What It Measures
Stress testing is a risk management tool that helps to identify how vulnerable a business, portfolio, or venture might be to unusual, negative circumstances. It may involve scenario analysis or simulation, based on hypothetical or historical data.
Why It Is Important
Stress testing is extremely useful to financial analysts because it provides them with additional information on potential portfolio losses in the event of extreme, but unlikely events. Crucially, stress testing enables risk managers to test how robust a particular investment or instrument will be in the event of a serious change in circumstances.
The current “credit crunch” is a perfect example of the importance of stress testing. Institutions that conducted thorough stress tests based on a simultaneous lack of credit coupled with increased exposure to collateral debt obligations have been less seriously affected, according to financial regulators.
How It Works in Practice
Stress testing is used to test instruments against scenarios such as: What happens if the market collapses by 60%? What happens if interest rates on our loans double? What if our lease costs increase by 75%?
The most common approach to stress testing is to use Monte Carlo simulation. This involves taking a number of randomized scenarios, ranging from modest to extreme outcomes (say a 90% drop in sales, versus a 90% increase in sales), and modeling the results on a probability distribution curve.
Most stress testing exercises involve multiple stressors. There is usually also the ability to test the current ability to cope with a known historical scenario. In recent years, many companies have tested their ability to cope with the kind of recession seen in Europe during the 1990s, for example.
There are three basic kinds of event a stress test can be applied to: Extreme event (current positions combined with historical event); risk factor shock; and external factor shock (shock of any external index factor, such as oil prices or exchange rates).
There are two basic sorts of stress test—sensitivity tests and scenario tests. A sensitivity test assesses the impact of large movement in financial variables on portfolio values without needing a specific reason. For example, you might test what happens if interest rates fall by 20%, or the cost of equity rises by 15%. Sensitivity tests can be run quickly and easily, but lack historical perspective.
Scenario tests are more costly and complex, but provide a better insight into long-term risk. They are constructed either within the context of a specific portfolio or based on a specific set of historical circumstances. Risk managers identify a portfolio’s key drivers and test what happens if those drivers are stressed beyond value at risk (VAR) levels. Many risk managers use a hybrid approach of scenario and sensitivity testing.
Tricks of the Trade
In many countries, stress testing is a regulatory requirement. In the United Kingdom, for example, the Financial Services Authority (FSA) requires certain financial institutions to conduct stress tests and ensure they have sufficient capital reserves available to handle extreme events.
Stress testing is mostly used to analyze market risk, and the impact of market changes on portfolios such as interest rates, equity, exchange rates, and commodity instruments. These portfolios are suitable for stress testing because their market prices are regularly updated, giving sufficient data to analyze.
VAR analysis is a very useful risk management tool, but it is not able to incorporate all possible risk outcomes, particularly sudden, dramatic changes in market circumstances.
Stress testing is often used as a tool to communicate risk to business leaders. Rather than the hypothetical probabilities of VAR, stress testing provides a response to a very specific set of circumstances.
Regulators commonly use stress testing to consider the vulnerability of entire financial systems. Stress tests were recently used by the Bank of England as part of a Financial Sector Assessment Program (FSAP), which considered how 10 large domestic banks would deal with a 35% decline in global stock prices, and a 12% decline in domestic property prices.