2. Soft Markets and Hard Truths in the Property/Liability Insurance Industry Steven N. Weisbart, Ph.D., CLU, Senior Vice President and 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 Alabama I-Day 2009 University of Alabama Tuscaloosa, AL October 7, 2009

3. Hard Truths about the U.S. Economy Soft Markets in the Property-Liability Insurance Industry Hard Truths in Property-Liability Insurance Operations Hard Truths About Property-Liability Investment Performance & Insuring Capacity Q&A Presentation Outline

4. Hard Truths about the U.S. Economy Been Down So Long It Looks Like Up to Me

5. Length of U.S. Business Cycle Contractions, 1945-Present* * As of September 2009, inclusive Sources: National Bureau of Economic Research; Insurance Information Institute. Current recession began in Dec. 2007 and is already the longest since the Great Depression. Duration (Months) Month Recession Started

6. Length of U.S. Business Cycles, 1945-Present* * As of April 2009, inclusive Sources: National Bureau of Economic Research; Insurance Information Institute. Duration (Months) Month Recession Started

7. Real Quarterly GDP Changes (annualized), 2005:Q3-2010:Q4F Sources: US Department of Commerce, Bureau of Economic Analysis (actual) at http://www.bea.gov/newsreleases/national/gdp/gdpnewsrelease.htm Blue Chip Economic Indicators 9/09 issue (forecasts). http://www.bea.gov/newsreleases/national/gdp/gdpnewsrelease.htm Spike due almost entirely to the weak dollar (growing exports and slowing imports) Red bars are actual; Yellow bars are forecasts/estimates The Q1:2009 decline was the steepest since the Q1:1982 drop of 6.4%

8. Employment

9. U.S. Nonfarm Private Employment, Monthly, Nov 2007 – August 2009 Seasonally adjusted. Source: US Bureau of Labor Statistics Millions The U.S. economy lost about 6 million jobs in 12 months Employment peak; recession starts

10. January 2000 through August 2009, seasonally adjusted U-6 went from 9.2% in April 2008 to 16.8% in Aug. 2009 Source: US Bureau of Labor Statistics; Insurance Information Institute. 9.7% Aug. 2009 unemployment rate (U-3) was the highest monthly rate since 1983. Peak rate in the last 30 years: 10.8% in Nov-Dec 1982. Unemployment and Underemployment Rates: Rocketing Up in 2008-9 Percent

11. U.S. Unemployment Rate Forecasts Quarterly, 2009:Q3 to 2010:Q4 Sources: Blue Chip Economic Indicators (9/09); Insurance Info. Inst. Unemployment is expected to peak in late 2009:Q4 or 2010:Q1. Rising unemployment will erode payrolls and workers comp’s exposure base.

12. Has the Contraction Ended and Recovery Begun? Source: http://www.calculatedriskblog.com/http://www.calculatedriskblog.com/

13. Will the Recession End Soon? Feb-Sep 2009 Initial Jobless Claims* *seasonally adjusted; state programs only Source: http://www.dol.gov/opa/media/press/eta/ui/current.htmhttp://www.dol.gov/opa/media/press/eta/ui/current.htm

14. Housing

15. Source: http://www.realtor.org/research/research/ehsdatahttp://www.realtor.org/research/research/ehsdata High Ratio of Unsold-Homes Inventory to Sales Will Likely Keep Prices Falling Millions of Homes, Annual Rate # of house sales fell; inventory wasn’t the problem # of house sales is rising but so is inventory

16. “ Shadow” Inventory of Unsold Homes: It’s Worse Than You Think “About 1.8 million homes are currently in foreclosure and they will continue to weigh on home prices for at least the rest of the year.” Zillow.com’s latest Homeowner Confidence Survey (published August 18, 2009) asked homeowners how likely they would put their homes on the market if they saw signs of a turnaround in the next 12 months:  Very likely, 8% (7.5 million homes)  Likely, 9% (7.5 million homes)  But Adam York, economist for Wells Fargo Securities, “contends that the amount of homes that have not yet been listed for sale could be around 4-5 million. Sources: Julie Haviv, “’Shadow’ inventory lurks over U.S. housing recovery,” Reuters, at http://www.reuters.com/article/GCA-Housing/idUSTRE56U5YZ20090731 and http://zillow.mediaroom.com/index.php?s=173 http://www.reuters.com/article/GCA-Housing/idUSTRE56U5YZ20090731 http://zillow.mediaroom.com/index.php?s=173

17. High Inventory Means Millions Fewer Private Housing Starts Measured by number of new units started, exposure growth for HO insurers is low. Housing start data also affects commercial insurers with construction risk exposure. I.I.I. estimate: each 100,000 decline in housing starts “costs” home insurers $90 million in gross premium. Estimated premium loss in 2008 vs. 2005: about $1 billion. Sources: US Department of Commerce; Blue Chip Economic Indicators (9/09); Insurance Information Inst. Millions of Units Housing bubble Recession

18. The “Silver Lining”? Inflation Might Be Relatively Low

19. After Recent Recessions, the Annual Inflation Rate Dropped Sources: US Bureau of Labor Statistics (actual, blue bars); Blue Chip Economic Indicators, 9/09 issue, (forecasts, yellow bars) Average inflation rate, 1992-2007: 2.67% Post- Recession

20. Continued high levels of unemployment/ underemployment Continued low levels consumer demand/ business investment  Significant increase in the personal saving rate, but virtually all used to pay down outstanding debt  Low- or no-growth in insured exposures Low levels of new borrowing  Affects housing, autos, other consumer durables Section Summary: Hard Truths About the U.S. Economy

21. Soft Markets in the Property/Liability Insurance Industry Primarily Driven by the Industry’s Underwriting Cycle, Not the Economy

22. Sources: A.M. Best, ISO, Insurance Information Institute 40 Years of Hard and Soft Markets 1975-781984-872000-03 Shaded areas denote “hard market” periods In 2007 net written premiums fell, the first decline since 1943

23. Year-to-Year Change in Net Written Premium, 2000-2009* Sources: A.M. Best (historical through 2008; ISO for 2009. *first quarter 2009 only P/C insurers are experiencing their slowest growth rates since 1943 Slow growth means retention is critical Protracted period of negative or slow growth is possible due to soft markets and slow economy

24. Does National Economic Growth Affect Real P/C Premium Growth? Not Much Sources: A.M. Best, US Bureau of Economic Analysis, Blue Chip Economic Indicators, 2/09; Insurance Information Inst.

25. *9-month data for 2008 Source: US Bureau of Economic Analysis; Federal Reserve Bank of St. Louis at http://research.stlouisfed.org/fred2/series/WASCUR; I.I.I. Fact Books http://research.stlouisfed.org/fred2/series/WASCUR Even When the Payroll Base is Rising, Soft Markets Reduce Workers Comp Premiums 7/90-3/91 Shaded areas indicate recessions 3/01-11/01 Wage & Salary Disbursement (Private Employment) vs. WC NWP $ Billions 12/07-? Weakening wage and salary growth is expected to cause a deceleration in workers comp exposure growth

26. Hard Truths in the Property/Liability Insurance Operations Primarily Driven by the Industry’s Underwriting Cycle, Not the Economy

27. The P/C Industry’s Combined Ratio Seems Headed Up, Again Sources: A.M. Best, ISO; III preliminary estimates. The industry’s combined ratio appears to be on a “cyclical upturn” dating to 2006. In 2008, even excluding net CAT losses (which added 3.4 points to the combined ratio vs. 2007) and M&FG losses (another 4.1 points vs. 2007), the 2008 ratio would have been 97.6. Combined Ratio 09:Q1 combined ratio was 98.4 excl. M&FG vs. 96.8 in 08:Q1

28. Medical Cost Inflation is One of the Most Serious Long-term Challenges Facing Casualty, WC, and Health Insurers Source: Department of Labor (Bureau of Labor Statistics); not seasonally adjusted http://www.bls.gov/news.release/pdf/cpi.pdf *through July 2009 http://www.bls.gov/news.release/pdf/cpi.pdf Since 1982-84, (index =100) the cost of medical care has nearly quadupled (to 375), while the overall cost of living merely doubled (to 215)

29. WC Medical Severity Rising Far Faster than Medical CPI Sources: Med CPI from I.I.I. Inflation Watch; WC med severity from NCCI based on NCCI states.

30. Since 2003, Tort System Cost Growth Has Moderated but Could Spike Again Sources: US Bureau of Labor Statistics, Tillinghast-Towers Perrin, 2008 Update on U.S. Tort Costs; Insurance Info. Inst.

31. Catastrophic Losses*: Was 2005 an Outlier or a Harbinger? *Excludes $4B-$6b offshore energy losses from Hurricanes Katrina & Rita. **2009:1H Note: 2001 figure includes $20.3B for 9/11 losses reported through 12/31/01. Includes only business and personal property claims, business interruption and auto claims. Non-prop/BI losses = $12.2B. Source: Property Claims Service/ISO; Insurance Information Institute $ Billions 30 Is $25 billion the new level of expected yearly CAT losses? Before 2001, CAT losses averaged about $8-10 billion per year.

32. Source: http://edr.state.fl.us/conferences/population/demographic.htmhttp://edr.state.fl.us/conferences/population/demographic.htm Data are from Feb. 18, 2009 Florida Demographic Estimating conference A Million More Florida Resident Households in the Next Decade? Millions of Households The State of Florida now (Feb 09) forecasts nearly 1 million more households by 2019 (up almost 13%). There will be more businesses, too. Hurricane Andrew Hurricane Wilma

33. High Population Growth in Coastal Alabama Puts More Insured Value at Risk Source: South Alabama Regional Planning Commission, “Coastal Alabama River Basin Management Plan”, January 2004.

34. High Population Growth Forecast for Coastal Areas in Baldwin County, AL, 2000-2030 High population growth in coastal Alabama puts more insured value at risk

35. High Mobile County Population Growth Forecast 2000-2030 is Near but Not Directly on the Coast Source: South Alabama Regional Planning Commission, “Coastal Alabama River Basin Management Plan”, January 2004.

36. Major (Category 3, 4, 5) Hurricanes Striking the US by Decade *Figure for 2000s is extrapolated based on data for 2000-2008 (7 major storms: Charley, Ivan, Jeanne (2004), Katrina, Rita, Wilma (2005), Ike (2008)). Sources: Tillinghast from National Hurricane Center: http://www.nhc.noaa.gov/pastint.shtm.; I.I.I. http://www.nhc.noaa.gov/pastint.shtm Mid 1920s – mid-1960s: AMO Warm Phase Mid-1990s – 2030s? AMO Warm Phase Colorado State team forecasts 3 more intense hurricanes in 2009

37. Inflation-Adjusted U.S. Insured Catastrophe Losses By Cause of Loss, 1987-2006¹ Source: Insurance Services Office (ISO).. 1 Catastrophes are all events causing direct insured losses to property of $25 million or more in 2006 dollars. Catastrophe threshold changed from $5 million to $25 million beginning in 1997. Adjusted for inflation by the III. 2 Excludes snow. 3 Includes hurricanes and tropical storms. 4 Includes other geologic events such as volcanic eruptions and other earth movement. 5 Does not include flood damage covered by the federally administered National Flood Insurance Program. 6 Includes wildland fires. Insured disaster losses totaled $297.3 billion from 1987-2006 (in 2006 dollars). Hurricanes and tornadoes accounted for approximately 75% of these.

38. Number of Tornadoes in Each Calendar Quarter, 2005–2009:Q2 Sources: US Dept. of Commerce, Storm Prediction Center, National Weather Service, at http://www.spc.noaa.gov/climo/torn/monthlytornstats.pdf 2009:Q2 is I.I.I. estimatehttp://www.spc.noaa.gov/climo/torn/monthlytornstats.pdf The first two quarters of 2009 were more typical of prior years than 2008.

39. Underwriting Expense Ratios Likely Will Rise as Premium Growth Lags *Ratio of expenses incurred to net premiums written. Sources: A.M. Best; ISO; Insurance Information Institute 2009:Q1 blended ratio was 29.1%

40. The financial crisis/great recession didn’t directly affect P-L underwriting performance; the underwriting cycle and catastrophes were 2008’s drivers Premium levels will likely continue to be soft, driven by low exposure growth, low inflation The potential for significant catastrophe losses remains high Expense ratios will likely grow unless expenses can be reduced Section Summary: Hard Truths about P-L Insurance Operations

41. Hard Truths about Property/Liability Investment Performance and Insuring Capacity

42. Asset Allocation of P/C Insurance Industry’s Cash & Invested Assets Portfolio Facts Invested assets totaled $1.26 trillion as of 9/30/08 Allocations were virtually unchanged from 12/31/07 Source: SNL Financial LC As of September 30, 2008 41

43. Bond Yields Tend to Follow Inflation, but the Relationship is a Loose One Sources: US Bureau of Labor Statistics (history); Blue Chip Economic Indicators, 8/09 (forecasts) Forecast July 1990- March 1991 recession March 2001- November 2001 recession

44. P/C Investment Income as a % of Invested Assets Follows 10-Year U.S. T-Note Sources: history: Board of Governors, Federal Reserve System; A.M.Best; Insurance Information Institute. Forecasts: Office of Management and Budget, Mid-Session Review, Fiscal Year 2010. Investment yield historically tracks 10-year Treasury note quite closely OMB forecasts that T-note yields won’t exceed 5.2% for the next decade. If recent history holds, P-C net investment yield will be somewhat lower.

45. P/C Industry Investment Income*, 1994-2008 *Primarily interest and stock dividends. ** Investment income (excluding one-time dividend) jumped in 2005 as insurers that had accumulated cash captured rising bond interest rates. Also, 2005 figure includes special one-time dividend of $3.2B. ***2009 figure is Q1 actual, annualized Sources: ISO; Insurance Information Institute. Investment income might moderate further if rates for new bond investments stay low and/or if insurers shift to shorter-maturity bonds and more US government notes. Investment income CAGR 1994-2007 was just 3.8%.

46. P/C Industry Net Realized Capital Gains and Losses, 1990-2009:Q1 Sources: A.M. Best, ISO, Insurance Information Institute. Nearly $9 billion in realized capital gains in 2007, but $-19.7 billion in 2008. $ Billions

47. Policyholder Surplus by Quarter, 2006:Q4 – 2009:Q1 Source: ISO Decline Since 2007:Q3 Peak 2009Q1: -$84.7B (-16.2%)

48. U.S. P/C Industry Premiums-to- Surplus Ratio: 1985-2009:Q1 Sources: A.M. Best, ISO, Insurance Information Institute. 1998 0.85:1–the lowest (strongest) P:S ratio in recent history. Premiums are a rough measure of risk accepted; surplus is funds beyond reserves to pay unexpected losses. The larger surplus is in relation to premiums—the lower the ratio of premiums to surplus—the greater the industry’s capacity to handle the risk it has accepted. 1.03:1 as of 3/31/09 Ratio at year-end

49. Largest Capital Events as a Percent of Surplus, 1989-present Ratio is for end-of-quarter surplus immediately prior to event. Date shown is end of quarter prior to event. Sources: PCS; Insurance Information Institute. The financial crisis now ranks as the largest “capital event” over the past 20+ years

50. Premium-to-Surplus Ratios Before Major Capital Events* *Ratio is for end of quarter immediately prior to event. Date shown is end of quarter prior to event. **Latest available Source: PCS; Insurance Information Institute. P/C insurance industry was better capitalized going into the financial crisis than before any “capital event” in recent history

51. In 2008, A.M. Best Affirmed or Upgraded 88% of P/C Insurers* *Through December 19. Source: A.M. Best. 50 In 2008, despite financial market turmoil, high cat losses and a soft market, A.M. Best lowered ratings on just 3.9% of P-C insurers. It placed another 4.4% under review

52. Reasons for US P/C Insurer Impairments, 1969-2008 Source: A.M. Best: 1969-2008 Impairment Review, Special Report, Apr. 6, 2008 Deficient loss reserves and inadequate pricing are the leading cause of insurer impairments, underscoring the importance of discipline. Investment catastrophe losses play a much smaller role.

53. 1975: 2.4% 1977:19.0%1987:17.3% 1997:11.6% 2006:12.2% 1984: 1.8% 1992: 4.5% 2001: -1.2% 10 Years 9 Years Note: 2008 result excluding Mortgage & Financial Guarantee insurers is 4.2%. Sources: ISO; A.M. Best (2009F); Insurance Information Institute. 2008: 0.3% P/C Insurance Industry ROEs, 1975 – 2009F* 52

54. Even though the industry has had its largest “capital event,” it is still well positioned to conduct business as usual Yields on bonds are expected to continued at low levels  Nearly a decade of investing at these levels means they’re “baked in” to investment performance for a long time Section Summary: Hard Truths about P-L Investments & Capacity

55. Insurance Information Institute On-Line If you would like a copy of this presentation, please give me your business card with e-mail address