European Re Insurance

Europe Equity Research
02 September 2010
Life Reinsurance
Unfolding the black box
• Life Reinsurance accounts for an average 38% of the premiums of European Insurance
reinsurers, 34% of our sum of parts. JPMC sees historic association with AC
Michael Huttner, CFA
asset risk during the sub-prime crisis, and current apprehension of (44-20) 7325-9175
underpricing => reinsurers trade at a 24% discount to target prices vs a 19% michael.huttner@jpmorgan.com
discount for the total insurance sector. J.P. Morgan Securities Ltd.
• Our report shows it actually has a fundamentally stable profile vs non -life Vinit Malhotra, CFA
(44 20) 7325-5321
(not a broker market, mainly mortality which is conservatively priced), and
vinit.malhotra@jpmorgan.com
has relatively high barriers to entry - so this is a business which should earn
J.P. Morgan Securities Ltd.
10-13% ROE we estimate. We see a strong chance of a return to
stabilisation as in the non-life reinsurance post the equity crash 01/03, albeit Dibin Korath
non-life re also benefited from a lift in pricing following 9/11 and also (91-22) 6157-3275
dibin.m.korath@jpmorgan.com
Hurricane Katrina.
J.P. Morgan India Private Limited
• The accounting is opaque and complex. We believe it is for that reason
the market is concerned that life re cannot pay its full share of dividends. Key points and tables:
Life re does only provide a payback after we estimate 5 years on average,
but we believe mature backbooks means this issue is overdone and our 1. Attraction of life re to the reinsurers –
diversification benefit example of Scor Table 2
dividend forecasts are safe (5.1% average yield vs 4.4% for the insurance
sector average) . 2. Implicit valuation, after deducting life re at 76%
of EV Table 4
• Valuation. Valuation of life reinsurance businesses calculated by backing
out life from life primary insurers using the market price if 76% of 10e 3. JPM sum of parts valuation showing life re Error!
Reference source not found.
embedded value. Life insurance assets are currently priced on 3.7x 11e P/E
at Hannover, 5.5x at Swiss, 6x at SCOR and Munich Re 9.3x 4. Business split of European listed reinsurers in
premiums Table 7
• Catalysts. Hannover (OW) targets 15% ROE, its life re profits are boosted
by its focus on impaired annuities, and offers 29% upside to our Dec11e 5. Accounting and reporting is complex and
target price. SCOR (OW) we forecast may announce at its 8 Sept investor opaque pp10-11
day some diversification into longevity life re, raising target ROE from 9%
6. Asset risk is the key source of life re earnings
above risk free to 10% above, and offers 25% upside. Swiss Re (N) offers no volatility – Scottish Re example Table 8
catalyst and its life re profitability is diluted by underpriced pre 2004
mortality, but offers deep value as it is trading at a 22% discount P/NAV 10e 7. Life re (ex asset risk) is less than half as volatile
vs 7% re average and wider insurance sector at 47% premium. as non-life re Table 20
Table 1: Reinsurance - Implied valuation for Reinsurers ex Life Re, LCm 8. Life re EV sensitivity is mainly mortality and
lapse risk – by contrast primary insurers is mainly
Implied valuation Munich Swiss $m Hannover SCOR
interest rate Table 30
Market cap 18,922 15,387 4,172 3,146
Market cap (ex Life Re) 15,100 7,453 1,629 2,026
Net profit '11e 2,473 1,820 648 478 9. life re is a relatively concentrated market –the
Net profit '11e (ex Life Re) 1,623 1,281 445 337 top 5 have 85% share globally figure16, and in US
life re Figure 22
MCAP/Net profit - group 7.7x 8.5x 6.4x 6.6x
Life Re - 76% of LifeRe EV/profit 4.5x 14.7x 12.5x 7.9x
MCAP/Net profit - (ex Life Re) 9.3x 5.8x 3.7x 6.0x 10. cash flow profile: in the case of Scor 39% of
total cash flow is in first 5 years Table 42
Source: J.P. Morgan estimates, MCAP from Bloomberg as on CoB 30th August
See page 91 for analyst certification and important disclosures, including non-US analyst disclosures.
J.P. Morgan does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may
have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their
investment decision.
Michael Huttner, CFA Europe Equity Research
(44-20) 7325-9175 02 September 2010
michael.huttner@jpmorgan.com
Table of Contents
Investment Thesis ....................................................................3
Overview......................................................................................................................3
Valuation gap...............................................................................................................6
High level summary with references ...........................................................................8
Nature of the life reinsurance market...........................................................................9
Five types of risk........................................................................................................10
Summary of Scottish Re profit and loss record 2005-1H10 – asset risk accounted for
85% of the earnings volatility in the period vs just 15% for mortality and other life re
operating risks............................................................................................................14
Life reinsurance is relatively stable (apart from asset risk), both under IFRS
accounting and US GAAP. ........................................................................................20
Comparison of the accounting standards ...................................................................22
Life re is more stable than non-life except for asset risk............................................27
Comparison of the European life reinsurers...............................................................31
Life reinsurers vs. life primary insurers – we believe the life reinsurance have more
attractive risk reward, particularly at current low levels of interest rates...................34
Life reinsurance products...........................................................................................37
RGA read across summary ...................................................40
November 2008 capital issue .....................................................................................41
Capitalisation of reinsurers ........................................................................................47
Key takeaways from the earnings of RGA pre transcript of RGA conference calls
3Q09-2Q10 ................................................................................................................47
Business model: RGA example .................................................................................51
Munich Re .................................................................................................................53
SCOR.........................................................................................................................65
Profitability of a typical life (mortality) reinsurance contract....................................73
Swiss Re.....................................................................................................................75
Life reinsurance – risk products and financing reinsurance
.................................................................................................83
Risk Reinsurance .......................................................................................................83
Financing reinsurance ................................................................................................85
Life reinsurance market shares ..................................................................................88
Valuation Methodology and Risks ........................................89
The contents of this report was partially based on conference calls with the main life
reinsurers at various levels to understand their business and the accounting,
comparison of the reported life re stats from presentations and annual reports. We
thank the reinsurers for their help and stress that this report, the views expressed and
the conclusions are all our own.
2
(44-20) 7325-9175 02 September 2010
Investment Thesis
Overview
The life reinsurance market is interesting for a number of reasons: (a) it is not
well understood, which presents an opportunity, in our view; (b) it has caused
significant earnings volatility, mainly we believe due to asset risk; and (c) it is a
focus area for most reinsurers in terms of growth and potentially M&A with the
planned disposal by Aegon of TransAmerica Re.
In this report our primary objective is to explain how life reinsurance works,
economically and from an accounting perspective. We argue that the market has a
fundamentally stable profile vs. non-life, in that it is not a broker market, is mainly
mortality where pricing has on the whole been relatively conservative, and has
relatively high barriers to entry due to the need to develop specialised underwriting
and relationships with cedants.
However, in the past asset risk (associated with the investing of the above
highlighted long term cash flows and also assumed asset risk from variable annuities)
has caused a problem, and resulted in huge earnings volatility. We conclude that
taking on asset risk is not a necessary by-product of the life re business, and thus we
see potential for earnings to stabilize just as non-life reinsurance did post the equity
crash 2001/03. For example RGA focuses mainly on YRT mortality (ie. pay as you
go) and longevity reinsurance is now increasingly on a swap basis with no asset risk
assumed.
The accounting dynamics are even more confusing - this is mainly because for US
GAAP and IFRS life reserves are set at historic cost on the basis of original pricing
and are only reset if the unit turns into loss and the adverse deviation reserves are
used up. Dividendable cash flow is still determined by statutory accounting, and
while embedded value provides a good view of expected profitability (value of new
business) and actual (experience variances) it does not guide to current cash flows.
Valuation. Assuming life is valued at 76% of 2010e embedded value, as is the case
for the European listed life primary insurers, then the valuation of the remaining units
is lowest at Hannover (3.7x 2011e earnings), next highest at Swiss and SCOR
(5.5x/6.0x), and highest at Munich (9.3x). Our unchanged Overweight
recommendations are Hannover and SCOR, Hannover for its 15% ROE target,
SCOR as we forecast it will lift its target ROE at 8 Sept investor day to 10% from
now 9% over risk free.
In our view, life reinsurance is, compared with non-life re, as reported in the
accounts of the main listed European life reinsurers, not very transparent, has
accounting that is still mainly historic based for liabilities, and reflects neither
underlying ‘dividendable’ cash flows (this is statutory earnings), nor true profitability
(embedded value does this better).
So we believe when earnings surprises do happen, as they did in the case of RGA at
1Q10 in the case of high adverse mortality, or for Swiss Re at 1H10 when underlying
profitability for quarterly earnings dropped from a year earlier SF300m per quarter to
less than SF210m (the equivalent of the stated less than $200m) per quarter, the
3
(44-20) 7325-9175 02 September 2010
market adjusts downwards quickly and is slower to look forward again to future
earnings development.
This note tries to analyse life reinsurance in greater detail and makes three main
points:
1. Asset risk
We believe that life reinsurance is by nature more stable than non-life, because it
consists of more layers of annual business (life reinsurance contracts last up to
around 30 years, so in theory excluding lapses, every year’s new business is just 3%
of the whole), is mainly directly negotiated so there is less price transparency
induced by brokers as in non-life re, and the largest risk is still mortality, which
benefits from a consistent improving trend which smoothes out most pricing
mistakes.
However, there is asset risk. In two cases, particularly Scottish Re and Swiss Re, we
believe that asset risk significantly increases volatility. For Scottish Re, we estimate
that 85% of the earnings changes 2004-9 were due to asset risk, just 15% due to
mortality and other operating risks. Effectively, Scottish Re’s venture as briefly the
fifth largest life reinsurer in the world, following its acquisition of the ING Life Re
block of business, ended when the subprime and Alt A investments of its funding
structures fell and it had to restructure. The operating business of Scottish Re is now
mainly with Hannover Re, where life profits account for an increasing part of group
profits.
In the case of Swiss Re we have estimated the impact on life re profitability had the
old style accounting continued. So, post 2007, we reallocated the surplus investment
returns to the operating units (the new style accounting just allocates risk free, and so
by definition removes most asset risk from the operating units). We estimate that life
re under the old style reporting would have reported a loss in 2008 due to investment
volatility, and that its earnings volatility would have been higher than that of non-life
(relative to premiums).
Life re asset risk exists wherever there are reserves and assets and it is only in one
business line that it is entirely absent: this is YRT (Yearly Renewable Term), which
is like a pay as you go life reinsurance contract, and where there is no asset risk.
2. Accounting
Life reinsurance for US GAAP and IFRS accounts for reserves using the original
pricing assumptions. There is no change made even if they are seen to be inadequate
(only the annual variance) unlike non-life, where reserves are best estimate and
constantly adjusted. This means that only if the life re division as a whole is about to
go into loss then the reserves are adjusted. It helps explain why Swiss Re explained
that part of the reason for the lower run rate of profits going forward is the current
dilutive influence of the business written in the US prior to 2004, when price
competition was particularly fierce.
Also, US GAAP and IFRS are a compromise form of accounting and all the
European reinsurers we spoke with actually rely more on embedded value to steer
their business and to set incentive pay (even RGA in the US which does not report
embedded value believes that it is a better guide to value). The two indicators most
life re units track is new business margin and experience variances.
4
(44-20) 7325-9175 02 September 2010
And finally the only measure of ‘dividendable’ cash flow comes from statutory
earnings, which are different again. Broadly, statutory earnings lead to delayed profit
recognition (less capitalisation of deferred acquisition costs), embedded value fronts
end more, with much of the value creation linked to the point of sale, and IFRS/US
GAAP are compromises.
3. Diversification
We believe that part of the attraction of life reinsurance is that, thanks to
diversification, a relatively modest outlay of capital in this area can help boost group
ROE. Also we believe that reinsurers also see life re as a relatively stable if, on
standalone basis, lower margin business compared with non-life re. We estimate that
the volatility of operating earnings in life re is 45% that of non-life re for all the
European reinsurers, on the basis of the ratio of standard deviation to premiums (the
ratio is 6.9% on average for the four listed European reinsurers for non-life, and
3.1% for life. We note that this comparison includes Swiss Re with the reported
accounting change in 2007, which now gives life re just the risk free return. We
believe this combination of perceived low risk and extra return for little extra capital
is the key reason the main European reinsurers all have substantial life reinsurance
operations.
Table 2: The attraction of life re is that it brings substantial diversification benefit - SCOR
€mn
Jan 2009 RBC Jan 2009 RBC Diversification
standalone diversified benefit
Non-life 2,800 2,400 14%
Life 1,900 1,000 47%
Total 4,700 3,400 28%
Life to total 40% 29%
Operating profit
2009
Non-life 188
Life 154
Group 342
2008 pretax ROE
Standalone Diversified
Non-life 6.7% 7.8%
Life 8.1% 15.4%
Group 7.3% 10.1%
Source: Company reports and J.P. Morgan estimates.
The above table shows how this works for SCOR. On a standalone basis for 2009
non-life earned 6.7% pretax ROE, and life 8.1%. However, when allocating the
diversification benefit life re earned 15.4%, and the group's total pretax ROE was
lifted from 7.3% to 10.1%.
Split of risk by segment: Example of Munich Re

We have shown below the type of risks for each of the products. Mortality accounts
for most of the biometric risk for Munich re at 68% (2009). Munich Re mostly
focuses on the mortality and morbidity risk.
5
(44-20) 7325-9175 02 September 2010
Table 3: Type of risks according to products: Munich Re mainly focus on the mortality and
morbidity
as stated
Products Ordinary life Group life Living benefits Annuity
Type of risks
Mortality Full cover Full cover
Morbidity Full cover
Longevity Full cover
Lapse Selective cover Selective cover Selective cover
Investment Selective cover Selective cover
Pct of total Ord and group life together 68% 29% 3%
Source: Munich Re presentation, Biometric risk portfolio in share of net premium, 2009
Valuation gap
We show below the summary valuation of life reinsurance within the four listed
European groups. We have used 76% of our forecast 2010e embedded value (EVM
in the case of Swiss Re) as a measure of the market value of life embedded value, s
this is the average multiple for the listed life insurers we follow.
Among these stocks our two Overweight recommendations are Hannover Re and
SCOR. Hannover appears as, implicitly, the most undervalued for the operations
other than life re, when life re is valued at 76% of embedded value. SCOR also looks
attractively valued. It is above Swiss Re in terms of other than life re valuation
multiples, but Swiss Re has a significant portfolio of pre 2004 mortality risks, which
were underpriced and which are diluting life reinsurance profitability.
On the assumption that the market would value life reinsurance on the basis of 76%
of embedded value, rather than just on reported IFRS/US GAAP earnings, then
Hannover Re is still most undervalued, followed by Swiss Re, then SCOR, and
finally Munich.
However, we believe that the market which invests in these reinsurers mainly on the
basis we believe of their ability to improve their non-life reinsurance and cash flows,
may be unwilling to value the life reinsurance on the same basis as standalone life
primary insurance groups.
In the case of Swiss Re, we believe this sum of parts approach using life embedded
value would be relatively unlikely for now; firstly because we believe that the market
cares about US GAAP earnings, where there is still dilution from pre 2004
underpriced mortality business, and secondly because we believe it would be
challenging to split life re out of Swiss Re.
There are three reasons we believe it would be challenging to split out life re from
Swiss Re, even in theory:
1. Business. Swiss Re runs its life reinsurance business co-mingled with its non-life
re unit. This means that in some geographies Swiss Re runs a composite business
within the same legal entities.
2. Diversification benefit. On Page 77 of its 2009 annual report Swiss Re shows the
capital requirement based on 99% tail VaR. This is SF25.9bn before
diversification, SF16.0bn after deducting SF9.9bn of diversification.
Diversification comes from all the four risk groups (non-life, life and health,
financial market, credit) but if we just allocate to life its share according to gross
6
(44-20) 7325-9175 02 September 2010
capital, then the capital used by life re as part of the group is we estimate SF5.5bn
- SF2.1bn share of diversification, ie SF3.4bn. So the life re unit as a standalone
would have to earn 62% more than as part of the group for the same economics.
3. Reporting. There is no separate balance sheet for life reinsurance.
Table 4: Group valuation ex Life Re business - €m and SFm in the case of Swiss Re
$m Munich Swiss Hannover SCOR
Market cap - from BBG as on cob 30th Aug 18,922 15,387 4,172 3,146
Less 76% of life EV/EVM 2010e -5,391 -7,934 -3,021 -1,486
Add back life re debt 1,569 0 477 366
Market cap excluding life re 15,100 7,453 1,629 2,026
Note 100% of life 2010e EV/EVM is JPMe 7,094 10,440 3,974 1,955
Debt:
Total sub debt FY09 4,790 7,172 1,481 629
Less life re sub debt -1,569 na -477 -366
Debt ex Life Re 3,221 7,172 1,004 263
Forecast 2011e net profit 2,473 1,934 648 478

Less life re taxed operating profit -850 -567 -203 -141
Forecast 2011e ex life re 1,623 1,367 445 337
Ratio market to 2011e net profit:

Total 7.7x 8.0x 6.4x 6.6x
Life re assuming 76% valuation to EV 4.5x 14.0x 12.5x 7.9x
Total excluding life re 9.3x 5.5x 3.7x 6.0x
We have also shown the split of our price target by line of business below.
Table 5: Reinsurers - Split of JPMe price target by line of business

%
Split of price target % Life Re % Non Life Re % Asst Mgt % others (incl debt)
Hannover Re 35.2% 79.6% 0.0% -14.8%
Munich Re 44.1% 62.0% 2.0% -8.1%
SCOR 34.1% 83.8% 0.0% -17.9%
Swiss Re 24.2% 91.9% 146.3% -162.4%
Source: J.P. Morgan estimates.
Summary split of business

We show below the split of business of the European listed reinsurers, which
includes the premiums and profit from life reinsurance. They all have sizeable life re
operations, but in the case of Munich this is diluted by their primary insurance units.
Table 6: Reinsurers - Split of 2010e operating profit by line of business

%
Life Re Non life Re AM Others
Hannover Re 27.9% 69.8% 0.0% 2.3%
Munich Re 28.2% 59.5% 1.0% 11.3%
SCOR 33.7% 66.3% 0.0% 0.0%
Swiss Re 30.5% 68.8% 163.5% -162.8%
Source: J.P. Morgan estimates
7
(44-20) 7325-9175 02 September 2010
Table 7: Reinsurance business split by premiums, 2010e

LCm
Health Non Life
Life Re Non Life Re Life Primary Primary Primary Others Total
Hannover 4,202 5,399 9,601
Munich 7,602 13,917 11,949 4,700 4,319 42,487
SCOR 2,744 3,221 5,965
Swiss Re 8,307 10,952 5 19,264
Source: J.P.Morgan estimates
High level summary with references

European listed reinsurers have significant life re business
All the European listed reinsurers have significant life re businesses, we believe
because this way the group as a whole benefits from diversification benefit, and
because life re excluding asset risk offers relative stability of earnings relative to
P&C. The pricing is stable due to the high entry barrier and the long term nature of
the Life Re contracts (in comparison with P&C). As an example, we note the long
term stable market share of Munich Re (see Figure 2).
Relative valuations of the remainder look attractive

The relative valuations of the remainder (Hannover most attractive, then SwissRe
and Scor and finally Munich) look attractive if assume life re valued at 76% of 2010e
embedded value (see Table 4).
Low level of transparency for Life Re in Europe, particularly compared with

RGA, the US listed pure life reinsurer
The Life Re segment has relatively low level of transparency and high degree of
accounting complexity, particularly compared with RGA, the US pure life listed
reinsurer. We have summarized the three different types of accounting & the way US
GAAP and IFRS work in terms of historic based reserves valuation with a buffer for
adverse deviation (see Table 17).
Asset risk of Life Re

We have summarized the asset risk from Scottish and Swiss Re. For Scottish Re
asset risk accounted for 85% of the earnings volatility during 2005-1H10 & just 15%
for mortality and other life operating risks (see Table 8). The only year operating risk
was significant was in 2006 and this was due to rating downgrades of Scottish Re
which started that year and which lead to an accumulation of negative reserve
adjustments. Swiss Re switched its accounting from 2007 and now allocates only risk
free rates (except for variable annuity & unlit linked) to the operating units. The
reason is that this is how the business is priced. We have shown (see Figure 4 &
Table 19) the drop in the investment income as well as the lower volatility in the new
style against the old style.
Sensitivity comparison
Life re is most sensitive to mortality and lapse risk, primary is mainly to interest rate
risk (refer Table 28 & Table 30). On an average for primary insurers, MCEV drops
by 14.6% for every 1% drop in interest rates and for reinsures increase 20.5% on a
5% drop in mortality/morbidity assumption.
8
(44-20) 7325-9175 02 September 2010
Swiss Re: uses risk free only to credit life re, has interest rate sensitivity due to
being short assets
US GAAP gives a relatively poor view of profits as it is mainly focused on baked in
assumptions on mortality, lapses and profitability. So has a very stable earnings
unwind, but says little about new business. The actual accounting that allocates only
the risk free to the operating units, life re’s standard deviation drops to SF0.4bn; in
relation to premiums its earnings volatility is now just one third that of non-life,
respectively, 2.5% and 7.9% (see Table 19). The main accounting conventions in life
are FAS60 (traditional life, constant emergence of margins, relative mainly to
premiums) and FAS97. The risk of write downs of the various intangibles (value of
acquired business in force, deferred acquisition cost) is low and we believe remote.
This is because, under US GAAP, the portfolio as a whole would have to show
negative returns (loss recognition is at the aggregated level), for this to happen.
We note that Swiss Re doest not report return on operating revenue (it stopped doing
so in 2007) as this is not a very consistent or comparable guide to underlying
profitability. We believe the measure that would best help understand profitability is
ROE for the life re unit alone, and none of the European listed reinsurers provide this
(see Table 54). We also note the comparison of the sensitivity of Swiss Re with peers
and primary insurers in Table 31.
Munich Re: positive interest sensitivity in life re

Munich’s split of business (as Dec09) in terms of net earned premiums was 68%
mortality, 29% living benefits including morbidity, 3% longevity and other (refer to
Table 37). Mortality, in particular, has little interest sensitive (morbidity has more as
we believe it has relatively more assets). Munich Re also have positive interest rate
sensitivity in Life Re (seeTable 28)
SCOR: possible venture into longevity

Excluding the equity indexed annuity business, which is a very low risk, high volume
and we believe relatively low margin product, and which SCOR has said they would
cut back sharply in 2010, life and financing, where the main risks are mortality and
lapses, accounted for 71% of total premiums in 2009. We believe this highlights the
predominance of traditional and relatively predictable business in SCOR's business
life reinsurance business mix (see Table 43)
Hannover
Hannover has a relatively high sensitivity to lapse and mortality risk compared with
its reinsurance peers. We believe this reflects its business mix, which we believe is
structured to use lapse risk and mortality risk to help provide front end commission
funding for primary insurers (see Table 31)
Nature of the life reinsurance market

• Committed long term partnerships between life insurer and reinsurer
• Strong client network needed
• Extremely knowledge and expertise driven industry
• Lower volatility of profits and higher predictability of results and cashflows
• Almost all business is directly written
9
(44-20) 7325-9175 02 September 2010
• Long lasting acquisition process for new business of 6 to 12 months

• Demanding regulatory requirements
• Local nature of the business supports the high barriers of entry
• High independencies of risks within homogeneous portfolios
• Correlation to non-life reinsurance only in a very limited number of event
scenarios – means high diversification benefit
• Excellent data availability increases statistical certainty
Five types of risk

Biometric risk – longevity, mortality, morbidity.
Mortality affects policyholders buying protection for their families or for their
mortgage finance, which is generally a younger age group than longevity, which is
effectively a risk that pensioners, ie generally older policyholders, live longer.
Mortality is often underwritten, particularly the case for RGA, on an YRT (yearly
renewable term) basis, which means there is no asset risk for the reinsurer. YRT
effectively functions on a pay as you go basis. Mortality, for the past ten years or so,
has been reinsured assuming the trend does keep improving, and normally the credit
given is 1/3 to 1/2 of the historic trend.
Longevity is mainly a pensions risk, and is present mainly in the US and the UK.
Longevity is relatively risky because there are few hedges, which is effectively a risk
that pensioners, ie generally older policyholders, live longer. For that reason the most
successful longevity reinsurers like Hannover reinsure niches like impaired annuities
(this is groups of pensioners who are expected to live shorter than the average) such
as typically Glasgow manual workers, in combination with the life company Just
Retirement. Despite its riskiness, longevity is becoming more attractive as there is
the possibility now to reinsure it on a swap basis, with no asset risk. Also arguably,
there is zero and possibly even small negative correlation with mortality risk. For that
reason (significant diversification benefit) we believe Swiss Re on 15 Dec 09 entered
into a longevity reinsurance agreement with 11,000 in payment pensions (total
SF1.7bn liabilities) for civil servants in the County of Berkshire in the UK. The
assumption is that, given the size of Swiss Re’s mortality book, there is some offset
for increased longevity.
Morbidity can fluctuate with the economy, as this is a category that can be used by
employers as a substitute for lay offs or early retirement. Morbidity so far continues
to be associated with asset risk.
Financial risk / wherever there is an accumulation of reserves to be invested.

The highest risk has been associated with variable annuities including GMDB where
the risk for the reinsurer fluctuates as the inverse of the account value (ie the lower
the available account value, the higher is the risk covered by the reinsurance
premium). Variable annuities in general pose the greatest financial risk as they are
difficult to hedge precisely, unless they are equity indexed annuities, where the
underlying asset is known and precisely hedgeable. Calculatory risk/behaviourial
risk for lapses. This is the risk of getting the behaviour of policyholders wrong. So
10
(44-20) 7325-9175 02 September 2010
credit life where the risk covered is payment protection against death, disability or
illness is a risk, which fluctuates with the economy. Also, long term care, where
recently new entrants in the US were affected by losses as they underestimated both
the incidence of claims, and also the length of time claimants would need cover.
Protection against catastrophes / pandemics. This is a risk which can partially be

hedged in the capital markets, although RGA has found there to be significant basis
risk since the capital markets structure is based on general population mortality not
the insured population or the actual lives insured by the respective company that
purchases the coverage. Swiss Re uses insurance linked securities, SCOR has a swap
with JPMorgan as reported by SCOR. . The risk when it is hedged is priced in terms
of deviation from normal mortality, with cover starting at 105% of norm, rising to
120% of norm, which would be a very high peak risk.
Accounting risk. Reinsurers in their presentations understate this risk, possibly

because life re is such a technical subject with mostly tailor made contracts relatively
limited transparency and limited cyclical volatility, that the actuaries who run life re
see relatively modest fluctuations in the underlying business and so perceive less
need to explain them than in non-life re. But there are three separate accounting /
regulatory conventions, which sometimes clash, statutory, IFRS/US GAAP and
Embedded value.
Statutory accounting: costs are mainly written off in the first year, or only deferred
on a very limited basis (zillmer adjustment in Germany). Statutory sets the
distributable cash flows as annual results.
Embedded value discounts these statutory cash flows over the life of the contract,
but says little about the amount of cash which can be distributed in the year. It
recognizes earnings upfront as the point sale is deemed to be the moment of value
creation, and future cash flows from the contracts sold are then discounted back.
IFRS (and also US GAAP) is a compromise solution, which sets the reserves and
the profit profile of the policy upfront based on the pricing assumptions of mortality,
lapses and investment return, and reports smoothed earnings using the mechanism of
deferred acquisition costs to smooth results. These earnings are not automatically
distributable and if the company got the pricing wrong, only the annual variance is
shown, after offset against adverse deviation reserves.
Part of the reason IFRS accounting for life reinsurance is a poor guide to profits is that
there is actually no set rule for the calculation of life liabilities. There is a new IFRS
accounting convention being prepared, which, will, we believe, provide a mark to
market approach to valuing life re liabilities, but it is still some time away. In the
meantime, the approach used is based on US GAAP: life re reserves are calculated
using the original pricing assumptions, and are only reset when the adverse deviation
reserves are exhausted. This can be for the global account as a whole, or as the case of
Swiss Re, for subsegments such as life, health and admin re. So life companies see
embedded value as a better guide of profitability. Embedded value is included as
available capital in Solvency 2, and so is one of the types of capital against which the
insurer can borrow. But the risk with embedded value is that it is highly volatile because
the calculation of embedded value under the MCEV convention is reset every year
using the point values at the year end of interest rate, yield curve, volatility of interest
11
(44-20) 7325-9175 02 September 2010
rates and equity prices. Embedded value for primary insurers halved in 2008 and almost
doubled in 2009 (however Embedded values for reinsurers were far less volatile).
There are three main points we believe about accounting risk:
a) There is no agreed measure of profitability under IFRS, as for example the

combined ratio in non-life. We believe investors would like to know the return on
equity, but reinsurers mainly report the ratio of operating profit to premiums (RGA,
SCOR, Hannover) or benefit ratio (SwissRe). This is not strictly comparable as when
a reinsurer is more exposed to FAS97 contracts then the denominator of premiums is
low compared to peers, and the profitability ratio of profit to revenues appears
relatively high. FAS97 splits off life reinsurance contracts where there is negligible
risk transfer to the insurer/reinsurer, which is the case for universal life or variable
annuities. Under FAS97, the accounting is that the premiums are treated like
deposits, and the main item of revenues is fees.
Another potential distortion is when there is a lot of growth from acquisitions, as the
acquired blocks of business are accounted under purchase GAAP. This means the
profits are mainly capitalised upfront with costs amortised over the duration of the
contracts, so the annual profitability appears lower than for organically grown
business.
b) Transparency is relatively low because life re is shown normally just as one

aggregated column in the accounts. The only life reinsurer that reports profitability
by type of product/risk is RGA, the specialist US life reinsurer. And even RGA
pointed out on the 2Q10 results call that it would improve transparency if variable
annuities were reported separately from other annuities in the asset intensive line. All
the other reinsurers aggregate all life reinsurance risks as one. This makes
understanding and explaining the risks very challenging in our view, as the risks and
flows and profit margins under longevity, critical illness, credit life and asset
intensive risks like variable annuities are quite different. We believe this is one
reason why investors are uncertain how to value life re.
c) The reporting is idiosyncratic. For example, Swiss Re does not allocate the
assets acquired for the life reinsurance business to life re, but instead to an Asset
Management function, and reallocates to life re the risk free current return, instead of
the actual return of the invested GAAP liabilities. As life re, particularly admin re
where Swiss Re is particularly active, invests substantially in corporate bonds, this
presentation tends to understate the total returns of life re when interest rates are
falling, and also tends to understate the volatility of the returns. This presentation
reflects pricing of the risk, and also how management is incentivised – life re is not
credited with asset risk returns.
Extra source of accounting noise. a further contrary element of financial hedging

costs, is that the cost of hedging is calculated for the purpose of the accounts using
the reinsurer’s own cost of debt. So, for example, variable annuity liabilities are
treated as embedded derivatives, which are measured at fair value. For example,
where these are reinsured by Swiss Re, which saw in 2008 its CDS spreads rise
sharply, there was an accounting reduction in the value of its liability (the accounting
assumes that Swiss Re effectively provides reduced cover due to its reduced implicit
credit rating as measured by the CDS spread) and so an increase in earnings. The
12
(44-20) 7325-9175 02 September 2010
effect is exactly opposite when CDS spreads narrow. Note, the actual hedging cost is
set by contract and does not vary.
The last accounting risk (which could also be treated as financial risk) is that on
contracts where the reserves are kept and invested by the primary insurer, and the
reinsurer operates on a coinsurance basis, the changes in value of the underlying
assets appear as mark to markets through the profit and loss account because they are
seen as embedded derivatives. This is called Modified Coinsurance or Coinsurance
with Funds Withheld. Under modco, liabilities and assets stay with direct company,
but under coinsurance with funds withheld, the liabilities move to the reinsurer, while
the assets stay with the direct company. Both structures create the embedded liability
B36. But if the reinsurer had insisted on taking the assets on its balance sheet, the
fluctuations in value would normally have been accounted for as available for sale, ie
mark to market through the balance sheet and not the profit and loss account.
Effectively, B36 means that the fair value mark to market is linked to the CDS of the
counterparty managing the investments.
Whilst statutory reporting is usually linked to explicit capital requirement this does
not hold for IFRS and EV. For EV publication purposes, companies usually take
statutory rating and economic capital requirements into account in order to take an
economic view. We believe that in order to make a like for like comparison, it would
be more appropriate to take a similar approach also for statutory and IFRS
accounting and compare accounting profits to the same level of capital requirement,
possibly adjusted for accounting differences only.
13
(44-20) 7325-9175 02 September 2010
Summary of Scottish Re profit and loss record 2005-1H10 –

asset risk accounted for 85% of the earnings volatility in the
period vs just 15% for mortality and other life re operating
risks
We wrote on the front page that life reinsurance, outside of asset risk, is relatively
dull and stable compared with non-life re. But then investors we have spoken to
asked, what about Scottish Re, the life reinsurance which for a period in 2008/9 faced
such losses that shareholders’ equity turned negative and it operated under direct
Order of Supervision of the Insurance Commissioner of Delaware.
The following section summarises the experience of Scottish Re, which we believe
supports our point that asset risk is the largest single risk for a life reinsurer, as this
accounted for 85% of annual earnings volatility in 2005-1H10. By contrast,
mortality, lapses and other operating risks accounted for just 15% on average we
estimate.
The following chart shows the experience of Scottish Re in 2005-1H10.
Figure 1: Scottish Re profit/loss and shareholders' equity 2005-1H10

$m
3,000
2,000
1,000
Shareholders' equity
0
Net income/loss:
2005 2006 2007 2008 2009 1H2010
-1,000
-2,000
-3,000
Source: company reports
The key events are as follows:
1. Purchase in Oct 04 of the ING Life re book of business in the US for a ceding
commission of $560m. In other words, ING paid Scottish Re to take the business
off its books. This ceding commission represents $200m for DAC write off
(effectively mortality losses), and the remainder was effectively a $360m
contribution to the capital required to run this business. In addition, Scottish Re
raised $230m to fund the regulatory capital base (RBC) of this business. The
issue with the ING life re business is that it was mainly written on level premium
terms, which is subject to the Valuation of Life Insurance Policies Regulation
XXX (Reg XXX).
14
(44-20) 7325-9175 02 September 2010
2. 2005 reasonable profit. During 2005 Scottish Re refinanced the collateral finance
initially provided by ING mainly by securitising collateral funding through a
number of special purpose vehicles. These special purpose vehicles were non
recourse and raised collateral funding, which was passed through to Scottish Re.
Scottish Re invested the proceeds mainly in Alt A and subprime for the yield
pickup (50% of all its investments in Dec06).
3. 2006-8 accelerating write downs on assets, triggered in 2006 by a write off of
deferred tax, and leading in 2008 to a very large trading loss on the underlying
subprime and Alt A assets.
4. 2009: $0.7bn gain on sale of ING Life Re to Hannover, and another $1.2bn
linked to this (Scottish Re effectively deconsolidated a unit whose sole purpose
was to provide reinsurance for part of the ING Life Re).
5. 1H10: modest gains on investments and final contingency profits on the sale to
Hannover Re.
We have gone through the filed reports of Scottish Re for the period and summarised
the main sources of loss below. The key point is that most of the earnings volatility is
due to asset risk, and only a relatively modest proportion to operating and mortality.
On average for the period, operating shifts accounted for just 15% of the change in
earnings, asset risk for the remaining 85%. The only year where operating risk was a
significant factor was 2006, and we believe this was due to the rating downgrades of
Scottish Re, which started that year, and which lead to an accumulation of negative
reserve adjustments: cedant true ups, experience refunds, premium accruals.
Reserve adjustments under US GAAP for life re only take place when the reporting
division as a whole is loss making. Here, the combination of various sources of
reserve adjustments did indeed push the earnings to a negative. One factor which
also pushed earnings to a negative is a write off of deferred tax as the State of
Carolina refused a change of accounting which Scottish Re requested and which
would have allowed Scottish Re to recognise the benefit of tax schemes.
Table 8: Scottish Re Net income and shareholders' equity 2005-1H10

$m
2005 2006 2007 2008 2009 1H2010 Average
Shareholders' equity 1,272 1,057 347 -2,410 -104 69
Net income/loss: 325 -377 -1,025 -2,710 2,305 178
of which realised and unrealised losses 4 -17 -927 -1,922 243 216
of which trading securities -1,875
of which impairment charge -777
embedded derivatives loss B36 -8 -16 -44 -200
of which gain/loss on sale to Hannover -120 704 60
of which deconsolidation gain Ballantyne 1,150
of which gain on repurchase of debt 20 19 253
of which deferred tax write off -203
sale of Life Re international -31
sale of Wealth Management -5
of which DAC write off -151
of which collateral unwind costs -10
Other (mainly operating) 329 -141 -74 -290 -45 -98
Pct of net profit swing due to operations 67.0% -10.3% 12.8% 4.9% 2.5% 15%
Pct of net profit swing due to asset risk 33.0% 110.3% 87.2% 95.1% 97.5% 85%
Source: Company reports.
15
(44-20) 7325-9175 02 September 2010
The interesting point is that as soon as accounting turned negative, then it seemed to
trigger in subsequent years an accumulation of further negatives. For example, as the
2006 loss eroded capital, the potential 2007 write down on Alt A and subprime could
no longer be argued away as just temporarily impaired, because Scottish Re no
longer had enough capital to guarantee that it would be willing and able to hold the
underlying assets to maturity. This impairment loss is the $777m item in 2007, the
main cause of the $1.0bn net loss in 2007. In 2008, the risky assets were treated as
trading instead of available for sale, which meant that the loss was taken direct to
earnings, without any review of the impairment logic.
And in 2009 the losses virtually all reversed thanks to the recovery of the value of the
subprime and Alt A assets and the sale of ING Life Re to Hannover Re. This in total
allowed Scottish Re to book a gain on sale of near $1.9bn, the main factor in the
$2.3bn net profit in 2009. We do note that in total for 2005-1H10, the cumulative net
loss is -$217m.
Where did these asset losses come from. Effectively, they were the result of an
investment policy guided by the need to overfund life reserves to their full statutory
level (this is higher than US GAAP due to the regulatory requirement to fund
mortality reserves on a very conservative basis, called Regulation XXX). So, in order
to pay collateral fees of around 1.5% and still make a profit, Scottish Re invested in
risky assets.
We show the investment asset split in the key quarters after the ING life re deal.
Mortgages and asset backed rose from 40% of total investments Dec05 to 54%
Dec06. At that Dec06 date, Alt A and subprime together accounted for 50% of
invested assets. And the yield pick up relative to US treasuries was maintained at 80-
90bps, sufficient we estimate to pay collateral fees and also generate a positive
margin.
Table 9: Scottish Re Summary investments in $m

Dec-05 Mar-06 Jun-06 Sep-06 Dec-06
Treasuries 110 267 180 111 127
Corporates 3,688 3,632 4,378 4,050 4,005
Munis 71 73 80 82 82
Mortgage and asset backed 3,745 4,464 5,408 5,276 5,708
Preferreds 137 135 131 133 136
Commercial mortgage loans 113 108 106 104 99
Cash 1,445 610 1,475 1,134 475
Total 9,309 9,289 11,758 10,890 10,632
Weighted average book yield 5.1% 5.3% 5.5% 5.6% 5.6%
Option adjusted duration 3.6 3.5 3.4 3.3 3.3
Yield for US govvies 3 year 4.3% 4.7% 5.1% 4.6% 4.7%
Yield pick up 0.8% 0.6% 0.4% 1.0% 0.9%
Worth 74 56 47 109 96
Source: Company reports, Bloomberg for govvies yield, JPMe for yield pickup.
We show the structure of Scottish Re’s balance sheet below. The key point is that
collateral finance facilities rose from 2% of total liabilities just when the ING Life re
closed, to 30% Dec06, from 20% of total capital in Dec04, to 320% in Dec06. And
we believe the investment policy for these assets was mainly asset backed, and in
particular Alt A and subprime, leading to significant asset volatility, which Scottish
Re was not able to offset.
16
(44-20) 7325-9175 02 September 2010
Table 10: Scottish Re Summary balance sheet in $m and debt to capital ratios
Dec-04 Dec-05 Mar-06 Jun-06 Sep-06 Dec-06
Assets
Investments own managed 4,328 6,893 6,952 9,883 9,078 8,871
Funds withheld 2,056 2,597 2,610 2,175 2,076 1,942
Total investments 6,384 9,490 9,562 12,058 11,154 10,813
DAC 417 595 620 643 620 619
Reinsurance balances 1,176 987 986 838 994 1,036
Other 192 283 308 289 319 285
Segregated assets 783 761 780 776 739 683
Total 8,952 12,116 12,256 14,604 13,826 13,436
Liabilities and capital
Reserves for policies 3,301 3,526 3,539 4,101 3,663 3,919
Interest sensitive contract liabilities 3,181 3,907 3,990 4,089 3,583 3,399
Collateral finance facilities 200 1,986 1,986 3,725 3,745 3,757
Long term debt 245 245 245 245 245 130
Segregated liabilities 783 761 780 776 739 683
Other 227 267 323 328 415 339
Total liabilities 7,937 10,692 10,863 13,264 12,390 12,227
Minorities 10 9 9 9 10 9
Mezzanine equity 142 143 143 143 143 143
Shareholders equity 863 1,272 1,241 1,188 1,283 1,057
Total capital 1,015 1,424 1,393 1,340 1,436 1,209
Total capital and liabilities 8,952 12,116 12,256 14,604 13,826 13,436
Ratio: Capital to liabilities 12.8% 13.3% 12.8% 10.1% 11.6% 9.9%
Collateral finance to total liabilities 2.5% 18.6% 18.3% 28.1% 30.2% 30.7%
Collateral finance to capital 19.7% 139.5% 142.6% 278.0% 260.8% 310.8%
Mortality was indeed negative, as the rise in the benefit ratio (ratio of claims to
premiums) shows in the following table.
Table 11: Scottish Re Adverse mortality - benefit ratios

2005 2006
US traditional 74.3% 85.0%
Total 70.5% 79.4%
Grand total 75.0% 87.0%
Source: Company reports
But the cost in terms of earnings was relatively modest, with total mortality variance
cost of $30m in 2006 and lapse cost of $27m.
Table 12: Scottish Re Increased mortality costs have a relatively modest impact on 2006 earnings
Increase in benefit ratio in 1Q06 in smaller claims by $16m.
Mortality also adverse in 4Q06 by $14m
Total 2006 mortality was 103% of expected.
Lapse assumptions cost $13m in 3Q and $14m in 4Q as lapses
rose on contracts with higher margins, fell on those with lower margins.
Collateral finance expenses rose 376% in 2006 vs 2005 due to increased facilities.
The following table shows more detail of the quarterly change in 2006 earnings, the
year that mortality and lapses had their greatest impact on earnings, accounting for
67% of the swing from profit to loss over 2005 to 2006.
17
(44-20) 7325-9175 02 September 2010
Table 13: Scottish Re Summary operating results quarterly in $m

Dec-04 Dec-05 Mar-06 Jun-06 Sep-06 Dec-06
Net operating result 27 51 14 -130 -24 -228
OF which deferred tax valuation allowance -112 -91
The 2006 loss was linked to the ratings downgrades which increased re costs, and the adverse effect of 2006
adjustments to previous estimates. Valuation allowances result from actual results of legal entity statutory income and
changes in statutory reserves combined with a reassessment of tax planning. 2006 saw $167m higher collateral
finance expenses and $34m goodwill write off. Modco impact was -$8m in 2005 due to change in value of embedded
derivatives and Modco impact was +$6m 2006. Modco is consequence of DIG B36 / SFAS133 which requires
bifurcation of embedded derivatives as pools of fixed income are deemed to contain embedded derivatives requiring
bifurcation. Interest on collateral financing is offset by investment income on fixed maturity assets. Net margin
represents the excess of the yield on earnings assets over the interest rate cost
Source: Company reports and J.P. Morgan estimates for summary of management discussion in SEC filing.
Our conclusion is that mortality and lapse risks are relatively manageable even as in
the case of Scottish Re, which bought a block of business from ING where mortality
was a risk due to historic underpricing, and which negotiated a $560m ceding
commission (discount) to help fund that risk. But, by comparison, asset risk leads to
swings that Scottish Re could only absorb after resort to special conditions from the
Insurance Commissioner of Delaware.
As part of the turnaround measures in 2008, Scottish Re obtained approval from the
Delaware regulator to limit funding with additional capital of its securitisation units
Orkney I and II, which were facing a shortfall due to the fall in the value of their
investments in subprime and Alt A. This was linked to an Order of Supervision by
the Delaware insurance department set to lapse in April 2009 and obtained
authorization to use different mortality tables which boosted capital and surplus by
$190m.
Table 14: Scottish Re Funding of the collateral requirements in $m

Date Amount Consolidated Comment
HSBCI Jun-04 189 Yes
Stingray Jan-05 265 Partly Up to $0.3bn
Orkney I Feb-05 850 Yes
Orkney II Dec-05 450 Yes
HSBC II Dec-05 529 Yes
Reinsurance facility Dec-05 n/a n/a Up to $1.0bn
Ballantyne Re May-06 1,739 Yes
Total 4,022
Total facilities 5,122
The above table shows how the collateral funding was achieved, to replace the
facility initially provided by ING at the sale of the business to Scottish Re. The need
for collateral funding is because the business was largely Regulation XXX and
Scottish Re was aiming to reduce the cost of the funding provided by ING.
This is how Scottish Re describes the collateral funding: at the time of the deal in
October 2004 ING was obligated to maintain collateral for the Regulation XXX and
AXXX reserve requirement of the acquired business (excluding the business
supported by other arrangements) for the duration. Scottish Re paid ING a fee based
on the face amount of the collateral.
In 2005 and 2006 Scottish Re completed three securitisations that collectively

provided approximately $3.7bn in collateral to fund Reg XXX from part of the ING
18
(44-20) 7325-9175 02 September 2010
life re business, and used reinsurance to complete the funding. These deals
extinguished ING's obligation to provide collateral. They resulted in a refund from
ING of fees of $6.2m in 2006 and $6.7m in 205.
Table 15: Summary of Reg XXX reserves for level premium term life (mortality cover
The Valuation of Life Insurance Policies Model Regulation, commonly referred to as Regulation
XXX, was implemented in the United States for various types of life insurance business
beginning January 1, 2000. Regulation XXX significantly increased the level of reserves that
United States life insurance and life reinsurance companies must hold on their statutory financial
statements for various types of life insurance business, primarily certain level term life products.
The reserve levels required under Regulation XXX increase over time and are normally in
excess of reserves required under Generally Accepted Accounting Principles in the United
States ("US GAAP"). In situations where primary insurers have reinsured business to reinsurers
that are unlicensed and unaccredited in the United States, the reinsurer must provide collateral
equal to its reinsurance reserves in order for the ceding company to receive statutory financial
statement credit. Reinsurers have historically utilized letters of credit or have placed assets in
trust for the benefit of the ceding company as the primary forms of collateral. The increasing
nature of the statutory reserves under Regulation XXX will likely require increased levels of
collateral from reinsurers in the future to the extent the reinsurer remains unlicensed and
unaccredited in the United States. We believe that funding long duration liabilities with shorter-
term funding facilities is not suitable or sustainable from a prudent asset liability management
perspective because it creates significant refinancing or rollover risk every year.
The increasing nature of the statutory reserves required under RegXXX is because
under level premium policies the mortality risk rises as the policyholders get older,
but the premium stays flat, so the financial risk increases with time.
Scottish Re entered into a number of financing deals in 2004-6 to secure longer term
funding for a large portion of our RegXXX collateral needs. .Scottish Re then used
this to fund investment in risky assets. It is this combination of leverage and asset
risk which we believe was a key factor in Scottish Re’s accumulation of losses in
2007-8.
Other reinsurers avoid the need to fund RegXXX reserves either because they are
affiliated, or because they choose to cover only mortality risk, via contracts which
are called YRT (yearly renewable term). Here, the premium is increased every year
using a pre agreed schedule, and in theory is enough to pay for the expected mortality
claims that year. With such a pay as you go system there are no assets and no asset
risk either.
19
(44-20) 7325-9175 02 September 2010
Table 16: Types of mortality cover written by Scottish Re -citation marks refer to the 2009 annual
report of Scottish Re
Scottish Re writes in its 2009 report: “we wrote reinsurance generally in the form of yearly
renewable term, coinsurance or modified coinsurance. Under yearly renewable term, we share
only in the mortality risk for which we receive a premium. In a coinsurance or modified
coinsurance arrangement, we generally share proportionately in all material risks inherent in the
underlying policies, including mortality, lapses and investments. Under such agreements, we
agree to indemnify the primary insurer for all or a portion of the risks associated with the
underlying insurance policy in exchange for a proportionate share of premiums. Coinsurance
differs from modified coinsurance with respect to the ownership of the assets supporting the
reserves. Under our coinsurance arrangements, ownership of these assets is transferred to us,
whereas in modified coinsurance arrangements, the ceding company retains ownership of these
assets, but we share in the investment income and risk associated with the assets. Modified
coinsurance is treated as an embedded derivative under SFAS113 B36 and leads to earnings
volatility in the accounts of the reinsurers because changes in the value of these assets come
through the profit and loss account”.
Source: Company reports, JPMorgan estimates for description of SFAS113 B36.
The above table explains the asset riskiness of the three main contracts: none under
yearly renewable term (YRT) which is like a pay as you go contract, higher under
coinsurance in that the asset ownership is to the reinsurer, and available for sale
accounting (ie unrealized gains through the balance sheet) is used, and highest under
modified coinsurance, as, because of SFAS133 B36, the contract is deemed to be an
embedded derivative, and the unrealised gains of the underlying assets go to the
profit and loss account of the reinsurer.
Life reinsurance is relatively stable (apart from asset risk),

both under IFRS accounting and US GAAP.
The reason life reinsurance profitability is relatively stable compared with non-life re
is three fold, in our view:
1. Business reason for higher barriers to entry. Life reinsurance is negotiated

directly with the primary insurance client for a relationship in mortality which
can last for up to 30 years and with pricing which is quite opaque and a life
reinsurer sits on an inventory of inforce business from many different pricing
eras. So there is less volatility in pricing than in non-life, where the risks are
normally 1-2 years, and where around half of total reinsurance is through brokers,
with low barriers to entry and relatively transparent pricing. The pricing point is
key: in non-life re a slip is prepared, which shows the risk, the pricing agreed by
the lead reinsurer, and the proportion of risk taken by each reinsurer, with
variations in say the commission rate paid. This is seen by each reinsurer. By
contrast, in life re the reinsurer only sees his portion of the pricing and volume,
and relies on the primary insurer to say if the pricing or conditions are inline or
very different from the other reinsurers.
In the chart below we can see the Munich Re market share in US is stable within the
10% range. The high cession rates in the peak years 2000-05 are, we believe, mainly
due to ‘the irrational pricing by reinsurers’ (citation from Munich Re October 2008
life re presentation). In those years competitors which have since exited include ING,
and competitors which have significantly repriced include Swiss Re
20
(44-20) 7325-9175 02 September 2010
Figure 2: Munich Re - market share and cession rate

%
70% 62% 59% 62% 60%
56%
60% 51% 52%
47%
50% 42% 40% 37%
40%
30%
20%
10% 12% 11% 10% 13% 12% 11%
6% 6% 9% 7% 9%
0%
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007
Cession rate Munich Re market share

Source: Munich Re Investor day presentation on Life reinsurance 2008
2. Accumulation reason. Life reinsurance contracts typically run for up to thirty

years. The reason is that they often have a substantial financing component,
where the stream of profits from mortality or morbidity is used to pre-finance an
large front end cash advance to the primary insurer, called the reinsurance
commission, which the primary insurer uses to fund large front end acquisition
costs to sales networks. Alternatively, reinsurance can reduce the statutory strain
at the cedant level on a non cash basis. For this pre financing to work without too
much strain, the contracts tend to run for well over ten years, and up to thirty
years normally for mortality deals. So any one year’s production accounts for
around one thirtieth of the total business which is generating profits. SCOR gave
the example at its 2008 investor day that the new business of the year 2007
contributed €15m to the 2007 financial life results, ie 9% of €167m. By
comparison, in non-life reinsurance where the contracts are for one to two years,
one year’s new business accounts for around 40% of the profits of that year
(another 40% from the previous year’s new business, and 20% from prior years
and in particular reserve releases or additions).
Figure 3: Profit trend lines for new business example
Indicative figures in Y axis, X axis is years
60
40
20
-20
-40
-60
2007 2012 2017 2022 2027
Distributable cashflow IFRS profits EV profits

Source: SCOR investor day presentation, 2008
21
(44-20) 7325-9175 02 September 2010
3. Accounting reason. Under IFRS and even more so under US GAAP, the profits
shown each year are effectively the equivalent of the profits assumed when the
contract is priced. So the amortization of the deferred acquisition cost asset
depends on the amount of profit expected that year. The financial year profits are
therefore the sum of the assumed profit development, and the variances for that
year. And there are buffers called adverse deviation reserves to smooth even
those variances. So the reported profits are relatively stable.
In addition, there is little adjustment possible to reset the original reserves in life re if
the pricing assumptions are seen to be too optimistic. To adjust for this under US
GAAP, the adverse deviation reserve for the whole unit’s results should be a loss;
under IAS there is more flexibility but it is still relatively limited. So, books of
business written under terms which produce low but still positive returns, like US
mortality 1997-2004, particularly for Swiss Re, just continue to dilute results going
forward. The only way of resetting the results is to sell the business and recognise the
difference in expected profit as a gain or loss on sale. Swiss Re reinsured part of its
US life re business to Berkshire in January 2010.
We note that Swiss Re has, according to the 2009 EVM reporting, $11bn of
reserving margin, which we believe corresponds to the adverse deviation reserves.
Swiss Re accounts by segment (life, health and admin re) and would only reset the
original reserves if the adverse deviation reserve for that segment is exhausted.
By comparison, in non-life reinsurance the reserves for outstanding claims are

constantly reviewed against current payment trends and reset, and this creates a
stream of profits or losses from reserve releases or additions.
Comparison of the accounting standards

There are three accounting standards under US GAAP 1) FASB60 which is long
duration contracts of traditional life; 2) FASB97 which is the universal life where
there is no transfer of risk; and 3) FASB113 which has no transfer of risk. We have
summarized the standards of each of these below:
FASB60 for long duration contracts of traditional life

We have shown the summary of the FAS 60 from the statement of financial
accounting standards.
“This Statement extracts the specialized principles and practices from the AICPA
insurance industry related Guides and Statements of Position and establishes
financial accounting and reporting standards for insurance enterprises other than
mutual life insurance enterprises, assessment enterprises, and fraternal benefit
societies.
Insurance contracts, for purposes of this Statement, need to be classified as short-

duration or long-duration contracts. Long-duration contracts include contracts, such
as whole-life, guaranteed renewable term life, endowment, annuity, and title
insurance contracts, that are expected to remain in force for an extended period. All
other insurance contracts are considered short-duration contracts and include most
property and liability insurance contracts.
22
(44-20) 7325-9175 02 September 2010
Premiums from short-duration contracts ordinarily are recognized as revenue over

the period of the contract in proportion to the amount of insurance protection
provided. Claim costs, including estimates of costs for claims relating to insured
events that have occurred but have not been reported to the insurer, are recognized
when insured events occur.
Premiums from long-duration contracts are recognized as revenue when due from
policyholders. The present value of estimated future policy benefits to be paid to or
on behalf of policyholders less the present value of estimated future net premiums to
be collected from policyholders are accrued when premium revenue is recognized.
Those estimates are based on assumptions, such as estimates of expected investment
yields, mortality, morbidity, terminations, and expenses, applicable at the time the
insurance contracts are made. Claim costs are recognized when insured events
occur.
Costs that vary with and are primarily related to the acquisition of insurance
contracts (acquisition costs) are capitalized and charged to expense in proportion to
premium revenue recognized.
Investments are reported as follows: common and nonredeemable preferred stocks at

market, bonds and redeemable preferred stocks at amortized cost, mortgage loans at
outstanding principal or amortized cost, and real estate at depreciated cost. Realized
investment gains and losses are reported in the income statement below operating
income and net of applicable income taxes. Unrealized investment gains and losses,
net of applicable income taxes, are included in stockholders' (policyholders')
equity.”
We have also shown the relevant paragraphs from the FASB60 accounting.
Short term & long term contracts (para 3 & 4): The short term contracts are like
the property and liability insurance contracts which cover claims costs only for a
short and fixed duration. The insurance company can normally cancel the contract or
revise the premium at the beginning of each contract period. Therefore, premiums are
earned and recognized as revenue as the protection is provided. For the long term
products (including life insurance contracts) the expected policy benefits do not
occur evenly over the period. The insurance companies provide insurance protection,
sales, premium collection, claim payment, investment & other functions and services.
The premiums normally exceed the benefits in the beginning and hence as the
premium revenue is recognized a liability for claims costs. Hence, a liability for the
expected cost is accrued over each period.
Recognition for long term contracts (para 10): The premiums are recognized
when due from the policy holders. From this premiums some amount is accrued as
liability for future claims costs which is = the present value of future benefits –
present value of future net premiums. Please find the text from the para 10 below
“Premiums from long-duration contracts shall be recognized as revenue when due

from policyholders. A liability for expected costs relating to most types of long-
duration contracts shall be accrued over the current and expected renewal periods of
the contracts. The present value of estimated future policy benefits to be paid to or
on behalf of policyholders less the present value of estimated future net premiums to
be collected from policyholders (liability for future policy benefits) shall be accrued
23
(44-20) 7325-9175 02 September 2010
when premium revenue is recognized. Those estimates shall be based on

assumptions, such as estimates of expected investment yields, mortality, morbidity,
terminations, and expenses, applicable at the time the insurance contracts are made.
In addition, liabilities for unpaid claims and claim adjustment expenses shall be
accrued when insured events occur”
Acquisition costs (para 11): Acquisition costs are capitalized and charged in
proportion to revenue recognized. Other costs are charged as incurred.
Estimation of liability for future policy benefits (para 21): The assumptions used
in calculating the liability are applicable at the time insurance contracts are made.
These same assumptions shall be used in the subsequent periods as well unless there
is premium deficiency. We have shown the text from the standards below.
“A liability for future policy benefits relating to long-duration contracts other than
title insurance contracts (paragraph 17) shall be accrued when premium revenue is
recognized. The liability, which represents the present value of future benefits to be
paid to or on behalf of policyholders and related expenses less the present value of
future net premiums (portion of gross premium required to provide for all benefits
and expenses), shall be estimated using methods that include assumptions, such as
estimates of expected investment yields, mortality, morbidity, terminations, and
expenses, applicable at the time the insurance contracts are made. The liability also
shall consider other assumptions relating to guaranteed contract benefits, such as
coupons, annual endowments, and conversion privileges. The assumptions shall
include provision for the risk of adverse deviation. Original assumptions shall
continue to be used in subsequent accounting periods to determine changes in the
liability for future policy benefits (often referred to as the "lock-in concept") unless a
premium deficiency exists (paragraphs 35-37). Changes in the liability for future
policy benefits that result from its periodic estimation for financial reporting
purposes shall be recognized in income in the period in which the changes”
Premium deficiency shall be recognized and charged to income and either the
unamortized cost is reduced or future liability benefits is increased.
FASB 97
We have shown the summary of the FAS 97 below. We have also shown the changes
with the FAS 60 as highlighted below.
“This Statement establishes standards of accounting for certain long-duration

contracts issued by insurance enterprises, referred to in this Statement as universal
life-type contracts, that were not addressed by FASB Statement No. 60, Accounting
and Reporting by Insurance Enterprises. The Statement also establishes standards of
accounting for limited-payment long-duration insurance contracts and investment
contracts and amends Statement 60 to change the reporting of realized gains and
losses on investments. New life insurance contracts have evolved since the
development of specialized insurance industry accounting principles and practices in
the early 1970s. Many of those new life insurance contracts have different provisions
than do the life insurance contracts to which Statement 60 applies. Those new life
insurance contracts are characterized by flexibility and discretion granted to one or
both parties to the contract. Statement 60 identifies but does not address those
contracts, noting that the accounting was under study by the insurance industry and
the accounting and actuarial professions. This Statement requires that the
24
(44-20) 7325-9175 02 September 2010
retrospective deposit method be used to account for universal life-type contracts.

That accounting method establishes a liability for policy benefits at an amount
determined by the account or contract balance that accrues to the benefit of the
policyholder. Premium receipts are not reported as revenues when the retrospective
deposit method is used. The Statement also requires that capitalized acquisition costs
associated with universal life-type contracts be amortized based on a constant
percentage of the present value of estimated gross profit amounts from the operation
of a "book" of those contracts. Any gain or loss resulting from a policyholder's
replacement of other life insurance contracts with universal life-type contracts is
recognized in income of the period in which the replacement occurs. This Statement
requires that long-duration contracts issued by insurance enterprises that do not
subject the enterprise to risks arising from policyholder mortality or morbidity
(investment contracts) be accounted for in a manner consistent with the accounting
for interest-bearing or other financial instruments. Payments received on those
contracts are not reported as revenue
This Statement also addresses limited-payment contracts that subject the insurance
enterprise to mortality or morbidity risk over a period that extends beyond the period
or periods in which premiums are collected and that have terms that are fixed and
guaranteed. This Statement requires that revenue and income from limited-payment
contracts be recognized over the period that benefits are provided rather than on
collection of premiums. This Statement amends the reporting by insurance
enterprises of realized gains and losses on investments. Statement 60 previously
required that realized gains and losses be reported in the statement of earnings on a
separate line below operating income and net of applicable income taxes. This
Statement requires that realized gains and losses now be reported on a pretax basis
as a component of other income and precludes the deferral of realized gains and
losses to future periods. This Statement is effective for financial statements for fiscal
years beginning after December 15, 1988. Accounting changes to adopt the
Statement should be applied retroactively through restatement of all previously
issued financial statements presented, or if restatement of all years presented is not
practicable, the cumulative effect of applying this Statement is to be included in net
income of the year of adoption.”
We have shown below an illustrative example of the computation of the estimated

gross profit.
25
(44-20) 7325-9175 02 September 2010
Table 17: Illustration of the computation of estimated gross profit

$
Investment
income Interest
Mortality Mortality Expense Recurring related to credited to Revised
Surrender assessmen cost Assessmen Expenses policy policy Estimated gross profit
Year charges ts incurred ts incurred balances balances gross profit at year 2
1 13,298 17,300 (3,685) 11,700 (12,176) 6,405 (5,490) 27,352 27,352
2 13,169 15,099 (3,541) 9,356 (9,669) 10,571 (9,061) 25,924 34,637
3 7,314 14,104 (3,627) 8,229 (8,476) 14,436 (12,374) 19,606 17,822
4 4,656 13,604 (3,866) 7,566 (7,781) 18,356 (15,734) 16,801 15,273
5 3,645 13,199 (4,107) 7,108 (7,309) 22,405 (19,204) 15,737 14,304
6 2,739 12,791 (4,330) 6,676 (6,866) 26,286 (22,531) 14,765 13,422
7 1,929 12,950 (4,513) 6,270 (6,449) 29,957 (25,677) 14,467 13,151
8 1,208 12,905 (4,690) 5,888 (6,057) 33,447 (28,669) 14,032 12,756
9 567 12,755 (4,865) 5,529 (5,688) 36,779 (31,525) 13,552 12,320
10 0 12,593 (5,003) 5,191 (5,340) 39,965 (34,256) 13,150 11,954
11-20 0 108,164 (55,512) 37,183 (38,270) 551,879 (473,039) 130,405 118,543
21-50 0 140,607 (88,833) 32,577 (33,712) 2,618,726 (2,244,622) 424,743 386,112
Total 48,525 386,071 (186,572) 143,273 (147,793) 3,409,212 (2,922,182) 730,534 677,646
Present value 180,944 176,087
Source: FASB 97 page 27
We have also shown below the illustration of the change in amortization of DAC
after the estimation of gross profit.
Table 18: Computation of amortization rate

$
Original estimate Revised estimate
Present value of estimated gross profit, years one to fifty x 180,944 176,087
(from table above)
Present value of capitalized acquisition costs y 90,986 90,986
Amortization rate (y) / (x) 50.284% 51.671%
Source: FASB97 page 28
FASB113
We have shown the summary of the FASB113 from the FASB accounting standard below.
“This Statement specifies the accounting by insurance enterprises for the reinsuring
(ceding) of insurance contracts. It amends FASB Statement No. 60, Accounting and
Reporting by Insurance Enterprises, to eliminate the practice by insurance
enterprises of reporting assets and liabilities relating to reinsured contracts net of
the effects of reinsurance. It requires reinsurance receivables (including amounts
related to claims incurred but not reported and liabilities for future policy benefits)
and prepaid reinsurance premiums to be reported as assets. Estimated reinsurance
receivables are recognized in a manner consistent with the liabilities relating to the
underlying reinsured contracts.
This Statement establishes the conditions required for a contract with a reinsurer to be
accounted for as reinsurance and prescribes accounting and reporting standards for
those contracts. The accounting standards depend on whether the contract is long
duration or short duration and, if short duration, on whether the contract is prospective
or retroactive. For all reinsurance transactions, immediate recognition of gains is
precluded unless the ceding enterprise's liability to its policyholder is extinguished.
Contracts that do not result in the reasonable possibility that the reinsurer may realize
a significant loss from the insurance risk assumed generally do not meet the conditions
for reinsurance accounting and are to be accounted for as deposits.
26
(44-20) 7325-9175 02 September 2010
This Statement requires ceding enterprises to disclose the nature, purpose, and effect
of reinsurance transactions, including the premium amounts associated with
reinsurance assumed and ceded. It also requires disclosure of concentrations of
credit risk associated with reinsurance receivables and prepaid reinsurance
premiums under the provisions of FASB Statement No. 105, Disclosure of
Information about Financial Instruments with Off-Balance-Sheet Risk and Financial
Instruments with Concentrations of Credit Risk. This Statement applies to financial
statements for fiscal years beginning after December 15, 1992, with earlier
application encouraged.”
Life re is more stable than non-life except for asset risk

One reinsurer we consulted suggested we check our assertion that life reinsurance is
more stable than non-life, except for asset risk.
The first company we checked was Swiss Re, and we believe our analysis there
supports our view that:
ASSET RISK IS THE MAIN SOURCE OF VOLATILITY IN LIFE RE
Swiss Re is a good example in our view, because the group switched its accounting in
2008. Previously, investment income was allocated to the individual operating units
according to the investments selected by that unit. So if life re invested in corporated
bonds, it was credited with the investment return on those corporate bonds.
In 2008, Swiss Re switched its accounting and now allocates only the risk free
(except for variable annuities and unit linked) to the GAAP liabilities of its operating
units. This means that life re gets credited (except for variable annuities and unit
linked) with the investment income of its various vintages of life re assets (ie its gets
credited with a weighted average of the risk free according to the age mix of its
portfolio).
The reason Swiss Re changed the accounting and now only allocates the risk free
return to the operating units is that this is how business is priced. Also, managers of
the operating units are incentivised on their pure operating returns, not whether the
underlying assets beat the risk free rate.
This means that with Swiss Re we can check how important asset risk is by
comparing the earnings volatility of life and non-life re using the two accounting
bases. (We estimated the old style reporting for 2008-9 by reallocating out of asset
management and legacy investment income to life and non-life re, and the
reallocation was on the basis of the invested technical reserves, in an average ratio of
40% non-life: 60% life re).
We show the results below. We estimate that if Swiss Re had continued its old style
accounting, allocating investment results according to the underlying assets of the
operating units life re and non-life re, then the standard deviation of life re would
have been SF1.7bn, non-life re SF2.1bn; to give a better idea we divided the standard
deviation by average revenues 2005-9, and this shows that life re had a slightly
higher earnings volatility, in relation to the size of its revenues, than non-life (10.6%
and 10.2% respectively).
27
(44-20) 7325-9175 02 September 2010
Table 19: Swiss Re: Average operating profit and their standard deviation by reporting unit in SFm 2004-9
Old style As reported
Average SD SD as Average SD SD as
Old style Old style pct of revs Reported Reported pct of revs
Non-life 3,333 2,070 10.2% 3,400 1,596 7.9%
Life 1,014 1,702 10.6% 1,209 400 2.5%
Financial services/asset management 3,925 3,119 60.6% 4,327 3,426 66.5%
Legacy n/a n/a n/a -1,209 2,375 -106.8%
Group items -1,489 1,080 499.9% -1,472 1,068 494.3%
Allocation -4,119 3,551 -86.9% -3,591 2,925 -71.6%
Total 2,665 2,704 7.4% 2,665 2,704 7.4%
Source: Company reports and J.P. Morgan estimates. Old style is based on actuals 2004-7 and JPMe estimates for 2008=9, where JPMe reallocated investment income from legacy and from asset
management to the two main operating units according to their invested technical reserves.
By comparison, using the actual accounting which allocates only the risk free to the
operating units, life re’s standard deviation drops to SF0.4bn; in relation to premiums
its earnings volatility is now just one third that of non-life, respectively 2.5% and
7.9% (see above table).
We also show this in terms of the operating earnings. The following chart shows the
operating earnings as reported, and life re is definitely smoother than non-life, which
had a significant dip in 2005.
Figure 4: Swiss Re operating results non-life and life re, 2004-6 old style, 2007-9 new style (only
risk free investment return allocated to operating units)
SFm
6,000
5,000
4,000
Non-life
3,000
Life
2,000
1,000
0
2004 2005 2006 2007 2008 2009
Old style Old style Old style New style New style New style
By comparison, under our estimates of the old style operating earnings, where
investment income is allocated according to the underlying assets, life re is more
volatile, as it had a loss in 2008 under that basis (see following chart).
28
(44-20) 7325-9175 02 September 2010
Figure 5: Swiss Re operating results of non-life and life re, old style, as reported 2004-7 and JPMe
2008-9 in SFm
7,000
6,000
5,000
4,000
3,000
Non-life
2,000
Life
1,000
0
actual actual actual actual JPMe JPMe
-1,000
2004 2005 2006 2007 2008 2009
-2,000
-3,000
More numbers on earnings volatility of life re vs non-life re – it is consistently

lower
We show below a comparison of the earnings volatility of life and non-life
reinsurance for the European listed reinsurers. For each company we calculated the
standard deviation of operating earnings pretax of life re and non-life re, and divided
by net earned premiums (gross premiums in the case of SCOR, as net earned
premiums data is less readily accessible). In each case, the ratio of standard deviation
to premiums is between half and one third in life re as it is in non-life re.
Table 20: Comparison between SCOR, Hannover Re, Swiss Re and Munich Re (2004-09) – on
reported basis life re operating earnings volatility is 3% vs 7% non-life re
€m, CHFm, %
For 2004-09 Hannover Re SCOR Swiss Re Munich Re
Standard deviation - operating profit 427 134 2,704 993
Non life 344 103 1,596 821
Life 112 51 400 174
Average premiums 7,637 4,920 28,084 36,564
Non life 4,216 2,562 16,746 13,765
Life 2,708 2,358 11,021 7,508
Standard deviation/avg. premiums 5.6% 2.7% 9.6% 2.7%
Non life 8.2% 4.0% 9.5% 6.0%
Life 4.1% 2.2% 3.6% 2.3%
Source: Company reports. For SCOR, we have used GEP. Average ratio standard deviation to premiums is 6.9% in non-life, 3.1% in
life re.
We show this data in the following two charts. The first shows non-life operating
profit in 2004-11e, and the data is relatively volatile with low earnings in 2005 for
all, and also relatively low earnings in 2008, particularly for SCOR and Hannover.
29
(44-20) 7325-9175 02 September 2010
Figure 6: Non life earnings (operating profit) for SCOR, Hannover Re, Munich Re and Swiss Re
(2004-11e)
€m, CHFm
6,000 5,613
5,000 4,471
3,820
4,000 3,486
3,068 2,989
2,727 2,746 2,822 2,681
3,000 2,277 2,242
1,944 1,817
2,000 1,025 1,284
463 331
670 410 657
731
323
692 510 762
1,000 153 160 (28) 202 2 213
-
-1,000
2004 2005 2006 2007 2008 2009 2010e 2011e
SCOR Sw iss Re Hannov er Re Munich Re

Source: Company reports and J.P. Morgan estimates. Swiss Re in SFm, SCOR, Hannover and Munich in €mn
We show the life operating profits of the same groups over the same period in the
chart below, and here the data is much smoother, although all groups did report lower
profits in 2008.
Figure 7: Life earnings (operating profit) for SCOR, Hannover Re, Munich Re and Swiss Re (2004-11e)
€m, CHFm
1,800 1,643
1,546
1,600
1,304 1,320 1,334
1,400 1,175
1,105 1,091 1,077
1,200
922 934
1,000 805
746 757
698 697
800
600 372
400 230 277 278
140 167 146 161 165 188
77 83 93 78 121
200 46
-
2004 2005 2006 2007 2008 2009 2010e 2011e
SCOR Sw iss Re Hannov er Re Munich Re

Finally, we show the detailed data for each reinsurer in the next four tables.
Table 21: SCOR - Earnings summary

€m
2004 2005 2006 2007 2008 2009 2010e 2011e
Gross earned premium 2,728 2,436 2,837 4,739 5,759 6,346 6,452 6,881
ow Non life 1,611 1,394 1,650 2,302 3,068 3,229 3,413 3,583
ow Life 1,117 1,042 1,188 2,437 2,691 3,117 3,039 3,297
Operating profit 199 242 408 577 348 372 488 697
ow Non life 153 160 331 410 202 213 323 510
ow Life 46 83 78 167 146 161 165 188
30
(44-20) 7325-9175 02 September 2010
Table 22: Swiss Re - Earnings summary

CHFm
2004 2005 2006 2007 2008 2009 2010e 2011e
Net earned premium 29,439 27,779 29,515 31,664 25,501 24,606 20,278 19,876
ow Non life 18,336 16,359 18,541 18,977 14,379 13,885 11,528 11,528
ow Life 10,205 10,512 10,974 12,665 11,090 10,679 8,744 8,307
Operating profit 3,367 1,973 5,856 5,187 (1,350) 954 2,640 3,158
ow Non life 2,727 1,025 5,613 4,471 2,746 3,820 1,817 2,681
ow Life 1,304 1,643 1,546 1,320 697 746 805 757
Table 23: Hannover Re - Earnings summary

€m
€m 2004 2005 2006 2007 2008 2009 2010e 2011e
ow Non life 3,456 3,922 3,914 4,498 4,277 5,230 5,399 5,507
ow Life 1,956 2,258 2,373 2,795 2,785 4,079 4,202 4,286
Operating profit 539 92 820 928 148 1,140 992 1,063
ow Non life 463 (28) 670 657 2 731 692 762
ow Life 77 93 140 230 121 372 277 278
Table 24: Munich Re - Earnings summary

€m
€m 2004 2005 2006 2007 2008 2009 2010e 2011e
ow Non life 14,181 13,565 13,795 13,507 13,448 14,096 13,917 14,334
ow Life 7,294 7,396 7,276 7,024 6,775 9,281 7,602 8,552
Operating profit 3,025 4,142 5,494 5,078 3,262 4,721 3,825 4,290
ow Non life 1,944 1,284 3,486 3,068 2,822 2,989 2,277 2,242
ow Life 698 1,105 922 1,091 934 1,175 1,077 1,334
Comparison of the European life reinsurers

Reinsurance life operating profit margin
We have shown in the table below the operating profit margin of the European
reinsurance companies for the life segment. We note that the larger reinsurers post
higher operating profit margins.
31
(44-20) 7325-9175 02 September 2010
Table 25: European reinsurers life operating margin

LCm, %
NEP 2004 2005 2006 2007 2008 2009
Swiss Re 10,205 9,638 10,974 12,665 11,090 10,679
Munich Re 7,294 7,396 7,276 7,024 6,775 9,281
Hannover Re 1,956 2,258 2,373 2,795 2,785 4,079
SCOR 1,080 1,010 1,121 2,191 2,429 2,779
Operating profit
Swiss Re (ex gains) 1,222 1,333 1,381 1,320 926 1,291
Munich Re 698 1,105 922 1,091 934 1,175
Hannover Re 77 93 140 230 121 372
SCOR 46 83 78 167 146 161
Operating profit % NEP

Swiss Re (ex gains) 12.0% 13.8% 12.6% 10.4% 8.3% 12.1%
Munich Re 9.6% 14.9% 12.7% 15.5% 13.8% 12.7%
Hannover Re 3.9% 4.1% 5.9% 8.2% 4.3% 9.1%
SCOR 4.3% 8.2% 6.9% 7.4% 6.0% 5.8%
Source: Company reports, Swiss Re operating profit is excluding gains. Note Munich Re includes life and health re. Note for Swiss Re
2007 onward is the new style accounting where life re gets just credited with the risk free return on its invested assets. 2007, the only
year of old and new reporting, had SF2744m life operating profit old style and SF1320m new style.
Reinsurance investment income yield in life

We note that the lowest investment yield was in the year 2008 for the life reinsurance
segment for the European reinsurers. For Swiss Re, annualised investment yield on
our estimates is declining from the 2007 level of 5.2%, reflecting the constant decline
in risk free rates since. However, there is accounting noise in the Swiss Re numbers,
which we cannot exclude with reported data, as they do include CVA/DVA (CVA is
credit valuation adjustment, DVA is debit valuation adjustment) valuation
adjustments which reflect the CDS spreads of Swiss Re and its cedants rather than
real investment cash flows. We believe this is the main reason why 2Q10 appears
strongly positive.
Table 26: Reinsurance investment income yield

LCm, %, Swiss Re CHFm during 2004-09 and 1Q10 and 2Q10 (translated back to SF at 0.95)
Investment income 2004 2005 2006 2007 2008 2009 1Q 10 2Q 10
Swiss Re 3,400 2,403 2,133 586 737
Munich Re 1,431 1,897 1,617 1,532 1,252 1,570 274 253
Hannover Re 222 275 305 294 246 520 101 128
SCOR 138 175 190 330 259 305 83 95
RGA 581 639 780 908 871 1,123 304 292
Life investments
Swiss Re 53,519 62,234 59,519 65,383 52,619 53,073 54,377 50,818
Munich Re 34,030 36,800 33,710 26,108 25,386 23,008 20,591 21,491
Hannover Re 7,600 8,857 10,909 11,351 12,440 15,886 16,728 18,176
SCOR 7,035 7,423 8,474 9,084 9,694
RGA 10,564 12,331 14,613 16,398 15,611 19,224 20,560 21,049
Life inv yield

Swiss Re 5.2% 4.6% 4.0% 4.3% 5.8%
Munich Re 4.2% 5.2% 4.8% 5.9% 4.9% 6.8% 5.3% 4.7%
Hannover Re 2.9% 3.1% 2.8% 2.6% 2.0% 3.3% 2.4% 2.8%
SCOR 4.7% 3.5% 3.6% 3.7% 3.9%
RGA 5.5% 5.2% 5.3% 5.5% 5.6% 5.8% 5.9% 5.5%
Source: Company reports. Note that the accounting split for Munich Re from 1Q10 is just life re, prior was life and health re together. Note fro 2007 onward Swiss Re only credits risk free to its
operating units. Investment income as we define it for Swiss Re consists of investment income and net gains on non-participating contracts. However, even this methodology has a weakness,
which is that this investment income does include adjustments for Credit Valuation Adjustments and Debit Valuation Adjustments, which mainly reflects the change of CDS costs for Swiss Re and
for its cedants where there is Modco business, which is pure accounting and does not reflect a cash cost. With reported data it is not possible to exclude these CVA/DVA adjustments.
32
(44-20) 7325-9175 02 September 2010
MCEV (Market Consistent Embedded Value) roll-forward comparison

Munich Re contributes maximum as by value of new business with 12.1% of the
beginning MCEV driven by the new business. Munich Re 2009 MCEV has also
benefited by the economic variances by 15% of the beginning MCEV against 10.6%
for Hannover Re.
Table 27: MCEV rollforward comparison of the 3 reinsurers

€m, %
2009 2009 2009 2009 2009 2009 2009 2009 2009
€m €m €m % % % Rank Rank Rank
MCEV roll forward comparison Munich Hannover SCOR Munich Hannover SCOR Munich Hannover SCOR
Opening MCEV 4,657 1,652 1,702 100.0% 100.0% 100.0% 1 1 1
Opening adjustments 306 (25) 0 6.6% -1.5% 0.0% 1 3 2
Adjusted opening MCEV 4,963 1,627 1,702 106.6% 98.5% 100.0% 1 3 2
Value of new business 562 84 113 12.1% 5.1% 6.7% 1 3 2
Expected return at reference rate 196 77 77 4.2% 4.6% 4.5% 3 1 2
Expected return in excess of reference rate 11 0.0% 0.7% 0.0% 2 1 2
Transfers from VIF and required capital to free
surplus 0.0% 0.0% 0.0% 1 1 1
Experience variances 145 (5) 19 3.1% -0.3% 1.1% 1 3 2
Assumption changes 113 41 (19) 2.4% 2.5% -1.1% 2 1 3
Other operating variance (126) (29) -2.7% -1.8% 0.0% 3 2 1
Operating MCEV earnings 891 179 191 19.1% 10.8% 11.2% 1 3 2
Economic variances 712 174 109 15.3% 10.6% 6.4% 1 2 3
Other non operating variance 13 0.3% 0.0% 0.0% 1 2 2
Total MCEV earnings 1,616 353 299 34.7% 21.4% 17.6% 1 2 3
Closing adjustments 194 231 (66) 4.2% 14.0% -3.9% 2 1 3
Closing MCEV 6,773 2,211 1,934 145.4% 133.8% 113.7% 1 2 3
Sensitivity
We have shown below the comparison of the 4 reinsurers on the sensitivity of the
various assumptions. We also note that the metrics are not exactly comparable since
the methodology used by each company is different. We note that Swiss Re EVM is
the most sensitive to the drop in interest rates by 1%. The reason was actually the
short duration at the end of 2009 and is less linked to the underlying life reinsurance
business.
We also note that Munich Re, SCOR & Swiss Re do not adjust the EV for the
liquidity premium. And finally we note that the largest exposure to mortality, in
terms of sensitivity to the base case embedded value, is Hannover Re at 35% vs 14-
20% for its peers.
33
(44-20) 7325-9175 02 September 2010
Table 28: Reinsurance sensitivity to assumptions

%
Year 2009 2009 2009 2009
Method MCEV EEV MCEV EVM
Sensitivity Munich Re SCOR Hannover Re Swiss Re
Interest rates & assets
Interest rates –100bp 3.9% 1.1% (3.4%) (3.5%)
Discount rate -100 bps NA 6.1% NA 4.2%
Equity/property values –10% 0.0% (0.7%) NA (3.1%)
Equity/property implied volatilities +25% 0.0% NA 0.0% NA
Expenses and persistency
Maintenance expenses –10% 1.1% 1.3% 2.0% NA
Lapse rates –10% (0.8%) 1.5% 15.7% 4.5%
Insurance risk
Mortality/morbidity (life business) –5% 19.5% 13.9% 35.0% 13.6%
Mortality (annuity business) –5% (0.1%) 0.0% (1.8%) (1.0%)
Liquidity premium None NA None None
Source: Company reports & J.P.Morgan estimates
MCEV by components: SCOR has the highest free surplus

We note that SCOR has the highest % of MCEV as the free surplus and Munich re
has the lowest free surplus at 8.3%. We have shown the detailed break up of the
segments in the table below.
Table 29: MCEV break up by components

€m, %
MCEV by components, €m Munich Re SCOR Hannover Re
Adjusted net worth 2,835 907 2,093
Free surplus 564 382 794
Required capital 2,271 526 1,299
VIF 3,938 1,027 2,461
PVFP (without options) 2,277 1,507 1,879
CoRNHR 1,252 421 490
FOGs 32 10 7
FCoRC 377 49 84
MCEV 6,773 1,934 4,554
MCEV by components, %
Adjusted net worth 41.9% 46.9% 46.0%
Free surplus 8.3% 19.7% 17.4%
Required capital 33.5% 27.2% 28.5%
VIF 58.1% 53.1% 54.0%
PVFP (without options) 33.6% 77.9% 41.3%
CoRNHR 18.5% 21.8% 10.8%
FOGs 0.5% 0.5% 0.2%
FCoRC 5.6% 2.5% 1.9%
MCEV 100.0% 100.0% 100.0%
Life reinsurers vs. life primary insurers – we believe the life

reinsurance have more attractive risk reward, particularly at
current low levels of interest rates
We believe that life reinsurance has a more attractive risk reward than life primary
insurance, particularly at the current low levels of interest rates. To support our view
we make two points. The first is our translation below of a 18 August interview of
the Munich Re CEO in the FTD, the second is a comparison of the embedded value
sensitivities of life primary insurance and life reinsurers, where the reinsurers have
sensitivity mainly to mortality and lapse, the primary insurers to interest rates.
34
(44-20) 7325-9175 02 September 2010
Interview with Munich CEO on 18 August in FTD

The following is our translation of an interview of Munich Re's CEO, Dr. Nikolaus
von Bomhard in the FTD on 18 August, and we have checked the reasonableness of
our translation with the company. We cite this interview because it supports our view
that life primary insurance is less attractive in terms of risk reward than life
reinsurance.
By comparison with life reinsurance, the economics in life primary are in our view
less favourable. This is in our view because life reinsurance is more sensitive to
mortality and biometric risks, which benefit from generally favourable trends over
time, while life primary is more sensitive to interest rate risk and investments, a
challenge in the current low interest rate environment.
Munich Re is not particularly concerned with the loss of market share in primary life
in Germany. Munich Re set fundamental concerns on the profitability of German life.
Seen economically this area is not very profitable. The business model relies on
paying long term guaranteed interest rates to clients, which are hard to earn in the
current period of low interest rates. Plus, the investment risk is with the insurer -
and at least 90% of investment return is to the policyholder.
Munich Re, by contrast, is very satisfied with ERGO's primary non-life which has
above average results. As for the life part of ERGO, part has been closed to new
business for now (Victoria) as it could not take sufficient investment risk to deliver
expected policyholder returns. Munich Re is also very optimistic in reinsurance
despite a shrinking topline, with strong regional growth in Asia, CEE and LatAm,
and also potential growth in life re.
The following table highlights the differences between reinsurers and primary
insurers in terms of embedded value sensitivities. The table is striking in our view in
three respects:
1. Asset risk seems mainly with the primary insurers. The interest sensitivity of the
reinsurers is on average near zero (this is because it is the average of positive
sensitivities for SCOR and Munich, and low negative sensitivities for Hannover
and Swiss) whereas on average for the primary insurers it is 15%. The main
reason for this sizeable difference we believe is that the primary insurers provide
minimum return guarantees to their policyholders (3.3% on average on the back
book in Germany, 2.25% for new business), and the reinsurers, except for some
small variable annuity shares, provide none.
2. By contrast reinsurers seem to have all the mortality risk. We believe this reflects
the reality that reinsurers do focus on mortality as their main offering. And typically
reinsurers price their mortality by giving away one third to one half the expected
trend improvement in mortality to the primary insurers. So for reinsurers we believe
the two main risks on mortality are: a) that the trend improvement slows or stops; b)
that there is a pandemic and that mortality for a year is not the 5% worse than the
sensitivity calculates, but 10% instead. This is the reason that Swiss Re, for
example, has hedged against pandemics via securitisations.
3. Also reinsurers have higher sensitivity to lapse risk. This reflects the business
model of reinsurers, which advance a high upfront commission to the primary
insurers, to help them pay for their marketing costs. And the reinsurers then try to
estimate, using lapse and mortality assumptions, that they will get sufficient
35
(44-20) 7325-9175 02 September 2010
premiums back to pay for the upfront commission. So, although the risks are
lapses and mortality, the business model is closely tied to making loans to
primary insurers, in a way which relieves their liquidity and solvency strain from
selling large new business volumes with large front end commissions to sales.
Table 30: Average sensitivities for the reinsurers and primary insurers (2009)
%
Sensitivity Primary Insurers Reinsurers
Interest rates –100bp (14.6%) (0.5%)
Discount rate -100 bps 5.2%
Equity/property values –10% (4.0%) (1.3%)
Equity/property implied volatilities +25% (1.9%)
Maintenance expenses –10% 2.7% 1.5%
Lapse rates –10% 1.8% 5.2%
Insurance risk
Mortality/morbidity (life business) –5% 1.6% 20.5%
Mortality (annuity business) –5% (0.9%) (0.7%)
Source: Company reports. Reinsurers is the average for SCOR, Hannover, Munich and Swiss. Primary insurers is the average for
Allianz, AXA, Generali and Munich Re primary life.
The following table shows the data in greater detail. We highlight three points:
1. The high interest rate sensitivity in life primary is due mainly to Allianz and
Munich Re primary life. This is because they are both significantly exposed to
German life, which currently has relatively high average guaranteed rates on old
business in Europe, at 3.3% on the backbook. Generali is around half their level,
and we believe this is mainly because Generali is less exposed to Germany, and
because in Italy the average guaranteed rate is 2%, and this is reduced to 1% after
the contractual deduction of 1.2% management fees.
2. Generali has the highest sensitivity to equity and property values at 6%, followed
by AXA and Allianz at 4%. Munich Re’s primary sensitivity is relatively modest
at 2%. We believe this simply reflects the more conservative investment strategy
of Munich, and the more aggressive strategy of Generali. In any case, the
numbers are all below 10%, implying that shareholders' net exposure to property
and equity values is less than one times life embedded value.
3. Hannover has a relatively high sensitivity to lapse and mortality risk compared
with its reinsurance peers. We believe this reflects its business mix, which we
believe is structured to use lapse risk and mortality risk to help provide front end
commission funding for primary insurers. A significant proportion of this is
securitised out using securitisation vehicles (L7 in 1Q09 securitised out €100m of
life embedded value, source Hannover website, L6 1Q06 also securitized out
€100m life EV)).
36
(44-20) 7325-9175 02 September 2010
Table 31: Re/Insurance sensitivity to assumptions

%
Year 2009 2009 2009 2009 2009 2009 2009 2009
Method MCEV EEV MCEV EVM MCEV EEV EV MCEV
Sensitivity Munich Re SCOR Hannover Re Swiss Re Allianz AXA Generali Munich Re primary
Interest rates –100bp 3.9% 1.1% (3.4%) (3.5%) (19.0%) (6.0%) (10.4%) (22.9%)
Discount rate -100 bps NA 6.1% NA 4.2% NA NA NA NA
Equity/property values –10% (0.0%) (0.7%) NA (3.1%) (4.0%) (4.0%) (6.2%) (1.8%)
Equity/property implied volatilities +25% (0.0%) NA (0.0%) NA NA (2.0%) (2.5%) (1.1%)
Maintenance expenses –10% 1.1% 1.3% 2.0% NA 3.0% 4.0% 2.4% 1.5%
Lapse rates –10% (0.8%) 1.5% 15.7% 4.5% 1.0% 3.0% 2.4% 0.6%
Insurance risk
Mortality/morbidity (life business) –5% 19.5% 13.9% 35.0% 13.6% 1.0% 2.0% 2.1% 1.1%
Mortality (annuity business) –5% (0.1%) 0.0% (1.8%) (1.0%) (1.0%) (1.0%) (0.7%) (0.9%)
Liquidity premium None NA None None NA NA NA NA
Life reinsurance products

We have summarized the characteristics of the 4 typical Life reinsurance products
and the risks associated with each product. Critical Illness is having the longest term
generally close to 25 years. Term life and Credit life does not have much interest rate
risk since they have low premium reserve.
Table 32: Life reinsurance products characteristics

as stated
Characteristics Term life Credit Life Financing Critical Illness
Biometric risk Constant sum upon death Decreasing sum upon death Typically contains Lump sum on industry standards
Under reassurance treaty (first year Under reassurance treaty (No first year Rate covered under reassurance
Lapse Risk financing) financing) Typically contains treaty
Interest rate risk None, due to low premium reserves None, due to low premium reserves
Major region US Europe
Type Individual Individual Individual usually
Premiums Stable Long term predictable Long term
Approximate term 10 years 20 years 10 years 25 years
Term life
Term life is a good and steady contributor of the premium and profits as can be seen
in the chart below. This is a typical US product sold to individuals. The negative cash
flow in the beginning year results from statutory reserve strain and/or a premium
commission paid to cedants to help finance the production. This is also a strong EV
generator since most of the EV is recognized in the first year.
37
(44-20) 7325-9175 02 September 2010
Figure 8: Term life: cash flow, IFRS profit & EV profit

10.0
8.0
6.0
4.0
2.0
0.0
-2.0
-4.0
-6.0
-8.0
1 2 3 4 5 6 7 8 9 10
Distributable cashflow IFRS profits IFRS premiums
EV profits Risk based capital
Credit life
This is a typical European product sold mainly to individuals for mortgage
protection. Also the cash flows and the IFRS profits are identical for these products
and are also very much predictable. This is also a strong EV generator. In addition, as
can be seen from the figure below the risk based capital is released quickly over time.
This product typically does not have negative distributable cash flow at the
beginning.
Figure 9: Credit life: cash flow, IFRS profit & EV profit

10.0
8.0
6.0
4.0
2.0
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
IFRS profits & distributable cash flow IFRS premiums

Financing reinsurance
In the financing reinsurance, like for other lines of business, the IFRS profit is spread
evenly over the total contract period. This combines traditional life reinsurance and
financial components. This usually contains biometric, mortality and lapse risks.
Also, all the life and health insurance can be combined with reinsurance.
38
(44-20) 7325-9175 02 September 2010
Figure 10: Financing reinsurance: cash flow, IFRS profit & EV profit
10.0
8.0
6.0
4.0
2.0
0.0
-2.0
-4.0
-6.0
-8.0
1 2 3 4 5 6 7 8 9 10
EV profits Risk based capital Carried-forw ard loss
Critical Illness
This is a good new business value generator creating EV profits. This product is also
often very long term as can be seen in the chart below. Hence this generates steady
premiums over the period. This is sold mainly to the individuals with life cover and
to protect mortgages.
Figure 11: Critical Illness: cash flow, IFRS profit & EV profit
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
0.0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
39
(44-20) 7325-9175 02 September 2010
RGA read across summary

We have spent some time analyzing the track record, risks and business model of
RGA (Reinsurance Group of America, covered by J.P.Morgan US Life Insurance
analyst, Jimmy Bhullar). The reason is that RGA is the only listed pure life reinsurer
in the world, it has very granular and transparent accounting and reporting, and we
believe its reporting, within the limitations of US GAAP, are a standard for the
sector.
The only key weakness in terms of reporting is that RGA does not publish its
tracking of embedded value, which it does, however, prepare internally. The reason
for so far non disclosure is that the US capital market so far has paid little or no
attention to embedded value for the primary life insurers, so there is no peer group
nor general understanding. However, we believe following our conversations with
RGA that it also (like its European peers) regards embedded value as the better
metric to track value creation.
Our analysis of RGA focuses on three areas:
1. The November 2008 $347m capital increase and the reasons for it. We note that
although RGA's business is primarily focused on mortality, that its prospectus
and SEC filings at the time stressed capital market type risks above other
operating risks. We believe this was partly a reflection of the capital markets
environment at the time, which was reaching new highs in terms of volatility. But
we believe it also highlights the issue that, unless a life reinsurer does just pure
mortality business such as YRT (effectively a pay as you go arrangement, with
zero investment risk), then investments are an issue and the volatility of
investment returns can potentially as in crisis years like 2008 overwhelm the
underlying operating profits. This means that, in the case of RGA, investments
have an asymmetric effect. They have the potential to be associated with
conditions where the company decides to make a capital increase, as was the case
in November 2008, whereas in more normal conditions assuming credit risk only
has the potential to raise ROE for the group from a 12-13% run rate to a 12-14%
level.
2. Profitability track record and US GAAP accounting. Under US GAAP (IFRS is
very similar) life re liabilities are valued on the basis of the original pricing
assumptions and are only ever reset if the life re division as a whole should fall
into loss. We believe this is the reason that, on the 2Q10 the conference call
RGA stated that they expect 2010 to be the bottoming out year in terms of
profitability, when the weak pricing years of up to 2004 have their maximum
impact. We note that in non-life should pricing for any individual risk be
understated then an adjustment is made immediately, as the accounting is best
estimate, not as life re effectively is, amortised.
3. Business model. We note that RGA is focused mainly on mortality and that it
prides itself on gaining market share not so much through pricing, but more in
terms of consistency and quality of service. We note also that Canada is
significantly more profitable than the US, and we believe this reflects the fact that
it is a more concentrated market.
40
(44-20) 7325-9175 02 September 2010
November 2008 capital issue

4 November 2008 RGA raised $346.9m gross by issuing 10.235million new shares at
$34.69 each, raising $331.9m net of $15m fees. Partly thanks to this on the 2Q09
conference call the company said it had over $0.4bn excess capital available for
deals, and that number at 2Q10 was over $0.5bn.
Main conclusions are:

1. For RGA reinvestment risk and investment income in general does not drive
profits and it uses only modest credit risk to help enhance profitability (ie 5.5%
returns). In the mortality risk business, especially YRT, investment income has
virtually no impact. RGA’s combined ratio is less than 100% on an expected basis
so it is really mortality that drives profitability. Investment income and reinvestment
yields affect RGA around the margins slightly. For example, because of the low
investment yields right now, RGA would expect to achieve an ROE of 12 to 13%
instead of 14-15%.
2. Mortality takes 4-5 years to build up; we believe the reason 2010 is cited as
the low point in US profitability by RGA is that this is when the underpriced period
2000-4 has the maximum weight in the portfolio as a whole.
3. The business is very finely analysed in the segments US traditional (mainly

mortality), financial and asset based, and even this does not seem enough as on the
2Q10 conference call there were very many questions about variable annuity vs
traditional fixed annuity earnings contribution, and RGA said that some break out of
VA profitability may be good. International and Asia are a little more homogenous.
International is mainly UK, Asia is mainly Australia.
4. The run rate of ROE is below 13% for now due to high mortality and dip in
reinvestment rate, trend is 14%. Last ROE low was 11% in 2005 when mortality was
high.
5. Swiss Re was very aggressive in the 97-03 period, now much less so (eg closed
Taiwan office). The most aggressive is Hannover.
6. Canada is highly concentrated high margin country, the US more fragmented;

profitability is adequate.
7. Volatility in US mortality has been in the $1-3m bracket, and has been above
expectations. The $250,000-$1,000,000 group is very stable.
8. Expectations of strong growth post 2008 capital markets crisis have not so far
materialised. So far very few M&A deals have occurred, which normally would tend
to lead to spikes in demand for reinsurance, and session rates are falling as primary
insurers have rebuilt their balance sheets and core growth has slowed.
9. The core mortality has baked-in premium growth for stable inforce as the
portfolios age and the premiums rise at each anniversary. The so called permanent
business (ie mortality is part of the contract and lapses are low) is higher margin and
sticky; pure term is more lapse prone.
41
(44-20) 7325-9175 02 September 2010
10. Mortality contracts run off over around 30 years. This means that the risk of
large claims can occur at any time for decades after the contract signing.
11. Underpricing in 2000-03 has the effect of cutting profitability to below 13%
ROE from a 14% run rate, and we estimate the relatively underpriced portfolio
accounts for around 20% of the total US book of business (assuming 30 year
durations); we estimate this means that it was written on ROEs of 5% below the run
rate, ie around 7-8%. For more highly capitalised reinsurers like Swiss Re we
estimate this means around 5% ROE run rate on that portfolio.
12. Mortality in US seems to have spiked in 2008-1Q10 vs very light in 2006-7.
13. RGA has around $500m capital it has not deployed. This is partly from the
4Q08 capital increase.
Risks as per RGA’s SEC filings around its 4 November 2008 capital rising. We
note that the listing of the risks would likely have been influenced by the capital
markets environment then, and more priority given to investment risks as a
result. The text in italics below is taken from the risk section of the 31/10/08
filing with the SEC.
1. Need for liquidity – this is a key reason given for the capital raising.
2. Volatile capital markets. Results of operations are materially affected by

conditions in the global capital markets and the economy generally.
a) These events and the continuing market upheavals may have an adverse
effect on us, in part because we have a large investment portfolio and are
also dependent upon customer behavior. Revenues may decline in such
circumstances and our profit margins may erode.
b) In addition, in the event of extreme prolonged market events, such as the

global credit crisis, we could incur significant losses.
c) Even in the absence of a market downturn we are exposed to substantial

risk of loss due to market volatility.
3. Collateral: some of our transactions with financial and other institutions specify
the circumstances under which the parties are required to post collateral. This may
increase under certain circumstances which could adversely affect our liquidity. In
addition we may be required to make payments to our counterparties related to any
decline in the market value of the specified assets.
4. Defaults on our mortgage loans and volatility in performance may adversely

affect our profitability. Mortgage loans are stated on our balance sheet at unpaid
principal balance adjusted for any unamortised premium or discount and are net of
valuation allocations. Such valuation allowances are based on the excess carrying
value of the loan over the present value of expected future cash flows discounted at
the loans original effective interest rate.
5. Accounting: Our principal investments are carried as follows.
42
(44-20) 7325-9175 02 September 2010
Fixed maturities and equities are classified as available for sale and reported at
their estimated fair value. Unrealised gains or loses are recorded as OCI.
Short term investments with remaining maturities of one year or less but greater than
three months are stated at amoritzed cost which approximates fair value.
Mortgage and policy loans are restated at unpaid principal balance, net of valuation
allowances.
Funds withheld at interest represent amounts contractually withheld by ceding

companies in accordance with reinsurance agreements. The value of the assets
withheld and interest income are recorded in accordance with specific treaty terms.
We use the cost method of accounting for investments in real estate joint ventures.
Investments not carried at fair value, principally mortgage loans, policy loans, real
estate joint ventures and other limited partnerships may have fair values which are
substantially different from the carrying value in our consolidated financial
statements. Each of such asset classes is regularly evaluated for impairments.
6. Valuation. Our valuation may include methodologies which are subject to

differing interpretations and could result in changes to investment valuations that
may materially adversely affect our results of operations. This is relation to level 1,
level 2 and level 3 as defined in SFAS 157.
Level 1 is quoted prices in active market for identical assets.
Level 2 are observable inputs such as quoted prices for similar assets, prices in
markets that are not active, or valuation methodologies with significant inputs that
are observable or can be corroborated from observable market data. This includes
primarily US and foreign corporate securities, Canadian provincial government
securities, RMBS and CMBS. We value most of these securities using inputs that are
market observable.
Level 3 are unobservable inputs that are supported by little or no market activity and
that are significant to the fair value of the related assets or liabilities. These
unobservable inputs can be based in large part on management judgment or
estimation and cannot be supported by reference to market activity. Even though
unobservable, management believes these inputs are based on assumptions deemed
appropriate given the circumstances and consistent with what other market
participants would use when pricing similar assets and liabilities. For our invested
assets, this category generally includes U.S. and foreign corporate securities
(primarily private placements), asset-backed securities (including those with
exposure to sub prime mortgages), and to a lesser extent, certain residential and
commercial mortgage-backed securities, among others. Additionally, our embedded
derivatives, all of which are associated with reinsurance treaties, are classified in
Level 3 since their values include significant unobservable inputs associated with
actuarial assumptions regarding policyholder behavior. Embedded derivatives are
reported with the host instruments on the condensed consolidated balance sheet.
Risks from October 2008 RGA prospectus

The following is a list of risks we extracted from the 29 October 2008 prospectus for
the $349m equity increase. We note that this was mainly on investment risks,
43
(44-20) 7325-9175 02 September 2010
relatively little on insurance and mortality. We believe, as noted above, that part of
the reason for the focus on investment risk is the environment in 4Q08, when capital
markets volatility rose to new peaks.
1. Adverse capital and credit market conditions may significantly affect our ability to
meet liquidity needs, access to capital and cost of capital.
2. Difficult conditions in the global capital markets and the economy generally may
materially adversely affect our business and results of operations and we do not
expect these conditions to improve in the near future.
3. There can be no assurance that actions of the U.S. Government, Federal Reserve
and other governmental and regulatory bodies for the purpose of stabilizing the
financial markets will achieve the intended effect.
4. The impairment of other financial institutions could adversely affect us.
5. Our requirements to post collateral or make payments related to declines in

market value of specified assets may adversely affect our liquidity and expose us to
counterparty credit risk.
6. Defaults on our mortgage loans and volatility in performance may adversely affect
our profitability.
7. Our investments are reflected within the consolidated financial statements utilizing
different accounting basis and accordingly we may not have recognized differences,
which may be significant, between cost and fair value in our consolidated financial
statements.
8. Our valuation of fixed maturity and equity securities may include methodologies,
estimations and assumptions which are subject to differing interpretations and could
result in changes to investment valuations that may materially adversely affect our
results of operations or financial condition.
9. Some of our investments are relatively illiquid and are in asset classes that have
been experiencing significant market valuation fluctuations.
10. The determination of the amount of allowances and impairments taken on our
investments is highly subjective and could materially impact our results of operations
or financial position.
11. Gross unrealized losses may be realized or result in future impairments.
12. Defaults, downgrades or other events impairing the value of our fixed maturity
securities portfolio may reduce our earnings.
13. The class A common stock is subject to a conversion proposal at a special

meeting of our shareholders on November 25, 2008.
Numerous important factors could cause our actual results and events to differ
materially from those expressed or implied by forward-looking statements including,
without limitation:
44
(44-20) 7325-9175 02 September 2010
• adverse capital and credit market conditions and their impact on our liquidity,
access to capital and cost of capital;
• The impairment of other financial institutions and its effect on our business;
• Requirements to post collateral or make payments due to declines in market value
of assets subject to our collateral arrangements;
• The fact that the determination of allowances and impairments taken on our
investments is highly subjective;
• Adverse changes in mortality, morbidity, lapsation or claims experience;
• Changes in our financial strength and credit ratings and the effect of such
changes on our future results of operations and financial condition;
• Inadequate risk analysis and underwriting;
• General economic conditions or a prolonged economic downturn affecting the
demand for insurance and reinsurance in our current and planned markets;
• The availability and cost of collateral necessary for regulatory reserves and
capital;
• Market or economic conditions that adversely affect the value of our investment
securities or result in the impairment of all or a portion of the value of certain of
our investment securities;
• Market or economic conditions that adversely affect our ability to make timely
sales of investment securities
• Risks inherent in our risk management and investment strategy, including
changes in investment portfolio yields due to interest rate or credit quality
changes;
• Fluctuations in U.S. or foreign currency exchange rates, interest rates, or
securities and real estate markets;
• Adverse litigation or arbitration results;
• The adequacy of reserves, resources and accurate information relating to
settlements, awards and terminated and discontinued lines of business;
• The stability of and actions by governments and economies in the markets in
which we operate;
• Competitive factors and competitors' responses to our initiatives;
• The success of our clients;
• Successful execution of our entry into new markets;
• Successful development and introduction of new products and distribution
opportunities;
• Our ability to successfully integrate and operate reinsurance businesses that
RGA acquires;
• Regulatory action that may be taken by state Departments of Insurance with
respect to RGA, or any of its subsidiaries;
45
(44-20) 7325-9175 02 September 2010
• Our dependence on third parties, including those insurance companies and

reinsurers to which we cede some reinsurance, third-party investment managers
and others;
• The threat of natural disasters, catastrophes, terrorist attacks, epidemics or
pandemics anywhere in the world where we or our clients do business;
• Changes in laws, regulations, and accounting standards applicable to RGA, its
subsidiaries, or its business;
• The effect of our status as an insurance holding company and regulatory
restrictions on our ability to pay principal of and interest on its debt obligations
US and Canada has higher profit margins

We have shown in the chart below the comparison of the profit margin (calculated as
operating income % NEP). We note that the US and Canada are having higher
margins historically when compared to the Europe and Asia Pacific. Also, the profit
margin of Canada has increased over the period 2003 to 2008 and overtook the US in
2009.
Figure 12: Operating income margin by regions for RGA

%
25.0%
20.1%
20.0%
15.4%
13.7% 13.5% 14.3%
15.0% 12.0% 12.9%
12.2% 12.5% 12.1%
11.2% 10.8%
10.5%
10.0%
9.5% 8.8%
8.4% 8.8% 8.5%
7.7% 6.5%
10.0% 6.3% 6.7% 5.7% 5.8% 6.5% 6.7% 7.2% 7.3% 7.5%
7.1% 7.0%
5.1% 5.9%
4.2%
5.0% 1.8%
0.0%
2003 2004 2005 2006 2007 2008 2009 2010e
Total US Canada Europe and SA Asia pac
Source: Company reports & J.P.Morgan estimates, Segment margins are calculated on the pre tax operating income; total margin is
calculated on the total operating income.
Premium growth is slowing down

The life reinsurance premiums growth has reduced from 27% in 2004 to 7% in 2009.
Among the drop the sharpest drop was in Asia Pacific where the growth has reduced
from 54% to 0% during the same period. We also note that the growth has increased
in 2009 for Canada and Europe/SA. J.P.Morgan estimates (this is drawn from the
research of our US colleague Jimmy Bhullar) 15% growth in the premiums in 2010e.
Table 33: Net premium growth is slowing down for the past 6 years
%
Premium growth 2004 2005 2006 2007 2008 2009 2010e
Total 27% 16% 12% 13% 9% 7% 15%
US 23% 10% 9% 8% 8% 7%
Canada 18% 35% 25% 13% 10% 15%
Europe and SA 32% 15% 6% 15% 4% 10%
Asia pac 54% 34% 26% 29% 16% 0%
46
(44-20) 7325-9175 02 September 2010
Capitalisation of reinsurers
Capitalisation of RGA from Sep 2008
RGA has increased its total capital by 47% from Sep 08 and this is mainly driven by
the rise in equity as can be seen in the table below. We also note that the current
excess capital is over $0.5bn (source: 2Q10 transcript).
Table 34: RGA capitalisation

$m, %
Sep 08 2Q 10 % change
Short term debt 95 -
Long term debt 923 1,216 31.8%
Prefs and collateralized 1,009 1,009 0.0%
Equity 2,607 4,443 70.4%
Total capital 4,539 6,668 46.9%
Key takeaways from the earnings of RGA pre transcript of

RGA conference calls 3Q09-2Q10
These are relevant texts parts from the transcripts filed with Bloomberg
2Q10 Results
From President & CEO Greig Woodring (in Italics):
Mortality in the US was slightly better than expected with strong mortality in Canada
and the UK. ROE was 13% annualised. Capital redundancy exceeded $500m.
Competition is ratcheting up in the US where it's a little more competitive. Growth in
Canada was mostly in creditor premium and traditional has been more muted. VA
business performed OK in the quarter. The new yield is 5.5% down from 5.8% and
there is pressure on investing new money and that is reflected in new business
pricing.
Pricing in 97-03 was quite a bit off, pricing in late 80s early 90s is very profitable. In
primary life the companies are [JPMe: we believe this refers to RGA’s primary life
insurance clients] looking for growth at the moment. UK is 80% of the segment which
consists of Europe, Africa and India and has strong growth at the moment.
Interest rates matter very little for a typical YRT mortality type quote; it really is not
that much of an issue. For the asset intensive businesses there are no new
transactions in the last year and that may be reflects our more conservative exposure
on investment income among other things. So interest rates have more of an effect on
in force business than for new pricing.
RGA has more STOLI stranger originated life insurance. The whole market was not
priced correctly for older age issuers. And this is when the market was underpriced
and that period ran up to maybe 2005.
For the period of time since say 2003, 2004 and later pricing and expectations are
pretty close, because we haven't seen a lot of evolution yet. From say 1997 to 1998
or so to 2003 our expectations today might be a little bit higher than we priced at
that time. That was a very competitive period of time and we have reassessed. Prior
our expectations are considerably below what we were pricing at because of
mortality and claims flows.
47
(44-20) 7325-9175 02 September 2010
The policies that were written sometime ago we priced at a level X. Mortality has
improved considerably. Our expectations would be that it would be less than one
times X mortality now. Why haven't they lapsed? Well, a lot of those products
remember are permanent products, they are not looking at replacing one set of
mortality charges with another, they are basically looking at a whole life level
premium or a universal life type product. So the products stay in force for a long
time. We have a lot of business in force from eras that precede 2003.
We had a return in 2009 of 13% on that mortality markets business. We expected

that to slightly fall in 2010 and that is based on the relative prominence or effect of
those more pressured issue years in the overall result that this will be the bottoming
out year and overall when we look back over time we will expect decent returns
overall.
1Q10 results
US ROE 10% in 1Q10 vs 13% in 2009; this is expected to fall a bit in 2010 and
bottom out (low double digits).We have been expecting that returns would be lower
because of the most competitive years from 98-03 having a bigger impact and will
bottom out about this year. 2010 we earned 13% on this business. We expect to come
maybe a little south of in 2010 and then marginally increase each year going
forward. The seasonal component is had more deaths over age 80 than would
normally expect. Excess capital at 1Q10 for deployment is $500-600m but we would
never want that down to zero. US mortality was about 1% above expected which is
about $10m
In 98-03 RGA's natural market share was 12-14% and due to extra competitive
market was struggling to capture 5-10%. This is the long cycle of life re. 'We don’t
have like the property casualty industry big ups and downs in claims where you
swing from positive results to negative results. But you have periods of time where
you're more competitive and periods of time where you’re less competitive and
margins are a little bit wider. So we blend all those together in each year's
experience. The swing in competitive intensity was at the reinsurance level.
FY09 results
The adverse US claims experience was 15-20m in 4Q09. The variation almost always
comes about because of large claims. The mortality by claims amount, by policy
count is pretty rock steady. {JPMe:This is around 86% benefits ratio like in 2008 and
2009}. By contrast the 2006-7 level was a lot better. (And of course 1Q10 was high
mortality 1% ahead, and 2Q10 was inline, so maybe there is a correlation here with
the economy in our view). By comparison 4Q09 mortality in Asia was $5m better
than expected. The US business in total in 2009 was performing better than priced.
And we get the large claims from business written 20 years ago.
Australia has been a good market, 10% annual growth, and is the leading reinsurer.
Credit life in Canada is written by the banks and is extremely profitable for them.
[JPMe: The banks choose to reinsure some of this risk and that help them on their tax
48
(44-20) 7325-9175 02 September 2010
position. It’s not extremely profitable to RGA in the way it is to the banks that have a
very large margin on that business. But it is safer business albeit less margin.]
3Q09 Results
Mortality was extraordinarily good in 2006-7; mortality on large cases has worsened
in the past 6-7 years compared to where it was 7-8 years ago. (ie 2000-1 trough).
Asset intensive guidance was up from $10m at 3Q09, then $12m 4Q09, then $16m 1-
2Q10.
Net premiums grow faster than inforce because it is mostly YRT mortality and as the
policies cross their anniversaries the premiums go up (ie not level premium
business). Taiwan Swiss Re closed their office in 2009, the biggest competitor is now
GenRe.
Variable annuities: the operating profit we believe is based on the spreads earned on
the portfolio of assets. Based on current returns on RGA’s portfolio of investment
assets, which includes corporate bonds and structured products, this is very positive;
we believe this is why asset intensive business profits (which consist of variable
annuities and fixed annuities mainly) keep surprising on the upside. The delta in
terms of performance has been over $9m a quarter thanks to higher account values
and higher fees. Fixed annuities the spreads are coming in.
Canada session rates are expected over time to fall to US levels (post 2010). In the
US it is hard to raise share above the current 20% level.
International growth is around 12%, US below 8%. In 2008-9 mortality in the $1-3m
range has been worse than the $250,000 to $1m range. The pricing implemented
today gets included in a new primary product with a 6 month lag, and to see the
claims rise is a 3-5 year lag. South Africa mortality is more volatile due to more
violent deaths but overall good returns on capital. UK is more critical illness but
overall good returns including mortality.
Rise of mortality in 1Q 2010 leading to drop in the share price

We quote the comments from our US life insurance Jimmy S. Bhullar, CFA (+1-212-
622-6397; jimmy.s.bhullar@jpmorgan.com) on the 1Q 2010 and 2Q 2010 results
which explains the rise in mortality.
1Q 2010 results
“Reported $1.25 operating EPS vs $1.58 JPMe and consensus $1.59 due to $0.07
negative tax and adverse mortality in the US. On a reported basis 1Q10 ROE was
10% despite high claims. Pricing is expected to remain tight. Mortality has been
adverse in the first quarter in the US for the past two years. As recently written
business, becomes a greater proportion of the overall book, RGA's margins should
improve.”
2Q 2010 results
“2Q10 results affirm our bullish stance on RGA as we expect positive pricing
conditions in the US life re market, strong foreign growth and deployment of capital
49
(44-20) 7325-9175 02 September 2010
to enable RGA to outperform. Mortality results were inline with expectations which
should help concern about mispricing in the business. Pricing to remain firm drive
by limited capacity as aggressive underwriters have either raised prices (Swiss Re),
exited (Scottish Re, Annuity and Life Re) or been acquired (ING Re, Lincoln Re).
Also dislocation in the securitisation markets boosts the appeal of reinsurance.
Finally concentration of share should increase, this was 76% in 2009 for the top five,
up from 61% 2000, vs 94% for the top three in Canada 2009 up from 69% “
We also note the following from the RGA press release in 2010 1Q.
“Regarding our earnings flow, we experienced the same sort of U.S. mortality
seasonality this quarter as we have in the first quarter in each of the last several
years, a pattern we expect. The first-quarter reporting period typically presents the
unfavorable combination of higher claims flows with the lowest quarterly premium
flows. Our U.S. traditional business reported some degree of additional higher-than-
expected mortality, while claim levels in Canada were also somewhat higher-than-
expected. Despite this claims experience, we still generated a consolidated
annualized operating return on equity in excess of 10 percent for the quarter.
We continually update our assessment of mortality trends affecting our business, and
use our findings to refine pricing on new business and expected future premium and
claims flow for our entire reinsurance portfolio. Though still subject to significant
volatility, this ongoing modeling forms the basis for our longer-term expectations.
Our return on the U.S. traditional mortality business was 13 percent in 2009. This
return has been somewhat depressed in both 2008 and 2009, due in part to the
influence of more competitively priced business and its relative contribution to
income. Our projection models indicate returns on the U.S. traditional business will
likely remain in low double digits in 2010 before gradually increasing in the
following years. That pattern has been anticipated, and we continue to target an
enterprise-wide return on equity of 13 percent."
We can note the drop in the share price post 1Q 2010 results in the chart below.
Figure 13: RGA share price drop post 1Q 2010 results

Indexed
120.0
115.0 1Q 10 results
110.0
105.0
100.0
95.0
90.0
Dec 09 Jan 10 Feb 10 Mar 10 Apr 10 May 10 Jun 10 Jul 10
REINSURANCE GROUP OF AMERICA

Source: Bloomberg
50
(44-20) 7325-9175 02 September 2010
Prepared with the kind help of Business model: RGA example

Jimmy Bhullar, covering analyst
for RGA, + 212-622-6397, RGA have five main operational segments segregated primarily by geographic
jimmy.s.bhullar@jpmorgan.com region: United States, Canada, Europe and South Africa, Asia Pacific, and Corporate
and other.
1. United States operations provide traditional life reinsurance, reinsurance of asset-

intensive products and financial reinsurance, primarily to large U.S. life insurance
companies. The focus is very much on mortality risk.
2. Asset-intensive products include reinsurance of annuities and reinsurance of
corporate-owned life insurance.
3. Canada operations provide insurers with traditional individual life reinsurance as
well as creditor reinsurance, group life and health reinsurance and non-guaranteed
critical illness products.
4. Europe and South Africa operations provide primarily reinsurance of traditional
life products through yearly renewable terms and coinsurance agreements and the
reinsurance of critical illness coverage that provides a benefit in the event of the
diagnosis of a pre-defined critical illness.
5. Asia Pacific operations provide life, critical illness, disability income,
superannuation, and non-traditional reinsurance.
6. Superannuation is the Australian government mandated compulsory retirement
savings program. Superannuation funds accumulate retirement funds for
employees, and in addition, offer life and disability.
RGA is number three in US with a 20% market share, and in Canada market leader.
North American markets are mature and session rates are actually dropping. RGA is
leader in Asia, particularly Taiwan, Australia, Hong Kong. In Asia life insurance is
still a growth industry with low reinsurance penetration.
The following are citations from RGA’s transcript on Bloomberg of an investor

conference filed on 9 September 2009 on Bloomberg, presented by RGA’s CEO
(A. Greig Woodring) & CFO (Jack B. Lay) in Italics.
Operating EPS growth has been 14% CAGR five years. ROEs are in the 14%
category, in 2006, 7 and 8 and book value per share growth has been in the 13% -
14% range compounded since the beginning of RGA's history.
The facultative business is about 270,000 hard to underwrite cases, of which 100,000
US and 170,000 outside the US. This is a barrier to entry as it needs substantial
actuarial expertise.
Financial reinsurance is a fee business. International life re is a growth area for

example in Asia which rebounded first from the recession. RGA is continuing to win
The asset portfolio is primarily investment grade, corporate debt, with some real
estate exposure, primarily through CMBS. It has improved significantly during 2009
thanks to spreads coming in. To the extent there are any credit concerns they are well
diversified.
51
(44-20) 7325-9175 02 September 2010
Note that RGA run capital in the good but not great range relative to ratings, towards
the lower end of the acceptable capital levels for the rating agencies and will continue
to run the company in such a way that we will not threaten ratings. Of the capital raised
in October 2008 only around $20m was used up in block transactions. Some of the
demand was from companies looking to acquire parts of AIG, which did not happen,
also the AIG subsidiaries which were looking to position themselves for an IPO.
In terms of pricing two years ago the UK was overheated, but not any more. It's not a
great market, it's not that bad. In the US the pricing is fine and a little bit better than
2008. No change since the improvement after 2003. In the US the session rates built
up to 70% then in 2008-9 trended down to just below 40% again. 2009-10 should
trend back up
Credit business in Canada for RGA

In 2Q 2010 growth in Canada was mostly in creditor premium. In 4Q 09 Canada
creditor business had higher premiums, lower profit margins but more stable profit
margins and because there is a lot of small policies it helped the Canada business
overall. Experience in Canada has been running for a year over the last several
years and 2009 was no exception. In terms of investment took off a little credit risk in
4Q09 and had a bit more liquidity. Creditor insurance in Canada had around 5%
margin in 3Q 09.
Reinsurance competitive factors

Reinsurers operate in a highly competitive industry, which could limit our ability to
gain or maintain market share. The reinsurance industry is highly competitive, and
we encounter significant competition in all lines of business from other reinsurance
companies, as well as competition from other providers of financial services.
Major players
Competitors vary by geographic market. Primary competitors (for RGA) in the North
American life reinsurance market are currently the following, or their affiliates:
Transamerica Occidental Life Insurance Company, a subsidiary of Aegon, N.V.,
Swiss Re Life of America and Munich American Reinsurance Company.
Primary competitors in the international life reinsurance markets are Swiss Re Life
and Health Ltd., General Re, Munich Reinsurance Company, Hannover Reinsurance
and SCOR Global Life.
Factors affecting the competitive position

Many of RGA’s competitors have greater financial resources. Ability to compete
depends on, among other things,
1. Ability to maintain strong financial strength ratings from rating agencies,
2. Pricing and other terms and conditions of reinsurance agreements, and
3. Reputation,
4. Service, and
5. Experience in the types of business that we underwrite.
52
(44-20) 7325-9175 02 September 2010
How RGA increases its competitive position?

However, competition from other reinsurers could adversely affect our competitive
position. RGA’s target market is large life insurers. RGA competes based on the
strength of underwriting operations, insights on mortality trends based on large book
of business, and responsive service. RGA believes their quick response time to client
requests for individual underwriting quotes and our underwriting expertise are
important elements to our strategy and lead to other business opportunities with
clients. Business will be adversely affected if RGA is unable to maintain these
competitive advantages or if our international strategy is not successful
Factors affecting the Reinsurance demand

We have noted below the factors that affect the Reinsurance demand. Reinsurance is
an arrangement under which an insurance company, the "reinsurer," agrees to
indemnify another insurance company, the "ceding company," for all or a portion of
the insurance risks underwritten by the ceding company. Reinsurance is designed to:
• reduce the net liability on individual risks, thereby enabling the ceding company
to increase the volume of business it can underwrite, as well as increase the
maximum risk it can underwrite on a single life or risk;
• stabilize operating results by leveling fluctuations in the ceding company's loss
experience;
• assist the ceding company in meeting applicable regulatory requirements; and
• Enhance the ceding company's financial strength and surplus position.
Impact of inflation
Inflation is a somewhat better scenario than deflation for us. In spite of unrealised
losses that may occur in the invested portfolio it’s - any extra yield that we make on
investments flows right to the bottom line. And so, on our mortality on our book of
assets supporting our mortality business we don’t have cash call on our assets in a
way that would force us to be unable to hold those assets to maturity. So the inflation
scenario as long as it’s not too bad is generally beneficial. Some moderate inflation
would profitably be good for us.
Munich Re
Executive summary - Munich Re life re
Munich as at the March 2010 Investor Day shows the split of business Dec09 in
terms of net earned premiums was 68% mortality, 29% living benefits including
morbidity, 3% longevity and other. Mortality in particular has little interest sensitive
(morbidity we believe has more as we believe it has relatively more assets). In
particular, mortality is mostly written on YRT basis, which means there is an annual
reckoning of claims and premiums, and there is little build up of reserves.
Most reliable indicator of profitability is MCEV earnings, which is the indicator

Munich uses and on which it incentivises management. Pricing did dip in mortality
in 1997-2003, and this was fully reflected in MCEV earnings in 2008 and 2009. By
comparison IFRS accounting has adverse development buffers, and there is no
reserve increase until this is used. So this creates smoothing and lags. Life re
typically has a relatively low 5-7 year payback, only financing re at on average 3-4
years is quicker.
53
(44-20) 7325-9175 02 September 2010
Mortality generally displays positive variances (which, however, are declining)

because pricing only assumes part of the improvements in mortality that finally come
about. Longevity is a very risky business to price because it is future statistics, and
there is no hedge. Munich prices it assuming zero correlation with mortality, unlike
some peers, where longevity developments have been challenging. There is no clear
line between mainly mortality business and mainly balance sheet support business, as
long as there is sufficient risk transfer. This is because the reinsurer still needs to
make assumptions and take on mortality and lapse risk. Only pure financing re is
different, where generally the main risk is lapses on unseasoned new blocks of
business.
All business at Munich is priced off risk free. A minimum return on risk allocated
capital needs to be achieved. The most stable experience has been in mortality, partly
thanks to the positive mortality trend over time (mortality is typically 30 year plus
contracts). Recent growth has been mainly in Asia and included deals to support
solvency during the capital markets crisis. Solvency 2 and IFRS Phase 2 may also
lead in theory to extra life re demand in mature Europe, but Munich is open to all
developments.
The IFRS ratio of operating profit to revenues is a relatively poor indicator of

profitability, in our view, as reinsurers with a lot of FAS97 business (accounted for
as deposits where only fees and not premiums appear as revenues) would have
consistently higher profitability ratios, while the return on allocated capital could be
the same or even lower.
Life reinsurance at Munich

The main indicator is embedded value, from 2001 traditional embedded value, since
2005 European Embedded Value and since 2008 Market Consistent Embedded Value
(MCEV). Management incentives in life re are all set relative to MCEV, in particular
the creation of new business value, changes in assumptions, and experience gains
and losses.
The October 2008 presentation, which shows the emergence of cash from capital
invested, is a good guide and the business structure has not changed much since. Life
re is very long term; break even is 5-7 years, often 30-40 year contract. This is
relatively little sensitivity to interest rates. Payback is only quicker for financial re
contracts, 3-4 years. Mortality pricing as observed by RGA dipped in 1997-03, and
this has depressed profitability. We believe this was recognised in MCEV in 2008
and 2009 at the latest. Life reinsurance pricing is based on risk free, not corporate
bonds, which means some contracts are lost. For MCEV, investment income is
neutral as it is based on replicating portfolios. Investment income is not used to
accelerate earnings. Of course, if current yields are higher, then MCEV is higher.
A way to assess the reliance on investment income of reinsurance portfolios is to

compare the sensitivity to moves in the risk free - the higher the sensitivity the more
asset intensive the business. So life re with low interest rate sensitivity would be like
Munich Re focused on YRT (pay as you go mortality). By comparison, AdminRe is
mainly invested in corporate bonds so is very sensitive to capital market
assumptions.
The use of profit-to-revenue as a measure of profitability would inflate businesses

which account using FAS 97, ie where there is little risk transfer and it is mainly
54
(44-20) 7325-9175 02 September 2010
accounted as deposits rather than premiums. There the revenue line is low, which
inflates investment returns. We believe a high share of this type of business is one
factor for Swiss Re's structurally higher reported ratio of life re operating profit to
revenues.
Life reinsurance is certainly more stable than P&C for example. This is not to say
that mortality is fully predictable, but that it tends to smooth out over time. Also in
mortality earnings the investment result is relatively modest. Morbidity can also be
subject to relatively adverse trends, linked to the economy. There is the example of
the hike in morbidity among doctors and dentists in the 1992-4 in the US, when the
government was in discussion a health reform which would have cut physicians'
income. There is a recent trend for worsening now too, but it is a drop in profitability
rather than a loss.
As a reminder, 68% of the business is mortality, 29% living benefits and 3%

longevity and other. A typical product has commission upfront, to replicate the
payment pattern to sales of the primary insurer. There are extensive checks on
underwriting rules and discipline. After 3-5 years there is enough evidence to judge if
initial pricing was adequate. If not, then Munich takes a charge against MCEV
earnings.
Under IFRS accounting, prudential buffers are set up, so that any adverse
development is only seen later in earnings. So for adverse deviation to be reflected in
a write off of say capitalised DAC then the adverse deviation has to first exceed these
prudential reserves including profit margins.
Mortality pricing up to ten years ago was very conservative, with little or no account
taken of the strong and continuous trend in improving mortality. This means that
there was steady positive variances from this, which are positive to IFRS earnings,
but as these blocks of business become relatively less important, the proportional
benefit is lower. More recently, mortality is priced in assuming around 30-50pct of
estimated future improvements in mortality. Munich prefers to use conservative
assumptions.
Short term mortality fluctuation becomes smoother over longer periods of time.
$250k-1m is the band where medical checks start for underwriting. The $10m plus
band of mortality risks has relatively few risks and so is statistically more volatile
(law of large numbers does not work so well). In mortality, one of Munich Re's
competitive advantages is to offer high €50m maximum retention on any one risk,
something few competitors can do. This also means that one or two claims in this
band will lead to much lower earnings.
Longevity offers not as much diversification as mortality business, as there is no

hedge. This means there is a limit on the capital Munich is willing to allocate to
longevity, both in absolute terms and relative to the rest of the life re portfolio. Other
large reinsurers have been more aggressive on longevity, and this has not necessarily
been a success. Longevity pricing is what you forecast 20-30 years from now on vs.
Mortality is rather priced on the basis of what you experienced currently.
Negative correlation between mortality and longevity is not proven scientifically,

Munich assumes a zero correlation, some peers assume modest negative correlation
of up to -0.5 (JPMe).
55
(44-20) 7325-9175 02 September 2010
Financing re
There are two types, pure liquidity, which is mainly exposed to lapse risk, and is for
new blocks of business, and which charges fees which are akin to liquidity financing
and balance sheet support.
On balance sheet support, this can be a block of morbidity, also including lapse risk,
and is normally a seasoned block, whereas financing is a new block of business with
great uncertainty over lapse risk.
But there is no clear distinction between life re, which offers balance sheet support
with some financing and that which is pure risk transfer. The main point due to the
different accounting regimes (FAS97 deposit accounting, where mainly fees are
treated as revenues vs traditional life re which shows premiums as revenues) is that
premiums are a relatively poor guide to development of a life re business.
The best benchmark is return on risk capital, in our view. Overall the pricing is set so
the returns on each business are subject to the same minimum requirements
(mortality, morbidity and financing). In practice, the positive development in
mortality means that this has had historically the more stable returns.
In MCEV there is a column for free surplus. However, to appreciate cash flow from
life re one should also consider required capital. We believe some groups also have
sufficient excess capital like Munich Re to leave excess capital in the life re business
until it can be used again.
Businesses which have consistently negative adverse experience are eventually sold
or stop writing business. This has, for example, been the case for some writers of
long-term care in the US, where all the key indicators turned negative: lower
investment returns, higher incidence of loss, longer period of care.
Growth prospects in life re are strongest in Asia. Moreover, the economic crisis
2008-9 helped by leading to demand for capital support and this led to substantial
extra business at Munich (€2bn premiums in block transfers). Also, in some markets
there is now greater acceptance of deals designed to support solvency. It is not clear
whether this will continue beyond the financial crisis. In the more mature markets
there is also the expectation that IFRS Phase 2 and Solvency 2 could give a boost to
life re business. For now this is not proven and we believe Munich is prepared for all
developments.
Munich Re Consolidated results contribution increasing from 2006 onwards

The % contribution of the life segment to the reinsurance has increased from 21.0%
to 28.5% during the period 2006-2009. In the year 2009 28.5% of the consolidated
results of the reinsurance were contributed by the life segment.
56
(44-20) 7325-9175 02 September 2010
Figure 14: % contribution of life increasing in the reinsurance

%
100.0%
26.0% 21.0% 22.0% 24.6% 28.5%
80.0%
70.0%
60.0%
40.0% 74.0% 79.0% 78.0% 75.4% 71.5%

20.0%
30.0%
0.0%
2004 2005 2006 2007 2008 2009
P&C Life
Source: Company reports, 2005 P-C negatively influenced by high NatCat claims (Katrina, Rita, Wilma). Note life includes health which
since Jan 2010 is separated out in a separate division.
Life reinsurance contributes to the diversification

We can note that there is a 9% (21% minus 12%) benefit by including the Life in the
reinsurance segment as can be seen in the chart below as per the 2007 data available.
The increased diversification is mainly driven by mortality, morbidity and longevity
risks, which are mainly uncorrelated to market & P&C risk. This is more important
due to the importance of diversification in Solvency II.
Figure 15: Munich re diversification benefit by including the Life in the Re segment
%
100%
12% 21%
80%
60%
40% 88% 79%
20%
0%
Ex cl Life, 2007 Incl Life, 2007
Div ersified ERC Div ersification effect

Source: Munich Re life investor day presentation 2008
Munich Re economic risk capital (pro rata, JPME)

We have estimated the economic risk capital by segments after diversification as
shown in the table below. Non-life re would be €7.5bn plus allocate the rest pro rate
between non-life re and life re (€3.2bn basic). So, the total would be €8.5bn non-life
re and €3.0bn life re. For non-life primary is €2.0bn and life primary €3.9bn.
57
(44-20) 7325-9175 02 September 2010
Table 35: Munich re capital by segments, 2009

€bn
Economic risk capital, 2009 Basic Pro-rata after diversification
Reinsurance 13.5 11.5
P&C 8.5
Life 3.0
Primary 6.7 5.9
P&C 2.0
Life 3.9
Total 20.2 17.4
Diversification effect (2.8)
Total after diversification 17.4
Life reinsurance market shares

We have shown in the chart below the market shares of the major life reinsurance
companies. We note that Munich Re & Swiss Re have a combined market share of
48% (2009 data) and they dominate the life reinsurance market.
Figure 16: Life reinsurance market share, 2009: dominated by top 2 players
%
Munich Re 27%
Sw iss Re 21%
RGA 12%
Hannov er Re 12%
SCOR 8%
GenRe 6%
Transamerica 5%
XL Re 1%
Partner Re 1%
Other 7%
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%

Life reinsurance - Global market share
Source: Munich Re presentation
We have shown the total GWP of life reinsurance in the chart below. The decline
2005 -7 was due to the planned recaptures of three large accounts in Canada and
Germany including we estimate Allianz Leben. Excluding this effect, the basic book
of business increased steadily during the period 2003-07
58
(44-20) 7325-9175 02 September 2010
Figure 17: Munich Re life reinsurance GWP increased in 2009

€m
10,000 9,742
9,000
7,811 7,665
8,000 7,540
7,299 7,130
6,876
7,000
6,000
2003 2004 2005 2006 2007 2008 2009
Life Re GWP
Source: Company reports. Note life includes health which since Jan 2010 is separated out in a separate division. For 2009 life
standalone was €6796m, the rest was health, and in 2008 the pure life re premiums was €5284m. The large difference in 2009 was
due to the writing of large life and health solvency relief contracts.
Operating EV earnings (as % of beginning EV) has increased in 2009

We can note from the chart below that % contribution of both operating EV earnings
as well as the total EV earnings has increased in 2009. This shows the improvement
in the profitability in the Life reinsurance segment.
Table 36: Munich Re EV earnings % EV beginning of the year

%
40.0% 34.7%
30.0%
19.1%
16.2% 16.6%
20.0% 13.3% 13.5% 14.4%
11.3% 11.8%
9.0% 8.9%9.8% 9.3%7.5%
10.0%
0.0%
2003 2004 2005 2006 2007 2008 2009
Operating EV earnings Total EV earnings

Source: Company reports, 2003 & 2004 EV are based on Traditional Embedded value; 2005 onwards
Market shares comparison of the Life reinsurance

Munich Re has significant market shares in Canada & Germany as can be seen in the
chart below. However Munich Re has low market share in Asia and USA and hence
scope for further improvement.
59
(44-20) 7325-9175 02 September 2010
Figure 18: Market share by regions - Munich Re

%
Global market 21.0%

Canada 53.0%
Germany 50.0%
UK 28.0%
Asia 14.0%
Rest of europe 10.0%
USA 9.0%
0.0% 10.0% 20.0% 30.0% 40.0% 50.0% 60.0%
Life Re market shares by region
Source: Munich Re Life investor day presentation 2008
We have shown below the type of risks for each of the products. Mortality accounts
for most of the biometric risk for Munich re at 68% (2009). Munich Re mostly
focuses on the mortality and morbidity risk.
Table 37: Type of risks according to products: Munich Re mainly focus on the mortality and morbidity
as stated
Products Ordinary life Group life Living benefits Annuity
Type of risks
Mortality Full cover Full cover
Morbidity Full cover
Longevity Full cover
Lapse Selective cover Selective cover Selective cover
Investment Selective cover Selective cover
Pct of total Ord and group life together 68% 29% 3%
Source: Munich Re presentation, Biometric risk portfolio in share of net premium, 2009
Munich re life strategic ambition taken to the 2009 VANB, JPME

We have taken the Munich re global life ambition of 2006 and assumed that the
strategic benefits have occurred in proportion to the year completed. This means that
of the 2006-2011 improvements 60% (corresponding to past 3 years) has occurred
already and 40% is left now. On this basis, we arrive at the new 2011e (JPME). For
the improvements in 11e to 15e we have assumed 100% is still left. So we have
added the 100% improvements during this period to get the 2015e JPME forecast.
60
(44-20) 7325-9175 02 September 2010
Table 38: Munich Re Global life strategic taken to the 2009 actual, JPME
€m, %
As of 2006 As of 2009
Global Life, €m Reported JPME Comment
VANB 228 562
Less exceptional level of sales 2009 312
VANB normalised 228 250
Improvement by 2011 220 80 40% left
Continuously improve core business 120 40 40% left
Grow non-traditional business 40 16 40% left
Expand business model 25 10 40% left
Superior financial model 35 14 40% left
Additional improvement by 2015 175 120 100% left
Continuously improve core business 95 70 100% left
Grow non-traditional business 65 40 100% left
Expand business model 5 5 100% left
Superior financial model 10 5 100% left
JPMe forecast by 2015 623 450
We have shown the JPME targets on the basis of the 2006 strategic improvements in
the chart below. 2009 Munich Re had an exceptionally strong year due to the
solvency relief deals, bringing in €562m value new business. We note that the target
for 2011e is unchanged, and merely adjust them for the 2009 change in methodology
from EEV to MCEV (change reflects also lower interest rates). The 2011e official
target is €330m (€440m previous method) and our forecast for 2015e is €450m.
Figure 19: Munich Re - VANB development

€m
600 562
500
356 330
400
277
300 228
200
100
0
2006 2007 2008 2009 2011e
Munich Re - VANB dev elopment

Source: Company reports. Munich Re Investor day presentation on Life reinsurance 2008
Geographic split of the global life reinsurance market

We have shown in the chart below the global life reinsurance market by regions for
the whole life reinsurance market (not only Munich Re). The US accounts for 50% of
the global market for life reinsurance in terms of life premiums and is the largest and
most developed market.
61
(44-20) 7325-9175 02 September 2010
Figure 20: US is the major market for life reinsurance globally, 2007
%
27%
7%
10%
6%
50%
USA Germany UK Canada Other

Source: Munich Re Life investor day presentation, 2008
Munich Re always had a good market share in Canada but the decline in the
premiums and market share is driven by the scheduled termination of one short-term
non-traditional business treaty.
Figure 21: Munich Re - Market share in Canada on the basis of GWP

%
65% 63% 63% 61%
60%
55%
53%
55%
50% 47% 48%

44%
45%
40%
2000 2001 2002 2003 2004 2005 2006 2007
Munich Re - Market share in Canada

We have shown below the US market share of the recurring new business in 2009.
This is a concentrated market with top 5 players controlling the 85% of the market.
62
(44-20) 7325-9175 02 September 2010
Figure 22: Market share - US recurring new business assumed (2009)

%
RGA Re. Company 22.42%
Sw iss Re 19.26%
Transamerica Re 18.10%
Munich Re (US) 13.52%
Generali USA Life Re 11.75%
Hannov er Life Re 3.25%
Canada Life 3.22%
SCOR Global Life (US) 2.94%
Ace Tempest 1.72%
General Re Life 1.69%
Wilton Re 1.20%
Optimum Re (US) 0.31%
RGA Re. Canada 0.07%
Employ ers Re Corp 0.06%
0.0% 5.0% 10.0% 15.0% 20.0% 25.0%
Market share - US recurring new business assumed
Source: Munich Re presentation
Differentiation factor for Munich Re

The ability to provide large capacity differentiates Munich Re, in our view. We have
quoted one innovation example for North America by Munich Re. Munich Re moved
to a US$50m per life retention, a limit few competitors can offer. The large case
program supports growth and the profit ambitions.
Table 39: Munich Re Innovation example in North America

As stated
Super pool program: innovation for North America
Purpose
Maximise benefits of Munich Re's increased per life capacity
Create sustainable competitive advantage for selected clients
Control access to larger retention for retention management purposes
Detail
Clients get access to increasing levels of Munich Re's US$50m capacity per life
Capacity: Tiered based on volume ceded
Capacity: Unencumbered by retrocession constraints
Program well on track in Canada & US
Source: Munich Re life investor day presentation, 2008
For Munich Re the individual life product is the core, but it also has leading position
in other lines of business. The major lines of business are shown in the table below.
Another advantage of having different lines of business is the diversification of value
added by new business.
Table 40: Munich Re main products

As stated
Munich Re - life products
Individual Life
Long term care (LTC) incl. LifePlans
Individual disability income
Group life and Group LTC
Critical illness
Source: Munich Re life investor day presentation 2008
63
(44-20) 7325-9175 02 September 2010
Munich Re holds the no.2 position in the Life reinsurance market in Asia with a 14%
share. The market is dominated by the top 3 global players; the domestic reinsurers
face tough competition but Korea Re has managed to maintain a strong domestic and
regional position.
Figure 23: Market share in Asia, 2008

%
RGA 19%
Munich Re 14%
Sw iss Re 12%
Gen Re 8%
Hannov er Re 7%
SCOR 3%
Transamerica 2%
Other 5%
Korea Re 15%
China Re 7%
Other 8%
0% 2% 4% 6% 8% 10% 12% 14% 16% 18% 20%
Market share in Asia
Source: Munich Re Investor day presentation on Life reinsurance, 2008
Key drivers of life reinsurance

Munich Re had an underwriting result of €430m in 2007 against a normalized level
of ~€300m. The key drivers for the steady increase from the normalized level are
given below.
• Steady increase in IFRS results – but slight volatility unavoidable

• Margins flowing through from historical superior business selection
• Exceptionally good mortality and morbidity experience in 2007
• Development in line with portfolio growth and value orientation
Table 41: Munich Re - Underwriting result and IFRS net profit

€m
Underwriting result incl technical interest IFRS net profit
2004 215 432
2005 140 976
2006 329 561
2007 430 725
2008 238 705
2009 228 728
Source: Company reports. Munich Re Investor day presentation on Life reinsurance 2008. IFRS net profit relates to life and health re
which since Jan 2010 report separately. Note €430m is in the life re day presentation for 2007 underwriting profit, 2008-9 are JPMe.
The 2008-2009 figures are for life and health.
We have shown in the table below the distributable earnings and the timing in 5 year
steps. The present value as at 2007 was €5.8bn.
64
(44-20) 7325-9175 02 September 2010
Figure 24: Munich Re - Distributable earnings (undiscounted), 5year steps as at 2007

€b
3.0 2.4
2.5 2.0 1.9
2.0 1.6 1.4
1.2
1.5 0.9 0.8 0.7
1.0 0.5 0.4 0.3 0.2
0.5 0.1 0.1 0.0
0.0
2008-12
2013-17
2018-22
2023-27
2028-32
2033-37
2038-42
2043-47
2048-52
2053-57
2058-62
2063-67
2068-72
2073-77
2078-82
2083-87
Distributable earnings
SCOR
General points for SCOR
Current ratio of debt to debt plus equity is 11%, peak was around 40% in the 1999-
2002 period. Prior to the repayment of the convertible this ratio was 16% which
could be (JPMe) a reasonable level of debt. The issue for this ratio, in our view, is
how to raise the debt, and what to do with the proceeds in order to remain value
creating.
We believe there is little interest in raising exposure by like Hannover using side
cars. The risk appetite of SCOR remains relatively low.
Non-life pricing for SCOR is rising on average because SCOR has a relatively small
presence in liability and in the US, both areas where rates are falling. Also because
SCOR is still relatively small (€3bn premiums, around one quarter the size of
Munich in just non-life re, one third that of Swiss Re) it is able to vary its portfolio
more. The A rating means that SCOR is regaining business it had lost, for example in
South Africa where it has now opened an office.
The appetite for risk appears relatively low at this stage for SCOR, but we believe
there is some appetite for longevity risk at the margin. SCOR's non-life portfolio is
very balanced, with no large bets, line sizes (ie premiums per type of risk) of around
€200m max. SCOR remains a committed supporter of 8 Lloyds syndicates, and is
still considering its strategy whether to 'build or buy' as it said at 1Q10 results
presentation.
JPMorgan executive Summary for SCOR on life re

• Within life mortality is the main risk driver.
• Writing some more longevity risk will increase diversification
• Rating: we believe the company sees the current rating of A as fully adequate,
with A+ offering some extra attraction
• New business analysis is done on a detailed basis, IFRS profits are used more as
an early warning system.
• Purchase GAAP (the value of acquired business in force assets) about half of the
book is under this: accelerates recognition of profits, means the subsequent
profitability in terms of ratio of reported profit to premiums is higher for organic
than acquired business.
65
(44-20) 7325-9175 02 September 2010
• For the industry in general, most recent deals on longevity have been swaps with
hardly any reinsurance premium. SCOR has not been active in the longevity
market recently
• Pricing - ROE safely above 900bps above risk free
• Financing and annuity business not much above
• Protection businesses generally have higher RoEs
• Services to clients (e.g. medical underwriting product development etc) are at the
core of the business model.
Details for SCOR on life re

Investment sensitivity
SCOR has much lower investment results than a primary insurer. This is because
SCOR’s Life business is mostly protection, i.e. biometric and lapse risk. SCOR stays
clear of investment related risks embedded in Life reinsurance contracts. SCOR does
have a small GMDB variable annuity exposure which is ten years old and was part of
the Converium acquisition. The only significant investment income sensitivity is that
SCOR has some life business with accumulated reserves (mostly from level premium
business) which are invested at low risk and produce investment income.
Diversification between life and non-life is a 28% benefit. This is allocated to both
P&C and life, and reduces the capital allocated to the respective business. Life in
terms of capital allocated and risk is much smaller than non-life, even though in
premium terms they are equal sized. It fluctuates a little, but in a narrow range.
Increasing the size of the life re portfolio would modestly but not significantly
increase the diversification benefit, we estimate. This is because the maximum
increase in diversification happens after adding a small portfolio (life) to a larger
portfolio (non-life). When they are equal sized there is no or little extra
diversification at the margin. So for example writing more P&C business would
actually reduce the diversification benefit. Longevity - this would add diversification
within the life re portfolio, as it is negatively correlated with mortality (even if the
correlation is very imperfect as they are different cohorts of policyholders). But
SCOR would still need to allocate extra capital, it is not as if writing a negatively
correlated risk would reduce total capital allocated to life.
SCOR's IFRS profitability is influenced by two factors relative to peers.

Firstly, SCOR takes very little investment risk, so compared to say Swiss Re
historically, it had one main source of profit margin, biometric risk, compared with
two for Swiss Re, biometric and investment risk.
Secondly, around half of SCOR's portfolio comes from acquisitions, half from
organic growth. On acquisitions, much of the profit is recognised upfront and the
value of business acquired asset created is then amortised. This tends to reduce the
ratio of annual profitability to premiums relative to the organic business. In the case
of the 2006 Revios deal for example, the recognition of this VOBA asset meant that
SCOR accounted for badwill upfront. In other words, purchase accounting limits the
reported IFRS profitability as a lot is already recognised and embedded in
assumptions.
By comparison, earnings from organic growth are more backended, which means
that the average profitability in terms of return on revenues will tend to be higher
66
(44-20) 7325-9175 02 September 2010
structurally. The accounting under IFRS is similar to US GAAP and is asymmetric. It

is possible to recognise worsening profitability and increase reserves set up at the
start of the contract, but very difficult to recognise improving profitability with a one
off release of reserves. Instead, this flows through over time as positive experience
variances. IFRS profitability is used as an early warning tool to recognise variances
from plan. For this IFRS is very well suited, in our view.
There is a third level of accounting (other than IFRS or embedded value) which is
used for recognition of cash flows and capital for dividends out of the life re division.
This is statutory accounting, because only statutory reserves can be released to cash
flow and dividends. Changes in IFRS reserves affect reported profits but cannot be
released to capital or cash flow. Over time embedded value, which is effectively the
discounted future statutory life cash flows, converges with statutory reporting, but
embedded value does not identify profits which can be released this year as a point
estimate. Only statutory accounting does this (however it is contained in EV reports
in the movements of ‘Adjusted Net Worth’ or ‘Adjusted net Asset Value’).
Fungibility and legal structures. Since the 2004 adoption of the EU Reinsurance
directive, the benefit of local regulatory regimes, particularly the strict UK FSA
environment, has been much reduced. So, primary insurers no longer have a strong
preference for any local carrier within the EU. For this reason SCOR has now
merged most of its European operations into one legal structure, and this improves
capital fungibility. Asia and Canada are accounted for as branches of this one
European legal structure, SCOR Global Life SE. The US is still different and there is
no prospect that this can be simplified, which means that this is a potential source of
capital friction, in our view (for all European Life reinsurers).
In terms of ranking of profitability in terms of profit divided by premium, pure

protection is higher than SCOR's average 6-7pct range, financing business is much
lower (most savings related are financing),. Overall, 99% of deals are written above
9% ROE, which means that the RGA range of profit to capital of 10-14% is also we
believe relevant to SCOR. For SCOR even the pure EIA equity indexed annuities
business was at 9%.
There is no split of margin between mortality, investment risk etc. In addition, much
of the add on services such as underwriting guidelines for mortality and critical
illness and other services such as medical checks by doctors which are actually paid
for by the reinsurers, are paid for out of the ROE, which is not just a reward for risk
taking, but also a reward for services, which also include claims management in
disability, critical illness and long term care. Around 71%of the portfolio is
mortality, and this is in terms of generations of contracts (refer Table 43: Split of
SCOR business).
Operating profit is a good measure, because the technical result can fluctuate and be
offset by opposite fluctuations in investment income. Technical result consists of
underwriting plus investment income on funds withheld. That return in Germany can
only vary in a range of 2.2-4% as there are very precise regulations to cap it. So it is
relatively low volatility. SCOR relies on embedded value for its life business to value
it, and many analysts use the same approach. By comparison, IFRS is seen as an
early indicator of claims volatility.
67
(44-20) 7325-9175 02 September 2010
Low life re interest rate sensitivity

SCOR’s relatively low interest rate sensitivity is supported by the following extract
page 51 of their 2009 reference document (the official annual report). Essentially the
point is that: a) in countries with products with minimum guarantees, that is the
valuation discount rate. So the liabilities are locked in, and the risk is that the
investment portfolio misses that minimum return. And b) for other countries SCOR
includes a provision for adverse deviation, which means that the sensitivity is
dampened by a relatively large buffer.
We cite below from SCOR’s official 2009 reference document.
Life interest rate sensitivities (2009 SCOR reference document page 251)
“Although in general all long term liabilities are discounted, in most cases there is no
immediate accounting impact from a 100 basis point change in interest for the
following reasons:
1. For the German, Italian, Swiss and Austrian markets, valuation interest rates are
typically locked-in at the minimum interest rate guaranteed by the ceding on the
deposited assets covering the liabilities.
2. For the business written in the United Kingdom, Scandinavia, United States
(traditional, non-savings products) and France (excluding Long Term Care),
valuation interest rates are locked-in based on a prudent estimate of the expected
rate earned on assets held less a provision for adverse deviation.
The life products with guaranteed minimum death benefits (GMDB) are not
materially sensitive to 100 basis point decrease in interest rates. An increase in
interest rates would not result in a decrease in the level of reserves as the interest
rates are locked-in.
The liabilities recorded for the annuity business would not change materially to a 100
basis point change in interest rates as they are linked to account values. However, the
shadow accounting would be impacted.
For Long Term Care products in France, ceding companies use valuation interest
rates established by French regulators which are linked to some extent, to market
rates. Reserve movements reported by ceding companies are influenced by numerous
factors, including interest assumptions, where are not distinguished separately.
SCOR does not actively revise the valuation interest rates during its reserving
process. Due to lack of direct data, the interest rate sensitivity cannot be precisely
analysed.”
Life – also low sensitivity to equity markets (official 2009 reference document page
251-252).
“In general, equity movements have no impact on the reported liabilities of the life
business as the underlying policies and reinsurance contracts are typically unrelated
to equity prices. For some risk premium treaties (where the underlying insurance
policies are unit-linked or universal life) the sums at risk and thus the expected
claims, vary with the movement of the underlying assets. However, under almost all
reinsurance programs, premiums are also linked to the sums at risk such that the
liability would not materially change.
68
(44-20) 7325-9175 02 September 2010
The premiums on the Guaranteed Minimum Death Benefit (GMBD) business

underwritten by the SCOR Group in the U.S. market vary with the value of the
underlying assets rather than the sum at risk. Thus, premiums would decrease under a
decline of the equity values whereas the expected claims would increase thus leading
to an increase in the liability. However, included in the reserve calculation is a
prudent margin for this fluctuation. Accordingly, the level of reserves recorded for
this business would remain unchanged in the event of a 10% decrease in equity
values. As the valuation assumptions are locked-in, an increase in equity values,
resulting in a decrease in the economic liability, would not impact the level of
reserves recorded in the financial statements.”
Distributable cash flow

Distributable cash flow (excluding the return of required capital) is less than
operating profit. Distributable cash flow expectation from 2009 in force business by
SCOR for 2010 is about €150m, which compares with EEV operating profit of
€190m for 2009 (which includes inforce block transaction for Ireland and UK
protection) and we forecast €160m 2010e (our 2010e forecast assumes the same
4.5% growth in EEV operating profit in 2010 as in 2009; this is conservative as at
1H10 SCOR reported 7.6% growth in underlying premiums excluding low margin
equity indexed annuities, and we exclude a repeat of the 2009 new business from
exceptional block transactions).
So we estimate distributable cash flow is 25% less than EEV operating profit. We
believe this supports our view of the life book, which is that it does produce
distributable cash flow, but more slowly than non-life, where the earnings are 100%
distributable in our view.
We show below SCOR’s reporting of distributable cash flow. The interesting point is
that this fits a curve with an annual decline of 5.8%, which effectively is the implicit
cash flow risk discount rate, including the risk free, and the inbuilt margins for lapse
risk and for mortality risk.
69
(44-20) 7325-9175 02 September 2010
Figure 25: Distributable cash flow in €mn

160
140
120
100
80
60
40
20
0
2005 2010 2015 2020 2025 2030 2035
Source: SCOR EV 2009 presentation. Distributable cash flow excludes free surplus and before release of required capital. €2285m
total JPMe.
Of the total €2285m distributable cash flow (and this is based on SCOR's reporting
of €1934m Dec09 embedded value), 43% is in the first eight years, 50% within the
first ten years, 67% within the first fifteen years, and 94% within the first 24 years.
Table 42: SCOR: Distributable cash flow in €mn, as pct of Dec09 EV €1934m, and as pct of total
€2285m JPMe distributable cash flow – 39% of EUR2.3bn total cashflow is in the first 5 years
€m
Distributable Pct of Pct of total
cash Dec-09 EUR2285m
flow EUR1934m EV cash flow
2010 150 7.8% 6.6%
2011 141 7.3% 6.2%
2012 132 6.8% 5.8%
2013 125 6.5% 5.5%
2014 118 6.1% 5.2%
2015 111 5.7% 4.9%
2016 105 5.4% 4.6%
2017 99 5.1% 4.3%
2018 94 4.9% 4.1%
2019 88 4.6% 3.9%
2020 83 4.3% 3.6%
2021 78 4.0% 3.4%
2022 73 3.8% 3.2%
2023 69 3.6% 3.0%
2024 65 3.4% 2.8%
2025 61 3.2% 2.7%
2026 58 3.0% 2.5%
2027 55 2.8% 2.4%
2028 52 2.7% 2.3%
2029 49 2.5% 2.1%
2030 47 2.4% 2.1%
70
(44-20) 7325-9175 02 September 2010
71% is traditional mortality and lapse risk

The following table shows SCOR’s split of life re business. Excluding the equity
indexed annuity business, which is a very low risk, high volume and we believe
relatively low margin product, and which SCOR has said they would cut back
sharply in 2010, life and financing, where the main risks are mortality and lapses,
accounted for 71% of total premiums in 2009. We believe this highlights the
predominance of traditional and relatively predictable business in SCOR's business
life reinsurance business mix.
Table 43: Split of SCOR business

2009 2008
Life 45% 52%
Financing 17% 18%
Annuities (equity indexed annuities) 13% 4%
Health 8% 7%
Disability 7% 7%
Long term care 4% 4%
Critical illness 4% 3%
Personal accident 2% 5%
Total 100% 100%
Total ex EIA 87% 96%
Life and financing 62% 70%
Life and financing to adj total 71% 73%
Source: SCOR 2009 embedded value slides
Main sensitivities of SCOR life re portfolio

The following table shows the main sensitivities of the life re portfolio of SCOR, in
terms of EEV. The main points are:
1. The single largest sensitivity is life mortality and morbidity. This shows SCOR’s
portfolio has a very traditional life re mortality exposure.
2. the annuities sensitivity to mortality and morbidity is just €0.6m. So far SCOR
does almost no longevity risk.
3. The impact of lower interest rates alone on future life earnings, excluding the
positive uplift from the use of a lower discount rate, is -€98m ie -5.1% in terms of
the base case of €1.9bn embedded value. This is significantly less than the €269m
sensitivity to mortality/morbidity, for a 5% worsening (the table gives the effect
of a 5% improvement in mortality and we have assumed the reverse).
Table 44: SCOR EEV sensitivities Dec09

€mn Amount Pct of base case
Base case 1,934.3
EEV sensitivities:
Mortality/morbidity (life) -5% 269.1 13.9%
No mortality improvements (180.4) -9.3%
Mortality/morbidity (annuities) -5% 0.6 0.0%
Lapse rates -10% 28.3 1.5%
Maintenance expenses -10% 24.9 1.3%
Discount rate -100bps 118.1 6.1%
Interest rates -100bps 20.3 1.0%
Equity and property capital values -10% (12.8) -0.7%
We were positively surprised by SCOR’s statement on page 251 of their 2009

official reference document (see above) that although life liabilities are discounted,
71
(44-20) 7325-9175 02 September 2010
there is little accounting impact in IFRS of a change in interest rates because: a) in

countries with minimum guaranteed rate, that is the discounting rate used at
inception, and not change; and b) in other countries there is a sizeable margin for
adverse deviation in the reserves to protect against this.
Table 45: SCOR Dec09 Investment Portfolio in €m

Pct of total €mn
Real estate 2% 399
Equities 4% 799
Alternatives 2% 399
Govvies and assimilated 23% 4,593

Covered and agency 5% 998
Corporate 13% 2,596
Structured products 4% 799
Funds withheld 39% 7,788
Cash 8% 1,598
Total 100% 19,969
To double check our understanding we compared the interest rate sensitivity of

SCOR’s fixed income assets to the reported IFRS sensitivity of a 100bps change in
interest rates.
We estimate the impact of a 100bps fall in interest rates on shareholders' equity at

€226m (see following table).
By comparison SCOR reports a sensitivity of €+204m to a 100bps fall in interest

rates for IFRS.
We believe the accounting reserve adjustment is the difference, and it is relatively

modest at €24m.
Table 46: JPMe estimate of interest rate sensitivity - our €226m estimate is only modestly above
the €204m reported, which we believe means the IFRS sensitivity of life reserves to interest rates
is relatively low (ie the difference of €22m)
€m
Fixed income 8,986
Duration 3.6
Tax 30%
Net 226
High mortality sensitivity, but low investment risk

Compared with primary life insurers, SCOR’s EEV has low sensitivity to investment
risk, and high sensitivity to mortality risk.
The following table shows that 100bps lower interest rates boosts EEV at SCOR by
1%, at Munich by 4%, and the peer group average is a negative -10% fall. This is
mainly due to the very high negative interest sensitivity of the German life insurers
Allianz and ERGO, where the combination of very long duration of the life contracts,
high guaranteed rates on the back book, and low current interest rates, means that this
is the single largest risk they face.
72
(44-20) 7325-9175 02 September 2010
SCOR’s +1% consists of a +6% positive due to the use of a 100bps lower discount
rate more than offsetting a -5% pure impact of lower interest rates on future
investment income and future cash flows.
And on mortality, a 5% drop in mortality would add 14% to SCOR’s EEV, 19% to
that of Munich, but just 1.5% for the primary life insurers.
The fact that reinsurers focus on mortality risk is supported by the sensitivity to the
assumption of no mortality improvement. As the reinsurers have priced in their
products assuming some continuing mortality improvement, no mortality
improvement would cut their embedded value (-9% at SCOR, -37% at Munich). By
comparison mortality is priced at current best estimates at the primary insurers, so
there is no sensitivity to this. This means that when primary insurers reinsure, they
lock in a particular rate of mortality improvement (around 50% of the historic trend
we estimate).
Table 47: SCOR and Munich Re have low investment risk compared with primary life insurers
Pct of base case 2009 EEV
Average
SCOR MunichRe primaries Generali ERGO Allianz AXA
Mortality/morbidity (life) -5% 13.9% 19.5% 1.5% 2.1% 1.1% 1.0% 2.0%
No mortality improvements -9.3% -37.1% not av not av 0.0% not av not av
Mortality/morbidity (annuities) -5% 0.0% -0.1% -0.9% -0.7% -0.9% -1.0% -1.0%
Lapse rates -10% 1.5% -0.8% 1.8% 2.4% 0.6% 1.0% 3.0%
100bps lower interest rates 1.1% 3.9% -14.6% -10.4% -22.9% -19.0% -6.0%
Equity and property capital values -10% -0.7% 0.0% -4.0% -6.2% -1.8% -4.0% -4.0%
Profitability of a typical life (mortality) reinsurance contract

The following table summarises the profitability of a typical life contract, whose
main feature is to provide financing to the primary insurer equal to the premium in
the first year, and to 5% of the annual premiums after.
There are three points to make:
1. Reported profits are reported last under statutory, most upfront in embedded
value, where the new business value allocates most of the return to the point of
sale, and there is a midway solution for IFRS, which smoothes profits using
deferral and amortization of acquisition costs..
2. Total profit is the same. There is some rounding in the model, but the numbers
add up to €516m for statutory, €479m for IFRS, €483m for embedded value.
3. ROE is highest on average under statutory, lower under embedded value, midway
under IFRS.
Table 48: Summary profitability under three different accounting regimes: statutory, embedded
value and IFRS
€mn
Statutory EEV IFRS
Average capital 52 48 48
Average annual profit 328 627 521
Average ROE 15.7% 7.7% 9.2%
Source: Company reports and J.P. Morgan estimates. Profits should be same in all three, some rounding differences.
73
(44-20) 7325-9175 02 September 2010
We show the allocated capital under the three approaches (statutory, EEV and IFRS
in the below chart. The main reason that the profitability under statutory is so much
higher than IFRS or EEV, is that statutory, although it requires capital upfront,
requires the least amount of capital on average. So the return on capital is the highest
under statutory. The total profit in all three regimes is the same, as they are the same
underlying cash flows.
Figure 26: Chart of allocated capital under statutory, embedded value and IFRS
€m
800
700
600
500
Statutory
400 EEV
IFRS (capped at 1/5x statutory)
300
200
100
0
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019
Source: SCOR investor day 2009 contract simulation and JPMe for capping the IFRS capital at 1.5x statutory. The dividends our are
deducted from both embedded value and IFRS capital.
In the model with IFRS capital detailed above and which we detail below we have
assumed that under IFRS cash can be ‘dividended’ out as soon as the IFRS
accumulated capital (initial capital plus profit for year) reaches 1.5x statutory capital,
which we estimate is sufficient for a rating range A-AA. We have deducted these
dividends from EEV capital and from IFRS capital.
74
(44-20) 7325-9175 02 September 2010
Table 49: Estimated ROE under statutory, EEV and IFRS accounting
€m
Average
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 ROE
Statutory:
Premium 1,000 900 828 770 724 688 653 621 590 560
Claims -310 -418 -481 -514 -532 -545 -560 -574 -589 -604
Commission (100% plus 5% renewal) -1,000 -45 -41 -39 -36 -34 -33 -31 -30 -28
Change in reserves -386 -213 -102 -30 20 61 101 142 183 224
Investment income 0 12 18 21 22 21 19 16 12 7
Pretax treaty profit -696 236 222 209 198 191 180 174 167 159
After tax treaty profit after 35% tax -452 153 144 136 129 124 117 113 108 103
Expenses pretax -40 -36 -33 -33 -31 -29 -28 -26 -24 -22
After tax expenses -26 -23 -22 -20 -19 -18 -17 16 -15 -15
Net profit -478 130 122 116 110 106 100 129 93 88
Required capital:
insurance risk 321 338 343 344 336 324 310 295 279 262
ALM risk 10 15 18 18 18 16 14 10 6 0
operational risk 0 21 20 18 17 16 16 15 14 13
Total 331 374 381 380 371 356 340 320 299 275
Statutory ROE -144.5% 36.6% 33.7% 31.9% 31.0% 31.1% 30.8% 42.0% 32.3% 32.9% 15.7%
EEV:
EEV operating profit 176 53 49 44 39 35 30 24 19 14
Embedded value:
Opening 478 654 707 756 800 839 874 904 928 947
EEV operating profit 176 53 49 44 39 35 30 24 19 14
Dividend 48 54 62 70 66 71 70 485
Closing 654 708 698 675 641 604 557 506 35 708
EEV ROE 31.1% 7.8% 6.9% 6.3% 5.7% 5.3% 4.8% 4.1% 3.6% 5.2% 7.7%
IFRS:
Statutory pretax -696 236 222 209 198 191 180 174 167 159
Less expenses -40 -36 -33 -33 -31 -29 -28 -26 -24 -22
Subtotal -736 200 189 176 167 162 152 148 143 137
DAC amortisation 816 -104 -99 -95 -92 -89 -87 -85 -83 -81
Operating result 80 96 90 81 75 73 65 63 60 56
Operating result to premiums 8.0% 10.7% 10.8% 10.5% 10.3% 10.6% 10.0% 10.1% 10.1% 10.0%
Taxed profit 52 62 58 52 49 47 42 41 39 36
IFRS capital (capped at 1.5x statutory)
Opening 478 497 561 572 570 557 534 510 480 449
Taxed IFRS profit 52 62 58 52 49 47 42 41 39 36
Dividend -2 48 54 62 70 66 71 70 485
Closing 497 561 572 570 557 534 510 480 449 0
IFRS ROE 10.7% 11.8% 10.3% 9.2% 8.6% 8.7% 8.1% 8.3% 8.3% 7.8% 9.2%
Source: Company reports from SCOR 2009 investor day slides and J.P. Morgan estimates. JPMorgan assumption for dividendable profit is that IFRS profit is capped at 1.5x statutory capital..
Swiss Re
The following is JPMorgan’s interpretation of Swiss Re’s view of life reinsurance
1a). Accounting - US GAAP is not very informative as it reflects mainly historic

assumptions
US GAAP gives a relatively poor view of profits as it is mainly about baked in
assumptions on mortality, lapses and profitability. So has a very stable earnings
unwind, but says little about new business.
By comparison EVM gives a complete picture of profitability, both value added by

new business and also experience variances and assumption changes. The part of
EVM which leads to high volatility is economic assumptions.
Finally dividends are still paid out of statutory earnings and there are three main
accounting conventions for this. Firstly in the US amortised cost is used for both
75
(44-20) 7325-9175 02 September 2010
reserves and assets, so sensitivities are moderate. UK in theory is market value on

both sides, but in times of volatility and stress FSA does exercise its influence to
moderate impacts, as in 2008 when it allowed the use of liquidity premiums. Finally
in Switzerland statutory is now consistent with Swiss Solvency Test, ie uses risk free
for both sides, assumes zero extra risk return on risky assets.
1b). Reported earnings vs cash flow – this is JPMorgan’s view of the reported
profits vs statutory earnings which set’ dividendable’ cash flows
In general, we believe that statutory earnings will be higher than Swiss Re's
convention of allocating just the risk free to life, as long as returns on risky assets are
positive, as they were in 2Q10 when corporate bond returns were strongly positive,
which means that the cash flow will be higher. This will reflect the superior cash
flow of admin re. Admin re cash flow in the first 6-12 months is a return of a large
part, we estimate around 30%, of the capital invested; so for example on the Barclays
Life £750m deal in November 2008 around £250m of the capital was expected to be
returned by May 2009. And admin re is more geared to asset values due to the value
of the unit linked accounts; as these are flat on November 2008, this means that the
fee income is less than expected, and this is not hedged. Whereas for mortality, US
GAAP gives about the same figure for profits as statutory cash flow, cash flow on
Triple x reserves business is negative compared with positive US GAAP
profitability.
Note that Swiss Re has no intention of changing this convention of just allocating the
risk free to life. The objective was to focus its underwriters on underwriting, where
they believe thanks to their proprietary LMS model that they have superior returns in
the US, and away from asset management, where the track record is poor. The risk
free rate in the currency is for the duration of the actual reserves. This understates a
little the profitability of admin re, as the last major admin re deal was Barclays Life
November 2008, and there have been no deals since.
1c) The main accounting conventions in life are FAS60 (traditional life, constant
emergence of margins, relative mainly to premiums) and FAS97 (mainly
universal life but also contracts where there is no risk to Swiss Re like unit
linked, with smooth emergence of economic margins, but more volatile
emergence of US GAAP margins relative to the underlying investment
performance including the change in unit linked account values).
2. Write down risk is remote we believe

Swiss Re explained that the risk of write downs of the various intangibles (value of
acquired business in force, deferred acquisition cost) is low and we believe remote.
This is because under US GAAP the portfolio as a whole would have to show
negative returns (loss recognition is at the aggregated level), for this to happen. Swiss
Re does it by choice across subsegment, life, health and admin re, all on a global
basis.
The mechanism is that Swiss Re checks actual variances against its Adverse
Deviation reserve, and when this is exhausted writes off intangibles and resets
reserves completely, segment by segment. And Swiss Re has stated that the
underlying return of the pre 2004 mortality book of business was relatively low but
still positive.
76
(44-20) 7325-9175 02 September 2010
Swiss Re has stated that the main way to check on profitability is to review the uplift
on the transition from US GAAP balance sheet to EVM. In 2008 at the peak of the
crisis this was a positive SF3.2bn for life, SF8.3bn in 2009. For the group in 2008 the
uplift was a negative SF1.9bn, a positive SF2.3bn in 2009.
Table 50: Swiss Re: EVM uplift from life reinsurance in Sbn
2007 2008 2009
Additional discounting -3.1 -3.7 1.4
Reserving basis:
GAAP margins 12.8 10.3 12.2
Other -0.3 -1.1 -1.0
Recognition differences 0.3 0.5 0.3
Extra tax -1.6 0.1 -0.9
Other 0.1 0.1
Frictional capital costs -4.2 -3.0 -3.7
Total EVM adjustments 4.0 3.2 8.3
In particular, the EVM 2009 presentation on slide 21 shows that the uplift from
revaluing reserves from US GAAP to EVM is SF12.2bn, which is the main positive
source of the SF8.3bn uplift of life EVM vs US GAAP. That figure was SF10.3bn in
2008 accounting again for the main positives underlying the net SF3.2bn uplift from
US GAAP to EVM capital in life.
Slide 11 from the 1Q09 Appendix slides highlights that the reserves are relatively
robust: the US GAAP life reserves contain SF10bn of uplift of which 2/3 is
traditional life (rest is equally health and admin re). These reserves can withstand as
a sensitivity 105% mortality vs assumptions, 100bps lower investment returns across
all segments, 10% higher lapses. By contrast, and this highlights the sensitivity of
FAS97 accounting (premiums treated as deposits, relates mainly to unit linked)
PVFP amortization, which negatively affects profit recognition, is accelerated by
SF0.1bn for 20% fall in equities, and by SF0.4bn for 20% one off lapse shock.
3. Key business factors for profitability

The key factor is risk free interest rates. This is a bigger factor than we had
previously thought, when Swiss Re gave figures of underlying which were based on
year earlier rates. So at 2Q09 when Swiss Re said the normalized profit figure for
that quarter was SF300m, this was based on 2008 risk free, which was 2.8% average
for the year for 5 year US government bonds. At 2Q10 when Swiss Re said the
normalised profit figure for that quarter was $200m (say SF200m, ie 33% less than a
year earlier) this was based on the rolling average risk free year yields which had
dropped significantly (the duration of the life book is 8-9 years we estimate). This
sensitivity implies that Swiss Re life margins are based on invested life assets of
around SF57bn, ie $54bn (SF100m quarterly earnings difference, SF400m for the
year / 70bps = SF57bn x 0.95 FX = $54bn). A cross check is that at 1Q10 Swiss Re
gave a figure of $240m per quarter ie SF250m, and this was based on 2009 risk free
rates. These were 2.5% for 5 year US government bonds at Dec09, ie just 30bps
lower; this would imply interest sensitive assets of SF66bn or $63bn. We believe this
provides a consistent view of the interest rate sensitivities. So equally, should the 5
year US government bond yield recover to say 4%, then profitability would be
significantly higher, we estimate around SF400m per quarter, ie double the 2Q10 run
rate of we estimate SF200m.
77
(44-20) 7325-9175 02 September 2010
Swiss Re has stated that the return on risky assets is less of a factor, as it has no
intention of re-risking other than adding around 1-2pct allocation to equities and 1-
2pct to bonds to be funded out of the cash of $36bn (short term plus cash is $50bn).
We do note that admin re's earnings are only positive if we allocate their share of
return on risky assets, which is most of the $18.6bn corporate bonds at 2Q10 (ie
around $3540m annualised uplift we estimate).
The other factors which Swiss Re tracks are mortality, lapses, and the distribution of
risks in the primary insurers' portfolios (age distribution, male/female distribution,
behaviour distribution, and association with different trends in terms of mortality and
lapses).
Swiss Re makes continuous assumptions about distribution (are risks male or female,
what age, what cohort in terms of mortality and lapses?) and we believe it took
negative model assumptions in 2Q10. This volatility was made worse in 2Q10 by
cedant changes in reserves, which are taken as they are communicated and are not
managed in any way.
The 2Q10 results reflect volatility due to cedant updates on reserves and accounting
for CVA (credit value adjustment) and DVA (debit value adjustment), which is the
volatility due to changes in the CDS of Swiss Re and its cedants, and Swiss Re is
determined to provide more transparency of life re, without, however, changing the
accounting, which is that life re only gets allocated the return on equivalent vintage
risk free assets, not the return on the actual underlying risky assets like the $17bn
corporate bond portfolio.
4. Actions taken in 2009 to deal with the crisis and which affected life re
profitability
a) Swiss Re repriced Triple X reserve business, where effectively Swiss Re now
charges a higher cost of funding (ie the cedant’s effective cost of funding) whereas
previously it had offered a cost of funding based on estimated cost of letters of credit,
the main alternative form of finance, and these were set relatively low. So Swiss Re
as a result is writing slightly shorter duration business as its rates are not attractive to
cedants for thirty year durations.
b) Swiss Re refocused on mortality business. This is positive for profitability longer

term we believe (new business has a only a gradual impact on the average of the
portfolio).
c) Swiss Re did no new admin re business, as did not find attractive business at its
pricing level. The likelihood, however, of some smaller $100-200m admin re deals
being closed now is higher. Negative for profitability we believe, as admin re has a
very quick payback, at least in terms of statutory earnings. This may have been the
miss factor which affected 2Q10 life earnings as Swiss Re had still signed no admin
re deals then.
d) In January 2010, Swiss Re reinsured part of their US life re backbook, with 2010
premiums of SF1.7bn, to Berkshire. This was because on the Swiss Re model of
investing at risk free rates profitability was very low, whereas the cash flow was
potentially worth more to Berkshire. The deal frees SF300m capital for Swiss Re,
and is small positive in terms of profitability as there is no loss recognised on the
78
(44-20) 7325-9175 02 September 2010
deal and the business reinsured had, we estimate, well below 6% ROE on the
assumption that the reserves are invested at the risk free rate.
5. Ranking of profitability.
Swiss Re's EVM based model shows that at the moment profitability is highest in
property non-life, next highest in admin re if we credit return on risky assets, next
highest (and least volatile) for traditional mortality, and lowest for casualty.
Swiss Re: general points about profitability and the 12% ROE target
Swiss Re's target 12% ROE assumes the following:
1. repayment of the Berkshire convertible at March 2011;

2. Also some capital management; Swiss Re’s priority is to return to a normal
dividend policy as soon as possible, and then, if there is still excess capital, it said
it would potentially consider other options.
3. Interest rates rise 100bps from their Dec09 levels, ie 140-150bps from now.
Swiss Re assumes that the annual cash flow from operations and from maturing
assets is reinvested at the higher yields. We assume, to simplify our models, that
there is a one off reinvestment of $10-11bn, which we estimate corresponds to the
short duration mismatch, as assets are 90% of the duration of liabilities (this is for
the group as a whole, and the mismatch is mainly in life). Plus, there the benefit
of 1-2pct each extra allocation for equities and corporate bonds, ie 2-4pct or $4-
6bn. So in total we believe the 12% target ROE assumes uplift in investment
income from current levels of we estimate $14-17bn x 140-150bps, ie $200m-
$270m extra.
4. Increasing the allocation of casualty from currently 16% of total group premiums
to we estimate 24% (but below the 28% peak of 2007). Casualty is potentially a
higher ROE business than property if premium rates recover and interest rates
rise. We estimate this adds $240m to profitability (assuming the capital allocated
rises by $1bn from we estimate $2bn to $3bn and the return on casualty rises
from we estimate 6% now to 12% ROE, ie from $120m to $360m ie an extra
$240m).
5. We estimate expected profits rise excluding the Berkshire Convertible by $440-
510m to $2.4bn from $1.9-2.0bn previously. As for S&P, Swiss Re said that it
had done all to merit a higher rating, including freeing up rating capital during
2Q10 by reducing the ALM duration mismatch to just 10% (90% asset to liability
duration ratio). Swiss Re stressed that buying back the Berkshire convertible was
a full commitment, whatever natural catastrophes may cost in the next six
months.
6. Organic growth in life and health follows the economic cycle and in particular the
housing market, which is when households buy more life insurance cover.
Longevity is assumed to increase gradually, but under the IFRS convention of
spreading earnings over the life of the contract this is not a significant contributor
to life profits.
7. Expense control is to remain below inflation, so improving operational leverage.
8. Swiss Re makes no assumption about positive reserve run off in non-life.
79
(44-20) 7325-9175 02 September 2010
View on the industry cycle:
1. Swiss Re hopes and expects premiums rates to rise end 2011, particularly in
casualty.
2. We believe the worse case for Swiss Re is more of the same, which means
sluggish growth, low interest rates, low reinsurance margins. Swiss Re is at a
disadvantage vs its peers in such a low inflation environment, in our view,
because the reserve run off profits up to the first SF2bn go to Berkshire.
Better than the current environment we believe would be a fall in interest rates, as
this would force a return to underwriting, where Swiss Re can compete with a good
track record, or a rise in interest rates, which would allow Swiss Re to return to
casualty which is currently just 16% of total group premiums down from a 28% peak
in 2007, and would boost the profitability of casualty from currently we estimate just
5% ROE.
Swiss re insurance risk stress test

We have shown below the Swiss Re’s exposure to major nat cat events net of
retrocession and securitization. The difference with 2009 and 2008 is due to higher
retrocession & securitization in California earthquake and decrease in the European
windstorm and Japanese earthquake.
The SF3.6bn figure for pandemic risks shows that peak mortality risk is a bigger risk
for Swiss Re than natural catastrophes: SF3.6bn risk from pandemic vs SF3.2bn for
the largest natural catastrophe.
Table 51: Insurance risk stress tests

CHF billion, %
Significant events with a 200 year return period 2008 2009 Change in %
Natural catastrophes
Atlantic hurricane -3.5 -3.2 -9%
Californian earthquake -3.9 -3.2 -18%
European windstorm -2.1 -2.7 29%
Japanese earthquake -1.7 -1.9 12%
Life insurance
Lethal pandemic -3.5 -3.6 3%
We show the history of EVM 2006-9 together with our forecast for 2010e. We
assume for our 2010e forecast:
1. the natural catastrophes and large claims in non-life reduce the EVM profit as
much (EVM and US GAAP are closest in non-life).
2. for asset management a small positive from a modest narrowing of spreads, more
than offset by a negative as SwissRe is short duration and interest rates are down.
Asset management is measured relative to benchmarks, so the normal run rate we
believe should be zero before capital costs and 2009 was quite exceptional.
3. legacy a relatively modest contribution as the run off is almost complete.
80
(44-20) 7325-9175 02 September 2010
As a result, our 2010e forecast of EVM profit before capital costs is much lower than
2009, which benefitted from a significant tightening of spreads in the asset
management unit.
Table 52: EVM reporting (the Swiss Re equivalent of embedded value) 2006-10e – before capital costs
SFm
2010e 2010e
Non-life Life Asset mgmt Legacy Group Total
New business profit 300 280 -700 60 -60
Previous years' profit/loss 370 300 -100 50 620
Total profit after capital costs 670 580 -700 -40 50 560
Release of capital costs 550 260 700 100 -100 1,510
Income before capital costs 1,220 840 0 60 -50 2,070
2009 2009
New business profit 1,002 311 4,162 213 432 6,120
Previous years' profit/loss 370 401 -309 498 960
Total profit after capital costs 1,372 712 4,162 -96 930 7,080
Release of capital costs 552 261 1,677 304 20 2,814
Income before capital costs 1,924 973 5,839 208 950 9,894
2008 2008
New business profit 658 595 -12,366 -2,396 -2,782 -16,291
Previous years' profit/loss 618 -141 -2,624 -296 -2,443
Total profit after capital costs 1,276 454 -12,366 -5,020 -3,078 -18,734
Release of capital costs 978 666 1,213 488 2,265 5,610
Income before capital costs 2,254 1,120 -11,153 -4,532 -813 -13,124
2007 2007
New business profit 1,787 1,247 -761 -1,534 45 784
Previous years' profit/loss 851 -42 0 0 100 909
Total profit after capital costs 2,638 1,205 -761 -1,534 145 1,693
Release of capital costs 1,601 1,136 728 69 -338 3,196
Income before capital costs 4,239 2,341 -33 -1,465 -193 4,889
2006 2006
New business profit 1,695 391 995 -71 3,010
Previous years' profit/loss 137 328 0 225 690
Total profit after capital costs 1,832 719 995 154 3,700
Release of capital costs 1,578 1,007 857 99 3,541
Income before capital costs 3,410 1,726 1,852 253 7,241
We show the relation between changes in total EVM net worth and EVM income in
the following table.
Table 53: Swiss Re change in life re economic net worth

SFm
ENW 2005 2006 2007 2008 2009 2010e
Brought forward 17,059 20,065 39,200 38,400 18,600 28,494
GEIS deal 1,989
Opening adjustments 161 -1,000 2,300
Income 1,720 2,367 4,900 -13,124 9,894 2,070
Dividend -785 -1,257 -1,200 -1,300 -340
Mandatory convertible and buybacks -1,300 -600 300 0
FX and other 1,910 -526 -2,200 -7,100 -300 -1,500
Carried forward 20,065 22,638 38,400 18,576 28,494 28,724
Finally we show the track record of Swiss Re life re profits in 1999-2009. We note
that Swiss Re doest not report return on operating revenue (it stopped doing so in
2007) as this is not a very consistent or comparable guide to underlying profitability.
81
(44-20) 7325-9175 02 September 2010
We believe the measure that would best help understand profitability is ROE for the
life re unit alone, and none of the European listed reinsurers provide this.
Table 54: Swiss Re reported life profits in SFm

1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Life premiums 7311 8330 8922 11275 10229 10205 9638 10974 12665 11090 10679
Life fees 881 879 955 808 916
Life claims -6119 -7448 -8502 -10084 -9085 -9331 -8669 -9594 -11112 -9065 -9348
Acquisition costs -1732 -1912 -2077 -2582 -2479 -2177 -2221 -2256 -2665 -2626 -2488
Investment income 2494 2983 3903 3448 3140 3311 5799 5353 6304 -1374 6654
Of which non- 1839 2530 3388 3476 3085 3178 3249 3016 4692 2632 2678
participating inv inc.
Of which non- 312 165 354 -229 -545
participating gains
Unit linked investment 9150 3019 2827 2120 -2822 4823
income/ gains and
losses
Other -741 -587 -704 -764 -983 -1308 -958 -844
Life operating profit 1505 1447 1682 1316 1643 1645 1645 1682 1320 697 746
Life operating result 850 994 1167 1344 1163 1222 1333 1381 2744 926 1291
ex gains
Allocated of -48 -208 -184 -135 -105 -165 -213 -452 -301 -368
overheads
Life operating result ex gains 946 959 1160 1028 1117 1168 1168 2292 625 923
after costs pretax
Life reserves 30339 40432 53851 51043 51041 54687 77851 102353 106723 87137 79223
Of which unit linked 1688 1964 1943 1799 2448 12629 14692 42834 41340 34518 37931
Estimate of capital required (4% 4856 5583 6098 5745 5262 6830 7682 8771 7139 6023
trad, 1% UL, 2X health prems)
Net of tax and of 14.6% 12.9% 14.3% 13.4% 15.9% 12.8% 11.4% 19.6% 6.6% 11.5%
central costs ROE
Return on operating 9.3% 9.2% 9.5% 9.1% 8.7% 9.1% 9.6% 9.2% 16.5% 6.4% 9.0%
revenue
Return on reserves 2.8% 2.5% 2.2% 2.6% 2.3% 2.2% 1.7% 1.3% 2.6% 1.1% 1.6%
Return on premiums 11.4% 10.7% 10.3% 10.0% 10.9% 12.1% 10.6% 18.1% 5.6% 8.6%
after allocation
Source: Company reports and J.P. Morgan estimates for expenses allocation and capital allocated
82
(44-20) 7325-9175 02 September 2010
Life reinsurance – risk products and

financing reinsurance
The life reinsurance product is one of two kinds: either: a) risk reinsurance or, b)
Financing reinsurance. The Risk reinsurance product makes money for the reinsurer
due to an information advantage, as the reinsurer can see data across several primary
insurers. The financing reinsurance product helps the primary insurer meet
acquisition costs, or as an alternative to securitization markets, particularly for
smaller insurers.
In this section we include a summary of discussions with Hannover Re and discuss in

some more detail the various kinds of life reinsurance products like YRT, Modco
(and the accounting of the B36 derivative).
Risk Reinsurance
The purpose of the buyer here is to transfer insurance risk to a third party, in this case
a reinsurer. These are mostly biometric risk factors – mortality, longevity etc. The
primary insurer may feel that they are too exposed to mortality risk so enters into a
reinsurance agreement, either in an excess of loss or a quota share basis.
Here the risk/reward profile is quite simple to understand – the reinsurer will
determine its price for the risks assumed in-line with mortality risks in-line with
other mortality exposures etc. Let’s assume the reinsurer charges 100 and the
expected claims cost is 90 if the expected claims cost materializes here then earns
10% of premium. If mortality is higher then the margin is less, and vice versa.
Reinsurers have established a fair amount of extra data and clients are in many cases
smaller companies, so there is an information advantage. They can compare the data
of one company with data in another company in same market which gives data
greater credibility. Primary insurers are also benefiting from a significant solvency
relief from the business reinsured, and that’s because the solvency margin is on the
net retained portfolio, not on the assumed written portfolio.
Risk reinsurance profits tend to tail off due to selection effect

In risk reinsurance, the mortality profits or risk profits of the reinsurers tail off over
the years and this has to do with the underwriting or selection process of the primary
insurance company. If we consider the case of a male aged 40 and let’s assume with
a typical mortality of 100% (i.e. average of all persons who have reached aged 40). If
the reinsurer then underwrites a person who is just turning 40 then the primary
insurer would have weeded out those who are about to die quickly. So the mortality
on the people who are reinsured is better than others, and this comes through.
YRT
One of the widely used methods of risk transfer reinsurance, particularly in the US
market is the Yearly Renewable Term (YRT) product.
The first concept we define here is the 'Net Amount at Risk': which is basically the
difference of the value of claim (in event of death) over the reserve which the
83
(44-20) 7325-9175 02 September 2010
insurance company has set up. The YRT agreement splits this net amount at risk
between the insurance company and the reinsurer.
This product is called the Yearly Renewable Term as the ceding company prepares a
schedule of ‘net amount at risk’ for every year of the policy and accordingly the
reinsurer develops a schedule of the premiums it will charge ever year.
It is interesting to note here that although the net amount at risk does decline steadily
every year, the insurer and the reinsurer agree to adjust the premium only at agreed
intervals to keep costs under control.
Coinsurance
The Coinsurance basis of reinsurance is the most comprehensive cover, as it transfers
all risks (either proportional or excess of loss basis). Typically, the reinsurer receives
a premium, and sets up a reserve, and then pays not just its share of claim, but also
any other benefits, along with commissions.
Coinsurance with funds withheld

Here the ceding company retains the assets which back the reserves of the ceded
portion of business and pays interest to the reinsurer on this book of reserves. There
could be two reasons why a ceding company chooses to retain the reserves.
a) it can attempt to maximise its investment income

b) this reduces its counterparty credit risk (removes the risk the reinsurer may not be
able to pay the claim when due).
Modified Coinsurance
The key difference between modified coinsurance (Modco) and the coinsurance with
funds withheld (above) is that the reinsurer only pays the ceding company for any
change in reserves after deducting the interest it would have otherwise received, thus,
implying that the ceding company is responsible for any changes in reserves simply
due to interest earned while all other changes belong to the reinsurer.
Interestingly, this ‘reserve adjustment’ interest rate is a key part of the negotiations
between the reinsurer and the cedant.
MODCO treaties were in focus when Hannover last year reported large unrealised
gains (and losses) on these treaties and recently changed the basis of calculation to
CDS from previous Option adjusted swaps (OAS) which did indeed reduce the
volatility on this line.
These unrealized gains or losses are due to the recognition of an ‘embedded

derivative’ and appear in the P&L due to the famous ‘B 36’ accounting rule
(Statement 133 Implementation Issue No. B36) which we discuss briefly below:
We discussed above that the interest rate that the ceding company pays to the
reinsurer on the ceded reserves is part of the negotiation and is in the reinsurance
agreement. This interest rate is typically based on a specified portion of the ceding
company’s return on its assets (or a specific book of cedant's assets).
84
(44-20) 7325-9175 02 September 2010
Thus the ability of the ceding company to pay interest (from the reinsurer's point of
view) is not linked directly to the risk profile of the ceding company, but rather on
the investment return the ceding company can earn on its assets. This introduces a
disconnect, which leads to the formation of the embedded derivative.
UK Enhanced annuities market

The UK protection market is the largest European reinsurance market, and the
primary insurers tender and re-tender their products every 12-18 months, which
means that the barriers to entry for re-insurers into this market are very low, which
means you get competition from new players, and even incumbents have become
aggressive recently.
Enhanced annuities market – Hannover Re is the only company which has credible
data. This covers only 25% of the population that is purchasing an annuity. This is
the population with the shortest longevity. They can predict clusters of annuities
ranging from illnesses, territorial distribution. Longevity in the UK is very different
in the North compared to the South. This is single premium immediate annuities –
the customer hands over a lump sum to the insurer and the price is quoted jointly by
the primary insurer with the reinsurer. There is no risk of lapse or persistency. Also
there is no risk of anti-selection going forward (in that good risks leave the portfolio
and bad risks remain). Not participating in the investment risk at all – in most cases
have arrangement with clients that guarantee certain interest rate.
In Germany the primary insurers charge 54% above actual mortality experience.
Further, on the accounting of life reinsurance, we flag off a key difference between
IFRS and US GAAP:
In IFRS you are forced every year to compare actual to expected and to immediately
reflect any deviation in your balance sheet and you have to prove to auditors that the
DAC asset is a real asset. Under US GAAP, if you receive 100 premium, you expect
90 of mortality but real claims experience in first year is 95, you are still able to show
this treaty with a profit of 10, because the underlying assumption in US GAAP is
correct and any short term deviation is just a random fluctuation.
Financing reinsurance
The demand here is driven not by consideration of offloading insurance risk but here
it is to manage the financials. Managing financials has a number of dimensions:
protect your solvency; protect your liquidity; protect your earnings under either
statutory or IFRS. Whatever they do companies have to include transfer of
underlying risk in order for it to qualify as insurance.
The typical financing reinsurance treaty is where the reinsurer helps the primary
company to meet the acquisition costs. You assume a life company is producing new
individual life business and this business creates a very substantial strain in the
balance sheet and also in the liquidity. The easiest to understand is the liquidity strain
– agents etc want cash up-front. There are also substantial internal acquisition costs –
policy issuance, compliance, etc – that only occur in Year 1. There are also solvency
costs – this triggers solvency requirements that have to be financed. This creates a
very substantial acquisition cost in Year 1. In this context, the primary insurer looks
for assistance from a reinsurer and in many of these
85
(44-20) 7325-9175 02 September 2010
The reinsurer participates in the cost structure of the primary insurance company.
The reinsurer puts money on the table to finance the acquisition of a life insurance
portfolio. They help the primary company to finance the acquisition of such a
portfolio. They share the initial expenses. This portfolio will generate future
earnings.
These cash flows are recovered pro-rata with the primary insurer. Many of the treaties
have terms and conditions that allow them to recover more than their share in the early
years. The main risk factor in financing business is the persistency risk – the key is to
keep the product on the books. We believe this is why lapse risk appears as a
significant factor in the embedded value sensitivities of the life reinsurers.
Financing reinsurance treaties also have a tailing off effect like the risk treaties, as is
evident from the IFRS profits of a financial re treaty, sourced from Hannover Re’s
Investor Day 2007.
Figure 27: IFRS profits of financing reassurance treaty

5.0
4.0
4.0 3.6
3.3
3
2.7 2.5
3.0 2.2 2.1
1.9
2.0 1.6 1.4
1.2 1.1
0.9 0.7
1.0 0.6 0.5 0.4
0.3 0.3 0.3 0.2 0.2 0.2 0.1
0.0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
IFRS result
If we compare financing reinsurance to alternative forms of ‘monetising' like

securitising, we see that the small and medium sized insurers (which make up the
bulk of the number of insurers worldwide) find it easier to use a financial reinsurance
product vs. tying up with the securitisation markets.
Block transfers
Another product of the financing reinsurance is the Block assumption transfer (BAT)
where Hannover assumes existing blocks of policies which are closed to new
business via quota share reinsurance. Hannover has indicated that demand for
monetizing the EV in this manner continues to remain strong and even in 2Q 10
helped towards the strong 7% growth seen in forex adjusted life re premiums.
We have shown below the example of the IFRS result of a block assumption
transaction. Year 3 and following years will look similar to the second year.
Table 55:Hannover’s example of block assumption transaction on traditional individual Life biz
from Germany, IFRS
Year 1 Year 2 Year 3+
Gross written premium 93.7 85.7 Similar to Year 2
Reinsurance commission (74.2) (1.6) Similar to Year 2
Change in DAC 64.1 (7.0) Similar to Year 2
IFRS result 4.0 3.6 Similar to Year 2
Source: Company reports (Hannover Re investor day presentation 2007 slide 122)
86
(44-20) 7325-9175 02 September 2010
Risk business is almost twice as profitable as Financing re

Overall, from a profitability point of view, we look at the IFRS new business margin,
as defined by Hannover: (IFRS results year 0 + present value of IFRS results)/ (IFRS
premiums year 0 + PV of IFRS premiums) and note that the risk reinsurance has a
9.8% margin vs. a 4.24% (Hannover Re data). Hannover does point out that its
Financial re business has an IRR of 11.5% so is still a profitable product.
Appendix I: Executive summary

• We believe life re has been under-researched by the market and under-reported in
terms of granularity and transparency by the listed European reinsurers. We
believe this is because by comparison with non-life re, life re is complex and
opaque.
• Non-life reinsurance is immediate and transparent: this year’s renewals affect
profits this and next year, negative claims trends have to be reserved
immediately, nat cats have an immediate effect, and because it is part a broker
market, pricing is relatively transparent. And finally the main annual reinsurance
conference is in glamorous Monte-Carlo.
• By contrast, life reinsurance is like treacle: long lasting, slow to change (new
business accounts for 3-10% of this year's profits), and with sticky accounting
(IFRS and US GAAP reserves are mainly historic). But, and this is the main
positive, in our view, in the main mortality business life re rarely produces losses.
This is because mortality tends to improve with time, so pricing errors tend to be
offset by positive variances, and we assume that companies have been reasonably
conservative in their pricing assumptions.
• The one exception to this is asset risk on the invested assets, which we believe
was a key factor in the upset at Swiss Re in 2008-9, which we believe caused
Scottish Re’s losses and forced closure, and which causes unusual volatility in
results on a quarterly basis elsewhere (eg ModCo at Hannover).
• This report goes through the main forms of life re, the different accounting and
reporting rules (IFRS, US GAAP, embedded value and statutory), the main trends
of each of the main life reinsurers in Europe. As a benchmark for reporting,
trends and profitability, we have used the example of RGA in the US, which is
the only company to report the main segments of life re: mortality, longevity,
asset intensive, Canada, other markets.
• Given our view that life re is a less volatile business, we should argue that life re
should be revalued. The difficulty is that investors who are used to non-life
reinsurance cycles, reporting and accounting, may be unwilling to pay up for life
reinsurance embedded value, particularly if there is the perception that life re
earnings are based on reserves which under different accounting would be
restated. We believe for example that the valuation discount of say Swiss Re
arising in life re may need a catalyst to be crystallized.
87
(44-20) 7325-9175 02 September 2010
Life reinsurance market shares

Munich Re & Swiss Re has a combined market share of 48% (2009 data) and they
dominate the life reinsurance market. The US accounts for 50% of the gross
premiums written and is the largest and most developed market.
Figure 28: Life reinsurance market share, 2009: dominated by top 2 Figure 29: US is the major market for life reinsurance, 2007
players
% %
Munich Re
27%
Sw iss Re 21%
RGA 12% 27%
Hannov er Re 12%
SCOR 8% 7%
GenRe 6%
Transameric 5% 10%
XL Re 1%
Partner Re 1% 6%
Other 7% 50%
0.0% 5.0% 10.0% 15.0% 20.0% 25.0% 30.0%
Life reinsurance - Global market share USA Germany UK Canada Other
Source: Munich Re life investor day presentation Source: Munich Re Life investor day presentation, 2008
Key points on the life re: We have shown below the US market share of the recurring new business in 2007.
This is a concentrated market with top 5 players controlling the 85% of the market.
1) Lower earnings volatility: Life
reinsurance has lower earnings Figure 30: Market share - US recurring new business assumed (2009)
volatility than Non life reinsurance. %
However the asset risk could be RGA Re. Company
22.4%
high, for example 85% Scottish Re Sw iss Re 19.3%
Transamerica Re 18.1%
earnings fluctuations were due to Munich Re (US) 13.5%
asset risk. Generali USA Life Re 11.8%
Hannov er Life Re 3.3%
Canada Life 3.2%
2) Relatively opaque accounting: SCOR Global Life (US) 2.9%
Life reinsurance accounting is Ace Tempest 1.7%
relatively opaque in comparison General Re Life 1.7%
Wilton Re 1.2%
with Non life Re. Optimum Re (US) 0.3%
RGA Re. Canada 0.1%
3) Diversification: Life re provides Employ ers Re Corp 0.1%
attractive diversification due to low
correlation of mortality & other life 0.0% 5.0% 10.0% 15.0% 20.0% 25.0%
risks with Non life. Market share - US recurring new business assumed
Source: Munich Re Investor day presentation on Life reinsurance
88
(44-20) 7325-9175 02 September 2010
Valuation Methodology and Risks

Hannover Re (Overweight; Price Target €45.70)
Valuation Methodology
We establish a Dec 11 SoTP based PT of €45.7. Our valuation is based on valuation
multiples of 8.7x for P&C and 9.1x for L&H representing a cost of equity of 11.5%
for P&C (1/11.5% = 8.7x) and 11.0% for Life (1/11% = 9.1x). We have changed the
cost of equity assumption for the life reinsurance unit to 11.0% vs. previously 10.0%
to reflect in our view investor concerns about impact of life reinsurance business
from low interest rates. We highlight we use net profit including realised gains which
we believe is appropriate given the low equity holdings at Hannover. We deduct the
face value of debt from the total value for our valuation.
Risks to Our View

Key downside risk to our OW rating and PT is continued softness in reinsurance
pricing leading to a deterioration in combined ratio in 2011 and 2012. Every 1%
worse 2012e COR impacts our PT by -5.6% (current Dec 11 PT of €45.7). Another
potential risk is continued pressure on earnings from nat cat losses, eg the increased
loss of €89m for the Deepwater Horizon oil rig.
SCOR (Overweight; Price Target €21.40)
Our sum of parts valuation uses the same methodology as for peers. We value non-
life using an 11.5% cost of capital which is equivalent to an 8.7x valuation PE, and
life 11.0% (as it is slightly lower risk) and this is equivalent to an 9.1x valuation PE.
We exclude from the non-life earnings we value the realised gains (we leave them in
the life valuation as our forecast is based on a total sustainable life margin) of we
forecast 6.65% as pct of net earned premiums (1H10 was 6.0%, part depressed by the
lower margin equity indexed annuity business). We deduct the debt at face value, and
we deduct both the €499m sub debt and the €196m financial debt, ie €695m in total.
Risks to Our View

The downside risks to our overweight recommendation are mainly two. The first is
deal risk, as SCOR has said that it is looking to build or buy in the Lloyds market,
and this could potentially could lead to dilution risk (albeit relatively modest as we
estimate SCOR has potential €0.55bn internal financing capacity, consisting of
€0.3bn potential extra debt and €0.25bn we estimate excess capital). The second is
that the non-life combined ratio disappoints due to high European natural
catastrophes, despite the improvement in pricing due to the January and July 2010
renewals.
Swiss Re (Neutral; Price Target SF 57.00)
Valuation: New SF57 Dec 11 SoTP based price target (previously SF61 Dec
2010)
Our Dec 2011 SOTP based price target is SF57. Our sum-of-parts valuation values
our forecast 2012E earnings on P/E valuation multiples of 8.3x for cyclical
reinsurance, life reinsurance and asset management. We also deduct -SF600m pre
tax, $430m net of tax (@95.5% FX rate and 25% tax rate) as the Berkshire
redemption penalty from our valuation. The valuation is based on our forecast
89
(44-20) 7325-9175 02 September 2010
operating earnings excluding realised gains and realised losses. The P/E multiple of
8.3x assumes a 12.0% cost of capital.
Risks to Our View

The upside risk is that Swiss Re remains very undervalued in terms of price to book
value of we estimate SF78 per share year end 2010 with a 39% discount and that it
starts a buyback earlier than the for now indicated earliest deadline of March 2011.
The downside risks are that non-life combined ratio, the competitive edge of the
group (we show in the chart below how Swiss Re is consistently since 2003 at or
below Munich Re), misses the new below 93% guidance going forward. The other
downside risk is that life reinsurance triggers a large assumption provision, which we
estimate could be up to around $1.6bn. .
90
(44-20) 7325-9175 02 September 2010
Companies Recommended in This Report (all prices in this report as of market close on 31 August 2010, unless
otherwise indicated)
Hannover Re (HNRGn.DE/€34.87/Overweight), SCOR (SCOR.PA/€17.16/Overweight), Swiss Re (RUKN.VX/SF 43.55
[01-September-2010]/Neutral)
Analyst Certification:
The research analyst(s) denoted by an “AC” on the cover of this report certifies (or, where multiple research analysts are primarily
responsible for this report, the research analyst denoted by an “AC” on the cover or within the document individually certifies, with
respect to each security or issuer that the research analyst covers in this research) that: (1) all of the views expressed in this report
accurately reflect his or her personal views about any and all of the subject securities or issuers; and (2) no part of any of the research
analyst’s compensation was, is, or will be directly or indirectly related to the specific recommendations or views expressed by the
research analyst(s) in this report.
Important Disclosures
• Market Maker/ Liquidity Provider: J.P. Morgan Securities Ltd. is a market maker and/or liquidity provider in Hannover Re,
SCOR, Swiss Re.
• Lead or Co-manager: J.P. Morgan acted as lead or co-manager in a public offering of equity and/or debt securities for Hannover Re
within the past 12 months.
• Beneficial Ownership (1% or more): J.P. Morgan beneficially owns 1% or more of a class of common equity securities of Swiss
Re.
• Client of the Firm: Hannover Re is or was in the past 12 months a client of JPM; during the past 12 months, JPM provided to the
company investment banking services, non-investment banking securities-related service and non-securities-related services. SCOR
is or was in the past 12 months a client of JPM; during the past 12 months, JPM provided to the company investment banking
services, non-investment banking securities-related service and non-securities-related services. Swiss Re is or was in the past 12
months a client of JPM; during the past 12 months, JPM provided to the company investment banking services, non-investment
banking securities-related service and non-securities-related services.
• Investment Banking (past 12 months): J.P. Morgan received, in the past 12 months, compensation for investment banking services
from Hannover Re, SCOR, Swiss Re.
• Investment Banking (next 3 months): J.P. Morgan expects to receive, or intends to seek, compensation for investment banking
services in the next three months from Hannover Re, SCOR, Swiss Re.
• Non-Investment Banking Compensation: JPMS has received compensation in the past 12 months for products or services other
than investment banking from Hannover Re, SCOR, Swiss Re. An affiliate of JPMS has received compensation in the past 12
months for products or services other than investment banking from Hannover Re, SCOR, Swiss Re.
Hannover Re (HNRGn.DE) Price Chart
Date Rating Share Price Price Target

N €37.7 N €41.3 N €24.1 (€) (€)
60
15-Nov-06 N 32.85 38.10
N €38.8 OW €43.3 N €23.2 N €38.9 OW €45.7 19-Dec-06 N 34.48 38.80
48
19-Mar-07 N 32.30 37.70
N €38.1 N €43.3 N €38.4 UW €26.5
UW €26.9
UW €36.4 OW €43.4 06-Jul-07 N 36.53 43.30
Price(€) 36 10-Jul-07 OW 36.18 43.30
05-Oct-07 N 36.71 41.30
24 11-Aug-08 N 29.60 38.40
29-Oct-08 N 15.70 23.20
12 04-Dec-08 N 19.60 24.10
16-Mar-09 UW 23.55 26.50
03-Jul-09 UW 26.38 26.90
0
Sep Jun Mar Dec Sep Jun
19-Oct-09 UW 33.92 36.40
06 07 08 08 09 10 09-Nov-09 N 33.15 38.90
05-Mar-10 OW 33.01 43.40
Source: Bloomberg and J.P. Morgan; price data adjusted for stock splits and dividends.
This chart shows J.P. Morgan's continuing coverage of this stock; the current analyst may or may not have covered it 05-May-10 OW 32.60 45.70
over the entire period.
J.P. Morgan ratings: OW = Overweight, N = Neutral, UW = Underweight.
91
(44-20) 7325-9175 02 September 2010
SCOR (SCOR.PA) Price Chart

40 OW €23.638 OW €23.3 N €21.1 (€) (€)
18-Oct-06 N 19.45 2.20
32 OW €2.45 OW €20 OW €18 N €21.4 N €23 14-Nov-06 OW 20.89 2.45
12-Feb-07 OW 21.73 23.64
N €2.2 OW €18.7 OW €16.2
OW €19.6 OW €21.4 N €22.6 OW €21.4 11-Mar-08 OW 13.89 18.70
24
Price(€) 25-Mar-08 OW 14.75 20.00
29-Oct-08 OW 11.04 16.20
16
17-Nov-08 OW 14.79 18.00
04-Dec-08 OW 14.93 23.30
8 12-Feb-09 OW 15.55 19.60
10-Jul-09 OW 14.58 21.40
07-Sep-09 N 18.20 21.40
0
13-Oct-09 N 18.96 21.10
06 07 08 08 09 10 15-Feb-10 N 17.45 22.60
04-Mar-10 N 18.45 23.00
Break in coverage Feb 26, 2004 - Oct 17, 2006. This chart shows J.P. Morgan's continuing coverage of this stock; the 27-Aug-10 OW 16.28 21.40
current analyst may or may not have covered it over the entire period.
Swiss Re (RUKN.VX) Price Chart

N SwF131 OW SwF130 OW SwF85
UW SwF26
OW SwF48 OW SwF57 (SwF) (SwF)
185
21-Nov-06 N 105.50 117.00
N SwF122 OW SwF141 OW SwF90
UW SwF55 N SwF37
OW SwF53 20-Feb-07 N 103.80 122.00
148
01-Mar-07 N 107.50 131.00
N SwF117 N SwF137 OW SwF120OW SwF66 N SwF28
OW SwF50 OW SwF61
N SwF57 09-Jul-07 N 111.10 137.00
Price(SwF) 111 05-Oct-07 OW 106.50 141.00
20-Nov-07 OW 85.10 130.00
74 06-May-08 OW 87.65 120.00
12-Aug-08 OW 70.25 90.00
37 10-Oct-08 OW 43.02 85.00
21-Oct-08 OW 47.50 66.00
25-Nov-08 UW 44.00 55.00
0
06-Feb-09 UW 18.60 26.00
06 07 08 08 09 10 09-Apr-09 N 23.20 28.00
08-May-09 N 38.52 37.00
This chart shows J.P. Morgan's continuing coverage of this stock; the current analyst may or may not have covered it 13-May-09 OW 38.68 48.00
over the entire period.
12-Aug-09 OW 45.26 50.00
25-Aug-09 OW 47.90 53.00
04-Nov-09 OW 45.10 57.00
22-Feb-10 OW 48.70 61.00
06-Aug-10 N 48.94 57.00
Explanation of Equity Research Ratings and Analyst(s) Coverage Universe:

J.P. Morgan uses the following rating system: Overweight [Over the next six to twelve months, we expect this stock will outperform the
average total return of the stocks in the analyst’s (or the analyst’s team’s) coverage universe.] Neutral [Over the next six to twelve
months, we expect this stock will perform in line with the average total return of the stocks in the analyst’s (or the analyst’s team’s)
coverage universe.] Underweight [Over the next six to twelve months, we expect this stock will underperform the average total return of
the stocks in the analyst’s (or the analyst’s team’s) coverage universe.] J.P. Morgan Cazenove’s UK Small/Mid-Cap dedicated research
analysts use the same rating categories; however, each stock’s expected total return is compared to the expected total return of the FTSE
All Share Index, not to those analysts’ coverage universe. A list of these analysts is available on request. The analyst or analyst’s team’s
coverage universe is the sector and/or country shown on the cover of each publication. See below for the specific stocks in the certifying
analyst(s) coverage universe.
92
(44-20) 7325-9175 02 September 2010
Coverage Universe: Michael Huttner, CFA: Allianz (ALVG.DE), CNP (CNPP.PA), Euler Hermes (ELER.PA), Generali
Deutschland (GE1G.DE), MLP (MLPG.F), Munich Re (MUVGn.DE), OVB Holding AG (O4BG.F), PZU (PZU.WA),
SCOR (SCOR.PA), Swiss Re (RUKN.VX), Zurich Financial Services (ZURN.VX)
J.P. Morgan Equity Research Ratings Distribution, as of June 30, 2010

Overweight Neutral Underweight
(buy) (hold) (sell)
J.P. Morgan Global Equity Research 46% 42% 12%
Coverage
IB clients* 49% 46% 31%
JPMS Equity Research Coverage 44% 48% 9%
IB clients* 68% 61% 53%
*Percentage of investment banking clients in each rating category.
For purposes only of FINRA/NYSE ratings distribution rules, our Overweight rating falls into a buy rating category; our Neutral rating falls into a hold
rating category; and our Underweight rating falls into a sell rating category.
Valuation and Risks: Please see the most recent company-specific research report for an analysis of valuation methodology and risks on
any securities recommended herein. Research is available at http://www.morganmarkets.com , or you can contact the analyst named on
the front of this note or your J.P. Morgan representative.
Analysts’ Compensation: The equity research analysts responsible for the preparation of this report receive compensation based upon
various factors, including the quality and accuracy of research, client feedback, competitive factors, and overall firm revenues, which
include revenues from, among other business units, Institutional Equities and Investment Banking.
Registration of non-US Analysts: Unless otherwise noted, the non-US analysts listed on the front of this report are employees of non-US
affiliates of JPMS, are not registered/qualified as research analysts under FINRA/NYSE rules, may not be associated persons of JPMS,
and may not be subject to FINRA Rule 2711 and NYSE Rule 472 restrictions on communications with covered companies, public
appearances, and trading securities held by a research analyst account.
Other Disclosures
J.P. Morgan ("JPM") is the global brand name for J.P. Morgan Securities LLC ("JPMS") and its affiliates worldwide. J.P. Morgan Cazenove is a
marketing name for the U.K. investment banking businesses and EMEA cash equities and equity research businesses of JPMorgan Chase & Co.
and its subsidiaries.
Options related research: If the information contained herein regards options related research, such information is available only to persons who
have received the proper option risk disclosure documents. For a copy of the Option Clearing Corporation’s Characteristics and Risks of
Standardized Options, please contact your J.P. Morgan Representative or visit the OCC’s website at
http://www.optionsclearing.com/publications/risks/riskstoc.pdf.
Legal Entities Disclosures
U.S.: JPMS is a member of NYSE, FINRA and SIPC. J.P. Morgan Futures Inc. is a member of the NFA. JPMorgan Chase Bank, N.A. is a
member of FDIC and is authorized and regulated in the UK by the Financial Services Authority. U.K.: J.P. Morgan Securities Ltd. (JPMSL) is a
member of the London Stock Exchange and is authorized and regulated by the Financial Services Authority. Registered in England & Wales No.
2711006. Registered Office 125 London Wall, London EC2Y 5AJ. South Africa: J.P. Morgan Equities Limited is a member of the Johannesburg
Securities Exchange and is regulated by the FSB. Hong Kong: J.P. Morgan Securities (Asia Pacific) Limited (CE number AAJ321) is regulated
by the Hong Kong Monetary Authority and the Securities and Futures Commission in Hong Kong. Korea: J.P. Morgan Securities (Far East) Ltd,
Seoul Branch, is regulated by the Korea Financial Supervisory Service. Australia: J.P. Morgan Australia Limited (ABN 52 002 888 011/AFS
Licence No: 238188) is regulated by ASIC and J.P. Morgan Securities Australia Limited (ABN 61 003 245 234/AFS Licence No: 238066) is a
Market Participant with the ASX and regulated by ASIC. Taiwan: J.P.Morgan Securities (Taiwan) Limited is a participant of the Taiwan Stock
Exchange (company-type) and regulated by the Taiwan Securities and Futures Bureau. India: J.P. Morgan India Private Limited is a member of
the National Stock Exchange of India Limited and Bombay Stock Exchange Limited and is regulated by the Securities and Exchange Board of
India. Thailand: JPMorgan Securities (Thailand) Limited is a member of the Stock Exchange of Thailand and is regulated by the Ministry of
Finance and the Securities and Exchange Commission. Indonesia: PT J.P. Morgan Securities Indonesia is a member of the Indonesia Stock
Exchange and is regulated by the BAPEPAM LK. Philippines: J.P. Morgan Securities Philippines Inc. is a member of the Philippine Stock
Exchange and is regulated by the Securities and Exchange Commission. Brazil: Banco J.P. Morgan S.A. is regulated by the Comissao de Valores
Mobiliarios (CVM) and by the Central Bank of Brazil. Mexico: J.P. Morgan Casa de Bolsa, S.A. de C.V., J.P. Morgan Grupo Financiero is a
member of the Mexican Stock Exchange and authorized to act as a broker dealer by the National Banking and Securities Exchange Commission.
Singapore: This material is issued and distributed in Singapore by J.P. Morgan Securities Singapore Private Limited (JPMSS) [MICA (P)
93
(44-20) 7325-9175 02 September 2010
020/01/2010 and Co. Reg. No.: 199405335R] which is a member of the Singapore Exchange Securities Trading Limited and is regulated by the
Monetary Authority of Singapore (MAS) and/or JPMorgan Chase Bank, N.A., Singapore branch (JPMCB Singapore) which is regulated by the
MAS. Malaysia: This material is issued and distributed in Malaysia by JPMorgan Securities (Malaysia) Sdn Bhd (18146-X) which is a
Participating Organization of Bursa Malaysia Berhad and a holder of Capital Markets Services License issued by the Securities Commission in
Malaysia. Pakistan: J. P. Morgan Pakistan Broking (Pvt.) Ltd is a member of the Karachi Stock Exchange and regulated by the Securities and
Exchange Commission of Pakistan. Saudi Arabia: J.P. Morgan Saudi Arabia Ltd. is authorized by the Capital Market Authority of the Kingdom
of Saudi Arabia (CMA) to carry out dealing as an agent, arranging, advising and custody, with respect to securities business under licence number
35-07079 and its registered address is at 8th Floor, Al-Faisaliyah Tower, King Fahad Road, P.O. Box 51907, Riyadh 11553, Kingdom of Saudi
Arabia. Dubai: JPMorgan Chase Bank, N.A., Dubai Branch is regulated by the Dubai Financial Services Authority (DFSA) and its registered
address is Dubai International Financial Centre - Building 3, Level 7, PO Box 506551, Dubai, UAE.
Country and Region Specific Disclosures
U.K. and European Economic Area (EEA): Unless specified to the contrary, issued and approved for distribution in the U.K. and the EEA by
JPMSL. Investment research issued by JPMSL has been prepared in accordance with JPMSL's policies for managing conflicts of interest arising
as a result of publication and distribution of investment research. Many European regulators require a firm to establish, implement and maintain
such a policy. This report has been issued in the U.K. only to persons of a kind described in Article 19 (5), 38, 47 and 49 of the Financial Services
and Markets Act 2000 (Financial Promotion) Order 2005 (all such persons being referred to as "relevant persons"). This document must not be
acted on or relied on by persons who are not relevant persons. Any investment or investment activity to which this document relates is only
available to relevant persons and will be engaged in only with relevant persons. In other EEA countries, the report has been issued to persons
regarded as professional investors (or equivalent) in their home jurisdiction. Australia: This material is issued and distributed by JPMSAL in
Australia to “wholesale clients” only. JPMSAL does not issue or distribute this material to “retail clients.” The recipient of this material must not
distribute it to any third party or outside Australia without the prior written consent of JPMSAL. For the purposes of this paragraph the terms
“wholesale client” and “retail client” have the meanings given to them in section 761G of the Corporations Act 2001. Germany: This material is
distributed in Germany by J.P. Morgan Securities Ltd., Frankfurt Branch and J.P.Morgan Chase Bank, N.A., Frankfurt Branch which are
regulated by the Bundesanstalt für Finanzdienstleistungsaufsicht. Hong Kong: The 1% ownership disclosure as of the previous month end
satisfies the requirements under Paragraph 16.5(a) of the Hong Kong Code of Conduct for Persons Licensed by or Registered with the Securities
and Futures Commission. (For research published within the first ten days of the month, the disclosure may be based on the month end data from
two months’ prior.) J.P. Morgan Broking (Hong Kong) Limited is the liquidity provider for derivative warrants issued by J.P. Morgan Structured
Products B.V. and listed on the Stock Exchange of Hong Kong Limited. An updated list can be found on HKEx website:
http://www.hkex.com.hk/prod/dw/Lp.htm. Japan: There is a risk that a loss may occur due to a change in the price of the shares in the case of
share trading, and that a loss may occur due to the exchange rate in the case of foreign share trading. In the case of share trading, JPMorgan
Securities Japan Co., Ltd., will be receiving a brokerage fee and consumption tax (shouhizei) calculated by multiplying the executed price by the
commission rate which was individually agreed between JPMorgan Securities Japan Co., Ltd., and the customer in advance. Financial Instruments
Firms: JPMorgan Securities Japan Co., Ltd., Kanto Local Finance Bureau (kinsho) No. 82 Participating Association / Japan Securities Dealers
Association, The Financial Futures Association of Japan. Korea: This report may have been edited or contributed to from time to time by
affiliates of J.P. Morgan Securities (Far East) Ltd, Seoul Branch. Singapore: JPMSS and/or its affiliates may have a holding in any of the
securities discussed in this report; for securities where the holding is 1% or greater, the specific holding is disclosed in the Important Disclosures
section above. India: For private circulation only, not for sale. Pakistan: For private circulation only, not for sale. New Zealand: This
material is issued and distributed by JPMSAL in New Zealand only to persons whose principal business is the investment of money or who, in the
course of and for the purposes of their business, habitually invest money. JPMSAL does not issue or distribute this material to members of "the
public" as determined in accordance with section 3 of the Securities Act 1978. The recipient of this material must not distribute it to any third
party or outside New Zealand without the prior written consent of JPMSAL. Canada: The information contained herein is not, and under no
circumstances is to be construed as, a prospectus, an advertisement, a public offering, an offer to sell securities described herein, or solicitation of
an offer to buy securities described herein, in Canada or any province or territory thereof. Any offer or sale of the securities described herein in
Canada will be made only under an exemption from the requirements to file a prospectus with the relevant Canadian securities regulators and only
by a dealer properly registered under applicable securities laws or, alternatively, pursuant to an exemption from the dealer registration requirement
in the relevant province or territory of Canada in which such offer or sale is made. The information contained herein is under no circumstances to
be construed as investment advice in any province or territory of Canada and is not tailored to the needs of the recipient. To the extent that the
information contained herein references securities of an issuer incorporated, formed or created under the laws of Canada or a province or territory
of Canada, any trades in such securities must be conducted through a dealer registered in Canada. No securities commission or similar regulatory
authority in Canada has reviewed or in any way passed judgment upon these materials, the information contained herein or the merits of the
securities described herein, and any representation to the contrary is an offence. Dubai: This report has been issued to persons regarded as
professional clients as defined under the DFSA rules.
General: Additional information is available upon request. Information has been obtained from sources believed to be reliable but JPMorgan
Chase & Co. or its affiliates and/or subsidiaries (collectively J.P. Morgan) do not warrant its completeness or accuracy except with respect to any
disclosures relative to JPMS and/or its affiliates and the analyst’s involvement with the issuer that is the subject of the research. All pricing is as
of the close of market for the securities discussed, unless otherwise stated. Opinions and estimates constitute our judgment as of the date of this
material and are subject to change without notice. Past performance is not indicative of future results. This material is not intended as an offer or
solicitation for the purchase or sale of any financial instrument. The opinions and recommendations herein do not take into account individual
client circumstances, objectives, or needs and are not intended as recommendations of particular securities, financial instruments or strategies to
particular clients. The recipient of this report must make its own independent decisions regarding any securities or financial instruments
mentioned herein. JPMS distributes in the U.S. research published by non-U.S. affiliates and accepts responsibility for its contents. Periodic
updates may be provided on companies/industries based on company specific developments or announcements, market conditions or any other
94
(44-20) 7325-9175 02 September 2010
publicly available information. Clients should contact analysts and execute transactions through a J.P. Morgan subsidiary or affiliate in their home
jurisdiction unless governing law permits otherwise.
“Other Disclosures” last revised September 1, 2010.
Copyright 2010 JPMorgan Chase & Co. All rights reserved. This report or any portion hereof may not be reprinted, sold or
redistributed without the written consent of J.P. Morgan.#$J&098$#*P
95

European Re Insurance

Transféré par

Informations du document

Description originale:

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

European Re Insurance

Transféré par

Droits d'auteur :

Formats disponibles

Europe Equity Research

Split of risk by segment: Example of Munich Re

Forecast 2011e net profit 2,473 1,934 648 478

Ratio market to 2011e net profit:

Table 5: Reinsurers - Split of JPMe price target by line of business

Summary split of business

Table 6: Reinsurers - Split of 2010e operating profit by line of business

Table 7: Reinsurance business split by premiums, 2010e

High level summary with references

Relative valuations of the remainder look attractive

Low level of transparency for Life Re in Europe, particularly compared with

Asset risk of Life Re

Munich Re: positive interest sensitivity in life re

SCOR: possible venture into longevity

Nature of the life reinsurance market

• Long lasting acquisition process for new business of 6 to 12 months

Five types of risk

Financial risk / wherever there is an accumulation of reserves to be invested.

Protection against catastrophes / pandemics. This is a risk which can partially be

Accounting risk. Reinsurers in their presentations understate this risk, possibly

There are three main points we believe about accounting risk:

a) There is no agreed measure of profitability under IFRS, as for example the

b) Transparency is relatively low because life re is shown normally just as one

Extra source of accounting noise. a further contrary element of financial hedging

Summary of Scottish Re profit and loss record 2005-1H10 –

The following chart shows the experience of Scottish Re in 2005-1H10.

Figure 1: Scottish Re profit/loss and shareholders' equity 2005-1H10

The key events are as follows:

Table 8: Scottish Re Net income and shareholders' equity 2005-1H10

Table 9: Scottish Re Summary investments in $m

Table 11: Scottish Re Adverse mortality - benefit ratios

Table 13: Scottish Re Summary operating results quarterly in $m

Table 14: Scottish Re Funding of the collateral requirements in $m

In 2005 and 2006 Scottish Re completed three securitisations that collectively

Source: Company reports

Source: Company reports, JPMorgan estimates for description of SFAS113 B36.

Life reinsurance is relatively stable (apart from asset risk),

1. Business reason for higher barriers to entry. Life reinsurance is negotiated

Figure 2: Munich Re - market share and cession rate

Cession rate Munich Re market share

2. Accumulation reason. Life reinsurance contracts typically run for up to thirty

Distributable cashflow IFRS profits EV profits

By comparison, in non-life reinsurance the reserves for outstanding claims are

Comparison of the accounting standards

FASB60 for long duration contracts of traditional life

Insurance contracts, for purposes of this Statement, need to be classified as short-

Premiums from short-duration contracts ordinarily are recognized as revenue over

Investments are reported as follows: common and nonredeemable preferred stocks at

“Premiums from long-duration contracts shall be recognized as revenue when due

when premium revenue is recognized. Those estimates shall be based on

“This Statement establishes standards of accounting for certain long-duration

retrospective deposit method be used to account for universal life-type contracts.

We have shown below an illustrative example of the computation of the estimated

Table 17: Illustration of the computation of estimated gross profit

Table 18: Computation of amortization rate

Life re is more stable than non-life except for asset risk

ASSET RISK IS THE MAIN SOURCE OF VOLATILITY IN LIFE RE

Source: Company reports.

More numbers on earnings volatility of life re vs non-life re – it is consistently

SCOR Sw iss Re Hannov er Re Munich Re

SCOR Sw iss Re Hannov er Re Munich Re

Table 21: SCOR - Earnings summary

Table 22: Swiss Re - Earnings summary