1. Homeowners Insurance: The National Market and Trends Residential Property Insurance Symposium Charlotte, NC May 31, 2013 Steven N. Weisbart, Ph.D., CLU, Senior Vice President & Chief Economist Insurance Information Institute  110 William Street  New York, NY 10038 Tel: 212.346.5540  Cell: 917.494.5945  stevenw@iii.org  www.iii.org

2. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 2 P/C Net Written Premiums, by Segment/Line, 2002 vs. 2012 Sources: A.M. Best; Insurance Information Institute research. Commercial Lines $205.6B/46% 2012 Pvt. Pass Auto $171.8B/39% Homeowners $66.6B/15% 2002 Commercial Lines $188.1B/50% Homeowners $40.3B/11% Pvt. Pass Auto $146.9B/39% Homeowners insurance NWP grew from 11% to 15% of total P/C premiums, growing faster than commercial insurance NWP (which shrank, as a percent of total P/C NWP, from 50% to 46%).

3. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 3 Homeowners Insurance Net Written Premium, 2000–2012 Sources: SNL Financial; Insurance Information Institute. $ Billions Homeowners insurance NWP resumed rising in 2010 after very low growth in 2007-09. Continued growth depends on exposure growth, the composition of the market (renters vs. owners), and other factors. Pct. Change

4. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 4 Average Premium for Home Insurance Policies** * Insurance Information Institute Estimates/Forecasts **Excludes state-run insurers. Source: NAIC, Insurance Information Institute estimates 2010-2012 based on CPI and other data. Home insurance premiums are rising in response to higher catastrophe losses

5. 5 Homeowners Insurance Profit, 2011, by State Sources: NAIC 2012 Profitability Report; Insurance Information Institute. Highest 25 States In 2011, Homeowners insurance produced double-digit percentage profits in just 13 states US average: -3.8%

6. 6 Homeowners Insurance Profit, 2011, by State Sources: NAIC 2012 Profitability Report; Insurance Information Institute. Lowest 25 States US average: -3.8%

7. 7 *Latest available. Source: NAIC. Return on Net Worth: Homeowners Insurance, 10-Year Average (2002-2011*) Hawaii was the most profitable state for home insurers from 2002-2011 due to the absence of hurricanes during this period (Percent) Top 25 States

8. 8 *Latest available. Sources: NAIC Hurricanes Katrina and Rita made Louisiana and Mississippi the least profitable states for home insurers from 2002-2011 Bottom 25 States (Percent) Return on Net Worth: Homeowners Insurance, 10-Year Average (2002-2011*)

9. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 9 Monthly Change* in HO-4 Insurance Prices, 1998–2013 *Percentage change from same month in prior year; through April 2013; not seasonally adjusted; first BLS data point is Dec 1998; Recessions indicated by gray shaded columns. Sources: US Bureau of Labor Statistics; National Bureau of Economic Research (recession dates); Insurance Information Institute. “Hard” market of 2002-03 A price change peak, Nov 2010, at a +4.3% rate Since 1998, the 12-month change in HO-4 prices has rarely been above 3% (and from Oct 2005 through Sept 2006 was negative!) Latest (Apr 2013) at 2.4%

10. 10 Homeowners Average Premium 2011: Highest 25 States Sources: NAIC 2012 Profitability Report; Insurance Information Institute.

11. 11 Homeowners Average Premium 2011: Lowest 25 States Sources: NAIC 2012 Profitability Report; Insurance Information Institute.

12. U.S. Residual Market: Total Policies In-Force at Year-End (1999-2011) Source: PIPSO; Insurance Information Institute Thousands of Policies in Force 4 Florida Hurricanes Katrina, Rita and Wilma In the 10-year period 2001-2011, the total number of policies in force in the residual market (FAIR & Beach/Windstorm Plans) roughly tripled. Gustav, Ike Irene

13. 13 U.S. Residual Market Exposure to Loss Source: PIPSO; Insurance Information Institute (I.I.I.). ($ Billions) In the 12-year period 2000 through 2011, total exposure to loss in the residual market (FAIR & Beach/Windstorm Plans) quadrupled (from $221.3 billion in 2000 to a record high of $884.7 billion in 2011). 4 Florida Hurricanes Katrina, Rita and Wilma

14. Homeowners Combined Ratios, 1993–2012F The industry-wide, nationwide Homeowners Combined Ratio since 1993 has been under 100 only 4 times in 20 years. Sources: A.M. Best (1990-2012E); Insurance Information Institute. 12/01/09 - 9pm 14

15. 15 Return on Net Worth: All P-C Lines vs. Homeowners, 1990-2011* *Latest available. **Excluding Hurricane Andrew (1992); Including 1992 produces an average homeowners RNW of 0.5%. Sources: NAIC; Insurance Information Institute. Average RNW: 1990-2011* All P-C Lines: 7.8% Homeowners: 3.2%** Homeowners Insurance Is Considerably More Volatile than the Market Overall Due to Coastal Exposure and Interior Wind/Hail Events Hurricane Irene Katrina, Rita, Wilma Texas “Mold” Crisis

16. Number of Insurers Competing Total Market Share of Top 5 Insurers HHI US 40046.7%668.9 North Carolina 6358.8%896.1 South Carolina 6855.2%943.6 Sources: SNL Financial; Insurance Information Institute The Marketplace for HO Premiums is Quite Competitive, 2012 Note: An HHI score under 1,000 is considered a competitive marketplace. Example: In a market, the top 5 insurers’ market share is 16%, 14%, 12%, 10%, and 8%. Total market share of these insurers is 60%. The HHI score for the state is likely to be 900-1000.

17. 17 Demographic and Macroeconomic Forces Affecting Homeowners Insurance Brighter Days Ahead, but Not Without Challenges

18. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 18 Projected Population Growth Rates (2010-2020) Vary Widely by State and Region* *based on 2000 census. Source: http://www.census.gov/population/projections/data/state/projectionsagesex.html (Table 7)http://www.census.gov/population/projections/data/state/projectionsagesex.html Southeast Mid-Atlantic New England U.S. population growth overall, 2010-2020, is projected to be 8.7%

19. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 19 Projected Population Growth Rates (2010-2020) Vary Widely by State and Region* (cont’d) *based on 2000 census. Source: http://www.census.gov/population/projections/data/state/projectionsagesex.html (Table 7)http://www.census.gov/population/projections/data/state/projectionsagesex.html SouthwestMountainFar WestGreat PlainsGreat Lakes U.S. population growth overall, 2010-2020, is projected to be 8.7%

20. 12/01/09 - 9pm 20 As the Population Ages, the Number of People per Household Shrinks (2012) Source: US Census Bureau at http://www.census.gov/hhes/families/data/cps2012.html Table AVG1.http://www.census.gov/hhes/families/data/cps2012.html As the “baby boom” ages and households get smaller, this will spur growth of smaller homes that are more suited to their requirements. Average Household Size Age of Householder The number of households in these age brackets will soar in the next 20 years

21. 12/01/09 - 9pm 21 Millions Changes in Household Composition, 2000-2010 Sources: U.S. Census Bureau, “Households and Families: 2010,” issued April 2012.; Insurance Information Institute. The number of traditional households (husband, wife, children at home) fell by 1.25 million in the first decade of the 21 st century. Multi-generation households rose by roughly the same number in that decade. Down 5% Up 14.4% Up 14.6% Up 30.8% Up 14.7%

22. 12/01/09 - 9pm 22 Millions of Units Private Housing Unit Starts, 1990-2014F Sources: U.S. Department of Commerce; Blue Chip Economic Indicators (5/13); Insurance Information Institute. Homeowners insurers are starting to see meaningful exposure growth for the first time since 2005. Commercial insurers with construction risk exposure, surety also benefit. Starts plunged 72% from 2005-2009 to lowest level since records began in 1959 Forecast range for 2013 is 0.90 to 1.35 million units Housing “Bubble”

23. So Far, the Pickup Is Mostly in Multi-Family Housing Starts *average of annualized seasonally adjusted January-April 2013 data; April is preliminary. Source: US Census Bureau at www.census.gov/construction/nrc/pdf/newresconst.pdf.www.census.gov/construction/nrc/pdf/newresconst.pdf Thousands of Units, Multi-Family 2013:Q1 multi-unit starts at a seasonally adjusted annual rate of 325,000, are nearly back to the average annual pre-recession rate of 339,000. Multi-family-unit starts rose in 2011, more in 2012, still more so far in 2013. Thousands of Units, Single Family Multi-family plunge did not begin until 2009 Single family plunge began in 2006

24. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 24 Number of Rental-Occupied Housing Units, Quarterly, 1990-2013 Sources: US Census Bureau; Insurance Information Institute. Trough in 2004:Q4 at 33.1 million units Since the Great Recession ended, renters occupied 3.6 million more units (+9.9%)—outstripping population growth (+2.9%). When will this end? 12/01/09 - 9pm 24 Millions Latest was 40.1 million units in 2013:Q1

25. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 25 Rental Vacancy Rates, Quarterly, 1990-2013 Sources: US Census Bureau; Insurance Information Institute. Peak vacancy rate 11.1% in 2009:Q3 Before the 2001 recession rental vacancy rates were 8% or less. Can they get there again? 12/01/09 - 9pm 25 Percent vacant Latest vacancy rate was 8.6% in 2013:Q1

26. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 26 Rental-Occupied Housing Units as % of Total Occupied Units, Quarterly, 1990-2013 Sources: US Census Bureau; Insurance Information Institute. Trough in 2004:Q2 and Q4 at 30.8% Since the Great Recession ended, renters occupied 3.6 million more units (+9.9%)—outstripping population growth (+2.9%). When will this end? 12/01/09 - 9pm 26 Millions Latest was 35.0% in 2013:Q1 Trend down began in 1994:Q3 from 36.2% in Q2

27. Housing Unit Starts: Building Momentum, Monthly, Jan 2011-Apr 2013* *at annualized rate, seasonally adjusted; Apr 2013 numbers are preliminary. Source: US Census Bureau at www.census.gov/construction/nrc/pdf/newresconst.pdf.www.census.gov/construction/nrc/pdf/newresconst.pdf Thousands of Units The number of units in multi-unit starts more than doubled from Dec 2011 to Dec 2012. Single family start rose nicely, too.

28. Average Square Footage of Completed New Homes in U.S., 1973-2011 Square Ft The trend toward building larger homes resumed in the second half of 2010. The average size of completed new homes often falls in recessions (yellow bars), but historically bounces back in expansions 28 Sources: U.S. Census Bureau at http://www.census.gov/construction/chars/pdf/medavgsqft.pdf and “Households and Families: 2010”, Census Brief issued April 2012; Insurance Information Institutehttp://www.census.gov/construction/chars/pdf/medavgsqft.pdf The number of householders living alone rose by nearly 4 million from 2000 to 2010 (+14.6%). Average household size and average family size didn’t change over that decade. Will we see a shift to smaller houses?

29. Highly Variable P/C Claims Drivers 29

30. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 30 Change* in the Consumer Price Index, 2004–2013 *Monthly, year-over-year, through April 2013. Not seasonally adjusted. Sources: US Bureau of Labor Statistics; National Bureau of Economic Research (recession dates); Insurance Information Institutes. Over the last decade, prices generally rose about 2% per year. 12/01/09 - 9pm 30 For two months in 2008, led by gasoline, the general price level was rising at a 5.5% pace But when gas prices dropped, the general price level was briefly lower than a year prior

31. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 31 Pct Changes in Price* for Residential Construction, 1990–2013 *Monthly vs. same month, prior year. Note: Recessions indicated by gray shaded columns. Sources: BLS Producer Price index ; National Bureau of Economic Research (recession dates); Insurance Information Institutes. Three times in the past decade the price of residential construction rose at a 7.5% annual rate, or more. Such spikes were probably not in the base rates. 12/01/09 - 9pm 31

32. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 32 Prices for Hospital Services: 12-Month Change,* 1998–2013 *Percentage change from same month in prior year; through April 2013; seasonally adjusted Sources: US Bureau of Labor Statistics; National Bureau of Economic Research (recession dates); Insurance Information Institute. Cyclical peaks in PP Auto tend to occur approximately every 10 years (early 1990s, early 2000s, and possibly the early 2010s) April 2013 Inpatient services +4.0, Outpatient services +4.5%

33. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 33 Change* in Price Index for Lumber: Sudden Spikes, 2008–2013 *Monthly, year-over-year, through April 2013. Softwood is seasonally adjusted; hardwood is not. January through April 2013 prices are preliminary. Sources: US Bureau of Labor Statistics, Producer Price Index series WPS0811 (softwood); WPU812 (hardwood). National Bureau of Economic Research (recession dates); Insurance Information Institutes. 12/01/09 - 9pm 33 The price of lumber dropped before and during the recession… But 30% spikes can happen suddenly Jul 2012 to Apr 2013

34. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 34 Change* in Price Index for Lumber: A Downward Trend but Sudden Spikes, 2004–2013 12/01/09 - 9pm 34 The price of lumber dropped before and during the recession… But 30% spikes can happen with no warning *Monthly, year-over-year, through April 2013. Softwood is seasonally adjusted; hardwood is not. January through April 2013 prices are preliminary. Sources: US Bureau of Labor Statistics, Producer Price Index series WPS0811 (softwood); WPU812 (hardwood). National Bureau of Economic Research (recession dates); Insurance Information Institutes.

35. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 35 Change* in Price Index for Plywood: A Recent Downward Trend but Spikes, 2004–2013 *Monthly, year-over-year, through April 2013. Not seasonally adjusted. Jan. through Apr 2013 prices are preliminary. Sources: US Bureau of Labor Statistics, Producer Price Index series WPU083; National Bureau of Economic Research (recession dates); Insurance Information Institutes. From the end of the recession (June 2009) to March 2013, the effect of the ups and downs of the price of plywood has resulted in a rise of 25.6%. 12/01/09 - 9pm 35 The price of plywood rose by 50% in late Spring 2004 over the prior year March 2013: +8.4%

36. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 36 Price Index for Waferboard, Monthly 2008–2013 Through April 2013. Not seasonally adjusted. Jan through Apr price changes are preliminary. Sources: US Bureau of Labor Statistics, Producer Price Index series WPU09220124; National Bureau of Economic Research (recession dates); Insurance Information Institutes. The prices of building materials such as waferboard and strandboard vary wildly and change levels rapidly. 12/01/09 - 9pm 36 In May 2010 the price of waferboard doubled (vs. May 2009) but then subsided The price of waferboard rose by 80% again in Apr 2013 (vs. 2012); will it drop again?

37. Catastrophes 37 12/01/09 - 9pm 37

38. Joplin Home Damage Claims Consumed a Large Share of MO’s Homeowners Premiums Sources: SNL Securities (Homeowners DPW for MO); Catastrophe loss data is from PCS as of May 10, 2012; Insurance Information Inst. ($ Millions) Insurers ROE in the MO Homeowners Line in 2011 Will be Negative by Several Hundred Percent 38 Catastrophe losses paid to homeowners arising from the May 22 Joplin tornado consumed 44% of all homeowners premiums written throughout the state in 2011

39. Sources: SNL Securities (Homeowners DPW for MO); Catastrophe loss data is from PCS as of May 10, 2012; Insurance Information Inst. ($ Millions) Insurers ROE in the MO Property/Casualty Insurance Market in 2011 Will be Very Low, If Not Negative 39 Catastrophe losses paid to Missouri policyholders arising from the May 22 Joplin tornado consumed 24% of premiums written for all p/c lines throughout the state in 2011 Joplin Tornado Claims Consumed a Large Share of MO’s P/C Insurance Premiums

40. 12/01/09 - 9pm 40 The Dozen Most Costly Hurricanes in U.S. History Insured Losses, 2012 Dollars, $ Billions *Estimate as of 12/09/12 based on estimates of catastrophe modeling firms and reported losses as of 1/12/13. Estimates range up to $25B. Sources: PCS; Insurance Information Institute inflation adjustments to 2012 dollars using the CPI. Sandy will likely become the 3 rd costliest hurricane in US insurance history Irene became the 12 th most expensive hurricane in US history 10 of the 12 most costly hurricanes in insurance history occurred in the past 9 years (2004—2012)

41. 12/01/09 - 9pm 41 If They Hit Today, the Dozen Costliest (to Insurers) Hurricanes in U.S. History Insured Losses, 2012 Dollars, $ Billions *Estimate as of 12/09/12 based on estimates of catastrophe modeling firms and reported losses as of 1/12/13. Estimates range up to $25B. Sources: Karen Clark & Company, Historical Hurricanes that Would Cause $10 Billion or More of Insured LossesToday, August 2012; I.I.I. When you adjust for the damage prior storms could have done if they occurred today, Hurricane Katrina slips to a tie for 6 th among the most devastating storms. Storms that hit long ago had less property and businesses to damage, so simply adjusting their actual claims for inflation doesn’t capture their destructive power. Karen Clark’s analysis aims to overcome that.

42. Hurricane Sandy resulted in an estimated 1.4 million privately insured claims resulting in an estimated $15 to $25 billion in insured losses. Hurricane Katrina produced 1.74 million claims and $47.6B in losses (in 2011 $) Superstorm Sandy: Number of Claims by Type* *PCS claim count estimate as of 11/26/12. Loss estimate represents high and low end estimates by risk modelers RMS, Eqecat and AIR. PCS estimate of insured losses as of 11/26/12 $11 billion. All figures exclude losses paid by the NFIP. Source: PCS; AIR, Eqecat, AIR Worldwide; Insurance Information Institute. 12/01/09 - 9pm 42 Sandy is a high frequency, (relatively low) severity event (avg. severity <50% Katrina)

43. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 43 US Insured Catastrophe Losses *As of 1/2/13. Includes $20B gross loss estimate for Hurricane Sandy. Note: 2001 figure includes $20.3B for 9/11 losses reported through 12/31/01 ($25.9B 2011 dollars). Includes only business and personal property claims, business interruption and auto claims. Non-prop/BI losses = $12.2B ($15.6B in 2011 dollars.) Sources: Property Claims Service/ISO; Insurance Information Institute. US CAT Losses in 2012 Will Likely Become the 2 nd or 3 rd Highest in US History on An Inflation-Adjusted Basis (Pvt Insured). 2011 Losses Were the 5 th Highest 2012 CAT losses were down nearly 50% from 2011 until Sandy struck in late October Record Tornado Losses Caused 2011 CAT Losses to Surge ($ Billions, 2012 Dollars) 12/01/09 - 9pm 43

44. Number Geophysical (earthquake, tsunami, volcanic activity) Climatological (temperature extremes, drought, wildfire) Meteorological (storm) Hydrological (flood, mass movement) Natural Disasters in the United States, 1980 – 2012 Number of Events (Annual Totals 1980 – 2012) Source: MR NatCatSERVICE 44 41 19 121 3 There were 184 natural disaster events in the US in 2012 There were over 150 natural disaster events in the US every year since 2006. That hadn’t happened in any year before.

45. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 45 Share of Flood Damaged Structures with Flood Insurance: Long Island Source: Newsday, 1/14/13 from FEMA and Small Business Administration. The Maximum FEMA Grant is $31,900. The Average Grant Award to Homeowners and Renters on Long Island is About $7,300 52,428 $15.0 43,106 $2.2 12/01/09 - 9pm 45 74,736 20,798 5,747 Only 37.5% of flood damaged buildings in Nassau County were insured for flood, 62.5% uninsured 27.6% of flood damaged buildings in Suffolk County were insured for flood, 73.4% uninsured 15,051 Number of buildings

46. Source: Wharton Center for Risk Management and Decision Processes, Issue Brief, Nov. 2012; Insurance Information Institute. Residential NFIP Flood Take-Up Rates in NY, CT (2010) & Sandy Storm Surge 46 Flood coverage penetration rates were extremely low in many very vulnerable areas of NY and CT, with take-up rates far below 50% in many areas

47. P/C Industry Homeowners Claim Frequency, US, 1997-2011 Sources: Insurance Research Council, “Trends in Homeowners Insurance Claims,” p.29; Insurance Information Institute Claims Paid per 100 Exposures

48. P/C Industry Homeowners Average Claim Severity, 1997-2011 Sources: Insurance Research Council, “Trends in Homeowners Insurance Claims,” p. 29, BLS inflation calculator, and Insurance Information Institute HO average claim severity is now three times what it was in 1997.

49. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 49 U.S. Employment in Insurance Claims Adjusting, Monthly, 1990–2013* Thousands *As of March 2013; Seasonally adjusted. Note: Recessions indicated by gray shaded columns. Sources: U.S. Bureau of Labor Statistics; National Bureau of Economic Research (recession dates); Insurance Information Institute. Katrina, Rita, Wilma Gustav, Ike Sandy

50. Investments: The Stark Reality 50 12/01/09 - 9pm 50

51. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 51 U.S. Treasury Security Yields*: A Long Downward Trend, 1990–2013 *Monthly, constant maturity, nominal rates, through Mar 2013. Sources: Federal Reserve Bank at http://www.federalreserve.gov/releases/h15/data.htm. National Bureau of Economic Research (recession dates); Insurance Information Institutes.http://www.federalreserve.gov/releases/h15/data.htm Yields on 10-Year U.S. Treasury Notes have been essentially below 5% for a full decade. Since roughly 80% of P/C bond/cash investments are in 10-year or shorter durations, most P/C insurer portfolios will have low-yielding bonds for years to come. U.S. Treasury security yields recently plunged to record lows 12/01/09 - 9pm 51

52. 12/01/09 - 9pm 52 Insurers Have Not Yet Fully Adapted to a Persistently Low Interest Rate Environment They Didn’t Expect Rates to be  Pushed to Such Low Levels  Pushed Down so Rapidly  Held to Such Low Levels for So Long  Suppressed via Unprecedented Aggressiveness of the Federal Reserve Ability to Release Prior Reserves Eased Urgency OFFSETTING FACTORS Capitalization Still Solid Emergence of Sophisticated Price Monitoring and Underwriting Tools

53. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 53 Distribution of Bond Maturities, P/C Insurance Industry, 2003-2012 Sources: SNL Financial; Insurance Information Institute. The main shift over these years has been from bonds with longer maturities to bonds with shorter maturities. The industry first trimmed its holdings of over-10-year bonds (from 24.6% in 2003 to 15.4% in 2012) and then trimmed bonds in the 5-10-year category. Falling average maturity of the P/C industry’s bond portfolio is contributing to a drop in investment income along with lower yields.

54. Purchasing Power of P/C Industry Investment Gains: 1994–2012F 1 In 2012 both investment income and realized capital gains were lower than in the comparable period in 2011. And because the Federal Reserve Board aims to keep interest rates exceptionally low until the unemployment rate hits 6.5%—likely at least another year off—maturing bonds will be re-invested at even lower rates. 1 Investment gains consist primarily of interest, stock dividends and realized capital gains and losses. *2005 figure includes special one-time dividend of $3.2B; 2012F figure is I.I.I. estimate based on annualized actual 2012:Q3 result of $38.089B. Sources: ISO; Insurance Information Institute. ($ Billions, 2012 dollars) Average yearly gain: $60.85B. We haven’t hit that average in the last 5 years.

55. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 55 P/C Net Premiums Written: % Change, Quarter vs. Year-Prior Quarter, 2002–2012 Sources: ISO; Insurance Information Institute. Finally! A sustained period (10 quarters) of growth in net premiums written (vs. same quarter, prior year), and strengthening. This upward trend is likely to continue as the economy’s recovery strengthens Most recent “hard market”

56. P/C Net Income After Taxes 1991–2012:Q3 ($ Millions) 2005 ROE*= 9.6% 2006 ROE = 12.7% 2007 ROE = 10.9% 2008 ROE = 0.1% 2009 ROE = 5.0% 2010 ROE = 6.6% 2011 ROAS 1 = 3.5% 2012:Q3 ROAS 1 = 6.3% Through the first three quarters of 2012, P/C Industry profits were up 222% from the comparable period in 2011, mainly due to lower CAT losses in 2012:Q2 and Q3 * ROE figures are GAAP; 1 Return on avg. surplus. Excluding Mortgage & Financial Guaranty insurers yields a 6.2% ROAS for 2012:H1, 4.6% ROAS for 2011, 7.6% for 2010 and 7.4% for 2009. Sources: A.M. Best; ISO; Insurance Information Institute.

57. Profitability Peaks & Troughs in the P/C Insurance Industry, 1975 – 2012:Q3* *Profitability = P/C insurer ROEs. 2012 is an estimate based on ROAS data. Note: Data for 2008-2012 exclude mortgage and financial guaranty insurers. 2012:H1 ROAS = 5.9% including M&FG. Sources: Insurance Information Institute; NAIC; ISO; A.M. Best. 1977:19.0% 1987:17.3% 1997:11.6% 2006:12.7% 1984: 1.8% 1992: 4.5% 2001: -1.2% 10 Years 9 Years 2011: 4.6%* History suggests next ROE peak will be in 2016-2017 ROE 1975: 2.4% 2012:Q3: 6.3%

58. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 58 Policyholder Surplus, 2006:Q4–2012:Q3 Sources: ISO; A.M.Best. ($ Billions) Surplus as of 9/30/12 was a new peak The industry now has $1 of surplus for every $0.80 of NPW, the strongest claims-paying status in its history Drop due to near-record 2011 CAT losses

59. 59 Key Takaways 12/01/09 - 9pm 59

60. 12/01/09 - 9pmeSlide – P6466 – The Financial Crisis and the Future of the P/C 60 HO Insurance Exposures Will Grow With the U.S. Economy and Population  In 2013-14, more growth in rental market than ownership market  Longer term, demographics might continue to shift the composition of the market toward smaller homes, more rentals HO Underwriting Profit is Highly Variable, Depends on Frequency, Severity of CATs  Frequency, Severity trends appear to be rising  Inflation in some claims drivers has been/can be severe, even when overall inflation is moderate Industry Capacity Hits a New Record by Year-End 2013 (Barring Meg-CAT) Pressure on Underwriting to Generate Profits as the Investment Environment Is/Remains Challenging  Most economists forecast interest rates to remain low for the next few years Takeaways

61. www.iii.org Thank you for your time and your attention! Insurance Information Institute