Financial Education

Portfolio monitoring includes monitoring changes in the characteristics of individual securities or asset classes. Over time, underlying average returns, volatility and correlations with other asset classes can change. Such changes alter the appropriate mix of assets for meeting an investor’s objectives and constraints. If the changes are perceived as temporary, they may also present opportunities to make tactical changes.

The market and economic environment also require monitoring. Of particular importance can be the yield curve, market risk premia, central bank policy, and unusual deviations from normal relationships between securities or asset classes.

Using the Monte Carlo method to estimate Value at Risk (VaR) produces a set of random outcomes reflecting the effects of particular sets of risks. Each set of outcomes is based on a probability distribution for each variable of interest. The distributions for each variable can be normal or non-normal.

Monte Carlo simulations are frequently the only method that provides a practical means to generate necessary risk management information. However, it can become quite a hog of computer resources for large portfolios.

A number of factors affect performance evaluation for hedge funds, particularly with respect to using the Sharpe ratio to measure risk-adjusted return.

Starting with return, typically monthly returns are compounded to an annualized rate of return. However, entry and exit opportunities may be permitted only quarterly or even less frequently. In addition, some measures of downside risk such as the maximum drawdown are not compounded. Measures comparing return (compounded) and drawdown (not compounded) may not fully reflect the risk/return profile.

The Sharpe ratio is defined as:

• Numerator is the difference between annualized return and the annualized risk-free rate
• Denominator is the annualized standard deviation of returns

The Sharpe ratio increases proportionately with the square root of time, and is not appropriate when returns are asymmetrical. In particular, the Sharpe ratio tends to be overestimated when returns are serially correlated or assets are illiquid. Furthermore, the correlations between the fund and an investor’s other portfolio assets are not considered.

There are a number of ways managers can “game” the Sharpe ratio, including:

• Lengthening the measurement interval
• Compounding monthly returns but calculating standard deviations without compounding
• Writing out of money put or call options to produce asymmetric returns
• Smoothing returns
• Using swaps to eliminate extreme outlying returns

In part because of these deficiencies, the Sharpe ratio has not been found to be a good predictor of hedge fund returns.

Venture capital investors typically receive convertible preferred stock when funding companies. If the company is forced to liquidate, the preferred shares will have precedence in receiving funds. The company founders will hold a residual stake of common shares. If subsequent funding rounds are provided, each round is typically senior to the previous.

Typically the preferred investors must see a return of capital and also some investment return (often a total of 2x the capital contributed) before any cash can be returned to common shareholders. This provides the founders with an incentive to earn the return required by their investors so they can reap their own rewards.

The preferred shares are also convertible to common shares, which is typically done when a corporate action (merger or IPO) creates liquidity for the common shares and an opportunity to cash out.

International bond managers can seek excess return from a variety of sources:

1. Bond market selection - choosing the best country in which to invest
2. Currency selection - deciding whether to hedge or retain currency risk
3. Duration/yield curve management - getting the most favorable returns within the selected market
4. Sector selection - choosing among government, corporate, local currency or dollar-denominated bonds
5. Issuer credit analysis - being able to identify improvement or deterioration in advance of changes in rating
6. Benchmark mismatches - investing in markets that are not included in the benchmark index

Changes in interest rates affect both the reinvestment rate earned on portfolio income (directly) and the value of the portfolio (inversely.) An immunization strategy is designed to lock in total return over a specified time horizon by creating a portfolio in which the two total return factors exactly offset each other.

To immunize a portfolio over a single period, the portfolio must have duration equal to the investment horizon and an initial present value of cash flows equal to the present value of the future liability.

Since yield changes over time will result in duration changes other than those caused solely by the passage of time, it is necessary to periodically rebalance an immunized portfolio. This must be done only if the benefits outweigh the costs. Some transaction costs must be borne in order to avoid duration mismatch, but some duration mismatch is needed to avoid transaction costs.

Typically the rebalancing will restore the dollar duration equivalency to the time horizon. Dollar duration is the product of Duration X Portfolio Value X 0.01.

To rebalance the dollar duration requires three steps:

1. Calculate present dollar duration based on the prevailing yield curve and time to maturity
2. Calculate the rebalancing ratio by dividing the original dollar duration by the new dollar duration. This can be expressed as the percentage change for each position by subtracting one and converting the result into percentage terms.
3. The new market value of the portfolio multiplied by the percentage change is the amount of cash that will be needed for rebalancing.

Investor Specific Concerns

Strategic asset allocation starts with the investor’s objectives, constraints, net worth and attitudes toward risk. A risk tolerance function can be applied to determine the investor’s specific risk tolerance.

Capital Market Situation

The current state of the market must be identified, and some procedure used to generate a set of expected returns, risks and correlations between asset classes.

Joint Investor-Market Relationship

Given the investor’s situation and the capital market outlook, an optimizer can be used to determine an allocation suited to the investor’s risk tolerance. The asset mix can be selected, and actual returns fed back into the process as feedback.

Purchasing Power Parity

Purchasing power parity (PPP) is based on the belief that movements in exchange rates should offset any difference in inflation between two economies. Over longer time horizons, purchasing power parity does tend to hold, in part because governments take the concept seriously and act to control relative inflation rates. In the short term, however, other factors such as capital flows can overwhelm the impact of PPP.

Relative Economic Strength

This concept focuses on investment flows rather than trade flows. It suggests that strong economies attract capital, which causes the currency to appreciate. Foreign investors must always weigh whether the higher yield offsets the risk of inflated currency values. Relative Economic Strength indicates how currencies should respond to economic news, but does not imply a “true” currency value. Because of this, many investors use it in conjunction with PPP to form a complete theory of interest rate movements.

Capital Flows

Focuses on expected capital flows, particularly with respect to long-term capital investments. Capital flows can reverse the normal relationship between currencies and short-term interest rates, as the stimulative effect of a lower interest rate may outweigh the lower yield.

Savings-Investment Imbalances

This theory explains currency movements in terms of savings relative to capital investment. If domestic savings are insufficient to cover capital investments, the remainder must come from foreign sources. To the extent that foreign investment is not offset by a trade imbalance, the currency must rise.

Suppose an analysis finds that semiconductor sales have a strong relationship with semiconductor equipment sales. There are three possible explanations for the relationship:

1. Higher semiconductor sales result in a need for more semiconductor equipment (A predicts B)
2. Having more equipment to make semiconductors results in higher sales (B predicts A)
3. Some other factor (such as economic conditions) results in higher need for both semiconductors and semiconductor equipment (C predicts A and B)

Without investigating and modeling the underlying linkages, using correlations relationships in a prediction model can lead to significant errors.

Equally important, suppose that no correlation is found between semiconductors and semiconductor equipment. They may still have a strong (but nonlinear) relationship that should be considered. Such relationships may be found by using multiple regression techniques.

For individuals drawing on retirement funds, a 4% withdrawal rate is generally recommended to result in only a small chance of the portfolio running out of money. In the October 2007 Journal of Financial Planning Spitzer, Strieter and Singh simulate thousands of 30-year periods to assess the overall probability of running out of funds.

They find that a standard 50/50 split between stocks and bonds can allow for a 4.4% withdrawal rate with just a 10% chance of depleting funds. Withdrawal rates of up to 6% can be supported with stock allocations of 75% or more.