Professional Documents
Culture Documents
▪ Duration is a first-order effect that accounts for changes in yield Yield curve flattener strategies
curve level.
▪ Convexity is a second-order effect that accounts for yield curve ▪ Duration neutral - Net zero duration position → Portfolio gain
slope and shape changes. from yield curve slope decrease
▪ All else equal, the expected price change of a bond with positive ▪ Bear flattener - Net negative duration position → Portfolio gain
convexity for a given rate decrease will be higher than the price from slope decrease and/or rising yields
change of an identical-duration, lower-convexity bond. ▪ Bull flattener - Net positive duration position → Portfolio gain
from slope decrease and/or lower yields
Yield curve strategies
Static yield curve Divergent yield curve shape view
Strategies for a static yield curve:
▪ Cash-based strategies
➢ Buy-and-hold – Constant without active trading
o Income: Coupon income
➢ Rolling down the yield curve - Constant, with Δ Price as
maturity shortens
o Income: Coupon income +/− Rolldown return
➢ Repo carry trade - Finance bond purchase in repo market
o Income: (Coupon income +/− Rolldown return) - Yield curve volatility strategies
Financing cost ▪ Bond call option
▪ Bond put option
▪ Derivatives-based strategies ▪ Callable bond
➢ Long futures – Purchase contract for future bond delivery ▪ Putable bond
o Targeted return: (Δ Price / Δ Bond yield) − Margin cost
Portfolio duration = sum of key rate durations Yield spreads compensate investors for assuming credit risk and
liquidity risk.
Key rate durations help:
▪ Credit risk for a borrower depends on the likelihood of default
▪ Quantify exposures along the curve and the loss severity.
▪ Identify a bond portfolio’s sensitivity to changes in the shape of ▪ Credit risk for a specific bond issuance also depends on the
the benchmark yield curve payment period, the debt seniority, and the repayment sources.
▪ Identify “shaping risk” – i.e., a bond’s sensitivity to changes in Liquidity risk: investor’s ability to easily buy or sell a specific
the shape of the benchmark yield curve security.
Active fixed-income management across currencies Default probabilities and recovery rates
Credit valuation adjustment (CVA) framework
The term structure of interest rates and exchange rates vary across
countries.
Yield curve strategies do not have to be limited to a single yield
curve.
A framework for evaluating yield curve strategies ASW (Asset Spread over MRR of fixed bond coupon
swap)
E(R) ≈ Coupon income +/− Rolldown return +/− E (Δ Price due to
benchmark yields) +/− E (Δ Price due to yield spreads) +/− E (Δ
Price due to currency value changes)
Changes in spread (DM or Z-DM) are the key driver of price change
QM (quoted margin): the yield spread over the MRR established
upon issuance to compensate investors for assuming the credit risk EffSpreadDurFRN = (𝑃𝑉)_−(𝑃𝑉)+/(2×(∆ 𝐷𝑀)(𝑃𝑉0)
of the issuer and fixed until bond maturity,
Bottom-Up credit analysis
DM (discount margin): the discount (or required) margin is the Bottom-up credit analysis involves an issuer-specific assessment
yield spread versus the MRR such that the FRN is priced at par on a and identifying suitable securities.
rate reset date. This changes if the issuer’s credit risk changes. ▪ For unsecured corporate bonds, factors like profitability and
leverage are used to analyze an issuer.
▪ For a sovereign borrower, the relevant metric is the economic
activity and the government’s ability and willingness to levy
taxes and generate sufficient revenue for its obligations.
▪ For a special purpose entity issuer with mortgage-backed bonds
or other securitized cash flows, factors considered are a credit
measure of both the residential borrowers and underlying
collateral value and internal credit enhancements.
Z-DM: Yield spread over MRR curve
Key Financial Ratios for Bottom-Up Credit Analysis
▪ For a bond portfolio, use market value-weighted averages for Z-Score Model = 1.2 × A + 1.4 × B + 3.3 × C + 0.6 × D + 0.999 × E
the duration and convexity measures.
▪ For lower-rated bonds, use the Duration Times Spread (DTS) A = Working Capital/Total Assets, B = Retained Earnings/Total
measure Assets, C = EBIT/Total Assets, D = Market Value of Equity/Total
▪ DTS ≈ (EffSpreadDur × Spread) Liabilities, E = Sales/Total Assets
▪ A portfolio’s DTS is the market value-weighted average of DTS of
its individual bonds Structural credit models use market-based variables to estimate
the market value of an issuer’s assets and the volatility of asset
value. The likelihood of default is defined as the probability of the
= 9 (= 50% × 5 + 50% × 13), which corresponds to an average CDS price change for a given CDS spread change ≈ − (∆(CDS
rating of Baa2/BBB. Spread) × EffSpreadDurCDS)
Using the weighted scale: portfolio average credit quality score Change in CDS contract value ≈ − (∆(CDS Spread) ×
EffSpreadDurCDS) x CDS notional amount
= 918 (= 50% × 70 + 50% × 1,766), nearer to Ba1/BB+.
Credit spread measures in top-down analysis: Credit spread Credit spread curve strategies
measures such as OAS are preferred to measure average portfolio Static Credit Spread Curve Strategies
credit quality. Spread-based rather than rating-based measures can If credit curves are expected to remain stable or unchanged over an
help determine portfolio value changes due to spread changes. investment horizon with low credit defaults and annual loss rates,
To calculate a portfolio’s average OAS, each bond’s individual OAS an active manager could use the following strategies:
is weighted by its market value. ▪ Lower portfolio’s average credit rating
▪ Add credit spread duration with similar rated longer-term
Liquidity Risk
bonds
Effective spread = Trade size × {Trade price – ((Bid + Ask))/
➢ Buy and hold
2 for buy order} and
➢ Rolling down
Trade size × {( Bid + Ask)/2 − Trade price for sell orders}
▪ Derivative-based credit strategies can add credit spread
Tail Risk duration or increase credit exposure.
Methods to Assess Portfolio Tail Risk ➢ Sell CDS single-name or index protection
➢ Use a long-short strategy
▪ VaR measures the minimum portfolio loss expected to occur at a
specific confidence level over a given time. Dynamic Credit Spread Curve Strategies
▪ Conditional value at risk (CVaR) or expected loss measures the
▪ Active strategies to capitalize on divergent market views given
average loss over a specific time conditional on that loss
anticipated credit curve changes across maturities and ratings.
exceeding the VaR threshold
▪ Active strategies can be cash-based or derivatives-based.
▪ Incremental VaR (or partial VaR) can be used to measure the
impact of adding or removing a bond in a portfolio.
▪ Relative VaR is used to measure the expected tracking error Global Credit Strategies
versus a benchmark portfolio by calculating VaR (or CVaR) Fixed Income Markets in Developed Countries:
based on a portfolio containing active positions minus the ▪ Broad range of private and public debt issuers
benchmark holdings under a market stress scenario. ▪ Bonds denominated in a freely floating domestic or other major
currency
Three models are used: ▪ Well-established and liquid derivatives
▪ Parametric method
▪ Historical simulation Fixed Income Markets in Emerging or Frontier Countries:
▪ Monte Carlo analysis ▪ Sovereign issuers, state-owned or controlled enterprises, banks,
producers operating in a dominant domestic industry
Synthetic credit strategies ▪ Bonds are denominated in a restricted domestic currency with
varying liquidity
Contract references a specific issuer (or issuers) and credit event
▪ Sovereign government and domestic issuers often issue global
terms that, when triggered, lead to a settlement payment equal to
bonds in US dollars or euros
the LGD multiplied by the contract notional amount from the seller ▪ Little to no diversification
to the buyer.
▪ Herfindahl–Hirschman Index (HHI) - a valid measure of stock- Two main approaches to style classification are:
concentration risk in a portfolio = ∑𝑛𝑖=1 𝑤𝑖 2 ▪ Holdings-based approaches: aggregate the style scores of
where wi = weight of stock i in the portfolio, n = no. of securities individual holdings
HHI of 1/n ⟶an equally weighted portfolio, HHI of 1.0 ⟶ ▪ Returns-based approaches: Compare the strategy’s returns to
portfolio concentration in a single security those of style indexes.
▪ Effective no. of stocks = 1/HHI
Active Equity Investing: Portfolio Construction
In factor-based strategies we try to replicate the performance of a
benchmark by creating a portfolio that has the same exposure to The four major building blocks used in portfolio construction
risk-factors as the benchmark. Some risk factors include: growth, are:
value, size, yield, momentum, quality, volatility etc. ▪ Factor weighting: This involves overweighting, underweighting
In market-capitalization-weighted indexing, we create or neutralizing rewarded factors.
portfolios with the same weights of constituent securities as the ▪ Alpha skills: This involves using expertise, experience and
benchmark index. superior analysis to identify assets that are
overpriced/underpriced. It also involves timing exposures to
In the full replication approach, all securities represented by the factors, securities and markets.
index are purchased. ▪ Position sizing: This refers to the choice between large positions
In the stratified sampling approach, we split the population into in a few stocks versus small position in many stocks.
▪ Breadth of expertise: It involves integrating the three building
strata (or sub-groups) and then sample from each strata.
blocks mentioned above. The breadth of expertise is high when
In the optimization approach, we try to maximize desirable
the number of independent decisions are high.
characteristics or minimize undesirable characteristics subject to Active share measures the extent to which the number and sizing
one or more constraints. of positions in benchmark differ from the portfolio. It is not
Attribution analysis refers to the analysis of sources of return of impacted by the correlation between securities, hence a portfolio
the portfolio and the underlying index and identifying the reasons manager has complete control on this measure.
It is calculated as:
xi = asset's weight in the portfolio 6. Multi-Manager Strategies Most investors invest in a range of
bi = benchmark weight in asset i hedge fund strategies
RCij = covariance of relative returns between asset i and asset j Three main approaches
▪ Creating one's mix of managers
Contribution of each asset to the portfolio active variance (CAVi): ▪ Fund-of-funds
CAVi = (𝑥𝑖 − 𝑏𝑖)𝑅𝐶𝑖𝑝 where: ▪ Multi-strategy funds
RCip = covariance of relative returns between asset i and the
portfolio Conditional Factor Risk Model applied to a hedge fund strategy's
returns is:
Long-only approaches rely on long positions only and no short (𝑅𝑒𝑡𝑢𝑟𝑛 𝑜𝑛 𝐻𝐹𝑖)𝑡 = 𝛼𝑖 + 𝛽𝑖,1 (𝐹𝑎𝑐𝑡𝑜𝑟 1)𝑡 + 𝛽𝑖,2 (𝐹𝑎𝑐𝑡𝑜𝑟2)𝑡 + ⋯
positions are taken.
+ 𝛽𝑖,𝐾 (𝐹𝑎𝑐𝑡𝑜𝑟 𝐾)𝑡 + 𝐷𝑡 𝛽𝑖,1 (𝐹𝑎𝑐𝑡𝑜𝑟1)𝑡
Long/short positions have both long and short positions. The + 𝐷𝑡 𝛽𝑖,2 (𝐹𝑎𝑐𝑡𝑜𝑟 2)𝑡 + ⋯ + 𝐷𝑡 𝛽𝑖,𝐾 (𝐹𝑎𝑐𝑡𝑜𝑟 𝐾)𝑡
positions can be negative and are not constrained to sum 1. + (𝑒𝑟𝑟𝑜𝑟)𝑖,𝑡
A long extension portfolio can be considered as a hybrid of long- Performance Contribution to a 60/40 Portfolio - Adding a 20%
only and long/short strategies. It is also called 'enhanced active allocation of a hedge fund strategy group to a traditional 60%/40%
equity' strategy. A popular example is the 130/30 strategy. portfolio typically results in:
▪ Lower total portfolio standard deviation
A market-neutral portfolio is another specialized form of
▪ Higher Sharpe and Sortino ratios
long/short portfolio construction. In dollar-neutral portfolios, ▪ Lower maximum drawdown.
the dollars invested in long positions are equal to the dollars
invested in short positions. In true market-neutral portfolio, the Asset Allocation to Alternative Investments
exposure to the market is cancelled out.
The Role of Alternative Investments in a Multi-Asset Portfolio
Alternative Investments Portfolio Management ▪ Higher risk-adjusted returns
▪ Achieve: capital growth, income generation, risk diversification,
Hedge Fund Strategies and/or safety
▪ Private equity
1. Equity Strategies: Invest in equity and equity-related ▪ Return enhancer
instruments. ▪ Illiquidity risk
Equity L/S Strategies Hedge funds
Dedicated Short Selling and Short Biased ▪ Spectrum: risk reducers to return enhancers.
Equity Market Neutral Long/short equity strategies provide equity-like returns but with
lower exposure to equity premium.
2. Event-Driven Strategies take positions in corporate securities Short-biased equity strategies try to generate alpha by going
and derivatives to profit from events such as mergers and short on overvalued securities.
acquisitions, bankruptcies, share issuances, buybacks, capital Arbitrage and event-driven strategies deliver equity-like returns
restructuring, reorganization, etc. with little to no correlation with traditional asset classes.
Merger Arbitrage Real assets
Distressed Securities ▪ Generally perceived to provide a hedge against inflation.
▪ Timber investments provide both growth and inflation-hedging
3. Relative Value Strategies earn good returns (credit, liquidity, properties.
or volatility premiums) during normal conditions. In a financial ▪ Commodities serve as a hedge against inflation and provide a
crisis, these strategies can result in losses. Equity market-neutral differentiated source of alpha.
investing is a relative value strategy.
The value of the tax-exempt account compounds using the pre-tax ▪ The inherent company-specific risk
returns, R. ▪ The decrease in portfolio efficiency caused by a lack of
diversification
FV = (1 + R)n ▪ The liquidity risk inherent in privately-held security or a
publicly-traded security
Approaches to mitigate the risks of a concentrated position: Introduction to Estate Planning: Wills, Probate, and Legal
Systems
▪ Sell and diversify
▪ Staged diversification ▪ Will: Legal document containing an individual’s instructions for
▪ Hedging and monetization strategies distributing his property after death.
▪ Tax-free exchanges ▪ Testator: The person transferring assets through a will is the
▪ Charitable giving strategies testator.
▪ Tax-avoidance and tax-deferral strategies ▪ Intestate: If an individual dies without leaving a valid will – a
common occurrence - she is declared to have died intestate, and
Tax-Optimized Equity Strategies – Equity Monetization, Collars the distribution of her assets is determined by court.
and Call Writing ▪ Probate: The legal process to confirm the validity of a will.
The case is based on the different life stages – from early career
phase to retirement of a married couple, the Schmitts.
In this case study:
▪ We identify and analyze the Schmitts' risk exposures and
observe how the types of risk exposure change across their 4. Risks Associated with Illiquid Asset Classes
different life stages. The analysis is conducted holistically
starting from the economic balance sheet, including human Key risks
capital. ▪ Uncertain cash flow pattern
▪ We recommend and justify methods to manage the Schmitt ▪ Smooth returns
family's risk exposures at different stages of their professional ▪ Rebalancing is costly
life. We use insurance, self-insurance, and adjustments to their
investment portfolio. Cash flow modeling
▪ We recommend and justify modifications to the Schmitts' life
▪ Drawdown structure
and disability insurance at different stages of the income
▪ Commitment strategy and liquidity requirements to meet capital
earners' lives.
calls
▪ Finally, we recommend and justify a plan to manage risk to the
▪ Too much → Severe liquidity risk; Too little → falling short of
Schmitts' retirement lifestyle goals.
return expectations
Human capital ⟶present value (PV) at a wage-risk adjusted
Addressing return smoothing behavior of illiquid asset classes
discount rate of the expected stream of income from employment
𝑝(𝑠𝑡)𝑤𝑡−1(1+𝑔𝑡 ) ▪ Using public market proxies in place of private asset classes.
𝐻𝐶0 = ∑𝑡=𝑁
𝑡=1 (1+𝑟 +𝑦)𝑡 ▪ Unsmoothing observed returns.
𝑓
Amount of coverage that the life insurance policy should Techniques used to unsmooth returns:
provide can be calculated using two methods. ▪ Geltner method - removes only the first-order serial correlation
▪ the human life value method ⟶ estimates the amount of future in observed returns.
earnings that must be replaced, ▪ Okunev and White method - extended the method of Geltner
▪ the needs analysis method ⟶ estimates the amount needed to (1993) to include higher-order serial
cover a survivor's living expenses. ▪ correlations.
▪ Getmansky, Lo, and Makarov (GLM) method - assumes that
Integrated Cases in Risk Management: Institutional observed returns for illiquid asset classes and hedge funds
follow a moving-average process.
Financial risks faced by institutional investors
5. Managing Liquidity Risk
1. Long-Term Perspective ▪ Establish liquidity risk parameters.
▪ Pension funds, sovereign wealth funds, endowments, and ▪ Assess the liquidity of current portfolio and monitor the
foundations take a long-term investment perspective evolution over time.
▪ Ability to invest in illiquid assets ▪ Develop a cash flow model and project future cash flows.
▪ Important to understand the financial risks that arise from ▪ Stress test liquidity needs and cash flow projections.
illiquid assets ▪ Develop an emergency action plan.
▪ Important to understand the interaction between liquidity risk
and market risk 6. Enterprise Risk Management for Institutional Investors
▪ These risks can pose ‘an existential threat’ to long-term
institutional investors
Composite returns
Sum of beginning assets and weighted external cash flows =
𝑛
𝑉𝑝 = 𝑉0 + ∑(𝐶𝐹𝑖 × 𝑤𝑖 )
𝑖=1
Beginning assets weighting method composite return =
𝑉0,𝑝𝑖
𝑟𝐶 = ∑ [𝑟𝑝𝑖 × 𝑛 ]
∑𝑝𝑖=1 𝑉0,𝑝𝑖
Beginning assets plus weighted cash flows method composite
return =
𝑉𝑝𝑖
𝑟𝑐 = ∑ (𝑟𝑝𝑖 × )
∑ 𝑉𝑝𝑖
Dispersion Measure
The standard deviation for a composite in which the constituent
portfolios are equally weighted =
∑𝑛 ̅̅̅̅ 2
𝑖=1(𝑟𝑖−𝑟𝐶 )
SC = √
𝑛
2
SCaw = √∑𝑛𝑖=1 [(𝑟𝑖 − 𝑟̅𝑝𝑟𝑜𝑥𝑦 ) × 𝑤𝑖 ]
𝑉0,
wi = 𝑖
𝑉0, 𝑇𝑜𝑡𝑎𝑙