Firm Level Analysis of Services Trade Restrictions in the Life Insurance Industry

Tamar Khachaturian and Sarah Oliver

**Accompanying Equations (HTML
Version)**

**Estimation
Framework**

The focus of this analysis is the effect of life insurance trade restrictions on market participation, as measured by the number of firms in each country, and on profit margins. The number of firms and average profit margins are calculated using the firm-level data described in section 4 and in the appendix. The main policy variable is the World Bank’s STRI for mode 3 life insurance. The index reflects licensing restrictions, foreign equity limits and other policies affecting the commercial establishment and operation of foreign life insurance firms.

First, a simple cross-sectional regression estimates the impact of the STRI level on the number of firms in each country. Equation (1) is the basic form of the model:

$\left(1\right)lnNumberoffirm{s}_{i2011}={\beta}_{1}+{\beta}_{2}STR{I}_{i2008}+{\epsilon}_{i2011}{}_{}$

The dependent variable is the number of firms in country *i* in 2011. Additional
controls are added in model 2: level of development, proxied by whether the
country is categorized as a high-income country, and the log of population.[1]
While the STRI is expected to have a negative impact on the number of firms,
level of development and country size are expected to have positive impacts:

$\left(2\right)lnNumberoffirm{s}_{i2011}={\beta}_{1}+{\beta}_{2}STR{I}_{i2008}+{\beta}_{3}HighIncom{e}_{i2011}{+}_{}{\beta}_{4}lnPopulatio{n}_{i2011}+{\epsilon}_{i2011}{}_{}$

The above models are estimated using both a linear regression and a non-linear Poisson model where the dependent variable is in levels rather than logs. In addition to the analysis of the total number of firms in a country market, these models are estimated separately for SMEs and large firms as well as for the ratio of SMEs to all firms in the country as the dependent variables.

Second, models estimate the impact of the STRI level on the average profitability of firms in a given market, measured as profit before taxes over net premiums written:[2]

$\left(3\right)AverageProfitMargi{n}_{i2011}={\gamma}_{1}+{\gamma}_{2}STR{I}_{i2008}+{\omega}_{i2011}{}_{}$

and

$\left(4\right)AverageProfitMargi{n}_{i2011}={\gamma}_{1}+{\gamma}_{2}STR{I}_{i2008}+{\gamma}_{3}HighIncom{e}_{i2011}{+}_{}{\gamma}_{4}lnPopulatio{n}_{i2011}+{\omega}_{i2011}{}_{}$

[1] Population data are from the World Bank, World Development Indicators (accessed June 30, 2015) https://data.worldbank.org/data-catalog/world-development-indicators.The dummy variable for level of development is from the World Bank’s categorization of income categories and equals 1 if the country is included in the high-income category; data are from the historical classification by income, available at World Bank, World Bank Country and Lending Groups (accessed September 1, 2016) https://datahelpdesk.worldbank.org/knowledgebase/articles/906519.

[2] This approach follows the unconditional average approach in Khachaturian (2015), with a simpler framework that excludes industry-specific and other controls.