Introduction

• There are policies that limit an organization’s willingness to assume risk in order to generate profit. To manage this, many organizations formally budget risk usage through asset allocation methods (e.g., mean-variance optimization techniques). The result yields a blend of assets that will produce a level of expected returns and risk consistent with policy guidelines.
• Risk, in financial institutions, is frequently defined as value-at-risk (𝑉𝑎𝑅). 𝑉𝑎𝑅 refers to the maximum dollar earnings/loss potential associated with a given level of statistical confidence over a given period of time. 𝑉𝑎𝑅 is alternatively expressed as the number of standard deviations associated with a particular dollar earnings/loss potential over a given period of time.
• Asset managers also use tracking error to gauge their risk profile relative to a benchmark. In the case of asset managers, clients typically assign a benchmark and a projected risk and return target vis a vis that benchmark. The risk budget is often referred to as tracking error, which is defined as the standard deviation of excess returns (the difference between the Portfolio’s returns and the benchmark’s returns), i.e.
• The ability to take risks away from the index is often referred to as active management.
  • Tracking error is used to describe the extent to which the investment manager is allowed latitude to differ from the index.
  • For the owner of capital, the 𝑉𝑎𝑅 associated with any given asset class is based on the combination of the risks associated with the asset class and the risks associated with active management.
• Risk management structure can be seen as a three legged stool. The 3 legs being –
  • The Risk Plan
  • The Risk Budget
  • The Risk Monitoring process

The Risk Plan

A risk plan should include 5 guide posts –

1. The risk plan should set expected return and volatility (e.g., 𝑽𝒂𝑹 and tracking error) goals for the relevant time period. The risk plan should use scenario analysis to explore those kinds of factors that could cause the business plan to fail (e.g., identify unaffordable loss scenarios) and strategic responses in the event these factors actually occur. The risk plan helps ensure that responses to events – be they probable or improbable – are planned and not driven by emotion. The planning process should explore the many “paths to the long term” and prepare the organization, and its owners and managers, for the bumps along the way.
2. The risk plan should define points of success or failure. Examples are acceptable levels of return on equity (𝑅𝑂𝐸) or returns on risk capital (𝑅𝑂𝑅𝐶). For the purposes of the planning document, risk capital might be defined using value-at-risk (𝑉𝑎𝑅) methods. Separate 𝑉𝑎𝑅 measures for different time horizons (monthly, quarterly, annually) should be explored. The 𝑉𝑎𝑅 (or risk capital) allocated to any activity should be sized in such a way that the exposures and upsides associated with the activity are at appropriate levels.
3. The risk plan should paint a vision ofhow risk capital will be deployed to meet the organization’s objectives. For example, the plan should define minimum acceptable 𝑅𝑂𝑅𝐶s for each allocation of risk capital. The plan should also explore the correlations among each of these 𝑅𝑂𝑅𝐶s as well to ensure that the consolidated 𝑅𝑂𝑅𝐶 yields an expected 𝑅𝑂𝐸, and variability around such expectation, that is at acceptable levels. Finally, the plan should also have a diversification or risk decomposition policy.
4. A risk plan should define the bright line between those events that cause ordinary damageand those that inflict serious damage. Strategic responses should exist for any threatening event – even if such events are low-probability situations. The risk plan should identify those types of losses that are so severe that insurance coverage (e.g., asset class puts) should be sought to cover the downside. For example, firms or plans with large equity holdings could face material loss and earnings variability in the event of protracted and substantial stock market losses. In this case, the risk plan should explore the potential merits of financial insurance (e.g., options on broad market indexes).
5. The risk plan should identify critical dependencies that exist inside and outside the organization, and describe the responses if there are breakdowns in such dependencies. Examples of critical dependencies include reliance on key employees and important sources of financing capacity. The risk plan should explore how key dependencies behave in good and bad environments. Frequently, very good or very bad events occur simultaneously with other material events. For example, in case of a pension plan, periods of economic downturn could coincide with lower investment performance, acceleration of liabilities, and a decreased capacity of the contributing organization to fund the plan. For this reason, scenario planning for the pension plan should explore what other factors affect the pension plan’s business model in both good and bad environments and develop appropriate steps to help the plan succeed.
6. An effective risk plan requires the active involvement of the organization’s most senior leadership. This involvement creates a mechanism by which risk and return issues are addressed, understood, and articulated to suppliers of capital (owners or beneficiaries), management, and oversight boards. The existence of a risk plan makes an important statement about how business activities are to be managed. It suggests that a higher standard of business maturity is present.

The Risk Budget

• The risk budget (often called asset allocation) should quantify the vision of the plan. Once a plan is put into place, a formal budgeting process should exist to express exactly how risk capital will be allocated such that the organization’s strategic vision is likely to be realized. The budget helps the organization stay on course with respect to its risk plan.
• For each allocation of risk budget, there should be a corresponding (and acceptable) return expectation. For each return expectation, there should be some sense of expected variability around that expectation. These should be consistent with the organization’s strategic objectives and risk tolerances.
• For risk budgets, a risk “charge” – defined as 𝑉𝑎𝑅 or some other proxy for “risk expense” – can be associated with each line item of projected revenue and expense. An 𝑅𝑂𝑅𝐶 can be associated with each activity as well as for the aggregation of all activities. 𝑅𝑂𝑅𝐶 must exceed some minimum acceptable levels so that the organization is sufficiently compensated – in cost/benefit terms – for the expenses and/or risks associated with generating revenues.
• Risk budgets should show a contribution to overall risk capital usage by activity. For example,  standard mean-variance optimization methods produce estimates of weights to be assigned to each asset class, in addition to overall estimates of portfolio standard deviation and the marginal contribution to risk from each allocation.
• 𝑅𝑂𝑅𝐶 should be estimated over all time intervals that are deemed relevant. For example, if investment boards meet monthly and are likely to react to short-term performance, monthly 𝑅𝑂𝑅𝐶 is relevant. Hence, management must define the time horizons over which risk budget allocations are to be spent and over which 𝑅𝑂𝑅𝐶 should be measured.
• Risk budgeting incorporates elements of mathematical modeling. Even though quantitative models are prone to failure at the worst possible moments, but they are still important for risk budgeting. Variances from budget can result from organization-specific factors (e.g., inefficiency) or completely unforeseen anomalies (e.g., macroeconomic events, wars, weather, etc.).

The Risk Monitoring Process

• Variance monitoring is a basic financial control tool. Since revenue and expense dollars are scarce, monitoring teams are established to identify material deviations from target. Unusual deviations from target are routinely investigated and explained as part of this process.
• Monitoring controls ensure that risk capital is used in a manner consistent with the risk budget.
• Material variances from risk budget are threats to the investment vehicle’s ability to meet its 𝑅𝑂𝐸 and 𝑅𝑂𝑅𝐶 targets. If excessive risk is used, unacceptable levels of loss may result. If too little risk is spent, unacceptable shortfalls in earnings may result.
• Risk monitoring is required to ensure that material deviations from risk budget are detected and addressed in a timely fashion.

Risk Consciousness

• There is an increasing sense of risk consciousness among and within organizations. This risk consciousness derives from several sources-

1. Banks that lend to investors increasingly care about where assets are placed.
2. Boards of investment clients, senior management, investors, and plan sponsors are more knowledgeable of risk matters and have a greater awareness of their oversight responsibilities. Especially as investments become more complicated, there is an increasing focus to ensure that there is effective oversight over asset management activities – either directly by an organization or delegated to an outside asset manager.
3. Investors themselves are expected to have more firsthand knowledge about their investment choices. Perhaps this has been driven by the notoriety of losses incurred by Procter & Gamble, Unilever, Gibson Greeting Cards, Orange County (California), the Common Fund, and others. Further, in the asset management world, asset managers increasingly must be able to explain, ex ante, how their products will fare in stressful environments.

Objectives Of Risk Management Units (RMUs)

• The RMU gathers, monitors, analyzes, and distributes risk data to managers, clients, and senior management in order to better understand and control risk. This mission requires that the RMU deliver the right information to the right constituency at the right time.
• The RMU helps the organization develop a disciplined process and framework by which risk topics are identified and addressed. The RMU is part of the process that ensures the adoption and implementation of best risk practices and consistency/comparability of approach and risk consciousness across the firm. As such it is a key promoter of an organization’s risk culture and internal control environment.
• To be vibrant, the RMU must be actively involved in setting and implementing the risk agenda and related initiatives.
• The RMU watches trends in risk as they occur and identifies unusual events to management in a timely fashion. It is more meaningful to identify a trend before it becomes a large problem.
• The RMU is a catalyst for a comprehensive discussion of risk related matters, including those  matters that do not easily lend themselves to measurement. For example, the RMU should be actively involved in the identification of and organizational response to low-probability yet high-damage events.
• The RMU is one of the nodes of managerial convergence – a locus where risk topics are identified, discussed, and disseminated across the organization and clients. In so doing, it helps promote enhanced risk awareness together with a common risk culture and vocabulary.
• As a part of the internal control environment, the RMU helps ensure that transactions are authorized in accordance with management direction and client expectations. For example, the RMU should measure a portfolio’s potential (i.e., ex ante) tracking error and ensure that the risk profile is in consonance with expectations.
• Together with portfolio managers and senior management, the RMU identifies and develops risk measurement and performance attribution analytical tools. The RMU also assesses the quality of models used to measure risk. This task involves back testing of models and proactive research into “model risk”.
• The RMU develops an inventory of quality and credible risk data for use in evaluating portfolio managers and market environments. It should be synthesized, and routinely circulated to the appropriate decision makers and members of senior management.
• The RMU provides tools for both senior management and individual portfolio management to better understand risk in individual portfolios and the source of performance. It establishes risk reporting and performance attribution systems to portfolio managers and senior management. In the process, the RMU promotes transparency of risk information.
• The RMU should not manage risk, which is the responsibility of the individual portfolio managers, but rather measure risk for use by those with a vested interest in the process. The RMU cannot reduce or replace the decision methods and responsibilities of portfolio managers. It also cannot replace the activities of quantitative and risk support professionals currently working for the portfolio managers. Trading decisions and the related software and research that support these decisions should remain the responsibility of the portfolio managers and their support staffs. The RMU measures the extent to which portfolio managers trade in consonance with product objectives, management expectations, and client mandates.

Risk Monitoring And Consistency

1. Forecast of tracking error should be consistent with the target –
   • The forecasted tracking error is an estimate of the potential risk that can be inferred from the positions held by the portfolio derived from statistical or other forward-looking estimation techniques.
   • An effective risk process requires that portfolio managers take an appropriate level of risk (i.e., neither too high nor too low) consistent with client expectations. This forecast should be run for each individual portfolio as well as for the sum of all portfolios owned by the client.
   • Tracking error forecasts should be compared to tracking error budgets for reasonableness. Policy standards should determine what magnitude of variance from target should be deemed so unusual as to prompt a question and what magnitude is so material as to prompt immediate corrective action. In this manner, unusual deviations across accounts will be easier to identify.
2. Risk capital should be properly allocated to the expected areas –
   • In financial variance monitoring, it is insufficient to know only that the overall expense levels are in line with expectations. Each line item that makes up the total must also correspond to expectations.
   • If there are material variances among line items that tend to offset each other, the person monitoring variances should be on notice that unusual activity may be present. As an example, if a department meets its overall expense budget but is materially over budget in legal fees (with favorable offsets in other areas), the reviewer might conclude that an event is present that might put future returns at risk.
   • The same principle holds for risk monitoring. Managers should be able not only to articulate overall tracking error expectations, but also to identify how such tracking error is decomposed into its constituent parts. If the risk decomposition is not in keeping with expectations, the manager may not be investing in accordance with the stated philosophy. This situation is referred to as “style drift”.
   • Examples of risk decomposition that a manager should be able to articulate and which the RMU should monitor might include:
     • The range of acceptable active weights (portfolio holdings less benchmark holdings) at the stock, industry, sector, and country levels.
     • The range of acceptable marginal contributions to risk at the stock, industry, sector, and country levels.

Quantifying Liquidity Concerns

• Since a portfolio’s liquidity profile can change dramatically during difficult market environments, tools that measure portfolio liquidity are an essential element of the stress analysis. For example, investors must be aware if a partial redemption could cause an illiquid asset to exceed some guideline.
• A tool used to assess the potential implications of illiquidity is the “liquidity duration” statistic. To calculate this statistic, begin by estimating the average number of days required to liquidate a portfolio assuming that the firm does not wish to exceed a specified percent of the daily volume in any given security. The point here is that it is important to estimate how long it would take to liquidate a portfolio’s holdings in an orderly fashion – that is, without material market impact. The liquidity duration for security can be defined as:

\( LD_i = \frac{Q_i}{\left( \frac{x}{100} \right) \cdot V_i} \)

where,

𝐿𝐷 is the liquidity duration statistic for security 𝑖, assuming that one does not wish to exceed 𝑥% of the daily volume in that security

𝑄_i is the number of shares held in security 𝑖

𝑉_i is the daily volume of security 𝑖

EXAMPLE –

• Calculate the liquidity duration for a given security, assuming that a risk manager does not wish to exceed 16% of the daily volume in that security, given that there are 200,000 shares held in that security and that the daily volume is 5,000 shares.

\( LD_i = \frac{Q_i}{\left( \frac{x}{100} \right) \cdot V_i} = \frac{200000}{0.16 \cdot 5000} = 250 \)

• An estimate of liquidity duration for the portfolio taken as a whole can be derived by weighting each security’s liquidity duration by that security’s weight in the portfolio.
• Liquidity duration is readily calculated for equity holdings, as volume data are easily available. In the case of fixed income securities, where volume information is not available, the estimate of the number of days required to liquidate a position – and an overall portfolio – in an orderly fashion (i.e., without a material adverse earnings impact) will likely result from discussions with portfolio managers.

Objectives Of Performance Measurement Tools

• The following are the primary objectives of performance measurement tools–
  • To determine whether a manager generates consistent excess risk-adjusted performance vis a vis a benchmark.
  • To determine whether a manager generates superior risk adjusted performance vis a vis the peer group.
  • To determine whether the returns achieved are sufficient to compensate for the risk assumed in cost/benefit terms.
  • To provide a basis for identifying those managers whose processes generate high-quality excess risk-adjusted returns. It is believed that consistently superior risk-adjusted performance results suggest that a manager’s processes, and the resulting performance, can be replicated in the future, making the returns high-quality.

Commonly Used Performance Tools And Techniques

• Commonly used performance tools and techniques include-

1. Green Zone
2. Attribution of Returns
3. The Sharpe and Information Ratios
4. Alpha versus the Benchmark
5. Alpha versus the Peer Group

Green Zone

• Each portfolio manager should be evaluated not only on the basis of ability to produce a portfolio with potential (i.e., forecasted) risk characteristics comparable to target, but also on the basis of being able to achieve actual risk levels that approximate target.
• At GSAM (Goldman Sachs Asset Management), a concept called the green zone has been developed to identify instances of performance or achieved tracking error that are outside of normal expectations.
• The green zone concept embodies the following elements –
  • For the prior week, month, and rolling 12 months, they calculate the portfolio’s normalized returns, which are defined as excess returns over the period minus budgeted excess returns over such period, all divided by target tracking error scaled for time. This statistic might be viewed as a test of the null hypothesis that the achieved levels of excess returns are statistically different from the targeted/budgeted excess returns.
  • For the prior 20 and 60 day periods, they calculate the ratio of annualized tracking error to targeted tracking error.
  • For each of the calculation in the point above, they form policy decisions about what type of deviation from expectation is large enough, from a statistical standpoint, to say that it does not fall in the zone of reasonable expectations that they call the green zone. If an event is unusual, but still is expected to occur with some regularity, they term it a yellow zone event. Finally, red zone events are defined as truly unusual and requiring immediate follow-up. The definition of when one zone ends and a second begins is a policy consideration that is a function of how certain one would like to be that all truly unusual events are detected in a timely fashion. For example, if the cost of an unusual event is very high, one would expect a very narrow green zone and quite wide yellow and red zones. In this case, one would expect to find more false positives.
  • The results of the green zone analysis are summarized in a green sheet document.

Attribution Of Returns

• A commonly used tool to measure the quality of returns is performance attribution. This technique attributes the source of returns to individual securities and/or common factors.
• One form of attribution, commonly called variance analysis, shows the contribution to overall performance for each security in the portfolio. This same kind of analysis can be performed at the industry, sector, and country levels, essentially by combining the performance of individual securities into the correct groupings.
• The RMU professional can use this analysis to ascertain whether the portfolio tended to earn returns in those securities, industries, sectors, and countries where the risk model indicated that the risk budget was being spent.
• The attribution process captures the weightings in various risk factors on a periodic basis and also accumulates the returns to such factors in order to produce a variance analysis expressed in factor terms.
• As a general rule, it is most meaningful to attribute returns on the same basis that ex ante risk for such returns is measured. For managers who think in factor terms, factor risk analysis and factor attribution will likely be more meaningful. For managers who think about risk in terms of individual securities, risk forecasting and attribution at the security level will likely be more relevant. This is not to say that risk should not be measured using a range of models.

The Sharpe & Information Ratios

• The Sharpe ratio divides a portfolio’s return in excess of the risk free rate by the portfolio’s standard deviation.

• The information ratio divides a portfolio’s excess returns (vis a vis the benchmark) by the portfolio’s tracking error.

• Both of these tools are designed to produce estimates of risk-adjusted returns, where risk is defined in standard deviation or tracking error space. In theory, two different estimates of standard deviation (or tracking error) could be used for these ratios – actual levels of standard deviation as well as forecasted levels.
• The Sharpe and information ratios incorporate the following strengths –
  • They can be used to measure relative performance vis a vis the competition by identifying managers who generate superior risk-adjusted excess returns vis a vis a relevant peer group. 𝑅𝑀𝑈s and investors might specify some minimum rate of acceptable risk-adjusted return when evaluating manager performance.
  • They test whether the manager has generated sufficient excess returns to compensate for  the risk assumed.
  • The statistics can be applied both at the portfolio level as well as for individual industrial sectors and countries.
• The Sharpe and information ratios incorporate the following weaknesses –
  • They may require data that may not be available for either the manager or many of his competitors. Often an insufficient history is present for one to be conclusive about the attractiveness of the risk-adjusted returns.
  • When one calculates the statistic based on achieved risk instead of potential risk, the statistic’s relevance depends, to some degree, on whether the environment is friendly to the manager.

Alpha Versus The Benchmark

• This tool regresses the excess returns of the fund against the excess returns of the benchmark. The outputs of this regression are –
• An intercept, often referred to as “alpha”, or skill.
• A slope coefficient against the excess returns of the benchmark, referred to as “beta”.
• Standard confidence tests can be applied to the regression’s outputs. The alpha term can be tested for statistical significance to see if it is both positive and statistically different from zero.
• This performance tool incorporates the following strengths –
• It allows management to opine whether skill is truly present or excess returns are happenstance. It tests whether the manager has generated excess returns vis a vis the benchmark.
• It allows management to distinguish between excess returns due to leverage and excess returns due to skill.
• The alpha and beta statistics, and tests of significance, are easy to calculate.
• The beta statistic shows if an element of the manager’s returns are derived from being overweight or underweight the market (occurs if the beta is statistically different from 1.0).
• This performance tool incorporates the weakness that there may not be a sufficient number of data points to permit a satisfactory conclusion about the statistical significance of alpha.

Alpha Versus The Peer Group

• This tool regresses the manager’s excess returns against the excess returns of the manager’s peer group. It is used to determine whether the manager demonstrates skill over and above what is found in the peer group. The peer group’s return is the capital-weighted average return of all managers who trade comparable strategies. The peer group is basically the manager’s competitors in his strategy.
• The outputs of this regression are:
  • An intercept, often referred to as “alpha”, or skill.
  • A slope coefficient against the excess returns of peer group, often referred to as “beta”.
• The alpha term represents the manager’s excess return against the peer group. The beta term measures the extent to which the manager employs greater or lesser amounts of leverage than do competitors.
• Standard confidence tests can be applied to the regression’s outputs. The alpha term can be tested for statistical significance to see if it is both positive and statistically different from zero.
• This performance tool incorporates the following strengths –
  • It allows management to opine whether skill is truly present or excess returns are happenstance. It tests whether the manager has generated excess returns vis a vis the peer group.
  • It allows management to distinguish between excess returns due to leverage and excess returns due to skill.
  • The alpha and beta statistics, and tests of significance, are easy to calculate.
• This performance tool incorporates the following weaknesses –
  • There may not be a sufficient number of data points to permit a satisfactory conclusion about the statistical significance of alpha or beta.
  • Returns of the peer group are biased due to the existence of survivorship biases.
  • There is often a wide divergence in the amount of money under management among the peers. It is often easier to make larger risk-adjusted excess returns with smaller sums under management than with larger sums.