How Do Insurers Set Risk Asset Allocations?

By Alton Cogert

All insurers face problems, some similar to other insurers and some unique to a given insurer. However, it is the problem of low rates for longer that is truly driving CEOs, CFOs and CIOs up the proverbial wall. With that in mind, we’ve seen an increasing number of articles, presentations and various forms of hand wringing attempts to guide insurers through a difficult environment.

But, investment portfolios can be improved by looking in more than one place. And that has caused most insurers to refocus on investments that are not investment grade fixed income. We like to call these Risk Assets, since they are investments that are much more subject to loss, and gain, than investment grade fixed income. And, of course, this must be accomplished while still taming surplus volatility that comes with Risk Assets.

Of course, different insurers tackle the question of what specific kinds of Risk Assets should be in the portfolio in different ways. However, at Strategic Asset Alliance, we were wondering how the overall allocation to Risk Assets was determined.

We could have attempted a survey of insurers, stratified the results by insurer size, line of business, etc. and tabulated what we were told. However, we all know that actions speak louder than words. So, we embarked on using the facts on hand (data from statutory filings) to determine how insurers set their allocation to Risk Assets.

At its core, an insurer is a risk bearing entity, focused on proper risk management. Thus, it makes good sense to assume that insurers attempt to balance various risks in an effort to maximize risk adjusted reward. Of course, that can mean different things to different insurers. However, statutory filings do give us a clue as to how to estimate some of the most important risks.

Specifically, we decided to compare the Risk Asset allocation (as a % of Surplus) to Underwriting Risk (for P/C and Health insurers, we used the Combined Ratio), Operating Leverage (Net Premiums to Surplus) and Financial Leverage (Total Liabilities to Surplus). This was segregated by very broad types of business: Life, Health and Property/Casualty.

We should expect there to be a relatively high correlation between these measures of risk and an insurer’s investment in Risk Assets. Thus, we performed a regression using these risk factors and the Risk Asset allocation as a percentage of surplus, based upon 2014 data provided by SNL Securities’ database of statutory filings.

For the largest twenty Life Insurers in assets, there is a 69% correlation between Financial and Operating Leverage and the Risk Asset allocation. Yet, for the entire Life Insurance Industry, there is virtually no correlation.

This should raise a few eyebrows. But, for now, let’s see what the Health Insurance Industry looks like:

Although the largest Health Insurers seem to have some correlation between these risk measures and their Risk Assets allocation, it is much lower than that found with Life Insurers. However, similar to the Life Industry, look at the overall Health Insurance Industry and there is virtually no correlation.

It gets even more interesting for the Property/Casualty Industry:

Once again the largest twenty companies in the industry exhibit some correlation of a company’s underwriting risk, operating leverage and financial leverage with its Risk Assets allocation. And, once again, the overall industry has virtually no such correlation.

And, it gets even more interesting, as we went deeper by looking at correlations specific to asset size and lines of business, using SAA’s SPG (Structured Peer Groups), where insurers (with less than $5 million in assets) are compared to insurers in a similar size range and similar line of business focus. Within every SPG it was fairly obvious that the general lack of correlation holds.

We even tested these results by lagging these risk ratios by a year. Once again, the results were depressingly similar.

What could be behind these results?

We did some more thinking on these mostly counterintuitive results and came up with ‘good’ reasons for what we saw and some ‘bad’ reasons. These will be the subject of an upcoming white paper that provides more details on this analysis.

Meanwhile, here is a preview of one possible good reason and one possible bad reason:

A good reason: What insurers really use to determine Risk Asset allocations have less to do with these factors and more to do with their sophisticated ERM analyses.

A bad reason:   Insurers are investing based mostly upon the asset-only ‘efficient frontier,’ which results in Risk Asset allocations without consideration of other items on the balance sheet, like company leverage, financial performance, nature of reserves, etc.

Where does your insurer fit in their Risk Asset allocation compared to other insurers of similar size with a similar line of business (much more detailed than just Life, Health or Property/Casualty)? (We have the data for such an analysis, and would be pleased to provide that comparison.)

And, where would your insurer fit among a list of potential ‘good’ and ‘bad’ reasons for this result?

Food for thought as you review how your company compares to peers in this important analysis and think about the current process for setting your insurer’s Risk Asset allocation.

We will be devoting this quarter’s research and discussions on the topic of Taming Surplus Volatility. If you would like to be the first to know when our research releases, please indicate the topics of most interest to you.