ارتباط بین فعالیت های بیمه زندگی و رشد اقتصادی: برخی شواهد جدید
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|16160||2013||23 صفحه PDF||سفارش دهید||10840 کلمه|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Journal of International Money and Finance, Volume 32, February 2013, Pages 405–427
This paper applies the panel seemingly unrelated regressions augmented Dickey-Fuller (SURADF) test to re-investigate the stationarity properties of real life insurance premiums per capita and real gross domestic product (GDP) per capita for 41 countries within three levels of income covering 1979–2007. Our empirical results first reveal that the variables in these countries are a mixture of I(0) and I(1) processes, and that the traditional panel unit-root tests could lead to misleading inferences. Second, for the estimated half-lives, the degrees of mean reversion are greater in high-income countries. Third, there is concrete evidence favoring the hypothesis of a long-run equilibrium relationship between real GDP and real life insurance premiums after allowing for the heterogeneous country effect. The long-run estimated panel parameter results indicate that a 1% increase in the real life premium raises real GDP by 0.06%. Finally, we determine that the development of life insurance markets and economic growth exhibit long-run and short-run bidirectional causalities. These findings offer several useful insights for policy-makers and researchers.
The importance of the insurance-growth relationship has risen over the past few decades due to the bigger makeup of insurance within the financial sector. The global insurance industry has seen an annual growth rate of over 10% since 1950, far exceeding that of global economic development (Dowling, 1982; Swiss Reinsurance Company, 1990; UNCTAD, 1972, 1991). This greater significance is also reflected in the business volume of life insurers.1 The rapid growth of life insurance premiums not only increases insurers' role as providers of risk transfer, but also raises their importance as institutional investors. In addition, a number of international life insurance firms, such as American International Group (AIG) and International Netherlands Group (ING), have experimented with various industrial and banking linkages (Wilkins, 2009). Such developments have a profound influence in that, while they may promote economic activities, they give rise to risk in financial markets at the same time. These ideas prompted the initial motivation for this study, where we investigate the link between life insurance activities and economic growth. It is essential to consider the relationship between life insurance and economic growth from both theoretical and empirical aspects. From a theoretical point of view, the relationship between life insurance and economic growth may run in either or both directions. The ‘supply-leading’ and ‘demand-following’ views as presented by Patrick (1966) postulate that economic growth (real income) can be enhanced either through growth in financial systems, or alternatively through growth in the economy, which brings about the development of financial activities. Based on the ‘supply-leading’ view, financial development enhances economic growth by transferring resources from traditional sectors to modern sectors and by promoting an entrepreneurial response in these modern sectors. In contrast, the ‘demand-following’ view indicates that a lack of financial development or institutions is due to a lack of demand for financial services. Thus, as the growth rate of real income rises, investors' and savers' demands for various new financial services materialize, hence leading to the creation of modern financial institutions, the supply of their financial assets and liabilities, and related financial services. For the empirical aspect, previous studies mainly utilize time-series or cross-sectional datasets to investigate the relevant issues of life insurance premiums and macroeconomics, e.g., Ward and Zurbruegg (2000) and Kugler and Ofoghi (2005), to mention a few. These empirical works concentrate on a small group of countries over fairly short or distant time spans and conceivably suffer from the “small sample” problem.2 However, researchers have recently been implementing panel data to analyze related issues (Beck and Webb, 2003; Arena, 2008; Haiss and Sümegi, 2008; Han et al., 2010; Lee, 2011, forthcoming; Chen et al., forthcoming).3 Therefore, this study employs panel unit root, panel cointegration, and panel causality tests to explore the relationship between per capita real gross domestic product (hereafter RPGDP) and per capita real life insurance premiums (hereafter RPLIP; insurance density). Previous studies lack a diagnostic analysis of the order of integration for variables entering a long-run relationship between one another, which could lead to spurious regression bias. The presence of a unit root in real income and life insurance premiums has crucial implications for modeling the insurance-growth nexus. Existing panel studies on insurance premiums do warn about the adverse effects of imposing homogeneity across countries and have employed a panel unit-root test combined with a panel cointegration test to exploit the extra power derived from combining cross-sectional and time-series data. Through understanding the order of integration of variables and controlling for country heterogeneity using the panel data approach, we believe this study contributes to providing not only a clear picture of the interrelationships between the life insurance market's development and economic growth, but also presents a more accurate inference than would be shown by time-series or cross-country data alone. The main purpose of this paper is to re-investigate the stationarity properties of RPLIP and RPGDP for 41 countries within three levels (high-, middle-, and low-) of income covering 1979–2007.4 We apply the panel seemingly unrelated regressions augmented Dickey-Fuller (Panel SURADF) test developed by Breuer et al. (2001, 2002), which allows us to account for possible cross-sectional effects and to identify how many and which countries within the panel contain a unit root.5 We proceed by measuring the half-lives and the corresponding confidence intervals when a variable's stationarity (integrated of degree zero; I(0)) is confirmed. Finally, after confirming the stationary process for the two series, we then investigate the relationships between the series based on the different orders of integration. For countries in which the two series are I(1), we employ the panel cointegration method to evaluate whether real GDP and real insurance premiums are cointegrated. In addition, we identify those countries for which the two series have different properties, i.e., where one is I(1) and the other is I(0). Under these conditions, real GDP and insurance premiums cannot be related to each other in the long-run and the panel cointegration analysis is inappropriate. To this end, we distinguish between long-run and short-run causalities. Why should we take different income levels into account? First, while insurance companies play an increasingly key role in the financial sector and their increased importance as providers of financial services and investment funds in capital markets is quite pronounced in developed economies,6 there are striking differences in many developing economies where insurance premiums remain low. Second, previous studies have argued that an increase in income results in greater demand for insurance, which then leads to higher premiums. Thus, an increase in income tends to raise people's purchasing power and living standards, thus resulting in growing demand for economic security, i.e. saving insurance and annuity. To partially resolve the problem of homogeneity in the panel data, we classify the panel data into three different sub-panels – namely, high-, middle-, and low-income groups – made up of 26, 8, and 7 countries, respectively (see Table A1 in Appendix). Another reason for this classification is that the relationships between the investigated variables under consideration could be sensitive to rich and poor countries. The remainder of this paper is organized as follows. Section 2 provides a literature review. Section 3 discusses the econometric methods. Section 4 presents the empirical results. Finally, Section 5 reviews our conclusions, while also outlining some of the implications.
نتیجه گیری انگلیسی
This paper's techniques are econometrically a significant improvement over existing studies on the link between life insurance activities and economic growth. This paper adopts the panel SURADF unit-root test of Breuer et al. (2001, 2002) to re-examine 41 countries' life insurance premiums and real GDP within high-, middle-, and low-income panel sets following the classification criterion of the World Bank and covering 1979–2007. Our results illustrate that the per capita real insurance premiums and real GDP per capita in the sample countries are a mixture of I(0) and I(1) processes, and that the generally used panel root tests could lead to misleading inferences. For the persistence measures, the half-lives of real insurance premiums per capita tend to cluster in the range of 1.24–6.40 years among high-income level countries, or much slower than the range of 3.99–4.98 years among middle-income level countries. For real GDP per capita the half-lives in Italy and Japan are respectively 4.81 and 7 years. These results suggest that high-income countries' economies grow slower, which is consistent with the convergence hypothesis. The panel cointegration test results herein provide substantive evidence of a fairly strong long-run cointegration relationship between real GDP and real life insurance premiums. The long-run estimated panel regression parameter results indicate that a 1% increase in real life premium raises real GDP by 0.06%. Finally, the causality results signify that there is a positive bi-causal relationship between the level of economic activity and life insurance markets in the long-run. In this sense, a high level of economic growth leads to a high insurance premium level and vice versa. In high-income countries, there seems to be a tendency to depend on insurance markets and that a sufficiently large amount of insurance activity seems to ensure a higher level of economic growth. This paper provides an empirical justification for strengthening both the insurance market and economic development. First, to achieve sustainable economic growth, it is advantageous to further reform the insurance market in order to improve information flow and to enhance competition. Second, to take advantage of the positive interaction between insurance market activities and economic growth, authorities should liberalize the economy while liberalizing the insurance sector – that is to say, strategies that promote development in the real economy should also be emphasized (Calderon and Liu, 2003 and Lee, forthcoming). If policy-makers wish to promote economic growth, they should focus attention on long-run policies – for example, the creation of modern financial institutions, i.e. banks, insurance firms, and so on, that will foster the most efficient allocation of resources through better risk management. Countries will benefit from strengthening their regulatory framework by creating a sound environment that facilitates insurance markets' development, which further stimulates economic growth.