We will examine the captive insurance from various points of view. In Chapter 1, we define the principles behind a captive insurance company. We will outline the core idea, and how this is realized in the various forms that a captive can take. We will then outline the advantages and disadvantages of managing risks through a captive, and the role it plays in a company's overall risk management strategy.
Chapter 2 deals with the issue of feasibility. The question we try to answer is, how can a company know if a captive is a good solution for them? How can one evaluate the financial impact of the captive in the risk management? We follow what is called in risk consulting a feasibility study, which is basically a risk consulting product, done to see whether the captive insurance is fit to a company. We outline the general ideas and the key steps of such studies.
Chapter 3 deals with legal, fiscal and regulatory issues. We would like to premise that each topic is very delicate and intricate by itself; the purpose of our treatment is not to be complete and exhaustive, but rather to show how the changes in the legislation can deeply affect the captive treatment. One of the topics of concern is the choice of the captive domiciliation- a crucial aspect, which alone could make the difference between a successful and a failing captive. This is very sensitive to changes in the legislation, either at the parent's country or at the captive's. We will outline some of the most popular domiciles nowadays, and explain why they are more or less attractive than others.
Another crucial issue is the impact of the upcoming directive Solvency II: as the directive is still under approval as of the day this paper was written, what will be said should not be regarded as definitive, but rather as what is agreed on as to what will the potential consequences be. Finally, we will briefly talk about the impact of CFC - controlled foreign companies- laws on captive management. While the American (as well as the English) situation is rather clear, the same does not hold true in Italy: being captives not so widespread, there is a lack of legislation as well.
1 Captive insurance companies
A captive insurance or reinsurance company, or "captive" in short, is "an insurance company that insures all or part of the risks of its parent company" [Baw]. Basically, it is an insurance provider owned fully by the parent company (which is not an insurance company itself), which insures only the risks arising from the parent company (or of their affiliates: more on this later). It is indeed a form of alternative risk transfer, and can be seen as a form of self-insurance, since it is actually owned by those whose risks it insures. There are various criteria to classify captive companies, as new forms continue to develop. We will take Tagliavini's book point of view [TAGLIA], which is substantially backed up by other references ([MEED], [CHA]) and allows us to examine several distinctive features of a captive, while maintaining a unitary definition. The distinguishing factors are essentially the following:
by the way they operate,
which risks they insure,
by whom they are owned and whom they insure.
1.1.a Insurance captives and reinsurance captives
The first distinction regards the relationship between the parent and the captive company. To this extent, the captive can be an insurance captive, if it acts as the primary insurer to the parent's risks. In this form, the captive is by all means a traditional insurance company, so it will directly take the risks of the company, it will write the insurance policy and collect the premiums, and it will be subjected to the capital requirements and legal obligations of the country they are domiciled in. For this reason, they are also called direct (writing) captives [MEED]. As an insurer, it may- or may not- transfer part of the risks to commercial reinsurers. Tagliavini points out [Taglia] that in this perspective, a captive works as a link between the parent and the reinsurance market. Tagliavini also distinguishes three stages in the captive insurance program:
the captive writes and words the policy, the parent pays the premium;
the captive accumulates the reserves necessary to cover up for the parent's claims. We should remember that, effectively, it is the parent company itself who allocates the necessary capital in order to make the captive company able to start operating. Unused funds are either used to enhance the capacity of the captive company to absorb risks, or to strengthen the financial capability of the parent;
the captive may on the other hand place the risks on the reinsurance market, through agreements meeting the specificity of the risks covered. It should be noted that typically, the reinsurance market offers more favorable conditions than the local traditional insurance.
The more typical case is, however, that of reinsurance captives. The regulatory regime for traditional insurance companies requires an allocation of a large capital in order to meet the solvency requirements, as well as large amounts of documentation necessary in order to obtain the authorization from the regulator; reinsurance companies are subjected to more loose treatment. The parent company thus decides to use a local third party traditional insurer to first take the risk, then transfer it entirely (or almost entirely) to the reinsurance captive, through a so called fronting agreement (the insurance company, in analogy, is called front company). Other than being lifted from the regulatory requirements, the reinsurance captive has another clear advantage: suppose, in fact, that the parent company has a number of subsidiaries in different countries. Many times the regulations prescribe that the insurance coverage has to be bought only from admitted carriers (that is, companies authorized to operate in the country). By underwriting a fronting agreement with a local insurer, the parent pays a fronting fee and transfers the risk to the front company, which on the other hand, cedes most of the risk to the reinsurance captive. Points 2) and 3) made before still hold for reinsurance captives: the reserves are set up, the claims are paid (to the insurer, who pays them to the parent), and the captive might further cede the risk in the reinsurance market. We will go further into the fiscal benefits arising from captives. We should point out, however, that in many countries insurance premiums are tax deductible: if the reinsurance captive is set up in countries with a favorable tax regime, the parent will factually collect back most of the premiums paid, and save a significant amount of taxes, since the captive will pay taxes on the premiums according to the domicile's regulation. captive reinsurance (better) Chartis.jpg
Figure Source: Chartis [CHA]
direct captive.jpg
Figure : Source: [MEED]
1.1.b Pure captives and open market captives
As already mentioned, the captive company is made to insure very specific risks, which are difficultly placed in the traditional insurance market. Therefore, the services of the captive insurance can be thought of as specific for the parent company. However, it could be that similar companies with similar risks to the parent's could access the services of the captive.
In this perspective, we say that a captive is a pure captive if it only covers the risks of the parent company and direct affiliates. When access is granted to a wider audience, it is called open market captive (or broad captive) [REF!], covering also third party risks. The latter case is however much less common than the former; pure captives are by far the more common type. Nonetheless, once the captive has reached financial and operational maturity, and if the parent's risks are well covered, adding additional third party risks could help diversify the insurance portfolio and increase the income from premiums, making the captive become an important source of income for the parent. Usually, this is done by cooperation among other captives, thus sharing parts of the market, or by coinsurance, with traditional insurance companies. See also [ART].
Tagliavini [TAGLIA] puts up a caveat: if the captive does not carefully examine the risks it takes, it might end up being overexposed to risks, which might pose a threat to the very survival of the captive itself; secondly, having to open the business to a wider market means losing the exclusivity of the captive-parent relationship in favor of a more commercial approach; therefore, the parent might lose control on the policies of the captive to favor its insurance market.
1.1.c Single parent and multi parent captives
Another classification can be made from the point of view of the ownership. If the captive is fully owned by one parent company, it is called single parent captive. This is the most common form of captive insurance, and they are usually pure captives, as their only activity is to cover one specific risk or a small cluster of risks. As it can be very specific in the insurance service it delivers, fitted exclusively on the parent's insurable risks, it is very flexible, highly customizable and it has the least costs.
If the owner is not only one company, we have a group captive or multi parent captive. Within the family of group captives we can highlight several different structures (see figure, adapted from [TAGLIA]). The risks insured by the captive can be homogeneous- which entails that its associates have a similar business and exposure to risks, or heterogeneous. According to this distinction, the captive can take different forms.
The "typical" case is that of a group of companies- or more in general, a group of individuals exposed to similar risks, which cannot find proper insurance coverage in the traditional market: maybe because the risks are too specific and traditional insurers cannot provide with the correct premiums, or if they do, they are generally much higher than the theoretical exposure to the risks. On the other hand, they would not have the means of gathering the capital necessary to set a captive company up by themselves; therefore, they decide join efforts to reach "critical mass" and share the costs of setting up a captive insurance company jointly, exploiting the advantages of economy of scale, into a so called association captive. A typical example, quite common in the United States, would be that of a captive insurance to cover the professional liability risk of surgeons, which are highly exposed to claims for medical malpractice. [MEED] distinguishes between association captives and industrial captives, which are captives set up by a group of similar industrial firms being exposed to the same type of risks. We shall however regard these two types as one.
Along similar premises is the risk retention group. It covers liability risks for the party members, but it specifically excludes first party coverages- that is, property, workers' compensation, and all personal lines. These were introduced by the Liability Risk Retention Act in 1986 as a response to the world wide crisis of third party liability insurance in 1985, when access to coverage reached unsustainable prices and restrictions.
A captive pool is an association of separated captive companies, joining forces to create a larger mass of risks (usually pure property risks) and increase diversification. Captives exchange with each other a share of the insurance premiums they collect from their parent companies, in return for an equal amount of premiums from non-affiliates. The nature of the risks is the same; however, the quantity, quality and geographical distribution will change. This will allow for phenomena of risk diversification in each captive portfolio (even if they will be handled in exactly the same way as a normal captive would). Normally, a captive pool is managed by either insurance companies, or specialized captive management societies.
An agency owned captive is not owned by a parent company, but rather is set up by a pool of insurance agents- which are, again, too small to set up their own captive. In this setup we can distinguish three actors: an insurance company, a group of agents, and a reinsurance company owned by the agents. The process follows this structure: the agent underwrites a certain volume of business with the insurance company. If the volume of contracts is sufficient, the insurance company transfers some of the underwritten risks to the reinsurance captive. A variation of this form is the joint venture captive, where the insureds are also shareholders of the captive.
Another captive solution dedicated to those enterprises who would not have the means of starting one by themselves is the rent-a-captive company. These are insurance companies, usually set up by sponsoring organizations (other insurance companies, or insurance brokers) who provide the necessary capital to start up the captive, then (as the name indeed suggests) rents it to the insured. The premiums received from each insured are capitalized separately from one another and are accounted for independently, so that the claims from one insured has to come from the fundings allocated for that insured only. This solution is advantageous for small and medium sized enterprises, who cannot setup a captive or just want to experiment with having one, without actually building it yet. The rent-a-captive owner, in turn, can gain from the renting fees and from the increase in premiums and diversification, due to the captive underwriting.
A Protected Cell Company (PCC) is somehow a generalization of the rent-a-captive mechanism. It is constituted as a single, standalone legal entity, made of a "kernel", or core, an entity containing the core assets, and a number of satellite cells, each one with their own cellular assets, and each works as a captive insurance company. The cells are managed centrally, by a board sitting at the core level, but each cell is completely disjoint from another, so that no cell can fund the liabilities of another one, but the core keeps all the assets which cannot be allocated to one cell alone. However, the PCC constitutes a single legal entity, so that the cells do not have a legal identity of their own. The creation of a new cell is decided by the core management, so that a potential new cell owner has to sign a cell management agreement with the board, tailored according to the requirements of the cell future owner. The clauses will include specific entry and exit regulations, as well as the compulsory regulatory requirements.
The main advantage of the PCC structure, as with many of the group captives, is that it allows smaller companies to access the captive solution without the large amounts of money required to start one up on their own: the "rent-a-captive" solution is usually structured as a PCC.
1.2 What a captive does: advantages and disadvantages
The formation of a captive is, essentially, an ERM and financial strategy. Understanding what a captive can do will also describe its advantanges and shortcomings.
1.2.a Insurance advantages
Industries facing very specific risks have a difficult time transferring them to the insurance markets. Very specific risks- to name a few, liability risks such as environmental liability, professional indemnity, or D&O liability for very exposed companies, property risks for complex industrial plants, are very difficult to place in the insurance market or require very high premiums, regardless of the claims history of the enquirer. This due to several concurrent reasons: first, the companies lack the actuarial knowledge of the risk necessary to quote the correct price for the coverage- if the risk is very new, or unusual, there is not enough experience to back the premium calculations up with statistical evidence, therefore the companies need to be conservative. Secondly, as it is known, the insurance market is cyclical: when the market is hard, coverage of certain risks is severely restricted and premiums increase, also dramatically. Captives can avoid or reduce cyclicality, since it is less influenced by external factors (fluctuations of interest rates, changes in the overall insurance market and so on) which have a dramatic impact on the market. Thirdly, a captive company is not exposed to moral hazard and adverse selection risks, which will therefore not penalize the evaluation of the losses.
A captive insurance, on the other hand, can be thought of as an entity of "dual" nature: it is a self insurance solution in the form of an insurance company. The parent company knows the risk perfectly (at least in principle!) as far as frequency, severity and consequences of claims go, it can put its expertise on the risk at the service of the correct premium calculation and the most appropriate reserving process to, basically, insure itself. Being also an actual insurance company, the captive is granted access to the international reinsurance markets, "the wholesale market of insurance" [Taglia]. As such, it tends to me more favorable than traditional insurance markets, and can afford to cover larger families of risks. Thomson et al. [TH] also point out that
"Reinsurers like to accept retrocessions from captives because they can take comfort from the fact that the insured is itself financially involved, through the captive, with its own risk."
Moreover,
"[…] most captives are managed by experts who themselves have ready access to both the traditional reinsurance market and also to captive retrocessionnaires, who provide an excess coverage which is some times difficult to achieve commercially."
One way of optimizing the risk retention/transfer strategy is to hire a broker, who, with the aid of his/her expertise and direct connection with the insurance companies, may help finding the best combination of coverage. However, the broker requires a commission which could reach 50% of the premium share: the saving is then severely damped. A captive on the other hand has direct access to the insurer itself, so it may strip down the premium of the brokerage expenses as well as of most of the commission expenses from the insurer itself. It should be kept in mind, however, that the captive also has wages and commissions: therefore, all these consideration hold true if and only if the captive expenses are significantly less than commissions for traditional insurers and brokers.
If a company operates internationally, the access to reinsurance market and the ability to tailor coverage to the specific risk make the captive a centralized risk control tool for all the branches of the company worldwide. The captive designs the basic insurance plan, called master policy, in a way so that (adapted and translated from [Taglia]) "it allows to integrate and bring to the desired level the multiplicity of clauses and conditions written by local policies". This is achieved by adding to the master policy a "difference in conditions" clause: by this clause, the master policy will integrate any gap contained in the local insurance policy which works unfavorably to the insured. As a side effect, moreover, "it can achieve these objectives in the currencies held most convenient for claim settling and premiums payment by the parent company."
1.2.b Risk management advantages
The ability of the captive to manage a risk is one of the key advantages of instituting one, since it improves the quality of loss monitoring and loss prevention. The actuarial activity connected to the premium evaluation (keeping track of claims frequencies and severities, reserving) is essentially a loss control activity. Since losses are ultimately retained, the captive will incentive building a stronger risk awareness and willingness to contain the risk itself. The very fact that the premium calculation and the reserving process requires a detailed knowledge of the risk should make the company more risk aware, thus mitigating the common effect of neglect mitigating a risk once it is insured. Thomson et al point out that
" a captive is a long term funding vehicle for which financial statements are prepared. […] Thus, the parent company's risk management programme can more easily be monitored and evaluated on a financial basis. […] Companies which have instituted safety and other loss prevention programmes have seen claims steadily reduce in size and frequency, with the result of either a build up of surplus, and therefore increased capacity in the captive, or a reduction in insurance premiums."
In virtue of what we observed before, they confirm that
"such companies have developed technical loss control expertise and in many cases significantly reduced the legal costs associated with several types of loss."
Of course a captive alone is not a sound enterprise risk management strategy. However, once embedded in a overall risk management program, it becomes a flexible and powerful tool. If the risk is evaluated correctly, the captive will increase its capacity over time, becoming able to cover bigger risks, therefore capable of offering more insurance solutions to the parent, as well as more profitable. The captive also allows [Thomson]
"to design allocation systems to distribute loss costs more equitably among profit centres, to implement uniform accounting procedures, to accumulate actuarial information, to design more effective claims handling, loss control and engineering programmes, and to unify their application throughout all divisions or subsidiaries of the company."
It is often pointed out ([Taglia], among others) that the risk management strategy of the use of a captive can be seen either as a risk retention strategy, or as a risk transfer strategy as well. It is risk retention if we consider that ultimately the risk never leaves the parent, since the captive is an affiliate to it, the premiums and the claims payments are still done within the company, resembling thus a "formalized mechanism to finance risk self insurance" [Taglia]. It is, however, risk transfer if we recall that the captive will transfer the risks undertaken to the reinsurance market. Bottom line, the central principle of transferring the risk to the captive is: retain the risk you are able to cope with, only transfer the rest at the best premium you can. Many firms can retain more risk than they do; if they perform better than average at managing the risks they face, there is no reason to transfer bearable risks to the insurance market (since the price will be higher than the expected retained cost); what exceeds the risk capacity can be transferred via Excess Loss or Stop Loss agreements.
Finally, a key advantage is given by the additional negotiation power the captive gives in the traditional market. A successful captive is capable of absorbing a large share of the parent's risks, at a price lower than what the market offers. Therefore, an insured can switch from the market to the captive whenever the premium gap is deemed as unacceptably high: thus, the insurer will lose the client- possibly forever. Wilmington Trust [WIL] paper registers that "it is not unusual to see an insurer adjusting its rates downward when faced with the likelihood of losing a piece of business to a captive, since once a risk is insured in a captive, it very rarely returns to the market." This effect makes it so that the parent has lower insurance prices not only because the captive charges lower premiums, but also because traditional insurer will charge lower premiums because the captive is there to begin with.
1.2.c Fiscal and legal benefits
A captive is also a financial vehicle. Since the insurance process is dealt with internally, the parent can benefit from more flexible and tailored premium payment and faster claim settling. The latter is often very relevant: especially when large claims occur, in traditional insurance it could take a long time before the claim is paid. With the captive, the settlement is done internally and much faster, thus dampening the impact of the claim to the cash flow levels. Moreover, since the premiums collected from the parent are actually still within the parent's assets, they can be invested and accumulate interest (while they still have to be readily available as reserves, they are not "lost"). In the reinsurance market, premiums are often paid in arrears, therefore by transferring to the reinsurance market through the captive one can further improve the cash flow levels.
The domiciliation of the captive entails further side advantages. One should always be well aware that a captive is an organic part of a risk management strategy and not merely for tax reasons; that said, a careful choice of the country is crucial to success. We will explain the reasons for this statement here; in Chapter 3, we will make an overview of the current legislation and list some popular choices for captive domiciles.
As a rule of thumb, domiciles are chosen among those countries where:
- insurance legislation is looser: when the control is lower a company can write more business, or use conditions that would not be permitted elsewhere;
- tax levels are lower: the development of current legislation (CFC laws, in particular) made it so that it is unlikely that tax benefits become a major advantage for captives. In the United States, for example, there have been court rulings saying that if the premium payment can be interpreted as self insurance. However, the location has to be chosen so that taxes do not conversely become a burden or rise problems.
- credit is easy to access: another rule to choose a captive domicile is to select those countries where loans are granted with lower interest rates and investment funds are readily available.
1.3 disadvantages of a captive
Forming, maintaining and operating a captive (especially at the beginning and when it is not yet financially autonomous) entails significant expenses. Usually parent companies do not know about captives, therefore they hire a consultant (most of the times, captive management specialists, or insurance brokers) to run a feasibility study to determine whether it is advisable to constitute a captive and how to do so. Then the funds necessary to meet the regulatory capital requirements to operate the captive need to be allocated. These funds will therefore not be available for the parent to invest or use.
Whatever the domiciliation might be, there will still be legal and regulatory requirements to be met, which will of course add costs. Moreover, upkeeping the captive brings management and organizational costs, domicile taxes and auditing fees, not to mention management time and possibly travel costs.
Once the expenses to start the captive are dealt with, it is fully operational, but not yet financially strong. If the claims follow the average trend outlined in the actuarial calculations and the captive is ran well, it should gradually gain strength and size, becoming a profit center for the parent. However, if the claims were evaluated incorrectly, or if several large claims happen one after another, the reinsurance treaties underwritten by the captive will have to cover the excess of losses. To quote Tagliavini:
"Broadly speaking, the reinsurance market reacts faster than direct insurance in adjusting rates when very large claims occur. Therefore, if the captive places the excess risk in the reinsurance market, periods with very high loss ratios will strongly impact on the reinsurance premiums, which are much more sensitive to losses."
This means that the impact of large losses on the captive activity will not be only temporary, as it will carry on for the following years, possibly resulting in losses. The parent's management needs to be aware of this possibility, keeping in mind that together with the captive goes the need of careful choice of the risks to transfer, a close attention to those risks, and a thorough loss control activity.
2. Setting up and monitoring a captive
From what we observed in the previous chapter, a successful captive relies on several prerequisites:
- a thorough knowledge of the exposure to risks;
- a correct risk transfer strategy;
-an assessment of how the captive will impact on the current insurance program;
- a solid captive structure and strategic management.
We will outline the steps a company should take to meet these requirement. The structure of the chapter will integrate parts of a mock-up feasibility test, which will hopefully serve as a beacon. The test will be simplified to its bare essentials, and consist of entirely made up data: no specific assumptions will be made about the company under study; there will be no particular specification of the industry (so the risk can be of any kind) and the historical loss figures will be made up. Due to structural limitations, the actuarial model will be also simplified (in particular, the loss severity distribution will be very simple, to ease the simulation process). The general references are [Wil], [Marsh], [Taglia] (chapter 5) and internal documentation, unless specific documentation is mentioned.
2.1 choosing the risks
The first step is to decide what risk(s) to transfer to the captive. In order to do so, however, it is essential that the company has a clear understanding of its risk exposure profile. A risk register; coupled with a risk map, can help better understand the risk situation. A risk register is basically a list of potential risks, categorized by several criteria such as the source of the risk, the type, the resources and stakeholders involved, and so on.
project_risk1.jpg
Figure , sample risk registry, source:http://projectshoptalk.files.wordpress.com/2012/04/project_risk1.jpg
A risk map is a classification of risks according to their frequency and severity. The assessment of these factors can be either based on historical losses, or based on expert judgment: either way, besides helping the visualization and the synthesis, the risk map helps determining whether a risk should be retained (green area in the graph below), controlled or mitigated (red area) or transferred (orange area).
WYCRR Heat Map.JPG
Figure , risk heat map. Source: http://www.westyorksprepared.gov.uk/files/WYCRR%20Heat%20Map.JPG
Many times companies hire professionals to collect this data; while it is certainly not necessary, it is crucial that the picture is as accurate and as complete as possible, since a rational choice of the risks to transfer will rely on these mappings.
Once the risks are identified, the next step is to examine carefully the financial statements, to picture the financial strength of the company, the availability of resources that would be needed to start a captive, and the exposure to risks of the key performance indicators, usually on a time horizon of five years in the past. Together with this, the current insurance program needs to be outlined and examined: the policy wordings, the risks covered and retained, the deductible and maximum coverages must be put together to portray the insurance plan (usually of the last five years). These figures are necessary to determine whether the company is sufficiently large, has enough financial stability, enough premium volumes to justify the need for a captive. This can be thought of as a pre-feasibility study. In lack of these essential prerequisites, there would be no need to further proceed.
The choice will depend upon the impact a risk has on the financial health of the company and on the current insurance programme structure. A reasonable choice is to transfer those risks whose frequency is low but have high severity, and are transferrable to the reinsurance market. We pointed out before that the captive is weak during its first years of life, therefore, as Pisani and Potrich [ART] advise, it would be wise to first transfer risks with low probability and medium to high impact at first, then, once it is well established, increase the deductibles, and add the very high losses and the long tailed risks (claims with medium to high severity but very long settlement time- useful because the long time allows for strategic allocation of reserves). Following Tagliavini, we outline another useful criterion to prioritize the risks to transfer to the captive. First, cover the risks that are not insurable in the traditional market, or are insured but with very high premium rates. This will spare from having to pay high taxes on the internal funds necessary to self insure the risk, and will give the immediate advantage of the better price for the coverage. The already insurable risks can be kept covered with traditional insurance at first, in order to allow time to build appropriate reserves; once the captive has gained strength, the parent can start expanding the coverage to those risks and build a more comprehensive coverage plan. The following two graphs below (source: Marsh) are the result of a survey indicating the most common risks transferred to a captive. As we can see, the largest part is made of property, liability and employees' benefits. and common sources from whom to seek consultancy. The common features of typical captive risks are, in essence, the following:
1. the volume of claims is sufficiently high, and within the primary layer or the working layers of coverage: the "best" risk is the one that claims (no claims mean no business for the captive, after all) but within the foreseeable limits;
2. the volume of premiums is sufficiently high, as to offset and justify the underwriting, management and maintenance expenses of the captive;
3. the risk is reinsurable, because of all the advantages of the reinsurance market over the traditional market we have underlined before.
graph.jpg
Figure , source: adapted from internal
As a rule of thumb, we can say that a volume of 2 million $ insurance premiums would be an adequate indicator of the lower bound for a company's activity. At the stage of the pre-feasibility study, the domiciliation of the captive also needs to be narrowed down. We leave a more detailed analysis of the most popular ones to the next chapter; however, we can start nailing down some of the issues involved in such an evaluation. We can outline the most common decision criteria to the following [source:Marsh, internal documentation]:
1) low capital and solvency margin level requirements;
2) low operational costs;
3) flexible capital structure and investment rules;
4) easy access to loan funds;
5) currency;
6) direct writing capability in relevant countries;
7) captives are able to write insurance and reinsurance business;
8) proximity to the parent company;
9) possibility of writing fronting treaties with the country;
10) strong insurance and reinsurance market;
11) low corporate taxation rate.
2.2 Risk tolerance and risk appetite assessment
The concepts of risk tolerance and risk appetite are increasingly popular in risk management and risk consulting. Risk tolerance can be defined as the amount of loss that a company can face without having to restructure or change the business activity. It can be quantified by examining the impact of losses on certain key performance indicators, that is, key financial figures (some of the more common are net income, leveraging, net debt, cash flow …) which are thought of as representative of the company's performance and financial health.
To make a concrete example, we quote some commonly accepted risk tolerances with respect to common items in the financial statements. They can also be found in [TAGLIA], but are also quite common in risk consulting [1] :
KPI
Low variance
High variance
Net working capital
3%
5%
Net income
0.1%
0.3%
Earnings per share
3%
5%
Net assets
0.1%
0.3%
Revenue
0.1%
0.3%
Risk appetite, on the other hand, is the amount of loss a company is willing to face: this can be thought of as a "gut feeling" of the management (the CEO and CFO and the risk manager, in particular); as such, it cannot be quantified using financial figures, but it is rather estimated by the managers themselves. While the two concepts are distinct, they should-in principle- point to the same range of values. The larger the difference, the more inconsistent is the gap between the risk attitude with the risk strategy.
In the field of captive management these two concepts play a very important part: in the previous section we showed how to identify the risks to transfer or not to the captive. However, not all risks are retainable in the same way by different firms. A very risk averse company, for example, will still prefer traditional insurance in exchange of a lower exposure to a certain risk; the insurance program (with or without captive) cannot overlook this aspect, as it would be meaningless to cover much more (or much less) than what a company would or could want to sustain. As such, the retention level of the captive will have to comply with the level of risk tolerance.
2.3 Loss analysis and loss projection
In the previous steps, the candidate risks for captive coverage were identified, and historical losses were gathered. The assessment of the risk tolerance formalized the ideal upper bounds to the risk retention levels, which will serve as a guide to design the overall insurance program later. The following step is the loss modeling. This step consists in the examining the time series of losses, the frequency and severity of claims and the identification of possible time trends. This step is a continuation of the risk register and risk map introduced before, and it is very useful regardless of the feasibility study, since it helps to profile the risk in a quantitative-rather than qualitative only- way and to detect possible alarms from increasing trends: it is an exercise both in actuarial technique and in risk management. As far as the captive study goes, the step forward is the loss projection: assuming that the historical losses portrait the risk profile of the company, using mathematical modeling we generate possible future loss scenarios arising from different initial assumptions: we can assume an "average loss" type scenario to determine the most likely exposure, then a "high loss" scenario to see what the impact of stress would be on the company.
Let us suppose that company X faces a risk whose loss history in the last 7 years period is given by the following graph:
From this first screening phase, some conclusions may already be drawn: for example, if risk Y in comparison has a loss history as follows:
While risk X averages 1920K$/year, risk Y amounts to about 90K$/year; most likely, the company would be advised to retain risk Y, since insuring them would bring an unjustifiably higher cost, while retaining the entire risk is likely to be bearable for the company. This simplified example is to show, that a close look at the loss history can eliminate some candidate insurable risk on an early phase. Let us assume from now on that risk Y will be retained fully and that risk X is the most likely candidate for captive transfer. The next question is, how to structure the possible new insurance program. In particular, the study needs to give an estimate of:
the deductible amount: so that the class of low, frequent claims is not transferred to the captive (which would imply a much higher premium and greater stress on the newborn captive financials) ;
a consistent layering of the loss amounts: presumably, the insurance/reinsurance program will work on multiple layers, some of which will be retained at the captive level while the loss in excess will be transferred to reinsurance; therefore, an "optimal" insurance plan will consider the layering structure that is both compatible with the market offering, and gives the best insurance premium and coverage;
the technical premium, which will be evaluated under the aforementioned assumptions, which will be the primary indication of the market pricing of the insurance program.
It is useful to divide the claims into layers, so to understand which levels need to be transferred, which will be retained by the company via the deductible, and which will be retained by the captive. To make a very simple example we refer again to risk X.
Looking at the graph, we can see that the low and frequent claims fall within the range of <250K. Therefore, a reasonable deductible would be in the range of 100-250K: let us assume it will be 200K. Moreover, there are no claims above 5000K: therefore, as a rule of thumb, we could reasonably expect a captive retention limit in the range of 5000-6000K, according to the availability of the market, the risk appetite of the company, and the current risk transfer plan. By no means these decisions are made only by looking at the historical claim assessment: we should always keep in mind that lower deductibles will make the insurance premium increase fast. On the other hand, very high levels of captive retention will lower the insurance premium but it will increase the capital requirements for the captive: many of the most popular captive domiciles are planning to adopt the upcoming Solvency II regime (more to come on this aspect on Chapter 3), Luxembourg, Dublin, Bermuda Islands to name the most relevant ones. As the value at risk increases, capital requirements will also grow, meaning that the company will have to retain higher margins in the captive as solvency capital. Therefore, a good choice of these levels cannot overlook these considerations.
Tagliavini points out that the convenience of establishing a captive is tightly related with the past claims/premium ratio. While there is no precise ideal ratios, several consultants identify the so called "club under 30", made of those company with claims/premiums ratio lower than 30%, who are the ideal candidates for captives, and the "club plus 100", who are on the contrary unfit to consider captives in the first place, as they would cost them more and more to maintain.
The next step is to evaluate the so called as-is versus the as-if scenarios. In practice, this amounts to estimating the losses that would occur if the insurance plan is kept the way it is (as is) against switching to constituting a captive (as if). the total (theoretical) price of the risk transfer. This will be made of two components:
Transfer price= Cost captive risk () + cost reinsurance transfer ()
As we mentioned, the term will depend on the historical losses and on the deductible, whereas the reinsurance transfer cost will depend again on the risk, and on the chosen transferred layer. If one were to carry on the study of our hypothetical company, he would make an actuarial model of the loss distribution, or use software tools like SAS, @Risk, and others to make a reliable fitting. It is however not our purpose: what interests us is the procedure, rather than the (still fascinating!) technicalities.
To evaluate and , we need to evaluate the expected losses and to make reasonable assumptions on the market pricing, such as the safety loading coefficient and the loading for expenses. There exists a range of values commonly accepted in the market, so these coefficients can be estimated given enough knowledge of the market (typically, a specialist would run these calculations, so they should be known). As far as the expected losses summand goes, one could use the same principles used in actuarial technique: model explicitly the severity and frequency of the claims, or generate statistics using Montecarlo simulations . Either way, the following are needed to calculate a likely estimate of the technical premium:
an expected cost per layer,
tail events, expected to occur, say, with probability 20%,10%,5% and 1%.
2.4 Business plan and differential analysis
Given the projection of losses in the future, the financials of the company, the choice of the domiciliation, we now see how those can be combined together to make a decision tool.
To do so, we generate mockup financial statements of the captive to be. The procedure will be no different from writing financial statements for a normal insurance company: what interests us here is, however, how the financials of the captive company will impact on the parent's company financials. Therefore, it could still be useful to go through the steps needed, highlighting where the study made in the previous steps comes in handy.
Revenue and profit and loss account
This part will consider:
Premium income: given by the algebraic sum of the gross premium income paid by the parent to the captive (estimated in the previous section); less the fronting fee; less the reinsurance premiums. The total is the net premiums written.
From the net premiums written we subtract: the amount of claims paid in the year (using the outputs of the simulations also made in the previous section), the claims reserve, and we add back the reinsurer share, and accruals.
The result is the net underwriting profit/loss. The profit and loss account starts from this figure, to which we add the investment income, and subtract all the other expenses: administration expenses, management costs, tax (based upon the choice of the domiciliation.
The (projected) result is the retained profit/loss profile for the time period under consideration, and it will serve as a basis to evaluate the performance of the potential captive, as well as the eventual needs for allocating capital to meet the solvency requirements.
Once the solvency capital is evaluated, if needed, we can write down a projection of the projected balance sheet. This part will be similar to any other insurance company's: the asset side will contain cash, deposits, fixed assets, and so on, whereas the liabilities part will contain mostly the claim reserves (as well of course as other eventual liabilities, debt, accruals and so on).
The resulting equity will be built of the share capital, plus the retained profit evaluated in the profit and loss account.
The final sheet is the actual benchmark: we project the income statement of the parent company in the time horizon basing upon the information gathered before and compare them with the as-is scenario. In the current situation, the company has the following expense for the insurance program:
,
Where is the total expense, is the amount of premium paid, are the taxes on the premium payment, is the expected risk retention from the parent and is the tax deduction. [2]
If there were a captive, we would have instead:
.
Where the notation is obvious [3] ; the new term is very important, because it represents the captive net income (which is calculated as retained by the company) calculated in the financial projections for the captive. Naturally, the captive is advantageous if There are two opposite trends: generally, , since the captive principle is to retain more risk within the firm, the insured layers will be higher, so will be the premium; for the same principle, and, since premiums are higher, also . On the other hand, : a crucial choice is that of the domiciliation; if the tax regime is favorable, the difference will be substantial. Moreover, the determining factor here is First of all, some of the paid premium returns to the parent through the captive; not all of it, because of the reserving and the expenses. Nonetheless, the gap between the old and the new insurance premiums is partially mitigated. Furthermore, if the claims are moderate, the parent will register a profit through the captive profit. Given the crucial impact of , several different scenarios are presented, typically answering to these questions: what would happen if we chose domicile A/B/C, assuming average/higher claim frequency? The difference will give a rather clear picture of what to expect from the captive performance. If different scenarios yield different answers, it is up to the consultant to suggest the best course of action.
2.5 Captive benchmarking
In this section we will briefly overview some leading perspectives on how to soundly evaluate the captive performance. We will limit ourselves to gathering some of the more used benchmarking criteria available in captive consulting: to name a few, AM Best ([AMBEST]), KPMG ([KPMG]) and Marsh all publish periodical reports on captive monitoring. The basic idea is quite simple: to evaluate the performance of a captive, compare it with the average of the market. The reference ratios are substantially the same as a regular insurance company: following [MARSH], we can identify the following indicators:
- loss ratio: defined by: (incurred losses+ expenses)/premiums earned. It measures the yearly impact of losses on the business;
-expense ratio: defined by expenses/premiums. It measures how efficient is the underwriting process;
-combined ratio: obtained by summing of the loss ratio with the expense ratio; it is one of the crucial measurement of performance of any insurance company, as a ratio above 100% means value destruction, below 100%, value creation; naturally, this also holds true for a captive;
-investment income ratio: the net investment income/net premiums earned; measures how efficiently are investments capitalizing;
-operating ratio: it is the loss ratio, less the investment ratio. The target value of this ratio is for it to be less than 100%, as this would imply value creation independently of profits due to well structured investments (therefore, good insurance performance);
-risk retention ratio: the amount of risk retained by the captive, divided by the captive available capital. This ratio is more captive related, in the sense that it measures the "amount of risk transfer". It depends also upon the risk itself; in this sense, it can be used as a specific benchmark for single risks. It is also affected by the age of the captive: a younger captive will have lower capital, thus a higher risk retention ratio; an older one is likely to be lower.
These ratios can be easily calculated through financial reports; to get a fruitful benchmark, the consultant has to monitor a large amount of captive portfolios so to create a stable reference to which compare the individual performance, as summarized by the graph below.
benchmarking.jpg
Figure : source [MARSH]
3. Legal and regulatory issues
3.1 domiciliation
In this section, we shall overview the popularity of some captive domiciliations. It is important to point out that there are very few "universal" rules to compare one country to another, as different parent company's countries may have different legal treatments on controlled foreign companies. As we can see from the pictures below, the situation is completely different if we consider only European companies, or if we consider American ones as well. worldcaptives.jpg
Figure : EU and US captive domiciles in 2010. source: [CD]
Figure : captive domiciliation world wide, source: [MARSH]captive domiciles.jpg
Figure : European captives as of 2011, source: [CD]eurocaptives.jpg
As we can see [4] , American companies have a wider set of choices, since federal laws may allow advantages in certain states (Vermont, Utah, Hawaii, and so on). For the purpose of our treatment, it is best to comment on the most common European domiciliations. The point of view will be that of a UK based parent company, but considerations such as location, local laws, Solvency II, will of course fit any European country. We shall follow the aforementioned [MARSH], [AMBEST] and additionally [WILLIS].
Dublin
Dublin is the capital of Ireland, which is a member of the European Union, therefore adopting the Euro currency. Since the institution of the International Financial Services Centre (IFSC) in Dublin, through the 1987 Finance Act Ireland has attracted many international financial services.
Dublin licenses both insurance and reinsurance companies, under the framework of the Irish Insurance Acts, which apply EU Insurance/Reinsurance directives. The authority requires yearly reports on sensible data such as business plans, financial projections, target markets, description of insurance products, policy wordings, reinsurance agreements, margin of solvency calculation.
Dublin hosts more than 120 captive insurance companies, most of which are of the form of single parent captive. While other structures are allowed, they require a case-by-case approval. Non-life captive insurers can write direct and reinsurance business, but it cannot write life business since a separate company is needed. Captive reinsurers on the other hand can write both life and non-life.
Ireland offers stable political environment, large availability of high quality professional insurance managers, and favorable regulatory and tax environment; moreover, it is conveniently located for all European companies. As a member of the European Union, it allows for writing insurance in all 27 EU countries- what is called Freedom of Service (FOS) basis. There are no specific legislations on PCC (protected cell companies).
The minimal capitalization level requirements, estabilished by the existing Solvency I directive and thus valid all across the European Union, are as follows: for writing direct insurance a minimum capital of €2.3m is needed, rising up to €3.5m if long tail risks are written. Starting from January 2013, these level will increase to €2.5m and to €3.7m for long tailed risks. For captive companies writing reinsurance business only, the minimum capital is €1.1m, going to €1.2m from January 2013. There are some restrictions on the quality and the mixture of assets that can be owned.
As we mentioned, the tax regime is very favorable, as trading profits and investment income taxing level (which is considered as trading profit for insurance undertakings) is at 12.5%.
Ireland, as a part of the EU, will apply the upcoming Solvency II directive.
Luxembourg
Since issuing a specific legislation in 1984, Luxembourg became a big center of attraction for reinsurance captives, hosting nearly 250 captives. Historically, most of these were reinsurance captives only. As a member of the European Union, it also adopts the Euro currency.
The restrictions on capital and investments structure are the same as Ireland, which also apply to all European countries. Again, Luxembourg can also write business on a FOS basis, for the same reason as Dublin. However the insurance commissioner of Luxembourg ("Commisariat aux Assurances, or CAA) has expressed willingness of restricting the license to certain requirements. Luxembourg has no legislation on PCC. The corporate net worth tax is 0.5%,while the corporate income tax rate is 29.34%.
Isle of Man
The Isle of Man, a small, self governing British Crown dependency in the Irish sea, has a long tradition as a captive domicile since 1982. The currency is linked to the British pound and valued 1:1. It offers very flexible and effective insurance legislation, and competitive capital requirements and taxation. The minimum capital to be held is to establish a captive in the isle of Man is £50,000, and a company is required to run a risk capital based assessment. The annual fees to be paid are also very low, and the taxation level is reported as "Nil" for captives and PCCs [WILLIS]. Access to credit is greatly facilitated, as loans can be given up to 100% of assets.
There are however limitation on the writing capability: since the European regulation does not allow for insurance to be written outside of the European Union, insurance may be written only a non admitted basis in the U.K. and wherever non admitted insurance is allowed. The Isle of Man has no intention of applying the Solvency II directive in the near future.
Guernsey
Guernsey is a small island in the coast of Normandy; like the Isle of Man, it is a self governing British Crown dependency. Their currency is the British Pound. Similar consideration as for the Isle of Man apply for Guernsey. The capital requirements for captive establishment are set to £100,000; however, in 2008 the Guernsey Financial Services Commissions introduced the Own Solvency Capital Assessment concept, under which the board of an insurance or reinsurance company needs to assess independently the solvency margin, with respect to all the risks (insurance and not insurance) that the company faces. As far as the writing capability, the taxation regime, the access to credit, and adhesion to the Solvency II directive, the same as for the Isle of Man applies.
3.2 solvency II
Recital 10 of the Solvency II directive states that captive will be treated:
"References in this Directive to insurance or reinsurance undertakings should include captive insurance and captive reinsurance undertakings, except where specific provision is made for those undertakings."
However, Recital 21 specifies that:
"This Directive should also take account of the specific nature of captive insurance and captive reinsurance undertakings. As those undertakings only cover risks associated with the industrial or commercial group to which they belong, appropriate approaches should thus be provided in line with the principle of proportionality to reflect the nature, scale and complexity of their business."
The principle is that the Solvency II capital requirements will be required to hold for all insurance companies insuring multiple lines of business for the whole market; captives, however, cover single lines of risks for specific companies. What is called the "Proportionality Principle" (ruled in Article 109 of the 2009/138/EC, Solvency II directive, and Article 77 of Level 2 directive) is precisely the assessment of risk by taking into account the less risky (and in the case of captives, "self contained" ) nature of a small company. Article 78 of the Level 2 directive specifies the range of application of the principle:
"Subject to the captive insurance or reinsurance undertaking complying with Article SCRS1, simplifications which are specifically made available to captive insurance and reinsurance undertakings shall apply only to captive insurance and reinsurance undertakings as defined in Article 13 of Directive 2009/138/EC that meet the following requirements:
in relation to the insurance obligations of the captive insurance undertaking, all insured persons and beneficiaries are legal entities of the group of the captive insurance or reinsurance undertaking and were also legal entities of that group at the time the relevant contract was entered into;
in relation to the reinsurance obligations of the captive insurance or reinsurance undertaking, all insured persons and beneficiaries of the insurance contract underlying the reinsurance obligations are legal entities of the group of the undertaking and were also legal entities of that group at the time the underlying contract was entered into;
the insurance obligations of the captive insurance undertaking and the insurance contract underlying the reinsurance obligations of the insurance or reinsurance captive undertaking do not relate to any compulsory third party liability insurance."
Aon [AON] observes that only 20% of European captives fall into this definition. As a result, as it emerged from QIS 4 (reported in [AMBEST]) nearly 25% of the participating captives did not meet the SCRs, and 7% did not meet the MCRs as well. CEIOPS however observed that 65 out of 99 captives that participated were situated in Luxembourg, where most captives are reinsurance ones, therefore skewing the sample. Moreover, they further elaborated that "companies should be able to anticipate the introduction of Solvency II […] reallocating funds between entities". A.M. Best observes that "this will make businesses think more carefully about where they need to domicile their captives, and […] there will be some businesses that think they may be required to hold onto an excessive amount of capital, and they may look at establishing themselves outside the EU".
Aon further comments that narrowing the scope of the captive insurance definition "is a clear breach of the principle of proportionality" and suggests a minor amendment in the Omnibus II directive to Article 29(4), to explicitly embrace all captive underwritings. Namely, the amendment suggested goes as follows (in clear, Article 29(4), in bold the amendment suggested):
"The Commission shall ensure that implementing measures take into account the principle of proportionality, thus ensuring the proportionate application of this Directive, in particular to captive insurance and reinsurance undertakings and small insurance undertakings."
There is no clear indication of which measures, if any, will be taken in this direction by EIOPA. It is clear, however, that even under proportional treatment, the Solvency II directive will have a deep impact on the world of captives. We quote directly from Ragge & Co., conjecturing the following consequences of some relevant articles from the directive:
"Each firm's regulator will be required to have regard to the nature, scale and complexity of the firm's activities when it establishes the minimum frequency and scope of the firm's supervisory reviews (see article 36 of the Solvency II Directive)
Each firm will be required to have a system of governance that is proportionate to the nature, scale and complexity of its operations (see article 41)
Each firm will be expected to conduct an "own risk and solvency assessment", which takes into account its specific risk profile, approved risk tolerance limits and business strategy. To enable it to do that, each firm must have processes in place that are proportionate to the nature, scale and complexity of the risks inherent in its business (see article 45)
A firm will be able to use simplified methods and techniques to calculate its technical provisions if that is necessary to ensure the actuarial and statistical methods it uses are proportionate to the nature, scale and complexity of the risks it insures (see article 86(h))
A firm will be able to use simplified calculations for specific sub-modules or risk modules in the standard formula, if the nature, scale and complexity of the risks it faces justify that and it would be disproportionate to require all insurers to apply the standard calculation (see articles 109 and 111(l))
(Subject to regulatory approval) a firm will be able to use "undertaking specific parameters" instead of the standard formula parameters to calculate its life, non-life and health underwriting risks (see articles 104(7) and 110 of the Solvency II Directive).
[…] in most c
