This dissertation is based on three papers concerning aspects of insurance market development, analyzed in the light of the economic function of insurance and with the methodological tools of Econometrics and based on evidence from Italy. The first deals with the development of the non-life market, the second analyzes the life market and the third a particular facet of, again, non-life: the influence of the legal environment. The abstracts for the three papers follow.
In the first, we analyze the consumption of non-life insurance across 103 Italian provinces in 1998-2002 in order to assess its determinants, in the light of the empirical literature. Using sub-regional data we overcome an important limitation of cross-country analyses, i.e. the systemic heterogeneity due to country-specific characteristics. Individual heterogeneity is accounted for through panel data techniques. However, considering spatial units within a single market raises issues of cross-sectional or spatial dependence, either due to common nation-wide and/or regional factors or to spatial proximity. We carefully assess spatial dependence, employing recent diagnostic tests, finding out that the regressors included in our specification successfully account for spatial dependence. Insurance turns out to depend on income, wealth and some demographics, as already established, but also on trust, judicial efficiency and borrowing conditions. This finding helps in explaining the gap between Central-Northern Italy and the South of the country.
In the second we analyze the consumption of life insurance across 103 Italian provinces in 1996-2002. Cross-country analyses of insurance development have suffered from the presence of too many idiosyncratic country-level determinants (like legal system, social security, taxation or inflation history). Panel studies in the field suffer from the curse of dimensionality as well. A regional analysis of Italy, a notoriously very
diverse country, provides an environment with less unobserved heterogeneity
while retaining a fair amount of variance in income, demographics etc.. An-
other limitation of cross-country studies of life insurance, the difficulty of observing prices, may be overcome as average policy loadings tend to be uniform across regions. On the converse, a regional study raises issues of cross-sectional dependence, either due to common nationwide factors or to spatial proximity.
We control for unobserved heterogeneity through macroregional and time fixed
effects. We develop a new spatial random effects model including both spatial
lags of the dependent variable and spatial and serial correlation in the errors; we estimate it by maximum likelihood through an algorythm implemented in the R language. Life insurance turns out to depend on economic development, savings and some demographics, as expected, but also on the general level of trust and the density of the distribution network. Life insurance is negatively correlated with education, supporting the view that better education fosters financial risk taking.
In the third, we start from the consideration that civil trials take far too long in Italy compared to other countries. Inefficiency of civil law in enforcing property rights is recognized as a limiting factor for economic development at large and for that of financial markets in particular. Of the three main elements of judicial inefficiency: unfair judgment, costly procedures and lengthy procedures, we concentrate on the third analyzing its
relevance for the insurance contract. We contend that the duration of civil trials is an important obstacle to non-life insurance, because it reduces the present value of the contingent claim held by the insured in case of litigation. Thus we expect non-life insurance consumption to be lower, all other things being equal, where judicial procedures are slower. We test our hypothesis on two datasets: a provincial panel dataset for the years 1998-2002 and a household survey comprising three waves for the years 1989, 1991 and 1993. We estimate a number of alternative specifications on the aggregate data, and probit and tobit models on household data, finding significant negative effects of judicial inefficiency on insurance consumption in both settings. We conclude that the excessive length
of civil trials is a depressing factor for the development of the Italian non-life insurance market.
