Insurance firms play a vital role in today's world. Whether it is one's life or a belonging, insurance firms assume significance by carrying these risks on its own shoulders. It is very simple to acquire insurance when one buys a car or a property. But one rarely ponders over how the insurance firm manages its own risk. To understand this, it is important to know the types of insurance companies. Insurance firms are commonly classified in two ways, firstly based on their structure of ownership and secondly based on their type of business. Based on their structure of ownership, it can either be a stock insurer or a mutual insurer or a reciprocal insurer. A stock insurance firm is owned by stockholders (who may or may not be policy holders) while a mutual insurance is owned by its policyholders. Reciprocal insurances operate in a unique way in which policy holders reciprocate in sharing their own risks and their existence is relatively very low. Based on their type of business, it can be classified as either a life insurer or a non-life insurer which are self explanatory.
The insurance business lately has become an increasingly risky one. Besides facing intense competition within the industry, insurance firms face threat from other financial institutions such as banks and mutual funds as well. Further, internationalization of financial markets and increased participation globally of insurance firms y has exposed them to foreign competition and raised the risk to uncomfortable levels. Add to this, the volatility of asset classes hasn't helped either.
The Need for Risk Management
Insurer's role in the economy cannot be neglected. They serve a few important functions in the economy. Firstly, insurers provide a way for common people and organisations exposed to risk of loss of life, belonging or property to transfer these risks to them in return for a premium. This is called as risk-bearing. Secondly, the insurer diversifies these risks by writing policies in large numbers for people, whose risks are more or less independent of each other. This is known as risk pooling. The third and most important function is known as financial intermediation. This involves raising funds through issuance of various specialized debt contracts and investing them in financial assets. For life insurers, these debt instruments consist of offering various types of products such as cash value life insurance, universal insurance and guaranteed investment contracts (GIC). Non-life insurers raise funds by writing various policies for people and businesses in return for a premium to cover several risks such as accidents, natural and man-made catastrophes, fires etc. Both life and non-life insurers invest the funds primarily in bonds and stocks. Life insurers also majorly invest in real estate assets, privately placed bonds and mortgages. Non-life insurers invest more in equity as compared to life insurers. The role of financial intermediaries gives rise to most of the need for risk management. This is because the cash flow arising from the issue of various debt instruments is different in pattern than the cash flow of the financial assets they invest in. This is in fact known as Asset-Liability mismatch. A common management technique known as Asset Liability Management is unique to the insurance industry. It involves matching the duration and convexity of assets with that of liabilities. Duration means the sensitivity of the price of an asset/liability to a change in interest rates whereas convexity means the sensitivity of duration of an asset/liability to a change in interest rates (Cummins, Phillips and Smith 1996).
Use of Derivatives
The use of derivative instruments as a risk management tool has exploded in recent years. This has been made possible through the introduction of variety of derivative products to solve the risk management needs of everyone and anyone. They range from standardized exchange products such as futures, options etc and tailor-made over the counter products such as forwards and swaps. The risk management options available through derivatives have provided insurers different techniques to manage their risks without sacrificing income. Besides the simple instruments, financial derivatives come in more complicated forms such as swaptions (blend of swaps and options) and more exotic forms such as Cat bonds. Cat Bonds are relatively newer derivative forms that allow transferring a set of risks from the sponsor to an investor. As an example, if a nonlife insurer has written policies for large number of properties in Florida, it might wish to survive after the occurrence of a destructive hurricane. It could sponsor a cat bond in consultation with an investment bank which would create a set of investors for the bond. In case if there is no hurricane in Florida, investors would receive healthy returns. But if hurricane occurs, the principal initially paid by the investors would be forgiven, and instead used to pay the claims of the policy holders. Life insurers face a lot of interest rate risk due to the nature of policies they provide. For example, embedded options in the policies allow policyholders to withdraw funds during changes in interest rates which expose life insurers to considerable amount of interest rate risk and they utilize derivatives such as interest rate swaps and futures to hedge its risk (Wikipedia c. 2008).
Reinsurance
Reinsurance is a mechanism by which insurance companies seek insurance for themselves. It is used mainly to protect itself from underwriting risk by sharing it with other insurance companies. Underwriting risk is mainly the risk of sudden losses occurring due to huge payments to its claimants during natural disasters like floods, earthquakes, fires etc (Investopedia c. 2008).
Case of Speculation
Derivatives are also used for income generation or speculation. The motives behind utilization of derivatives are not hedging but maximize profits. There is a general concern that speculation might increase the risk of insolvency among financial institutions especially after the huge losses suffered by Barings Bank, Procter and Gamble and many more. This has led to increased reservations on their use and changes in regulatory laws on disclosures of derivative trading. As Colquitt and Hoyt (1996) observe that shareholders, policyholders, industry regulators and others with a direct interest in the operational performance of the insurance industry have come to recognize not only the benefits of derivatives trading . . . but also the potential misuse and abuse.
Regulations
The Insurance companies regulations folded out in UK in 1994 require insurers to disclose the number, type and value of derivative contracts in a form named 13A (Gallen and Kipling 1996). Under the regulations, UK insurers are not allowed to use derivatives for speculator purposes. However, no clarity by the regulators provides a scope for UK insurers to speculate. In the US, regulations are made at the state level and thus vary state by state. In the US, insurers are allowed to speculate legally (NAIC n.d.).
Derivative Market in US and UK
Previous studies have revealed that the extent of use of derivatives in UK is much lower than in US. However, the total notional value of derivatives used by insurers in UK is not trivial. According to the Association of British Insurers (2005), UK insurance industry is the third largest provider of insurance policies in Europe (after France and Germany) and is one of the biggest in the world with US and Japan occupying the first two positions.
Having been actively involved in an insurance company, owned by my uncle, for over a year in the US, the use of derivatives to hedge risks has certainly caught my attention especially after the increased fluctuations the financial markets have seen. US and UK have lucrative insurance markets but is unexplored by researchers for quite sometime. The previous studies based on these markets are either too old or have unexplored key aspects of derivatives use. Further, the period of reservations on their use has passed and derivative markets are growing by the day. Also, there hasn't been a comprehensive comparison between the two countries. These factors have motivated me to undertake a research to study the use of derivatives by insurers in US and UK and compare the findings.
Research Questions
Q.1: What are the different derivatives instruments used by insurance firms in UK and US?
Q.2: What are the different motives behind utilization of derivatives by insurance firms in UK and US?
Q.3: What are the reasons for non-usage of derivatives among insurance firms in US and UK?
Research Objectives
The objective of this study is to examine the latest trends employed by insurance firms in derivative use as a risk management tool and using this as a benchmark, identify the various characteristics and features specific to each type of insurer (such as stock, mutual, life, non-life, large size and small size). This enables to understand the operations undertaken by each type of insurer and consequently the various risks arising from the same.
Derivatives can be and are widely believed to be used for speculation purposes with the aim of enhancing profits. The second objective is to investigate the motives for using derivatives besides risk management.
The use of derivatives in the insurance industry is not as extensive as in other industries. Hence, the third objective is to investigate the determinants of non-usage of derivatives among insurers.
The final objective of this research study is to compare the findings of US and UK to gain a further insight on the differences and the social environment they operate in.
Flow structure of the study
The discussion proceeds as follows: In Chapter 2 provides an overview of the prior related studies on the subject. Chapter 3 discusses the research methods employed, sampling strategy and issues of validity, reliability, ethical considerations etc. Chapter 4 presents the findings and discusses the implications of the same. Chapter 5 concludes by summarizing and answering the research questions. Bibliography and appendices follow.
Review of the Related Literature
Several researches on derivative usage by financial firms such as banks (Shanker 1996), savings and loans (Brewer et al. 1996) and insurers (Colquitt and Hoyt 1997) have been conducted. Although the types of risks vary largely between different industries, there exists commonality in the rationale of derivative use and techniques employed.
Cummins, Phillips and Smith (2000) observe that the modern finance theory seems inconsistent with use of derivatives. The theory states that investors which operate in complete and frictionless markets have diversified portfolios and thus can eliminate their non-systematic risks through their portfolio choices. Hence, the theory actually states that derivative use destroy shareholder value. However in the real world, the incompleteness and market frictions have motivated value-maximizing managers to use derivatives. Cummins, Phillips and Smith (2000) identify costs of financial distress, agency costs due to information asymmetries between managers and investors and underinvestment problem (increased leverage) as major market imperfections. Richard (1996) states that insurers basically have two rationales for derivative use: 1) Shareholder vale maximization and 2) Managerial Motivations. The first rationale suggests that insurers will try and increase shareholder value by increasing income and avoiding financial distress costs (for e.g. bankruptcy costs) arising due to various risks undertaken. For an insurer, these risks include foreign exchange risks, asset price volatility risk and interest rate risks. The second rationale suggests that risk-averse managers use derivatives to enhance and produce better results to win incentives or please its shareholders.
Findings by different researchers on derivative use
Bouzouita and Young (1998) conducted a survey study on non-life insurers based in USA. His research was based on 177 firms which included mutual as well as stock firms. A very low percentage (12%) of insurers used derivatives. Further, stock companies (about 65%) were found to use derivatives more than mutual (35%). Also, non-life insurers used more number of futures and options contracts, followed by swaps. Forwards were found to be the least preferred. They found an increased use by larger firms than smaller firms. Firstly, the study conducted by Bouzouita and Young (1998) was only based on non-life firms which do not provide a deep insight on the subject. Secondly, the survey research is based in 1998. Derivative markets have increased three-fold since then and although Bouzouita and Young (1998) revealed considerable information, their findings might significantly be completely different from the situation today. Cummins, Phillips and Smith (1996) conducted a broader research on life as well as non-life firms. Their sample consisted of 1207 life insurers and 2063 non-life insurers and data was based on the regulatory annual statements filed with National Association of Insurance Commissioners (NAIC) in the US. Their research aimed at identifying the trends in insurance companies regarding derivatives and also studying the factors behind its use. Their finding was consistent with that of Bouzouita and Young (1998) as only 12% of life insurers and 7% of non-life insurers were found to use derivatives. Further, stock firms were more in number than mutual firms. However, swaps were used more than futures and options which is inconsistent with the other. While, interest rate swaps were used in large number by life insurers, forward contracts were more popular among non-life insurers. Further, they state that economies of scale for larger firms accounts for more usage than smaller firms. The data used was based on annual statements filed until 1994. Again, the findings lose validity as they are old. Also, both the researches discussed above fail to segregate firms based on their motives for usage (Hedging or speculation). A further study undertaken by Colquitt and Hoyt (1997) based on data from 571 US-based life insurers investigate determinants of specifically the use of futures and options. The results of the study indicate that size, leverage and mismatch of assets and liabilities are positively related to the use of futures and options. Consistent with the findings of above researchers, it also finds a positive relationship between organisational form (mutual or stock) and usage of futures and options. Another interesting outcome of the study is that the larger the firm's ratio of separate account assets to total assets, the greater the likelihood that the firm will hedge with futures and options. The study by Colquitt and Hoyt (1997) has similar limitations as the two previous researchers. Raturi (2005) fill the gap by conducting a research based on relatively recent data. His sample included 1939 non-life and 1013 life insurers and data was based on annual statements filed with NAIC. His findings confirmed the researches earlier. He found more number of stock firms using derivatives than mutual firms. Further, derivative use was more prevalent among life than non-life insurers. Around 16% of life insurers were found to use derivatives. But he makes an interesting point. The numbers of derivative users were decreasing as compared to the situation in 1994. He found that lack of financial expertise on derivatives and regulatory restrictions prevented most from using derivatives. Also, lack of familiarity with regulatory and accounting treatment of derivatives was cited as another reason. Further, an interesting finding was made by him. Life insurers were using derivatives more for hedging purposes as compared to non-life insurers which used more for speculation purposes. He argued that this could be because P-L companies have more equity investments than life insurers and thus may be writing covered calls for income generation. Though Raturi's (2005) data was relatively newer, it did not cover the various types of derivative instruments used and preferences among life and non-life insurers.
Hardwick and Adams (1999) were the first to conduct a study on the UK insurance industry. They conducted an extensive study on only the life insurance firms based in the UK. Their sample consisted of 88 firms selected randomly and data was obtained from Thesys insurance company database for 1995. These were annual statements filed with the regulatory authority (Form 13A) consisting of statutory returns and data on derivatives. The findings were revealed striking details. Around 57% of life insurance firms were found to use derivatives. This revelation was substantially different from researches conducted in US. However, their finding on stock and mutual firms was on similar lines as US. Stock firms (around 70%) accounted for more number among the users. They did not investigate extensively on types of derivatives and only mention that futures and options are majorly preferred. Further, they also reveal that the propensity to use derivatives is positively influenced by a firm's size, leverage and international links. Hardwick and Adams (1999) provide surprising information on use of derivatives by mutual firms. They find that 'capital poor' mutual firms use derivatives as a substitute for capital. They call for tighter external monitoring of internal control for these firms. Again, the findings are too old and the case of speculation is completely is ignored. Moreover, their findings are only concentrated on life insurers. Yung-Ming Shiu (2007) further conducts a research on the UK non-life insurance industry and empirically analyse the relationship between usage and insurer organizational characteristics with panel data for the period 1994 to 2002. His sample consisted of 360 insurers and data was obtained from Synthesys Non life insurance database. His findings showed consistency with US. Only about 12% were derivative users and was showing a downward trend through 2002. Futures were found to be majorly used, except in 2001 when options were found to be used more. The graphs below summarize his findings.
Yung-Ming Shiu (2007) reveals firm characteristics such as size, leverage, liquidity, interest rate risk exposure, line of business concentration and organisation form as major determinants for use of derivatives. Limited study on types of derivatives other than futures and options and concentration on only non-life insurers constitute the limitations of the study.
Moving outside US and UK, Xu Beilin (2007) investigates pension funds and life insurance industry based in Holland. He assesses the different approaches available to Liability driven investment (LDI) in varying interest rate scenarios and differing time horizons. His study concentrates basically concentrates on LDI strategies that use swaps and swaptions. He collects data from 1983 quarter one to 2006 quarter three. Xu Beilin (2007) reaches a few important conclusions. Firstly, swap and swaptions add value to add value to Asset-Liability management and the strategy to be used is highly dependent on the interest rate environment. Furthermore, swap and swaptions, with structured dynamic strategies, produce similar performance in changing and constant interest rate environments. He also concludes that premiums paid by policy holders can significantly reduce through the use of derivatives providing insurers with a competitive advantage in the market. De Ceuster et al. (2003) conduct an empirical analysis of the determinants of derivative usage as well as the extent of usage among Australian insurers for the period 1997-1999. They study a sample of 131 life insurers and 350 non-life insurers. The data is obtained from annual reports filed with the insurance regulating authority of Australia. They recognise 57% of total life insurers as derivative users. It is much larger than the US market but closely resembles the UK market. In the non-life sector, 13% of them are found to use derivatives significantly higher than the US figures. De Ceuster et al. (2003) reveal mixed results on the determinants of derivative use as compared to the findings of US and UK markets. While the determinants of derivative usage were size and leverage for life insurers, size and extent of long tailed lines of business written were major determinants of non-life insurers. In long tailed lines of business the length of time between the incident and settlement of the resultant claim is long. For example, worker's compensation claims continue for years before they are settled. Again, the determinants of the extent of derivative usage for life insurers were size and asset-liability duration mismatches while they were size and extent of long tailed lines of business written for non-life insurers. Their study, unlike previous researches, reveals an important point that determinants of derivatives differ for life and non-life insurers.
Ladekarl et al. (2007) go further in analysing the use of derivatives to hedge embedded options in the Danish market. The research published by the World bank performs a case study on three Danish insurers and uses their income statements and balance sheet for the period 2000-2005. After the introduction of various hedging instruments in 2001, pension institutions (includes insurance) have been largely successful in reducing their interest rate and equity price risk. They have been used significantly to reduce the asset-liability mismatch, however it is noted that it has increased exposure to counterparty credit risk. Ladekarl et al. (2007) also note that in early 2000s the creation of a comprehensive framework in risk management was imminent. However, the new regulations have impacted in an opposite and there are risks found in their operations than what was existent before the changes. She concludes that the changes have reduced the risk-taking capabilities of the pension institutions. But overall, use of derivatives has allowed them to expand product offerings consisting of embedded options without assuming a very risky position and exposure to interest rate risks. However, being a relatively new phenomenon in Danish industry, the authors' point that effective derivative markets, sensible practices and effective regulations would be vital in deciding the survival and sustainability of the approach.
A study of the Spanish life insurance industry by Lopez and Gonzalez (2005) was carried out using a sample 28 firms. They conduct an empirical analysis on the determinants with an emphasis on agency costs and maximization of the value of the firm. The information and data on asset-liability management in insurance firms were obtained from Insurers Cooperative Research Association. Though the sample was relatively small, they claim that it represents about 50% of the total Spanish market. Their findings identify 50% of the total sample as derivative users and a staggering 99% of them use OTC markets. Since most operations are long term for life insurance firms, organised exchanges are not preferred by Spanish life insurers. Practically all trading occurs with swap and forward rate agreements. Like Hardwick and Adams (1997) for the British market, Cummins et al. (1996) for the U.S. market, and De Ceuster et al., (2001) for the Australian market, they verify the existence of a positive and statistically significant relation between the size variable and the hedging decision, implying a greater propensity on the part of large life insurance companies to hedge with derivatives. They conclude that derivative use is not popular in the Spanish industry and most of the usage concentrates in hedging the risks involved in long term operations associated with life insurers. Finally, the results on the two phenomena 1) agency costs and 2) shareholder maximization are unclear. Lopez and Gonzalez (2005) observe that on occasions, it is difficult to establish a clear separation between the two, as the corporate risk management could reduce the costs of financial distress at the same time as it alleviates the underinvestment (drop of a positive NPV project due to leverage and conflicting opinions between bondholders and shareholders) problems. The above analyses by various researchers on markets, other than US and UK, mostly concentrate on the determinants and bear more or less the same results. Overall, there has been relatively very low emphasis on detailed analysis of different types of derivatives used by both the life as well as non-life insurers.
A theoretical and empirical study of the relationship between reinsurance and derivatives by Song and David (2008) describes two sources of capital shocks for an insurer namely fluctuation in value of the assets invested in and volatility in loss claims. Volatility in asset values shouldn't seem to pose problems for the insurer as majority of the funds are invested in bonds. But adverse changes in interest rates can hurt the cash flow of the insurer. On the other hand, sudden increase in claims due to catastrophes and disasters can substantially lower the available capital of the insurer. Song and David (2008) suggest external capital, through equity and bond offerings, as one choice to restore capital. The cost of raising capital through equity offerings is the most expensive, with Lee et al (1996) estimating an expense of 7% of the total proceeds. Though bond offerings are relatively cheaper, bondholders demand a higher premium after a major disaster. These costs can slide the insurer into further financial distress. Internal accumulated capital can be a choice; however Froot et al. (1993) calculate that the implied cost is the foregone positive NPV projects due to capital constraints which can be as costly as raising external capital. The further speculate that carrying surplus capital would be one way to avoid the constraint problem. Current regulations require that insurers hold a minimum amount of capital based on the underwriting risk they are exposed to. Hence for property and casualty insurers it is usually high. However, insurers are not ready to bear the cost of holding large multiples of optimal capital and face financial distress during and after a major catastrophe. In their empirical study of non-life firms based in US and data from 2000 to 2005, they find a substitution effect between reinsurance and derivatives. Another interesting result of their study shows positive relationship between credit ratings of the insurer and use of derivatives.
Going through the related literature on risk management strategies in the insurance industry, various research gaps are identified. Firstly, none of the researchers concentrate their study on the various types of derivatives used. As it is known, various derivative contracts such as futures, options, forwards and swaps exist that can be further classified into equity futures, bond futures, foreign exchange forwards, commodity forwards, interest rate swaps, foreign exchange swaps , swaptions and many more. A detailed study would help understand different forms/type of insurers more; life, a non-life, a stock, mutual and reciprocal. It will provide an insight to the degree of exposure to different types of risk specific to each type of insurer, the manager's concerns and tools utilized to hedge these risks. Further the derivative world has increased beyond imagination. Most of the previous studies are based on relatively old data which undermines the validity of the findings. Also, various new exotic forms of derivative instruments have evolved since then and hence it calls for a study to capture the new trends. It is also observed that most of the researchers have actually not tried to explore the motives behind derivative use which could be either hedging or speculation. They have carried out research with the assumption that derivatives are used as risk management tools. Further, the average extent of derivative use is found to be very low among insurance firms. While previous studies have explored the determinants of derivative usage, the determinants of their non-usage is mostly untouched.
This study aims to contribute in three ways. Firstly, this study explores the various types of derivative instruments used and consequently identify the characteristics specific to the insurer. Secondly, it explores the extent of use of derivatives for speculation purposes. This will further enhance an understanding of the firm's characteristics. Thirdly, it explores the determinants of non-usage and establishes relationships. To achieve the research objectives, an extensive survey research was carried out in the UK and US insurance markets. Since US and UK are two of the few countries where derivative markets are huge and available in various forms and insurance companies are allowed to use them extensively, this research study through the comparison of US and UK insurance firms will deliver more elaborate results and hopefully, provide a basis for future research studies.
RESEARCH METHODOLOGY
Definition of Conceptualization
Conceptualization involves three tasks: (1) identifying the variables for your research; (2) specifying hypotheses and relationships; and (3) preparing a diagram that visually represents the theoretical basis of the relationships you will examine (Hair et al. 2007)
Defining Variables
Variables are objects or characteristics/attitude that is measured quantitatively or qualitatively. In order to achieve the research objectives, several variables are analysed and subjected to various measuring techniques, which has been discussed later. The variables integral to this study are listed below:
Organisational form: - An insurer can be classified as either stock, mutual or reciprocal depending on the ownership structure.
Business Function: - An insurer can be classified as either life or non-life insurer depending on the type of insurances they offer.
Firm Size: - The firm size (in terms of total value of assets) influences various decisions of the company.
Users/Non-users:- The number of users and non-users of derivatives.
Various derivative instruments such as swaps, futures, forwards, options etc and their sub classes.
Motives: Insurers use derivatives either to hedge or speculate.
Determinant for non-usage: Various reasons such as unfamiliarity, fears or preference of reinsurance could lead to non-use of derivatives.
Hypothesis Development
Development of hypotheses involves the identification of the sample under study and measurement of appropriate variables. Based on the previous researches, the following hypotheses can be constructed:
Stock insurers have greater propensity to use derivatives.
More number of life insurers use derivatives as compared to non-life insurers.
Larger firms are more likely to use derivatives.
Futures and Options are preferred by UK insurers whereas Swaps are more preferred by US insurers.
Research Perspective
The research study takes positivist as well as an interpretivist approach. Positivist approach is like working with an observable social reality (Remenyi et al. 1998:32). It is positivist in the way that it researches the use of derivatives which is observable and real. Further, the researcher is independent of and neither affects nor is affected by the subject of the research (Remenyi et al. 1998:33).
Interpretivism advocates that it is necessary for the researcher to understand differences between humans in our role as social actors (Saunders, Lewis and Thornhill 2007:106). This research examines the thought process and motives of insurers controlled by individuals in taking decisions on the subject of derivatives.
Research Approach
The approach of this study can be classified as a deductive. It tests the hypotheses that were developed based on previous research studies by examining a sample of population of insurers based in UK and US. Based on the tests, the hypotheses are validated or invalidated and if required, modify it to reflect the findings. An important characteristic as observed by Saunders, Lewis and Thornhill (2007:117) is that there is a search to explain casual relationships between variables. Further, they state that the deductive approach means operationalising so as to present facts quantitatively. Both characteristics have implications for this research study.
Research Design & Methods
This study initially employs the cross-sectional design. Bryman and Bell (2007:55) define cross-sectional research design as It entails the collection of data on more than one case (usually quite a lot more than one) and at a single point in time in order to collect a body of quantitative or quantifiable data in connection with two or more variables, which are then examined to detect pattern of association. The research design is also commonly referred to as social survey research. This research conducts a survey among insurers in UK and US at a single point of time to detect relationships between various variables such as ownership structure, firm size, types of derivatives etc. and describe variables individually. However, a good response rate and a representative sample are the basic criteria for enhancing the external validity of the research. This research subsequently adopts a comparative design typically cross-national research design as it compares the findings between two different countries and seeks similarities and differences to achieve a deeper understanding of the subject.
The research uses both quantitative as well as qualitative techniques. Bryman and Bell (2007) discusses quantitative approach in a step manner. Initially, it uses existing theory and previous findings to build a hypothesis. The hypothesis is then put to test by analysing the data obtained. This research follows a quantitative approach by identifying the pattern of usage of various types of derivatives among different insurers (for e.g. life and non-life) and analysing them. This enables to identify the various functions performed by both life and non-life insurers and the resulting risks and then relating it to the usage of derivatives. The qualitative approach focuses on using the data on usage by stock, mutual, large sized and small sized firms and investigating the mindsets, motives and factors behind the findings. Also, examining the reasons behind non-usage of derivatives takes more of a quantitative-cum-qualitative approach.
Data Collection Methods
One of the main decisions before conducting a research is to select the most suitable data collection method that would improve the credibility of the research (Hair et al. 2007). The data was gathered using primary as well as secondary sources.
Primary Data
Self-completion questionnaires (Refer Appendix) were adopted for data collection as they are more standardised and allow easy comparison. Further, survey questionnaires allow more control over the research process and collection of large amount of data in an economical way (Saunders, Lewis and Thornhill 2007:138). Further, shorter questionnaires reduce the risk of development of boredom among respondents. Bryman and Bell (2007) observe that questionnaires are more convenient to respondents as compared to structures interviews.
During the summer of 2008, a request to participate in the survey along with the questionnaire was sent through express post to 256 insurers based in UK, of which 112 were life insurers with the rest being non-life. The post was directed to the Accounting/Finance/Operations department of the insurance firms. Subsequent follow up requests to 55 randomly selected firms (due to time and cost constraints) and initial post yielded a response rate of 18.75%. A few kind insurance firms wrote to us back informing of their inability to take part in the research due to various reasons.
During the same period, questionnaires were sent through post to 302 insurance firms based in US, of which 134 were life insurance firms with the rest being non-life. Following initial posts and subsequent follow up requests to 51 randomly selected firms yielded a response rate of 21.19%. Again, a few insurance firms kindly wrote back citing their inability to take part in the research due to various reasons.
The questions in the survey aimed to collect information such as organisational structure, firm size, use of various derivatives in the past one year and motives to analyse and achieve the research objectives. The questionnaire comprised of eight questions accompanied with instructions that were easy to follow and would take 5-10 minutes to complete. Five out of eight questions were closed ended with the rest being combination of closed as well as open. This avoided the possibility of absence of the respondents' choice in the listed ones.
Though the response rate was on the lower side, the respondents fairly portray a representation of the insurance industries in the two countries. This is argued on the fact that number of life and non-life respondents from both the countries did not differ largely. Further, though many respondents fell in the middle asset category of 250-1000m/$, the overall distribution was satisfactory.
Secondary Data Sources
Secondary data provides the researcher access to good quality data without investing much in cost and time (Bryman and Bell 2007). Secondary data provide useful information relating to the research and moreover, can be analysed and reanalysed as it is available faster.
The secondary sources used for this research were majorly reports published by Association of British Insurers, National Association of Insurance Commissioners and UK insurance index. These sources provided additional information on the performance of insurers and competition between them. It should be noted that validity and reliability issues relating to these secondary sources were not substantial since the authors were either direct representatives or regulators of the insurance industry in their respective countries.
Data Analysis and Presentation
The quantitative analysis of data was carried out using SPSS 15.0 and Microsoft Excel 2007 software for windows. Several variables were defined and subsequently subjected to univariate as well as bivariate analysis. The value labels, as inserted in SPSS software, for various variables are given below:
Organisational form: -
Stock Insurer - '1'
Mutual Insurer - '2'
Reciprocal Insurer - '3'
Business Function: -
Life - '1'
Non-Life - '2'
Firm Size: - The firms sizes were labelled from '1' to '7' in the same order as listed in the questionnaire (Refer Appendix)
Users/Non-users of derivatives:-
Non-users - '0'
Users - '1'
Participants in various individual derivatives were labelled as '1' and non-participants as '0'.
Motives: Hedging was labelled as '1', speculation and both were labelled as '2' and '3' respectively
Determinant for non-usage: The reasons (Refer Appendix) were labelled from '1' to '5' in the same order as listed in the questionnaire
A univariate analysis is carried out to calculate the number of stock/mutual/ reciprocal or life/non-life insurers using derivatives. Subsequently bivariate relationships are examined between different variables to understand the nature of risks and functionalities of insurers. For example, firm size is correlated with the users/non-users of derivatives to investigate if derivatives are preferred by a certain asset group. Similarly, correlations between individual derivatives and life/non-life insurers are examined to detect any possible directions or trends.
Qualitative data analysis employs the grounded theory approach. Bryman and Bell (2007:585) define it as theory that was derived from data, systematically gathered and analysed through the research process. It relates to the grounded theory that derivatives are effective risk management tools but is also used for speculation or income enhancement. Further the use of certain derivative is related to the degree of exposure to corresponding risk. The study provides a logical reasoning on the new trends and motives behind top level managers' decisions to engage in derivatives.
Sampling Strategy
According to Saunders, Lewis and Thornhill (2007:237), Purposive sampling enables the researcher to use their judgement to select cases that will best enable them to answer their research questions and research objectives. This technique is used to select the countries for comparison purposes. UK and USA are two of the highly liberalised financial markets of the world. They are regarded as economic powers of the world. Insurers are several in numbers, not restricted in their use of risk management tools, easily accessible and efficiently regulated in the two countries.
A simple random sampling technique, form of probability sampling, is then adopted to select the samples. The samples are selected from the database of UK Insurance Index and National Association of Insurance Commissioners for insurers based in UK and US, respectively. Simple random sampling was repeated, in case the address detail of insurance firm was not available. It should be noted that simple random sampling technique used here does not ensure generation of a representative sample since the samples are heterogeneous. Sample size was selected based on time and cost constraints and is subject to sampling error.
Validity, Reliability and Generalisability
Saunders, Lewis and Thornhill (2007) regard reliability, validity and generalisability as the basis for improving and achieving credibility. It informs whether the data collection methods and analysis procedures yielded consistent findings. The study investigates the use of derivatives, which is a critical issue and hence, the probability of incorrect answers is very low. Moreover, the respondents would not like to downgrade themselves by providing incorrect information since it can be also obtained through their annual reports and filings.
Bryman and Bell (2007:164) state that The issue of validity has to do with whether or not a measure of a concept really measures the concept. Validity associated with survey research design is devalued by few researchers who argue that various casual inferences can be unambiguously constructed. The variables used in this study are clear in the sense that they are what they seem to be. The correlations found between different variables in this study do not mislead or provide results that would be actually invalid. Since relationships are supported with logical reasoning.
This study does not attempt to generalise the findings to other parts of the world since regulations, social and economic factors vary widely. However, the implications of the findings cannot be discarded and may definitely find commonalities with studies based in different countries.
Ethical Considerations
The BES Ethics student handbook has been carefully read and the policies and principles mentioned have been noted.
The collected data has been treated confidential and secure. The respondents were thoroughly informed about the research objectives through a covering letter. They were informed and assured that the data provided would be treated as confidential at the highest level and would only be shared with the university staff. Further, the letter stated that by completing and returning the questionnaire, they agreed in principle, to participate in the survey. This study, through its findings, does not intend to defame an organisation or an individual but only aims to gain an in-depth knowledge on the subject.
I, in my full conscious state of mind, declare that the issues pertaining to ethics and privacy were understood and endured and shall not act or conduct myself in anyway that would hurt or dishonour an organisation or individual.
Research Limitations
Data Analysis and Discussion
UK insurance industry
The data collection process in the UK can be deemed as partially successful, keeping in mind the constraints involved in accessing and convincing the insurance firms to take part in the survey research. Out of the 256, 48 insurance firms took some time off from their gruelling schedule to answer the questionnaire set. Hence the response rate was approximately 18.8%. A further 19 were kind enough to send us note citing inability to take part in the research due to varying reasons. A mere 13 out of the 48 firms claimed using derivatives in the past one year. In percentage terms, approximately 27% of the insurance firms used derivatives in the past one year (Refer Appendix).
Stock vs. Mutual
Out of the 13 firms that used derivatives, eight were stock firms with the rest being mutual firms. In percentage terms, 25.81% of stock insurers (including non-users) used derivatives whereas it was 31.25% for mutual insurers (Refer Appendix). Though stock users of derivatives were larger in number, mutual insurers were found to have greater propensity to utilize derivatives. The result disapproves of the managerial discretion hypothesis formulated by Mayers and Smith (1988) which states that managers of stock insurers are given greater discretion to pursue risky projects and thus utilize derivatives to hedge, since the owners of stock insurers can enforce greater control over the managers through mechanisms available to them. They argue that owners can mitigate the risk of aberrant behaviour of managers through increased monitoring and control. It can be also argued that modern market imperfections would compel stock insurers to hedge risks so that benefits to overall firm value exceed transaction and management costs. Also, incentives in the form of stock options could compel managers of stock insurers to hedge risks. However, this results enforces that risk-averse managers who are weakly controlled by their owners and are not subject to the rules of market for corporate control, like in the case of mutual insurer, utilize derivatives with more freedom to minimize risks that threaten their job or to pursue projects which are highly risky but with positive net present values or enhance year end profits. Further, unlike stock insurers whose portfolios are well diversified and have more access to equity capital, mutual insurers might not have enough capacity and equity capital to survive adverse economic or nature's shocks and hence, would compel them to take measures to manage risks efficiently.
Life vs. Non-Life Insurers and Different Types of Derivatives
Out of the 13 derivative users, 8 were classified as life insurers and the rest non-life insurers. In percentage terms, a staggering 61.5 % were life insurers (Refer Appendix). It implies that more number of life insurers is generally looking to hedge risks as compared to non-life insurers. The usage of interest rate swaps is highest among life insurers (Refer Appendix). The major functions of a life insurer compel it to hedge interest rate risks. Interest rate risk is created when the durations of assets and duration of liabilities is mismatched. Interest rate sensitivity has increased considerably in the last ten years due to changing inflation rates and credit crunches. Furthermore, life insurances have introduced various new products in their portfolio of offerings such as universe life insurance, variable life insurance and annuities. They face the risk that policyholders will terminate policies or withdraw funds to take merry of attractive yields elsewhere. For example, cash value life insurances contain options that make insurers liable to prepayments due to competition from other financial intermediaries such as banks. Another form of debt instruments called Guaranteed Investment Contracts (GIC), bought largely by institutional investors, incorporate embedded options which when utilized during fluctuations in interest rate and economic conditions harm insurers to a large extent. It is evident from the findings as 88.11% of interest rate swap users are in the life business. A very little occurrence of interest rate swap usage (Refer Appendix) is found among non-life insurers since their liabilities are shorter term and do not face the risk of disintermediation. Moreover, since both life and non-life users primarily invest in public traded bonds, the long-tailed business of non-life insurer serves as a natural hedge against the assets, because as interest rates go up, the market value of liabilities goes down and so does the market value of assets. This doesn't occur in life insurance business due to the presence of embedded options. Hence, there is no scope for interest rate risk management in the case of non-life insurers. No use of interest rate futures and bonds is found since swaps are easier and more liquid than futures. Moreover, there are no margin calls and offers more flexibility in terms of the size of the contract.
Though, the use of foreign exchange swaps and forwards is higher among life insurers but similar relative frequency of usage which is on the higher side, is found for both types of insurer. While 6 out of 8 life insurers engaged in foreign exchange swaps (or forwards), non-life insurers were not far behind with 3 out of 5 using them (Refer Appendix). These figures are evident of rapid globalization of markets where insurers are selling and investing in foreign markets through foreign subsidiaries, to effectively diversify and capture the lucrative yields. As Berkman and Bradbury (1996) observe firms with overseas subsidiaries are more likely to use derivatives to manage foreign currency exposures. The norms of conventional theory calls for retaining risks which hold comparative advantage whereas hedging those risks which do not hold a comparative advantage. This directly applies to the case being discussed in the sense that domestic operators of foreign land are bound to hold a better risk-return profile than foreign operators and thus calls for hedging the foreign exchange risk. Foreign exchange forwards are preferred over foreign exchange swaps. Forwards market is believed to be bigger than swaps market as far as currency hedging is concerned.
Both types of insurers engage in a great deal of activity in the futures market (bonds as well as equity) (Refer Appendix). A benefit of futures can be related to, for example, the heavy investments of life insurers in Collateralized Mortgage Obligations (CMO). CMOs are special purpose entities that consists a large pool of mortgages. Investors in CMOs buy bonds and receive payments based on definite rules. One interpretation of these results is that insurers are trying to hedge the duration gap between CMOs and the relatively short term GICs. CMOs offer attractive yields but are generally illiquid. GIC policies are generally very highly sensitive to interest rates during the corresponding period. In the case of non-life insurers, most of the contracts are relatively shorter such as car insurances which are highly profitable lines of business. This could lead to a very broad equity duration gap and might be the reason for high usage of futures markets. Besides, large investments in equity also raise the need for futures.
The use of financial options is very little among the firms. It suggests that insurance firms find other derivatives more effective than options. Certainly, this is not true since options offer attractive returns while their downside is limited. This would mean that options markets are not yet fully explored by insurance firms. Other interesting observation is that no one has reported the use of complex derivatives such as swaptions, swap forwards, swap futures etc. Furthermore, not even one insurer has utilised catastrophe bonds or side cars. This implies that insurers like to keep it simple with the use of derivatives.
Large Size vs. Small Size firms
A positive correlation is found between the use of derivatives and the firm size. Larger firms tend to use more derivatives than smaller firms (Refer Appendix). This could be due to the fact that larger firms are more likely to take on riskier and wider variety of investments and hence, require a more effective risk management system. Furthermore, larger firms utilize economies of scale and hence have greater resources to provide training to personnel on the use of derivatives. Another interpretation could be the agency costs involved. Agency costs results from the contractual relationship and different in interests between shareholders, bondholders and managers. A broader definition would also include employees and sometimes stakeholders. As a firm grows, the operations become more complex and hence, information asymmetries between shareholders and managers are expected to increase. Nance et al. (1993) observes that hedging could be a cost effective as well as convenient way to reduce agency conflicts. As a classic example of agency conflicts is the underinvestment problem. This problem arises due to highly leveraged capital structure. The organisation would have to drop a positive net present value project since shareholders would contend that the profits of the project would largely benefit debt holders rather than themselves. Hence derivatives could be used to capture the value at least partially that would have been otherwise lost due to underinvestment.
A very interesting observation is that larger firms are found to utilize foreign exchange swaps and forwards to a greater extent. Larger sized firms are more likely to expand their base and offer products globally. Larger investments in foreign markets would call for increased derivative use for hedging currency risks. Interest rate swaps, futures and options show a more or less a similar inclination towards larger firms as well as smaller firms.
Hedging vs. Speculation
Most of the firms claimed to use derivatives for hedging purposes except for one which used it for speculation purposes as well. UK regulations forbid insurance firms to use derivatives for speculatory purposes. However Hardwick and Adams (1999) observe that the lack of transparency in the U.K. associated with the accounting for financial derivatives and their disclosure in annual reports, there is scope for life insurers to speculate on derivatives markets and for such trading to remain undetected. Hence, the choices stated by the insurers could be far from truth maybe especially in the case of larger firms.
Determinants of Non-usage of Derivatives
As discussed above, a positive correlation exists between firm size and use of derivatives (Refer Appendix). It suggests that smaller firms avoid the use of derivatives. It could be because of lack of resources to train and employ personnel to manage the derivative markets. However, a near zero correlation exists between the firm size and reasons for non-usage as outlined in the questionnaire. This suggests that reasons for non-usage are varied and is not associated with the firm size. However, a very strong positive correlation exists between business form and the reasons. It implies that for non-life insurers, reinsurance is preferred over derivatives. This is largely because non-life insurers are exposed to underwriting risks at a much greater extent than life insurers. The volatility in cash outflows due to property catastrophes (earthquake, tropical storm or terrorism), liability lawsuits and other events is high which could explain the motive behind preference of reinsurance over derivatives. Although catastrophe bonds form as substitutes, they have not captured the imagination of non-life insurers.
Further a high number of non-users cited Unfamiliarity with derivatives as the reason. Five of the 10 users have net assets below 50m while the other 5 lie between 250 to 1000m. It could be possible that the latter five might lie in the lower end of the asset range 250-1000m. In this case, it can be said that smaller firms are largely ignorant of the benefits and volume of derivative markets. However, this is only being speculated.
US Insurance Industry
The data collection process in the US was marginally better than in UK, with 64 out of 302 insurance firms taking part in the survey research. The response rate stood at 21.19%, which is partially successful. Further, 12 firms were kind enough to send us a note citing inability to take part in the research due to varying reasons. 39 out of 64 firms claimed using derivatives in the past one year. In percentage firms, approximately 60% of the insurance firms used derivatives in the past one year (Refer Appendix).
Stock vs. Mutual
Stock firms constituted 24 out of the 39 firms that used derivatives. In percentage terms, exactly 60% of stock firms (includes non-users) used derivatives whereas 62.5% of mutual firms used derivatives (Refer Appendix). Despite number of mutual firms using derivatives were less than stock firms, mutual firms have greater propensity to use derivatives. As in the case of UK, the result disapproves of the managerial discretion hypothesis of Mayers and Smith (1988). Risk-averse managers, as in mutual firms, look to utilize derivatives more because they are weakly controlled, not subject to market control and also fear loss of job.
Life vs. Non-life and Different Types of Derivatives
A staggering 71.79% of the firms which utilized derivatives were life insurers (Refer Appendix). This reinforces the fact that life insurers are exposed to financial risks to a greater degree than non-life insurers. Segregating into different types of derivatives instruments, life insurers utilize interest rate swaps to a much greater degree as compared to non-life insurers. Both life and non-life insurers use other derivatives to more or less the same extent (Refer Appendix) with no special preferences found for the two types of insurers.
Interest rate swaps are widely used by life insurers (as in the case of UK) to hedge the interest rate risks arising from the asset-liability mismatch due to the various policies offered by them. These include GICs, universal life insurances and cash value life insurances which provide policy holders various privileges. These privileges can be detrimental to life insurers as interest rates change and due to competition from other financial intermediaries. Swaptions, an exotic form of derivative, was also used by a few life insurers. Besides limiting the downside risk, it suggests that US life insurers are ready to explore alternatives for their risk management purposes.
Foreign exchange swaps and forwards are highly used by both life and non-life insurers (Refer Appendix). The rapid internationalization of markets and liberalization of other economies have exposed insurers to a high foreign currency risk. Foreign exchange swaps are preferred over forwards (unlike the case in UK).
A few life insurers use commodities forwards in the past one year (Refer Appendix). Since no direct relation is detected with the operations of life insurers, it seems to be used for speculation purposes.
A high frequency of use of futures and options is found in life as well as non-life insurers (Refer Appendix). The same implications as discussed in the case of UK can be applied here. Besides, as life and non-life insurers invest huge amounts in bonds and equity, futures and options could also be used for speculation (Discussed below in detail).
A few non-life insurers also used CAT bonds in the past one year (Refer Appendix). As CAT bonds are meant to hedge the risk of large property loss due to sudden natural or man-made catastrophes, their use is not unexpected. Frequent hurricane in parts of US, such as Hurricane Katrina, could motivate non-life insurers to utilize the benefits of CAT bonds.
Larger vs. Smaller Size firms
A very strong positive correlation is found between firm size and derivative use (Refer Appendix). This result confirms the findings in UK. Larger firms tend to have more appetite for risk, better resources to train/employ personnel and more agency conflicts.
Unlike in UK, foreign exchange swaps and forwards is not used largely by bigger firms. It shows a mixed pattern. Again, the motives behind it could be speculation.
Hedging and Speculation
Unlike in the UK, there is no regulation on the use of derivatives for speculation. Hence as expected, many insurance firms used derivatives for speculation (Refer Appendix). Though hedging motives dominated speculation motives, but the extent to which the firms engage in speculation cannot be ignored. Insurers are looking for various other strategies to increase income and they are totally justified. No particular trend for speculation is noticeable here with regards to firm size, business function or organizational ownership structure. It implies that decisions to use derivatives for speculatory purposes are purely managers' call based on his inside information and gut feeling.
Determinants of Non-usage of Derivatives
A very striking determinant of non-usage in US is Fears due to derivative related scandals. Various derivative related bankruptcies and corporate scandals in US and around the world have seemed to influence the mindset of US insurers. Interestingly, not many insurers in UK cited fears as their concern. An interpretation could be that scandals were highly publicized in US as compared to UK. However, as in the case of UK, many non-life insurers preferred reinsurances to hedge their risks. This is primarily because non-life insurers are more worried about underwriting risks due to the nature of their policies (Refer Appendix). Further, more insurers with asset value less than 250m$ came under the category of non-users. As discussed above, use of derivatives is positively correlated to size of the firm.
Comparison of findings with previous research studies
This finding reveals that mutual firms have a greater propensity to use derivatives. Previous studies (Cummins, Phillips and Smith (1996); Bouzouita and Young (1998) and Hardwick and Adams (1999)) have found stock insurers to use derivatives more extensively as compared to mutual firms. Further, consistent with the findings of Cummins, Phillips and Smith (1996), interest rate swaps are widely used by life insurers to hedge most of their interest rate risk. However, the usage of options was very low in UK which is inconsistent with the general observation by Hardwick and Adams (1999) i.e. futures and options are widely preferred by UK insurers. Raturi (2005) concluded that lack of financial expertise was cited as the number one reason for non-usage of derivatives. Inconsistent with the same, findings reveal that reinsurances are preferred hedging tools especially among non-life insurers. However, it should be noted that the previous studies were conducted using a large sample of insurers over a long period of time. This study limits its research to last one year and suffers from a much smaller sample due to time constraints and low response rates.
The descriptive statistics used do not exhibit large errors to alter an argument or opinion of a certain individual. The rounding of percentage and correlation values to two decimals is not expected to cause significant deviations in the interpretation of the results.
Conclusions
The objectives of this research are firstly to identify the trends in use of derivative instruments by insurance firms, identify the various characteristics or features specific to each type of insurers, investigate the determinants of non-usage of derivatives and finally to compare the findings of US and UK.
Overall, a greater percentage of insurers use derivatives in US as compared to UK. The derivative outlook certainly seems better for US insurer when compared to
