Keeping faith: IAIS rebuffs view G-Sii concept will die

But firms see workplan announcement as step away from focus on high-risk entities

If certain global insurers are correct, the idea of designating a handful of firms as systemically risky is facing an inevitable demise.

On 28 February, the Switzerland-based International Association of Insurance Supervisors (IAIS) announced it was developing an activities-based approach to assessing systemic risk in the insurance sector, as part of the next three-year cycle of its supervision efforts.

Insurers have seized on the idea this could mean the list of nine global systemically risky insurers – a key part of plans by the IAIS and the Financial Stability Board (FSB) to combat systemic risk – will be torn up. Regulators insist the designations are not going away.

The chief risk officer at a big US insurer tells Risk.net: “The approach IAIS had been taking for systemic risk was inappropriate. Their announcement is basically: ‘We’re walking away, we’re withdrawing from thinking about insurance companies as systemic, and [we’re] thinking that activities might be systemic instead.’”

The IAIS plans to develop a methodology to identify systemically risky activities at the sector level by 2019, with a public consultation in 2018.

The body will include the new approach in its Common Framework for the Supervision of Internationally Active Insurance Groups (IAIGs). It will inform IAIS’s ongoing efforts to design a higher loss absorbency (HLA) requirement for global systemically important insurers (G-Siis) and an insurance capital standard (ICS) for a wider group of international insurers, the implementation of which has been pushed back to 2022.

The announcement comes at a time of fading confidence in the industry about the future regulation of G-Siis and about the progress of the ICS project.

“The general expectation is the [US] local designation process will be revoked, so the whole notion of designating individual companies because of their size has become much harder to argue,” says the head of regulation at a European insurer.

“The political environment has made the current approach much more difficult. I think that’s acknowledged in the IAIS… Within the IAIS, quite a few supervisors were a bit sceptical about the current approach.”

Yet it remains unclear how far the announcement represents a meaningful break with existing policy.

The general expectation is the [US] local designation process will be revoked, so the whole notion of designating individual companies because of their size has become much harder to argue

Head of regulation, European insurer

IAIS secretary-general Yoshihiro Kawai emphasises the new activities-based approach is intended to complement, and not replace, the G-Sii rules: “If we only focus on the entity-based approach, there is a gap because those systemically important activities could be carried out by other entities that are not systemically relevant.”

Sticking with an entity-based approach also protects against big insurance company failures, he argues: “To make it clear, the entity approach is relevant because failure is very much entity-based. The company fails and that has a huge systemic impact, as we have seen in the financial crisis. So we are keeping our entity-based approach, but it will be complemented by an activity-based approach.”

The IAIS currently dubs nine insurers G-Siis: Aegon, Allianz, AIG, Aviva, Axa, MetLife, Ping An, Prudential Financial and Prudential.

“Designating G-Siis is still the end goal, but they will be identified differently,” says Julie Mix McPeak, a member of IAIS’s executive committee and commissioner of the Tennessee Department of Commerce & Insurance. McPeak also serves as president-elect of the US’s National Association of Insurance Commissioners (NAIC), which represents state insurance regulators.

“We were relying on a relative weighting formula, kind of assuming that some of the largest companies would have been some of those most likely to contribute to systemic risk. Several members, including those in US, didn’t feel like that was the best assessment of systemic risk, and there could be activities going on at any size company that might be problematic from a regulatory perspective,” she adds.

Dual approach

Some in the industry see such a twin-track approach as unlikely to last.

“Doing two things in parallel makes sense for the transition period, but it doesn’t make sense for good,” says Lutz Wilhelmy, head of regulatory risk management at Swiss Re. “When it turns out the activities-based systemic risk assessment is superior – and I’m convinced it will be superior – then it should really replace [designations].”

Lutz Wilhelmy, Swiss Re

But many insurers are simply unsure what to make of the IAIS’s latest announcement and the delay in implementation of international capital rules.

The chief risk officer at one G-Sii says: “Currently, the only implications of being a G-Sii are if they have ICS and HLA in place; until that, there are none. If they’re moving towards systemically important activities, presumably they might walk away from the systemically important companies, or maybe they’ll designate companies with big activities that are deemed to be systemic, we just don’t know yet.”

Even representatives at some supervisors say they “cannot tell” whether a list of G-Siis would last by the end of the process, although they point to the suggestion that entity- and activities-based regulation will be conducted in parallel. “How would you apply an HLA on the basis of an ICS if it is not entity-based?” asks one.

“At this stage, the [activities] framework is complementary. The IAIS is not going to abandon G-Siis. The FSB is happy with the list, and replacing the list would require FSB approval,” says a European lobbyist.

More regulation or less?

The range of interpretations of IAIS plans encompasses some who think they could even lead to more regulation, not less.

Rob Curtis, global insurance regulatory head at KPMG Australia, who led the UK Financial Services Authority’s Solvency II programme, thinks activity-based regulation will be the start of a system of national regulation of smaller insurers.

“It’s probably a natural extension of the G-Sii route,” he says. “Most supervisors will probably start to say who is domestically significant for them – not necessarily globally significant, but significant for a particular [home] market. They might be some very niche players… As with all things regulatory, it starts with the larger players, then it filters down.”

A market with two or three providers of a key type of insurance might mean they are deemed domestically significant, for example. Curtis points to other niche insurers, such as professional indemnity insurers, that may not have to raise capital, but might need to show recovery and resolution planning.

Every market has its own systemic risk concerns, he says: “If there were difficulty with any of the large annuity providers, [for example], that would likely cause some grief for the UK government.”

Many non-G-Siis will have been pressed by national regulators to plan for recovery and resolution, even if this has not been made public, he thinks: “Recovery analysis is now thought of as just good risk management.”

Whether entirely clear or not, the IAIS’s intentions are being welcomed by the industry as a step towards a more nuanced assessment of systemic risk.

Tobias Buecheler, head of regulatory strategy at Allianz, says: “Allianz has been supportive of an activities-based approach for a while. I think the perspective will change from saying this [insurer] is a big ugly gorilla to [asking] which part of the animal will do harm.”

Buecheler points to the US savings and loans crisis in the 1980s as an example of how groups of smaller enterprises might collectively create systemic risk. “It was the biggest state-sponsored bank bailout before the recent financial crisis, with over $500 billion of taxpayers’ money poured into those mid-sized savings and loans banks,” he says. “They all had essentially the wrong business model. None of them individually as a company would have mattered, but all of them together were a major problem.”

At Insurance Europe, Cristina Mihai, head of prudential regulation, says: “We tentatively support an activities-based approach, but are waiting for more details. We have always been against an entity-based approach. Systemic risk in insurance comes from certain activities done in a certain way, not from size or interconnectedness.”

Development timeline

Meanwhile, the IAIS has laid out a generous timeline for development of its activities-based methodology.

The assessment methodology will be available for public consultation by the end of 2018, with adoption forecast for 2019. At this point, the methodology will be incorporated into IAIS’s HLA requirements as part of ICS Version 2.0, which will be implemented beginning in 2022.

That might still prove ambitious if barriers to the ICS’s development cannot be surmounted. “It’s still a mystery to me where the IAIS is at the moment. I’m not sure how you can promulgate an HLA in the absence of a global framework. I can’t see that happening any time soon,” says KPMG’s Curtis.

“I don’t see how different regulators are going to implement something which is un-implementable at the moment.”

When it turns out the activities-based systemic risk assessment is superior – and I’m convinced it will be superior – then it should really replace [designations]

Lutz Wilhelmy, Swiss Re

“First of all, we have to get the ICS right. To base the HLA on something that isn’t there is like buying the cat without seeing it,” says Swiss Re’s Wilhelmy. “Everyone in the IAIS has to agree with it.”

“Confidential reporting for ICS Version 1 should have started by the middle of this year, and late in the process that was [changed] to extended field testing, which means we’re in another loop. There is growing scepticism over valuation in the ICS, and internal models need to be addressed.”

“You cannot expect a score of European insurers using internal models to willingly adopt a standard approach, and you cannot expect the European Union regulators that have just approved internal models for Solvency II to disallow them for ICS,” says Wilhelmy, who sees a lot more conflict ahead.

“I think the ICS is currently on life support, anyway. We will get a review of Solvency II in 2018, so the appetite for any lawmakers in Europe from 2021 to switch to ICS in 2022 may be quite limited,” he says.

However, IAIS secretary-general Kawai asserts the body’s plan is clear. The activity-based approach will be developed and incorporated with the entity-based approach by 2019, after which, it will be used to help the FSB create a new list of systemically relevant companies in 2020. The ICS and HLA will apply to those G-Siis in 2022.

Kawai says completing these two work streams – the activities-based approach and the HLA – will be key to the finalisation of the ICS, which he says “is now the centre of the universe of our activities”.