 10% annualized portfolio volatility and sizes each position in a hypothetical trending quarter by weight = target vol ÷ the asset's own realized vol:

Each weight is simply 10% ÷ realized vol (crude: 10/30 = 33.33%; S&P: 10/16 = 62.50%; Treasury: 10/7 = 142.86%); each contribution is signal × weight × period return, with the short Treasury position's negative asset return (prices fell as yields rose) turning into a positive contribution because the position profits from the decline. Total basket contribution this quarter: +11.27% of NAV, and (the point worth pausing on) the summed gross notional exposure across just three positions is 238.7% of NAV, entirely because the low-volatility Treasury future needed a very large notional (142.86% of NAV on its own) to contribute a normal-sized slice of risk. A headline "238% gross exposure" sounds aggressive; volatility targeting shows it is simply the mechanical consequence of equalizing risk contribution across instruments of very different natural volatility, not a statement about conviction.

WE7: Relative value: a convertible-bond delta-neutral hedge (US $)

Silverline Technologies (synthetic) has a convertible bond outstanding: par $1,000, conversion ratio 20 shares, currently trading at $1,050.00 (105.0% of par) while the common stock trades at $48.00. Parity value (what the bond would be worth if converted right now) = 20 × $48.00 = $960.00; the bond trades at a $90.00 premium over parity (9.38%), reflecting the embedded option's remaining time value plus the bond floor. The option's delta is 0.60 (given, per DV1.02's Greeks, not re-derived here).

The hedge. Shares to short per bond = delta × conversion ratio = 0.60 × 20 = 12.0 shares.

The stock moves from $48.00 to $52.00 (+$4.00, +8.33%). Using the delta approximation: bond P&L ≈ delta × Δstock × conversion ratio = 0.60 × $4.00 × 20 = +$48.00. Short-stock P&L = −12.0 shares × $4.00 = −$48.00. Net hedged P&L = $0.00, by construction, since the short position was sized to be exactly delta-offsetting for a small move. An unhedged holder of the same bond would have simply pocketed the full +$48.00, carrying the stock's entire equity beta. The point of the trade is exactly this contrast: the arbitrageur gave up that $48.00 of directional upside (and, symmetrically, would have been protected from an equivalent downside move) on purpose, in order to isolate whatever else the position is actually harvesting, the credit/carry spread embedded in the bond and the gamma-trading profit from continuously rebalancing the hedge as delta itself shifts with the stock, which is precisely where DV1.02's fuller options treatment picks up.

Practice set

Work each problem before reading its solution. P1–P6 are guided, P7–P13 independent, P14–P16 comprehensive/timed. All fund names are synthetic; all figures are computed exactly as shown: reproduce them.

P1 (guided: management fee only). A fund manages ₹800 cr of AUM and charges a 2% annual management fee. Compute the annual fee and the monthly accrual.

Solution. Annual fee = 2% × ₹800 cr = ₹16.00 cr. Monthly accrual = ₹16.00 cr ÷ 12 = ₹1.3333 cr. The management fee is charged on AUM regardless of that year's performance, it is not conditional on any return being earned at all.

P2 (guided: hard hurdle test). NAV starts at $100.00, prior high-water mark $100.00. Gross return for the year is 9%. Management fee 2%, hard hurdle 5%, incentive fee 20%. Compute the incentive fee and net return.

Solution. Gross NAV = $109.00. Management fee = 2% × 100 = $2.00 → NAV after mgmt fee = $107.00 (a 7.00% return). 7.00% clears the 5% hurdle and $107.00 clears the $100 high-water mark, so incentive fee = 20% × ($107.00 − $100×1.05) = 20% × $2.00 = $0.40. NAV after incentive fee = $106.60 → net return 6.60%.

P3 (guided, high-water mark blocks a fee despite clearing the hurdle). NAV $100.00, high-water mark $100.00, Year 1 gross return +20%; Year 2 gross return −10%. Same fee terms as P2 (2% mgmt, 5% hurdle, 20% incentive, hard hurdle). Walk both years.

Solution. Year 1: gross NAV $120.00; mgmt fee $2.00 → NAV after mgmt fee $118.00 (18.00% return, clears hurdle and HWM) → incentive fee 20% × (118.00 − 105.00) = $2.60 → NAV after incentive fee $115.40 → net return 15.40%; new HWM $115.40. Year 2: gross NAV = 115.40 × 0.90 = $103.86; mgmt fee 2% × 115.40 = $2.308 → NAV after mgmt fee $101.552 (a −12.00% year), below both the hurdle and the $115.40 high-water mark, so incentive fee = $0.00 regardless. Net return −12.00%; HWM stays at $115.40. (This is a smaller, two-year rehearsal of Worked Example 1's central lesson.)

P4 (independent, full three-year waterfall, a different fund). NAV ₹250 cr, high-water mark ₹250 cr; 2% management fee, 4% hard hurdle, 20% incentive fee. Gross returns: Year 1 +10%, Year 2 +18%, Year 3 −20%. Compute each year fully.

Solution. Year 1: gross NAV 275.00; mgmt fee 5.00 → NAV after mgmt fee 270.00 (8.00% return, clears 4% hurdle and the HWM) → incentive fee 20% × (270.00 − 250×1.04) = 20% × 10.00 = 2.00 → NAV after incentive fee 268.00 → net return 7.20%; new HWM 268.00. Year 2: gross NAV 268.00 × 1.18 = 316.24; mgmt fee 5.36 → NAV after mgmt fee 310.88 (16.00% return) → incentive fee 20% × (310.88 − 268×1.04) = 20% × 32.16 = 6.432 → NAV after incentive fee 304.448 → net return 13.60%; new HWM 304.448. Year 3: gross NAV 304.448 × 0.80 = 243.5584; mgmt fee 6.089 → NAV after mgmt fee 237.469 (a −22.00% year), below hurdle and below the 304.448 HWM, so incentive fee 0.00; net return −22.00%; HWM unchanged at 304.448.

P5 (independent: exposure calc). A book has AUM ₹700 cr, long positions ₹620 cr, short positions ₹260 cr. Compute gross and net exposure.

Solution. Gross = (620 + 260) ÷ 700 = 125.71%. Net = (620 − 260) ÷ 700 = 51.43%. This is a fund running substantial leverage (125.71% gross) that is still, on net, more than half again long the market at over 51% net exposure, a book that is far from "market-neutral" despite carrying two-sided positions.

P6 (independent: pair trade P&L). ₹15 cr long one name, ₹15 cr short another. The long returns +6%, the short returns −2% over the holding period. Compute the net P&L in ₹ and as a percentage of the capital deployed to the pair.

Solution. Long P&L = 15 × 6% = ₹0.90 cr. Short P&L = −15 × (−2%) = +₹0.30 cr (the short gains because the shorted name fell). Net P&L = 0.90 + 0.30 = ₹1.20 cr = 1.20 ÷ 15 = 8.00% of the pair's capital, the trade captured the full 8-percentage-point spread between the two names' returns, regardless of which direction either name moved on its own.

P7 (independent: merger arb naive spread). A target trades at $57.60 against a $60.00 cash offer, with the deal expected to close in 4 months. Compute the gross spread, the simple return, and the annualized return.

Solution. Spread = $60.00 − $57.60 = $2.40. Simple return = 2.40 ÷ 57.60 = 4.1667%. Annualized = 4.1667% × (12 ÷ 4) = 12.5000%.

P8 (independent: merger arb with deal-break risk). Same deal as P7 ($60.00 offer, $57.60 current price), but the market assigns an 85% probability of completion and, if it breaks, the stock is assumed to fall to $46.00. Compute the expected-value gain and return, and compare to P7's naive figure.

Solution. EV = 0.85 × 60.00 + 0.15 × 46.00 = 51.00 + 6.90 = $57.90. EV gain = 57.90 − 57.60 = $0.30. EV return = 0.30 ÷ 57.60 = 0.5208%, a small fraction of the naive 4.1667% simple return (or the 12.50% annualized figure) computed in P7, because the 15% break probability combined with a punitive $46.00 break price drags the expected value down sharply. This is the same lesson Worked Example 4 draws, at a different deal's numbers.

P9 (independent: carry trade). Domestic short-tenor rate 6.5%, foreign rate 4.5%, and the domestic currency depreciates 1.5% against the foreign currency over the period. Compute the carry and the net unhedged return.

Solution. Carry = 6.5% − 4.5% = 2.00%. Net return = 2.00% − 1.50% = 0.50%. The FX move ate three-quarters of the carry, leaving a thin but still-positive return.

P10 (independent: volatility-targeted sizing). Target portfolio volatility 12%, an asset's realized volatility 24%, the asset's period return +9%. Compute the position weight and its contribution to NAV.

Solution. Weight = 12% ÷ 24% = 0.5000 (50.00% of NAV). Contribution = 50.00% × 9.00% = 4.50% of NAV.

P11 (independent, convertible delta hedge on a falling stock). A convertible bond has conversion ratio 25 and delta 0.55. The underlying stock falls from $40.00 to $36.00. Compute the shares to short, the bond P&L (delta-approximation), the short-stock P&L, and the net hedged P&L.

Solution. Shares to short = 0.55 × 25 = 13.75. Δstock = 36.00 − 40.00 = −$4.00. Bond P&L ≈ 0.55 × (−4.00) × 25 = −$55.00. Short-stock P&L = −13.75 × (−4.00) = +$55.00. Net hedged P&L = −55.00 + 55.00 = $0.00, the hedge protects on the downside exactly as it did on the upside in Worked Example 7; a short-stock hedge earns money on the way down precisely to offset the convertible bond's loss in value.

P12 (independent: equalization, conceptual with numbers). Investor A subscribed at the fund's inception, NAV $100.00. Investor B subscribed later, at NAV $115.00. By year-end the fund's NAV (net of management fee, before incentive fee) is $126.50. Using a 20% incentive rate and each investor's own subscription price as their personal high-water mark, compute each investor's incentive-fee liability.

Solution. Investor A's incentive base = 126.50 − 100.00 = $26.50 → fee = 20% × 26.50 = $5.30. Investor B's incentive base = 126.50 − 115.00 = $11.50 → fee = 20% × 11.50 = $2.30. Same fund, same NAV, same day, but Investor B, having bought in at a higher NAV, owes less than half of Investor A's fee, because Investor B has "earned" less new gain relative to their own entry point. Neither number is wrong; they are answering different questions (gain since inception vs. gain since B's own subscription).

P13 (independent: gate proration). A fund's NAV is $300 million with a 10% quarterly gate. Redemption requests total $65 million. Compute the gate cap, the proration factor, and what an investor requesting $1,000,000 actually receives this quarter.

Solution. Gate cap = 10% × $300,000,000 = $30,000,000. Proration factor = 30,000,000 ÷ 65,000,000 = 46.15%. The $1,000,000 requester receives $1,000,000 × 46.15% = $461,538.46 now, with the remaining $538,461.54 carried to the next redemption window.

P14 (timed, 10 min: side pocket). A fund's total NAV is ₹90 crore across 9,00,000 units, of which a side-pocketed illiquid position is worth ₹9 crore. Compute NAV per unit, the side-pocket value per unit, the liquid NAV per unit, and what an investor redeeming 5,000 units receives in cash now versus as a stub interest.

Solution. NAV/unit = ₹90,00,00,000 ÷ 9,00,000 = ₹1,000.00. Side-pocket/unit = ₹9,00,00,000 ÷ 9,00,000 = ₹100.00 (exactly 10% of NAV). Liquid NAV/unit = 1,000.00 − 100.00 = ₹900.00. Redeeming 5,000 units: cash now = 5,000 × ₹900.00 = ₹45,00,000; stub interest (paid on eventual realization) = 5,000 × ₹100.00 = ₹10,00,000.

P15 (timed, 10 min: soft lock-up penalty). An investor redeems $3,500,000 while still inside a soft lock-up carrying a 2.5% early-redemption penalty. Compute the penalty and the net proceeds.

Solution. Penalty = 2.5% × $3,500,000 = $87,500. Net proceeds = $3,500,000 − $87,500 = $3,412,500.

P16 (timed, 12 min, comprehensive: fee decomposition across a loss year and a strong recovery). NAV $500.00, high-water mark $500.00; 2% management fee, 5% hard hurdle, 20% incentive fee. Year 1 gross return −8%; Year 2 gross return +22%. Compute both years fully, and state explicitly which fee is charged in the loss year and why.

Solution. Year 1: gross NAV 500 × 0.92 = $460.00; management fee 2% × 500 = $10.00 (charged in full, despite the loss, the management fee is asset-based, not performance-based) → NAV after mgmt fee $450.00 (a −10.00% year), below hurdle and HWM, incentive fee $0.00. Net return −10.00%; HWM stays $500.00. Year 2: gross NAV 450 × 1.22 = $549.00; management fee 2% × 450 = $9.00 → NAV after mgmt fee $540.00 (a 20.00% return, clears the 5% hurdle and the $500.00 HWM) → incentive fee 20% × (540.00 − 450×1.05) = 20% × 67.50 = $13.50 → NAV after incentive fee $526.50 → net return 17.00%; new HWM $526.50. Across the two years, the management fee was charged both times ($10.00 then $9.00); the incentive fee was charged only in Year 2 ($13.50), and only once NAV had cleared both the hurdle and the high-water mark reset by Year 1's loss.

P17 (independent: the second fee layer, priced). An underlying hedge fund returns 14.00% gross and charges 2 and 20 with no hurdle. A fund of funds holding it charges 1 and 10 on what it receives. (a) Compute the investor's net return through both layers. (b) State the total fee load in percentage points and as a share of the gross return. (c) State what the second layer alone cost. (d) Name one thing the second layer buys that the first cannot.

Solution. (a) Layer one: 14.00 − 2.00 = 12.000% after the management fee; performance fee = 20% × 12.000 = 2.400%; the underlying fund passes up 9.600%. Layer two charges on that 9.600%, not on the original 14.00%: 9.600 − 1.000 = 8.600%; performance fee = 10% × 8.600 = 0.860%; investor net 7.740%. (b) Total load = 14.000 − 7.740 = **6.260 percentage points**, or 6.260/14.000 = 44.71% of the gross return. (c) A single-layer investor in the same manager would have kept 9.600%, so the second layer cost 1.860 percentage points. (d) Manager selection and monitoring, operational due diligence on managers the investor cannot inspect directly, access to funds closed to new capital, or a minimum ticket small enough to build a diversified book. Whether any of those is worth 1.86 points a year is the question AL1.04 asks with evidence.

Applied mini-project: build and verify a real fee term sheet

The task (≈3 hours, two parts). This mini-project pairs a from-scratch model you build yourself with a real, primary-sourced document, the discipline this module's R10 warning exists to install.

Part A, build your own fee/HWM/hurdle model on a real return series. Pick a real, publicly available index return series as your stand-in "gross fund return" (the exact instrument does not matter; using a real index return path, rather than an invented one, is the point), the Nifty 50 for an India-domiciled fund or the S&P 500 for a US-domiciled fund, covering a period that includes a genuine drawdown and recovery (2020's COVID crash and rebound, or 2008–09, both work well and are easy to source from any free index-data provider or the exchange's own historical data page). Build, in a spreadsheet or in Python: (1) the year-by-year (or quarter-by-quarter) NAV walk applying a 2% management fee, a hurdle you choose and state (5–8% is typical), and a high-water mark; (2) run it once with a hard hurdle and once with a soft hurdle with 100% catch-up, and report the net-of-fee return difference between the two conventions over your chosen period; (3) identify, explicitly, every period in which the high-water mark blocked an incentive fee that the hurdle test alone would have allowed.

Part B, pull one real fund's actual fee terms from a primary source. Using the SEC's Investment Adviser Public Disclosure (IAPD) site (adviserinfo.sec.gov), find one real, SEC-registered private-fund adviser and open its Form ADV Part 2A brochure; extract the actual disclosed management fee, incentive fee, and any hurdle or high-water-mark language, quoting the exact clause and page/section. If you can also access an Indian Category III AIF's placement memorandum or SEBI disclosure (several fund managers publish summary term sheets), extract the equivalent terms there for comparison. Write two paragraphs: what surprised you about how the real disclosure differs from the "clean" 2/20-with-hurdle template this module teaches (real term sheets are often messier, tiered fee schedules, share classes with different terms, fees that step down at scale), and what you could not verify from public sources alone (almost certainly: actual net performance), naming, explicitly, why that gap exists (R10: no public financials, self-reported performance databases).

Scoring rubric (max 20; ≥16 to pass).

Case Lab: run case CL-AL.1 ("Fund of One: Sizing a Long/Short Sleeve") in the app's Cases tab, size a long/short allocation sleeve within a larger portfolio using the gross/net exposure discipline this module teaches, in both an India and a US context.

Reading & resources

Everything below is an anchor to verify against and learn from, never a source to quote a fund's actual performance from, for that, there is usually no substitute for Form ADV, a placement memorandum, or an audited statement you hold directly.

• SEC Investment Adviser Public Disclosure (IAPD), adviserinfo.sec.gov, the free, public, primary source for any US-registered adviser's Form ADV, including Part 2A fee disclosures. [Free] [Beginner]
• SEC, Form PF overview and the Investment Advisers Act of 1940 / Investment Company Act of 1940 §3(c)(1) and §3(c)(7), read the SEC's own summaries of the exemptions this module relies on. [Free] [Intermediate] (check the current thresholds)
• SEBI (Alternative Investment Funds) Regulations, 2012, and subsequent SEBI circulars on Category III AIF leverage: sebi.gov.in, the primary source for India's closest hedge-fund analogue. [Free] [Intermediate] (as of 2026-07: check the current circular)
• IFSCA (International Financial Services Centres Authority) fund-management regulations, the primary source for GIFT City master-feeder-style structures domiciled in India. [Free] [Advanced]
• Sebastian Mallaby, More Money Than God: the standard narrative history of the hedge-fund industry; strong on strategy evolution and on LTCM. [Paid] [Beginner–Intermediate]
• Roger Lowenstein, When Genius Failed: the LTCM collapse in full, the canonical real-world lesson in relative-value leverage risk this module's Common Mistakes section draws on. [Paid] [Beginner]
• President's Working Group on Financial Markets, Hedge Funds, Leverage, and the Lessons of Long-Term Capital Management (1999), the primary-source government report on LTCM's balance sheet and leverage. [Free] [Advanced]
• CFA Institute curriculum readings on Alternative Investments (Hedge Fund Strategies), the standard institutional-exam-level treatment this module is calibrated against. [Paid] [Intermediate–Advanced]
• Nareit / CAIA Association introductory materials on hedge-fund fee structures and fund terms, free primer content on hurdles, high-water marks, and lock-ups from an industry body. [Free] [Beginner]
• Going deeper (the ILPA (Institutional Limited Partners Association) principles on fund terms and fee transparency) written for private-equity LPs but directly applicable to reading any hurdle/catch-up/high-water-mark clause with a skeptical, institutional eye. [Free] [Advanced]

Teach it back & journal

Feynman prompt. Explain to someone who has only ever bought mutual funds why a hedge fund charging "2 and 20" is not simply "an expensive mutual fund." Use the alpha/beta framing (a long-only fund's return is mostly beta; a hedge fund's toolkit (shorting, leverage, derivatives) exists to attempt alpha instead), and then walk them through Worked Example 1's Year 3 specifically: a fund that returned 10% and yet the manager was paid no performance fee at all, because of a high-water mark. If your listener can, unprompted, explain why that is fair to the investor rather than a trick, you have installed the actual mechanism, not just the vocabulary. Then add: why would the exact same fund, under a soft hurdle with a catch-up provision instead, potentially leave that investor with less net return at some performance levels than a hard hurdle would, despite both being fully disclosed, standard, legitimate fee conventions?

Journal prompt. Think of a real or hypothetical allocation decision (your own capital, a family office, an institution you have read about) that put money into (or considered) a hedge fund, private fund, or any vehicle with fees, lock-ups, or gates you did not fully understand at the time. Write for fifteen minutes: what would you have needed to check, and against which primary source, before committing capital? Which of this module's five strategy buckets would that vehicle have fallen into, and would you have been able to state its actual return driver and its characteristic risk in one sentence, the way this module asks you to? Close with this question, and answer it in writing: the next time an AI tool tells you confidently what a fund "returned last year" or "charges," what is the one habit (from this module specifically) that stops you from repeating that number before you have traced it to Form ADV, a placement memorandum, or an audited statement you hold yourself?

This module's flashcards and mastery quiz are wired into the app: see the node's Quiz and Reviews.