Financial Planners: Identify and Manage Risk

Banks, life insurance companies and financial planners are all in the business of identifying and managing risk for their clients. There are a lot of similarities between how banks and life insurance companies manage risk and the tools financial planners use to help clients manage their financial futures.

We’d like to compare them and explain how financial planners can take advantage of the methods that banks and insurance companies have learned to manage financial risks.

Short-term view vs. long-term view

When it comes to risk management, most banks focus on managing short-term market risk such as a significant drop in equity returns or a sharp change in interest rates over a short period of time. This short-term risk focus is a result of two important drivers: (1) Banks are required to value their assets and liabilities to reflect current market conditions, an approach referred as market consistent valuation. This valuation approach can potentially result in big swings of a bank’s asset, liability and capital positions when equities, interest rate and other market variables move significantly. (2) Most bank liabilities are very liquid; deposits can be withdrawn at any time at their market value, creating the risk of a run on the bank particularly when a stress event occurs. As a result, regulators and bank executives are very focused on the impact of short-term severe market changes on a bank’s solvency. They need to ensure that banks maintain strong capital levels in case an unexpected stress event happens tomorrow.

Life insurance companies are in slightly different risk management situations than banks: (1) Life insurance companies are allowed to value some of their assets and liabilities at book value, an approach reflecting the purchase cost rather than current market conditions. (2) Many of the insurance company’s liabilities are not as liquid. For example, a death benefit may not be payable for many years, significantly mitigating the “run on the bank” type of risk. As a result, insurance companies have a bit more tolerance to short term market movements than banks. But insurance companies are very exposed to risk over a long term horizon — potentially 30 or more years out.

Financial planners need to be concerned with both the short-term and the long-term when assessing their clients’ financial needs. There are many factors where a short-term change can make a big impact on a client’s financial plan — some might be disruptions in the financial markets while others might come from changes in a client’s personal circumstances. But advisors also help their clients evaluate the distant future. Much of retirement planning is less about managing the impact of a short term market event and more about analyzing the financial viability of a strategy over a long time horizon.

Stress Test vs. Monte Carlo Simulation

Most of the risk measurements used by banks are developed through stress tests due to the short term focus. For example, the Federal Reserve has developed stress tests for SIFI’s (Systemically Important Financial Institutions) that involve severe market shocks to several key economic measures: equity performance, interest rates, credit spreads, and GDP growth, over four to eight consecutive quarters.

In contrast, life insurance companies are more focused on measuring and managing long term risk. The NAIC (National Association of Insurance Commissioners) developed reserves and capital for a long-term contract such as a guaranteed lifetime income benefit of a variable annuity. This approach uses Monte Carlo simulations over a long time horizon (30 years or longer) and focuses on the tail risk — that is, the outcomes of some of the worst-case scenarios.

From this perspective, retirement planning risk evaluation is more like the financial and risk management of a life insurance company. The long-term retirement planning horizon requires financial planners to use simulations to quantify risk for their clients. Probability of success is a measurement used by most financial planners in measuring the tail risk in retirement.

Combining Monte Carlo simulations with stress tests

Monte Carlo simulation is a powerful approach to quantify long term risk over many different possible outcomes. However, it does have its limitations. Most Monte Carlo simulations only focus on market risks such as equity returns, interest levels and credit spreads.

Recognizing that some of the key variables are not modeled in the Monte Carlo simulation, life insurance company risk experts capture the risk of those key variables by applying stress tests to their simulation. To do this they stress their mortality and/or expense assumptions for a certain set of policies, before rerunning the simulation and analyzing the results. In fact, state insurance regulators now require life insurance companies to run stress tests on the Monte Carlo Simulation and submit those stress results for regulatory review as part of their annual AAT (Asset Adequacy Test) filing.

Stress test that financial advisors can consider

For retirement planning, we believe that it is very important for financial advisors to borrow from the insurance companies and apply stress tests to the baseline retirement Monte Carlo simulation. Most of the simulations used in financial planning capture market risks including equity and fixed income returns. Unfortunately, too often many key variables are left out of the simulation model such as taxes, social securities, longevity, inflation and health care cost. What will happen to client’s probability of success if tax rate is higher, client lives longer, inflation is higher, social security is reduced or retirement health care cost is higher?

Many financial planning software come with what if functions that help advisors to understand the impact of stressful event. It is very important for advisor to consistently and systematically utilize stress test capabilities to create a holistic view. Let’s look through a hypothetical example how advisor can use stress test to create a more robust retirement plan for his/her client.

Let’s start with a retirement plan created by an advisor for the hypothetical client: the retirement has 90% probability of success, output from a typical retirement Monte Carlo simulation that includes stochastic asset returns. The chart below shows some simulation details with the ranges of portfolio value cross scenarios and time period. Clearly, the client has adequate asset to last till the end of planning horizon in most of the scenarios — not a bad! However, it is very important to be aware that the Monte Carlo simulation only includes stochastic asset returns. The probability of success only captures the asset return risk. Risks associated with other key variables such as inflation are essentially left out of the simulation model and not captured in the retirement plan.

Now let’s look at stress test to understand the potential risk exposure this retirement plan has to variables other than asset returns. In this hypothetical example, the financial advisor applied six different stress test to the base retirement plan including a 30% immediate equity market drop, 20% higher future tax expenses, a 20% reduction in social security payment, 5 year longer planning horizon, 20% higher retirement health care expense as well as a 1% higher inflation. The results of the stress test are shown in the chart below.

For this particular client, the 1% higher inflation stress has significantly reduced the probability of success from 90% to 41%. Higher tax rate impacts the retirement plan in a very meaningful way, lowering the probability of success to 66%. Other key variables also impact the probability but to a less extend comparing to inflation and taxes. The stress test information is very valuable as it now provides a holistic view of client’s retirement plan, capturing the key risks that are not included in the baseline Monte Carlo simulation model. After these risk exposures are identified and quantified via stress test, advisor can consider potential options, design specific strategies to mitigate the risk exposure, communicate the stress test and any adjustment to retirement plan to his/her client. Without going through the stress testing process, advisor and client may have continued to focus on the 90% success probability and lost sight of the potential significant risk exposure to inflation, taxes and other variables.

Summary

Banks, life insurance companies and financial advisors are in the business of identifying and managing risks for clients. Traditional retirement planning shares a lot of similarities with how life insurance companies manage risk.

Retirement planning is a process to identify and manage risks that emerge over a long time horizon, and it is very important to use simulations as part of that process. The key risk drivers that are not simulated in the Monte Carlo exercise need to be captured through stress tests to complete a thorough planning exercise.