Skip to main content
  • This website sets cookies on your device. To find out more about how we use cookies please refer to our Privacy and Cookie Policy. By continuing to use the site, we’ll assume that you are content for us to set these on your device.
  • Close
Home > News and Publications > International Capital Standards for Insurers - speech by Victoria Saporta
 

International Capital Standards for Insurers - speech by Victoria Saporta

12 November 2015

​Given at the Insurance ERM, London on 5 November 2015

All references and charts relating to this speech can be found in the speech PDF.

Thank you for inviting me to your annual conference.  It is a pleasure to be here. 

In my talk today I plan to cover some of the main international initiatives on insurance capital regulation.  I will start with the case for an international insurance capital standard before moving to basic capital requirements (BCR) and higher loss absorbency (HLA).
 
The case for international insurance capital standards
 
The International Association of Insurance Supervisors (IAIS) has embarked on a journey of delivering the first international capital standard (ICS) for internationally active insurance groups (IAIGs) across the world.  It issued a consultation paper at the end of 2014 and plans to publish proposals on the first version of the ICS by mid-2017 for implementation to IAIGs from 2020 after refinement and final calibration in 2018 and 2019. This forms part of a revamped common framework for the regulation of IAIGs (Comframe). 
 
But why do we need an international capital standard for IAIGs?  
 
Banking regulators have been developing minimum banking capital standards to level the playing field for internationally-active banks for the best part of 30 years.  But this comparison must be surely irrelevant, if one thought that that the insurance sector was more domestically-orientated than banks. 
 
But is it?
 
To answer this question, it is useful look at the research and the data.
 
Using data from 2000, an early study examined the 53 largest financial groups worldwide and found that insurance groups were more international than banking groups measured by the geographic source of their revenues (van der Zwet (2003)).  Employing 2005 data, another study found that the 25 top insurance groups in the EU were more internationally orientated than the top 30 banking groups with on average 54% of their activities conducted outside their home countries, compared to 45% for banks (Shoenmaker et al (2007)). 
 
And the trend towards greater internationalisation amongst EU global insurance firms seems to be continuing unabated by the financial crisis (in contrast to banking1).  A recent and comprehensive study (Shoenmaker and Sass (2014)) finds that in aggregate the top 25 insurers in Europe wrote 58% of their premiums outside their home country in 2012, up from 50% in 2000.  The same study reports that these insurers write less than 50% of their premiums in their home country, with nine of them writing the majority of their international premiums outside the EU.  Looking at the EU as a region and at the top 20 EU companies, 27% of the premiums written by the largest EU insurers in 2012 come from outside the EU, up from 22% in 2000.  This compares with a more volatile series from North America, centring around a roughly comparable 23% of the gross premiums of the top 20 companies being written outside the region.  By contrast, almost all of the gross premiums written by the top Asian insurance firms is written within the region (Chart 1).
 
The figures quoted so far look at the international activity of firms from the perspective of the group’s home regulator.  Looking at the data from a host perspective, we find that the amount of gross premiums, in inflation adjusted terms2, written by foreign-controlled entities in particular regions, has increased over the last 25 years in all regions of the world 3 4    (Charts 2 and 3), with the more striking increases having occurred in Asia.
 
More specifically, foreign- controlled entities write a significant share of non-life business in each region, ranging from 10% to 30% depending on the region.  This is not surprising given the opportunities of non-life insurers to diversify risk across regions, as well as by type of risk (Chart 4).   What is perhaps more surprising is that this degree of openness to trade is also reflected in life insurance, where differing national tax and welfare regimes may have been thought to dampen the scope for foreign expansion (Chart 5). 
 
In summary, existing studies have found that the largest insurance companies in the world tend to be more international than the largest banks in the world and that in Europe the trend towards greater internationalisation has been growing steadily over the past decade.  At the same time, host supervisors have seen a growing share of policies written in their countries by foreign-controlled entities.
 
It is also plausible to expect that internationalisation is likely to grow as global companies seek to exploit profit opportunities in growth markets elsewhere in the world.  Swiss Re Economic Research and Consulting has found that an S-curve is a good way to illustrate the average relationships between economic development and insurance penetration.  More specifically, insurance spending as a share of GDP grows in line with income for low income countries and for high income countries (tracing the two flat parts of the S-curve), on average, but faster than income for economies with GDP per capita in the range of $10,000 to $35,000 (tracing the increasing part of the S-curve), suggesting more attractive growth opportunities in these areas of the world (Chart 6).  
 
Is internationalisation to be welcomed?  I would say “yes, but…”
 
The internationalisation of insurance business could give rise to substantial risk-sharing benefits.  After all, the underlying rationale for insurance provision arises from the law of large numbers and the efficiency benefits from pooling and diversification of risk.  Competing companies that have the ability to access a larger number of policyholders in international markets are able to diversify risks for all policyholders, domestic and foreign, lowering costs and enabling more efficient consumption smoothing across states of the world and across time. 
 
Here’s the “but”.  As the financial crisis illustrated with great force, financial internationalisation needs to be supported by a robust international regulatory framework so as to mitigate the risk of shocks buffeting operations of global firms in one part of the world spilling over and threatening safety and soundness and policyholder protection in other parts of the world.  A key element5 of such a framework is an international capital standard – a common metric to measure and assess solvency in support of safe financial integration and cooperation.  If anything the trends shown above, suggest that this development is rather overdue.
 
An appropriately designed international capital standard will furnish home and host supervisors with a common language to assess the solvency of insurance groups operating in their jurisdictions, enhancing supervisory cooperation between home and host supervisors and helping to build trust.  Such a standard should also enable investors to assess and compare the solvency of insurance firms competing internationally, enhancing market discipline and reducing regulatory arbitrage.  Over time, such a standard should also reduce the need for multiple overlapping and conflicting local practices in measuring the same risks, reducing costs for firms, which in turn should deliver a better proposition for policyholders. 
 
The need for such a standard is independent of the separate debate of whether certain insurance firms or the sector as a whole can pose systemic risk.  It is much more basic than this.  It’s a need for home and host regulators, the market and rating agencies, to be able to compare apples with apples rather than apples with oranges and hence to help achieve a level-playing field for insurance firms operating on a global scale. 
 
Developing such a single minimum standard is no easy task, given the different starting points and regulatory capital regimes in place across different jurisdictions.  This is why the IAIS has decided to deliver the ICS in stages with identified key milestones: version 1, version 2 and an ultimate goal.  Version 1 of the ICS is planned for completion by the middle of 2017.  It will set out a standard method for calculation of the capital requirements for confidential reporting to supervisors.  Following this, version 2 of the ICS is planned for completion in 2018 and aims to improve the level of comparability of version 1.  Some aspects of the methodology, such as the valuation basis, may however still contain differences between jurisdictions.  ComFrame will include the implementation of the ICS version 2 for IAIGs, and is planned for adoption in the IAIS General Meeting in 2019.  In the longer term, the ultimate goal is the development of a single ICS to apply across jurisdictions.
 
It is worth noting that since the IAIS is an international standard setting body, similar to the Basel Committee on Banking Supervision (BCBS) and the International Organisation of Securities Commissions (IOSCO), it has no role in enforcing implementation of the ICS.  The adoption of the ICS will therefore be the responsibility of national authorities, supported by peer reviews.  Further details of how the IAIS foresee the ICS to be implemented as a minimum standard will be covered in future consultations.
 
Insurance and systemic risk: GSIIs, BCR and HLA
 
As I noted earlier, the case for the development of a common metric to assess capital adequacy for internationally active firms does not rely on whether some internationally active firms pose or can amplify systemic threats.  Most of the public debate on the subject of global capital standards, however, has focussed on this issue. 
 
This is unsurprising.  The lessons from the circumstances under which AIG and a number of other insurance groups in the US had to access public funds are well known and so are the lessons from the contribution of financial guarantor monolines in exacerbating the crisis.6   Given the circumstances, it was natural for the FSB to ask the IAIS to provide it with a methodology that helps identify insurers whose failure or distress could have systemic consequences – the so-called Globally Systemically Important Insurers (GSIIs) – and then to develop international capital standards at group level for them.
 
Colleagues from the Bank of England have recently examined how the failure or distress of one or more insurers can have systemic consequences.  They set out the transmission channels through which insurance can cause adverse effects to financial stability (French et al (2015)).  Diagram 1 below, reproduced from their paper, provides a summary view of their analysis.  First, the distress or failure of one or several insurance companies could affect financial stability if it led to the disruption of critical services that insurance companies provide (Diagram 1, first column).  The strength of this channel depends on the degree of substitutability between providers, which is lower if market shares are highly concentrated.  Insurers could also contribute to systemic risks if they behaved in ways which propagated financial system shocks.  This could occur if their actions threatened the resilience of one or several systemically important financial counterparties (Figure 1, column 2) or if their actions contributed to the disruption of the functioning of systemically important financial markets (Figure 1, column 3).  The article contains data and case studies that suggest that both of these channels have been operative in the past and might be operative in the future.
 
This analysis is, in my view, entirely consistent with the methodology adopted by IAIS in identifying GSIIs based on criteria including size and substitutability, interconnectedness with other significant financial institutions and markets, as well as the extent of potentially systemically activities taken7.  That said, and as the IAIS have made clear, methodologies for GSII identification can be improved over time, as both data and our understanding of the drivers of systemic risk improves.  This is the reason why the IAIS intends to publish a proposal for an improved and revised identification methodology later this year.
 
Before I conclude, it is worth recognising the progress that has already been made by the IAIS towards the development of group-wide international capital standards for GSIIs: 2014 and 2015 respectively saw the completion of the Basic Capital Requirement (BCR) and the Higher Loss Absorbency (HLA) requirements for GSIIs8 9.  These are two important milestones.  Together they provide a globally comparable group capital requirement for GSIIs, designed to appropriately target those activities deemed to be of greater systemic relevance.  The HLA augments the BCR to increase requirements in respect of activities deemed systemically important.  Once completed, the ICS will replace the BCR as the basis for the HLA. 
 
The development of the ICS is already well under way, with the first field-testing exercise being completed by volunteer firms in September this year10.  The work of the IAIS is not, however, confined to the group of GSIIs and IAIGs.  The Insurance Core Principles (or ICPs) apply to insurance supervision in all jurisdictions regardless of the level of development of the market and the types of products on offer.  They set out a globally accepted framework for the supervision of the insurance sector in its entirety, prescribing the essential elements that must be present in the supervisory regime in order to promote a financially sound insurance sector and provide an adequate level of policyholder protection.
 
Conclusions
 
Insurance is a global business.  This benefits us all.  The recent financial crisis, however, once again brought the collective nature of our economic fate into sharp relief.  The increasing internationalisation of insurance firms and their capacity to transmit risks across borders has reinforced the need for a comprehensive set of global standards for the insurance sector. 
 
The IAIS has made remarkable progress in the past few years.  It has developed a methodology to identify globally systemically important insurers, and to apply higher loss absorbency requirements to activities that could pose systemic risk.  Its next goal is to develop a common standard for capital requirements to apply to internationally active insurance groups.   The next few years will be just as crucial a period for securing the many benefits of internationalisation.

 

1 BIS (2015) Quarterly Review – International banking and financial market developments, September 2015
2 2010 prices.
3 Subsidiaries and branches.  Note data for Latin America and the Caribbean does not extent back to 1989.
4 Interestingly, from a jurisdiction or host regulator perspective as well, and looking only at the EU, the evidence suggests that insurance (measured by the share of premiums written by foreign-controlled branches and subsidiaries) is more international than banking (measured by the share of assets written by foreign-controlled branches in subsidiaries (Shoenmaker and Sass (2014) – report a figure of 36% for insurance vs 25% in banking).
5 Other important elements are robust judgement-based supervision, policyholder compensation schemes and recovery and resolution plans.
6 Carney (2014) makes similar points.
7 IAIS (2013), Global systemically important insurers: initial assessment methodology, Policy paper, available at http://iaisweb.org/index.cfm?event=openFile&nodeId=34257
8 IAIS (2014), Basic Capital Requirements (BCR) for Global Systemically Important Insurers (G-SIIs), Policy paper, available at http://iaisweb.org/index.cfm?event=openFile&nodeId=34540
9 IAIS (2015), Higher Loss Absorbency (HLA) requirements for G-SIIs, Policy paper, available at http://iaisweb.org/index.cfm?event=openFile&nodeId=57131
10 IAIS (2015), Risk-based Global Insurance Capital Standard (ICS), Consultation document, available at http://iaisweb.org/index.cfm?event=getPage&nodeId=56993

Share