Risk metrics for climate change
What metrics are most useful to insurance companies when considering the risks and impacts of climate change?
Announcer: This podcast is intended solely for educational purposes and presents information of a general nature. It is not intended to guide or determine any specific individual situation and persons should consult qualified professionals before taking specific action. The views expressed in this podcast are those of the speakers and not those of Milliman.
Tony Dardis: Hello and welcome to Critical Point, a podcast brought to you by Milliman. I'm Tony Dardis and I'll be your host. On today's episode we're talking about COVID-19 and Enterprise Risk Management or ERM for life insurance companies. At the heart of effective risk management is the ability to manage the tail events and in regards to that the COVID-19 pandemic is raising some quite profound questions for risk practitioners. In this podcast we'll discuss what we're seeing as the emerging issues for life companies. Joining me today are my colleagues Ariel Weis--
Ariel Weis: Hello, everyone.
Tony Dardis: And Chloe Lau.
Chloe Lau: Hi, there.
Tony Dardis: All right. First let's talk about risk analytics and I'm thinking here about stress testing which is really the mainstay of the risk practitioner's work when it comes to analyzing tail risk. Ariel, what do you think COVID-19 is bringing to our attention in terms of the types of stresses we look at?
Ariel Weis: So that's a really good question to start with, Tony. I think we can divide this in two ways and when we've talked to clients and when we've talked to other risk practitioners in the industry, we're finding that they're thinking about it two ways as well. There's the impact of COVID-19 on assumptions for stresses and then there's the impact of COVID-19 on the types of stresses. So if your question is about the types of stresses, traditionally a lot of stress testing has been done around a best estimate scenario or a best estimate assumption set and then stressing that with different sets of assumptions which creates mathematically interesting results that doesn't necessarily come back to real life results. With COVID-19 what we're seeing a lot more and what we're thinking about more as well is in the realm of narrative stress testing or narrative scenarios as we call them or reverse stress testing. Narrative scenarios are interesting because instead of shocking a set of assumptions on a best estimated basis, you're really thinking about conceptually scenarios like the 1918 Spanish Flu pandemic and how would those translate under a COVID-19 episode and what types of historical information do we have to supplement our judgment in creating those scenarios for a stress test and therefore having a narrative that ties together the results? From a reverse stress testing perspective, and this is done a little bit less but it's quite interesting too, every company that we talk to has a set of explicit or implicit risk GAAP-like statements in terms of where they're willing to cross a line for important metrics that they measure and how far they're willing to go before they say no more of these. And reverse stress testing really goes to the heart of those lines and trying to define what type of assumptions and model behavior will break those limits so that the company can understand how bad things have to get before management actions are required and management has to step in. Does that sound clear?
Tony Dardis: Very interesting, Ariel, and I think we're going to see a lot more of reverse stress testing in the near future. I think also related to stress testing that I'm very interested in is the question of what metric should we be testing. What are your thoughts on that topic, Chloe?
Chloe Lau: Thanks, Tony. It's definitely an interesting question. I mean, in this pandemic you're definitely looking at both qualitative and quantitative type of metrics. But I think first thing the more important thing is to make sure that your metrics are actually consistent with what your definition of a risk appetite is within your company. You know, in some way risk appetites somewhat dictate the metrics that you should be looking at to begin with so that you know what needs to be measured and what their limits are and what actions you can trigger when they actually do reach the limits. So obviously there are the traditional stress testing metrics that focus on the company's financial positions that we should be looking at currently like, you know, impacts on regulatory and economic capital, statutory or GAAP earnings. You know, these metrics are certainly useful, especially during this volatile market environment. Regulators, shareholders and obviously senior management, they're all going to be interested in the financial stability of the company, so metrics like available capital versus required capital as well as liquidity ratios are going to be on everyone's mind, right. And another thing is profitability. You know, that's another metric that is going to be scrutinized quite a bit during this volatile financial time. So you know, metrics like ROE, ROI should also be monitored as well. The other thing to keep in mind though is that, you know, this pandemic COVID-19 has been affecting companies not only from a financial or reputational sense but also having severe impacts on their workforce management and operational structure and generally other areas within the companies that are not being measured when conducting the stress testing because they might not be easily quantifiable and they might need somewhat measured in a more qualitative manner, for example, operational risk. It could include a very broad bag of risks that impact a company's resilience, even survival. But these risks are a lot harder to quantify. You know, there's some aspect of operational risk that could be easily put a dollar amount to it, such as loss of revenue due to business interruptions or loss of new business. But a lot of the unquantifiable risk sometimes get overlooked. They're not traditionally metrics that we look at and we should definitely keep them at the forefront of our mind during the stress testing in this environment as well.
Tony Dardis: Chloe, you mentioned capital aspects and that gets on to another element of stress testing which is around economic capital and most companies now in the U.S. are doing an economic capital calculation, what's often dubbed the quantification of ERM. I'm interested, Ariel, why do companies do an economic capital calculation and what role should that be playing, especially in the context of the current environment?
Ariel Weis: Yeah and that's a good point that really ties to Chloe's comments. Economic capital has traditionally been defined as a quantification of the amount of capital that a company thinks it needs to hold in order to survive its worst-case scenario. When you think about capital you're thinking about really the amount of additional assets on top of the liabilities that a company's holding and economic capital differs from regulatory capital in that it's not really tied to specific regulations that may have a specific objective. For instance, statutory capital in the U.S. is really focused on ensuring that life insurance companies stay solvent to pay future claims from policyholders. But economic capital really focuses on the best estimate view that the company has and what's going to happen to it from an economic basis and those risks that are arising and what capital might need to be required. And it's generally broken down between market or financial, underwriting or product, strategic operational, those kinds of categories. In terms of why it comes in for COVID-19 is that we are really in one of those stress environments that could represent the worst-case scenario. And it's important for companies to understand how much of the available capital has been consumed by this economic capital that's required, how much is left and in the stress environment that could come in the months ahead from the moment of this recording anyway, how much more capital will the company need to hold or even go out and get by whatever means are available to the company to be able to survive and thrive in the recovery that hopefully is to come after the event.
Tony Dardis: You raise a number of interesting points there, Ariel. I'll also add in the context of economic capital, to the extent you're trying to make assumptions about what's happening to risks that entail and also what their interrelationships are, it's an incredibly difficult thing and I think COVID-19 may be raising questions as to whether such assumptions need to be reevaluated and recalibrated. If you think about Solvency II in Europe, that's calibrated to catch a 1-in-200 events. It raises the question around COVID-19 and its implications for the economy and the financial markets. Is this a 1-in-200 event? Or is there worse that could come? So you think about, say, the Spanish Flu, that's about 100 years behind us. The financial and economic events of 2008-2009, not even 20 years behind us. So where does today's environment play or fall on your company's economic capital probability spectrum?
Tony Dardis: Another thing I'll mention around economic capital is: are models perhaps a little bit too simplistic in how they capture interrelationships between risks in the tail? So we all know about the concept of all correlations going to one in times of market distress; thus you start to lose your diversification benefits. But are such interpretations of correlations a little bit too simplistic? You think about a pandemic that may cause a market crash. Sounds a little bit similar to what we're seeing at the moment, doesn't it? But does a market crash cause a pandemic? So the relationships don't work in opposite directions in that regard. So this seems like something that maybe companies might want to explore a little bit more carefully in the future.
Ariel Weis: One thing to add to that in the sense of the 1-in-200 event or not, it's interesting that you point out that we had the Spanish Flu 100 years ago and we had financial events that had been of a, so far, worse magnitude than what we're seeing these days, not even 20 years ago. It's really the combination and that correlation in the tail that could lead to a 1-in-200 event. Even though as of this point in time we may not be in 1-in-a-200 year event yet with COVID-19, especially for life insurance companies. I think different industries could have a different view on that and what this-- what disruption this is causing as a whole. The other thing I’ll add to economic capital and this is an interesting tidbit is that it's not a new tool. Economic capital has been around for a few thousand years and there's records of Phoenicians actually calculating some very rudimentary versions of economic capital when they invented their versions of insurance back in the day.
Tony Dardis: There's lots for sure we can say about economic capital. One last point I'd like to make is the question of getting replenished numbers out for management review and action in circumstances such as the current environment. You think about companies who look at economic capital from the market value perspective and they're in a position where the excess of available to required capital is likely to be diminishing significantly under the current environment. Seems to me there the details of such is something that senior management would for sure be very keen to be aware of sooner rather than later. So that asks the question how quickly can companies generate their economic capital numbers so that management has an up to date perspective on things and how COVID-19 is affecting the economic health of the company. So we're certainly aware of best practice companies in the market who are being able to get numbers out quickly. They've been successful in operationalizing their economic capital calculation so they can produce replenished numbers extremely quickly, you know, maybe within hours, maybe within a day or two. You know, certainly very different to what we're seeing other companies maybe taking weeks to produce those numbers.
Tony Dardis: Let's move on to talk about risk strategy. And what is the current environment? What is COVID-19 making companies think about in terms of potential risk strategies? Are they beginning to think about possibly doing some things a little bit differently to manage the risks. Chloe, from an operational risk perspective, what are you seeing as the biggest concerns and risks that companies are facing and how are they being managed?
Chloe Lau: So Tony, I think this is actually quite an interesting question because, you know, operational risk is an area that a lot of risk professionals are aware of, but given how broad operational risks are, you know, certain aspects are bound to be under-measured for sure. And this is one of those situations where operational risk is getting hit on multiple sides now and it's affecting pretty much everybody. You know, operational scenarios considered unthinkable before, you know, like a shutdown of all physical offices or you know, entire company relying on the same network to access your share drive and working from home. These are now very, very real and realistic scenarios that, you know, many people might not have thought would have happened before. So even companies who didn't have cybersecurity at the forefront of their mind before and now are thinking how to protect themselves when they're not given enough time to properly vet all the employees' home computers or to ensure that all the employees can work from home remotely and securely. And on top of all these IT risks that are probably coming up, you know, business disruption is probably one of the bigger operational risk categories during this pandemic. I think a lot of life insurers still have functions that rely on paper transactions or in-office engagement. Some insurers don't necessarily have call centers that are set up to allow employees to deal with customers remotely. You know, all these might hugely impact their claims management, their sales and even reputation or regulatory actions. In terms of strategies to manage the operational risks for the future, it really depends on how prepared a company is I think. Many companies use business continuity planning and disaster recovery policies to help them understand how prepared they are. Even just before the wider lockdown for COVID-19 I know a lot of the companies actually did test runs on having the majority of their organization work from home for one day or two just to see whether their server or the network has the capacity to handle everybody's requests and to be able to get everybody online. You know, it's obviously difficult to think of all the potential extreme scenarios like this in the future. But I think this pandemic has showed us how quickly a company needs to change its operational process to adapt to the extreme external environment. So preparedness and flexibility I think are definitely keys in managing operational risks. In addition to having policies in place that we can put in, company-wide business continuity drills during a normal environment would probably be something that companies might want to look into in the future just to test whether the entire company can be actually offline and remote work for the entire time, making sure that there a lot of the operational processes have dual policies in place where they could be in the office and also work remote or have rotating shifts and all that. So that would be some of the strategies that maybe some companies could consider.
Tony Dardis: So certainly a lot to think about on the operational risk side of things. And of course there's very significant risks on the financial side as well and maybe I'll say a little bit about that. So obviously, equity market volatility and if you could have imagined a couple of months ago further declines in interest rates it seems incredible. One topic in particular I think that's worth mentioning and getting increased attention is the threat and possibility of negative interest rates in the U.S. including long-term rates and that just exacerbates and gives more urgency to addressing what was already a looming reinvestment crisis for life insurers. We've seen at least one company outside of the U.S. bite the bullet and purchase a very large interest rate hedge for protection against their expectation of lower rates for longer. Now of course there's a cost associated with hedging in that fashion but given where we are and the short- to medium-term outlook, maybe we'll see more companies adopting that kind of strategy. Negative rates as well creates a modeling nightmare for life offices. How do you model policyholder and ensure a behavior in such an environment? And are your modeling platforms doing the right thing? Are you buying and selling assets as you might expect? And then how much weight do you want to give to the fact that so long as paper money continues to exist then you can earn more by storing cash under your floorboards like my grandma used to do versus keeping your money in the bank. So there's all kinds of interesting modeling issues to worry about in the context of negative rates. Another thing I'll mention in terms of financial risks is that in the context of investments I think the time has come for broader enterprise risk aspects to be brought into the asset allocation decisions if not already. So for example when making an investment decision what are the impacts from an economic capital perspective? So we're seeing more companies giving consideration to that. There has been a trend in recent years along those lines and I think the current environment, if anything, is going to give more impetus to that. So not only does that give you more comfort around the investment decisions you are making, but also potentially opens the door for investment in wider asset categories maybe to pick up a bit of extra yield and at the same time get a little bit more diversification. So I think some really interesting things from an investment perspective coming out of the current environment.
Tony Dardis: To wrap things up, let's talk a little bit about how we're seeing the Chief Risk Officer getting involved in critical management decisions as the COVID-19 crisis unfolds. Ariel, your thoughts on that?
Ariel Weis: Yeah. I think there's a two-fold aspect to how CROs and their departments are getting involved during COVID-19. And I think you both touched on them in your points that you've made earlier. There's a point of the Office of the CRO generally being an aggregator, if not directly a calculator and an owner, of an economic capital model, and therefore being in a prime position to very quickly analyze what the results of the distress in the markets and the potential impact of the pandemic could be on the company results and its capital position. Sort of being a first line even though it's in a second line role for helping others within the company understand the impact of the changes in the environment and helping senior management quickly make decisions that matter and will have direct profound effects in the profitability of the company in the near future as well as in the medium term. There's also the other aspect that I think that CROs are getting involved in in terms of making sure that they had been second-line to more of the operational aspects of the company continuing to work through business-as-usual plans and now there's such contingencies making sure that first line is activating those as appropriate and where they aren't invoking those plans themselves and where any hiccups are found, because of the unique situation that they find themselves in, also working with first line to smooth things out and have a consistent response for crisis management across the entire organization.
Tony Dardis: Very good to see the CRO certainly at the table in making some of these strategic decisions in the current environment. And I think it's been an emerging trend over time that the CRO has been getting increasingly involved in using ERM to help with strategic decisions.
Chloe Lau: And I think one other thing to add to both of your points is that, you know, Ariel mentioned that the CRO is getting in the position to helping others, you know, gather information. And I wonder if a lot of these companies now are global companies and with a pandemic that's happening across the globe, it's hitting different regions at different times and there are different CROs at different regions and I think they could really act as to coordinate across a different region as they usually talk to each other quite frequently at a group risk level where first lines might not have the same luxury. So definitely the risk function is able to help in that respect as well to gather more of, like, a consistent global response for a company. And also lessons learned that other regions might have gotten hit earlier than some others as well.
Tony Dardis: Well, thank you to you both for your perspectives. You've been listening to Critical Point, presented by Milliman. To listen to other episodes of our podcast, visit us at Milliman.com or you can find us on iTunes, Google Play, Spotify or Stitcher. We'll see you next time.
Critical Point Episode 22: COVID-19 and life insurance risk management
At the heart of effective risk management is the ability to manage the tail events, and the COVID-19 pandemic is raising some quite profound questions for risk practitioners in the life insurance industry, as Milliman consultants discuss for this episode of Critical Point.