Title: Does the insurance reform promote the development of life insurance sector in India? An empirical analysis.

Dr. Amlan Ghosh*

Affiliation Dr. Amlan Ghosh* University of North Bengal, amlanpost@gmail.com

Abstract

Reforms in the Indian life insurance sector began in 1999 and since then the growth of the life business has been impressive despite some restriction. Whether the reforms in this sector have helped the industry to grow or not, is an empirical matter. We, therefore, studied the relationship between life insurance sector reforms in India and the growth of life business in post reform period. At the empirical level, we first construct an index to measure the reforms and then used VAR-VECM model to find out the long run relationship. The Granger causality test suggests that life insurance sector reforms improved the overall development of life insurance development in the recent years in India.

Key Words Life Insurance Reforms, VAR, VECM, Indian Life insurance, Granger Causality, India.

Introduction Liberalization of the domestic financial market has been a common characteristic of a number of economies since late 60s. This was particularly true in case of industrially advanced countries like Australia, Japan, UK, and France. However, this was not been confined to these industrially developed countries only. In recent years, many LDCs have taken macroeconomic reforms, which involve structural adjustment programme. Main concentration was towards the financial system, especially banking and insurance sectors, which typically either owned or controlled by the state itself. The developing country like India along with other semi-industrialized countries has opened up their financial sector1. The New Economic Policy (NEP) introduced in India in June 1991 by the then newly elected government and the process of liberalization of Indian financial sector is part of that new policy. The main thrust of reforms in the financial sector was the creation of efficient and stable financial institutions and markets. Reforms in the banking and nonbanking sectors focused on creating a deregulated environment, strengthening the prudential norms and the supervisory system, changing the ownership pattern, and increasing competition. The main idea is globalization, privatization, deregulation and liberalization2. With the paradigm shift in the development strategy, the economy is increasingly opening up and there is a step forward towards market orientation. Consequently, some financial markets such as capital market, for-ex market and banking sector have reformed subject to various degrees of level. The insurance sector yet to receive the reform

initiatives to get the benefit out of the global changes that occurred in the recent past. The Uruguay Round of GATT (now WTO) also advocated the removal of restrictions and non-

tariff trade barriers to free flow of international services across countries so that domestic market of LDCs improve its efficiency and competitiveness and eventually improve economic growth. It is against this backdrop that many countries have deregulated its insurance sector and countries, which already allowed private insurance business further deregulated their reinsurance business such as Brazil (1991) and Peru (1991)3.

In India, the reforms in the insurance sector (Life and General) commenced with the setting up of the Committee on Reforms on Insurance Sector under the chairmanship of Dr. R.N.Malhotra, the ex- governor of RBI, by the GOI in April 1993 for examining the structure of insurance industry. The recommendations of the Committee was submitted in 1994 which was accepted in principle by the government and started implementing the recommendations since December 1999, thus heralding an era of liberalization in the countrys insurance sector4. The setting up of Insurance Regulatory and Development Authority (IRDA) and opening up of Insurance Business (life and general) to foreign capital up to 26 per cent were the initial steps in this direction. It is widely acknowledged that the opening up of the insurance sector has been aimed at ushering in greater efficiency in the insurance business by maximising productivity and minimising transaction cost. Competition is believed to bring a wider choice of products at lower prices to the consumers, larger coverage of population, better customer service, superior information technology, higher returns to the policyholders, and so on.

Fig: 1

Indian Life Insurance Market Share (2008-09)

6.92 4.79 3.25 2.22 2.5 2.06 1.23 1.05 5.06

LICI ICICI Bajaj SBI Reliance HDFC Birla Tata Kotak Others

70.92

At present there are 21 private life insurers are operating in the Indian life insurance market along with the only state own life insurer Life Insurance Corporation of India (LICI) and at the end of the financial year 2008-09. The total volume of premium reached to Rs. 221,791 crore in 2008-2009 from Rs. 24,630 crore in the year 1999-2000 which is little more than 800% increase by 22 numbers of insurers (including LICI) in India. In India, private life insurers are slowly gaining the momentum to penetrate the market with their new products, services and the global knowledge of expertise in doing life business. This can be witnessed from their growing market share statistics which shows (Fig: 1) nearly 30 percent of the Market are in their hands at the end of 2008-09 financial years. Most important aspect is that their acceptability is on the rise though it is an urban phenomenon. The prominent private players operating actively are ICICI Prudential Life (6.92%), Bajaj Allianz Life (4.79%), SBI Life (3.25%), HDFC Standard Life (2.50%),

Birla Sun Life (2.06%), Reliance Life (2.22%), Max New York Life (1.73%), and TATA AIG Life Insurance Company (1.23%)5.

Literature Review The role of financial development and economic growth has been well established by the researchers and economic analysts in their empirical studies [Levine and Zervos6 (1998), Levine7 (1990), King and Levine (1993(a)8 and (b)9 Levine et al.10 (2000), and Beck et al.11 (2000)]. These studies established the role of financial institutions and financial intermediaries in fostering the economic growth by improving the efficiency of capital accumulation, encouraging savings and ultimately improving the productivity of the economy. Now the research has shifted from established link between financial development and economic growth to understand factors that affects the overall financial services, thereby the underlying factors that lead to improve the financial development. Insurance is one of the important financial services that can trigger the growth in an economy by channelising the long-term savings for the productive purpose and providing a shield before the risk associated with any activity related to productivity, assets or life. Recent studies show that the insurance industry can improve the economic growth through financial intermediation, risk aversion and generating employment. For example, we can highlight the studies of Outreville12 (1990 b), Catalan et. al.13 (2000) and Ward and Zurbruegg14 (2002). By identifying, the macro-economic factors that promote the demand for life insurance it would be possible to find out the factors actually work as a catalyst in 6

promoting financial development and thereby economic growth. For example, recent empirical work on insurance market by Browne and Kim15 (1993), Browne et. al.16 (2000), Ward and Zurbruegg17 (2002), Beck and Webb18 (2003) and Esho et al.19 (2004) have shown that the level of insurance demand can be influenced by the economic, demographic and legal factors. Despite the findings of several influencing factors affecting the life insurance demand and the promotion of life insurance development, there is meek guidance for the policy makers to focus on specific factor/s to foster the life insurance development. More importantly, the focus on demand side has neglected the supply side of the life insurance market. The causal relationship between insurance development and the economic development has well being studied by Arena20 (2008) and found that the insurance activity does promote the economic development. Another study by

Vadlamannati21 (2008) shows that the insurance reforms have positive affect on the economic development in India. But The objective of this study is to evaluate the life
insurance reform, which is one of the factors leading to the increase of number of players in the market along with the array of product choices for the consumers in India, and to investigate the effects of these reforms empirically on the total development of life insurance business in India and not the economic development. No such study has yet been published so far on Indian life insurance market after the implementation of reforms in this sector in 1999 and to the best of authors knowledge this is first such an attempt to measure the effects of life insurance reforms empirically. The prime objective of this paper is to find out the causal relationship between life insurance reforms and the overall life insurance development in India and by doing so we would be able to answer whether we need more reforms in this sector or not. This study is also significant as the new Insurance Bill which

will allow more reforms (e.g. increasing the FDI cap of foreign insurers from present level of 26% to 49%) in the insurance sector is pending with the parliament in India. Measuring Life Insurance Reforms In this section we will try to evaluate empirically the effect of life insurance reforms on the overall development of the life insurance business in India to strengthen our earlier findings in this section. As far as our knowledge is concern no such attempt has been made so far, at least at the time of writing this section, by any researcher to investigate empirically the effects of reforms in the life insurance sector and its implications in the market in India. And to do so, the first problem we face is that how to measure the reforms in the life insurance sector. There is no such accepted measure is available to be used in our study. Therefore, we considered to construct a composite index of life insurance reforms which can be used in our study to find out the existing relationship between reforms and the development of the life market in India. To construct the index which has been named as Life Insurance Reforms Index (LIRI), we have considered the fundamentals which are post reform phenomenon, i.e., those elements which manifest the reforms initiatives in this sector. We have considered major policy reforms and regulatory reforms in constructing the LIRI. The following are the main categories which has been measured to construct the LIRI,

(a) FDI (Foreign Direct Investment) in life insurance business, and (b) Regulatory reforms in life insurance sector.

FDI in life insurance business

In the post reform period India witnessed joint ventures in the life insurance industry with foreign companies bringing maximum of 26 % capital which is stipulated by the regulator IRDA22. Since, there is a cap on the FDI in India, foreign companies cant operate individually in the insurance market in India. Due to this regulation foreign companies need to collaborate with a domestic company to enter into the life market. This FDI cap reduces the operational ability of foreign companies in India and therefore, we believe that the volume of FDI in every year would not show the exact picture of the life insurance industry in India as it would be always 26% of the total capital along with the foreign assets in the new life insurance companies. Using market share of the new private foreign life companies also not convincing as a measure of FDI in life insurance sector in India as these companies are very new and whatever the market share they have achieved are mostly concentrated in the urban areas which represents the only urban market of India and not the huge untapped rural market. Thus we come up with a simple measure which defines the FDI in the life insurance sector. We have presumed the number of new entrant of foreign and domestic companies every year to proxy the FDI in life insurance sector. The scores in the FDI in life insurance sector would calculated as under, FDI in life insurance sector = 1 for every one new entrant in this sector.

= 2 for every two new entrant in this sector, and so on.

That is, if in any given year there are six new foreign entries, then score would be 6 (six) in that particular year. Regulatory reforms in life insurance sector

It is very demanding and complicated task to quantify the regulatory reforms process in life insurance sector and it is more difficult when these reforms are in nascent stage. We, therefore, come up with a scoring system which will, eventually, define the regulatory reforms initiated and taken by the government. To measure the regulatory reforms the following scoring system is being applied in this study, Life insurance Regulatory reforms = 0 for no reforms initiatives and steps; = 1 for setting up of any committee; = 1 for report submitted by any committee; = 1 for any report accepted by the government; = 1 for passing any bill in the parliament; and = 0.25 for every new regulations framed under the IRDA Act, 1999, till date. As far as regulations are concerned, forming a committee and subsequently submitting the report of that committee in the parliament is quite long and a democratic process which is very important in implementing laws and reforms in any sector in India. Therefore we have given importance to the committee formations and reports and passing out of those reforms bill in the Parliament. Vadlamannati, (2008) also used similar kind of parameters in his study. In calculating the life insurance regulatory reforms, we have started from 1993 when the first step towards opening of the insurance sector was taken by the formation of the Malhotra Committee23. Thus we have given score of 1(one) in that year until any other major steps are taken. In 1994, the committee submitted its report which fetches another one point in the total score. Again, in 1995, another committee, named Mukherjee Committee was formed in the insurance sector 24. This also adds up one more to

10

the score of the life insurance regulatory reforms. This is how we have continued and come up with a score to quantify LIRI. For the total development of the life insurance industry see the chronology of reforms in the life insurance sector and regulations framed under the IRDA Act 1999, in the appendix.

Data and methodology We have used two variables in our study to analyze the reforms initiative in India. We used the total life insurance premium volume (LIP) as a measure of development of life insurance business in India and a composite index (LIRI) to measure the life insurance reforms in India. To eliminate the heteroscedasticity we have used the natural logarithm of life insurance premium in our study25. The specified variables denoted as

Lt = log LIP and Rt = LIRI.

In this study we first check the stationary properties of the variables since the non stationary time series variable might give spurious results26. We will use Augmented Dickey Fuller (ADF) test27 and Philips Perron (PP) test28 to verify the stationary time series variable. Non stationary variables may be used in our model provided the series are cointegrated. Therefore co-integration study also been done to verify this property. We will use Engle-Granger29 (1987) co-integration test. We also check the short run dynamics of our model by using the VAR-VECM technique30.

Stationarity tests

11

Standard regression with non-stationary data leads to spurious relationship with erroneous conclusion. It is, therefore, becomes pertinent to study the nature of the time series data involved in our study. In our study one macro economic data series (total life insurance premiums) are used which generally follow the random walk. The stationarity of both the series has been checked by the unit root test which involves Augmented Dickey Fuller (ADF) tests and Philips Perron (PP) tests.

The results of the unit root tests are very sensitive to the assumptions about the time series under test, e.g. trend, intercept or both trend and intercept. To understand the importance of the nature of the series under the unit root test, we plot them graphically at their level values and after differencing.
Fig: 2 Graphical Presentation of Life Premium and Life Insurance Reforms Index at Levels
50 40 30 20 10 0 88 90 92 94 96 LIP 98 00 02 04 06 08

L IRI

Fig: 3 Graphical Presentation of Life Premium and Life Insurance Reforms Index at First Difference

12

1 2 1 0 8 6 4 2 0 -2 8 8 9 0 9 2 9 4 9 6 9 8 0 0 0 2 0 4 0 6 0 8

DLIP

D LIRI

From Fig.2, we can see that both the time series have some trend and intercept at their levels. Considering the particular nature of trend in both the series, we have differenced the data series once and the trends have been removed but the intercept remained which can be seen in the (Fig: 3). Based on these characteristics the ADF test and PP test are performed. The results of both the tests are summarised below in the table 1 and 2. Table: 1 ADF UNIT ROOT TEST Lag Length: 1 (Automatic based on Modified AIC, Maximum Lag =4) Variables L L R R Null Hypothesis L has a unit root
(intercept & trend)

ADF test Stat. -1.6384 -3.1677 -2.2708 -2.8234

Prob* 0.7381 0.0391 0.4281 0.0747

DW stat 1.8596 2.0366 1.7873 1.8772

Critical Values 1% 5% 10% -4.5325 -3.6736 -3.2773 -3.8573 -3.0403 -4.5325 -3.6736 -3.8573 -3.0403 -2.6605 -3.2773 -2.6605

L has a unit root

(intercept)

R has a unit root

(intercept & trend)

R has a unit root

(intercept) * Mac Kinnon (1996) one-sided p-values.

13

It is clear from the ADF test (Table: 1) that both the series (life insurance premiums and life insurance reforms index) have unit root at their level values at 10%, 5% and 1% significance level. That is, the series are non-stationary. The same properties of both the series are confirmed by the PP test which showed in (Table: 2). Table: 2 PHILIPS-PERRON UNIT ROOT TEST Bandwidth: 2 (Newey-West using Bartlett Kernel) Variables L L R R

Null Hypothesis L has a unit root

(intercept & trend)

ADF test Stat. -1.0116 -3.0325 -1.9603 -2.4972

Prob* 0.9949 0.0498 0.5865 0.1316

DW stat 1.5461 1.8324 1.3819 1.7182

Critical Values 1% 5% 10% -3.8085 -3.0206 -2.6504 -3.8315 -3.0299 -4.4983 -3.6584 -3.8315 -3.0299 -2.6551 -3.2689 -2.6551

L has a unit root

(intercept)

R has a unit root

(intercept & trend)

R has a unit root

(intercept) Mac Kinnon (1996) one-sided p-values.

After the first differencing, the hypothesis of unit root is rejected in both series (see ADF test in Table: 2). That is, both the series become stationary after first differencing. So, they are integrated of order one, i.e., I(1). These findings also confirmed by the PP test except in case of R. but the correlogram, which shows Autocorrelation Functions (ACF) and Partial Autocorrelation Function (PACF) at different lags (fig: 4 to fig: 7) confirms our findings.

Co-integration Co-integration tests are conducted to ascertain any long run equilibrium relationship between these two series. The basic purpose of the co-integration test is to determine whether a group of non- stationary variables are co-integrated or not. Engel and

14

Granger (1987) points out that the two non-stationary variables can be used in regression if the linear combination of the two non-stationary variables are stationary. In such cases, the variables are said to be co-integrated. For two series to be co-integrated, both need to be integrated in the same order. Since the two variables in our study are non- stationary and integrated of order I(1), we have used the Engel-Granger co-integration test for the cointegration study. In order to test the co-integration of the series Lt and Rt, we have estimated the following two equations [equation (1) and (2)] and the residual series Ut and Vt of each estimated equation.

Lt = + Rt + Ut Rt = + Lt + Vt

---------------------- (1) --------------------- (2)

The results of the estimated equations are as follows, Lt = 8.5379 + 0.0846 Rt S.E. (0.120) (0.005) t (71.085) (15.244) --------------------- (3)

Rt = -92.0701 + 10.9240 Lt --------------------- (4) S.E. (7.142) (0.716) t (-12.890) (15.244)

After we obtain the residuals, we plot them graphically (Fig. 8) to see whether they contain any trend or not and then we examined the same with the help of ADF test (Table: 3) and PP test (Table: 4) to check the unit root property.

15

Fig: 8
Graphical Presentation of Residual Series Ut and Rt
10 8 6 4 2 0 -2 -4 -6 88 90 92 94 96 98 UT 00 VT 02 04 06 08

The ADF test and the PP test on the residual series indicate that both the series are stationary at 5 % and 10 % level. Therefore, both the life insurance premiums and the life insurance reforms are co-integrated in the long run. The correlogram of the residual (Fig: 9 and Fig: 10) series also confirms that they are stationary, i.e., I(0). Now we can say that there is a stable long run relationship between insurance reform and the development in the life insurance sector. Table: 3 ADF UNIT ROOT TEST Lag Length: 1 (Automatic based on Modified AIC, Maximum Lag =4) Variables Null Hypothesis ADF test Stat. Prob* Critical Values 1% 5% 10% Ut Ut has a unit root -2.6488 0.0111 -2.6923 -1.9601 -1.6070
(intercept)

Vt

Vt has a unit root

(intercept)

-2.2807

0.0253 -2.6923 -1.9601 -1.6070

Table: 4 PHILIPS-PERRON UNIT ROOT TEST 16

Variables Ut Vt

Bandwidth: 2 (Newey-West using Bartlett Kernel) Null Hypothesis ADF test Stat. Prob* Critical Values 1% 5% 10% Ut has a unit root -2.6247 0.0115 -2.6857 -1.9590 -1.6074
(intercept)

Vt has a unit root

(intercept)

-2.2874

0.0248 -2.6857 -1.9590 -1.6074

Vector Error Correction Model (VECM) In this model, both the series become stationary after first differencing. But differencing may result in loss of information in long run relationship among the variables. Even if there exists a long run equilibrium relationship between the two series, there may be disequilibrium in the short run. EngelGranger identifies that the co-integrated variables must have an ECM (Error Correction Model) representation and a VAR model can be reformulated by the means of all level variables. The Vector Error Correction specification restricts the long run behaviour of the endogenous variables to converge to their cointegrated relationships while allowing a wide range of short run dynamics, hence, one can treat the error terms (ET) as the equilibrium error31. Through the co-integration term, the deviation from the long run equilibrium is corrected gradually in the course of a series of short run adjustments. Therefore, VECM gives us important information about the short run relationships between these two co-integrated variables. The general form of this modified equation by employing variables of our study is presented below,

Lt = 1 + 1 ET1t-i +

i =1 n

Lt-i +

i =1

Rt-i + t

--------------- (5)

Rt = 2 + 2 ET2t-i +

i =1

Rt-i +

i =1

Lt-i + t

---------------- (6)

17

Where, t and t are white noise error terms, and ET is equal to ET = [Lt-i (Rt-i)] which is the long run effect and lagged independent variables are short run effect. That is, changes in the dependent variables are effected by the ET, Lt-i, and Rt-i. . Before estimating the VEC Model with the co-integrated vectors it is necessary to identify and select the optimal lag length of initial VAR32. Therefore, different information criterias were computed for different time lags 33. Based on the results of different information criteria (AIC, SIC, HQ, LR, FPE) we have selected optimal lag 4 in our study. RESULTS OF VECM COEFFICIENTS ESTIMATION:Table: 5 Co-integrating Vector Coefficients Variables Coefficients t statistics Standard Errors Lt-1 1.0000 Rt-1 -0.1290 -32.3301* 0.0039 C -7.8592 *Null hypothesis that estimated coefficient is equal to zero can be rejected at 1% level. Table: 6 (a) VECM Coefficients Explanatory variable Coefficients Constant 0.5715 ET1t-1 -0.1145 Lt-1 -1.1102 Lt-2 1.5126 Lt-3 -0.0992 Lt-4 -1.6077 Rt-1 0.0258 Rt-2 -0.0471 Rt-3 -0.0362 Rt-4 0.0261 Table: 6 (b) VECM Coefficients Explanatory variable Coefficients Constant 8.7015 ET2t-1 2.6264 Lt-1 -12.6635 Lt-2 84.0103 Lt-3 -36.2187 Lt-4 -70.2233

Dependent variable Lt

t Statistics 5.3572* -3.3304* -2.4900** 1.9861** -0.2299 -2.5831** 2.9012** -3.7163* -1.7153*** 1.3413

Standard Errors 0.1066 0.0343 0.4458 0.7616 0.4315 0.6224 0.0089 0.0126 0.0211 0.0194

Dependent variable Rt

t Statistics 1.9722 1.8466*** -0.6867 2.6671** -2.0294*** -2.7280**

Standard Errors 4.4119 1.4223 18.4395 31.4979 17.8466 25.7412 18

Rt-1 Rt-2 Rt-3 Rt-4

0.7069 -1.8771 -0.4905 2.4960

1.9212*** -3.5797* -0.5619 3.0984*

0.3679 0.5243 0.8728 0.8055

Note: *, **, *** indicates significant at 1%, 5% and 10% level.

Findings from VECM From VECM, the estimated equation functions has the following forms Lt = - 0.1145 (Lt-1 - 0.1290Rt-1- 7.8592) - 1.1102Lt-1 + 1.5126Lt-2 - 0.0992 Lt-3 1.6077 Lt-4+ 0.0258Rt-1-0.0471 Rt-2- 0.0362 Rt-3+ 0.0261 Rt-4+ 0.5715238422 ----- (7)

Rt = 2.6264 (Lt-1 - 0.1290Rt-1- 7.8592) - 12.6635Lt-1 + 84.01039 Lt-2 - 36.2187Lt-3 70.2233Lt-4+ 0.7069 Rt-1- 1.8771Rt-2- 0.49051Rt-3+ 2.4960 Rt-4+ 8.7015 ----- (8)

From the above results we can observed that the co-integrating vector coefficients in the long run of in both the equations are significant at 5 % level. This indicates that the system is in the state of short term dynamics. In the short run, in case of equation (5), the lagged values of Rt variable of consecutive three years has significant influence on Lt (life insurance premium volume) along with the lagged values of first, second and fourth year of Lt . On the other side, in equation (6), the dependant variable Rt significantly dependent on second, third and fourth year lagged values of Lt and first, second and fourth year lagged values of Rt itself while other variables do not affect the life insurance reforms in short run. The positive sign of ET2t-1 shows that the change in the value of Rt (insurance reforms) positively depends on past errors.

The causal relationship

19

A long run relationship implies that there must be at least one causal relationship exists among the variables. Therefore, the next step is to find out whether reforms in the life insurance sector promotes the development of life insurance business in India or the overall development in the life insurance sector helps to increase the reform process in life insurance sector. Since the series in our study are I(1) and co-integrated, the proper statistical inference can be obtained by analyzing the causality relationship on the basis of error correction model (ECM) as the simple F statistic in the traditional Granger causality test does not have a standard distribution. The result of the VEC Granger causality test {in Table. 7(a) and (b)} shows that the relationship between the two variables in India is bidirectional which means life insurance reforms in India improves the total development in the insurance sector and the development in the insurance sector also promote the overall reforms in India.

VEC Granger Causality Table: 7(a) Dependent variable: L Excluded R All Table: 7(b) Dependent variable: R Excluded L All
* Significant at 1% level.

Chi-sq 19.44915* 19.44915

df 4 4

Prob. 0.0006 0.0006

Chi-sq 18.14406* 18.14406

df 4 4

Prob. 0.0012 0.0012

20

Conclusion The effect of the insurance sector reforms on the development of life insurance sector is very important and highly debatable in India. In an attempt to shed light on this issue the present study investigates the relation between life insurance sector reforms and the overall development of life insurance business in the recent years in India by applying VARVECM econometric methodology. The ADF test and the PP test statistics were used to test the unit root properties of the variables. It is clear from the above empirical study that the life insurance sector reforms improved the overall development of life insurance development in the recent years in India. The VEC Granger causality test shows that the relationship between the insurance sector reforms and development of life insurance sector in India is bi-directional. This is probably due to the huge potentiality of the life insurance market which is still under served and the untapped market itself works as a catalyst in improving the reforms in this sector.

In summary, the results of this study supports the preposition that the reforms in the insurance sector improve the overall development of this sector and if we could improve upon the reform process in the life insurance sector, we will be able to see more development in this segment. It would be interesting to know further whether the development of the insurance sector has any impact on the economic development in the post reforms period in India. If we could improve upon the reform process in the life insurance sector, we will be able to see more development in this segment and ultimately an improvement in the economy. Therefore, policy makers should improve upon the reforms/reform process in life insurance sector for the development of life insurance sector

21

itself and for the development of the Indian economy due to the important role played by the insurance industry.

References (1) Rao, D. T, (2000), Privatisation and Foreign Participation in (Life) Insurance Sector, Economic and political Weekly, March 25, pp. 1107-1120.

22

(2)

Bhole, L.M., (1999), Financial Institutions and Markets: Structure, Growth and Innovations (Third Edition), Tata McGraw-Hill Publishing Company Limited, New Delhi, India.

(3)

Palande, P.S, Shah, R.S and Lunawat, M.L, (2003), Insurance in India: Changing Policies and Emerging Opportunities, Response Books, Sage Publications, New Delhi, India, p.62.

(4)

Government of India, (1994), Report of the Committee on Reforms in the Insurance Sector, Ministry of Finance, New Delhi, India.

(5)

Annual Report of IRDA, (2000-2001) and (2007-2008), IRDA, Hyderabad, India; (www.irdaindia.gov.in).

(6)

Levine, R. and S. Zervos, (1998), Stock Markets, Banks and Economic Growth, American Economic Review, Vol. 88, No.3, pp. 537-558.

(7)

Levine, R., (1997), Financial Development and Economic Growth: Views and Agenda, Journal of Economic Literature, June, Vol. 35, No. 2, pp 688-726.

(8)

King, R. G., and R. Levine., (1993a), Finance and Growth: Schumpeter Might be Right, Quarterly Journal of Economics, Vol. 108, No.3, pp. 717-737.

(9)

King, R. G. and Levine, R., (1993b), Finance, Entrepreneurship and Growth: Theory and Evidence, Journal of Monetary Economics, Vol. 32, pp. 513-542.

(10) Levine, R., N. Loayza, and T. Beck, (2000), Financial Intermediation and Growth: Causality and Causes, Journal of Monetary Economics, Vol. 46, No.1, pp. 31-77. (11) Beck, T., Levine, R. and Loayza, N., (2000), Finance and the Sources of Growth, Journal of Financial Economics, Vol. 58, No.1-2, pp. 261-300. (12) Outreville, J. F., (1990b), The Economic Significance of Insurance Markets in Developing Countries, Journal of Risk and Insurance, Vol.18, No.3, pp. 487-498.

23

(13) Catalan, M., Impavido, G and Musalem, A.R., (August, 2000), Contractual Savings or Stock Markets: Which leads? Financial Sector Development Department, the World Bank. (14) Ward, D., and R. Zurbruegg., (2000), Does Insurance Promote Economic Growth? Evidence from OECD Countries, Journal of Risk and Insurance, Vol. 67, No.4, pp. 489-507. (15) Browne, M. J. and Kim, K., (1993), An International Analysis of Life Insurance Demand, Journal of Risk and Insurance, Vol. 60, No. 4, pp. 616-634. (16) Browne, M.J., Chung J., and Frees E. W. (2000) International Property-Liability Insurance Consumption Journal of Risk and Insurance, Vol.67, No.1, pp. 73-90. (17) Ward, D., and R. Zurbruegg., (2002), Law, Politics and Life Insurance Consumption in Asia The Geneva Papers on Risk and Insurance, Vol. 27, pp. 395-412. (18) Beck, T. and Webb, I., (October 2002), Economic, Demographic and Institutional Determinants of Life Insurance Consumption across Countries, Working Paper: World Bank and International Insurance Foundation. (19) Esho, N., Kirievsky, A., Ward, D. and Zurbruegg, R., (2004), Law and the Determinants of Property Casualty Insurance, Journal of Risk and Insurance. Vol. 71, No.2, pp. 265-283. (20) Arena, M, (2008), Does Insurance Market Activity Promote Economic Growth? A Cross Country Study of Industrialized and Developing Countries, Journal of Risk and Insurance. Vol. 75, No.4, pp. 921-946.

(21) Vadlamannati, K., (2008), Do Insurance Sector Growth and Reforms Affect Economic Development? Empirical Evidence from India, Margin: The Journal of Applied Economic Research; Vol. 2; pp.43-86.

24

(22) Gazette of India Extraordinary Part III Section 4; Insurance Regulatory and Development Authority (Investment) Regulations, 2000; (www.irda.org.in.). (23) Government of India, (1994), Report of the Committee on Reforms in the Insurance Sector, Ministry of Finance, New Delhi, India. (24) Sinha, T., (2004), The Indian Insurance Industry: Challenges and Prospects, Swiss Re Visiting Professor, Institute of Insurance and Risk Management, Hyderabad, India, p.24. (25) Gujarati, D.N and Sangeetha., (2007), Basic Econometrics (Fourth Edition), Tata McGraw-Hill Publishing Company Limited, New Delhi, pp. 396-450. (26) Gujarati, D.N and Sangeetha, (2007), op. cit, pp. 821-825. (27) Dickey, D. A. and Fuller, W.A., (1979), Distribution of the Estimators for Autoregressive Time Series with a Unit Root. Journal of the American Statistical Association, Vol. 74, pp. 427-431. (28) Phillips, P. C. B. and Perron, P., (1987), Testing for a Unit Root in Time Series Regression. Biometrica, Vol. 75, No. 2, pp. 335-346. (29) Engle, R and Granger, C. W. J., (1987), Co-integration and Error correction: Representation, Estimation and Testing, Econometrica, vol. 55, pp.251-276. (30) Enders, W., (2004), Applied Econometric Time Series (Second Edition), John Wiley & Sons, Wiley India (P.) Ltd., pp. 320-362. (31) Gujarati, D.N and Sangeetha, (2007), op. cit., p. 843. (32) Enders, W., (2004), op. cit., pp. 335-339. (33) Skrabic, B. and Tomic-Plazibat, N., (2009), Evidence of the Long-run Equilibrium between Money Demand Determinants in Croatia, Proceedings of World Academy of Science, Engineering and Technology, Vol. 37, January, ISSN 2070-3740, pp. 578-581. 25

Annexure:

Chronology of Reforms in the Life Insurance Sector 1993 Setting up of The Malhotra Committee 1994 Recommendations of the Committee released 1995 Setting up of the Mukherjee Committee 1996 Setting up of (interim) Insurance Regulatory Authority (IRA). 1997 Mukherjee Committee report submitted but not made public. A Standing Committee on reforms created. Govt. gives greater autonomy to LICI 1998 Cabinet decides to allow 40% foreign equity in private insurance companies -26% to foreign companies and 14% to NRIs, OCBs and FIIs. Setting up of the TAC 1999 The Standing Committee headed by Mr. Murli Deora decides that the foreign equity in private insurance companies should be limited to 26%. The IRA Act was renamed as The Insurance Regulatory and Development Authority (IRDA) Act. Cabinet clears the IRDA Act 2000 The President gives assent to The IRDA Act.
Source: IRDA annual reports, various years.

Calculation of Life Insurance Reforms Index (LIRI) in India Years 1988-89 FDI (Total No. of Comp.) 01 Regulatory Reforms 0 LIRI Index 1.00

26

1989-90 1990-91 1991-92 1992-93 1993-94 1994-95 1995-96 1996-97 1997-98 1998-99 1999-00 2000-01 2001-02 2002-03 2003-04 2004-05 2005-06 2006-07 2007-08

01 01 01 01 01 01 01 01 01 01 01 11 14 13 14 15 16 17 21

0 0 0 1 2 3 4 7 9 11 14.5 15 17.5 17.75 18.50 19.25 19.25 20 21.25

1.00 1.00 1.00 2.00 3.00 4.00 5.00 8.00 10.00 12.00 15.50 26.00 31.50 30.50 32.50 34.25 35.25 37.00 42.25

Fig 4 Autocorrelation . |****** | . |***** |

Correlogram of Lt (lag=20) Partial Correlation AC . |****** | 1 0.843 . *| . | 2 0.688

PAC 0.843 -0.074

Q-Stat 17.145 29.194

Prob 0.000 0.000

27

. |**** | . |*** | . |**. | . |* . | . |* . | . | . | . *| . | .**| . | .**| . | ***| . | ***| . | ***| . | ***| . | ***| . | ***| . | ***| . | .**| . | . *| . | Fig 5 Autocorrelation

. | . . *| . . *| . . *| . . *| . . | . . | . . *| . . *| . . *| . . *| . . | . . | . . | . . | . . | . . | . . |* .

| | | | | | | | | | | | | | | | | |

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

0.553 0.425 0.305 0.187 0.069 -0.033 -0.117 -0.192 -0.261 -0.321 -0.370 -0.398 -0.401 -0.393 -0.369 -0.323 -0.252 -0.141

-0.027 -0.059 -0.061 -0.082 -0.099 -0.050 -0.049 -0.068 -0.076 -0.078 -0.071 -0.039 -0.012 -0.037 -0.013 0.020 0.057 0.141

37.395 42.535 45.345 46.475 46.639 46.678 47.232 48.858 52.138 57.662 65.917 76.842 89.805 104.71 121.14 137.94 153.26 162.89

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Correlogram of Rt (lag=20) Partial Correlation AC PAC Q-Stat Prob

. |*******| . |****** | . |***** | . |**** | . |*** | . |* . | . | . | . *| . | .**| . | .**| . | ***| . | ***| . | ***| . | ***| . | ***| . | ***| . | .**| . | .**| . | . *| . | . *| . |

. |*******| . | . | . *| . | . *| . | . *| . | . *| . | .**| . | . *| . | . |* . | . | . | . | . | . *| . | . | . | . | . | . | . | . *| . | . | . | . | . | . | . | . | . |

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

0.875 0.753 0.626 0.485 0.334 0.185 0.017 -0.136 -0.235 -0.309 -0.369 -0.416 -0.436 -0.418 -0.377 -0.335 -0.280 -0.220 -0.156 -0.087

0.875 -0.055 -0.091 -0.139 -0.140 -0.105 -0.211 -0.100 0.085 0.004 -0.052 -0.087 -0.020 0.056 -0.012 -0.073 0.011 0.004 -0.016 -0.014

18.499 32.917 43.429 50.106 53.471 54.574 54.584 55.272 57.494 61.690 68.279 77.579 89.077 101.13 112.57 123.40 132.88 140.66 146.55 150.18

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Fig 6 Autocorrelation . |**. . *| . | |

Correlogram of Lt (lag=20) Partial Correlation . |**. .**| . | | 1 AC 0.280 PAC 0.280 Q-Stat 1.8198 2.6459 Prob 0.177 0.266

2 -0.184 -0.285

28

. *| . . *| . . |* . . |* . . | . . *| . . *| . . *| . . *| . . | . . *| . . *| . . | . . | . . |* . . | . . | . Fig 7

| | | | | | | | | | | | | | | | |

. | . . *| . . |**. . *| . . |* . . *| . . | . . *| . . *| . . *| . . *| . . | . . *| . . | . . | . . | . . | .

| | | | | | | | | | | | | | | | |

3 -0.176 -0.036 4 -0.175 -0.182 5 6 0.195 0.311 0.068 0.048 0.154 -0.135

3.4503 4.2952 5.4070 6.1524 6.1754 6.7287 7.7670 7.9412 8.1487 8.2732 8.8988 9.5987 9.6379 9.6571 10.417 10.538 10.539

0.327 0.368 0.368 0.406 0.519 0.566 0.558 0.635 0.700 0.763 0.781 0.791 0.842 0.884 0.885 0.913 0.938

7 -0.026 9 -0.161

8 -0.123 -0.171 10 -0.063 -0.166 11 -0.065 -0.068 12 -0.048 -0.103 13 -0.100 -0.114 14 -0.098 -0.053 15 -0.021 -0.074 16 17 18 19 0.013 0.023 0.002 0.024 0.004 0.001 0.072 -0.010

Correlogram of Rt (lag=20) Partial Correlation AC PAC Q-Stat Prob

Autocorrelation

. |*** . *| . . | . . |* . . *| . . *| . . |* . . *| . . *| . . *| . . *| . . *| . . | . . | . . | . . | . . | . . | . . | .

| | | | | | | | | | | | | | | | | | |

. |*** .**| . . |**. . *| . . *| . . |* . . | . . *| . . | . .**| . . | . . *| . . |* . . | . . | . . | . . | . . *| . . | .

| | | | | | | | | | | | | | | | | | |

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

0.371 -0.077 0.062 0.067 -0.176 -0.059 0.125 -0.074 -0.166 -0.179 -0.127 -0.153 -0.015 0.049 0.002 -0.024 -0.029 -0.046 -0.051

0.371 -0.249 0.235 -0.105 -0.168 0.143 0.014 -0.157 0.009 -0.275 0.059 -0.132 0.093 -0.046 -0.018 -0.028 -0.042 -0.076 0.011

3.1958 3.3394 3.4392 3.5638 4.4692 4.5781 5.1104 5.3139 6.4138 7.8313 8.6190 9.9067 9.9209 10.093 10.094 10.158 10.282 10.751 11.906

0.074 0.188 0.329 0.468 0.484 0.599 0.646 0.724 0.698 0.645 0.657 0.624 0.700 0.755 0.814 0.858 0.891 0.905 0.890

Fig 9

Correlogram of Ut (lag=20)

Autocorrelation . |****** | . |*** |

Partial Correlation . |****** | .**| . |

1 2

AC PAC Q-Stat 0.718 0.718 12.437 0.384 -0.269 16.193

Prob 0.000 0.000

29

. |* . . *| . .**| . ***| . ***| . ***| . .**| . .**| . . *| . . |* . . |**. . |**. . |* . . |* . . | . . *| . . *| . . | .

Fig 10

| | | | | | | | | | | | | | | | | |

. *| . . *| . . *| . . *| . . | . . *| . . *| . . *| . . | . . |* . . |* . .**| . . *| . . | . . *| . . | . . | . . |* .

| | | | | | | | | | | | | | | | | |

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

0.118 -0.113 -0.309 -0.416 -0.408 -0.375 -0.313 -0.225 -0.082 0.096 0.265 0.233 0.133 0.066 -0.022 -0.100 -0.118 -0.033

-0.086 -0.185 -0.186 -0.105 -0.025 -0.154 -0.082 -0.098 0.033 0.087 0.104 -0.313 -0.090 -0.013 -0.078 -0.019 0.006 0.087

16.567 16.931 19.807 25.389 31.127 36.365 40.298 42.519 42.845 43.342 47.591 51.342 52.767 53.184 53.241 54.838 58.227 58.739

0.001 0.002 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Correlogram of Vt (lag=20)

Autocorrelation . |****** | . |*** | . |* . | . | . | .**| . | ***| . | ****| . | ****| . | ***| . | ***| . | .**| . | . | . | . |**. | . |**. | . |* . | . |* . | . |* . | . | . | . | . | . | . |

Partial Correlation . |****** | .**| . | . *| . | .**| . | .**| . | . *| . | . *| . | .**| . | . | . | . *| . | . | . | . |* . | . |* . | ***| . | . *| . | . | . | . *| . | . | . | . | . | . | . |

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

AC 0.750 0.449 0.191 -0.054 -0.280 -0.419 -0.460 -0.479 -0.430 -0.342 -0.191 0.001 0.199 0.213 0.157 0.124 0.067 0.004 -0.023 0.022

PAC Q-Stat 0.750 13.594 -0.260 18.720 -0.094 19.699 -0.204 19.782 -0.210 22.148 -0.091 27.789 -0.066 35.083 -0.195 43.607 -0.056 51.060 -0.127 56.181 0.025 57.943 0.067 57.943 0.085 60.326 -0.358 63.461 -0.128 65.453 -0.045 66.932 -0.077 67.474 -0.032 67.476 -0.020 67.600 0.039 67.834

Prob 0.000 0.000 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

30

31