Portfolio measurement is usually performed by the calculation of returns. But returns alone do not describe the performance of a portfolio. It is essential to calculate portfolio risk in order to understand the degree of risk a portfolio manager takes in order to achieve those returns. In this article we will explore two different timings for calculating portfolio risk: ex-ante and ex-post along with the uses for each.
Ex-ante risk is calculated based on the securities in a portfolio prior to the start of the measurement period. For a portfolio that which contains a single asset class, such as a Canadian equities portfolio, the risk measure can be as simple as beta, the sensitivity of the portfolio relative to the index. Further complexity can be added by breaking down the portfolio by sector or market capitalization. When determining the risk of a pension fund portfolio with multiple asset classes, the popular approach is to use Value-at-Risk (VaR).No matter what measure we use, it will be based on historical volatilities and correlations of the individual securities in the portfolio.
Measuring risk at the start of the period can perform a multitude of functions. For a single asset class portfolio, such as a Canadian equities portfolio, the owner of the assets can measure how much risk the portfolio manager is taking both on an absolute basis and relative to the benchmark. For example, if one relies on an asset manager to outperform the S&P/TSX Composite, then one can measure how much the portfolio expects to vary from that index. By monitoring the risk, one may avoid uncomfortable discussions at the end of the period determining how the portfolio could have underperformed the benchmark by such a large magnitude.
Another use for ex-ante risk measurement could be the determination of asset mix for a fund. Central banks, pension funds and asset managers use Value-at-Risk (VaR) to calculate a probability distribution for various asset mixes over time periods such as one day, one month or one year. Based on the output from such a model, one may derive useful statistics such as the probability of a negative return in the portfolio cannot be greater than 1%.Thus, the board overseeing the management of the portfolio can be assured that of a close match between assets and liabilities.
A third use of ex-ante risk is the verification of fund strategy. For example, a DC plan may offer several different investment vehicles with styles denoted by terms such as conservative, aggressive, or ultra aggressive. With the use of ex-ante risk one can formulate an opinion as to whether these funds live up to their descriptions.
In each of these examples, the use of ex-ante risk is crucial to monitoring risk, formulating optimal asset allocation decisions and determining appropriateness of risk.
In contrast, ex-post risk is the measurement of experienced volatility and performance attribution at the end of the measurement period. The calculated volatility of the daily (or weekly or monthly) fund value can then be used to determine the Sharpe ratio, which is a measure of return relative to risk. As with ex-ante risk, ex-post risk can be calculated on an absolute or benchmark-relative basis.
Leading fund managers recognize that use of both ex-ante and ex-post risk can add tremendous value to portfolio oversight. For example, if a portfolio manager chooses to invest in volatile internet equities yet they go through a period of relative quiet, then the ex-ante risk reports will show expectations of portfolio volatility even if was not realized. On the other hand, an investment in low volatility financial equities in 2007-2008 would seem to be benign from the ex-ante reports although high volatility was realized from the investment. With the use of both ex-ante and ex-post risk, these leading managers can measure the risk exposures of the portfolio and how it can expect to perform under various stress scenarios at the beginning of a period. Then, at the end of the period, they can measure actual portfolio performance, break down the sources of return relative to the risks taken, and determine whether the portfolio was paid for those risks. By using both ex-ante and ex-post risk measurement, fund managers are presented with a more complete picture of risk and returns.