Volume 1 Flashcards
Business cycle phases & analysis ?
1) Initial Recovery = low yields, stocks rally, riskier assets outperform: STEEP YIELD CURVE.
2) Early Upswing = Short rates rise, stocks rise: YIELD CURVE BEGINGS TO FLATTEN.
3) Late Upswing = rates rise, stocks rise, volatility rise.
Monetary policy shifts from neutral to tight.
4) Slowdown = short rates topping, bonds begin to rally, stocks are soft/falling.
YIELD CURVE MAY INVERT.
5) Recession = Yields drop, stocks drop.
Taylor Rule ?
Appropriate Policy interest rate
= ST neutral real rate + θ + 0.5(θ - θ *) + 0.5(Y-Y *).
Avec θ = inf forecats; θ* = target inf; Y = GDP forecast; Y* = Potential GDP; ST neutral real rate + θ = nominal rate.
Long-run CME = use r_neutral.
ST CME = must consider the expected path of short rates (dictated by Yield Curve) donc on rajoute θ.
E(avg rates 10yr｜known fiscal&monet policy) ?
CME framework in the PF Mgmt process ?
1) Specify the set of asset classes included, expectations needed, including the time horizon(s) to which they apply.
2) Research the historical record
3) Specify the methods/models & their information requirements
4) Determine best sources of information
5) Interpret Current investment environment (= levels of returns more important than precision!!! eg: r_ai > r_e > r_b)
6) Produce the set of expectations
7) Monitor actual outcomes.
When interpreting Current investment environment, what is the most important ?
Levels (osef de la précision).
eg: r_ai > r_e > r_b)
Challenges of economic market forecast ?
A) Limitation of economic Data
B) Data Measurement errors&biases
C) Limitations of Historical estimates
D) Ex-post risk can be a biased measure of ex-ante risk
E) Bias in analyst’s methods (eg. data mining bias & time-period bias).
F) Failure to account for conditioning info (solution = estimate re for different business cycles)
G) Misinterpretation of correlation (spurious correl & does not imply causation)
H) Psychological traps (anchoring, overconfidence, status quo, prudence, confirmation bias, availability).
Availability bias = the tendency to be overly influenced by events that have left a strong impression and/or for which it is easy to recall an example
I) Model Uncertainty
A) Limitations of economic data ?
1) Definition (changes)
2) Construction (re-basing)
3) Timeliness (lag)
4) Accuracy (Revisions)
B) Data Measurement errors & biases ?
1) Transcription errors (1.3 instead of 3.1)
2) Survivorship Bias
3) Appraisal (smoothed) data = Produces lower volatility and correlation measures.
C) Limitations of Historical Returns ?
b.Discuss challenges in developing capital market forecasts
1) Regime changes (tech, political, legal)
2) Long Data Series: multiple regimes, may not be available, missing&asynchronous data:
High-frequency (daily) data produce more precise variances and co-variances (and less precise means). High-frequency data are more sensitive to asynchronism (missing data).
I) Model Uncertainty (is the model correct?) ?
1) Parameter uncertainty = estimated with error.
2) Input uncertainty = proxies.
What can exogenous shocks affect ?
a) Trend Growth = long term GDP growth.
b) Cyclicality = business cycle (short-intermediate term).
What’s an exogenous shock ?
Events from outside the economic system that affect its course.
May be short-lived or drive changes in trend growth.
Can be positive or negative.
Exogenous shock types ?
1) Policy changes
2) New products/technology
3) Geo-politics
4) Natural disasters
5) Natural resources/critical inputs
6) Financial Crises.
Economic growth trend factors to formulate Capital Market expecations ?
LT GDP growth trend =
1) growth from labour inputs (potential labour force + participation rate)
2) Growth from labour productivity (growth from increasing Capital Inputs + growth in TFP)
Capital Deepening ?
= Labor productivity growth - TFP growth
What will the avg level of GOV bond yields (~10y) be pulled towards ?
Towards the real LT trend GDP growth.
Expected required return on equity ?
Capital Appreciation ?
Δ(E/GDP) = (%ΔE – %ΔS) = the growth rate of earnings per share
Three approaches to economic forecasting ?
1) Econometric Modelling = Quant estimates as variables change.
2) Economic indicators = leading indicators: focus on identifying turning points
3) Checklist Approach = flexible¬ complex.
subjective, judgmental, no consistency of analysis across items or at different points in time.
Strengths and weaknesses of Economic Forecasting VS Econometric model forceasting ?
Inflation impact on asset classes ?
Cash = Inflation protected.
Bonds: a) if inf is as expected, already reflected in bonds ytm.
b) if inf > expected; lower real yields.
c) deflation = rising real rates; positive for IG bonds returns, negative for HY bonds returns.
Stocks: a) if inf is as expected, little effect.
b) if inf > expected; negative effect unless company has pricing power.
c) if inf < expected or deflation; decline in asset prices & sales prices (= decline in stock returns, then HY bonds’ returns).
Real Estate: Rents & Property value will increase with inflation.
Inverse is true.
Low inflation impact on cash, bonds, equities and real estate ?
How does Inflation behave with the business cycle ?
It’s pro-cyclical.
Expectations of inflation are also pro-cyclical.
Shape of the yield curve and stage of the business cycle ?
1) Initial Recovery = Steep YC
2) Early Expansion = Front Steepening, back-half flattening
3) Late Expansion = Flattening from long inwards
4) Slowdown = Flat to inverted
5) Contraction = Steepening.
Macroeconomic international linkages ?
1) (X-M) = (S-I) + (T-G)
2) ｜Current Account｜ = ｜Capital Account｜
As the current account changes, interest rates, fx-rates and asset prices, change to keep the capital account in balance.
Interest rate / Exchange Rate Linkages ?
A country can’t simultaneously:
1) allow unrestricted capital flows
2) maintain a fixed rate
3) pursue independent monetary policy.
In response to the projected cyclical decline in the Eastland economy and in private sector borrowing over the next year, Hadpret expects a change in the monetary and fiscal policy mix. He forecasts that the Eastland central bank will ease monetary policy. On the fiscal side, Hadpret expects the Eastland government to enact a substantial tax cut. As a result, Hadpret forecasts large government deficits that will be financed by the issuance of long-term government securities.
Discuss the relationship between the shape of the yield curve and the monetary and fiscal policy mix projected by Hadpret.
Approaches to set expectations for FIXED INCOME returns ?
1) DCF
2) Building Block approach = nominal rf + term premium + credit premium + liquidity premium.
rf is one period (holding period, typically 30d or 90d though); term premium = maturity premium = duration.
The liquidity premium can be estimated from the yield spread between the highest-quality issuer (typically a sovereign bond) and the next highest-quality large issuer of similar bonds (often a government agency). A widening yield spread indicates an increase in the liquidity premium and required rate of return.
Term premium drivers ?
1) Level dependent inflation uncertainty (uncertainty in level of π, pas e(inf) θ parce que ça c’est contenu dans le nominal rf)
2) Ability to hedge recession risk (if good ability, alors low TERM PREMIUM)
3) Supply/demand of bonds @ various maturities.
4) Cyclical effects = business cycle
Credit Premium (ig et hy) ?
IG: reflects risk of credit migration.
HY: reflects default risk.
Both are countercyclical.
Liquidity Premium ?
Lower for:
- new bonds priced near PAR
- large well-known issuers
- high quality
- simple structure
What are Credit Spreads ?
= Credit premiums + default rate.
Driven primarily by credit premiums & financial market conditions. Secondarily by default rate and consequently E(losses).
Based on Exhibit 2 and the anticipated effects of the monetary policy change, the expected annual return over a three-year investment horizon will most likely be:
A.lower than 2.00%. B.approximately equal to 2.00%. C.greater than 2.00%.
Criterias (numerical) that make debt repayment difficult for a country (emerging) ?
ABILITY
- Fiscal deficit/GDP > 4%
- Debt/GDP > 70%
- real GDP growth rate < 4%
- Current ACCOUNT Deficits > 4% GDP
- Foreign Debt > 50% GDP
- Foreign exchange reserves < 100.00% of short-term debt
Criterias (litteral) that make debt repayment difficult for a country ?
ABILITY
- Concentration of wealth = less diverse tax base.
- Dependence on commodity exports, cyclical industries.
- Restrictions on trade, capital flows, currency conversions.
- Reliance on Foreign Borrowing.
Risks that affect a country’s WILLINGNESS TO REPAY DEBT ?
Political & Legal risks:
- weak property rights.
- weak enforcement of contract law.
Approaches to setting expectations for Equity Returns ?
1) Historical statistics approach (weak?)
2) DCF: r = (D1/P0) + g. g is the LT GDP g rate.
3) Grinold-Kroner Model.
4) Risk Premium Approach.
Current or historical data to calculate returns ?
Si on calcule average historical equity risk premium, on prend historical average data de bond yield gov 10yr, et de equity mkt return.
Si on calcule expected equity risk premium, on prend expected equity mkt return, et current bond yield gov 10yr.
Grinold-Kroner Model (re = equity returns) ?
A Long-terme:
%ΔE = LT trend GDP growth rate.
%ΔS = 0
%ΔP/E = 0.
((D/P) - %ΔS) = income component of re.
Avec %ΔE = nominal earnings growth. !!!⚠!!!! The share of corporate profits in GDP is also included in this component (@ LT = fixed = 6%; can vary @ST).
The Grinold–Kroner model states that the expected return on equity is the sum of the expected income return (2.4%), the expected nominal earnings growth return (7.3% = 2.3% from inflation + 5.0% from real earnings growth) and the expected repricing return (−3.45%). The expected change in market valuation of −3.45% is calculated as the percentage change in the P/E level from the current 14.5× to the expected level of 14.0×: (14 − 14.5)/14.5 = −3.45%. Thus, the expected return is 2.4% + 7.3% − 3.45% = 6.25%.
Risk Premium approach (to determine re = equity returns) ?
a) ERP. On y rajoute rf pour avoir Re.
b) Equilibrium Approach (Singer-Terhaor):
fully integrated with global economy, or fully segmented.
ERP formula ?
Risk premium fully integrated market (Singer-Terhaor) ?
le ^G c’est juste une notation, on ne compute pas RP (risk premium) à la puissance.
SRg = Sharpe ratio de G = Risk premium de G / E(std dev) de G
Risk Premium fully segmented Market (Singer-Terhaor) ?
le ^S c’est juste une notation, on ne compute pas RP (risk premium) à la puissance.
Final Risk Premium de i local equity market ?
WORKS for commo, real estate, bonds, etc. NOT ONLY EQUITY, any asset class.
- w = [0.5; 0.75] for equity emerging MKTS.
- w = [0.75; 0.9] for Developed equity MKTS.
- w = lower for RE, higher for COMMOs.
E(Re) avec RPi ?
+ illiquidity premium
β de asset class ?
= Covar_return(i),return(glob) / [σ(Glob)]
Risks for Equity Emerging MKTS ?
Same as for fixed income +
generally less fully integrated = country risk (local economic & market factors exert greater influence on risk&return).
In addition to the economic, political and legal risks faced by fixed income investors, equity investors in emerging markets face corporate governance risks. Their ownership claims may be expropriated by corporate insiders, dominant shareholders or the government. Interested parties may misuse the companies’ assets. Weak disclosure and accounting standards may result in limited transparency that favors insiders. Weak checks and balances on governmental actions may bring about regulatory uncertainty, seizure of property or nationalization.
Historical returns estimation for future returns of REAL ESTATE ?
BAD!!
- trade infrequently
- appraisal data return series too smooth
- understates volatility and correlations
Cap rate approach to value Real estate ?
r = NOI / P
E(Re) real estate ? (discount rate)
= NOI/P + g.
g is the LT trend growth rate of GDP = NOI growth rate.
E(Re) ST real estate ? (discount rate)
= NOI/P + g - %ΔCapRate.
Cap rate = NOI/P
Low cap rate properties ?
high quality + long leases.
High cap rate properties ?
low quality + short leases.
Cap rates relationship with credits spreads, interest rates & credit availability ?
- Cap rates = positively related to interest rates, credit spreads (the two are negatively related so the effects somewhat offset each other).
- Cap rates = negatively related to credit availability.
E(Re) with risk premium for Real estate ?
E(Re) = one period rf + term premium + credit premium + equity premium + illiquidity premium.
term premium = long-lived assets.
credit premium = Tenants.
equity premium = Owner bears risks of ownership.
illiquidity premium = 2-4%
REITs relationship with equities and direct real estate on the long-run ?
@Short-term: REITs = correlated with equities
@LT: REITS = correlated with direct real estate
How to ““forecast”” exchanges rates (spoiler: on peut pas, c’est un random walk with a drift)
1) Trade flows (don’t exert a significant influence).
2) Relative PPP: S_p/b = π_p - π_b
PPP is a poor predictor of fx-rates in the short/medium term.
3) Competitiveness/sustainability of current account:
high CA deficits = ok if investment is high or CA deficit < 2% GDP. Else, not sustainable = upward pressure on interest rates.
4) Uncovered interest rate Parity: %ΔS_p/b = r_p - r_b.
Higher yielding currency should depreciate. But it does not hold IRL and carry trades are profitable.
5) Capital flows.
6) Portfolio Balance Approach.
Capital flows mechanisms on fx-rates ?
The deeper the central bank goes in the asset classes for its foreign currency PF, the more supportive they are of the local currency (because you can’t move out of the less liquid assets of the country as easily).
If p country offers a higher risk-adjusted E(return), the domestic currency will appreciate.
Public debt makes up the majority of Country C’s currency portfolio, which is the least supportive flow (or holding) to a currency. Public debt is less supportive because it has to be serviced and must be either repaid or refinanced, potentially triggering a crisis. Some types of flows and holdings are considered to be more or less supportive of the currency. Investments in private equity represent long-term capital committed to the market and are most supportive of the currency. Public equity would likely be considered the next most supportive of the currency. Debt investments are the least supportive of the currency.
Shrinkage estimation of population VCV matrix (to determine volatility) ?
= weighted average of correlations derived from multifactor models and historical correlations (or between two other sources)
De-smoothing ?
Re-state all returns as a weighted average of some true return and a previous observed return, then estimate volatility.
How to prepare an economic balance sheet for a client (+ implications for asset allocation) ?
Conventional A/L
+ extended A/L = PV(human capital) + PV(pension income) + PV(expected inheritances) + PV(future consumption)
PV(human capital) behaves ~70% like BOND, 30% like equity.
for an endowment, includes everything (restricted to scholarship funds, and unrestricted).
a) Strategic Asset Allocation ?
Core Exposure to Asset Classes + subclasses = tactical allocation, dynamic allocation, etc.
SAA is target allocations for each asset class. Upper and lower BOUNDS.
eg. SAA - target 10% +/- 5%
Rebalancing types ?
1) Calendar Rebalancing (periodic basis).
2) Percentage Rebalancing = % range around the target (standard (up graph) or proportional (down graph)).
Can rebalance back to:
- target
- range end
- midpoint between target & range end
Taxes influence on rebalancing ?
Asymmetric bands + wider.
Because a gain is taxable (outflow cash), but a loss offsets a gain so it’s a cash inflow.
Mean-Variance Optimization (Asset-only context) ?
Objective of MVO = Max E(Rp) for a given σp.
Requires estimates of E(R), E(σ) & CORREL_A,B.
Efficient frontier & cash: two types ?
Criticisms of MVO ?
1) Outputs (weightings) highly sensitive to small Δinputs (E(R), E(σ) & CORREL_A,B).
2) Allocations tend to be highly concentrated in a subset of available asset classes; especially when we have inconsistencies between Return and Risk.
Solution = Use Reversed Optimization or Black-Litterman (cf graph).
3) investors are concerned with more than just E(R) & σ² (eg. skewness & kurtosis).
4) sources of risk may not be diversified (use of factors VS asset classes).
5) Most portfolios exist to pay for a liability or consumption series (liability or goals-based asset allocations).
6) MVO is a single period framework that does not take account of trading/rebalancing/taxes.
What are the output and inputs of reversed optimization ?
Output = E(return) per class
Inputs = β per asset class + ERP + rf (CAPM) = risk and correlation/covar inputs.
Et Market Cap asset class weightings.
Historical data MVO, VERSUS
Reversed Optimization or Black-Litterman ?
MVO model using historical data inputs. MVO tends to result in asset allocations that are concentrated in a subset of the available asset classes.
Compared to the use of historical inputs, the Black–Litterman and Reverse-Optimization models most likely would be less concentrated in a few asset classes and less distant from the global weights.
Risk Budgeting ?
- Identify total desirable σp
- Allocate σp among risk classes efficiently.
The goal of risk budgeting is to maximize return per unit of risk. A risk budget identifies the total amount of risk and attributes risk to its constituent parts. An optimum risk budget allocates risk efficiently.
MCTR (marginal contribution to total risk) ?
marginal contribution to total risk
IF portfolio is optimal, the ratio of the excess return to the marginal contribution to total risk (MCTR) is equal to the Sharpe ratio for the portfolio
ACTR (absolute contribution to total risk) ?
absolute contribution to total risk
Liability relative AA vs Asset-only AA ?
SO-AA: Surplus Optimization asset Allocation (portfolio)
AO-AA: Asset Only Asset Allocation (portfolio)
H/RS-AA: Hedging return Seeking Asset Allocation (portfolio)
Hedging-Return seeking PF Asset Allocation ?
Goals-Based Asset Allocation ?
Heuristic Asset Allocation ?
- endowment model = Yale model = emphasizes active management of large allocations to non-traditional investments, seeking to earn illiquidity premiums.
Canada model is the same, but internally managed portfolio. - The 60/40 stock/bond heuristic is not an optimization model.
- 1/N rule asset allocation heuristic involves equally weighting allocations to assets; 1/N of wealth is allocated to each of N assets available for investment at each rebalancing date. All assets are treated as indistinguishable in terms of mean returns, volatility, and correlations.
On discount par le return afférent à la goal based allocation selected (highest return given a certain amount of risk, for a certain time horizon)
What’s the probability ratio ?
proba ratio = [E(R_P) – R_L] ÷ σ_P
governance ?
After-tax returns interests ?
After-tax returns equity ?
After-tax portfolio optimization with 4 asset classes ?
8 locations (taxable/non-taxable).
Affected by 8* [E(r), E(σ)].
UNAFFECTED BY COVARIANCES/CORRELATIONS.
change in goals ? Constraints ? Beliefs ?
A change in constraints relates to material changes in constraints, such as time horizon, liquidity needs, asset size, and regulatory or other external constraints. In this case, Young’s circumstances have changed; she is considering accepting the offer and retiring five years sooner than she originally anticipated.
A change in an investor’s personal circumstances that may alter her risk appetite or risk capacity is considered to be a change in goals. In this circumstance, Young’s risk appetite or risk capacity have not changed, whereas the time horizon associated with her goals has.
A change in the investment beliefs or principles guiding an investor’s investment activities is considered to be a change in beliefs. In this circumstance, Young’s guiding principles have not changed.
Sharpe ratio and Efficient frontier ?
The Sharpe ratio is the slope of the line drawn from the risk-free rate to a particular portfolio. The two portfolios of interest are the policy portfolio and the TAA portfolio because both are indicated as being efficient. The diagram to the right indicates that the policy portfolio/risk-free combination has a higher slope than the TAA/risk-free combination.
