pricing insurance risk
TRANSCRIPT
Pricing Insurance RiskModule C: Historical US Property-Casualty Profitability and Volatility
Stephen J. MildenhallCreated 2021-11-15 16:30:43.766628
independent | informed | imaginative
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 1
Table of Contents
Module C Contents
Section C.1 Premium to GDP Ratio, 1923-2020Section C.2 Surplus to GDP Ratio, 1931-2020Section C.3 Industry Volatility and Profitability Metrics, 1985-2020Section C.4 Direct and Net Volatility and Profitability by Major Line, 1992-2020Section C.5 Direct Premium Growth by Major Line, 1992-2020Section C.6 Implications
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 2
C.01. Premium to GDP Ratio,1923-2020
Premium to GDP Ratio
1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
NW
P to
GD
P R
atio
Figure 1: The ratio of US statutory net written premium (NWP) to nominal gross domestic product (GDP) gives a good measureof the insurance market cycle. It measures the penetration of the insurance industry into the economy. GDP and NWP are bothare on a nominal basis; inflation cancels in the ratio. If GDP in 2020 had grown inline with 2019 over 2018, then premium toGDP ratio in 2020 would be unchanged. Source: NAIC US Statutory Compbined P&C Industry premium and GDP from FRED.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 3
Premium to GDP Ratio: Cyclical Growth Between 1947 and 1997
1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
NW
P to
GD
P R
atio
Figure 2: Secular growth from the end of WW2 through 1986 has given way to retrenchment and stabilization. The 1986watershed saw tax reform, which introduced loss reserve discounting for tax purposes, as well as the claims-made form andabsolute pollution exclusion. The period since 2008 is particularly stable.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 4
Premium to GDP Ratio: 1968 Watershed
1920 1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
NW
P to
GD
P R
atio
Great D
epression
Pearl Harbor
Oil C
risis
LMX Spriral
AndrewN
orthridge
WTC
KatrinaG
FC
HIM
After 1968, all years above 2.8%Until 1968, all years below 2.8%
Figure 3: The 2.8 percent level (green lines) separates 1968 and prior from all subsequent years. The ratio was less than 2.8percent in every year until 1968, while it has been greater than 2.8 percent in every year since. Significant insurance-relatedevents are called out on the x-axis.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 5
Annual Change in Statutory Surplus, 1932 to 2020
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
-30%
-20%
-10%
0%
10%
20%
30%
40%
Cha
nge
in S
urpl
us
1974
2001
2008Change in PHSmean change 0.0785Upper CI z=2.3, SD=0.102Lower CI
Figure 4: The three most significant post-1968 surplus declines have coincided with significant asset value declines: 1974, 2001and 2008. 2000-02 is the only consecutive three year decline.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 6
Annual Change in Statutory Surplus, 1932 to 2020 With 10- And 25-Year Rolling SD
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
-30%
-20%
-10%
0%
10%
20%
30%
40%
Cha
nge
in S
urpl
us
Change in PHSmean change 0.0785Upper CI z=2.3, SD=0.102
Lower CIRolling 10 year SDRolling 25 year SD
Figure 5: Rolling standard deviation reasonably steady over long-term history, and currently at historically low levels.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 7
Multi-Year Change in Statutory Surplus, 1932 to 2020
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
-20%
-10%
0%
10%
20%
30%
1 Ye
ar C
hang
e in
Sur
plus
1 Year Change in Surplus, cagr=7.8%
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
-20%
0%
20%
40%
2 Ye
ar C
hang
e in
Sur
plus
2 Year Change in Surplus, cagr=8.0%
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
-20%
0%
20%
40%
60%
80%
3 Ye
ar C
hang
e in
Sur
plus
3 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
20%
40%
60%
80%
100%
120%
4 Ye
ar C
hang
e in
Sur
plus
4 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
25%
50%
75%
100%
125%
150%
5 Ye
ar C
hang
e in
Sur
plus
5 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
6 Ye
ar C
hang
e in
Sur
plus
6 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
7 Ye
ar C
hang
e in
Sur
plus
7 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
250%
8 Ye
ar C
hang
e in
Sur
plus
8 Year Change in Surplus, cagr=8.0%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
250%
300%
9 Ye
ar C
hang
e in
Sur
plus
9 Year Change in Surplus, cagr=8.1%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
250%
300%
10 Y
ear
Cha
nge
in S
urpl
us
10 Year Change in Surplus, cagr=8.1%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
50%
100%
150%
200%
250%
300%
350%
11 Y
ear
Cha
nge
in S
urpl
us
11 Year Change in Surplus, cagr=8.2%
1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0%
100%
200%
300%
400%
500%
12 Y
ear
Cha
nge
in S
urpl
us
12 Year Change in Surplus, cagr=8.2%
Figure 6: Surplus doubles every nine years, on average. Current period shows historically low long-term growth rates. Notadjusted for inflation.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 8
Premium to Surplus Ratio, 1932 to 2020
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
0
0.25
0.5
0.75
1
1.25
1.5
1.75
2
2.25
2.5
2.75
3
NWP to Surplus RatioFive-year rolling average
Figure 7: Industry leverage ratios have been decreasing since their mid-1970s high of 2.7 to 1. They have been remarkablystable over the last decade. Current leverage ratios were also seen in the 1930s.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 9
Premium to GDP vs. Suprlus to GDP
1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 4.5%Surplus to GDP
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
Prem
ium
to G
DP
1968-69average
193119511968199220052020
Figure 8: Period since 1968, above the dotted line, has seen much greater swings in surplus to GDP (east west direction) thanthe earlier period. Current premium penetration is below historical highs but current surplus to GDP ratios are at historic highs.Note relatively high surplus levels in 1930s (yellow): the industry did not return to 1930s levels of relative capitalization until1995.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 10
Three Phases of the Market 1931-2020
1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 4.5%Surplus to GDP
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
Prem
ium
to G
DP
1931
19681969
1985
1986
20012020
Phase I
Phase IIPhase III
193119511968199220052020
Figure 9: Phase I, 1931-1968: establishment, depression, WW2, and recovery. Phase II, 1969-1985: growth and expansion,industry looking for ways to find coverage. Tax code and high interest rate favorable to casualty business. Phase III,1986-present: retrenchment and responsibility, secular decline in interest rates and revised tax code emphasize underwritingprofits, and growing casualty reserves increase capital intensity.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 11
Market Dynamics Since 1986 Explained by Prior Year Surplus Levels
1.0% 1.5% 2.0% 2.5% 3.0% 3.5% 4.0% 4.5%Prior Year-End Surplus to GDP
2.8%
3.0%
3.2%
3.4%
3.6%
3.8%
4.0%
Prem
ium
to G
DP
1986
1992
2001
2005
2009
2018
2020
198620012020
Figure 10: Regression of premium to GDP on prioryear surplus to GDP. Anticlockwise cycle between1994 and 2008!
Results: Ordinary least squares===================================================================Model: OLS Adj. R-squared: 0.862Dependent Variable: PG AIC: -365.3573Date: 2021-11-15 14:51 BIC: -362.2466No. Observations: 35 Log-Likelihood: 184.68Df Model: 1 F-statistic: 214.0Df Residuals: 33 Prob (F-statistic): 5.54e-16R-squared: 0.866 Scale: 1.6216e-06--------------------------------------------------------------------
Coef. Std.Err. t P>|t| [0.025 0.975]--------------------------------------------------------------------Intercept 0.0474 0.0011 44.8331 0.0000 0.0453 0.0496pSG -0.4717 0.0322 -14.6297 0.0000 -0.5374 -0.4061-------------------------------------------------------------------Omnibus: 3.263 Durbin-Watson: 1.031Prob(Omnibus): 0.196 Jarque-Bera (JB): 2.006Skew: 0.368 Prob(JB): 0.367Kurtosis: 3.913 Condition No.: 150===================================================================
• Very strong correlation of premium penetration against prior year-end surplus ratio, regression R2 = 0.87• Suggests current premium levels will persist while industry capitalization remains strong
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 12
C.02. Surplus to GDP Ratio, 1931-2020
Surplus to GDP Ratio, 1931-2020
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
PHS
to G
DP
Rat
io
Figure 11: The surplus to GDP ratio shows that relative surplus levels have steadily increased since their recorded low in 1974 of1.1% (red dot). YE 2020 surplus estimated at 2 percent above Q3. Last year’s level is the all-time high of 4.6%.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 13
The Surplus to GDP Ratio Has Increased During the Modern Period, 1968-2020
1930 1940 1950 1960 1970 1980 1990 2000 2010 2020Year
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
PHS
to G
DP
Rat
io
Figure 12: Linear regression has R2 = 0.892, though horrible error auto-correlation until the mid-2000s.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 14
C.03. Industry Volatility and ProfitabilityMetrics, 1985-2020
Industry Net Volatility and Profitability Metrics, 1985-2020
Table 1: Industry operating and profitability metrics
Quantity Kind Full Name Definition
NEP USD000 Net earned premiumNIL USD000 Net incurred lossExp USD000 Expenses All underwriting expenses and other income, excluding LAEUWGain USD000 Underwriting gainNII USD000 Net Inv Inc Net investment income, excluding gain or lossesOP USD000 UW Operating result UW Gain plus net inveestment incomeGL USD000 Inv Gain or Loss Realized and unrealized investment gain, pre-tax 2005-19NI USD000 Net income Statutory net income, page 4DF USD000 Dividends and Fin. Dividends net of capital contributions (generally negative)PHS USD000 Polholder Surplus Policyholder surplus, page 4
NIIR Net Inv Income Ratio Net investment income to earned premiumLR, ER ratio Loss, Expense Ratio Loss and expense ratio, to earned premiumCR, OR ratio Combined, Operating Ratio UW Gain, operating result to earned pmreium
NII_PHS ratio Net Inc Inc / PHS NII / PHSTIG_PHS ratio Total Inv Gain / PHS Total investment income and GL to PHS (tax mis-match)NI_PHS ratio Net Inc / PHS NI / PHS, quasi-ROE, includes realized gainsOP_PHS ratio Op Res / PHS Operating result to PHS, “operating-ROE”ROEci ratio Comp Inc / PHS Change in PHS net of dividends and financing / PHS, a
comprehensive income ci ROE
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 15
Insurance Volatility and Profitability Metrics Showing Underwriting and Operating
statistic
NetEarned
Premium
Net In-curredLoss Expenses
UndewritingGain
UW Op-eratingResult
Net InvIncomeRatio Loss Ratio
ExpenseRatio
CombinedRatio
OperatingRatio
min 133.342G 118.572G 39.761G -50.309G -11.448G 0.0832 0.652 0.263 0.929 0.8mean 363.201G 272.050G 103.009G -8.799G 33.397G 0.123 0.765 0.283 1.05 0.915max 643.004G 450.916G 187.661G 34.753G 88.101G 0.154 0.889 0.3 1.19 1.04sd 139.202G 91.640G 40.254G 17.410G 24.162G 0.0214 0.0587 0.00936 0.0591 0.0481cv 0.383 0.337 0.391 -1.98 0.723 0.174 0.0767 0.033 0.0564 0.0525growth -0.043 -0.0358 -0.0428 -0.507 1.09 0.0188 0.00867 847.159u 0.00612 0.00507r2 0.966 0.958 0.976 0.196 0.507 0.736 0.501 0.0588 0.441 0.139rse0 26.001G 19.077G 6.265G 15.837G 17.220G 0.0112 0.042 0.00921 0.0448 0.0453rse 26.418G 19.383G 6.366G 16.091G 17.496G 0.0113 0.0427 0.00936 0.0455 0.046slope 0.0358 0.0313 0.0367 -0.0832 0.0489 -0.0142 -0.00515 760.683u -0.00355 -0.00186year -25.642T -16.775T -7.458T -1.474T -3.235T 3.62 8.66 -0.148 8.51 4.32ar r2 0.991 0.966 0.993 0.255 0.517 0.839 0.428 0.6 0.425 0.238ar rse0 13.366G 16.574G 3.303G 15.331G 16.691G 0.00871 0.0426 0.00586 0.0422 0.0398ar rse 13.366G 16.574G 3.303G 15.331G 16.691G 0.0813 0.167 0.113 0.165 0.18ar slope 1.02 0.999 1.03 0.511 0.713 0.95 0.621 0.754 0.604 0.45ar intercept 8.584G 9.693G 1.316G -3.599G 11.277G 0.00436 0.285 0.0694 0.41 0.5
Table 2: statistics shown for the total US statutory industry. Columns are described on previous slide. Rows are described below.Source: S&P Global Market Intelligence, Statutory Accounts page 4.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 16
Insurance Volatility and Profitability Metrics: Investments and Returns
statisticNet InvIncome
Inv Gainor Loss
NetIncome
Dividendsand Fin.
PolholderSurplus
Net IncInc / PHS
TotalInv Gain/ PHS
Net Inc/ PHS
Op Res/ PHS
Comp Inc/ PHS
Pre-TaxIncome
min 19.508G -98.513G -5.786G -36.665G 75.511G 0.0575 -0.0976 -0.0192 -0.0435 -0.0996 -5.882Gmean 42.196G 17.064G 31.102G -10.247G 411.178G 0.134 0.178 0.078 0.0819 0.089 37.886Gmax 56.981G 121.267G 70.061G 9.159G 929.932G 0.258 0.4 0.136 0.176 0.246 89.243Gsd 10.261G 34.761G 21.541G 14.922G 241.158G 0.0606 0.0974 0.0355 0.0476 0.0796 25.711Gcv 0.243 2.04 0.693 -1.46 0.587 0.451 0.547 0.455 0.581 0.894 0.679growth -0.0259 -1.92 0.147 0.0346 -0.0665 0.0477 0.0913 0.165 1.17 -0.681 -0.082r2 0.855 0.0817 0.562 0.752 0.965 0.903 0.501 0.0402 0.00436 270.752u 0.516rse0 3.967G 33.798G 14.468G 7.533G 45.602G 0.0191 0.0698 0.0353 0.0482 0.0808 18.148Grse 4.031G 34.340G 14.700G 7.654G 46.334G 0.0195 0.0709 0.0358 0.049 0.0822 18.439Gslope 0.0213 0.0553 0.0493 0.12 0.0547 -0.0407 -0.0368 -0.00865 -0.00364 0.00144 0.0463year -1.761T -1.871T -3.038T 2.450T -44.622T 11.07 13.28 1.43 0.68 -0.167 -3.472Tar r2 0.904 237.963u 0.483 0.748 0.983 0.969 0.293 0.0812 0.156 0.00665 0.45ar rse0 3.033G 35.777G 15.522G 7.549G 31.402G 0.0104 0.0776 0.0339 0.0402 0.0807 18.996Gar rse 3.033G 35.777G 15.522G 7.549G 31.402G 0.04 0.213 0.196 0.183 0.255 18.996Gar slope 0.896 -0.0158 0.696 0.888 1.04 0.944 0.5 0.276 0.353 0.0806 0.664ar intercept 5.334G 17.495G 10.889G -2.324G 10.202G 0.00192 0.0818 0.0579 0.0565 0.0797 14.413G
Table 3: Statistics shown for the total US statutory industry. Columns are described on above. Rows are described below.Source: S&P Global Market Intelligence, Statutory Accounts page 4.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 17
Insurance Volatility and Profitability Metrics
Gloss for rows• growth is the average anuual growth rate• r2, rse0, slope, year are the R2, residual square error, slope and intercept for a regression of each variable
against time (year)• rse equals rse0 plus the standard error of the slope parameter, giving a better indication of uncertainty• ar_r2, ar_rse0, ar_rse, ar_slope, ar_intercept are the same statistics for an AR(1) autoregressive model
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 18
Time Series of Volatility and Profitability Metrics
1985 1990 1995 2000 2005 2010 2015 2020Year
0
100
200
300
400
500
600
×109Net Earned Premium; SD=1.39e+11linear fit, r2=0.966, rse0=2.6e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0
100
200
300
400
×109Net Incurred Loss; SD=9.16e+10
linear fit, r2=0.958, rse0=1.91e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0
25
50
75
100
125
150
175
×109Expenses; SD=4.03e+10
linear fit, r2=0.976, rse0=6.27e+09
1985 1990 1995 2000 2005 2010 2015 2020Year
40
20
0
20
×109Undewriting Gain; SD=1.74e+10ar fit, r2=0.255, rse0=1.53e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0
20
40
60
80
×109UW Operating Result; SD=2.42e+10
ar fit, r2=0.517, rse0=1.67e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0.08
0.09
0.10
0.11
0.12
0.13
0.14
0.15
Net Inv Income Ratio; SD=0.0214ar fit, r2=0.839, rse0=0.00871
1985 1990 1995 2000 2005 2010 2015 2020Year
0.65
0.70
0.75
0.80
0.85
0.90
Loss Ratio; SD=0.0587linear fit, r2=0.501, rse0=0.042
1985 1990 1995 2000 2005 2010 2015 2020Year
0.265
0.270
0.275
0.280
0.285
0.290
0.295
0.300
Expense Ratio; SD=0.00936ar fit, r2=0.6, rse0=0.00586
1985 1990 1995 2000 2005 2010 2015 2020Year
0.95
1.00
1.05
1.10
1.15
1.20
Combined Ratio; SD=0.0591linear fit, r2=0.441, rse0=0.0448
1985 1990 1995 2000 2005 2010 2015 2020Year
0.80
0.85
0.90
0.95
1.00
Operating Ratio; SD=0.0481ar fit, r2=0.238, rse0=0.0398
1985 1990 1995 2000 2005 2010 2015 2020Year
0
10
20
30
40
50
×109Net Inv Income; SD=1.03e+10
ar fit, r2=0.904, rse0=3.03e+09
1985 1990 1995 2000 2005 2010 2015 2020Year
100
50
0
50
100
×109Inv Gain or Loss; SD=3.48e+10
linear fit, r2=0.0817, rse0=3.38e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0
10
20
30
40
50
60
70
×109Net Income; SD=2.15e+10
linear fit, r2=0.562, rse0=1.45e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
30
20
10
0
10×109
Dividends and Fin.; SD=1.49e+10linear fit, r2=0.752, rse0=7.53e+09
1985 1990 1995 2000 2005 2010 2015 2020Year
0
200
400
600
800
×109Polholder Surplus; SD=2.41e+11
linear fit, r2=0.965, rse0=4.56e+10
1985 1990 1995 2000 2005 2010 2015 2020Year
0.050
0.075
0.100
0.125
0.150
0.175
0.200
0.225
0.250
Net Inc Inc / PHS; SD=0.0606linear fit, r2=0.903, rse0=0.0191
1985 1990 1995 2000 2005 2010 2015 2020Year
0.1
0.0
0.1
0.2
0.3
0.4
Total Inv Gain / PHS; SD=0.0974linear fit, r2=0.501, rse0=0.0698
1985 1990 1995 2000 2005 2010 2015 2020Year
0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14
Net Inc / PHS; SD=0.0355ar fit, r2=0.0812, rse0=0.0339
1985 1990 1995 2000 2005 2010 2015 2020Year
0.05
0.00
0.05
0.10
0.15
Op Res / PHS; SD=0.0476ar fit, r2=0.156, rse0=0.0402
1985 1990 1995 2000 2005 2010 2015 2020Year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25
Comp Inc / PHS; SD=0.0796ar fit, r2=0.00665, rse0=0.0807
Figure 13: Time series of the variables from the previous table. Declining investment income and increasing volatility of gainsand loss are particularly evident. See gloss on next slide.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 19
Time Series of Volatility and Profitability Metrics
Title decoder: Net Earned Premium; SD=1.39e+11 linear fit, r2=0.966, rse0=2.6e+10• CR is the variable name, e.g., combined ratio• SD standard deviation• If the variable appears to have a linear trend show the R2 and residual standard error of a regression against
time• If it appears autoregressive ar show statistics for a fit against the lagged variable• Linear trend is selected if the linear model R2 is > 0.9 or is greater than the ar model R2
Line legend• Solid line is the statistic• Horizontal line is its mean value• Dotted line is the linear (straight) or ar fit (not straight, and appears lagging)
Interpretation• rse significantly less than SD suggests potentially explainable variation• Expenses (Exp) and surplus (PHS) have a much lower uncertainty than their SD suggests because they have a
clear trend• Almost all the variability in UWGain is unexpected• CR may have a slightly predictable component—corresponding to the autoregressive underwriting cycle
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 20
C.04. Direct and Net Volatility andProfitability by Major Line, 1992-2020
Direct and Net Calendar Year Loss Ratio Average, Standard Deviation, Coefficient of Variation,Skewness, and Kurtosis by Major Line
Direct Net DiffLine Average SD CV Skew Kurt Average SD CV Skew Kurt Average SD CV Skew KurtPersonal Auto 0.768 0.041 0.053 -0.416 0.889 0.765 0.040 0.052 -0.424 0.687 0.003 0.001 0.001 0.007 0.202total 0.752 0.065 0.087 0.475 0.484 0.753 0.057 0.075 0.529 0.153 -0.000 0.009 0.011 -0.054 0.331Other Comm 0.65 0.072 0.111 0.632 1.25 0.65 0.054 0.083 0.803 0.443 0.000 0.019 0.029 -0.171 0.81Comm Auto 0.751 0.081 0.108 -0.013 -0.946 0.75 0.073 0.097 -0.039 -1.005 0.002 0.008 0.010 0.026 0.059Inland Marine 0.551 0.082 0.149 0.664 -0.151 0.568 0.059 0.105 -0.016 -1.192 -0.018 0.022 0.044 0.68 1.04CMP 0.716 0.109 0.152 0.418 0.062 0.717 0.088 0.123 0.178 -0.874 -0.001 0.020 0.029 0.24 0.936Workers Comp 0.772 0.106 0.138 0.025 -0.472 0.768 0.1 0.13 -0.121 -0.558 0.005 0.006 0.007 0.146 0.086Homeowners 0.763 0.163 0.214 1.66 4.35 0.766 0.131 0.171 1.66 5.24 -0.003 0.033 0.044 0.003 -0.884Liability 0.857 0.16 0.187 0.703 -0.496 0.833 0.158 0.19 0.769 -0.593 0.024 0.002 -0.003 -0.066 0.097Med Mal 0.837 0.214 0.256 0.734 -0.265 0.834 0.2 0.24 0.825 0.044 0.003 0.014 0.016 -0.091 -0.309Comm Property 0.691 0.314 0.454 1.56 2.03 0.676 0.188 0.278 1.6 4.21 0.015 0.126 0.176 -0.042 -2.189Reinsurance nan nan nan nan nan 0.813 0.318 0.391 2.2 7.63 nan nan nan nan nanFin Guaranty 0.631 0.709 1.12 2.31 5.51 0.627 0.627 1 2.06 4.07 0.004 0.082 0.122 0.251 1.44
Table 4: The average, standard deviation, coefficient of variation, skewness and kurtosis of direct and net loss ratio, by majorline. Sorted by Net CV. Exact grouping of statutory lines shown on next slide. Source: Industry Combined Direct Statutory IEE,1996-2017.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 21
Direct and Net Calendar Year Combined Ratio Average, Standard Deviation, and Coefficient ofVariation by Major Line
Direct Net DiffLine Average SD CV Average SD CV Average SD CVPersonal Auto 1.01 0.044 0.044 1.01 0.039 0.039 0.002 0.005 0.005Other Comm 0.937 0.073 0.078 0.973 0.046 0.048 -0.036 0.026 0.030total 1.02 0.068 0.067 1.04 0.057 0.055 -0.014 0.011 0.012Comm Auto 1.04 0.084 0.080 1.05 0.077 0.073 -0.007 0.007 0.008CMP 1.05 0.114 0.109 1.08 0.094 0.088 -0.027 0.020 0.021Inland Marine 0.854 0.092 0.108 0.918 0.081 0.089 -0.064 0.011 0.020Workers Comp 1.05 0.117 0.112 1.05 0.101 0.096 -0.007 0.016 0.016Homeowners 1.06 0.167 0.158 1.07 0.135 0.126 -0.016 0.032 0.032Liability 1.14 0.163 0.143 1.14 0.156 0.137 -0.001 0.007 0.007Med Mal 1.06 0.21 0.199 1.07 0.189 0.177 -0.015 0.021 0.022Comm Property 0.986 0.32 0.325 1.02 0.205 0.202 -0.029 0.116 0.123Reinsurance nan nan nan 1.16 0.37 0.319 nan nan nanFin Guaranty 0.818 0.889 1.09 0.923 0.601 0.65 -0.106 0.288 0.437
Table 5: The average, standard deviation, and coefficient of variation of direct combined ratio, by major line. For most linescombined ratio SD is higher than loss SD, indicating pro-cyclical expenses. LAE included with loss. Sorted by Net CV. Directminus net equals assumed minus ceded, and so does only proxies ceded. Broadly, expect net CR to be greater than direct,reflecting the cost of reinsurance, and SD to be lower, reflecting its volatility reduction. This is true for all lines execpt PersonalAuto. Personal Auto cessions tend to involve high loss ratio business ceded to residual market pools. Exact grouping of statutorylines shown on next slide. Source: Industry Combined Direct Statutory IEE, 1996-2017.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 22
Direct and Net Average Loss Ratio and Combined Ratio by Line
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15 1.20Combined Ratio
Fin Guaranty
Inland Marine
Other Comm
Comm Property
Personal Auto
total
Comm Auto
Workers Comp
CMP
Homeowners
Med Mal
Liability
Reinsurance
DirectNet
0.50 0.55 0.60 0.65 0.70 0.75 0.80 0.85 0.90Loss Ratio
Fin Guaranty
Inland Marine
Other Comm
Comm Property
Personal Auto
total
Comm Auto
Workers Comp
CMP
Homeowners
Med Mal
Liability
Reinsurance
DirectNet
Figure 14: Lines sorted by descending net combined ratio.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 23
Direct and Net Average Combined Ratio vs. Standard Deviation Combined Ratio
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40SD CR
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20Av
g D
irec
t CR Personal
Auto
OtherComm
total
CommAuto
CMP
InlandMarine
WorkersComp
Homeowners
Liability
MedMal
CommProperty
0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40SD CR
0.80
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
Avg
CR
PersonalAuto
OtherComm
total
CommAuto
CMP
InlandMarine
WorkersComp
Homeowners
Liability
MedMal
CommProperty
Reinsurance
DirectNet
Figure 15: Left: direct, right: direct and net, with joining line. Expect imppact of reinsurance to be in the northwest direction:lower standard deviation but higher combined ratio. This is the case for most lines. Risk benefit most evident for commercialproperty.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 24
Mapping Statutory Lines to Major Lines
US Statutory lines grouped as follows• CMP
• Commercial Multiple Peril (Liability Portion), Commercial MultiplePeril (Non-Liability Portion)
• Comm Auto• Commercial Auto Liability, Commercial Auto Physical Damage
• Comm Property• Allied Lines, Earthquake, Fire, Glass
• Fin Guaranty• Financial Guaranty, Mortgage Guaranty
• Homeowners• Farmowners Multiple Peril, Federal Flood, Homeowners Multiple Peril
• Inland Marine• Inland Marine
• Liability• Other Liability, Other Liability - Claims-Made, Other Liability -
Occurrence, Products Liability
• Med Mal• Medical Professional Liability
• Other Comm• Aggregate Write-Ins For Other Lines Of Business, Aircraft (All Perils),
Boiler And Machinery, Burglary And Theft, Credit, Credit A & H,Fidelity, Group A&H, International, Multiple Peril Crop, Ocean Marine,Other A&H, Private Crop, Surety, Warranty
• Personal Auto• Private Passenger Auto Liability, Private Passenger Auto Physical
Damage
• Reinsurance• Reinsurance-Nonproportional Assumed
• Workers Comp• Excess Workers’ Compensation, Workers’ Compensation
• total• Totals (Lines 1 Through 34)
• Federal Flood included with Homeowners• Liability computed as total minus all other lines owing
to Occurrence vs. Claims Made coding change inoriginal data source
• Earthquake with Comm Property: Northridge was apersonal lines event but the CEA greatly reducedpersonal lines quake premium
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 25
Loss Ratio Correlation by Major Line
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
WorkersComp total
CMP 1 0.539 0.731 -0.213 0.754 0.668 0.586 0.65 0.538 0.312 0.416 0.898Comm Auto 0.539 1 0.306 -0.480 0.021 0.213 0.285 0.544 0.39 0.599 0.080 0.407Comm Property 0.731 0.306 1 -0.196 0.634 0.741 0.35 0.347 0.354 0.224 0.121 0.712Fin Guaranty -0.213 -0.480 -0.196 1 0.018 -0.083 -0.373 -0.539 -0.228 -0.072 0.116 -0.059Homeowners 0.754 0.021 0.634 0.018 1 0.639 0.385 0.306 0.279 0.069 0.313 0.769Inland Marine 0.668 0.213 0.741 -0.083 0.639 1 0.237 0.283 0.47 0.052 0.213 0.632Liability 0.586 0.285 0.35 -0.373 0.385 0.237 1 0.809 0.534 0.249 0.492 0.703Med Mal 0.65 0.544 0.347 -0.539 0.306 0.283 0.809 1 0.662 0.279 0.517 0.662Other Comm 0.538 0.39 0.354 -0.228 0.279 0.47 0.534 0.662 1 0.222 0.509 0.596Personal Auto 0.312 0.599 0.224 -0.072 0.069 0.052 0.249 0.279 0.222 1 0.418 0.486Workers Comp 0.416 0.080 0.121 0.116 0.313 0.213 0.492 0.517 0.509 0.418 1 0.614total 0.898 0.407 0.712 -0.059 0.769 0.632 0.703 0.662 0.596 0.486 0.614 1
Table 6: Pearson correlation coefficient between direct loss ratios by major line.
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
Reinsur-ance
WorkersComp total
CMP 1 0.539 0.637 -0.147 0.793 0.764 0.615 0.559 0.598 0.385 0.353 0.343 0.852Comm Auto 0.539 1 0.32 -0.489 0.132 0.283 0.113 0.441 0.307 0.644 -0.020 -0.134 0.358Comm Property 0.637 0.32 1 -0.091 0.522 0.557 0.443 0.195 0.27 0.248 0.488 -0.015 0.59Fin Guaranty -0.147 -0.489 -0.091 1 0.189 0.236 -0.305 -0.542 -0.311 -0.166 -0.194 0.226 0.023Homeowners 0.793 0.132 0.522 0.189 1 0.675 0.573 0.332 0.39 0.28 0.383 0.479 0.855Inland Marine 0.764 0.283 0.557 0.236 0.675 1 0.358 0.201 0.361 0.159 0.367 0.332 0.663Liability 0.615 0.113 0.443 -0.305 0.573 0.358 1 0.649 0.724 0.26 0.679 0.369 0.75Med Mal 0.559 0.441 0.195 -0.542 0.332 0.201 0.649 1 0.881 0.339 0.39 0.409 0.597Other Comm 0.598 0.307 0.27 -0.311 0.39 0.361 0.724 0.881 1 0.329 0.445 0.557 0.692Personal Auto 0.385 0.644 0.248 -0.166 0.28 0.159 0.26 0.339 0.329 1 -0.063 0.313 0.582Reinsurance 0.353 -0.020 0.488 -0.194 0.383 0.367 0.679 0.39 0.445 -0.063 1 0.196 0.465Workers Comp 0.343 -0.134 -0.015 0.226 0.479 0.332 0.369 0.409 0.557 0.313 0.196 1 0.598total 0.852 0.358 0.59 0.023 0.855 0.663 0.75 0.597 0.692 0.582 0.465 0.598 1
Table 7: Pearson correlation coefficient between net loss ratios by major line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 26
Loss Ratio, Premium Change, and Loss Change Correlation
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
Direct Loss Ratio
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
Direct Premium Change
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Workers Comp
total
Direct Loss Change
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
Net Loss Ratio
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
Net Premium Change
CMPComm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
CMP
Comm Auto
Comm Property
Fin Guaranty
Homeowners
Inland Marine
Liability
Med Mal
Other Comm
Personal Auto
Reinsurance
Workers Comp
total
Net Loss Change
0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.0
0.2
0.4
0.6
0.8
1.0
0.2
0.0
0.2
0.4
0.6
0.8
1.0
Figure 16: Premium and loss change is year over year change in total losses. It is another measure of the degree to which thecycle for a line is correlated with the overall market. Numerical values given below. Direct and net views.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 27
Correlation Between Annual Change in Losses, by Major Line
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
WorkersComp total
CMP 1 0.303 0.814 0.312 0.867 0.721 0.164 0.103 0.46 0.13 0.38 0.903Comm Auto 0.303 1 0.181 -0.231 0.138 0.216 0.256 0.517 0.43 0.513 0.145 0.41Comm Property 0.814 0.181 1 0.266 0.733 0.752 0.136 -0.087 0.355 0.122 0.118 0.78Fin Guaranty 0.312 -0.231 0.266 1 0.319 0.226 -0.015 -0.147 0.124 -0.237 -0.128 0.311Homeowners 0.867 0.138 0.733 0.319 1 0.694 0.022 -0.146 0.307 0.091 0.332 0.862Inland Marine 0.721 0.216 0.752 0.226 0.694 1 0.209 -0.063 0.424 -0.019 0.333 0.753Liability 0.164 0.256 0.136 -0.015 0.022 0.209 1 0.549 0.102 0.011 0.169 0.31Med Mal 0.103 0.517 -0.087 -0.147 -0.146 -0.063 0.549 1 0.366 0.228 0.41 0.176Other Comm 0.46 0.43 0.355 0.124 0.307 0.424 0.102 0.366 1 0.238 0.217 0.527Personal Auto 0.13 0.513 0.122 -0.237 0.091 -0.019 0.011 0.228 0.238 1 0.111 0.346Workers Comp 0.38 0.145 0.118 -0.128 0.332 0.333 0.169 0.41 0.217 0.111 1 0.427total 0.903 0.41 0.78 0.311 0.862 0.753 0.31 0.176 0.527 0.346 0.427 1
Table 8: Pearson correlation between annual change in direct losses, by major line.
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
Reinsur-ance
WorkersComp total
CMP 1 0.186 0.668 0.368 0.812 0.666 0.379 0.214 0.274 -0.067 0.547 0.41 0.817Comm Auto 0.186 1 -0.004 -0.302 0.046 0.055 0.249 0.322 0.36 0.505 0.204 0.083 0.343Comm Property 0.668 -0.004 1 0.294 0.627 0.556 0.071 -0.147 0.125 -0.074 0.577 -0.049 0.62Fin Guaranty 0.368 -0.302 0.294 1 0.45 0.564 0.112 -0.106 0.131 -0.282 0.094 -0.108 0.34Homeowners 0.812 0.046 0.627 0.45 1 0.68 0.324 0.003 0.189 0.016 0.544 0.366 0.836Inland Marine 0.666 0.055 0.556 0.564 0.68 1 0.366 -0.123 0.316 -0.095 0.497 0.184 0.639Liability 0.379 0.249 0.071 0.112 0.324 0.366 1 0.299 0.34 -0.011 0.46 0.18 0.478Med Mal 0.214 0.322 -0.147 -0.106 0.003 -0.123 0.299 1 0.28 0.143 -0.100 0.451 0.213Other Comm 0.274 0.36 0.125 0.131 0.189 0.316 0.34 0.28 1 0.277 0.317 0.044 0.459Personal Auto -0.067 0.505 -0.074 -0.282 0.016 -0.095 -0.011 0.143 0.277 1 -0.090 0.148 0.318Reinsurance 0.547 0.204 0.577 0.094 0.544 0.497 0.46 -0.100 0.317 -0.090 1 0.002 0.616Workers Comp 0.41 0.083 -0.049 -0.108 0.366 0.184 0.18 0.451 0.044 0.148 0.002 1 0.407total 0.817 0.343 0.62 0.34 0.836 0.639 0.478 0.213 0.459 0.318 0.616 0.407 1
Table 9: Pearson correlation between annual change in net losses, by major line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 28
Correlation Between Annual Change in Premium, by Major Line
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
WorkersComp total
CMP 1 0.757 0.745 0.149 0.664 0.768 0.753 0.793 0.025 0.503 0.575 0.913Comm Auto 0.757 1 0.421 0.093 0.306 0.696 0.626 0.512 0.016 0.706 0.474 0.841Comm Property 0.745 0.421 1 0.212 0.408 0.496 0.604 0.774 0.166 0.215 0.157 0.646Fin Guaranty 0.149 0.093 0.212 1 0.264 0.141 0.103 0.161 0.433 0.193 -0.111 0.257Homeowners 0.664 0.306 0.408 0.264 1 0.305 0.498 0.58 -0.130 0.369 0.359 0.614Inland Marine 0.768 0.696 0.496 0.141 0.305 1 0.479 0.523 0.176 0.482 0.564 0.746Liability 0.753 0.626 0.604 0.103 0.498 0.479 1 0.561 -0.108 0.335 0.472 0.788Med Mal 0.793 0.512 0.774 0.161 0.58 0.523 0.561 1 0.086 0.477 0.271 0.742Other Comm 0.025 0.016 0.166 0.433 -0.130 0.176 -0.108 0.086 1 0.091 -0.121 0.142Personal Auto 0.503 0.706 0.215 0.193 0.369 0.482 0.335 0.477 0.091 1 0.262 0.717Workers Comp 0.575 0.474 0.157 -0.111 0.359 0.564 0.472 0.271 -0.121 0.262 1 0.606total 0.913 0.841 0.646 0.257 0.614 0.746 0.788 0.742 0.142 0.717 0.606 1
Table 10: Pearson correlation between annual change in direct premium, by major line.
name CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability
MedMal
OtherComm
PersonalAuto
Reinsur-ance
WorkersComp total
CMP 1 0.718 0.711 0.11 0.752 0.535 0.466 0.725 -0.001 0.506 0.257 0.644 0.886Comm Auto 0.718 1 0.343 0.060 0.434 0.563 0.331 0.403 0.18 0.72 0.023 0.547 0.818Comm Property 0.711 0.343 1 0.247 0.567 0.386 0.486 0.682 -0.132 0.322 0.299 0.257 0.647Fin Guaranty 0.11 0.060 0.247 1 0.231 0.308 0.596 0.127 0.328 0.222 -0.012 -0.071 0.215Homeowners 0.752 0.434 0.567 0.231 1 0.378 0.381 0.592 -0.163 0.474 -0.091 0.569 0.686Inland Marine 0.535 0.563 0.386 0.308 0.378 1 0.437 0.475 0.165 0.493 0.317 0.451 0.677Liability 0.466 0.331 0.486 0.596 0.381 0.437 1 0.418 0.013 0.36 0.341 0.252 0.558Med Mal 0.725 0.403 0.682 0.127 0.592 0.475 0.418 1 -0.003 0.365 0.57 0.405 0.723Other Comm -0.001 0.18 -0.132 0.328 -0.163 0.165 0.013 -0.003 1 0.042 0.031 -0.115 0.114Personal Auto 0.506 0.72 0.322 0.222 0.474 0.493 0.36 0.365 0.042 1 -0.076 0.437 0.77Reinsurance 0.257 0.023 0.299 -0.012 -0.091 0.317 0.341 0.57 0.031 -0.076 1 0.055 0.288Workers Comp 0.644 0.547 0.257 -0.071 0.569 0.451 0.252 0.405 -0.115 0.437 0.055 1 0.669total 0.886 0.818 0.647 0.215 0.686 0.677 0.558 0.723 0.114 0.77 0.288 0.669 1
Table 11: Pearson correlation between annual change in net premium, by major line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 29
Direct Loss Ratio Statistics by Major Line of Business
index CMPCommAuto
CommProperty
FinGuaranty Homeowners
InlandMarine Liability Med Mal
OtherComm
PersonalAuto
WorkersComp total
min 0.543 0.624 0.365 0.0791 0.565 0.393 0.678 0.548 0.522 0.663 0.577 0.639mean 0.716 0.751 0.691 0.631 0.763 0.551 0.857 0.837 0.65 0.768 0.772 0.752max 0.964 0.897 1.59 3.21 1.33 0.74 1.24 1.34 0.853 0.853 0.993 0.921sd 0.109 0.0808 0.314 0.709 0.163 0.082 0.16 0.214 0.0723 0.0409 0.106 0.0653cv 0.152 0.108 0.454 1.12 0.214 0.149 0.187 0.256 0.111 0.0533 0.138 0.0868growth 573.984u 0.00221 0.126 0.208 0.00546 0.0224 -0.00404 -0.00346 0.00437 -0.00479 -0.0126 -0.0046r2 0.281 0.00215 0.0793 0.0174 0.168 0.0329 0.546 0.272 0.118 0.0185 0.324 0.342rse0 0.094 0.0822 0.307 0.715 0.152 0.0821 0.11 0.186 0.0692 0.0413 0.0891 0.0539rse 0.0961 0.084 0.314 0.731 0.155 0.0839 0.112 0.19 0.0707 0.0422 0.0911 0.0551slope -0.00678 -439.525u -0.0104 0.011 -0.00787 -0.00175 -0.0139 -0.0131 -0.00292 -654.337u -0.00711 -0.00448year 14.32 1.63 21.51 -21.42 16.55 4.06 28.72 27.17 6.51 2.08 15.03 9.75ar r2 0.262 0.78 4.630u 0.471 0.00922 0.00367 0.663 0.79 0.272 0.434 0.69 0.226ar rse0 0.0878 0.0393 0.323 0.534 0.126 0.0826 0.0949 0.0994 0.0633 0.0319 0.0589 0.055ar rse 0.24 0.131 0.517 0.676 0.272 0.283 0.207 0.187 0.229 0.202 0.169 0.216ar slope 0.464 0.883 0.00213 0.686 0.0716 -0.062 0.804 0.866 0.517 0.758 0.836 0.443ar intercept 0.376 0.0888 0.682 0.203 0.688 0.581 0.161 0.104 0.313 0.183 0.116 0.414
Table 12: Direct calendar year Loss ratio statistics by major line of business. See below for gloss.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 30
Net Loss Ratio Statistics by Major Line of Business
index CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine Liability Med Mal
OtherComm
PersonalAuto
Reinsur-ance
WorkersComp total
min 0.568 0.633 0.387 0.0657 0.589 0.451 0.623 0.574 0.575 0.662 0.374 0.582 0.653mean 0.717 0.75 0.676 0.627 0.766 0.568 0.833 0.834 0.65 0.765 0.813 0.768 0.753max 0.895 0.873 1.33 2.77 1.24 0.658 1.17 1.32 0.787 0.838 2.05 0.971 0.884sd 0.0884 0.0728 0.188 0.627 0.131 0.0595 0.158 0.2 0.0538 0.0399 0.318 0.1 0.0567cv 0.123 0.0971 0.278 1 0.171 0.105 0.19 0.24 0.0827 0.0521 0.391 0.13 0.0754growth -0.00217 964.957u 0.0511 0.217 -0.00478 0.00924 -0.002 -999.599u -0.00182 -0.00512 0.0612 -0.0145 -0.00612r2 0.388 0.00228 0.0611 0.0124 0.273 0.112 0.474 0.218 0.346 0.0644 0.0798 0.282 0.405rse0 0.0704 0.0741 0.186 0.635 0.113 0.0571 0.117 0.18 0.0443 0.0393 0.311 0.0863 0.0446rse 0.072 0.0757 0.19 0.649 0.116 0.0584 0.12 0.184 0.0452 0.0401 0.318 0.0882 0.0456slope -0.00647 -408.233u -0.00546 0.0082 -0.00802 -0.00233 -0.0128 -0.011 -0.00372 -0.00119 -0.0105 -0.00624 -0.00424year 13.69 1.57 11.62 -15.82 16.86 5.25 26.52 22.84 8.11 3.15 21.97 13.29 9.26ar r2 0.466 0.746 0.0144 0.441 0.125 0.0368 0.479 0.731 0.663 0.447 0.0597 0.683 0.368ar rse0 0.0618 0.038 0.192 0.487 0.0904 0.0593 0.114 0.105 0.0321 0.0306 0.318 0.0538 0.0422ar rse 0.194 0.137 0.385 0.633 0.221 0.254 0.251 0.204 0.145 0.198 0.507 0.161 0.185ar slope 0.63 0.864 0.119 0.664 0.252 0.194 0.668 0.833 0.807 0.766 0.243 0.802 0.555ar intercept 0.259 0.102 0.591 0.211 0.556 0.456 0.268 0.132 0.124 0.175 0.609 0.14 0.329
Table 13: Net calendar year Loss ratio statistics by major line of business. See below for gloss.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 31
Loss Ratio Statistics by Major Line of Business
Key• min, mean, max, sd, and cv are the minimum, mean, maximum, standard deviation and coefficient of variation of
loss ratio• growth is the arithmetic average annual growth rate• r2, rse0, slope, and year are the R2, residual standard error, slope coefficient and year (time) coefficient for a
regression of loss ratio against time; rse equals rse0 plus the standard error of the slope parameter• ar r2, ar rse0, ar intercept, and ar slope are the R2, residual standard error, slope coefficient and year
(time) coefficient for an autoregression with lag 1; ar rse equals ar rse0 plus the standard error of the autoregressive parameter
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 32
Direct Loss Ratio Time Series by Major Line
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
CMP; SD=0.109 (0.0645), cor=0.898ar fit, r2=0.262, rse0=0.0878
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
Comm Auto; SD=0.0808 (0.0507), cor=0.407ar fit, r2=0.78, rse0=0.0393
1995 2000 2005 2010 2015 2020year
0.25
0.50
0.75
1.00
1.25
1.50
Comm Property; SD=0.314 (0.154), cor=0.712ar fit, r2=4.63e-06, rse0=0.323
1995 2000 2005 2010 2015 2020year
0
1
2
3
Fin Guaranty; SD=0.709 (0.561), cor=-0.0588ar fit, r2=0.471, rse0=0.534
1995 2000 2005 2010 2015 2020year
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Homeowners; SD=0.163 (0.106), cor=0.769ar fit, r2=0.00922, rse0=0.126
1995 2000 2005 2010 2015 2020year
0.4
0.5
0.6
0.7
Inland Marine; SD=0.082 (0.0425), cor=0.632ar fit, r2=0.00367, rse0=0.0826
1995 2000 2005 2010 2015 2020year
0.75
1.00
1.25
Liability; SD=0.16 (0.126), cor=0.703ar fit, r2=0.663, rse0=0.0949
1995 2000 2005 2010 2015 2020year
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Med Mal; SD=0.214 (0.159), cor=0.662ar fit, r2=0.79, rse0=0.0994
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
Other Comm; SD=0.0723 (0.0419), cor=0.596ar fit, r2=0.272, rse0=0.0633
1995 2000 2005 2010 2015 2020year
0.7
0.8
Personal Auto; SD=0.0409 (0.0266), cor=0.486ar fit, r2=0.434, rse0=0.0319
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
1.0
Workers Comp; SD=0.106 (0.0695), cor=0.614ar fit, r2=0.69, rse0=0.0589
1995 2000 2005 2010 2015 2020year
0.7
0.8
0.9
total; SD=0.0653 (0.0429), cor=1ar fit, r2=0.226, rse0=0.055
Figure 17: Direct calendar year Loss ratio time series by major line. See gloss below.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 33
Net Loss Ratio Time Series by Major Line
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
CMP; SD=0.0884 (0.0645), cor=0.852ar fit, r2=0.466, rse0=0.0618
1995 2000 2005 2010 2015 2020year
0.7
0.8
Comm Auto; SD=0.0728 (0.0507), cor=0.358ar fit, r2=0.746, rse0=0.038
1995 2000 2005 2010 2015 2020year
0.4
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Comm Property; SD=0.188 (0.154), cor=0.59ar fit, r2=0.0144, rse0=0.192
1995 2000 2005 2010 2015 2020year
0
1
2
Fin Guaranty; SD=0.627 (0.561), cor=0.0228ar fit, r2=0.441, rse0=0.487
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
1.0
1.1
1.2
Homeowners; SD=0.131 (0.106), cor=0.855ar fit, r2=0.125, rse0=0.0904
1995 2000 2005 2010 2015 2020year
0.5
0.6
Inland Marine; SD=0.0595 (0.0425), cor=0.663ar fit, r2=0.0368, rse0=0.0593
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
1.0
1.1
Liability; SD=0.158 (0.126), cor=0.75ar fit, r2=0.479, rse0=0.114
1995 2000 2005 2010 2015 2020year
0.5
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Med Mal; SD=0.2 (0.159), cor=0.597ar fit, r2=0.731, rse0=0.105
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
Other Comm; SD=0.0538 (0.0419), cor=0.692ar fit, r2=0.663, rse0=0.0321
1995 2000 2005 2010 2015 2020year
0.7
0.8
Personal Auto; SD=0.0399 (0.0266), cor=0.582ar fit, r2=0.447, rse0=0.0306
1995 2000 2005 2010 2015 2020year
0.25
0.50
0.75
1.00
1.25
1.50
1.75
2.00
Reinsurance; SD=0.318 (0.268), cor=0.465ar fit, r2=0.0597, rse0=0.318
1995 2000 2005 2010 2015 2020year
0.6
0.7
0.8
0.9
Workers Comp; SD=0.1 (0.0695), cor=0.598ar fit, r2=0.683, rse0=0.0538
Figure 18: Net calendar year Loss ratio time series by major line. See gloss on next slide.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 34
Loss Ratio Time Series by Major Line
Title decoder: CMP; SD=0.109 (0.0645), cor=0.898 ar fit, r2=0.262, rse0=0.0878• Line; standard deviation• Down-side semi-deviation is shown in parenthsis (explain)• Correlation of the line with total on the first line• (second line) shows the R2 and residual standard error of an autoregressive loss ratio model
Interpretation• When the rse is much lower than SD it suggests the market cycle is predictable• Tends to occur in casualty lines (e.g., commercial auto, medical malpractice, private passenger auto, and workers
compensation)• The cycle for property lines tends to be idiosyncratic, for obvious reasons.
Line Legend• Thin gray line in each plot shows the total loss ratio, for context• The horizontal lines show the mean (thicker) and mean ±Φ−1(22/23) = ±1.71 standard deviations
• If the loss ratios were normally distributed we expect all observations from 22 years (1996-2017) to fall within thesetram lines
• They provide a surprisingly good estimate of the range of loss ratio, except for Financial Lines (which uses adifferent tick spacing, note).
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 35
Total Loss Ratio Time Series, Direct and Net
1995 2000 2005 2010 2015 2020year
0.7
0.8
0.9
Total Direct and Net; SD=0.0567 (0.0429), cor=1ar fit, r2=0.368, rse=0.185
viewDirectNet
Figure 19: Average direct loss ratio is 0.752 with standard deviation 0.065. Average net loss ratio is 0.753 with standarddeviation 0.057. Remember gross equals direct plus assumed minus ceded, so the difference between direct and net is not thesame as ceded.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 36
Direct Loss Ratio Time Series by Major Line, Loss Ratio Scale 0 to 150 Percent
1995 2000 2005 2010 2015 2020year
46.7%
71.6%
96.6%
CMP; SD=0.109 (0.0645), cor=0.898ar fit, r2=0.262, rse0=0.0878
1995 2000 2005 2010 2015 2020year
56.6%
75.1%
93.7%
Comm Auto; SD=0.0808 (0.0507), cor=0.407ar fit, r2=0.78, rse0=0.0393
1995 2000 2005 2010 2015 2020year
69.1%
141.0%
Comm Property; SD=0.314 (0.154), cor=0.712ar fit, r2=4.63e-06, rse0=0.323
1995 2000 2005 2010 2015 2020year
63.1%
Fin Guaranty; SD=0.709 (0.561), cor=-0.0588ar fit, r2=0.471, rse0=0.534
1995 2000 2005 2010 2015 2020year
38.8%
76.3%
113.7%
Homeowners; SD=0.163 (0.106), cor=0.769ar fit, r2=0.00922, rse0=0.126
1995 2000 2005 2010 2015 2020year
36.3%
55.1%
73.9%
Inland Marine; SD=0.082 (0.0425), cor=0.632ar fit, r2=0.00367, rse0=0.0826
1995 2000 2005 2010 2015 2020year
49.1%
85.7%
122.4%
Liability; SD=0.16 (0.126), cor=0.703ar fit, r2=0.663, rse0=0.0949
1995 2000 2005 2010 2015 2020year
34.6%
83.7%
132.9%
Med Mal; SD=0.214 (0.159), cor=0.662ar fit, r2=0.79, rse0=0.0994
1995 2000 2005 2010 2015 2020year
48.4%
65.0%
81.6%
Other Comm; SD=0.0723 (0.0419), cor=0.596ar fit, r2=0.272, rse0=0.0633
1995 2000 2005 2010 2015 2020year
67.5%76.8%86.2%
Personal Auto; SD=0.0409 (0.0266), cor=0.486ar fit, r2=0.434, rse0=0.0319
1995 2000 2005 2010 2015 2020year
52.9%
77.2%
101.6%
Workers Comp; SD=0.106 (0.0695), cor=0.614ar fit, r2=0.69, rse0=0.0589
1995 2000 2005 2010 2015 2020year
60.3%
75.2%
90.2%
total; SD=0.0653 (0.0429), cor=1ar fit, r2=0.226, rse0=0.055
Figure 20: Same as the previous plot, but using the same y axis scale on all plots, to emphasize inter-line behavior.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 37
Net Loss Ratio Time Series by Major Line, Loss Ratio Scale 0 to 150 Percent
1995 2000 2005 2010 2015 2020year
51.5%
71.7%
92.0%
CMP; SD=0.0884 (0.0645), cor=0.852ar fit, r2=0.466, rse0=0.0618
1995 2000 2005 2010 2015 2020year
58.3%
75.0%
91.7%
Comm Auto; SD=0.0728 (0.0507), cor=0.358ar fit, r2=0.746, rse0=0.038
1995 2000 2005 2010 2015 2020year
24.5%
67.6%
110.7%
Comm Property; SD=0.188 (0.154), cor=0.59ar fit, r2=0.0144, rse0=0.192
1995 2000 2005 2010 2015 2020year
62.7%
Fin Guaranty; SD=0.627 (0.561), cor=0.0228ar fit, r2=0.441, rse0=0.487
1995 2000 2005 2010 2015 2020year
46.6%
76.6%
106.5%
Homeowners; SD=0.131 (0.106), cor=0.855ar fit, r2=0.125, rse0=0.0904
1995 2000 2005 2010 2015 2020year
43.2%
56.8%
70.5%
Inland Marine; SD=0.0595 (0.0425), cor=0.663ar fit, r2=0.0368, rse0=0.0593
1995 2000 2005 2010 2015 2020year
47.0%
83.3%
119.6%
Liability; SD=0.158 (0.126), cor=0.75ar fit, r2=0.479, rse0=0.114
1995 2000 2005 2010 2015 2020year
37.6%
83.4%
129.3%
Med Mal; SD=0.2 (0.159), cor=0.597ar fit, r2=0.731, rse0=0.105
1995 2000 2005 2010 2015 2020year
52.7%65.0%77.3%
Other Comm; SD=0.0538 (0.0419), cor=0.692ar fit, r2=0.663, rse0=0.0321
1995 2000 2005 2010 2015 2020year
67.4%76.5%85.6%
Personal Auto; SD=0.0399 (0.0266), cor=0.582ar fit, r2=0.447, rse0=0.0306
1995 2000 2005 2010 2015 2020year
8.5%
81.3%
Reinsurance; SD=0.318 (0.268), cor=0.465ar fit, r2=0.0597, rse0=0.318
1995 2000 2005 2010 2015 2020year
53.8%
76.8%
99.7%
Workers Comp; SD=0.1 (0.0695), cor=0.598ar fit, r2=0.683, rse0=0.0538
Figure 21: Same as the previous plot, but using the same y axis scale on all plots, to emphasize inter-line behavior.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 38
Some Lines Are More Volatile Than Others. . .
1995 2000 2005 2010 2015 2020year
46.7%
71.6%
96.6%
CMP; SD=0.109 (0.0645), cor=0.898ar fit, r2=0.262, rse0=0.0878
1995 2000 2005 2010 2015 2020year
56.6%
75.1%
93.7%
Comm Auto; SD=0.0808 (0.0507), cor=0.407ar fit, r2=0.78, rse0=0.0393
1995 2000 2005 2010 2015 2020year
69.1%
141.0%
Comm Property; SD=0.314 (0.154), cor=0.712ar fit, r2=4.63e-06, rse0=0.323
1995 2000 2005 2010 2015 2020year
63.1%
Fin Guaranty; SD=0.709 (0.561), cor=-0.0588ar fit, r2=0.471, rse0=0.534
1995 2000 2005 2010 2015 2020year
38.8%
76.3%
113.7%
Homeowners; SD=0.163 (0.106), cor=0.769ar fit, r2=0.00922, rse0=0.126
1995 2000 2005 2010 2015 2020year
36.3%
55.1%
73.9%
Inland Marine; SD=0.082 (0.0425), cor=0.632ar fit, r2=0.00367, rse0=0.0826
1995 2000 2005 2010 2015 2020year
49.1%
85.7%
122.4%
Liability; SD=0.16 (0.126), cor=0.703ar fit, r2=0.663, rse0=0.0949
1995 2000 2005 2010 2015 2020year
34.6%
83.7%
132.9%
Med Mal; SD=0.214 (0.159), cor=0.662ar fit, r2=0.79, rse0=0.0994
1995 2000 2005 2010 2015 2020year
48.4%65.0%81.6%
Other Comm; SD=0.0723 (0.0419), cor=0.596ar fit, r2=0.272, rse0=0.0633
1995 2000 2005 2010 2015 2020year
67.5%76.8%86.2%
Personal Auto; SD=0.0409 (0.0266), cor=0.486ar fit, r2=0.434, rse0=0.0319
1995 2000 2005 2010 2015 2020year
52.9%
77.2%
101.6%
Workers Comp; SD=0.106 (0.0695), cor=0.614ar fit, r2=0.69, rse0=0.0589
1995 2000 2005 2010 2015 2020year
60.3%75.2%90.2%
total; SD=0.0653 (0.0429), cor=1ar fit, r2=0.226, rse0=0.055
Figure 22: Same as the previous plot, but not clipping loss ratios at 150 percent for Commercial Property, Liability and FinGuaranty.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 39
Reinsurance More Effective for Some Lines Than Others. . .
1995 2000 2005 2010 2015 2020year
51.5%
71.7%
92.0%
CMP; SD=0.0884 (0.0645), cor=0.852ar fit, r2=0.466, rse0=0.0618
1995 2000 2005 2010 2015 2020year
58.3%
75.0%
91.7%
Comm Auto; SD=0.0728 (0.0507), cor=0.358ar fit, r2=0.746, rse0=0.038
1995 2000 2005 2010 2015 2020year
24.5%
67.6%
110.7%
Comm Property; SD=0.188 (0.154), cor=0.59ar fit, r2=0.0144, rse0=0.192
1995 2000 2005 2010 2015 2020year
62.7%
Fin Guaranty; SD=0.627 (0.561), cor=0.0228ar fit, r2=0.441, rse0=0.487
1995 2000 2005 2010 2015 2020year
46.6%
76.6%
106.5%
Homeowners; SD=0.131 (0.106), cor=0.855ar fit, r2=0.125, rse0=0.0904
1995 2000 2005 2010 2015 2020year
43.2%56.8%70.5%
Inland Marine; SD=0.0595 (0.0425), cor=0.663ar fit, r2=0.0368, rse0=0.0593
1995 2000 2005 2010 2015 2020year
47.0%
83.3%
119.6%
Liability; SD=0.158 (0.126), cor=0.75ar fit, r2=0.479, rse0=0.114
1995 2000 2005 2010 2015 2020year
37.6%
83.4%
129.3%
Med Mal; SD=0.2 (0.159), cor=0.597ar fit, r2=0.731, rse0=0.105
1995 2000 2005 2010 2015 2020year
52.7%65.0%77.3%
Other Comm; SD=0.0538 (0.0419), cor=0.692ar fit, r2=0.663, rse0=0.0321
1995 2000 2005 2010 2015 2020year
67.4%76.5%85.6%
Personal Auto; SD=0.0399 (0.0266), cor=0.582ar fit, r2=0.447, rse0=0.0306
1995 2000 2005 2010 2015 2020year
8.5%
81.3%
Reinsurance; SD=0.318 (0.268), cor=0.465ar fit, r2=0.0597, rse0=0.318
1995 2000 2005 2010 2015 2020year
53.8%
76.8%
99.7%
Workers Comp; SD=0.1 (0.0695), cor=0.598ar fit, r2=0.683, rse0=0.0538
Figure 23: Reinsurance on Commercial Property, typically excess, reduces volatility very effectively whereas net loss more volatilefor liabiltiy lines, often protected with quota shares.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 40
Direct Premium and Loss Dynamics
1995 2000 2005 2010 2015 2020
15
20
25
30
35
40
45
×109 CMPEP, rse=0.0688linear, r2=0.937IL, rse=0.115linear, r2=0.759
1995 2000 2005 2010 2015 2020
15
20
25
30
35
40
45
×109 Comm AutoEP, rse=0.06ar, r2=0.956IL, rse=0.0683ar, r2=0.951
1995 2000 2005 2010 2015 2020
5
10
15
20
25
30
35×109 Comm Property
EP, rse=0.115linear, r2=0.925IL, rse=0.385linear, r2=0.425
1995 2000 2005 2010 2015 2020
0
5
10
15
20
25
30
×109 Fin GuarantyEP, rse=0.123ar, r2=0.889IL, rse=1.29ar, r2=0.376
1995 2000 2005 2010 2015 2020
20
40
60
80
100
×109 HomeownersEP, rse=0.05linear, r2=0.989IL, rse=0.199linear, r2=0.817
1995 2000 2005 2010 2015 2020
5
10
15
20
25
×109 Inland MarineEP, rse=0.118linear, r2=0.936IL, rse=0.207linear, r2=0.822
1995 2000 2005 2010 2015 2020
20
30
40
50
60
70
80
×109 LiabilityEP, rse=0.146linear, r2=0.874IL, rse=0.129ar, r2=0.851
1995 2000 2005 2010 2015 2020
5
6
7
8
9
10
11
12
×109 Med MalEP, rse=0.075ar, r2=0.916IL, rse=0.139ar, r2=0.746
1995 2000 2005 2010 2015 2020
10
15
20
25
30
35
40
×109 Other CommEP, rse=0.1linear, r2=0.845IL, rse=0.155ar, r2=0.646
1995 2000 2005 2010 2015 2020
75
100
125
150
175
200
225
250
×109 Personal AutoEP, rse=0.0708linear, r2=0.943IL, rse=0.0681linear, r2=0.944
1995 2000 2005 2010 2015 2020
20
25
30
35
40
45
50
55
×109 Workers CompEP, rse=0.113linear, r2=0.805IL, rse=0.0785ar, r2=0.832
1995 2000 2005 2010 2015 2020
200
300
400
500
600
700
×109 totalEP, rse=0.0623linear, r2=0.963IL, rse=0.0827linear, r2=0.917
Figure 24: See below for gloss. rse is residual standard error without parameter uncertainty, normalized by the mean.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 41
Net Premium and Loss Dynamics
1995 2000 2005 2010 2015 2020
15
20
25
30
35
40
×109 CMPEP, rse=0.0663linear, r2=0.94IL, rse=0.0819linear, r2=0.858
1995 2000 2005 2010 2015 2020
15
20
25
30
35
40×109 Comm Auto
EP, rse=0.0573ar, r2=0.951IL, rse=0.0701ar, r2=0.935
1995 2000 2005 2010 2015 2020
5
10
15
20
25
×109 Comm PropertyEP, rse=0.0939linear, r2=0.946IL, rse=0.242linear, r2=0.694
1995 2000 2005 2010 2015 2020
0
5
10
15
20
25×109 Fin Guaranty
EP, rse=0.113ar, r2=0.893IL, rse=1.13ar, r2=0.4
1995 2000 2005 2010 2015 2020
20
40
60
80
100×109 Homeowners
EP, rse=0.053linear, r2=0.987IL, rse=0.136linear, r2=0.896
1995 2000 2005 2010 2015 2020
2
4
6
8
10
12
14
×109 Inland MarineEP, rse=0.0962linear, r2=0.929IL, rse=0.136linear, r2=0.851
1995 2000 2005 2010 2015 2020
20
30
40
50
60
70
80
×109 LiabilityEP, rse=0.136linear, r2=0.883IL, rse=0.121ar, r2=0.865
1995 2000 2005 2010 2015 2020
4
5
6
7
8
9
10
×109 Med MalEP, rse=0.0656ar, r2=0.938IL, rse=0.143ar, r2=0.705
1995 2000 2005 2010 2015 2020
10
15
20
25
30
35
×109 Other CommEP, rse=0.0663ar, r2=0.899IL, rse=0.111ar, r2=0.743
1995 2000 2005 2010 2015 2020
75
100
125
150
175
200
225
250×109 Personal Auto
EP, rse=0.0752linear, r2=0.936IL, rse=0.0739linear, r2=0.933
1995 2000 2005 2010 2015 20204
6
8
10
12
14
16
×109 ReinsuranceEP, rse=0.152linear, r2=0.217IL, rse=0.268ar, r2=0.337
1995 2000 2005 2010 2015 2020
20
25
30
35
40
45
50×109 Workers Comp
EP, rse=0.0703ar, r2=0.915IL, rse=0.0808ar, r2=0.819
Figure 25: See below for gloss. rse is residual standard error without parameter uncertainty, normalized by the mean.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 42
Direct Premium and Loss Dynamics
CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine
index EP IL EP IL EP IL EP IL EP IL EP ILcv 0.270 0.231 0.285 0.306 0.413 0.499 0.385 1.586 0.470 0.458 0.456 0.482cvlr 0.149 0.149 0.106 0.106 0.446 0.446 1.103 1.103 0.211 0.211 0.146 0.146r2 0.937 0.759 0.740 0.688 0.925 0.425 0.447 0.036 0.989 0.817 0.936 0.822rse 0.069 0.115 0.148 0.174 0.115 0.385 0.291 1.585 0.050 0.199 0.118 0.207ar r2 0.982 0.661 0.956 0.951 0.969 0.132 0.889 0.376 0.998 0.700 0.987 0.826ar rse 0.036 0.138 0.060 0.068 0.073 0.476 0.123 1.292 0.021 0.257 0.053 0.205
Liability Med MalOtherComm
PersonalAuto
WorkersComp total
index EP IL EP IL EP IL EP IL EP IL EP ILcv 0.405 0.331 0.263 0.269 0.249 0.267 0.291 0.283 0.251 0.185 0.319 0.282cvlr 0.183 0.183 0.252 0.252 0.109 0.109 0.052 0.052 0.135 0.135 0.085 0.085r2 0.874 0.646 0.473 0.051 0.845 0.521 0.943 0.944 0.805 0.538 0.963 0.917rse 0.146 0.200 0.195 0.267 0.100 0.188 0.071 0.068 0.113 0.128 0.062 0.083ar r2 0.955 0.851 0.916 0.746 0.930 0.646 0.988 0.961 0.937 0.832 0.991 0.904ar rse 0.086 0.129 0.075 0.139 0.063 0.155 0.031 0.056 0.065 0.078 0.030 0.088
Table 14: See below for gloss.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 43
Net Direct Premium and Loss Dynamics
CMPCommAuto
CommProperty
FinGuaranty
Home-owners
InlandMarine
index EP IL EP IL EP IL EP IL EP IL EP ILcv 0.265 0.214 0.256 0.273 0.396 0.430 0.365 1.409 0.451 0.414 0.355 0.345cvlr 0.121 0.121 0.095 0.095 0.273 0.273 0.982 0.982 0.168 0.168 0.103 0.103r2 0.940 0.858 0.724 0.671 0.946 0.694 0.478 0.042 0.987 0.896 0.929 0.851rse 0.066 0.082 0.137 0.159 0.094 0.242 0.268 1.405 0.053 0.136 0.096 0.136ar r2 0.977 0.803 0.951 0.935 0.964 0.513 0.893 0.400 0.997 0.856 0.968 0.860ar rse 0.040 0.096 0.057 0.070 0.075 0.305 0.113 1.126 0.026 0.160 0.064 0.130
Liability Med MalOtherComm
PersonalAuto
Reinsur-ance
WorkersComp
index EP IL EP IL EP IL EP IL EP IL EP ILcv 0.391 0.333 0.269 0.259 0.223 0.229 0.292 0.280 0.169 0.318 0.233 0.185cvlr 0.187 0.187 0.236 0.236 0.081 0.081 0.051 0.051 0.384 0.384 0.128 0.128r2 0.883 0.650 0.627 0.185 0.740 0.393 0.936 0.933 0.217 0.006 0.770 0.441rse 0.136 0.201 0.167 0.238 0.116 0.182 0.075 0.074 0.152 0.323 0.114 0.141ar r2 0.918 0.865 0.938 0.705 0.899 0.743 0.987 0.958 0.102 0.337 0.915 0.819ar rse 0.110 0.121 0.066 0.143 0.066 0.111 0.033 0.057 0.156 0.268 0.070 0.081
Table 15: See below for gloss.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 44
Premium and Loss Dynamics
• Loss ratio uncertainty combines loss uncertainty and premium uncertainty• Loss uncertainty = event risk, catastrophes, unexpected trends, estimation error, etc.• Premium uncertainty = market or cycle uncertainty; ability to charge technical rate in market• rse is residual standard error without parameter uncertainty
• Looking at premium and loss time series separately illuminates the particular dynamics of each line• Each plot shows premium and loss time series (thicker) and a linear or AR(1) fit (thinner)
• Cat-driven loss dynamics stand-out for CMP, Comm Property, Homeowners, Other Commercial and to a lesserextent, Inland Marine
• Non-cat losses tend to be smoother, as events bleed in through development: Comm Auto and Workers Comp
• A priori expect premium in most lines to grow smoothly, approximately with GDP, as seen in Homeowners• However, most lines exhibit excess premium volatility
• Personal Auto and Workers Comp premium are more uncertain than loss as measured by regression residual standarderror
• Comm Auto premium exhibits huge cyclic behavior
• Loss ratio uncertainty is a multifaceted and subtle pheneomenon
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 45
Calm Surface Masks Inner Turmoil
1992 1996 2000 2004 2008 2012 2016 2020year
0.20%
0.22%
0.24%
0.26%
0.28%
CMP
1992 1996 2000 2004 2008 2012 2016 2020year
0.16%
0.18%
0.20%
0.22%
0.24%
0.26%
Comm Auto
1992 1996 2000 2004 2008 2012 2016 2020year
0.10%
0.11%
0.12%
0.13%
0.14%
0.15%
0.16%
0.17%Comm Property
1992 1996 2000 2004 2008 2012 2016 2020year
0.02%
0.03%
0.04%
0.05%
0.06%
0.07%
Fin Guaranty
1992 1996 2000 2004 2008 2012 2016 2020year
0.35%
0.40%
0.45%
0.50%
0.55%
Homeowners
1992 1996 2000 2004 2008 2012 2016 2020year
0.08%
0.09%
0.10%
0.11%
0.12%
Inland Marine
1992 1996 2000 2004 2008 2012 2016 2020year
0.20%
0.25%
0.30%
0.35%
0.40%
0.45%
Liability
1992 1996 2000 2004 2008 2012 2016 2020year
0.05%
0.06%
0.07%
0.08%
0.09%
Med Mal
1992 1996 2000 2004 2008 2012 2016 2020year
0.20%
0.22%
0.24%
0.26%
0.28%
0.30%
0.32%
Other Comm
1992 1996 2000 2004 2008 2012 2016 2020year
1.10%
1.15%
1.20%
1.25%
1.30%
1.35%
Personal Auto
1992 1996 2000 2004 2008 2012 2016 2020year
0.25%
0.30%
0.35%
0.40%
0.45%
0.50%
0.55%Workers Comp
1992 1996 2000 2004 2008 2012 2016 2020year
3.00%
3.10%
3.20%
3.30%
3.40%
3.50%
3.60%
3.70%
total
Figure 26: Direct premium to GDP by major line shows a wide variety of behaviors. Broadly, property has increased, whileliability coverages have decreased.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 46
The Underwriting Cycle Is Driven by Commercial Lines
1992 1996 2000 2004 2008 2012 2016 2020year
1.2%
1.4%
1.6%
1.8%
2.0%
2.2%Personal
1992 1996 2000 2004 2008 2012 2016 2020year
1.2%
1.4%
1.6%
1.8%
2.0%
2.2%Commercial
1992 1996 2000 2004 2008 2012 2016 2020year
3.0%
3.2%
3.4%
3.6%
3.8%
total
Figure 27: Premium to GDP for personal lines vs. commerical shows the cycle is more driven by commercial. All y axes have thesame range but different locations. Personal lines (blue) shown on commerical plot to compare relative sizes. Commercial lineshas been bigger than personal since 2001
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 47
Direct and Net Premium and Loss
1992 1996 2000 2004 2008 2012 2016 2020
15
20
25
30
35
40
45
×106 CMPDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 202010
15
20
25
30
35
40
45
×106 Comm AutoDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
5
10
15
20
25
30
35×106 Comm Property
Direct premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
0
5
10
15
20
25
30
×106 Fin GuarantyDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
20
40
60
80
100
×106 HomeownersDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
5
10
15
20
25
×106 Inland MarineDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
20
30
40
50
60
70
80
×106 LiabilityDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
4
6
8
10
12
×106 Med MalDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
10
15
20
25
30
35
40
×106 Other CommDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
75
100
125
150
175
200
225
250
×106 Personal AutoDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 202015
20
25
30
35
40
45
50
55
×106 Workers CompDirect premiumDirect lossNet premiumNet loss
1992 1996 2000 2004 2008 2012 2016 2020
200
300
400
500
600
700
×106 totalDirect premiumDirect lossNet premiumNet loss
Figure 28: Direct and net premium and loss by major line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 48
Direct and Net Combined Ratio and Loss Ratio by Line by Year
1992 1996 2000 2004 2008 2012 2016 2020
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
CMPDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 20200.6
0.7
0.8
0.9
1.0
1.1
1.2
Comm AutoDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
Comm PropertyDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
2
1
0
1
2
3
Fin GuarantyDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.6
0.8
1.0
1.2
1.4
1.6
HomeownersDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.4
0.5
0.6
0.7
0.8
0.9
1.0
Inland MarineDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 20200.6
0.8
1.0
1.2
1.4
LiabilityDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.6
0.8
1.0
1.2
1.4
1.6Med Mal
Direct CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 20200.5
0.6
0.7
0.8
0.9
1.0
1.1
Other CommDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.7
0.8
0.9
1.0
1.1
Personal AutoDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.6
0.7
0.8
0.9
1.0
1.1
1.2
1.3
Workers CompDirect CRDirect LR
Net CRNet LR
1992 1996 2000 2004 2008 2012 2016 2020
0.7
0.8
0.9
1.0
1.1
1.2total
Direct CRDirect LR
Net CRNet LR
Figure 29: Direct and Net Combined Ratio and Loss Ratio by Line by Year.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 49
Net Operating Results
nameAvg UWMargin
WtdAvg UWMargin
SD UWMargin
Avg OpMargin
Wtd AvgOp Margin
SD OpMargin Avg II MSD
Adj OpMargin
Med Mal -0.072 -0.044 0.189 0.117 0.128 0.181 0.188 0.644 0.155Inland Marine 0.082 0.096 0.081 0.104 0.117 0.080 0.023 1.308 0.156Workers Comp -0.053 -0.037 0.101 0.088 0.100 0.097 0.141 0.911 0.123Liability -0.140 -0.103 0.156 0.037 0.058 0.148 0.177 0.250 0.053Fin Guaranty 0.077 -0.030 0.601 0.035 -0.106 1.052 -0.042 0.033 0.018Personal Auto -0.007 -0.004 0.039 0.027 0.028 0.040 0.034 0.676 0.035total -0.036 -0.026 0.057 0.025 0.030 0.060 0.061 0.418 0.035Comm Property -0.016 0.015 0.205 0.018 0.047 0.203 0.034 0.090 0.019Reinsurance -0.161 -0.136 0.370 0.013 0.033 0.341 0.174 0.038 -0.574Comm Auto -0.052 -0.048 0.077 0.009 0.010 0.072 0.061 0.129 0.013Other Comm 0.027 0.030 0.046 -0.003 -0.011 0.512 -0.030 -0.006 -0.009CMP -0.076 -0.059 0.094 -0.008 0.003 0.079 0.068 -0.106 -0.014Homeowners -0.071 -0.038 0.135 -0.041 -0.012 0.129 0.030 -0.320 -0.061
Table 16: Insurance operating result, with allocated investment income from policyholder funds, by major line. Weighted Wtdaverage margins are higher than straight Avg averages because bad results are followed by increasing rates and volume, thusgood years receive relatively greater weight. Weighted also biases towards the most recent year. The straight average is a moreappropriate measure of average achieved profitability. Sorted by descending average operating margin, most profitable first. AvgII shows implied investment income. MSD shows (straight) average to standard deviation ratio for operating margin. Adj OpMargin shows the operating margin adjusted for average expenses in the line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 50
Average Margins, All Lines Combined
Historical average margins
index UW Margin Op MarginIEE Op
Ins Funds IEE Total Op
Inc StmtPre-Tax To-tal Return Adj Op Margin
Average -0.0366 0.0905 0.0276 0.106 0.108 0.126Wtd Avg -0.0259 0.0951 0.0313 0.111 0.112 0.133Drag 0.0107 0.00459 0.00367 0.0052 0.00387 0.00641
1992 1996 2000 2004 2008 2012 2016 2020Year
0.20
0.15
0.10
0.05
0.00
0.05
0.10
0.15
0.20
UW MarginOp Margin
IEE Op Ins FundsIEE Total Op
Inc Stmt Pre-Tax Total Return
Figure 30: Average margins over time, all lines combined
Target average margins• Table shows historical average pre-tax
margin to earned premium• UU margin, with no investment income• Op margin, with income but no gains• IEE margin, with investment gains on
insurance funds• IEE margin, wtih all invsetment gains• Income statement, with all gains• THe last two should be the same
• Weighted return greater than averagebecause of cycle effect
• Adj Op Margin adjusts for expenses
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 51
C.05. Direct Premium Growth by MajorLine, 1992-2020
Average Direct Premium Growth Rates by Line (Percent)
index Mean SD CAGR AdjWorkers Comp 1.53 6.62 1.33 1.32Med Mal 2.82 8.14 2.53 2.49CMP 3.30 3.86 3.23 3.23Comm Auto 3.67 5.73 3.52 3.51Personal Auto 3.82 2.90 3.78 3.78Other Comm 3.97 7.29 3.73 3.70total 4.03 3.30 3.98 3.98GDP 4.28 2.19 4.25 4.25Inland Marine 5.62 5.22 5.49 5.48Liability 5.63 12.26 5.00 4.87Comm Property 5.70 9.07 5.34 5.28Homeowners 6.14 2.70 6.11 6.11Fin Guaranty 6.86 14.73 5.96 5.77
Table 17: Average direcet premium growth rates by line of business, with (nominal) GDP included for reference. Lines sortedfrom slowest to fastest growth. Insurance grows broadly with the economy, as the premium to GDP analysis showed. Since 1992,premium growth has slightly lagged GDP growth. Mean shows the average annual growth rate and SD its standard deviation.CAGR is the compound average growth rate. Adj shows the arithmetic mean adjusted for volatility using the lognormal µ+ σ2/2formula. It is very close to the actual compound rate. Source: GDP from FRED.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 52
Annual Change in Direct Premium by Line
1995 2000 2005 2010 2015 2020year
0.2
0.1
0.0
0.1
0.2
0.3
0.4
prem
ium
, pct
cha
nge
CMPComm_AutoComm_PropertyHomeowners
Inland_MarineLiabilityMed_MalOther_Comm
Personal_AutoWorkers_ComptotalGDP
Figure 31: Change by line correlated to the economy (GDP), but displays considerable variation by line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 53
Annual Change in Direct Premium Detailed by Line
1995 2000 2005 2010 2015 2020year
0.025
0.000
0.025
0.050
0.075
0.100
0.125
0.150CMP
CMPtotalGDP
1995 2000 2005 2010 2015 2020year
0.05
0.00
0.05
0.10
0.15
Comm_Auto
1995 2000 2005 2010 2015 2020year
0.1
0.0
0.1
0.2
0.3
Comm_Property
1995 2000 2005 2010 2015 2020year
0.2
0.1
0.0
0.1
0.2
0.3
0.4
0.5
0.6Fin_Guaranty
1995 2000 2005 2010 2015 2020year
0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
0.14Homeowners
1995 2000 2005 2010 2015 2020year
0.05
0.00
0.05
0.10
0.15
Inland_Marine
1995 2000 2005 2010 2015 2020year
0.2
0.1
0.0
0.1
0.2
0.3
0.4
Liability
1995 2000 2005 2010 2015 2020year
0.05
0.00
0.05
0.10
0.15
0.20
0.25
0.30
Med_Mal
1995 2000 2005 2010 2015 2020year
0.05
0.00
0.05
0.10
0.15
0.20
Other_Comm
1995 2000 2005 2010 2015 2020year
0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
Personal_Auto
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
Workers_Comp
1995 2000 2005 2010 2015 2020year
0.02
0.00
0.02
0.04
0.06
0.08
0.10
0.12
total
Figure 32: Degree of correlation to the economy (GDP) varies by line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 54
Annual Change in Direct Premium by Line Common Scale
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25CMP
CMPtotalGDP
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Comm_Auto
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Comm_Property
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Fin_Guaranty
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Homeowners
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Inland_Marine
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Liability
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Med_Mal
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Other_Comm
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Personal_Auto
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25Workers_Comp
1995 2000 2005 2010 2015 2020year
0.10
0.05
0.00
0.05
0.10
0.15
0.20
0.25total
Figure 33: Magnitude of premium changes varies considerably by line.
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 55
Premium and GDP Growth Since 1992 (1992=1.0)
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
Workers CompWorkers ComptotalGDPEMP
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
Med Mal
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
CMP
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
Comm Auto
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
Personal Auto
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
Other Comm
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
total
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Inland Marine
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
3.5
4.0Liability
1995 2000 2005 2010 2015 2020year
1.0
1.5
2.0
2.5
3.0
3.5
4.0
Comm Property
1995 2000 2005 2010 2015 2020year
1
2
3
4
5
Homeowners
1995 2000 2005 2010 2015 2020year
1
2
3
4
5
6
7
8Fin Guaranty
Figure 34: Lines sorted from slowest to fastest average growth rate, since 1992. The workers compensation plot includes theoverall level of employment EMP from the FRED series CE16OV. WC premium has declined as productivity has improved. Broadly,labor inputs have become less important in higher-risk segments of the economy. Commercial auto is particularly interesting.Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 56
C.06. Implications
Implications
Loss ratio and margin expectations• Adjusted Insurance Operating Ratios averages 13%
• Net basis• Pre-tax• Including all investment income
• Considerable variation in underlying loss ratio by line
• By line asymptotic (large portfolio) direcet loss ratioCVs range from 5% for personal auto to 45% forcommercial property
• Smaller portfolios will have larger CVs• Total industry loss ratio CV is 9% direct and 8% net
Margin expectations• Average historical pre-tax, net margin based on
industry data 1992 to 2020• -4% underwriting income to net earned premium• 9% excluding capital gains and• 11% including all capital gains
• Over period, investment income declined from over20% to less than 10% of net earned premium
• A pricing model producing a gross margin of 8 to 10%is consistent with realized experience 1996-2020
Pricing Insurance Risk Course by Stephen J Mildenhall is licensed under Creative Commons Attribution 4.0 International 57