Home State Loyalty: An Examination of Investments in States of Property-Casualty Insurers Dara Marie Marshall, Ph.D. Miami University Willie Dion

Transcription

1 Home State Loyalty: An Examination of Investments in States of Property-Casualty Insurers Dara Marie Marshall, Ph.D. Miami University Willie Dion Reddic, Ph.D. DePaul University

2 Background Investment decisions should be based on risk and return characteristics of securities Behavioral finance literature has documented that individual investors do not invest in this rational way but invest in securities based on their familiarity with firms. Property and casualty insurers derive the majority of their income from investments and thus should make their investment decisions based on risk and return and due to the nature of their business, should make conservative investment decisions

3 Research Questions Do we observe familiarity based investment when there are high disincentives to do so? More specifically, do property and casualty insurers make investment decisions based on familiarity?

4 So why is this important? Property and Casualty insurers need to remain liquid Underwriting usually generates losses which make investment income all the more important. Insurers must be liquid in order to pay claims Familiarity theory has mostly been shown at an individual investor level with a few institutional studies Individual investors do not face the same incentives as insurers to make rational investment decisions Trade/Business relationship as measure of familiarity true decrease in information asymmetry?

5 Summary of Results On average, we find that insurers are more likely to invest in municipal securities where they are home domiciled and where they write relatively more policies.

6 Prior Related Research Insurers investment decisions Insurers underwriting income and investment income are negatively correlated Lambert and Hofflander (1966); Lambert and Hofflander (1967); Johnson (1972); Kahane and Nye (1975); Fairley (1979); Vaughn (1998) Risk Preferences Insurers tend to hold portfolios with less risk than the market Monroe and Trieschmann (1972) ; Cummins and Grace (1994); Petroni and Wahlen (1995)

7 Prior Related Research Familiarity theory and Investment decisions: Psychology Theory competence hypothesis Heath and Tversky (1991) Rational Decision making based on imperfect information Individual investors Massa and Simonov (2004) Grinblatt and Keloharju (2001) Karlsson and Nordén (2007). Institutional investors Lütje & Menkhoff (2007) Anderson, Fedenia, Hirschey, & Skiba (2011)

8 Predictive Validity Framework Familiarity Investment Portfolio Allocation DirectPremiums Written SameState Domicile ActualCosts i,j GFBal SIZE DebtPerCapita GeneralRevOwn- SourcesPerCapita Surplus

9 Hypotheses H1(null form): An insurer s proportion of direct premiums written within a particular state has no effect on the insurer s proportion of state investment portfolio allocated to municipal securities in that state. H2(alternative form): Insurers domiciled in the same state as the investment state have a higher proportion of their state investment portfolio allocated to municipal securities in their domicile state. H3(alternative form): Insurers with a higher proportion of direct premiums written in their domicile state have a higher proportion of their state investment portfolio allocated to municipal securities in that state.

10 Research Design ActualCosts j,i = β 0 β 1 0 : H1 β 2 > 0 : H2 β 3 > 0 : H3 + β 1 DirectPremiumWritten j,i + β 2 SameStateDomicile j,i + β 3 [DirectPremiumWritten j,i X SameStateDomicile j,i ] + Σ βicontrols j,i

11 Sample Selection Full Sample Insurer-state-years in Schedule T 575,484 Insurer-state-years in Schedule D 36,909 Insurer-state-years where both DirectPremiumsWritten is not 0 and ActualCosts are > Insurer-state-years where DirectPremiumsWritten = 0 and ActualCosts are > Insurer-state-years where both DirectPremiumsWritten is not 0 and ActualCosts are = Insurer-state-years where DirectPremiumsWritten = 0 and ActualCosts is missing Insurer-state-years where both DirectPremiumsWritten and ActualCosts are missing 0

12 Sample Selection Panel E: Insurer-state-years (Schedule D and Schedule T) After D and T Merge All merged no missing values Change analysis ,901 21, ,955 21,767 19, ,935 23,119 20, ,966 24,516 21,508 Total 102,757 90,495 60,651

13 Comparison of State Investments by Insurers' Domicile State Different Domicile State Same Domicile State mean observations mean observations p ActualCosts , , DirectPremiumsWritten , , Observations 90,495

14 Descriptive Statistics count mean p25 p50 p75 ActualCosts 90, DirectPremiumsWritten 90, SameStateDomicile 90, NetAdmittedAssets 90, GFBal 90, SIZE 90, DebtPerCapita 90, GeneralRevOwnSourcesPerCapita 90, Surplus 90,

15 Results Tobit Levels Regression Pred. Sign Tobit Tobit Model with Interaction b/t b/t Intercept (1.52) (1.44) DirectPremiumsWritten? *** *** (-10.12) (-18.41) SameStateDomicile *** 0.190*** (55.80) (20.42) DirectPremiumsWritten X SameStateDomicile *** (26.53) Sigma 0.235*** 0.232*** (227.61) (227.69) Year Indicators Yes Yes State Indicators Yes Yes Chi square p(chi2) Pseudo R-square Number of observations * p<0.10 ** p<0.05 *** p<0.01

16 Selection Bias vs endogeneity Domicile Incentives Familiarity Investment Portfolio Allocation DirectPremiums Written SameState Domicile ActualCosts i,j GFBal SIZE DebtPerCapita GeneralRevOwn- SourcesPerCapita Surplus

17 Matched Sample - Characteristics SameStateDomicile = 0 SameStateDomicile = 1 mean sd mean sd Diff. t statistic Premium_Tax_Rate (-0.05) LocalInvestmentCredit (-0.77) RetalitoryTaxCredit (0.21) NetAdmittedAssets (0.17) SIZE (-0.02) Observations t statistics in parentheses * p<0.10 ** p<0.05 *** p<0.01

18 Tobit Levels Regression Propensity Score Matched Sample (abbreviated) Pred. Sign Tobit Tobit Model with Interaction b/t b/t model Intercept (-0.94) (-0.86) DirectPremiumsWritten? 0.255*** (8.24) (0.04) SameStateDomicile *** 0.200*** (9.23) (7.61) DirectPremiumsWritten X SameStateDomicile *** (3.61) StateBondTaxCredit 0.157*** 0.155*** (4.22) (4.20) Chi square p(chi2) Pseudo R-square Number of observations * p<0.10 ** p<0.05 *** p<0.01

19 Contribution Prior subjects: individual investors, mutual funds and fund managers P&C Insurers invest to hedge Familiarity based on geography, heritage, culture, saliency We add direct trade/business relationship

20 Future Extensions So what if insurers do this. What are the consequences? Examine Portfolio Performance Do insurers portfolios based on familiarity underperform insurers portfolios that don t exhibit familiarity based investment Examine Publicly traded insurers Do investors penalize insurers for inefficient investment behavior Political implications Are there changes in investment based on party control of a state Are right-to-work states viewed less risky than unionized states?

21 Thank you!

22 Pooled OLS on Changes Intercept Pred. Sign Base Model Base Model with Interaction Base Model b/t b/t b/t b/t Base Model with Interaction *** *** (-3.69) (-3.69) (-0.64) (-0.63) DirectPremiumsWritten? ** ** ** ** (-2.03) (-2.03) (-2.01) (-2.01) SameStateDomicile *** 0.163*** 0.162*** 0.162*** (10.31) (10.28) (10.31) (10.28) DirectPremiumsWritten X SameStateDomicile (0.75) (0.78) Year Indicators Yes Yes Yes Yes State Indicators No No Yes Yes F Degrees of Freedom R-square Number of Observations * p<0.10 ** p<0.05 *** p<0.01

23 Tobit Bootstrapped Standard Errors Pred. Sign Tobit Tobit Model with Interaction Tobit Tobit Model with Interaction b/t b/t b/t b/t model Intercept *** *** *** *** (-69.85) (-66.99) (-57.83) (-55.98) DirectPremiumsWritten? *** *** *** *** (-3.82) (-3.61) (-3.10) (-3.73) SameStateDomicile *** 0.214*** 0.387*** 0.213*** (33.94) (28.31) (36.84) (19.33) DirectPremiumsWritten X SameStateDomicile *** 0.396*** (12.21) (10.58) sigma Intercept 0.244*** 0.242*** 0.244*** 0.242*** (123.94) (106.15) (92.48) (118.51) Year Indicators No No Yes Yes Chi square p(chi2) Pseudo R-square Number of observations * p<0.10 ** p<0.05 *** p<0.01