Generally speaking, though, a gradual tapering and a gradual rise in interest rates are a good thing for insurance companies, ignoring any other changes that may occur in the rest of the economy. I think the first thing you would see is a restoration of some of the margins that have been squeezed in the past few years as insurers have come under pressure due to the low-interest-rate environment and guarantees on their policies. As rates rise and their portfolio turns over, their portfolio yields will increase and their margins will expand, and generally that will be a good thing. After they recover their margins, I would expect to see interest rate increases being passed on to policyholders through increased credited rates, and I think we might see some of that margin used to support growth as well.

T&R: Are there also risks to insurance companies?

GM: Well, this scenario could be altered depending on what happens in the economy at large as a result of Fed tapering and interest rate hikes. People might reduce the level of insurance they're buying or other things might happen that could put pressure on insurance companies' margins. And if rates were to rise significantly and suddenly, that could put liquidity pressure on the industry as policyholders look for alternatives. In many cases, insurance companies are carrying their assets at an amortized cost rather than marking them to market, and that difference might give rise to mark-to-market pressure for companies with longer-duration assets and liabilities if rates rise significantly and suddenly.

T&R: What should companies do to monitor the health of their insurance companies through the tapering process?

GM: Certain types of products, such as separate account products that are invested in specialized mutual funds, are marked to market, so there is no concern with those types of funds. But for other products, I would recommend looking to the rating agencies, which do monitor these things as part of their process.

T&R: And what should a company do—an insurance company or other type of business—to determine where it faces risks related to Fed tapering and the prospect of rising interest rates?

GM: I'm a fan of stress testing, or scenario planning, the process by which people develop an understanding of how their company's financials would react under a set of targeted scenarios. What would happen if rates rose at different rates of increase? You want to understand how different levels of interest rates could impact each of your business units' risks, the interactions among the businesses, and the company overall. If interest rates rise gradually—say, 100 basis points a year for the next two or three years—what would happen? And what if rates rise twice that fast?

You want to assess those outcomes relative to the company's risk appetite and risk limits, its projected capital, and projections of liquidity. You have to make sure you're comfortable with the results of the stress test. That's a good way of doing basic risk management for an insurance company.

T&R: Isn't this also a risk management best practice for companies outside the insurance sector?

GM: Yes, as a general rule, companies are going to find that it's very helpful to have an understanding of how their businesses would react in different rate environments. They should pick a few scenarios and run stress tests. Even though I use the term 'stress test,' I don't mean companies should assume there will be a major change in rates. They just have to decide what they think is a reasonable set of scenarios and understand how those different scenarios might impact their business.

The key is to make sure you're looking at the risks that are important to your mix of businesses and that you understand what the full range of impacts from those could be. This is fundamental to any good risk management program.

T&R: What are the key considerations for companies in determining their risk appetite?

GM: The appetite for risk tends to come from the top. A chief risk officer or other executive responsible for risk management would talk to members of the board and senior management to get a feel for the kinds of risks or financial movement that would cause concern. This is typically an iterative process, and different organizations do it in different ways.

Board members and senior management may be more responsive to specific scenarios, so it may be helpful to take a draft proposal of a framework for their company's risk appetite to the organization's leaders and say, 'Hey, if this event happened and we had a loss of X dollars, how would you feel about that?' Then they have to triangulate with people because different members of the board and senior management will have different views on risk. The goal is to bring all those different people together, reach consensus, and get a set of parameters that everyone is relatively comfortable with.

As part of the process, some companies will perform stochastic analyses and consider results at a targeted level such as a 90th or a 99th percentile as part of the development of their stress tests. These stochastic models are complex, with some subjective elements, and they may not be effective for all risks. For an annual model, this could result in a risk constraint expressed as not wanting to lose more than X dollars once every 10 years or 100 years. In the end, there should be consideration of both magnitude and likelihood of potential risk.

T&R: So, where should companies start the stress testing for the current round of Fed tapering? Is it reasonable to evaluate the effects of an interest rate increase of 100 basis points? 200 basis points?

GM: Companies need to determine what they think is likely in terms of a modest increase based on what they've heard. Different people will have different perspectives on what's reasonable. I'm not going to speculate about what the government's going to do nor about how interest rates are going to respond, but I don't think most people would expect a gradual tapering to result in rates that are rising by 300 basis points a year for five years.

Now, there may be people who want to include a stress test with that kind of rate increase anyway because they're concerned about their business and want to know what would happen in the event of a rapid rise in rates. They might come up with a scenario or two that represent a much larger increase than they actually expect to happen, a scenario that isn't likely but is possible. They might think, 'Rates have gone up this much in the past. It's not clear that conditions now warrant this size of increase, but if it did happen, what might it mean for us? How should we set ourselves up to deal with this scenario, should it occur?'