Efficiency Performance of GCC Insurance Sector

The objective of this study is to analyse the efficiency performance of Takaful and conventional in GCC insurance sector for the period 2015 to 2019. Data Envelopment Analysis (DEA) were used to estimate the efficiency level. Premium and investment income are taken as common output variable while labour, total fixed assets, business services, and equity capital as major input. The results show that the Takaful firms are equally efficient compared to the conventional firms. The Takaful when compared to conventional insurance is highly technical and pure technical efficient however it is moderately cost efficient and there is a large opportunity for improvement.


Introduction
The insurance sector plays a pivotal role in the prosperity and development of the socioeconomics of the society by mitigating the risk of all economic activities (Lee, 2019). Conventional insurance has been in existence for many centuries so it is deeply rooted thus it benefits from its evolution process, reliability, technical and product range superiority whereas Takaful is of a recent origin with facing issues of inefficient use of technology, product range, inadequate Re-Takaful firms, regulations inconsistencies, different organizational structure, ineffective use of funds and ineffective management. Thus, the objective of this study is to analyse efficiency performance for Takaful and conventional in GCC countries. Since its introduction in 1979, Takaful has developed steadily in terms of both volume and demographics, with the estimated Takaful assets being around 33 billion USD (Rahman, 2009). According to Al-Amri (2015), Saudi Arabia and Malaysia are the dominant countries in the GCC and South East Asian markets, respectively, where the Takaful industry is a highly concentrated market. Therefore, the objective of THIS study is to analyze the efficiency performance of the takaful and conventional insurance companies in GCC countries.

Literature Review
The most used and common efficiency estimation techniques is Data Envelopment Analysis (DEA). Other efficiency estimation techniques are Free Disposable Hull analysis (FDH), the Stochastic Frontier Approach (SFA), the Thick Frontier Approach (TFA), and the Distribution-Free Approach (DFA).
DEA is a non-parametric approach with linear programming to measure the relationship of produced goods and services (output) to assigned resources (input) and determine the efficiency score. Using data and the DEA methodology, the efficiency is measured for Takaful and conventional insurance providers in GCC countries in respect of technical efficiency(TE), scale efficiency (SE), pure technical efficiency (PTE), cost efficiency (CE) and allocative efficiency (AE). The operating efficiency performance has been examined by many studies relating to Takaful and conventional insurance albeit not for the GCC countries. Kader et al., (2010) investigated the cost efficiency in respect of General Insurance companies. The efficiency of life insurance in Malaysia was examined by Saad et al. (2006). Technical efficiency basically measures how effectively the technology is being used when it comes to the inputs to arrive at a certain level of output. Allocative efficiency symbolises how skilful and effective the management chooses the mix of inputs at the given input prices. A cost frontier, however, symbolises the minimum cost required for producing any given quantity of output to reflect the characteristics of a perfectly efficient firm. In an interesting study on the efficiency for the coexistence of Takaful and conventional insurance, it was found that Takaful has lower efficiency when compared to conventional (Ismail et al., 2011). A very important and recent contribution was made by Roziana (2013) with respect to the cost efficiency. The study highlighted that there is a significant variation in cost efficiency when a comparison is made between the Takaful and conventional insurance industry.

Methdology Data
An unbalanced set of 140 Takaful and conventional insurance companies was selected as the sample population over the period 2015 to 2019. The sample size represents about 90 percent of the GCC market and reflects the fact that the selected sample is the most representative of the GCC. The study is based largely on the primary data collected from the GCC insurance market. These comprises of but not limited to annual financial and audit reports, balance sheet and assets liabilities reports, and income statements of the selected companies.

Data Envelopment Analysis
DEA is a linear programming procedure based on a non-parametric piece-wise surface (frontier) over the data. Efficiency measures are then computed relative to this surface (Fare et al. 1983). The two categories of envelopment surfaces are constant returns to scale (CRS) and variable returns to scale (VRS) models. CRS suggests that if a firm experiences a surge in its inputs or resources, it will experience a proportional surge in its production or outputs. The VRS basically shows that an increase in input does not result in a proportional change. In DEA we use several Decision Making Units (DMUs). They are normally denoted as n DMUs to be assessed with a varying number of diverse inputs to the different outputs. The DEA efficiency score is acquired by taking the maximum ratio of weighted outputs to weighted input. This measurement allows multiple outputs and inputs to be reduced to single 'virtual' input (xi) and single 'virtual' output (yi) by optimal weights. The following equitation is highlighted by Battese and Coelli (1992): The xi variable indicates the number of inputs and yi indicates the number of outputs. Basically, the xi and yi represent data for all the n DMUs. The vector (u' yi / v' xi) highlights the ratio of all outputs over inputs. The efficiency for the ith DMU is maximized by finding the values for u and v (where u is an M  1 vector of output weights and v is a K  1 vector of input weights) , subject to the constraints that all efficiency measures must be less than or equal to 1. The efficiency for the ith DMU is maximized by finding the values for u and v (where u is an M 1 vector of output weights and v is a K 1 vector of input weights) , subject to the constraints that all efficiency measures must be less than or equal to 1, thus, leading to an infinite number of solutions. To avoid the problem of an infinite number of solutions a constant constraint ( ' xi = 1) is imposed on Equation 1 above.
Max (µ'yi), µ,φ φ'xi = 1 µ'yi-φ'xj ≤ 0 , j=1,2,3…, N, µ,φ ≥ 0 (2) The notations and indicate the transformation of u and v. The efficiency measure is then a function of the multipliers of the "virtual" input-output combination, as in Equation 2. It also indicates the multiplier form of the DEA linear programming problem. The envelopment form of the linear programming problem is shown below: min Ө, Ө, λ yi+Yλ ≥ 0 Өxi-Xλ ≥ 0 λ≥0 (3) where is a scalar and is an N 1 vector of the constant. The value of is the efficiency score for the ith DMU; it should be solved N times, one for each DMU. Table 1 Average

Efficiency of Takaful Firms in GCC Countries 2015 -2109
The average efficiency results measured took into consideration 101 Takaful firms in GCC countries performing during the period of 5 years from 2015-2019. The efficiency results are mixed. The average TE is 82.06%, which indicates that Takaful is efficient in general. The average PTE is on the lower side in terms of standard deviation and minimum value but greater in terms of mean. The average cost efficiency for the period from 2015-2019 is 61.04, which highlights that Takaful firm's average CE is still not up to the mark and has ample room for improvement in coming years of time.  As far as the average efficiency taken for the 5 years from 2015 to 2019 for all the conventional firms; technical and pure technical efficiency are on the higher side showing commitment from the management to achieve their task efficiently according to the task given. Whereas CE and AE on the lower efficiency side. The mean is on the higher side whereas standard deviation is on the lower side.

Conclusions
Over the past few decades, numerous empirical studies have intensively examined the performance of insurance firms. Insurance firms, whether Takaful or conventional, have a similar transaction, that is, the selling of insurance protection to the consumers for a consideration. This study focuses efficiency of the conventional insurance and Takaful market in GCC countries.
The empirical results of this study are categorised based on the objectives. The first empirical finding is the efficiency of the GCC insurance market between Takaful and conventional using the Data Envelopment Analysis (DEA) approach. Firstly, the average technical efficiency (TE), pure technical efficiency (PTE), and scale efficiency (SE) for Takaful firms are 82%, 87% and 77%, respectively, hence, there is room for improvement. Secondly, the empirical results show that conventional firm's TE, PTE, and SE are 86%, 89% and 88%, respectively, thereby indicating that conventional firms are effectively utilising the resources through best practice and technological advancements. The efficiency level of conventional firms is slightly higher than that for Takaful firms because conventional firms adopt a proactive approach compared to Takaful firms. In terms of cost efficiency (CE) and allocative efficiency (AE), Takaful firms are 61% and 69%, respectively, while conventional firm's CE and AE are 59% and 65%, respectively.