Curriculum Part 5 Chapters 25–34 · the largest part

Hedge Funds & Managed Futures

Strategy by strategy: what the legs are, which risk premium is harvested, what the strategy is implicitly short, and how it is benchmarked, replicated, accessed and regulated.

Managed futures: structure & foundations

The industry. Commodity trading advisers trade listed futures and forwards across equity indices, rates, currencies and commodities, typically through managed accounts or funds, with a regulatory perimeter distinct from equity-focused hedge funds. Two features shape everything: instruments are exchange-cleared and liquid, and positions can be long or short with equal ease.

Four core dimensions classify any managed futures programme:

Data sources
Fundamental — economic forecasts, supply-and-demand estimates, crop and inventory data. Technical — historical price and volume.
Implementation style
Systematic — rules determine entry, exit, scaling and sizing, with execution automated; typically broader across markets and strategies. Discretionary — the manager decides, often informed by models, so the boundary is blurred in practice.
Strategy focus
Momentum, mean reversion, global macro, relative value, carry trading, multi-strategy.
Time horizon
High-frequency, short-term, medium-term and long-term — the dimension that most affects capacity, turnover and transaction-cost sensitivity.

Why trend following works, when it works. Momentum is documented across asset classes and horizons; the behavioural explanations are under-reaction to news followed by herding, and the structural explanations include hedging demand and the slow diffusion of information. The payoff profile is the important part: a trend follower is long a straddle in effect — it loses small amounts in choppy, range-bound markets and makes large gains in sustained directional moves, giving positive skew and long-volatility, crisis-friendly behaviour. Distinguish time-series momentum (each market against its own history, the CTA norm) from cross-sectional momentum (relative winners against losers, the equity-factor norm).

Portfolio construction in a systematic programme is largely risk management: volatility-scaled position sizing so each market contributes comparable risk, correlation-aware diversification across sectors, explicit stop and risk limits, and a rebalancing rule. Capacity is set by liquidity in the underlying futures, and turnover determines how much transaction cost erodes the signal.

Investing in CTAs

Benefits claimed and testable: low average correlation with equities; a tendency to perform well in sustained equity drawdowns because trends develop in rates, currencies and commodities as well; genuine liquidity and transparency relative to other alternatives; and no credit or valuation opacity, because positions are exchange-priced daily.

The honest caveats: correlation is unstable and can be positive when a rally is trending; performance in sharp, V-shaped reversals is poor because trend signals are whipsawed; long flat or negative stretches are normal and test governance more than they test the strategy; and dispersion across managers is wide, so manager selection matters as much as the allocation decision.

Risk measurement for CTAs needs care: leverage is notional rather than balance-sheet, so exposure must be measured in risk terms (volatility contribution, margin-to-equity, stress loss) rather than as gross assets. Because returns are close to symmetric but fat-tailed, drawdown and tail measures are more informative than volatility alone.

Three approaches to benchmarking CTAs: peer-group indices of manager returns (contaminated by self-selection, survivorship and backfill bias, and by heterogeneous strategies inside one label); systematic rule-based indices that mechanically implement a trend strategy, giving a replicable, investable standard for what generic trend earns; and factor or regression-based benchmarks that decompose returns onto time-series momentum factors across asset classes. The last is the most analytically satisfying and the basis for arguing that a given manager's alpha is or is not generic trend.

Managed accounts and platforms are the structural counterpart: the investor owns the assets, sees positions daily, can impose risk limits and terminate quickly, and avoids commingled-fund contagion — at the cost of operational build, minimum size, and potential tracking error against the manager's flagship fund.

Relative value strategies

Limits to arbitrage is the organising idea. A mispricing can persist because arbitrage requires capital, financing and time: fundamental risk (the trade may simply be wrong), noise-trader risk (the gap can widen before it closes, triggering margin calls and redemptions), implementation costs (borrow cost and availability, transaction costs, short-sale constraints), and model risk. The practical consequence — an arbitrageur with a finite horizon and outside capital may be forced to close a correct position at a loss — is the mechanism behind most relative-value blow-ups, and the phrase to reach for is funding rather than valuation.

Convertible arbitrage. The convertible is a bond plus an equity call plus, typically, an issuer call and other features. Decompose it into investment value (the straight-bond floor), conversion value, and the option premium. The classic trade is long the convertible and short delta shares of the underlying, capturing cheap implied volatility and earning gamma by re-hedging the delta as the stock moves, while carrying credit exposure that may be hedged with CDS and interest-rate exposure that may be hedged with rates. Sources of return: static income (coupon plus short-rebate less borrow cost), volatility capture through re-hedging, and cheapness convergence. Risks: credit deterioration, a collapse in implied volatility, rising borrow cost or a short-selling ban, and — decisively — withdrawal of repo and prime-broker leverage, which is what turned a valuation opportunity into a liquidation in 2008.

Pairs trading and market neutrality. Pairs trading takes offsetting positions in two related securities when their spread deviates from its historical relationship, relying on cointegration rather than correlation — a distinction the exam tests, because correlated series can drift apart permanently while cointegrated ones share a long-run equilibrium. Then be precise about neutrality: dollar neutral (equal long and short market value) is not beta neutral (equal beta-weighted exposure), and neither guarantees sector, factor or currency neutrality. Most "market-neutral" disappointments are residual factor exposure that nobody neutralised.

Directional strategies

Financial economics of directional strategies. Returns come from a mix of market beta, factor exposures and genuine selection or timing skill, and the analytical task is separating them. A long/short manager with a persistent net long position has an equity beta; one that varies its net exposure has a conditional beta that static regression will misprice. This is where rolling betas and conditional models earn their place.

Equity long/short. Return sources are stock selection on both sides, net exposure management, and the short rebate. Key measures to keep straight: gross exposure (long + short, a leverage measure), net exposure (long − short, a directional measure), and beta-adjusted net exposure, which is the one that actually predicts market sensitivity. Short selling has asymmetric practical risks — unlimited loss potential, borrow recall, squeeze risk, dividend and corporate-action liabilities — which is why the short book is usually more diversified than the long. Dedicated short bias exists as a category and is structurally disadvantaged by the upward drift of equity markets. Variants to know: sector or region specialists, quantitative equity market neutral, and activist strategies, where the return driver is influencing the company rather than predicting it.

Global macro. Top-down positioning across rates, currencies, equity indices, credit and commodities, expressed in liquid instruments and frequently with asymmetric option structures. Distinguish discretionary macro (thematic, judgement-driven, concentrated, hard to replicate) from systematic macro (rules-based, diversified, closer to a CTA). Return drivers are macro forecasting, policy analysis, and identifying unsustainable pegs, imbalances or valuations; risks are timing (correct thesis, wrong entry), leverage, liquidity in stress, and the fact that a small number of large positions produces lumpy returns and severe path risk. Because positions are liquid, macro funds typically offer better liquidity terms than most hedge fund strategies — a genuine portfolio advantage worth stating in an essay.

Credit strategies

Economics of credit risk. A spread compensates for expected loss (default probability × loss given default), a default risk premium for bearing the systematic component, plus liquidity and tax components. Because spreads are set under a risk-neutral measure, the implied default probability exceeds the physical one — the most reliably examined distinction in this chapter.

Merton model
Equity is a call on firm assets struck at the face value of debt; risky debt equals a risk-free bond less a put. Default happens only at maturity when asset value falls short. Inputs: asset value and volatility (unobservable, inferred from equity), leverage, horizon. Predicts near-zero short-horizon spreads — the credit spread puzzle — because default is predictable and there is no jump.
KMV extension
Practitioner structural model: computes distance to default from an estimated default point (typically between short-term and total liabilities) and maps it empirically to an expected default frequency, which loosens Merton's distributional assumption.
Reduced-form models
Default is the first jump of a Poisson process with hazard rate λ, calibrated from market spreads; survival probability over t is e−λt. Fits observed prices exactly and handles entities with no observable asset value, but offers no economic account of why default occurs, and λ and recovery are not separately identified from spreads alone.
Empirical models
Statistical scoring on accounting and market variables (the Altman Z-score tradition). Practical, transparent, atheoretical; vulnerable to changing accounting practice and regime shifts.

Distressed debt. Two orientations: passive/trading — buying claims cheaply and selling into recovery, and active/control — accumulating the fulcrum security to influence or take equity in the reorganised company. Identifying the fulcrum security, the class that converts into post-reorganisation equity, is the core analytical act, and it requires an enterprise-value estimate plus the absolute-priority ordering of claims. Bankruptcy regime matters enormously: creditor-friendly versus debtor-friendly regimes differ on whether management stays in control, how quickly a plan can be imposed, and whether cross-border assets can be consolidated — so the same balance sheet has different recovery expectations in different jurisdictions. Valuation risk is severe: marks are model-based, timelines are uncertain, and legal outcomes are negotiated rather than computed.

Asset-based lending completes the chapter: loans secured on specific collateral — receivables, inventory, equipment, real estate, royalties, life settlements — where underwriting is collateral valuation and control rather than cash-flow projection. Return comes from a high contractual coupon plus fees; risk is collateral valuation error, perfection of security interests, servicing capability, and the borrower population's adverse selection.

Volatility, correlation & dispersion

Volatility as a risk factor. The volatility risk premium exists because investors pay for protection: implied volatility persistently exceeds subsequently realised volatility, so systematic option sellers are paid for supplying insurance and bear negatively skewed returns. Related premia sit in the same family — the correlation risk premium (index options are expensive relative to a basket of single-name options because index protection is what buyers want) and the variance risk premium.

Managing exposure with options. Covered calls monetise the premium and cap upside; protective puts and collars buy convexity at a carry cost; put spreads cheapen protection while capping it. The exam's framing is always cost per unit of protection and what is given up.

Modelling volatility. Volatility clusters and mean-reverts, which is what GARCH-family models capture; stochastic volatility models add a second random driver and can generate the smile; implied volatility is forward-looking but contains a risk premium, so it is a biased forecast of realised volatility. Know that the three volatilities — realised, implied, forecast — answer different questions.

Products. Variance swaps deliver pure exposure to realised variance against a strike with no delta hedging, and are convex in volatility, so a short position loses quadratically — this convexity is why variance swaps, not volatility swaps, are the classic blow-up instrument. Volatility swaps are linear in volatility but harder to hedge. VIX futures reference expected 30-day implied variance, not spot VIX, so a constant-maturity long position pays roll cost whenever the curve is in contango; VIX options and exchange-traded products inherit that structure, and daily-reset leveraged products compound the drag.

Dispersion trading sells index volatility and buys single-name volatility (or the reverse), monetising the correlation risk premium. It is short correlation: it profits when constituents move independently and loses sharply when everything moves together — precisely in a crisis. That is a compact example of the general lesson that most volatility-selling strategies are short a tail.

Volatility hedge funds divide into relative-value volatility (arbitraging the surface across strikes, maturities and underlyings), tail-risk funds (systematically long convexity, reliably negative carry, designed to pay in a crash), and short-volatility carry funds (steady returns, severe drawdowns). Sizing and governance matter more than forecasting for all three.

Hedge fund replication

The case for replication rests on the claim that much of average hedge fund return is systematic exposure rather than skill. If so, a liquid replication delivers most of the return at lower fees, with daily liquidity, full transparency, no lock-up or gate, no manager-specific operational risk, capacity that scales, and — importantly for institutions — no adverse selection in access.

Factor-based
Regress index or fund returns on liquid factors and hold the estimated exposures. Simple and transparent; backward-looking, so it lags exposure changes, and it inherits omitted-factor error. Works best on strategies with stable betas.
Payoff-distribution
Replicate the distribution of returns rather than the path, typically with options on a reserve asset. Matches moments; does not match timing, so it is a poor hedge for a specific fund and a reasonable substitute for an allocation.
Algorithmic (bottom-up)
Implement the strategy's mechanics directly — a rule-based merger-arbitrage or trend-following programme. Highest fidelity where the process is genuinely systematic; not feasible for discretionary skill.

Limits. Replication cannot capture idiosyncratic skill, illiquidity or private information, and the empirical record is mixed and strategy-dependent — strongest for trend and long/short equity, weakest for distressed, activist and other strategies whose returns come from control and negotiation. Related wrappers extend the same idea: alternative mutual funds and UCITS offer liquid, regulated access with leverage and liquidity constraints that dilute the strategy, and alternative ETFs add intraday liquidity and rules-based transparency at the cost of strategy simplification.

Funds of funds & multi-strategy

Access routes: direct single-manager investing, funds of hedge funds, multi-strategy funds, managed accounts and platforms, and replication products. Each answers a different constraint — capital, staffing, governance speed, and liquidity requirements.

Funds of hedge funds provide manager selection and diligence, diversification across strategies, access to closed managers, and monitoring for investors without an internal team. Costs: a second fee layer that compounds against net return, potential mismatch between the fund's liquidity terms and the underlying funds' terms, reduced transparency at the position level, and dilution toward index-like returns as the number of managers grows.

Portfolio construction inside a fund of funds involves strategy allocation, manager count (diversification against dilution), position sizing on a risk basis, liquidity laddering so redemptions can be met without gating, and cash management for subscription and redemption flows. Value is added through selection, strategy timing, access, and risk management — and the exam expects an honest note that the evidence for the first two is weaker than the marketing.

Fund of funds versus multi-strategy. A multi-strategy fund allocates internally across strategies run by its own teams: capital can be moved faster, netting of fees across strategies is more favourable to investors, and there is a single operational infrastructure to diligence — but concentration of business and operational risk in one firm is total, and internal transparency depends on one governance structure. A fund of funds diversifies firm risk and can access external specialists, but reacts slowly and pays a fee layer for the privilege.

Hedge fund indices are the measurement backdrop and carry the well-known biases: self-selection in reporting, survivorship, backfill, instant history, and heterogeneous strategy definitions across providers — so index comparisons overstate returns and understate risk. Investable indices reduce some biases and introduce capacity and selection constraints of their own.

Hedge fund operational due diligence

How it differs from private equity ODD. Hedge funds trade continuously, so the review focuses on trading operations, cash movement, valuation frequency and counterparty arrangements; private equity ODD leans on legal documents and long-horizon firm viability. Both are independent of the investment team and, in strong programmes, hold a veto.

Trading procedures
Trace the trade lifecycle: order generation and authorisation, execution and allocation across accounts, confirmation and settlement, and reconciliation — daily, and against the administrator and prime broker, not internally only. Test who can override, and what happens on a break.
Cash management
The highest-risk area. Who may instruct a wire, is there dual authorisation, are third-party payments restricted, and are bank accounts in the fund's name and reconciled independently? Sole authority in one person is disqualifying in practice.
External parties
Administrator, auditor, custodian, prime brokers, valuation agents, legal counsel — verified directly, with attention to independence, capacity relative to fund size, and turnover history.
Compliance
Registration status, compliance manual and testing, personal-account dealing and pre-clearance, insider-information controls and restricted lists, expense allocation, marketing accuracy, and regulatory examination history.
Documentation & decision
Findings recorded so the decision is reviewable, with severity ratings, remediation commitments and re-review dates; the output is approve, approve with conditions (managed account, side letter, limits), or decline.
Investigative diligence
Background and litigation checks on principals, verification of biographies and credentials, and public-record searches. Misrepresented history is a strong predictor of other misrepresentation.

Resource allocation is itself examinable: build an internal team, outsource to a specialist, use a hybrid, or rely on the fund of funds or consultant — with the trade-offs in cost, independence and depth. Governance completes the picture: an independent, competent, non-overcommitted board of directors for offshore funds, with the authority to challenge the manager on valuation and gating; and insurance coverage (directors' and officers', errors and omissions, fidelity) whose exclusions matter more than its headline limits.

Regulation & compliance

Three foundational principles of financial market regulation: protecting investors; ensuring markets are fair, efficient and transparent; and reducing systemic risk. Every specific rule you meet should be attributed to one of these — that framing is what an essay answer needs.

United States. The architecture rests on the securities and commodities statutes plus the adviser regime: private funds rely on offering exemptions that restrict them to accredited and qualified investors and limit general solicitation (relaxed for certain offerings under the JOBS Act, subject to verification requirements); investment advisers above thresholds register and are subject to examination, compliance-programme, custody and reporting obligations; systemic-risk reporting was introduced post-crisis; derivatives reforms brought central clearing, trade reporting and margin requirements for standardised swaps; and commodity pool operators and trading advisers face their own registration regime. Insider trading and market manipulation enforcement applies with full force to private funds.

Europe. The alternative investment fund managers regime is the centrepiece: authorisation of managers, a depositary requirement, leverage and liquidity reporting, remuneration rules, valuation independence, and marketing rules via passport or national private placement. UCITS provides the retail-eligible wrapper with strict eligible-asset, diversification and liquidity rules — which is precisely why UCITS versions of hedge fund strategies are diluted. Distribution rules and disclosure obligations add cost, but also give European investors a depositary and reporting regime that US private funds do not have.

Asia. Regulation is jurisdiction-by-jurisdiction rather than unified: licensing of managers, distinctions between professional and retail investors, and marketing restrictions differ substantially across the major centres, so the practical answer to "can we market this fund there?" is always jurisdiction-specific.

What this means for an allocator: regulatory regime is a diligence input, not a substitute for diligence. Registration does not verify valuations; a depositary does not guarantee performance; and an unregulated offshore fund with excellent independent governance can be operationally safer than a registered one with weak controls.

Confusion pairs

Time-series vs cross-sectional momentum
Each market against its own history (CTA style, long-volatility payoff) vs relative winners against losers (equity factor style).
Systematic vs discretionary
Rules determine entry, exit and sizing, broadly diversified vs manager judgement, usually more concentrated and harder to replicate.
Gross vs net exposure
Long + short (leverage) vs long − short (direction). Beta-adjusted net is the one that predicts market sensitivity.
Dollar vs beta neutral
Equal market values vs equal beta-weighted exposures. Neither implies sector, factor or currency neutrality.
Correlation vs cointegration
Co-movement of returns vs a shared long-run equilibrium in levels. Pairs trading needs the second.
Merton vs KMV
Closed-form structural model with default only at maturity vs a practitioner variant mapping distance to default onto an empirical default frequency.
Variance vs volatility swap
Convex in volatility, quadratic losses on a short vs linear in volatility, harder to hedge.
Fund of funds vs multi-strategy
External managers, firm-risk diversification, second fee layer, slow reallocation vs internal teams, fast reallocation and fee netting, concentrated operational risk.
Dispersion trade direction
Short index vol / long single-name vol is short correlation — it loses when everything moves together.

Practice

Six multiple-choice questions in exam style, with the reasoning — not just the letter.

1. A trend-following CTA's return profile is best described as similar to:

A. A short straddle
B. A long straddle
C. A covered call
D. A short put on equities
B. Small, frequent losses in choppy markets and large gains in sustained moves in either direction — positive skew, long-volatility behaviour. That is why trend often performs in extended equity drawdowns and poorly in sharp reversals.

2. A convertible arbitrage portfolio suffers severe losses despite no change in the issuers' credit quality or implied volatility. The most likely cause is:

A. Delta drift
B. Withdrawal of repo and prime-broker financing, forcing deleveraging
C. A fall in risk-free rates
D. Dividend increases on the underlying shares
B. The strategy's binding constraint is funding liquidity. When leverage and stock borrow are withdrawn, positions are liquidated regardless of valuation — the limits-to-arbitrage mechanism in its purest form.

3. Which statement about reduced-form credit models is most accurate?

A. They derive default from the firm's asset value falling below a barrier
B. They treat default as an exogenous jump and can fit observed spreads exactly, but cannot separately identify λ and recovery
C. They require an estimate of asset volatility
D. They predict higher short-horizon spreads than observed
B. A and C describe structural models; D describes the opposite of the credit spread puzzle, which is that structural models predict spreads that are too low at short horizons.

4. A dispersion trade that is short index volatility and long single-name volatility is:

A. Long correlation, and profits in a crisis
B. Short correlation, and loses when constituents move together
C. Neutral to correlation
D. Equivalent to a long variance swap on the index
B. It monetises the correlation risk premium and therefore loses when realised correlation rises toward one — exactly what happens in a market-wide shock.

5. For which strategy is factor-based replication likely to work least well?

A. Managed futures trend following
B. Long-biased equity long/short
C. Activist and distressed control strategies
D. Equity market neutral
C. Returns generated by control, negotiation and legal process are idiosyncratic and cannot be reproduced with liquid factor exposures. Trend and long/short equity have relatively stable, identifiable exposures.

6. During hedge fund ODD, the single most serious finding among the following is:

A. The fund uses two prime brokers rather than one
B. The CIO has sole authority to instruct wire transfers
C. The administrator is based in a different jurisdiction from the manager
D. Compliance manual last updated eighteen months ago
B. Unilateral cash-movement authority in the hands of the investment decision-maker removes the key control against misappropriation. The others are ordinary findings requiring follow-up, not disqualifying ones.

Constructed-response practice

Write these under time. Each outline is the shape the rubric rewards, not a model answer to memorise.

Prompt A (15 minutes). An endowment holds a fund of hedge funds and is considering replacing it with a combination of three direct managers and a liquid replication product. Evaluate the proposal.

Outline:
  • State what the fund of funds provides: selection, diligence, access, diversification, monitoring — and its cost, the second fee layer plus liquidity mismatch and reduced transparency.
  • Assess direct investing against the governance budget: staffing for ODD and monitoring, decision speed, minimum sizes, and concentration risk with only three managers.
  • Assess replication honestly: fee, liquidity and transparency benefits; strategy-dependent fidelity; no idiosyncratic skill, no illiquidity premium.
  • Recommend a structure — e.g. replication for generic beta-like exposure, direct managers where skill is verifiable, retained external help for diligence — and name the condition under which the recommendation changes (adding investment staff).
  • Close on measurement: how you would benchmark each sleeve, including factor-adjusted attribution rather than peer indices.

Prompt B (12 minutes). A short-volatility carry fund reports a five-year Sharpe ratio of 2.1 with no monthly loss exceeding 2%. Explain why this understates risk and describe the diligence and risk measures you would apply.

Outline:
  • Identify the payoff: short optionality earns the volatility risk premium; returns are negatively skewed with a rare severe tail the sample may exclude.
  • Explain why Sharpe is the wrong statistic — it uses only two moments and rewards a smooth series produced by selling insurance.
  • Name better measures: CVaR/expected shortfall, stress and scenario losses on a volatility spike with correlation going to one, maximum drawdown, and convexity/gamma exposure.
  • Diligence: valuation of positions, margin and financing terms, leverage measured in risk not notional terms, position limits, and behaviour in the worst historical episodes.
  • Conclude with sizing and portfolio context: cap the allocation, and note the pairing argument with a long-convexity or trend allocation.
← 12 · Commodities 14 · Structured Products →