Buy-side Risk Management - What You Need to Know
The roots of risk management can be traced back to the need for banks and large financial institutions to manage their economic and regulatory capital at the holding company or consolidated level. Thus, tools such as value at risk (VAR) and daily earnings at risk (DE@R) were born. Over the last 10 years or so, buy-side organisations have adapted and customised some of these same practices; however their objectives, approaches, methods and implementation are very different.
In my opinion, effective buy side risk management should focus on three key areas: measurement of performance and investment success, return contribution/attribution analysis and exposure/risk analysis. The effective implementation of these should help senior management and portfolio managers answer the following questions:
Before any portfolio manager begins his or her fiduciary responsibility of managing money, it is important that clear investment objectives are agreed, including a clear understanding of investment success. Regardless of whether performance is measured relative to a benchmark, absolute return target or peer group, it should be analysed for every account with regular frequency.
The investment objective and/or performance targets should help dictate how much risk a portfolio ultimately assumes. Some create performance objectives by first selecting a risk tolerance (e.g. volatility level) and backing into a return target by reverse engineering an Information or Sharpe ratio. This is also an acceptable way of determining an appropriate definition of investment success. Ultimately, it is critical to a) understand how investment success is measured, b) define clear portfolio risk parameters, and finally, c) measure the product against both factors with regular frequency. At the end of the day, clients pay us to take and understand risk and generate performance.
Another important piece of the puzzle lies in understanding market events and how they have impacted a portfolio. This is where attribution and contribution analysis fit in. By analysing performance results, the strengths and weaknesses of the investment capabilities can be identified, and efforts (and risk budget) can be focused on the sweet spot. Are we great security selectors? Strong sector or asset allocators? Market timers? etc. I believe that some risk functions fail to recognise the importance of realised performance analysis, as it is not as sexy as stress testing or calculating a new methodology for measuring spread risk on a mortgage portfolio. Results of this type of analysis can be used to ensure that portfolio managers are following their articulated investment processes. For example, a portfolio manager may describe his or her investment process as being one that primarily focuses on security selection rather than sector allocation. This can easy be tested via simple Brinson style performance attribution. Risk budgets can also be established around core competencies. In my experience, portfolio managers who stray from their investment process can, and sometimes do, wind up in trouble. Understanding their selection and construction processes, and monitoring the results, is very important.
While performance and attribution analysis are backward looking, exposure and risk analysis should tell us where we are headed. Generally, I think about risk in the following buckets: current exposure and allocation; factor sensitivities; portfolio or macro level risk (standard deviation or volatility; active/total risks, VAR); scenario analysis; and stress testing. Many great books have been written on these subjects and it is easy to get lost in the mathematics.
Focus should be placed on:
Traditional risk exposure and factors that should be assessed on an ongoing basis include but are not limited to: equity beta (and active risks); interest rate and key rate duration; spread duration (credit); currency deltas; option Greeks (vega, gamma, etc.); liquidity; security concentrations and geographic/sector allocations. It is difficult to define a ‘one size fits all’ list, since there are so many different strategies – commodities, real estate, private equity, power trading, distressed debt, emerging markets, etc. Each strategy requires focused attention, customisation and diligence in order to identify the relevant risk measures. While these factors can be measured using the same tools as those found on the sell side, one key difference may be the application. In managing long-only portfolios, relative risk measures such as active duration or ex-ante active risk take priority over absolute portfolio risks. This is a key difference between buy- and sell-side investments. Hedge funds tend to think about risk more like sell-siders than buy-siders.
Scenario and stress testing analysis is critical in understanding the impact of economic and/or macro events on a portfolio. The implications of yield curve flattening/steeping, credit spread widening, liquidity evaporating, currencies devaluating, historical correlations diverging, major countries defaulting and more, all need to be understood.
Finally, the absolute and total relative risks need to be understood – active risks, tracking errors, relative VARs provide good understanding on the overall level of portfolio risks.
Risk management for buy side firms should focus on providing transparency around investment processes, performance and forward looking risks to both senior and portfolio management. There are other very important risk areas, which have not been addressed, such as operational risk, which also require diligence and attention.