Lecture 13: Portfolio Management

Portfolio Management Overview

Introduction to Portfolio Management

• Jake Xia introduces the topic of portfolio management, emphasizing a practitioner's perspective rather than a purely theoretical one.

• The course has been taught since 2012 and recorded for MIT OpenCourseWare in 2013, becoming one of the top five viewed courses.

Lecture Outline

• The lecture will cover basic concepts of portfolio construction, the endowment model, and portfolio theory with simple illustrations.

• Discussion on limitations of traditional theories proposed in the 1950s and advancements made to address these shortcomings.

Understanding Portfolio Construction

Initial Exercise

• An exercise is introduced where participants create a hypothetical investment portfolio with $10,000.

• Participants are encouraged to use intuition without overthinking their choices while considering various investment criteria.

Key Considerations for Investment Choices

• Important questions include objectives for returns, time horizons, loss tolerance, and ability to predict market movements.

• Participants must also consider diversification levels and how to size each investment within their portfolios.

Decision-Making Process in Investments

Factors Influencing Investment Selection

• The decision-making process involves selecting markets and instruments from which investments will be chosen.

• Emphasis on data collection, signal derivation, model building, factor consideration, predictions assembly into strategies.

Capital Allocation Strategies

• Discusses optimizing capital allocation across selected investments while managing associated risks effectively.

Risk Assessment in Portfolio Management

Understanding Risk Measurement

• Volatility is initially used as a measure of risk; however, it is noted that this may not be the best indicator for assessing risk profiles.

Investment Portfolio Analysis and Strategies

Overview of Investment Portfolios

• The discussion begins with an analysis of a portfolio consisting of 50% in S&P straddles (an options strategy), 30% in VIX-related ETFs, and 20% in dollar index futures, indicating familiarity with option strategies.

• Another portfolio is highlighted, featuring 70% bonds and 30% stocks, which deviates from the typical equity-heavy allocations seen in standard portfolios like the 60/40 or 70/30 models.

• A simple portfolio structure is noted: 100% S&P representing average market performance, alongside various combinations of cash and US Treasury bonds.

Trends in Investment Choices

• Notably, there has been a shift away from cryptocurrencies compared to previous years; this year’s focus appears to be on index investing through ETFs.

• Cash investments are viewed as low volatility options with minimal returns. Current interest rates around 4%-5% make cash a reasonable choice for capital preservation.

Understanding Risk and Return

• The speaker emphasizes that large-cap tech stocks significantly influence index performance; being long on QQQ or S&P means exposure to these high-performing stocks.

• Private equity and venture capital are discussed as alternatives to public market investments. These involve higher risk but can yield greater returns over time due to their illiquid nature.

Objectives and Constraints in Portfolio Management

• Maximizing return while minimizing risk is crucial; investors should aim for a favorable return-to-volatility ratio while considering personal constraints such as ethical investment choices (e.g., avoiding fossil fuels).

• Liquidity needs must also be factored into investment decisions. For instance, if an investor requires access to funds regularly, they may avoid illiquid assets.

Inflation and Relative Purchasing Power

• Inflation impacts purchasing power significantly; wealth isn't just about total dollars but what those dollars can buy relative to others.

• The importance of absolute returns versus beating benchmarks (alpha generation) is highlighted as a measure of investment skill.

Asset Liability Matching

• Proper asset-liability matching is essential for maintaining financial stability during downturns. Past experiences have shown that mismatches can lead to severe budget cuts or operational disruptions when facing drawdowns.

Investment Time Horizon and Portfolio Management

Understanding Time Horizons in Investment

• The speaker emphasizes the importance of defining the time horizon for an investment portfolio, questioning whether it is intended for one day, a year, or ten years. This distinction is crucial as it influences investment strategies.

• Despite the focus on long-term returns for serious investors, short-term performance metrics often dominate discussions due to peer pressure and career risks faced by investment managers.

Personal Income and Spending Patterns

• The speaker illustrates how personal income peaks around mid-career (approximately age 50), correlating with increased spending during this phase. Younger individuals typically earn and spend less.

• As individuals approach retirement, both income and spending decrease. This necessitates careful planning of personal portfolios based on one's life stage.

Endowment Portfolios: Long-Term Strategies

• Endowments operate with a perpetual portfolio aimed at long-term growth. They must generate returns that outpace inflation (targeting an 8% nominal return), which can be challenging in low-interest environments.

• Universities increasingly rely on endowment returns to fund their operating budgets, with about 40% of these budgets sourced from endowment spending due to declining research grants.

Investment Strategies Employed by Endowments

• A variety of investment strategies are utilized by endowments including cash, government bonds, corporate bonds, hedge funds focused on macro markets, quantitative funds employing statistical arbitrage techniques, and private equity investments.

• Real asset investments such as real estate and natural resources are also part of the strategy mix alongside newer assets like cryptocurrencies and intellectual properties.

Focus on Manager Selection in Investments

• The endowment model prioritizes hiring external managers who specialize in both public and private investing across various domains while focusing on achieving absolute returns relative to benchmarks.

• The classic portfolio construction problem involves determining optimal investment percentages based on return objectives, loss tolerance, predicted volatility, and correlation among investments to maximize overall portfolio return while minimizing risk.

Portfolio Construction and Risk Management

Understanding Portfolio Construction

• The portfolio construction problem is fundamentally about sizing investments across various levels, closely related to asset allocation issues.

• Asset classes can be grouped into 10 to 15 categories; however, this complicates optimization. Simplifying to 3 to 5 major risk factors (e.g., equity, bonds, inflation) can ease the process.

• Factor analysis allows for grouping investments in multiple ways, potentially expanding to hundreds of factors for better optimization.

Risk Management in Portfolios

• The risk management aspect mirrors portfolio construction as it also involves sizing but adds considerations like avoiding concentration and managing illiquidity.

• Concentration risks and unacceptable losses must be addressed when constraining a portfolio.

Portfolio Theory Basics

• In a two-asset model, weights (w1 and w2) must sum to one. The overall portfolio return (Rp) is derived from the weighted returns of each asset.

• When assets are perfectly correlated (rho = 1), the portfolio volatility simplifies to a linear combination of individual volatilities.

Special Cases in Correlation

• Perfect negative correlation (rho = -1) leads to distinct branches in solutions for portfolio volatility based on asset weightings.

• If one asset has zero volatility (sigma1 = 0), the overall portfolio volatility becomes dependent solely on the second asset's weighting.

Capital Allocation Line and Efficient Frontier

• A riskless asset or cash with zero volatility influences the capital allocation line, which represents optimal portfolios combining risk-free assets with existing portfolios.

• The efficient frontier illustrates optimal return versus volatility trade-offs; points along this line represent improved portfolios through effective risk management strategies.

• The Sharpe ratio quantifies performance by comparing excess return over volatility. It relates alpha and beta concerning benchmarks while considering correlation effects.

Portfolio Theory and Risk Management

Leveraging Portfolio for Increased Returns

• Discusses the concept of adjusting portfolio weights by borrowing against a risk-free asset, allowing for increased investment in a risky asset.

• Highlights that this strategy is common in risk parity portfolios, aiming to balance risk contributions from different assets.

Understanding Risk Parity Portfolios

• Compares traditional 60/40 equity-bond portfolios with risk parity approaches, emphasizing equal risk contribution from both asset classes.

• Introduces the importance of rebalancing within portfolios to maintain desired risk levels and optimize returns.

The Free Lunch of Diversification

• Illustrates a two-year scenario with two assets where one doubles and then halves while the other does the opposite, demonstrating how rebalancing can enhance returns.

• Explains that without rebalancing, despite diversification, the portfolio could yield zero return after two years due to dominance by one asset.

Rebalancing Strategy Considerations

• Questions whether maintaining equal weighting is justified based on ongoing projections of performance and risks associated with each asset.

• Emphasizes that if an investor believes in equal probability distributions for both assets, they should continue to rebalance their portfolio accordingly.

Limitations of Modern Portfolio Theory (MPT)

• Shifts focus to MPT's limitations, noting its reliance on capital market assumptions which can be sensitive to small changes.

• Points out that MPT often requires artificial constraints due to its numerous potential solutions, making practical application challenging.

Alternative Approaches to Portfolio Construction

• Introduces research into alternative methods like gain-loss ratios as a response to MPT's shortcomings.

• Critiques volatility as a measure of risk; highlights how it may not accurately reflect true investment scenarios depending on market conditions.

Investment Sizing and Optimization

Understanding Expected Gain and Loss

• The speaker introduces a new method for comparing investments using expected gain (G) and expected loss (L), emphasizing that these are absolute numbers.

• The formula for investment sizing is presented as G minus L divided by G plus L, which aligns with Kelly's criteria in binary betting, providing a bounded range from -1 to +1.

• A coin flip example illustrates how to calculate expected gain and loss based on probabilities of outcomes, making the concept more tangible.

• In continuous scenarios, the use of probability density functions is necessary to derive expected gain and loss through integration over all possible outcomes.

• The optimization problem shifts focus from return versus volatility to maximizing gains while minimizing losses, highlighting that higher volatility can be beneficial in certain contexts.

Practical Application in Investment Games

• Participants are encouraged to track daily gains and losses during an investment game, summarizing results at the end to derive G and L values.

• A discussion arises about selecting high-volatility stocks; while they may yield high returns, they also affect the overall quality of investment ratios negatively.

• Students inquire about estimating expected gains; the speaker clarifies that it involves similar effort as traditional methods but focuses on half distributions instead of whole ones.

Modern Portfolio Theory Challenges

• The speaker addresses predicting future market behavior based on past data, noting human tendencies to extract patterns complicate this process due to behavioral factors.

• A typical thinking process in finance is outlined: observing data, quantifying it, recognizing patterns, building models for predictions, and iterating for calibration.

• Human behavior adds complexity to financial modeling since historical patterns may not repeat consistently due to crowding effects among investors.

Understanding Crowd Behavior and Market Dynamics

The Swaying of the Millennium Bridge as a Metaphor for Market Behavior

• The Millennium Bridge opened in 2000, where initial swaying caused people to synchronize their steps, illustrating how panic or greed can amplify market movements.

• This synchronization leads to unpredictable crowd behavior, akin to market bubbles and crashes driven by collective emotions.

Feedback Loop in Crowd Modeling

• A feedback loop is introduced with agents' actions (S), observations (O), and an amplifier (A), showing how agent actions influence observations and vice versa.

• Under normal conditions, agents are rational; however, increased volatility causes them to react more strongly, leading to a reactive state among the crowd.

Order Parameter and Synchronization

• The order parameter measures synchronization within a crowd: close to 1 indicates full synchronization while near 0 indicates disarray.

• Increased reactivity among agents raises the order parameter, indicating greater synchronization; higher random noise decreases this likelihood.

Simulation of Bubble Dynamics

• Simulations show that external shocks can lead to peaks in reactive states among agents, resulting in synchronized behavior and unstable systems.

• If external shocks persist at high levels, all agents may enter a reactive state leading to significant instability—characteristic of market bubbles.

Behavioral Finance vs. Fundamental Analysis

• Understanding bubble formation is crucial for portfolio management; it transcends traditional fundamental analysis focusing on economic indicators or company performance.

Power Law Distribution: Insights into Wealth Inequality

Characteristics of Power Law Distribution

• Power law distribution describes phenomena where a small percentage of agents hold the majority share—illustrated by wealth distribution where 20% possess 80% of wealth.

• This scale-free property persists across various domains such as venture capital returns and city sizes due to network effects favoring larger nodes.

Examples and Implications of Power Law

• Notable examples include wealth concentration among top-tier venture capital funds and city growth dynamics influenced by connectivity preferences in networks.

Nature vs. Social Phenomena: Distribution Differences

• Unlike social phenomena exhibiting power law distributions, natural traits like height follow normal distributions centered around average values without skewed extremes.

Causes Behind Skewed Wealth Distribution

• The feedback mechanisms observed earlier contribute significantly to power law distributions in social contexts but are absent in natural growth processes.

Understanding Heterogeneous Agents in Financial Markets

The Role of Different Agents

• Financial markets are influenced by various agents with differing powers, leading to the concept of heterogeneous agents.

• Wealth concentration occurs as richer agents can leverage their advantages to gain more power over time.

Power Law Distributions

• The relationship between random variables and their ranks is illustrated through charts, emphasizing how power law distributions differ from normal distributions.

• Key parameters of power law distributions (Cumulative Density Function and Zipf's Law) are interconnected, influencing agent behavior in financial markets.

Feedback Mechanisms and Market Dynamics

• Agents with greater power can amplify their influence when they make successful bets based on market observations, contributing to wealth accumulation.

• The emergence of "super agents" may destabilize the system; this raises questions about order and stability within financial contexts.

Key Takeaways for Portfolio Management

Investment Strategies

• Effective portfolio management requires clear objectives, understanding loss tolerance, and proper sizing of investments.

• Diversification is essential but must be coupled with regular rebalancing to manage risk effectively.

Understanding Risk

• Traditional metrics like Sharpe ratio may not adequately capture investment skill or provide effective sizing strategies; a new ratio is suggested for better assessment.

• Recognizing powerful entities such as government agencies (e.g., the Fed) is crucial for understanding market dynamics and potential influences on investments.

Questions on Risk Measurement

Tax Considerations in Investments

• Taxes on capital gains significantly impact investment decisions; family offices often prefer long-term private equity due to tax benefits.

Measuring Risk Effectively

• Expected loss should be prioritized over worst-case scenarios when measuring risk; understanding potential downside protection is vital for investment strategy.

Stop Loss Strategies

• Stop losses are typically set closer than expected losses to mitigate risks quickly; they aim to create asymmetric payouts that protect against significant downturns.

Investment Strategies and Hedge Fund Management

Asymmetric Optionality in Investments

• The concept of asymmetric optionality is introduced, where investors are encouraged to take risks on investments that have a favorable probability distribution.

• Investors should consider the worst-case loss against its low probability when evaluating potential investments, weighing the significant upside against possible downsides.

Hedge Fund Dynamics

• Discussion on hedge funds highlights their profit-sharing model, where managers benefit from profits but do not bear the downside risk, effectively acting as if they hold a free call option.

• This dynamic creates an incentive for many individuals to aspire to become hedge fund managers due to the lucrative nature of this arrangement.

Performance Tracking and Learning Opportunities

• Emphasis on tracking G and L (gains and losses) numbers post-November 15 as part of performance evaluation in investment activities.

• A poll is suggested regarding preferences for learning formats: paper trading games versus lectures focused on research processes like data collection, signal generation, backtesting, portfolio optimization, or managerial skills such as fundraising and team building.