Sellers and potential buyers of structured settlement annuity (SSA) portfolios could find themselves handicapped by outdated and inaccurate mortality projections.
Trading activity in SSA portfolios has been on the increase in recent years, stimulated by buyers interested in longevity risk and reinvestment opportunities and more insurers becoming motivated or needing to sell long-tail liabilities. Over the last two years alone, we’ve seen over $25 billion in liabilities placed on the market.
And while the underlying financial criteria for a continuing stream of deals remain largely positive, one factor has risen to the surface that could potentially stem the flow – mortality projections.
Structured settlement annuities (SSAs)
SSAs are court-ordered awards, placed typically by a property and casualty insurer with a life insurer, to provide an income stream to a claimant or dependents after an event such as a workers’ compensation claim. A distinct characteristic of the market is that beneficiaries can vary considerably in age and degree of impairment, and include young children, meaning that in some cases, payments can be expected to continue for many decades. Since the market took off in the early 1980s as a result of favorable tax treatment, many SSA writers have found these liabilities progressively more difficult to manage and to account for the risks to financial markets and other stakeholders. However, SSAs appeal to financial investors because of the long-term nature of the liabilities and opportunities to reinvest the assets backing reserves.
The predominant approach used to project mortality for SSAs has been “rated age.” This approach involves assigning a shorter life expectancy than standard mortality would imply (and thus a shorter stream of annuity payments) based on the degree of the individual’s impairment, from which actuaries have set mortality assumptions for SSA pricing, reserving and projections based partially or wholly on the rated age.
Over time, several flaws have appeared in the rated age approach. Firstly, the underwriting was only as good as the information available at the time of issue, and may have been influenced by pressure to produce winning bids in what, in past years, was a very competitive market. In most cases, insurers have not refreshed their approach to expected future mortality. Moreover, the rated age methodology hasn’t accurately accounted for medical advances or even the seasoning of the business.
Figure 1 illustrates the three most common current approaches to projecting impaired mortality. The difference between rated age mortality and actual age mortality increases dramatically over time. In this example, life expectancy at age 31 is identical under the rated age and the constant extra deaths approaches, and yet clearly the two mortality curves are dramatically different, illustrating one flaw of relying heavily on life expectancy and rated age set at issue.
The issue is even more salient because our analysis of what we believe to be the only predictive model of SSAs covering over 5,000 deaths, using our Emblem software, shows that SSA mortality experience doesn’t conform to standard annuitant mortality, or even population mortality tables. For example, we have seen an as yet unexplained pattern of dependents who receive SSA payments dying sooner than expected. Equally, there is evidence of the impaired lives living longer. Consequently, industry perspectives on structured settlement mortality are frequently out of date and unsupported by emerging data.
As it stands therefore, some potential SSA portfolio transactions will be relying – and perhaps stalling – on what an underwriter saw as the prevailing mortality picture 30 years ago. For buyers and sellers alike, such imperfect information is unlikely to help them make the most educated decisions about a target block of business or, critically, to understand the risks better and access or demonstrate hidden value.
Given that mortality is the only variable in SSA liabilities, better projections of life expectancy in a portfolio could make the difference between good and bad deals. Transparency of mortality assumptions can also go a long way towards shortening the process (sellers have more realistic expectations, and buyers are quicker to gain comfort).
Lori Helge is a senior life insurance consultant in Willis Towers Watson’s Insurance Consulting and Technology practice.