Captive insurance is the most popular form1 of alternative risk financing due to the myriad of benefits, both economic and noneconomic, that can be achieved by its utilization. The benefits of captive insurance compared to commercial insurance include:
- Stabilization of costs: Captives are not subject to the underwriting cycle. Therefore, potential changes in premium year over year will not be subject to the commercial rate changes of an insurance company’s book of business. Rather, premium changes will be based primarily on the captive’s loss experience. Assuming stable loss experience, premiums paid into the captive will be consistent.
- Earn investment income: Captives can earn investment income on their loss and unearned premium reserves. A guaranteed cost policy purchased from a commercial insurer would not provide this additional income to the insured.
- Capture underwriting profits: If the premium paid into the captive is greater than the loss and expense costs that are ultimately incurred by the captive, then the captive is able to retain these underwriting profits. These underwriting profits can be transferred back to the parent in the form of a dividend or can help build up the capital and surplus. A captive is also able to earn investment income on this excess capital and surplus. These profits would not be realized were the parent to purchase a guaranteed cost policy.
- Provide coverage not available in the market: Some coverages may have extremely high premiums, making them essentially unavailable in the traditional market. A captive can provide these coverages at a rate that, when possible, is based on the insured’s loss experience. Additionally, a captive can provide coverage for atypical exposures that are not offered by traditional insurers.
- Improved claims handling and risk control: In certain situations, a captive allows for greater control of the claims handling process. Potentially large claims can be identified and handled early on to reduce costs, as incurred losses on claims tend to increase the longer that they stay open and unsettled.2
- Potential federal tax benefit: A captive may be able to deduct both loss payments and reserve estimates for federal tax purposes, whereas self-insurance only allows for deductions as losses are paid. This accelerated timing of deductions allows for investment income to be earned on the taxes temporarily saved. For large captives and captives that write long-tailed coverages, these savings can be considerable, particularly in a high interest rate environment.
- Capturing the total cost of risk: While many of the benefits described above can be achieved through self-insurance, a captive is an excellent risk management tool for tracking the total cost of risk. According to a survey of parent companies in the Marsh 2020 Captive Landscape Report, the top reason why parent companies use a captive is to act as a formal funding vehicle to insure risks that the parent has decided to self-insure.3 With a captive, the costs of self-insuring can easily be tracked through the captive’s financial statements.
These benefits, among others, are why nearly all Fortune 500 companies, and many midsized companies, utilize some form of captive insurance.4 The most popular form of captives is single parent captives, which insure the risks of a single company, called the parent. Single parent captives accounted for nearly 80% of gross written premium of the captive market in 2019.5
Although captives are set up for different reasons than other types of insurance, we were curious as to whether single parent captive insurers themselves are earning higher profits than other forms of insurance. We reviewed historical single parent captive financial statement data from over 60 single parent captives over the past five years (2016-2020), along with annual statement data from S&P Global Market Intelligence for traditional commercial insurance companies and for risk retention groups (RRGs), a more regulated form of captive insurance, to assess whether single parent captive insurers are more profitable than other forms of insurance. Throughout the remainder of this article, any reference to “captives” refers to single parent captives. Group captives, insuring the risks of multiple companies (and RRGs are a form of group captive), have different goals, objectives, and benefits.
We looked at various financial ratios that can be derived easily from the balance sheets and income statements of insurers. We did not adjust for any accounting biases and took the values as is. The data used for the traditional insurers and RRGs is derived from the sum of the top 100 (using calendar year 2020 net earned premium) for each company type.
In comparing profitability, arguably the best metric to look at is the combined ratio. We calculated combined ratios as the ratio of incurred losses, incurred loss adjustment expenses (LAE), and underwriting expenses to earned premium, all derived from the income statement. The graph in Figure 1 shows the historical combined ratios for each of the past five years for each insurance entity.
Figure 1: Combined Ratios – 2016 to 2020 by Company Type
Captives have considerably lower combined ratios than both traditional insurers and RRGs. Traditional insurers and RRGs have combined ratios hovering around the breakeven mark of 100%. In 2019 and 2020, the RRG combined ratios are slightly over 100%. On the other hand, the five-year average for captives is 83%. Because 100% less the combined ratio represents the percentage of premium that is earned as underwriting profit (ignoring investment income), captives are earning profits of nearly 20% of premium! These profits can be used to increase surplus, which can provide a cushion in the event of adverse loss development. They can also be used to pay a dividend back to the parent.
While the results of reviewing historical combined ratios certainly are telling about a captive insurer’s profitability, we dug further into these profit differences. In a simplified world, an insurer’s premium is the sum of three items: projected future expenses, projected future losses, and a profit and contingencies provision. The first component of premium that we compared between company types is expenses. Historical expense ratios were calculated as the ratio of current year underwriting expenses to current earned premium. The graph in Figure 2 details the historical expense ratios for each of the past five calendar years for each type of company.
Figure 2: Underwriting Expense Ratios – 2016 to 2020 by Company Type
Captive underwriting expense ratios have averaged around 9% over the past five years, whereas traditional insurers have averaged around 31% and RRGs 21%. Captives have little to no overhead costs. They typically do not have dedicated staff and office space. Claims handling is usually handled by a third-party administrator (TPA), and actuarial services are provided by a consulting firm. Some captives may also run a portion of the expenses through the parent company, rather than through the captive, which may be a contributing factor to the lower ratios. Traditional insurers, in addition to paying for staff and corporate office space, also have other costs such as agents’ commissions.
Comparing traditional insurers to RRGs, RRGs typically have much smaller staffs, so the overhead costs will be less. Additionally, commission and brokerage expenses contribute to the difference in expense ratios between traditional insurers and RRGs. Using the ratio of net commission and brokerage expenses to earned premium, the 2016 to 2020 average for traditional insurers is 13% compared to 0.5% for RRGs. Many RRGs even had zero net commission and brokerage expenses, so the difference in commissions is quite substantial.
Another expense factor is premium taxes. Premium tax rates are lower for captives than traditional insurance companies. In addition, some states such as Arizona and Utah only have a low annual fee compared to the percentage of premium charged in most other states. RRGs are subject to the same premium taxes imposed on traditional insurers,6 so there is no benefit there.
However, not all captives have lower expense ratios. We categorized our sample of 60 captives into three groups based on premium volume. Small captives, those with less than $1 million in annual premium volume, have a five-year annual average expense ratio of 42%. Medium-sized captives, those with annual premium volume between $1 million and $10 million, have expense ratios closer to RRGs, with a five-year average at 17%. Large captives, those with annual premium volume greater than $10 million, benefit from economies of scale and have an average expense ratio of 8%. These large captives are the primary driver of the low expense ratios in Figure 2 above, and account for the majority of the captives surveyed. Some extremely large captives even have expense ratios less than half of 1%. The table in Figure 3 displays the expense ratios and the percentages of captives surveyed by year and size.
Figure 3: Captive Expense Ratios – 2016 to 2020 by Captive Size Band
|% of Captives Surveyed
|% of Captives Surveyed
|% of Captives Surveyed
Because the historical expense ratios for captives are significantly lower than those for the other company types, the lower expenses must be the driver of the differences in combined ratios in Figure 1 above, right? Not so fast! Let’s examine captive expenses more closely. If we return to our simplified premium example, compared to projecting loss and LAE, expenses are very predictable. Captives incur expenses for services such as captive management, actuarial, audit, legal, and board meetings. These expenses are mostly fixed and can be easily projected into the future. The one variable expense that captives incur is premium taxes, and this expense is very predictable due to its simple addition to the overall premium. In estimating premium, captives will generally know what their expenses will be, and price these expenses in their policies accordingly. Similarly, commercial insurers and RRGs are generally able to predict the expense component of their premiums with a high degree of accuracy.
Therefore, because it is not the expenses themselves that are driving the lower expense ratios, it must be the larger premium base that is the driver. Let’s keep investigating the components of the premium to provide more clarity.
The next component of premium we investigated was the loss and LAE. We calculated the loss and LAE ratios as the ratio of current year incurred losses and LAE to earned premium from the income statement. The graph in Figure 4 displays these historical loss and LAE ratios by year for each type of company.
Figure 4: Loss and LAE Ratios – 2016 to 2020 by Company Type
Historically, traditional insurers have had lower loss and LAE ratios than both captives and RRGs. The traditional insurer ratios are slightly below 70% on average, whereas captives are around 75% and RRGs are around 82%. There are a few reasons for these differences. The first is that because captives and RRGs have lower expenses, losses and LAE make up a larger portion of the premium charged. The second reason is the profit and contingencies provision. Traditional insurers typically include a profit and contingencies provision in their rates. While some captives and RRGs do include a profit provision in their rates, captives and RRGs are primarily formed to provide insurance “at cost.” The profit provision included in a captive’s or RRG’s premium is to provide a buffer in the case of adverse loss development, rather than trying to earn a profit. This provision is typically less than that of a traditional insurer. The higher profit provision for traditional insurers will contribute to a higher premium and a lower loss and LAE ratio, all else equal.
The final and most important reason why captives are earning higher profits than traditional insurers and RRGs is that actual loss activity generally emerges less than expected or priced. Going back to our simplified premium example, premium consists of projected losses, projected expenses, and a profit and contingencies provision. Because the expenses are fairly consistent year over year and easy to project for the upcoming policy year, the higher premium is due to the estimated losses and LAE used in the premium calculation being greater than the losses and LAE that actually develop.
In recent years, captives have had a conservative approach in charging premium, and as losses develop less than projected, the captive generates underwriting profits. The conservative approach can be a function of several factors. First, rather than charging premium based on losses at the expected level, some captives include losses at higher probability levels, such as the 75th probability level in the premium calculation, to include a buffer in the event of adverse loss development. These higher probability levels lead to higher premium, all else equal. Second, captives may reserve recent policy years more conservatively. Because recent policy years are relatively immature, the reserve estimates rely on the initial expected losses. In addition, if the estimates for the initial expected losses rely on averages based on more recent years, there is an ongoing cycle of funding and reserving the captive more conservatively. As these recent policy years mature, the captive releases reserves that do not emerge and the captive eventually earns a profit. These reserve releases represent another source of profit.
Comparing RRGs to traditional insurers and captives, RRGs are only allowed to write liability coverages, which are long-tailed in nature and, in many cases, can include risks that are low in frequency but high in severity, such as errors and omissions and directors and officers liability.7 These risks are highly variable in nature and can make estimating premium more difficult. Additionally, liability claims are often litigated and the potential impact of factors such as social inflation and landmark judicial decisions can make projecting losses more challenging. While captives similarly write long-tailed liability coverages for the potential federal tax benefit, they also write potentially more profitable coverages such as workers' compensation and property. The workers' compensation industry as a whole has been very profitable in recent years,8 potentially contributing to lower loss and LAE ratios for captives and traditional insurers relative to RRGs. Property exposures are short-tailed, so they have the ability to release reserves sooner in the event of favorable experience. These could be a few of the many potential reasons why the RRG loss ratios and combined ratios are higher.
Another important metric in analyzing the profitability of insurance companies is return on equity (ROE). ROE is one of the most important ratios used in rate filings to ensure that rates are reasonable and adequate. We calculated ROE as the ratio of the current year net income from the income statement (including investment income) to the prior year’s surplus and equity from the balance sheet. The graph in Figure 5 displays historical calendar year ROE for each type of company.
Figure 5: Return on Equity – 2016 to 2020 by Company Type
The historical average ROE for traditional insurers is generally lower than captives and higher than RRGs. Traditional insurers’ ROE on average is around 8%, with captives averaging around 13% and RRGs around 6%. Traditional insurers’ ROE is lower than what other industries are earning, as over the past 10 years the S&P 500 has had an annual average return of 13.6%.9 According to the New York University (NYU) Stern School of Business, the average levered beta (a measure of systematic risk compared to the market) for property and casualty insurance companies as of January 2021 was 0.64.10 This could be a driver as to why traditional insurers are earning less than other industries, and the lower beta should be reflected in the stock price for traditional insurers that are publicly traded.
The lower ROE for RRGs can be explained in a few ways. For one, the comparably higher loss and LAE ratios lead to a lower net income, all else equal. Second, as mentioned previously, RRGs often write low-frequency/high-severity liability coverages, for which claims costs are highly variable, and they may also be subject to large claims. RRGs may need to maintain higher amounts of surplus to account for this. As mentioned above, the primary goal of RRGs is to provide insurance “at cost,” so they may not be trying to earn a profit and may not be focused on cost of capital and ROE.
Captives are a different type of company compared to RRGs and traditional insurers, and captives also have different goals. A parent company has the choice of placing its capital in either the company itself or the captive. Ignoring potential tax benefits, it would not be wise for the parent to continue to place capital into a captive if it were earning a lower ROE than the corporate ROE the parent is achieving (assuming the risks for the captive are similar to the parent’s business). While ROE is an appropriate metric for traditional insurers, one could argue that it is not appropriate for captives. The conservative approach of captives in charging premium, along with their requirements of excess capital and surplus, is usually not efficient. The parent could use these funds to invest and grow its own business, which would normally provide a better return on its capital. The real ROE for a captive is arguably the sum of the potential federal tax benefit and the investment income earned on the funds in the captive, combined with any other economic benefits afforded by the captive such as underwriting profits not available to self-insurers, all as a percentage of surplus. This will normally be below the targeted rate of return on the parent’s core business, but there are examples where the captive’s ROE as defined above would exceed that of the parent company.
We also compared investment returns between the three types of companies to see whether this contributed to a difference in profitability. We calculated this metric as the ratio of the current year net investment income from the income statement to the prior year’s assets from the balance sheet. The results are displayed in Figure 6.
Figure 6: Investment Return on Assets – 2016 to 2020 by Company Type
For all three types of companies, investment return has been generally between 1.5% and 3.0%. Traditional insurance companies have had the highest and most consistent return at 2.7% on average, while RRGs and captives are slightly lower and more volatile at 2.1% on average. All three types of companies need to invest in relatively liquid assets in order to pay claims, so we would not expect the ratios to differ too much. One potential explanation for why traditional insurers have slightly higher investment returns is that they may have investment staff in-house, whereas RRGs and captives will typically incur a fee for using an outside investment firm. Because we used net investment return in the ratio, this cost leads to lower net investment returns, all else equal. Additionally, RRGs are owned by their members, so they may be more risk-averse.
As displayed through the various metrics, over the past five years single parent captives have been more profitable than commercial insurers and RRGs. These profits can be used to either help build up a healthy surplus or released as a dividend back to the parent (or both). While profits are primarily driven by a potentially more conservative approach to running the captive, a parent company can benefit in a variety of ways, including having independence from the commercial insurance cycle; insuring additional coverages and/or higher layers of loss; improving claims handling and risk management; and quantifying the total cost of risk.
2Sheils, L.C. (May 22, 2017). Claims Handling in Captives. Captive.com. Retrieved September 19, 2021, from https://www.captive.com/news/claims-handling-in-captives.
4Tucciarone, J.W. & Biscotti, L. (December 2018). Captive Insurance Companies: A Common Sense Approach to Improved Risk Management. CPA Journal. Retrieved September 19, 2021, from https://www.cpajournal.com/2018/12/19/captive-insurance-companies.
9Knueven, L. (June 14, 2021). The average stock market return over the past 10 year. Business Insider. Retrieved September 19, 2021, from https://www.businessinsider.com/personal-finance/average-stock-market-return.
10NYU Stern School of Business (January 2021). Total Betas by Sector (for computing private company costs of equity) – US. Retrieved September 19, 2021, from https://pages.stern.nyu.edu/~adamodar/New_Home_Page/datafile/totalbeta.html.