sigma
No 7/2006 Securitization – new opportunities
for insurers and investors




3 Executive summary

4 Introduction

8 Characteristics of life insurance
securitizations

17 Characteristics of P & C
securitizations

24 Market overview

31 Prospects

37 Appendix




New sigma premium figures:
www.swissre.com/sigma
sigma No 5/2006,
December update
Published by:
Swiss Reinsurance Company
Economic Research & Consulting
P.O. Box
8022 Zurich
Switzerland

Telephone +41 43 285 2551
Fax +41 43 285 4749
E-mail: sigma@swissre.com

New York Office:
55 East 52nd Street
40th Floor
New York, NY 10055

Telephone +1 212 317 5135
The editorial deadline for this study was
Fax +1 212 317 5455
7 November 2006.
Hong Kong Office:
sigma is available in English (original lan-
18 Harbour Road, Wanchai
guage), German, French, Italian, Spanish,
Central Plaza, 61st Floor
Chinese and Japanese.
Hong Kong, SAR
sigma is available on Swiss Re’s website:
Telephone +852 2582 5691
www.swissre.com/sigma
Fax +852 2511 6603
The internet version may contain slightly
Authors:
updated information.
Rainer Helfenstein
Telephone +41 43 285 5977
Translations:
Swiss Re Group Language Services
Dr Thomas Holzheu
Telephone +1 212 317 5190
Graphic design and production:
Swiss Re Logistics/Media Production
Editor:
Dr Kurt Karl
© 2006
Telephone +1 212 317 5564
Swiss Reinsurance Company
Zurich
sigma co-editor:
All rights reserved.
Aurelia Zanetti
Telephone +41 43 285 2544
The entire content of this sigma edition is
subject to copyright with all rights reserved.
Managing editor:
The information may be used for private
Thomas Hess, Head of Economic Research
or internal purposes, provided that any
& Consulting, is responsible for the sigma
copyright or other proprietary notices are
series.
not removed. Electronic reuse of the data
published in sigma is prohibited.

Reproduction in whole or in part or use
for any public purpose is permitted only
with the prior written approval of Swiss Re
Economic Research & Consulting and if
the source reference “Swiss Re, sigma
No 7/2006” is indicated. Courtesy copies
are appreciated.

Although all the information used in this
study was taken from reliable sources,
Swiss Reinsurance Company does not ac-
cept any responsibility for the accuracy
or comprehensiveness of the information
given. The information provided is for in-
formational purposes only and in no way
constitutes Swiss Re’s position. In no
event shall Swiss Re be liable for any loss
or damage arising in connection with the
use of this information.
Executive summary



Securitization is a financial technique that pools assets together and, in effect,
Securitization transforms assets into
tradable securities. turns them into tradable securities. Financial institutions and businesses of all
kinds use securitization to reduce their capital requirements and immediately
realize the value of cash-producing assets by creating securities separate from
the debt and equity securities of the sponsor. Securitization has evolved from its
beginnings in the 1970s and is today used heavily by banks along with deriva-
tive techniques that have evolved from securitization. Securitization in the insur-
ance industry is relatively new, with the first transaction having taken place in
the early 1990’s. The number of transactions has increased steadily in recent
years, as has the use of associated derivative techniques such as cat swaps.

Life insurers are securitizing parts of their business, improving return on equity
Life securitizations provide many benefits
to their sponsors, including improved through capital efficiency. By selling risks to investors, insurance companies re-
profitability.
duce their need to hold capital and increase their ability to write new business.
Some securitization structures also provide ancillary tax advantages. Securitiza-
tion allows (re)insurers to focus on underwriting, structuring and passing risks
directly to the debt capital markets, as well as improving their return on equity
(ROE). Ultimately, (re)insurers should be able to share these benefits with their
clients, increasing the demand for transferring risks. Life bonds also monetize in-
tangible assets, fund regulatory capital requirements, and transfer catastrophic
risks, eg mortality, to the bond market.

Catastrophe bonds, the primary P & C securitization approach to date, provide
P & C securitizations to date mostly transfer
extreme risks to the debt capital markets. many benefits to re(insurers) as a source of (re)insurance capacity. They serve
as collateralized protection for extreme event risk, which eliminates counter-
party risk, at a multi-year fixed price. In addition, they augment traditional ca-
pacity, since cat bond investors for the most part do not also provide traditional
(re)insurance protection.

Fixed-income investors are increasingly interested in insurance-linked securities
Interest in the market continues to grow
within the investment community. and related risk-taking, because they provide exposure to specific insurance
risks, such as the risk of an earthquake in a specific area, resulting in a “pure
play” investment; have their funds held in trust, so the investor faces no counter-
party risk with the bond’s sponsor, the (re)insurer; and have a low correlation
with equity and credit markets, making them a diversifying asset class.

In the past five years, the outstanding volume of P & C securities has doubled,
The overall market is expected to grow
very rapidly over the next 10 years. while the volume of life bonds has tripled, taking the total outstanding volume of
securities to about USD 23 billion. In the past two years, issuance has accelerated
rapidly. Primarily driven by the funding needs of the life industry, the volume of
outstanding ILS is expected to grow to USD 150 to 350 billion by 2016. The fu-
ture size of the market is uncertain due to its nascent state and because there
are other ways for investors to provide capital to the (re)insurance industry. Nev-
ertheless, ILS and related solutions will become an increasingly important source
of capacity for the (re)insurance industry.




3
Swiss Re, sigma No 7/2006
Introduction



Insurance-linked securities (ILS) are a means of ceding insurance-related risks to
There are P & C and life insurance-linked
securities, cat and non-cat bonds. the capital markets. In this sigma, ILS are categorized in two ways: by risk type –
property/casualty (P & C) and life risks – and by catastrophe (cat) and non-cat
risk. To date, P & C bonds have tended to transfer peak risks, while life bonds
have usually provided financing backed by future premium flows, though some
also transfer peak mortality risks. A cat bond transfers the risk of extreme events
– such as hurricanes and earthquakes in densely populated areas or sharp in-
creases in mortality – to the capital markets. A non-cat bond, usually for life in-
surance books of business, is more of a financing vehicle.

P & C bonds originated in the hard market of the early 1990s, after hurricane
The first P & C cat bond was issued after
hurricane Andrew. Andrew, when reinsurance capacity for catastrophes was limited and expensive.
The earliest forms provided a simple mechanism to transfer catastrophic risks to
capital markets, easing industry capacity constraints. In a typical transaction, a
special purpose vehicle (SPV) enters into a reinsurance contract with a cedent
and simultaneously issues cat bonds to investors.¹ The reinsurance is usually an
excess of loss contract. If no loss event occurs, investors receive a return of prin-
cipal and a stream of coupon payments that compensate them for the use of
their funds and their risk exposure. If, however, a pre-defined catastrophic event
does occur, investors suffer a loss of interest, principal, or both. These funds are
transferred to the protection buyer or cedent, in fulfillment of the reinsurance
contract.

Most life bonds differ from P & C bonds in a very crucial respect – they are usually
Life bonds are usually a financing tool.
a financing tool. Life bonds typically securitize the flow of future premium pay-
ments of traditional life insurance policies. In a legal sense, risk is not fully trans-
ferred, since the life insurance company will always retain the obligation of its
policies. However, the burden of life insurance risks, such as mortality and lapse
risk, are assumed by the investors. For these bonds, investors and protection
buyers share the benefits and losses in the development of the underlying poli-
cies which have been securitized.

Some life bonds are cat bonds – transferring extreme risks to the capital markets.
Some life bonds transfer extreme
mortality risk. These bonds are very similar to P & C cat bonds and, so far, have been based on
mortality indexes, though longevity bonds as well as morbidity bonds are also
feasible. If mortality develops as expected, the investors collect the designated
interest and, at maturity, the principal is returned to them. On the other hand, if
mortality increases substantially, triggering the bond, the investors suffer a loss
of interest, principal, or both.




¹ A SPV is also referred to as a “bankruptcy-remote entity” whose operations are limited to the acquisi-
tion and financing of specific assets. The SPV is usually a subsidiary company with an asset/liability
structure and legal status that makes its obligations secure even if the parent company goes bankrupt.




4 Swiss Re, sigma No 7/2006
 Structuring basics

In a typical ILS structure as used for cat bonds, the Reinsured enters into a finan-
cial contract with a Special Purpose Vehicle (SPV), (see (1) in Figure 1). The SPV
hedges the financial contract by issuing notes to investors in the capital markets
(see (2) below). Proceeds from the notes are invested in high-quality securities
and held in a collateral trust (3). Finally, investment returns are swapped to a
LIBOR-based rate by the Swap Counterparty (4).


Figure 1
Investments 3
Typical ILS structure

Investment Principal and interest
earnings
1 2
Premium Notes
Investors
SPV
Reinsured
Contingent Cash proceeds
claim payment
Investment Scheduled
earnings interest
Swap
4
Counterparty

Source: Swiss Re Capital Markets




Some ILS benefit from the participation of a credit insurer. Many life securitiza-
Investors can be protected from the credit
quality of the sponsoring company and tions include credit protection, in which a bond insurer (often called a “monoline”
the underlying insurance risk by credit
or “wrapper”) guarantees the interest and principal payment on the underlying
insurance, and …
securities. The guarantee backs up the security with a high rating (usually AAA).
The issuing company pays a premium to the bond insurer for providing the credit
enhancement, or credit “wrap”.

There are two common transaction structures, “accounting only” and “legal sep-
… by legal separation.
aration.” In accounting only, the business continues to reside in the insurance
company which directly issues the bond. More typically, however, the bonds are
issued on a non-recourse basis, with legal separation. In this case, the business
is reinsured to a bankruptcy-remote special purpose vehicle (SPV), which issues
the debt. In other words, if the sponsoring insurance company goes bankrupt,
the bond is not affected, since the underlying assets, usually AAA government
bonds, are in the SPV. If the bond has no credit wrap and is subject to recourse,
the investor bears the risk of the sponsoring company becoming insolvent.




5
Swiss Re, sigma No 7/2006
Introduction




Transparency for investors
Figure 2
The transparency and basis risk
for various types of triggers
Pure
Parametric parametric
index


Modeled
Industry
loss
index
MITT

Indemnity


Basic risk to issuer

Source: Swiss Re Capital Markets




While investors prefer to maximize the transparency of the trigger for cat bonds,
There are six types of triggers for cat bonds.
sponsors would like to minimize the basis risk. Transparency, however, as shown
in Figure 2, is often associated with higher basis risk. Cat bonds have used a va-
riety of triggers to manage this tradeoff:
 An indemnity transaction is based on the actual losses of the sponsor.
 An industry index transaction is based on an industry-wide index of losses
(eg, Property Claim Services (PCS) in the United States).
 In a Modeled Industry Trigger Transaction (“MITT”)², industry index weights
are set post-event using modeled loss techniques.
 A pure parametric trigger is based on the actual reported physical event
(eg, magnitude of earthquake or wind speed of hurricane).
 A parametric index is a more refined version of the pure parametric trigger
using more complicated formulas based on more detailed measurements.
 In a modeled loss transaction, losses are determined by inputting actual
physical parameters into an agreed-upon, fixed model which then calculates
the loss.

Trigger types vary for non-cat transactions. For an embedded-value transaction,
“Triggers” for non-cat bonds are generally
based on developments of the underlying for example, the trigger is the quality of the premium payments and investment
policies.
returns relative to the expected payment flow. For example, losses occur when
mortality rates are higher than expected and lapses are earlier than expected.

Over the years, the type of triggers for cat bonds has shifted from indemnity and
parametric index triggers to modeled loss and industry index triggers, although
this trend has stabilized in recent years, reflecting the varying needs of potential
transaction sponsors.




² MITT, patent pending, was developed by Swiss Re.




6 Swiss Re, sigma No 7/2006
 The costs of securitizations can be divided into capital costs and structuring
The cost of a securitization consists of
capital and structuring costs. costs. The capital costs are the costs at which the issuing company is able to
raise capital, generally at LIBOR plus a risk premium. The risk premium depends
on the nature of the underlying secured business or risk as well as the structure
of the security. It will be lower for business with a low risk profile (such as level-
premium term life or fee business, eg variable annuities) and higher for more
complex businesses (such as universal life). Likewise, the risk premium of cat
bonds will be commensurate with the probability of the bond being triggered.

The structuring costs are composed of fees to lawyers and the costs of advisors,
Structuring costs are the legal, advisory,
actuarial and other service fees and per- actuarial consultants, rating agencies, structuring and placing the security. Gen-
sonnel costs to create the security.
erally, the average costs and fees for lawyers, actuarial consultants and advisors
are subject to the complexity and size of the transaction and will decline with
the number of transactions done. Expenses for rating agencies and securities
structuring and placement, in contrast, may be unrelated to the number and size
of the transactions. Expertise in the field of structuring and placement of insur-
ance securitizations is scarce and may represent a significant portion of the costs.

The success of a securitization can also depend on the jurisdiction chosen, crite-
It is important to choose the jurisdiction
that best fits the transaction. ria being the type of security and the tax and regulatory treatment. For example,
South Carolina (US) is often used for Triple X securitizations, because these
bonds are based on U.S. life business and South Carolina allows organizations
to form special-purpose financial captives.

(Re)insurers with a large book of business can use the strength of their balance
Transformation benefits all parties.
sheet to transform risks via securitization. Risks are assumed according to the
clients’ needs and can be securitized in different tranches and/or different trig-
gers in accordance with investors’ risk appetite. The “transformer” retains basis
risk or (lower) layers of risk that diversify well with their more traditional book of
business.

Transactions need to be large to be economical. For the life bonds, transactions
Transaction size matters.
are complex and need to be at least USD 200 million to complete the transac-
tion efficiently. P & C bonds are often smaller, but still need to be about USD 100
million. Nonetheless, as with securitizations of other instruments, as the market
matures smaller transactions become feasible.

This sigma updates and expands on an earlier sigma on ILS.³ This sigma exam-
Plan of this sigma
ines developments in the P & C and life insurance securitization markets, focus-
ing on sponsor motivation. It clarifies recent market trends for ILS and explores
the interests of investors. Finally, it considers the prospects for the development
of the market for ILS and related solutions. Because the life bonds have been on
the market for a shorter time and their issuance has recently surged, this sigma
focuses more heavily on life bonds.




³ Swiss Re sigma No 3/2001, “Capital market innovation in the insurance industry”.




7
Swiss Re, sigma No 7/2006
Characteristics of life insurance securitizations



Securitization of life insurance business is a capital management tool for
Securitization is a capital management
tool ... (re)insurers. Securitizations allow a (re)insurer to:
 monetize intangible assets via embedded-value (EV) securitizations,
 fund regulatory capital requirements, or
 transfer catastrophe risks.

Life securitizations have various positive benefits. Embedded value and funding
… providing many benefits.
of capital requirement transactions increase the return on equity (ROE). These
transactions also may provide tax advantages by deferring realized income into
the future. Mortality bonds protect (re)insurers against extreme catastrophic
events, such as a pandemic or widespread death from any other cause.


Embedded-value (EV) securitizations

Securitizations can monetize intangible assets, such as deferred acquisition costs
Securitization can monetize intangible
assets, such as deferred acquisition (DAC) and the present value of future profits (PVFP). To acquire new business,
costs, and ...
insurers must pay commissions to agents or brokers. A part of these acquisition
costs are generally capitalized as an asset, or “activated.” In the following years,
part of the premium paid by the policyholder is used to write down the deferred
acquisition costs. A securitization allows an insurer to monetize their DAC, since
the risk of early lapse is partially transferred to investors.

When a company acquires a life insurer or a closed life insurance portfolio, the
… the present value of future profits.
acquiring company usually activates a part of the present value of future profits
of the acquired company. A securitization allows an insurer to monetize the
present value of these future profits. These securitizations are viewed differently
by internal accountants, regulators and rating agencies (see box: The account-
ing, statutory and rating agency views on PVFP).

This type of transaction provides financing. It can finance the cash strain associ-
These securitizations provide financing
and can improve ROE. ated with writing new business; generate cash for further acquisitions; and free
up capital from life insurance business for other corporate purposes, eg, an ac-
quisition, share buyback, or writing P & C business in a hard market. The financing,
appropriately structured, can also increase the return on equity, provided the
bond is issued at a favorable interest rate, below the return on the book of busi-
ness securitized (see box: How can securitization increase ROE?).




8 Swiss Re, sigma No 7/2006
 The GAAP accounting, statutory accounting and rating agency views
on PVFP

GAAP accounting, statutory accounting and rating agencies have differing
views on the present value of future profits:
GAAP accounting view: Currently, the two dominant international accounting
frameworks (IFRS and US GAAP) allow companies to recognize the present
value of future profits (PVFP).
Solvency view: In the EU, for example, life insurers are able to include part of
the PVFP in the calculation of their available solvency capital.⁴ In the EU, 50% of
the PVFP net of taxes can be used for available solvency capital. However, this
amount must not exceed the lower of 25% of the required solvency capital or
the average profits of the last 5 years multiplied with the factor 6. From 31 De-
cember, 2009 onward, life insurers will not be able to use the PVFP item for
their solvency capital calculation. Some market observers believe that this may
increase the attractiveness of EV securitizations, since companies will even have
more incentives to replace the PVFP item on their balance sheet with a cash
item.
Rating agency view: Rating agencies have a view of securitizations similar to
insurance regulators. For example, the S & P rating model also takes 50% of the
PVFP net of taxes into account for the rating capital calculation. A monetization
of PVFP will increase the rating capital and can improve the credit standing of a
life (re)insurer.


Because seasoned books of life business provide transparency for investors,
Investors bear risks, such as mortality
and lapse risk, ... they are very comfortable with EV securitizations. However, investors do bear in-
surance risks with these bonds. For example, investors partially bear the mortal-
ity and lapse risk of the securitized block of policies. On the one hand, if the
cash flows from the policies are less than expected, due to higher-than-antici-
pated mortality or early lapses, then the payments to investors will be, delayed
and possibly reduced. In the extreme, for example with a sharp increase in mor-
tality, the payments are reduced to zero. Other risks may also be transferred to
the investors, depending on the type of policies covered. For example, some in-
terest-rate risk may be transferred to investors if some of the policies securitized
include interest-rate guarantees and the set-aside assets do not fully match
these liabilities. On the other hand, if mortality rates and lapse rates prove to be
more beneficial to the cash flows than expected, the investor receives the full
value of the expected flows, but the life insurance company captures all the ex-
cess benefits.




⁴ Switzerland has adopted this EU Directive. Therefore the same principles apply in Switzerland.




9
Swiss Re, sigma No 7/2006
Characteristics of life insurance securitizations




The investors’ risk can be mitigated through the provision of financial guarantees,
… but these risks can be mitigated.
over-collateralization, and other standard credit enhancement techniques for as-
set-backed securities. For example, with over-collateralization, the insurer
pledges excess cash flows which are first affected by adverse developments
affecting the cash flow. After this layer of the security is depleted, the lowest lay-
er held by investors is affected next, and so on. If the excess cash flows are not
needed, a residual is effectively recaptured with bond repayment.


How can securitization increase ROE?

In an embedded-value transaction, the issue can achieve a higher ROE through
earning a spread. The underlying business was written at a certain internal rate
of return (eg 10%). The sponsoring company passes a lower return on to inves-
tors (eg 7%). The sponsoring company increases its profitability by earning the
spread. However, achieving a consistently higher ROE requires continuous deal
flow and accurate pricing.

In addition to this, an EV securitization will generally increase the available sol-
vency capital, facilitating share repurchases (increasing ROE) or other investment
undertakings (such as new business financing, acquisitions, etc.). Also, the
overall capital management may become more efficient, assuming the cost of
securitization is below the cost of raising new capital or issuing hybrid debt.


A typical embedded-value securitization

A life insurer seeks to securitize a USD 400 million block of business. This block
Insurers use securitization for capital relief
in pursuing their growth strategies. includes a few business lines, such as traditional life, interest-sensitive life, de-
ferred annuities, and corporate-owned life insurance. The insurer seeks capital
relief to pursue its growth strategy. A structured solution is created with losses
first affecting the retained residual, then the BB layer, then the BBB layer, etc.
The retained residual ensures that the sponsor suffers the first loss on the bond,
providing protection to the investors owning the higher layers of the bond. The
security provides risk transfer, reducing the required reserves the insurer holds
and allowing the pursuit of new business. The insurer receives money today in
return for the flow of premiums over the life of the policies. The investors receive
an income flow with a reasonable risk/return relationship. The flows include in-
terest payments and the return of principal over the life of the bond. This asset
helps diversify the investor’s fixed-income portfolio, since the bond is uncorre-
lated with equity or conventional fixed-income securities. However, the investor
must be familiar with the risks involved, which include mortality risk, lapse risk,
and sometimes other risks, such as interest-rate risk with the interest-sensitive
life policies.




10 Swiss Re, sigma No 7/2006
Figure 3
A typical embedded-value securitization
structure
AAA 200 mm/50%
wrapped




AAA 100 mm/25%
wrapped

30 mm/7.5%
BBB
30 mm/7.5%
BB
40 mm/10% Retained




Securitizations that fund regulatory capital requirements

In the US, securitizations can be used to mitigate pressure due to stringent stat-
In the US, securitization is used to fund
regulatory capital requirements. utory reserve requirements. Many companies have used securitizations to lower
their reserve strain. With this approach, “redundant” reserves are securitized.

Triple X/AXXX reserves securitization. In the US, companies and investors re-
The reserve requirements for XXX and
AXXX business are considered to be gard the reserves required by Regulations XXX (for level-premium term life) and
excessive.
AXXX (for universal life with guarantees) to be much higher than economically
justified. One way to alleviate the problem is to issue a bond equal in value to
the redundant reserves (the difference between the statutory reserves and what
would be economically justified) from a special purpose vehicle (SPV). The in-
vestors’ funds reside in the SPV and serve as collateral for the bond and the
redundant reserves. In other words, instead of the life company providing the
collateral for the reserves, it is provided by investors.


Triple X and AXXX business in the US

In the US, the Valuation of Life Insurance Policies model regulation (better known
Guideline Triple X, for level-premium term
life products, requires excess reserves. as Guideline Triple X, or XXX) became effective in 2000 and requires the pre-
funding of future liabilities of term life products with a guaranteed, or level, pre-
mium. However, the required mortality assumptions are outdated and do not
reflect today’s best practices. Companies writing term life business need to hold
significantly more capital to cover these conservative estimates than is econom-
ically necessary. The National Association of Insurance Commissioners (NAIC)
lowered reserve requirements, somewhat, for Triple X business with the intro-
duction of the 2001 CSO mortality table (replacing the outdated 1980 CSO
tables), which contain slightly lower mortality assumptions. However, the 2001
CSO tables are also considered conservative.




11
Swiss Re, sigma No 7/2006
Characteristics of life insurance securitizations




A variety of approaches to relieving Triple X reserve strain have been tried. The
To relieve the reserve strain, reinsurance —
backed by a LoC — has been common. most common approach involves reinsuring business (directly or indirectly) to
an offshore reinsurer not subject to US reserve requirements. The offshore rein-
surer uses a letter of credit (LoC) to provide required security for the difference
between XXX reserve and local reserve. However, there are three concerns
about the LoC market. First, most LoCs are short duration, whereas these life
policies are typically for 10 to 20 years.⁵ Second, the prices have risen along
with demand. Finally, the LoC market has limited capacity. The increased use
of securitizations to finance excess reserves has apparently eased some of the
pressure on the LoC market. Shorter, and also longer, duration LoCs are now
easier to secure and more available.

Guideline AXXX is linked to Guideline Triple X. It mandates additional reserves
Guideline AXXX is for universal life policies.
for many types of universal life insurance policies that contain so-called “no
lapse” guarantees.⁶ The additional reserves can be very significant and many
market observers regard these as largely redundant.⁷ Late in 2006, the first
securitization of AXXX business was issued.


If the mortality experience deteriorates and the company needs to build up ad-
Investors are protected by various buffers.
ditional reserves, initially the losses from the adverse mortality experience (ie
higher mortality experience than was assumed when pricing the business) will
be withdrawn from the economic reserves of the company. This is the retained
layer of the bond, similar to the equity layer of a collateralized loan obligation.⁸
If that is exhausted, then payments will come from the reserves placed with the
SPV. These bonds are often credit-enhanced, so investors are also protected by
the monoline insurers in the unlikely event that these reserves are insufficient.

The main benefit of a Triple X securitization is that it provides an alternative to
XXX securitizations also mitigate the tax
disadvantages associated with Triple X. LoCs. While short-term LoCs are less expensive (at least on a pre-tax basis),
they do not match the duration of the life policies. LoCs matching the duration
of the life policies are not less expensive. A securitization eliminates the re-pric-
ing risk of short-term LoCs. Another important result of Triple X securitizations is
to preserve the tax consolidation of the underlying block of business and allow
the continued utilization of the reserve deductions for tax purposes.




⁵ Level-premium term life products typically cover a person for a fixed number of years, usually 10 to 20
years. The insured is required to pay the same premium each year of the contract, reducing lapse risk.
⁶ What the no-lapse guarantee policy adds to the traditional universal life design is a secondary guaran-
tee (ie, in addition to any guarantee with respect to crediting rates, mortality costs or expense loads)
that if a certain premium or premiums are paid in the front-loaded manner specified in the contract, the
policy will not lapse. In a policy which does not contain a “no-lapse” guarantee, the policy can lapse if
the cash value of the policy becomes smaller than the shadow-account value. The shadow account is a
hypothetical account, which is augmented by premium payments and interest income on the cash value
and reduced by mortality phases and administration costs.
⁷ For example Tillinghast “Life insurance securitization expanding”, in: Emphasis 2/2004
⁸ A collateralized loan obligation bundles bank loans into a security for sale to the capital markets. The
bank or a specialized investor retains the first loss, or equity, layer.




12 Swiss Re, sigma No 7/2006
 The typical XXX securitization is constructed by forming a downstream subsidi-
How does a XXX securitization change
the tax treatment for the business? ary and reinsuring the subject business to this newly formed subsidiary. This
structure is often designed to maintain tax consolidation of the subject block of
business while financing the redundant statutory reserves. As the reserves
build, the parent sponsor continues to utilize the substantial reserve deductions
on its consolidated tax return. As the reserves peak and begin to decline, the
parent sponsor is typically responsible for the taxes on the substantial income
reported by the captive. The impact of this reserve ramp-up and subsequent re-
versal of the redundant statutory and tax reserves on a US GAAP basis is the
recognition of a deferred tax liability that builds as the redundant reserves build
and declines as the redundant reserves decline. While the overall tax burden is
not altered by the securitization, the timing of the tax payment is deferred (up to
10 to 15 years relative to the LoC solution) and brought into line with the origi-
nal tax treatment which would apply in the absence of reinsurance. If the tax en-
vironment changes over this time horizon, it could change the economics of the
securitization transaction. Similarly, changes in actuarial methodology for XXX
reserves could also affect the beneficial tax economics. This latter risk is likely to
be greater than the risk of changes in the tax environment. Neither risk is large
for in-force business, but both are possible.


Transaction costs of securitization

The cost of a XXX securitization is difficult to estimate and compare with alter-
natives (such as reinsurance and LoCs). It is particularly difficult to assess the
size of the cost advantage. Transamerica Re⁹ has provided a simple illustration
of principles involved in Triple X securitizations. The coupon to investors de-
pends on the risk profile of the underlying business. The pre-tax cost is about 95
basis points (bp) over and above LIBOR. The tax impact may lower this cost. Of
the 95 bp, 50 bp is the coupon for investors, 35 bp is the financial guarantor
premium and 10 bp is the cost of capital for debt issued (assuming a 1% RBC
charge on the debt issued).


Extreme mortality or longevity securitizations

Life (re)insurers can hedge against a pandemic flu or a sharp increase in longev-
Securities can also transfer risks, such
as mortality and longevity, to the capital ity through securitizations. If the security has a parametric trigger, the transfer of
markets.
catastrophe risks works well for large diversified (re)insurers with a well diversi-
fied mortality or longevity portfolio, because the basis risk is smaller. For indem-
nity-triggered bonds, investors would prefer a geographically well-diversified
portfolio underlying the security. However, no indemnity-triggered bonds have
been issued yet.




⁹ “Triple X securitization: a review of transactions to date”, in: The Messenger, April 2005.




13
Swiss Re, sigma No 7/2006
Characteristics of life insurance securitizations




Recently issued mortality bonds

Swiss Re’s Vita I and Vita II and Scottish Re’s Tartan Capital are the only mortal-
Mortality cat bonds have been
successfully issued. ity catastrophe bonds to be issued so far. Vita Capital Ltd, which was issued in
November 2003, was a USD 400 million facility with a three-year term that will
pay Swiss Re in the event that a predefined population mortality index should
exceed 130% of its 2002 level. Vita Capital II Ltd, issued in April 2005, raised
USD 362 million. The first tranche attaches at 110% of expected mortality for
any consecutive two-year period over a five-year term. Scottish Re’s Tartan
Capital’s risk coverage raised USD 155 million, has a two-year period and has
its principal at risk if its US mortality index exceeds predefined percentages of
the expected mortality level, 115% for Class A Notes, and 110% for Class B
Notes.




Table 1 Strengths Weaknesses
SWOT analysis of L & H securitizations  Market is growing, liquidity improving,  Many are mostly financing, rather than
spreads are narrow; inexpensive risk transfer, though some risk is trans-
 Seasoned book, generally accepted by ferred
capital markets  Expensive for catastrophic risk, mortality
 Can be used to monetize intangible and longevity
assets, fund regulatory capital require-
ments, transfer cat risks, eg mortality,
(perhaps longevity and morbidity?)
 Improved capital efficiency from, eg,
securitizing XXX reserves
 Favorable tax, regulatory treatment
 50% of CFOs surveyed by Tillinghast
indicated they would consider securiti-
zation over the next 2–3 years¹⁰
 Securitization creates associated solu-
tions and unlocks additional traditional
capacity

Opportunities for protection buyers Threats to protection buyers
 Manages concentration of mortality risk,  Regulatory risk
reduces capital needs  Rising cost of issuance, if market be-
 Hedges pandemic exposure comes saturated?
 Unloads risk/capital requirement for  Some companies may use the securitiza-
closed book tions to price more aggressively, increas-
 Increases ROE ing the risk of a soft market
 Invites new competition (eg banks)?
 There may be unexpected correlations
of basis risk with capital markets




¹⁰ Tillinghast, “Life insurance CFO survey No 11: managing current and future demands on capital”,
August 2005.




14 Swiss Re, sigma No 7/2006
Table 2
Risks to sponsors and investors from
the three types of securities


(Re) insurer risk stemming from the security Investor Risk
Type of security
Type of risk Embedded Value XXX and AXXX Extreme Mortality
Interest-rate If the security includes Some interest-rate risk None, unless the security With all three types of transactions,
risk blocks of business that may exist. Reserves are is issued with a floating the investor faces a reinvestment
are interest-rate sensi- mainly held in fixed- rate. risk. Interest payments may only
tive, due to guarantees income instruments. be reinvested at lower rates. Also,
or floating-interest-rate AXXX universal policies many issues are floating-rate.
policies, the interest-rate usually have interest-rate
risk can be substantial. guarantees, so these
liabilities must be
matched with assets that
fund those guarantees.
Equity-market Dependent on the amount None, if insurance reser- None, if insurance With all three types of deal struc-
risk of equities held in the ves are entirely held in reserves are entirely held tures, the equity risk is limited.
asset portfolio. If invest- fixed-income instruments in fixed-income instru- However, embedded-value deals
ment returns – gains on (which is most often the ments (which is most could have equity assets in the
the equity portfolio – are case). often the case). reserves supporting the policies.
higher than expected, the Also, equity risk could come
(re)insurer may earn an through counterparty risk – if the
excess return. Usually, insurance company becomes
the sponsoring company insolvent.
keeps the upside.
Credit risk If the bond is credit-wrapped: A downgrade of the None The investor has the risk that the
sponsoring company may allow the monoline insurer (re)insurer becomes insolvent and
to charge higher guarantee fees. The higher fees may is not able to pay interest and/or
change the economics of the transaction. Generally, face amount. However, with many
the sponsoring company pays the monoline insurer a securitizations, the interest
quarterly fee (based on the amount of the outstanding payments and the principal are
loan). protected by a monoline insurer,
reducing or eliminating the
counterparty risk for the investor.
Mortality risk The payback of the bond If mortality experience is If mortality index does not Embedded value: If mortality
may vary depending on adverse, it will first deplete exceed the agreed thres- increases above what was
the mortality experience. the economic reserves of hold, the (re)insurer has assumed when pricing the bond,
If mortality experience is the life company and then to pay back the full face the payback period will lengthen
lower (higher) than was the redundant reserves in amount of the bond to and the principal may be reduced.
assumed when pricing the SPV. investors. However, if the Triple X, AXXX: If mortality ex-
the bond, the payback mortality index exceeds perience is adverse, it will first
will be accelerated the pre-determined thres- deplete the economic reserves
(slowed). hold, the adverse mortality of the life company and then the
experience will reduce redundant reserves in the SPV
the payment of principal (ie into the collateral which
to investors (in the extreme investors have provided).
case, none of the principal Investors have the risk that part
will be repaid upon the (or the entire) face value will not
bond’s maturity). be paid back at maturity. If credit-
wrapping exists, the depletion
of the face value will be covered
by the bond insurers.
Mortality cat bond: If mortality
index exceeds the threshold, the
principal reverts (partially or
fully) to the (re)insurer.




15
Swiss Re, sigma No 7/2006
Characteristics of life insurance securitizations




(Re) insurer risk stemming from the security (continued) Investor Risk
Type of security
Type of risk Embedded Value XXX and AXXX Extreme Mortality
Lapse risk Higher lapses than Same as for Embedded Limited. For parametric Embedded value/Triple X, AXXX:
expected will reduce the Value triggers, lapse is irre- Early surrender of a savings
cash flow generated by levant. If an indemnity policy will create an immediate
the embedded-value trigger, a higher lapse cash demand for repayment and
business. In more rate would reduce the deprive the insurer of at least
extreme cases, the probability of the bond some expected future income.
principal may be reduced. being triggered. If it is an Funds will become available more
industry loss trigger, quickly, thus the pay-back period
the lapse rate of the may shorten. However, some of the
issuing (re)insurer might future expected profits may not
affect the industry loss, be realized.
again changing the Mortality cat bond: None, if a
probability of the bond parametric trigger. Higher lapses
being triggered. would reduce the probability of
bonds based on insurance losses
being triggered.
Legal risk Minor If taxation framework None None
changes during duration
of bond, this can substanti-
ally impact the economics
of the transaction. Spon-
soring company bears
legal risk.
Reputation If bond performs poorly, Same as for Embedded None. These bonds are None
risk the issuing (re)insurer Value fully backed in an SPV
may have difficulty by fixed-income assets.
issuing future bonds.
Also, poor bond perfor-
mance may reflect and
call attention to other
underlying problems
the company faces.




16 Swiss Re, sigma No 7/2006
Characteristics of P & C securitizations

P & C risk transfer and capital management options

For P & C insurance, there is a variety of mechanisms for transferring risk and
For P & C insurers, there are many ways
to transfer risk and manage capital. managing capital. This wide variety of capital management structures and tools
reflects the volatility of the P &C industry, relative to the life insurance industry.
The tools include traditional reinsurance, collateralized reinsurance, cat bonds
(excess of loss), cat swaps, industry loss warranties, contingent capital and
side-cars (which have evolved from quota share cat bonds and collateralized re-
insurance). In addition, exchange-traded insurance options can provide protec-
tion to (re)insurers (see box: Exchange-traded options). Cat bonds complement
or substitute for these various transactions. The structure chosen depends on
the specific needs of the protection buyer and the availability of fixed-income
investors to support the structure. The characteristics of each of these types of
tools are explained below and shown in Table 4.

Traditional reinsurance
Traditional reinsurance involves a reinsurer agreeing, for a premium, to indemnify
Traditional reinsurance indemnifies
the insurer. the ceding insurer (the cedent) against all or part of the loss the insurer may
sustain under the covered policies it has issued. Typically this has no pre-event
collateral, and recovery depends on the ability of the reinsurer to pay claims
when due.

Catastrophe bonds
Unlike life bonds, which mostly provide financing, P&C insurance-linked securi-
P&C cat bonds pay on the occurrence
of a pre-defined event. ties typically transfer peak risks to the capital markets. The bulk of insurance
securitization transactions to date have involved catastrophe bonds (popularly
known as cat bonds). In a typical transaction, an SPV enters into a reinsurance
contract with a cedent and simultaneously issues cat bonds to investors. If no
loss event occurs, investors receive a stream of coupon payments and a return
of principal that compensate them for the use of their funds and their risk expo-
sure. If, however, a pre-defined catastrophic event does occur, investors suffer a
loss of interest, principal, or both. These funds are transferred to the cedent, in
fulfillment of the reinsurance contract.

Catastrophe swaps
Another common way to transfer catastrophe risk is through a swap transaction,
Cat swaps exchange fixed payments for
floating payments which depend on the in which a series of fixed, predefined payments is exchanged for a series of
occurrence of insured event(s).
floating payments whose values depend on the occurrence of an insured event.
The cedent can enter into the swap directly with counterparties or through a
financial intermediary. Swaps, by design, offer benefits over catastrophe bonds.
They are simpler to implement and entail fewer fixed costs. Unlike cat bonds,
they are usually not collateralized and do, therefore, entail credit risk. The re-
duced liquidity relative to a tradable security increases the cost of protection,
often outweighing the lower out-of-pocket costs.




17
Swiss Re, sigma No 7/2006
Characteristics of P & C securitizations




Exchange-traded options

Although efforts to date to develop exchange-traded catastrophe options have
In the past, insurance options have
been traded on exchanges. not been successful, exchange-traded instruments may eventually become a
widely accepted means of transferring insurance risk to capital markets. PCS
exchange-traded catastrophe call options were standardized contracts that pro-
vided the purchaser with a cash payment if an index measuring catastrophe
losses exceeds a certain level, known as the strike price. A key shortcoming of
these options was that they were based on broad regional indices that intro-
duce too much basis risk to effectively hedge an insurer’s cat exposures.

Futures and options contracts based on the initial version of the Chicago Board
Exchange-traded options have failed
due to lack of activity. of Trade (CBOT) cat index began trading in December 1992, but there was
little activity in the market, and trading was halted. Other attempts have also
been made, only to fail again. Recently, a new exchange has been created,
HedgeStreet.com, but so far it has attracted limited interest.


Industry Loss Warranties
Industry loss warranties (ILWs) are typically structured as indemnity insurance
Industry loss warranties are over-the-
counter contracts which pay the protec- or reinsurance contracts covering specified insurance losses. The defining fea-
tion buyer on the occurrence of a pre-
ture of an ILW contract is the presence of a condition for indemnification which
defined event.
is linked to an industry loss metric. The threshold for the first condition (the actu-
al insured loss of the (re)insurer) is set so low that it is virtually certain to occur
if the industry loss is triggered. As such, the price of the ILW is based on the risk
associated with the industry losses or index. Hence, they are easier to under-
write and can be offered by hedge funds. A second typical feature of an ILW is
the backing-up of the policies by letters of credit from global commercial banks
for the limit of liability written. This enables unrated players like hedge-fund-
owned reinsurers to offer attractive solvency protection to potential buyers. Sim-
ilar to index-based ILS, ILWs have basis risk. They are most attractive to compa-
nies with diversified portfolios whose risk profile is similar to the overall market.
This is true for large (re)insurers, which explains the growing role of ILWs in the
retrocession market in recent years. The recent focus of rating agencies and reg-
ulators on basis risk in ILWs (and index-based cat bonds) may reverse this trend
although it is too early to tell.

Contingent capital
Contingent capital addresses capital needs through risk financing rather than
Contingent capital provides a loan, or
other financing, upon the occurrence risk transfer and is based on the mechanics of “put options”.¹¹ These agree-
of predefined event.
ments are more complex than traditional line-of-credit deals through commer-
cial banks. Contingent capital instruments provide the buyer with the right to
issue and sell securities at a fixed price for a fixed period of time if a predefined
event occurs. These securities may be equity, debt, or some hybrid. Contingent
capital differs from insurance (whether provided from a reinsurer or via cat bond
capacity) in that it does not provide indemnification, but instead provides ac-
cess to capital that either dilutes equity or must be repaid.


¹¹ “An option contract giving the owner the right, but not the obligation, to sell a specified amount of an
underlying security at a specified price within a specified time. This is the opposite of a call option,
which gives the holder the right to buy shares.” (Source: www.investopedia.com)




18 Swiss Re, sigma No 7/2006
 Side-cars
Side-cars are special purpose vehicles that provide additional capacity to a
Side-cars provide capacity of limited
duration. sponsoring (re)insurer via partially collateralized quota share. Side-cars typically
have a lifetime of a few years with a pre-defined divestment procedure. They are
set up at the beginning of a hard market with the intent to be divested prior to
the next soft market. Side-cars are funded largely by third-party capital seeking
to participate in the business of the sponsor. The capital consists of equity and
collateralized loans and covers some modeled aggregate-loss exposure. In
some cases, the sponsor writes business, and then cedes it via quota share – or
some other reinsurance agreement – to the side-car. In other cases, the sponsor
effectively acts as a managing general underwriter for the side-car, and the risk
is written directly by the side-car. Investors with reinsurance subsidiaries may in-
vest in the side-car or assume the same risk using quota share reinsurance.

Side-cars enable the sponsors to leverage access to business and underwriting
Side-cars are used to accelerate the
balance sheet capabilities on a broader capital base without raising expensive equity capital.
Risk-bearing capital can be kept off balance sheet in the side-car vehicle. The
underwriting reinsurer is able to accelerate the balance sheet and earn superior
returns on capital over the cycle due to higher leverage. The sponsors also
charge fees for the underwriting and administrative expenses and receive a
ceding commission. As a capital management tool, side-cars offer the advan-
tages of tailor-made protection without material basis risk and with favorable
regulatory, accounting, and tax treatment. The typical side-car structure does
require the sponsor to retain the tail risk at the top of the partially collateralized
structure, which may be set at, for example, a 250-year event.

A dozen side-cars were formed after hurricane Katrina, supporting Bermuda-
based sponsors. Interestingly, four side-cars in 2006 were for start-up insurers
of the class of 2005 (Harbor Point, Validus, Lancashire, and Flagstone). It is im-
portant to remember that side-cars are actually “old wine in a new bottle” and
are much the same as securitized quota shares such as 1996’s Georgetown Re,
which effectively launched the cat bond market after the 1994 Northridge
earthquake.


Table 3 Side-car Capital, USD million Sponsor/investor
List of side-cars formed post-Katrina Bay Point Re 150 Harbor Point
Blue Ocean Re 300 Montpelier
Concord Re 750 Lexington
Cyrus Re 525 XL Capital
Flatiron Re 800 Arch Capital
Helicon 145 White Mountains
Mont Fort Re 60 Flagstone
Petrel Re 200 Validus
Rockridge 91 Montpelier
Sirocco 95 Lancashire
Starbound Re 310 Renaissance Re
Timicuan Re 50 Renaissance Re

Source: A.M. Best




19
Swiss Re, sigma No 7/2006
Characteristics of P & C securitizations




Table 4
Summary of capital management
innovations

Industry loss Contingent
Cat bonds Cat swaps warranties capital Side-cars
Compensation/financing Compensates Compensates Compensates Provides financing Compensates reinsured
buyer against buyer against buyer against on pre-agreed against losses
losses losses losses terms in case of
loss event. No
earnings relief
Basis risk/tail risk Minimal if indem- Present in deals Significant Minimal if indem- Significant tail risk
nity trigger, MITT, with trigger nity trigger; signi- with some basis
significant if index-, based on index ficant if index-, risk created by
model-, or para- model-, or para- structural mitigants
metric-based metric-based
Moral hazard Low if index-based/ Low if index-based/ Low; index- Low; index- Medium; mitigated
parametric, medium parametric, medium based based via contract design
if indemnity-based, if indemnity-based;
mitigated via mitigated by
contract design contract design
Counterparty risk Minimal. Capital Yes Depends on Depends on Depends on side-car
is invested in whether limit is whether pre- structure and collateral
safe securities collateralized funded or arrangements for any
held by trustee unfunded “pass through” quota
share reinsurance
Liquidity for risk taker Medium for rated Low Low Low Limited to retrocession
transactions; market
same as similarly
rated corporate
and ABS bonds
sold in a Rule
144A or similar
private placement
Regulatory/accounting/ Varies No favorable Varies No favorable Well established +
tax (RAT) rules for RAT treatment, RAT treatment favorable RAT treatment
cedent subject to penalty for
tail risk and basis risk
created by structural
mitigants
Capacity providers Institutional fixed- Large primary or Reinsurers, hedge Reinsurers, hedge Institutional fixed-
income investors, reinsurers funds, institutional funds, commercial income investors,
hedge funds investors banks institutional financial sponsors
investors (private equity), hedge
funds
Buyers of protection Large primary Large primary Reinsurers, hedge Primary insurers, Predominantly
insurers, reinsurers, insurers or rein- funds reinsurers, corpo- reinsurers
corporates, and surers rates, government
government entities entities
Intermediation Investment banks The counterparties, Reinsurance Direct, Direct,
brokers broker reinsurance reinsurance
broker broker
Standardization Customized Customized Customized Customized Customized
Complexity of High, expected to High, expected to Low, based on High Varies by book of
underwriting decrease as firms decrease as firms market risk only insured business
gain experience gain experience




20 Swiss Re, sigma No 7/2006
 Benefits for P & C (re)insurers and investors of cat bonds

Sponsoring and accessing capacity via a cat bond may provide a P & C re(insurer)
with many benefits, but these may come with some drawbacks, such as basis
risk.

In sponsoring a cat bond, a (re)insurer can potentially improve both its risk and
Issuing a cat bond increases capital
efficiency and, like other forms of rein- capital management effectiveness. Cat bonds provide an additional avenue to
surance, improves return on equity.
hedge underwriting risk – especially risk related to low-frequency, high-severity
events – by transferring the risk from the (re)insurer’s balance sheet (supported
in large part by equity capital) to the broad fixed-income market, reducing peak
risk to the (re)insurer. Securitization also adds flexibility to a reinsurer’s access to
capacity. As with traditional reinsurance, it adds to rating agency capital-ade-
quacy requirements and may improve ROE and other performance measures.
Unlike traditional reinsurance, there is usually no credit risk for the issuing (re)in-
surer, since the cover is fully collateralized.

Fixed-income investors also benefit from investing in cat bonds. The bonds allow
Investors benefit from the low correla-
tion with other fixed-income markets … investment in specific insurance risks without exposure to other risks carried by
the (re)insurer that come with equity investment. Moreover, because of the low
correlation of defaults between debt capital markets and cat risks, investors can
improve their portfolio risk/return profile.

Cat bonds may pay a higher interest rate than similarly rated corporate debt and
… and from reasonably high returns.
traditional asset-backed securities (eg MBS, credit card receivables etc). These
higher spreads compensate investors for the perceived illiquidity of cat bonds
and the non-traditional nature of the securities (ie the novelty premium). These
spreads have narrowed substantially since the early 1990s, when the first cat
bond was issued, as more fixed-income investors and broker-dealers have be-
come involved in the market.


Weaknesses

Cat bond protection buyers generally face more basis risk than do buyers of
Cat bonds usually carry basis risk.
traditional reinsurance, since investors prefer index- or model-based triggers
for these bonds.¹² This is because deals linked to synthetic portfolios (such as
industry loss indexes), unlike those with indemnity triggers, are not subject to
moral hazard problems.

Rating agencies have recently focused on the basis risk introduced by cat bonds
Rating agencies are focusing on
basis risk. and ILWs in an effort to accurately reflect the advantages and disadvantages
of these products in their financial strength models. These efforts are expected
to evolve and be refined considerably in coming years as the financial strength
models develop in concert with regulatory initiatives such as Solvency II.




¹² Traditional reinsurance treaties will typically have numerous sublimits, exclusions, terms and conditions
that can introduce basis risk and run counter to the “follow the fortunes” fundamental of treaty reinsur-
ance. Nonetheless, these features are often minor relative to the basis risk in a cat bond or ILW.




21
Swiss Re, sigma No 7/2006
Characteristics of P & C securitizations




Moreover, cat bonds have predominantly been issued for short-tail risks. Since
Most cat bonds are for short-tail risks.
books must be closed when bonds mature, investors and sponsors are unwilling
to carry the potentially severe run-off risk associated with longer-tail risks. Since
they are fully collateralized, standard cat bond deals require the special purpose
vehicle (SPV) to hold the full cover in highly rated investments. Hence, ILS inves-
tors cannot benefit from the built-in leverage available by investing in (re)insur-
ance shares.¹³

Finally, there is some question about whether a mega cat is really a zero-beta
Is a mega cat really a zero-beta event?
event, or whether there can be default correlation with other asset classes. For
example, Katrina affected energy prices. Much of the analysis of the correlation
between cat bonds and other asset classes is based on lower-layer, higher-fre-
quency insurance risks that are currently not securitized.


Impediments to growth

A major impediment to growth in the cat bond sector is that rating agency and
Rating agency and regulatory capital
models do not reward the full benefits regulatory capital models understate some of the major benefits of cat bonds.
of cat bonds.
While collateralization of reinsurance recoverables receives favorable treatment,
buying reinsurance from either a fully collateralized cat bond SPV (or a highly
rated reinsurer) is given little if any incremental benefit over buying reinsurance
from a lower-rated reinsurer. Counterparty risk from reinsurer insolvency be-
comes much greater for the relatively remote events for which cat bonds pro-
vide capacity.

The relatively less favorable regulatory capital treatment of (re)insurers com-
Insurers face more rigorous risk transfer
tests than banks. pared to banks for economically equivalent transactions impedes the develop-
ment of some types of transactions, such as on motor insurance. Most US ABS
deals that achieve meaningful capital relief would fail risk transfer tests imposed
on insurers to qualify as reinsurance, if measured with a similar approach. Given
an equal playing field, (re)insurer usage of these transactions could expand sig-
nificantly.

The ability to obtain multi-year fixed-price capacity also receives no benefit in
The favorable diversification aspect of
multi-year cat bonds is not recognized. the current models, even though it creates substantial value. Sometimes during
the term of a major cat bond, a major event changes the price of the underlying
risk. The sponsor of the cat bond does not face any re-pricing or refinancing risk
with its bond, but market participants with annually renewed protection do. This
diversification of protection with multi-year bonds and annual renewal reinsur-
ance is not recognized by rating agencies and analysts. However, this type of
diversification is well-recognized in the case of debt financing – a mix of short-
and long-term debt is viewed favorably.




¹³ See Swiss Re technical publishing, “The economics of insurance”, 2001, for a discussion of how
insurers create value for shareholders.




22 Swiss Re, sigma No 7/2006
 Demand for cat bonds has often been stronger than supply. One reason for the
Cost may also be a partial impediment
to growth. shortage of supply appears to be the relatively high cost to the sponsor. The
leverage over surplus inherent in the capital structure of a diversified reinsurer
gives it a significant capital-efficiency and pricing advantage over cat securitiza-
tion for non-peak cat risks that is unlikely to be overcome. Even for peak risks,
investors are still demanding novelty and liquidity premiums for non-standard
risks while a liquid and tradable market develops. However, this premium has
declined substantially over time, as fixed-income investors have become more
familiar with cat bonds and their financing techniques and as liquidity has in-
creased. Previously, a drawback for sponsors was the sometimes unclear regu-
latory, accounting, and tax (RAT) treatment for cat bonds. Properly structured
reinsurance from indemnity cat bonds now receives accounting treatment simi-
lar to reinsurance. Reinsurance from a properly structured non-indemnity cat
bond may also achieve substantially equivalent RAT treatment with the assist-
ance of an experienced advisory team in the structuring process.

Another impediment stems from the cap that rating agencies typically place on
The rating agency cap on cat bonds has
slowed the development of the market. cat bond ratings. The actual ratings, therefore, effectively overstate the expected
loss on the transaction. This is particularly true for the most remote layers, where
the benefit of collateralization for the industry would be the greatest.

The lack of standardized cat bond contracts has reduced liquidity and slowed
Standardization of cat bond contracts
would facilitate growth. the development of a secondary market. For most cat insurance risks, defining a
standardized, broadly accepted trigger is still a challenge. The development of
transparent, objective, consistent, credible, frequent and timely industry loss in-
dexes for a wide variety of cat risks could be very beneficial. Currently, no such
index exists outside of the US, which has Property Claim Service (PCS) data for
industry loss indexes. Industry loss indexes for European wind and flood and for
Japanese earthquakes would increase issuance of cat bonds.¹⁴ Any standardi-
zation, however, would need to take into account the evolving views of regula-
tors and rating agencies on retained basis risk.




¹⁴ Market participants have begun to explore new possibilities of an independent company providing
the loss data. It would be based on a survey of the largest insurers for particular risks, such as Europe
windstorm and/or Japanese earthquake.




23
Swiss Re, sigma No 7/2006
Market overview

Key Developments

For the past few years, the issuance of life bonds exceeded that of cat bonds.
Issuance of life bonds has been robust,
driven mostly by monetizing embedded Interest in life securitization has increased due to the need to (1) fund growth of
value and, in the US, by funding regula-
new business and regulatory reserve requirements (XXX); (2) monetize embed-
tory reserves.
ded value of defined blocks of business; and, (3) transfer catastrophic mortality
risk to capital markets. The total outstanding volume of insurance-linked securi-
ties has grown to nearly USD 23 billion, from about USD 6 billion in 2001. It is
important to remember that the USD 23 billion is only for bonds. Hence, it sub-
stantially understates the participation of fixed-income investors in the market
because it excludes swaps, ILWs, and private transactions.


Figure 4 24 in USD bn
Total ILS issued and outstanding,
1998–2006: the market is taking off
20

16

12

8

4

0
1998 1999 2000 2001 2002 2003 2004 2005 2006*

New issues Outstanding from previous years

* data through August 25, 2006

Source: Swiss Re Capital Markets




The life bond market

The total volume of life securitizations placed in the fixed-income markets
Total issuance of life securitizations was
USD 15.9 billion between 1996 and 2006. amounted to USD 15.9 billion between 1996 and June 2006. More than half
of the transactions were embedded-value deals. The bulk of the remaining vol-
ume was Triple X securitization in the US. To date, only Swiss Re and Scottish
Re have placed excess mortality bonds. The face amount of the two Swiss Re
transactions was USD 762 million, and of the Scottish Re transaction USD 155
mn. So far, no pure longevity transactions have been closed, even though there
is an active life settlements market which absorbs some longevity risk.¹⁵ Since
2001, the outstanding volume of life insurance-linked bonds has more than
quadrupled, from USD 3.4 billion to USD 15.6 billion (August 2006).




¹⁵ A life settlement is the sale to an investor of an existing life insurance policy for more than its cash
surrender value but less than its net death benefit. The investors in these policies then become the
beneficiaries of the policies, paying the premiums until the death of the insured.




24 Swiss Re, sigma No 7/2006
 In XXX securitizations, coupons to investors may fall after a company’s first
Spreads appear to decline after
first issuance. transaction. In contrast, spreads for AAA-indexed bond funds were relatively
constant from February 2005 to May 2006.

Costs are shrinking as the market matures. When Barclays Life issued GBP 400
The market continues to develop.
million of floating-rate notes backed by life policies in November 2003, the issue
had an average life of 2.1 years and was priced at 40 basis points over LIBOR.
Friends Provident›s December 2004 two-tranche issue had average lives of 2.9
and 5.8 years, respectively, and the tranches were priced at tighter spreads –
20 and 23 basis points, respectively – over LIBOR.¹⁶

Structuring costs have also been lowered through “shelf” registrations. Shelf-of-
Structuring costs have been lowered
through “shelf” securitizations. fering programs are structured in such a way that all the legal, modeling, rating
and other structuring costs are done for a very large bond issue. However, not
all of the bond capacity is issued initially, some “sits on the shelf” and is issued
at any time that the capacity is needed by the protection buyer and the market
is willing to absorb the extra risk. After the initial bond issue, subsequent issues
are released without additional structuring cost, lowering the cost of issuance
and, more importantly, reducing the time between a decision to access the mar-
ket and closing.

Often, bonds contain a credit wrap. The issuing company has to weigh the costs
Credit-wrap costs are also falling.
(premium paid to the bond insurer for providing the credit protection) and bene-
fits (lower coupon of bond) of a credit wrap. Intense competition has resulted in
costs charged by monoline insurers for XXX securitizations falling from 50–55 bp
(on excess reserves securitized) to 25–30 bp.¹⁷

The market is developing its capacity to absorb risk. Barclays Life and Friends
Unwrapped bonds have also been
issued – a sign of the market’s Provident backed their embedded-value securitizations with credit wraps, but
developing risk appetite.
Swiss Re’s Queensgate transaction of USD 245 mn of life-insurance-backed
bonds had no credit wrap. The Swiss Re security was able to raise relatively more
capital, 87% as a portion of the amount of its value in force, than the Barclays
Life and Friends Provident deals, which raised only 50 to 60% of future profits.¹⁸




¹⁶ “Securitization comes to life”, in: Institutional Investor, 12 May 2005
¹⁷ Smith, B, “Securitization of Excess Reserves”, Society of Actuaries Annual meeting, Nov 2005
¹⁸ “Securitization comes to life”, in: Institutional Investor, 12 May 2005




25
Swiss Re, sigma No 7/2006
Market overview




Scottish Re issued a 30-year bond of USD 850 million in January 2005. This is
In the US, life securitizations have been
mainly on excess XXX reserves. the first securitization of excess reserves arising from XXX to be completed for
a reinsurance company. Some primary insurers had already used this capital
management tool for XXX reserves (Banner Life, Genworth). Nearly half of the
outstanding life securitizations are XXX transactions (see Figure 6).

In October 2006, First Colony Life Insurance, a wholly-owned subsidiary of
The first AXXX securitization was
recently issued. Genworth, issued the first AXXX securitization, a private placement. Few details
are available, but it is known that the issue was for USD 315 million and it’s a
floating-rate bond maturing in 2050.


in USD bn
16
Figure 5
Newly issued and outstanding life
14
securitizations, 1998–2006
12

10

8

6

4

2

0
1998 1999 2000 2001 2002 2003 2004 2005 2006*

New issues Outstanding from previous years

Some issues have already been paid back. 2006*: year to date

Source: Swiss Re Capital Markets




in USD bn
16
Figure 6
Outstanding life bonds by type,
14
1998–2006
12

10

8

6

4

2

0
1998 1999 2000 2001 2002 2003 2004 2005 2006*

Outstanding – EV Outstanding – Reg. XXX
Outstanding – Excess mortality Outstanding – Life settlement

Some issues have already been paid back. 2006*: year to date

Source: Swiss Re Capital Markets




26 Swiss Re, sigma No 7/2006
 The P & C bond market

The volume of outstanding and newly issued P & C cat bonds increased robustly
The volume of outstanding P & C cat
bonds has more than doubled in five in 2005 and 2006, particularly after Katrina. Since 2001, the outstanding vol-
years to USD 8 billion.
ume of cat bonds has more than tripled, from USD 2.4 billion to USD 7.7 billion
(August 25, 2006). Demand for cat bonds from investors remains strong, de-
spite the potential triggering of one rated bond after Katrina.


in USD bn
8
Figure 7
P&C bonds issued and outstanding,
7
1998–2006
6

5

4

3

2

1

0
1998 1999 2000 2001 2002 2003 2004 2005 2006*

New issues Outstanding from previous years

* data through August 25, 2006

Source: Swiss Re Capital Markets




In addition to ILS, which use fully-funded SPVs, there is a large volume of over-
Industry loss warranties are a large and
growing segment of ILS, with hedge the-counter swaps, industry loss warranties (ILW), and collateralized reinsur-
funds selling protection to insurers.
ance arrangements. Hedge funds are very active in all these instruments. The
volume of these undisclosed deals is assumed to be roughly the same as the ca-
pacity of non-life ILS. Estimating the market may become more complicated,
since some money managers are selling, via reinsurers, ILW-type protection
through over-the-counter derivatives. In recent years, ILWs have played an in-
creasing role in providing retrocession cover to reinsurers, although this may
change with the new, less favorable, rating agency treatment for ILWs.

Recent bond issues have extended the boundaries of ILS. In July 2005, Oil Ca-
New types of P & C bonds involve trans-
fer of liability risk, motor-book risk, and sualty Insurance, an energy-industry-owned mutual liability insurer set up a
credit risk.
USD 405 million pure casualty catastrophe bond (claims-made indemnity trig-
ger). In December 2005, AXA securitized 85% (USD 234 million) of their French
personal-motor-book risk (indemnity trigger). In January 2006, Swiss Re securi-
tized USD 252 million of credit risk (indemnity trigger).¹⁹




¹⁹ See box: Three innovative P & C securitizations on the following page for more details on this transaction.




27
Swiss Re, sigma No 7/2006
Market overview




Three innovative P & C securitizations

In an innovative deal, AXA transferred part of the risk of its French motor insur-
Risks from motor insurance policies have
been transferred to capital markets. ance policies, covering about three million vehicles with EUR 1 billion of premi-
um income, to the capital markets. The transaction is triggered when the loss
ratio of this book of business rises over the trigger rate in any of four one-year
periods. If the trigger threshold is confirmed to have been exceeded, the losses
above the trigger ratio are deducted, up to the full EUR 200 million of the con-
tingent deposit, before the securities’ funds are returned to investors. Thus, if
claims jump unexpectedly on the book of business, AXA will be partially pro-
tected from excessive motor insurance losses. This protects AXA from, for exam-
ple, rising damages being awarded for personal injury cases. The AXA deal is
unique in a few ways. First, this is the first time that exposure to high-frequency,
low-severity risk has been transferred to the capital markets, even though it would
likely take a low-frequency, highly severe event to trigger the bond. Second, it is
the first transaction primarily involving motor insurance. Finally, it stabilizes their
loss ratio on a particular book of business.

In January 2006, Swiss Re completed the first indemnity-based credit reinsur-
Credit reinsurance risk has also been
transferred. ance securitization. The EUR 252 million retrocession transaction covered the
aggregate loss in excess of a first loss retention over a three-year period and
was tailored to provide optimal economic and rating capital relief. The underlying
risk is linked to claims and reserves on Swiss Re’s credit reinsurance business
for underwriting years 2006, 2007 and 2008. Thus, the transaction allows in-
vestors to participate in an actively managed, broadly diversified credit book for
three years. The first loss retention and the participation proportional to the en-
tire portfolio create financial incentives for Swiss Re to maintain its performance.

In another innovative deal, Oil Casualty Insurance, Ltd. (OCIL) received reinsur-
Even liability risk has been successfully
securitized. ance coverage from Avalon Re, a special purpose vehicle, for its excess general
liability book of business. Oil Casualty Insurance is a mutual with 80 sharehold-
ers from the energy sector. The securitization has an indemnity trigger and the
cover is per occurrence and in aggregate over the three-year life of the security.
The first attachment point is at USD 300 million and is for 90% of the next USD
150 million of losses (B rating); the next is at USD 450 million for 90% of the
next USD 150 million of losses (BB+); the final tranche attaches at USD 600
and is for 90% of the next USD 150 million (BBB+). The securitization covers
events within a three-year period. This transaction was the first to transfer liability
risk to the capital markets and it provides protection to OCIL similar to excess of
loss reinsurance, but without the counterparty risk.
Sources: Tillinghast, Swiss Re, Lane Financial LLC




28 Swiss Re, sigma No 7/2006
 Prior to 2005, the P & C bond market was exclusively for large natural catastrophe,
Most P & C bonds are still cat bonds for
wind and earthquake. predominantly wind and earthquake. About 85% of the current outstanding
volume is still this type of bond. The remaining 15% is split fairly evenly between
liability, credit, auto and other miscellaneous risks.

Excluding US wind peril, on a risk-adjusted basis, spreads on cat bonds have
Spreads have generally narrowed.
narrowed since 2002. In addition, structuring costs have fallen, making it more
attractive for sponsors.

The cat bond market responded to Katrina in an orderly manner, with the bond
The cat bond market was well-behaved
after the 2005 hurricane season. potentially affected falling in value, while other bonds were only marginally af-
fected. Spreads have increased for non-indemnity deals due to higher frequency
and severity expectations. Spreads for US indemnity deals rose more, due to
additional fears of unmodeled exposures to secondary perils. There was a wave
of new issues post-Katrina with many new sponsors (Munich Re, AXA, Hartford,
Montpelier, and PXRE), which put downward pressure on prices as a normal
market reaction to increased supply, thereby raising spreads.


Who buys insurance-linked securities and why?

Investors prefer non-indemnity triggers to indemnity triggers for property cat
Investors find ILS bonds attractive.
risks to reduce moral hazard risk and avoid the modeling uncertainty associated
with secondary perils – especially with commercial portfolios – that can come
with indemnity triggers. Due to basis risk, non-indemnity triggers are more
acceptable for large diversified (re)insurers than for clients with a narrow risk ex-
posure. Despite a post-Katrina hike in spreads and higher uncertainty regarding
the underlying models, cat bonds remain attractive for investors, given their rel-
ative returns and low correlation to other fixed-income investments.

Cat bonds yield reasonable returns, usually with less volatility than comparable-
Cat bonds are a diversifying asset class.
quality corporate bonds, especially if the seasonal volatility of hurricane and
windstorm bonds is excluded (ie, year-over-year returns). The returns on cat
bonds may still have some “novelty” premium, earning a higher return for the
same rating. Also, since corporate bonds carry credit risk, while cat bonds carry
natural catastrophe risk, the correlation of these two different fixed-income as-
set classes is negligible. Hence, the addition of cat bonds to a portfolio improves
the performance and lowers the risk of the portfolio.²⁰

The investor base, which was insurance and reinsurance companies and money
Dedicated cat funds and hedge funds
have increased their participation in the managers in 1999, has shifted dramatically to dedicated cat funds (28%, up
sector in recent years.
from 5% in 1999) and hedge funds (31%, up from 5% in 1999). The other major
segment is money managers, which has been fairly stable – 29% recently and
30% in 1999. Life insurers still invest but for the most part do so indirectly via
cat funds and hedge funds.




²⁰ The analysis here is specifically for natural catastrophe bonds.




29
Swiss Re, sigma No 7/2006
Market overview




Hedge funds and the insurance industry

Hedge funds purchase ILS, but also provide protection directly to (re)insurers
Hedge funds are providing protection to
(re)insurers. through ILWs and collateralized reinsurance. In addition, hedge funds have
many other connections to the insurance industry. For example, they purchase
the stocks and hybrid capital issued by (re)insurance companies, set up reinsur-
ance vehicles and companies and fund side-car capital arrangements, and take
other non-cat risks (life, life settlement, aviation, terrorism, etc), whether or not
in rated bond form.


Cat bonds tend to be less volatile than corporate bonds with the same rating.
Cat bonds increase return and reduce
volatility when added to a fixed-income Hence, adding cat bonds to a fixed-income portfolio reduces the standard devia-
portfolio.
tion of returns. In addition, if it is assumed that corporate bond returns have a
normal probability distribution, that they are uncorrelated with insurance-linked
securities, and that the individual perils of the cat bonds are uncorrelated, then it
can easily be shown that the Sharpe ratio improves as cat bonds are added to a
speculative-grade corporate bond portfolio. The Sharpe ratio adjusts the returns
of a portfolio for its volatility. A higher Sharpe ratio implies better risk-adjusted
performance – higher returns with the same or less volatility. Adding a slightly
high-return, but much more volatile asset to a portfolio will lower the Sharpe ra-
tio, but adding a higher return, less volatile asset – such as a cat bond – to a
speculative-grade corporate bond portfolio will increase the Sharpe ratio.²¹

Traditional money managers are very interested in ILS securities, but have been
Life bonds have been particularly well
received by investors. particularly receptive to financing life bonds, such as embedded value and
those concerning Guideline Triple X. The life bonds tend to have high ratings
and are usually based on seasoned policies. Because there is a track record for
the performance of the policies, the investors can become familiar with the as-
set by analyzing the history of the policies. This is very similar to other asset-
backed securities, such as mortgage-backed securities, or an ABS of credit card
debt. This explains the narrow spreads for life bonds, most of which have low-
volatility premium flows. Mortality bonds have also been well received, because
data on mortality is readily available, very detailed and transparent.




²¹ The Sharpe ratio is a measure developed by Nobel Laureate William Sharpe to assess risk-adjusted
performance. It is calculated by subtracting the risk-free rate from the rate of return for a portfolio and
dividing the result by the standard deviation of the portfolio returns.




30 Swiss Re, sigma No 7/2006
Prospects


Market potential for life bonds

According to a Tillinghast survey of the CFOs of twenty-eight large US insurers,
More US insurance CFOs are considering
securitization. securitization is showing a dramatic increase in popularity. Tillinghast found that
while only 4% of respondents are currently using securitization to address capi-
tal needs, 50% said they will consider it in the next two to three years, ranking it
second only to reinsurance.²² Of the respondents, 55% are currently using or
exploring securitization of term business associated with Regulation XXX, while
66% are considering securitizations for universal life business associated with
AXXX.

The introduction of Solvency II may increase the use of securitization in Europe.
Solvency II may increase use
of securitization. Solvency II is expected to treat all risk-mitigating instruments such as reinsur-
ance, hedging and securitization, in a consistent manner. Solvency II is likely to
accept a wider spectrum of risk-hedging and risk-transfer instruments than Sol-
vency I, which permits a uniform capital reduction for the use of reinsurance.
Solvency II is based on economic principles and therefore securitizations are
likely to receive appropriate credit, which is not always the case under the Sol-
vency I framework. Solvency II may therefore facilitate a substantial expansion
of P & C securitization beyond cat bonds.

Given investors’ preference for well-diversified mortality portfolios, global life
Global life (re)insurers are well positioned
to securitize mortality risk. (re)insurers are well positioned to transfer some of their mortality risk to the cap-
ital market. These companies will also be able to bundle a portfolio with a criti-
cal size.

Life insurers and pension funds are increasingly faced with longevity risks. In
There have been no excess-longevity-risk
transactions so far. some markets (such as in the US, Canada, the UK, Switzerland and the Nether-
lands), pension schemes are predominantly privately arranged. In other markets
in Europe and Asia, private and occupational pension policies are increasingly
in demand, raising the longevity risk accumulated by life insurers and pension
funds. Securitization would allow some mitigation of longevity risk, however, no
longevity bond has yet been placed because it has been difficult to match sell-
ers and buyers. Global longevity risk is estimated to be very large. Life insurers’
technical reserves for private annuities in payout were probably in excess of
USD 600 billion at end of 2004. It is difficult to estimate an equivalent figure for
pension funds. However, pension funds in OECD and select non-OECD coun-
tries had assets of USD 15.6 trillion in 2004.²³ The rapidly developing life settle-
ment market, via which fixed-income investors absorb substantial longevity risk,
provides hope that a pure longevity solution is possible.




²² “Life insurance CFO survey No 11: managing current and future demands on capital”, Tillinghast,
August 2005
²³ “Pension markets in focus”, OECD, December 2005. Pension schemes in some countries have signifi-
cant deficits. For example, Lane Clark & Peacock (LCP) estimates that the overall deficit under FRS17 for
the UK defined benefit pension schemes of FTSE 100 companies was GBP 37 billion as of July 2005,
which was equivalent to 12% of assets (source: Accounting for pensions 2005, Lane Clark & Peacock).




31
Swiss Re, sigma No 7/2006
Prospects




The potential global market for embedded-value securitizations (defined as de-
The market for embedded-value
transactions is USD 400 to 500 billion. ferred acquisition costs (DAC) and present value of future profits (PVFP)) is esti-
mated to be USD 400–500 billion for primary life (re)insurers. Compared to the
current outstanding volume of EV transactions (USD 6.3 billion), there is a large
upside potential for further transactions. The top 10 European and US life insur-
ers alone, with market share of 23.7%, have total intangible assets of USD 173
billion, of which USD 100 billion are DAC.

In the US, life insurers’ regulation on XXX reserves will be a focus for the securi-
The market for Triple X business
is USD 86 billion by 2016. tization markets in the medium term. Securitization offers large insurers a flexi-
ble capital management tool which can potentially improve profits. To date, USD
6.1 billion of Triple X bonds have been issued. Triple X reserves in the US were
USD 52 billion in 2005 according to Milliman.²⁴ It is assumed that around two-
thirds of these are redundant reserves (USD 34 billion), thus only 18% of the re-
dundant reserves are currently securitized. Based on Millman’s projections of
triple X reserves, redundant reserves are estimated to be USD 72.5 billion in
2010 and USD 86.3 billion in 2016, while S&P believes they may even amount
to USD 100 billion in a few years. This leaves ample potential for Triple X securi-
tizations. Though only one securitization of AXXX business has been issued to
date, there is substantial potential in such transactions, although the structuring
is more complex.

The combined volume of extreme mortality bonds issued so far is USD 0.9 bil-
The market potential for excess-mortality
risk is difficult to estimate. lion, which is tiny compared to the global sums assured. However, it may only
make sense to securitize extreme mortality risk, not the entire sum assured, so it
is difficult to accurately estimate the market potential for this type of securitiza-
tion. Securitization of extreme mortality risks is an option for large and well di-
versified life (re)insurers, since investors seem to prefer a diversified mortality
book. Life reinsurers can play a major role in this business, since they can bun-
dle business from various geographical areas and reach the diversification level
desired by investors.




²⁴ Burden, J and G Kelly and B Smith, “XXX Implications,” Society of Actuaries, Reinsurance News
issue No 54, August 2004




32 Swiss Re, sigma No 7/2006
 Embedded value Triple X Extreme
Table 5
(PVFP and DAC) a) redundant reserves b) mortality c)
Estimation of current and potential In USD billion
market size Outstanding bonds,
as per June 2006, USD billion 6.3 6.1 0.9
Potential size of market,
USD billion 400–500 34.4 5 500
Volume of outstanding bonds 18% (1)
in % of current size of market ~1.5% 7.1% (2) 0.02%
Estimated potential market
size in 2010, USD billion N/A 73 7 000

(1) Market share on the basis of the current size of the market (34.4 USD billion).
(2) Market share on the basis of the estimated size of the reserves in 2016 (86.3 USD billion) for the
current business. Reserves for the business today will gradually increase over the next decade.

a) The global embedded value (here interpreted as DAC and PVFP accounted for in the balance sheet) is
based on 23 of the largest European and US life insurance companies, accounting for 42% of the
global life insurance premium volume. These companies reported approximately USD 175 billion of
embedded value in their balance sheets for 2005. Japanese life insurers do not currently activate
DAC or PVFP on their balance sheet.
b) Volume and projections are from Milliman. There are no projections available for AXXX .
c) The extreme-mortality-risk market is derived from a pandemic scenario which assumes that the
current population mortality doubles in every country in a given year, leading to an additional loss
of 56 million lives and a loss of population income of USD 5 500 billion.




Market potential for non-life bonds

Five years ago, cat bonds provided USD 2 billion capacity, which was equiva-
The market for cat bonds is expected to
grow four- to five-fold. lent to some 3% of the traditional reinsurance market. To date, the volume of
outstanding cat bonds is USD 8 billion²⁵, which is equal to 6% of the aggregate
global cat reinsurance capacity (exposure) of USD 124 billion. Within the next
ten years, the global cat capacity is expected to almost double to about USD
230 billion. If the penetration of cat bonds in relation to the traditional reinsur-
ance capacity doubles, we can expect outstanding cat bonds of USD 30 billion,
an almost four-fold increase. If the penetration tripled, the outstanding capacity
would grow to USD 44 billion, a more than five-fold increase. Given past devel-
opments, a USD 30 to 44 billion market by 2016 appears to be most likely.


Table 6 USD billions 2001 2006 * 2016 2016 2016
Estimating the potential for cat bonds Global cat reinsurance covers 83 124 234 234 234
Outstanding cat bonds 2 8 15 30 44
As a % of traditional capacity 3% 6% 6% 13% 19%
Compared to 2006 same double triple
penetration penetration penetration

(*) year-to-date




²⁵ As of August 25, 2006.




33
Swiss Re, sigma No 7/2006
Prospects




The securitization potential for other non-life risks is more difficult to estimate,
Motor insurance provides a huge pool of
risk to be potentially securitized. since the markets are nascent and there are only few initial transactions. Motor
insurance has been discussed for a while as a risk with good potential for secu-
ritization, and the first deal was finalized in 2005. The potential for motor bonds
is large, due to the huge volume of insured motor risk. In 2006, motor insurance
incurred estimated losses of USD 350 billion. This is expected to grow to USD
660 billion by 2016. Securitization of 3% of the claims – which is equivalent to
the penetration of property cat risks five years ago – would require a volume of
USD 20 billion of outstanding motor bonds, while a penetration comparable to
property cat today would result in motor bonds of USD 42 billion. This would
presuppose overcoming the regulatory impediments to growth raised above.

Complementary solutions could develop along with ILS securitizations, as in the
banking industry. For example, banks use the credit default swap market in tan-
dem with the securitization market to manage their credit exposures, and some-
thing similar could evolve along with the development of the motor insurance
securitization market or other ILS techniques.


Table 7 2006 2016 2016 2016
Estimating the potential for motor bonds Global motor claims 350 660 660 660
Outstanding bonds 10 20 42
As a % of traditional capacity 2% 3% 6%
Compared to property cat half same same
penetration penetration penetration
as 2001 as 2001 as 2006




General implications

Investor appetite for life securitizations could change rapidly if, for example, the
Demand for life securitizations could
change with adverse development for mortality experience of a transaction proves worse than expected. It is a rela-
these bonds or excessive issuance.
tively new market, and poorly structured or inadequately supported bonds could
undermine confidence in the market. Also, if supply increases rapidly, with a
large number of life (re)insurers sponsoring bonds, it could outstrip demand. In-
vestors would require wider spreads in both cases, at least until their interest
catches up with supply.

Capital markets require at least USD 200 million in the issuance of a life insur-
Only large insurers can currently use
securitization for capital efficiencies. ance-linked security and USD 100 million for a cat bond. Hence, in the near fu-
ture only large insurers will securitize their blocks of business. However, as the
process becomes more acceptable and standardized, smaller-scale issuance
will become possible. In any event, reinsurers can and have pooled and securi-
tized the risks of a number of their clients, so this scale issue is not really a sig-
nificant impediment to growth.




34 Swiss Re, sigma No 7/2006
 Securitizations could impose market discipline on the industry by providing an
Securitizations tend to keep the
market competitive and complement ongoing window into pricing of risk, potentially reducing the volatility of insur-
(re)insurance.
ance pricing cycles. It is unlikely, however, to replace or substitute for reinsur-
ance or insurance products. Instead, it complements the industry and expands
it by making capital available for high-frequency risks and protecting against
extreme losses from low-frequency, high-severity risks.

Life securitizations completed to date have tended to focus on very low-risk
Securitization of riskier tranches
of business may be limited to large cash flows. The ability to securitize riskier tranches of business could be limited
(re)insurers.
to the very largest life insurers, since only these companies would be able to
provide a sufficiently diversified pool of risks to satisfy capital markets› require-
ments.

The cost of capital of a funded securitization is lower than that of financial rein-
Funded securitizations have a lower cost
of capital than financial reinsurance. surance. However, the rigidities are greater, since securitizations are structured
transactions and long-term in nature. Financial reinsurance may therefore be
more suitable for flexible and short- and medium-term transactions.

Nevertheless, more work needs to be done to support this nascent market:²⁶
The insurance industry, governments
and regulators, and rating agencies  The insurance industry needs to increase the transparency of these types of
can facilitate the development of the
transactions. This includes clarification of the risks transferred through im-
ILS market.
proved data, modeling, and documentation. In addition, more standardization
of contracts, special purpose vehicles and triggers would help. One reason
the L & H bonds have proved so successful with investors is that their cash
flows are better documented and available than for P & C cat bonds, making
the modeling easy and transparent on seasoned books of business. Finally,
transparency would be facilitated by making the price comparison with
(re)insurance contracts explicit. This would make the decision to securitize
versus insure more straightforward, though an estimate of the basis risk im-
bedded in most securitizations would be necessary.
 Governments and regulators could facilitate the market in many ways, also.
First, it will be necessary to recognize the risk transfer involved with securiti-
zations and allow capital relief for this commensurate with the value of the
transfer. Second, securitizations are a risk management tool and should be
recognized as such in the qualitative assessment of (re)insurers. Properly con-
stituted SPVs must be accepted as counterparties providing capital relief and
subject to no more stringent solvency requirements than other financial enti-
ties. Finally, no unwarranted restraints should be imposed on institutional in-
vestments in ILS securities.
 The rating agencies also play a key role in the development of this market. In
addition to recognizing the risk transfer, risk management, and counterparty
issues of securitizations, the rating agencies must clarify precisely how they
will incorporate securitizations into their ratings.




²⁶ See the Group of Thirty, “Reinsurance and International Financial Markets,” Washington, DC, 2006
for a similar list of recommendations.




35
Swiss Re, sigma No 7/2006
Prospects




All parties – the insurance and reinsurance industry, government and regula-
Insurance-linked securities will help to
develop and expand the insurance market. tors, rating agencies, other legal and financial sectors, and the public – need to
cooperate in developing a framework for facilitating securitizations. Currently,
some peak risks – such as global pandemic risk, Japanese earthquake risk,
European windstorm and flooding risk, Florida windstorm risk, and California
earthquake risk – are underinsured. One part of the solution to this social prob-
lem is to offload more of these risks to the fixed-income market, which has enor-
mous capacity compared to the (re)insurance industry. Also, just as the mort-
gage-backed security market facilitated the development of the US housing
market, insurance-linked securities will be able to improve capital and risk man-
agement for insurers, lowering the cost of insurance to consumers. Regulators
can play an important role by creating a level playing field between (re)insurers
and banks in using securitization techniques: equivalent transactions should re-
ceive equivalent treatment. Finally, securitizations are capital and risk manage-
ment tools for (re)insurers, which will help develop this industry by making the
market more complete and efficient. History has shown that the development of
the financial sector – banking and insurance – has promoted economic growth
and development.




36 Swiss Re, sigma No 7/2006
Appendix



Table 8
Life insurance securitizations since 2005, excluding private placements

Maturity Size
Sponsor SPV Issue date (years) USD m Rating Coverage
Banner Life Potomac Trust Capital 05/01/2005 20 49 AAA/Aaa Reg XXX
Banner Life Potomac Trust Capital 10/01/2005 20 49 AAA/Aaa Reg XXX
Swiss Re Queensgate 12/01/2005 20 175 A+/A1 Embedded Value
Swiss Re Queensgate 12/01/2005 20 45 BBB/Baa1 Embedded Value
Swiss Re Queensgate 12/01/2005 20 25 BB/Ba1 Embedded Value
Genworth INC Money Markets 19/01/2005 30 100 AAA/Aaa Reg XXX
Banner Life Potomac Trust Capital 24/01/2005 20 49 AAA/Aaa Reg XXX
Banner Life Potomac Trust Capital 24/01/2005 20 49 AAA/Aaa Reg XXX
Genworth INC Money Markets 28/01/2005 30 100 AAA/Aaa Reg XXX
LILAC Patrons’ Legacy 01/02/2005 19 100
Scottish Re Orkney Holdings 04/02/2005 30 850 AAA/Aaa Reg XXX
Banner Life Potomac Trust Capital 06/02/2005 20 49 AAA/Aaa Reg XXX
Swiss Re Vita Capital II Ltd. 13/04/2005 5 62 A-/Aa3 Excess Mortality
Swiss Re Vita Capital II Ltd. 13/04/2005 5 200 BBB+/A2 Excess Mortality
Swiss Re Vita Capital II Ltd. 13/04/2005 5 100 BBB-/Baa2 Excess Mortality
Genworth INC Term Securities 09/06/2005 28 200 AAA/Aaa Reg XXX
Genworth INC Term Securities 04/10/2005 30 300 AAA/Aaa Reg XXX
Scottish Re Orkney Re II 21/12/2005 30 383 AAA/Aaa Reg XXX
Scottish Re Orkney Re II 21/12/2005 30 43 A-/Aa2 Reg XXX
Scottish Re Orkney Re II 21/12/2005 30 30 BBB+/Baa2 Reg XXX
Swiss Re ALPS Capital II 23/12/2005 20 220 AAA/Aaa Embedded Value
Swiss Re ALPS Capital II 23/12/2005 20 90 AAA/Aaa Embedded Value
Swiss Re ALPS Capital II 23/12/2005 20 30 BBB/Baa1 Embedded Value
Swiss Re ALPS Capital II 23/12/2005 20 30 BB/Ba1 Embedded Value
Scottish Re Tartan Capital 04/05/2006 3 75 Aaa/AAA Excess Mortality
Scottish Re Tartan Capital 04/05/2006 3 80 Baa3/BBB Excess Mortality
Scottish Re Ballantyne Re 02/05/2006 30 250 Aa2/A-/AA Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 500 Aaa/AAA/AAA Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 500 Aa1/AAA/AAA Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 400 Aaa/AAA/AAA Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 10 Baa1/BBB+/BBB+ Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 40 Baa1/BBB+/BBB+ Reg XXX
Scottish Re Ballantyne Re 02/05/2006 30 50 NR/NR/NR Reg XXX
RGA Timberlake Financial 28/06/2006 30 850 AAA/Aaa Reg XXX

Source: Swiss Re Capital Markets




37
Swiss Re, sigma No 7/2006
Appendix




Table 9
P & C insurance securitizations since 2005, excluding private placements

Maturity Size
Sponsor SPV Issue date date USD m Rating Trigger Peril
Swiss Re Arbor I Series VIII 3/15/2005 3/15/2007 20 B Parametric Multiperil
Index
USAA Residential Re 5/31/2005 6/6/2008 91 BB Indemnity Multiperil
2005 Class A
USAA Residential Re 5/31/2005 6/6/2008 85 B Indemnity Multiperil
2005 Class B
FM Global Cascadia 6/7/2005 6/13/2008 300 BB+/BB+ Pure PNW EQ
Parametric
Swiss Re Arbor I Series IX 6/15/2005 6/15/2007 25 B Parametric Multiperil
Index
OCIL Avalon Re Class A2 6/30/2005 6/6/2008 135 B+/BB- Indemnity Industrial
Accident
OCIL Avalon Re Class B 6/30/2005 6/6/2008 135 CCC/CCC Indemnity Industrial
Accident
OCIL Avalon Re Class C 6/30/2005 6/6/2008 135 CCC-/CCC- Indemnity Industrial
Accident
Zurich American* KAMP Re 7/28/2005 12/14/2007 190 CC Indemnity Multiperil
PXRE Atlantic & Western Re
Class A 11/8/2005 11/15/2010 100 BB+/BB Modeled Loss Multiperil
PXRE Atlantic & Western Re
Class B 11/8/2005 11/15/2010 200 B+/B Modeled Loss Multiperil
Munich Re Aiolos 11/18/2005 4/8/2009 130 BB+ Pure Euro Wind
Parametric
AXA FCC SPARC Class A 12/9/2005 7/15/2011 126 AAA/AAA N/A Auto
AXA FCC SPARC Class B 12/9/2005 7/15/2011 76 AA N/A Auto
AXA FCC SPARC Class C 12/9/2005 7/15/2011 32 BBB/BBB- N/A Auto
Swiss Re Arbor I Series X 12/15/2005 12/15/2006 18 B Parametric Multiperil
Index
PXRE Atlantic & Western Re II 12/21/2005 1/9/2009 125 BB+ Modeled Loss Multiperil
Class B
PXRE Atlantic & Western Re II 12/21/2005 1/9/2007 125 BB+ Modeled Loss Multiperil
Class A
Montpelier Re Champlain Re Class A 12/22/2005 1/7/2009 75 BB- Modeled Loss Multiperil
Montpelier Re Champlain Re Class B 12/22/2005 1/7/2009 15 B+/B- Modeled Loss Multiperil
Swiss Re Crystal Credit Class A 1/13/2006 12/31/2008 131 BBB-/Baa2 N/A Credit
Reinsurance
Swiss Re Crystal Credit Class B 1/13/2006 12/31/2008 98 BB/Ba2 N/A Credit
Reinsurance
Swiss Re Crystal Credit Class C 1/13/2006 12/31/2008 76 B/B2 N/A Credit
Reinsurance
Swiss Re Australis 1/26/2006 2/3/2009 100 BB Parametric Multiperil
Index
Undisclosed Redwood VII 2/9/2006 1/9/2008 160 BB+ Industry Index CA EQ
third party*
Undisclosed Redwood VIII 2/9/2006 1/9/2008 65 BB+ Industry Index CA EQ
third party*
Hartford Fire Foundation Re Class D 2/17/2006 2/24/2010 105 BB Industry Index Multiperil
FONDEN* CAT-Mex Class A 5/11/2006 5/19/2009 150 BB+ Pure Mexico EQ
Parametric
FONDEN* CAT-Mex Class B 5/11/2006 5/19/2009 10 BB+ Pure Mexico EQ
Parametric
ACE* Calabash Re Class A1 5/24/2006 6/1/2009 100 BB MITT US Wind
USAA Residential Re 5/31/2006 6/5/2009 48 B Indemnity Multiperil
2006 Class A




38 Swiss Re, sigma No 7/2006
 Maturity Size
Sponsor SPV Issue date date USD m Rating Trigger Peril
USAA Residential Re 2006 5/31/2006 6/5/2009 75 BB+ Indemnity Multiperil
Class C
Swiss Re Successor Cal 6/6/2006 6/6/2008 48 BB/Ba3 Parametric CA EQ
Quake Parametric Index
Class A I
Swiss Re Successor 6/6/2006 6/6/2008 97 BB/Ba3 Parametric Euro Wind
Euro Wind Class A I Index
Swiss Re Successor 6/6/2006 6/6/2008 19 BB-/B1 Parametric Euro Wind
Euro Wind Class B I Index
Swiss Re Successor 6/6/2006 6/6/2008 111 B/B3 Parametric Euro Wind
Euro Wind Class C I Index
Swiss Re Successor II Class A I 6/6 /2006 6/6/2008 73 B/B3 Various Multiperil
Swiss Re Successor II Class E I 6/6/2006 6/6/2008 154 Various Multiperil
Swiss Re Successor III Class A I 6/6/2006 6/6/2008 7 Various Multiperil
Swiss Re Successor IV Class A I 6/6/2006 6/6/2008 30 B/B3 Various Multiperil
Swiss Re Successor 6/6/2006 6/6/2008 103 BB/Ba3 Modeled Loss JP EQ
Japan Quake Class A I
Swiss Re Successor 6/6/2006 6/6/2008 26 BB-/B1 Modeled Loss JP EQ
Japan Quake Class B I
Swiss Re Successor 6/6/2006 6/6/2008 71 B/B3 Modeled Loss JP EQ
Japan Quake Class C I
Swiss Re Successor 6/6/2006 12/6/2007 14 BB-/B1 Industry Index US Wind
Hurricane Industry
Class B I
Swiss Re Successor 6/6/2006 12/6/2007 7 B/B2 Industry Index US Wind
Hurricane Industry
Class C I
Swiss Re Successor 6/6/2006 12/6/2007 34 B Industry Index US Wind
Hurricane Industry
Class D I
Swiss Re Successor 6/6/2006 12/6/2007 5 Industry Index US Wind
Hurricane Industry
Class E I
Swiss Re Successor 6/6/2006 12/6/2007 54 B/B2 Industry Index US Wind
Hurricane Industry
Class F I
Swiss Re Successor 6/6/2006 12/6/2007 42 BB-/B1 Modeled Loss US Wind
Hurricane Modeled
Class B I
Swiss Re Successor 6/6/2006 6/6/2007 3 BB/Ba3 Parametric Euro Wind
Euro Wind Class A II Index
Swiss Re Successor 6/6/2006 6/6/2007 3 B/B3 Parametric Euro Wind
Euro Wind Class C II Index
Swiss Re Successor 6/6/2006 6/6/2007 10 B Industry Index US Wind
Hurricane Industry
Class D II
Swiss Re Successor 6/6/2006 6/6/2007 35 Industry Index US Wind
Hurricane Industry
Class E II
Swiss Re Successor 6/6/2006 6/6/2007 3 B/B3 Modeled Loss JP EQ
Japan Quake Class C II
Munich Re Carillon Class A1 6/19/2006 3/30/2007 51 B+ Industry Index US Wind
Munich Re Carillon Class A2 6/19/2006 3/30/2007 24 B+ Industry Index US Wind
Munich Re Carillon Class B 6/19/2006 3/30/2007 10 B Industry Index US Wind
Balboa Vasco Re 2006 6/21/2006 6/5/2009 50 BB+ Indemnity US Wind
Liberty Mutual Mystic Re Class A 6/21/2006 5/31/2009 200 BB+ Industry Index US Wind




39
Swiss Re, sigma No 7/2006
Appendix




Maturity Size
Sponsor SPV Issue date date USD m Rating Trigger Peril
Dominion DREWCAT Capital 6/30/2006 12/28/2006 50 BB- Pure US Wind
Resources Parametric
Hannover Re Eurus 7/28/2006 4/8/2009 150 BB Parametric Euro Wind
Index
Endurance Shackleton Re Class B** 8/1/2006 8/1/2008 60 Industry Index US Wind
Endurance Shackleton Re Class C** 8/1/2006 8/1/2008 50 Industry Index Multiperil
Endurance Shackleton Re Class A 8/1/2006 2/7/2008 125 BB/Ba3 Industry Index CA EQ
Tokio Marine* Fhu-Jin Class B 8/3/2006 8/3/2011 200 BB+ Parametric Japan
Index Typhoon
Swiss Re Successor 8/4/2006 1/5/2007 50 Industry Index US Wind
Hurricane Industry
Class E III
FM Global Cascadia II 8/25/2006 8/31/2009 300 BB+/BB+ Pure PNW EQ
Parametric

* Transformed by Swiss Re
** In loan format




40 Swiss Re, sigma No 7/2006
Recent sigma publications

No 7/2006 Securitization – new opportunities for insurers and investors
No 6/2006 Credit and surety: solidifying commitments
No 5/2006 World insurance in 2005: moderate premium growth, attractive profitability
No 4/2006 Solvency II: an integrated risk approach for European insurers
No 3/2006 Measuring underwriting profitability of the non-life insurance industry
No 2/2006 Natural catastrophes and man-made disasters 2005: high earthquake casualties,
new dimension in windstorm losses
No 1/2006 Getting together: globals take the lead in life insurance M & A


No 5/2005 Insurance in emerging markets: focus on liability developments
No 4/2005 Innovating to insure the uninsurable
No 3/2005 Insurers’ cost of capital and economic value creation: principles and practical implications
No 2/2005 World insurance 2004: growing premiums and stronger balance sheets
No 1/2005 Natural catastrophes and man-made disasters in 2004: more than 300 000 fatalities,
record insured losses


No 7/2004 The impact of IFRS on the insurance industry
No 6/2004 The economics of liability losses – insuring a moving target
No 5/2004 Exploiting the growth potential of emerging insurance markets – China and India in the spotlight
No 4/2004 Mortality protection: the core of life
No 3/2004 World insurance 2003: insurance industry on the road to recovery
No 2/2004 Commercial insurance and reinsurance brokerage – love thy middleman
No 1/2004 Natural catastrophes and man-made disasters in 2003: many fatalities, comparatively moderate
insured losses


No 8/2003 World insurance in 2002: high premium growth in non-life insurance
No 7/2003 Emerging insurance markets: lessons learned from financial crises
No 6/2003 Asia’s non-life insurance markets: recent developments and the evolving corporate landscape
No 5/2003 Reinsurance – a systemic risk?
No 4/2003 Insurance company ratings
No 3/2003 Unit-linked life insurance in western Europe: regaining momentum?
No 2/2003 Natural catastrophes and man-made disasters in 2002: high flood loss burden
No 1/2003 The picture of ART


No 7/2002 Bancassurance developments in Asia – shifting into a higher gear
No 6/2002 World insurance in 2001: turbulent financial markets and high claims burden impact premium growth
No 5/2002 Third party asset management for insurers
No 4/2002 Global non-life insurance in a time of capacity shortage
No 3/2002 The London market in the throes of change
No 2/2002 Insurance in Latin America: growth opportunities and the challenge to increase profitability
No 1/2002 Natural catastrophes and man-made disasters in 2001: man-made losses take on a new dimension




3
Swiss Re, sigma No 7/2006
Swiss Reinsurance Company
Economic Research & Consulting
Mythenquai 50/60
P.O. Box
8022 Zurich
Switzerland

Telephone +41 43 285 2551
Fax +41 43 285 4749
sigma@swissre.com