Chapter 12

The Beginning of the Entrepreneurial Years

Insurance

You don't need to pray to God any more when there are storms in the sky, but you do have to be insured.

—Bertolt Brecht

Berkeley in the 1960s ushered in a dramatic transformation felt throughout the rest of the country. It was the epicenter of free love, free speech, drugs, and radicalism. To many it was a period of catastrophic social change. At the time, however, I had a very different notion of catastrophic change.

My interest in insurance began when Robert “Bob” Goshay, a gregarious colleague at Berkeley and an insurance expert, asked one day, “Do you think catastrophic risks could be transferred using futures markets?”

Bob believed the idea was transformational and I thought it was feasible. It was that conversation, and a subsequent one in 1970 when we discussed the 1966 earthquake in Parkfield, California,¹ which inspired our research. We collaborated on a paper that explored the feasibility of a reinsurance futures market, published in The Journal of Business Finance in 1972.²

In 1973, Bob took a year's sabbatical at Lloyd's of London, attempting to convince its management and membership that these new insurance risk transfer mechanisms would enhance underwriting profits and reduce risk. Primary insurance firms at the time hedged themselves using reinsurance companies, which insured insurance companies that had too much risk in their books. Post sabbatical, Bob reported, “Lloyd's is living in the nineteenth century. There's no way we can get this cozy club to embrace our new alternative to reinsurance.”

I had always harbored a romantic image of the group of insurance syndicates, mainly because of the 1936 movie Lloyd's of London.³ The movie painted the members of Lloyd's as innovative risk takers who insured the British merchant fleet during the Napoleonic wars. Lloyd's had been instrumental in helping Lord Nelson win the battle of Trafalgar. I was crestfallen and disillusioned when Bob described to me the actual lack of imagination of many of the syndicates. I found it paradoxical that those who were willing to underwrite nontraditional risks would not consider new risk management tools.

I had mentioned my interest in catastrophe futures to Henry Hall Wilson when first interviewed at the CBOT. However, insurance futures took a backseat when it became evident that interest rate futures would become a reality. The idea lay dormant for more than a decade before Les Rosenthal brought it to light again in the late 1980s. Les was adventurous and liked change. Anxious to get back into developing new products, he called and asked, “Whatever happened to your idea on insurance derivatives?” I explained that my former colleague's sabbatical at Lloyd's had not been positive, and told him that I still believed the industry could use alternative ways to hedge. Les proceeded to champion the idea at the CBOT, urging the exchange to rekindle my early efforts.

Commoditizing Catastrophes

After Billy O'Connor became chairman of the CBOT board in 1990, he formed an insurance futures committee. Les was appointed chairman of the committee, and I was named vice chairman.

I was at Kidder and organized a meeting of industry leaders at the office headquarters in Hanover Square, a triangular park located in the financial district in lower Manhattan. It became clear from industry meetings that the CBOT should focus on catastrophic events like hurricanes and earthquakes instead of the conventional kinds of insurance like medical and auto initially recommended by the staff of the exchange.

In the late 1980s and 1990s, the United States was experiencing a high incidence of catastrophes that should have been, according to executives in the industry, once-in-a-century events. After three consecutive events of this nature, one CEO we talked to still insisted matter-of-factly that the probability of such events occurring was one in a hundred and that there was really no need for the insurance industry to hedge. I never understood how you could have three one-in-a-hundred events in five years.

The more recent losses incurred by the insurance industry were attributed to the increased frequency and severity of natural calamities, but also to the fact that the U.S. population had been migrating toward coastal regions, where the threat of natural disasters posed much greater damage risks. Nearly 70 percent of the U.S. population was now concentrated in these areas. In effect, these coastal regions became hothouses for the development of infrastructure, expensive dwellings, and human capital.

In 1989, Hurricane Hugo caused $5.9 billion in damages in the Carolinas.⁴ Furthermore, California's burgeoning population was hit by the Loma Prieta earthquake in 1989, resulting in property losses of up to $6 billion.⁵ The populations in the Gulf Coast spanning from Texas to Florida, along with the entire East Coast, continued to grow. In 1938, a major hurricane struck Long Island, causing $306 million in damages, an amount equal to $4.77 billion today, to the potato farms and small number of dwellings that existed at the time.⁶ It was appropriately dubbed the “Long Island Express.” Today, given the development of the Hamptons and other expensive resorts in the area, a similar event could easily result in a record level of damages. We had a close call in 2011 when Hurricane Irene hit the East Coast.

The insurance industry was and still is undercapitalized for catastrophic events. The arithmetic simply did not add up. In 1997, U.S. primary insurance companies had a total capital of $239 billion, while reinsurance companies had another $52 billion.⁷ This was supplemented by another $10 billion to $15 billion⁸ in the retrocessional market.⁹ All of this was supposed to underpin an economy of $8.3 trillion¹⁰ once including all products and service liabilities, such as toxic spills, director's and officer's liabilities, medical malpractices, legal malpractices, and personal injuries. It also needed to support $12 trillion to $15 trillion in property.¹¹ The potential for catastrophic losses was $20 trillion, and any single catastrophic event in areas like Miami, Florida or Orange County and Los Angeles, California, could have resulted in insured losses totaling $50 billion to $100 billion.¹² In fact, it was the catastrophes of the late 1980s and early 1990s that generated support for the idea of a futures contract in hurricanes, tornados, and earthquakes. In 1992, Hurricane Andrew struck well south of Miami and left a trail of insured damages totaling $15.5 billion, or $20 billion today.¹³ At the time, it was the greatest single cost of any catastrophic event in U.S. history. While the U.S. insurance markets seemed vulnerable to phenomenal losses, the worldwide capital markets, with equities and fixed income totaling more than $30 trillion, could easily absorb a hefty portion of risk from the insurance markets.

Les and I turned our attention to natural disasters in 1990. We spent the rest of the year developing the terms for a catastrophe contract, along with marketing materials. In writing articles and developing marketing materials for Kidder, I applied the same formula for sales brochures, speeches, and seminars that we had used in financial futures, and worked closely with CBOT staff.

Selling Catastrophe Futures

Selling insurance derivatives was more difficult than selling interest rate futures. Although a series of catastrophes had provided the structural change needed to catalyze the development of insurance derivatives, we now had to standardize the product.

Insurance policies were contracts of indemnification, where an insurance company paid the insured for any incurred damages. It was a costly process. It was also a business that required negotiations after the policy was written. I was struck by ads in the trade press that described how the claims department of an insurance company should be run as if it were a profit center instead of cost center. Not only did this seem like a violation of the insurance contract, it was incredibly inefficient.

However, if an insurance policy were a contract for differences instead of indemnification, the insurance payments could be easily determined. The claims department would no longer be incentivized to maximize profits and be at odds with customers. For example, a reinsurance company could offer a policy that was event-based. An earthquake in the City of San Francisco with a magnitude of 7.0 on the Richter scale would trigger the policy payment. The payment would be based on the premium to be paid by the insurance company. Assuming the San Francisco scenario described above, the reinsurance policy would simply state a $200,000 premium for $10 million in coverage. The $10 million would be paid if a previously approved verifier reported an earthquake in a predetermined geographical area of 7.0 or higher magnitude.

I didn't realize it at first, but we even had standardization: a small market had emerged in London for industry loss warranties (ILWs). ILWs were a type of reinsurance or derivatives contract through which one party purchased protection based on the total loss arising from an event incurred by the entire insurance industry rather than by the party itself. For example, the buyer of a $100 million U.S. Wind ILW pays a premium and receives $100 million in coverage if total losses to the insurance industry from a single U.S. hurricane exceed $20 billion. The industry loss, $20 billion in this case, is often referred to as the trigger. The $100 million, the amount of protection offered by the contract, is referred to as the limit.¹⁴ ILWs were contracts for differences and not for indemnities.¹⁵ The industry's loss might not correlate with an individual company's losses, thereby triggering basis risk for the reinsured. This risk was higher for companies whose exposure concentration was further away from the industry averages. ILW covers were therefore typically bought by companies whose portfolios closely followed the market.¹⁶

In the meetings I had organized for industry members, there also seemed to be a demand from the primary insurance companies in the United States. Chicago-based property and casualty companies enthusiastically supported our efforts, as did others. The reception by the reinsurance companies, however, was less than enthusiastic. A transparent market that offered a hedging alternative to reinsurance was not welcome. It was particularly threatening to companies with very strong balance sheets. A cleared hedging vehicle with no counterparty risk was a nightmare. It was the same old story.

In spite of the primary insurance industry's support for the concept, we still met internal opposition. One director at the exchange mounted a campaign to kill the concept. He thought the new contracts posed too much of a financial risk for the clearinghouse. Billy O'Connor and Les Rosenthal convinced the board that this wasn't true and gave the contract its final push. New markets required champions. Leaders willing to take personal risks were indispensable in organized markets, particularly at the CBOT where the political process determined the outcome of research and development. After significant debate, the board of directors finally voted to launch the new insurance futures contract.

Our timing could not have been better. There were rumors that Allstate would violate insurance capital requirements if it failed to hedge the disastrous consequences of another major catastrophe. Industry buzz alleged that Berkshire Hathaway had written Allstate a catastrophe cover of $500 million for another Hurricane Andrew at a 40 percent rate on line,¹⁷ and in doing so received $200 million in premiums. If the rumors were true, this was a great deal for Berkshire Hathaway. If another Hurricane Andrew occurred in the short balance of the hurricane season, the maximum loss for the firm was $500 million. Since the firm had received $200 million in premiums, its exposure was only $300 million. It was an unbelievable risk-reward ratio. I was convinced that there was enough capital available to underwrite risk at this price.

Insurance catastrophe futures contracts commenced trading at the CBOT on December 11, 1992. I felt buoyed with excitement as the opening bell rang, and was reminded of creating the first interest rate futures contract. Like the GNMA futures, insurance futures were a whole new field and asset class. The first trade was made between Les Rosenthal and Billy O'Connor, who sought to set an example for the membership.

The new futures contracts were based on the loss ratios of the catastrophe insurance industry—quarterly insured catastrophic losses nationwide and in three U.S regions: Eastern, Midwestern, and Western. The price of these futures contracts at any given time reflected the market's expectation of the quarter's catastrophe loss incurred by a sample of companies.

The basis of these futures was an index tracking the losses of almost 25 property and casualty insurers, who reported their loss data quarterly to Insurance Services Office, Inc. Unfortunately, the CBOT insurance contracts were not without problems. As argued by Bouriaux and Himick (1998), first, the loss development period of six months was insufficient in capturing the true losses experienced by the industry. Second, the ISO-issued reports were only updated twice during each contract period, preventing them from achieving the continuous pricing that was the hallmark of exchange markets. Finally, we learned that the index was flawed and did not reflect the true industry loss ratios when a catastrophe occurred.¹⁸ The reason was that large industry players such as Allstate and State Farm did not report to the ISO. Consequently, ISO used to have to make adjustments to account for these major players. The inaccuracy of these adjustments became clear after the 1994 Northridge earthquake,¹⁹ when the ISO index severely underestimated the loss index ratio. The contract failed, and we tried another approach.

The CBOT catastrophe insurance futures didn't mimic reinsurance. Option call spreads better simulated the reinsurance layers that the insurance industry was accustomed to.²⁰ Consequently, the exchange redesigned the contracts and began trading options contracts on September 29, 1995, using the Property Claims Services' (PCS) loss estimates instead of those reported by the ISO. The PCS reported cumulative insurance industry losses in a timelier and more accurate fashion, providing additional liquidity. Furthermore, the contract design was simplified and the development period was extended to 12 months. Later on, due to the lack of industry demand, PCS-indexed insurance futures were dropped entirely. Only cash options on PCS industry estimates were offered for trading.²¹

Entering the Insurance Space as an Entrepreneur

Despite the many benefits associated with PCS options, initial trading in PCS options was not very liquid. One reason was the steep learning curve associated with this financial product. Both insurance and reinsurance companies needed to be comfortable with the language of derivatives and basis risk, the risk that the loss incurred by insurers or reinsurers may not be correlated with the losses of the industry as a whole. The issue of liquidity was more of an issue for capital market professionals, who were accustomed to being able to enter and reverse a trade at a moment's notice. Insurance and reinsurance companies did not put as much emphasis on easily getting in and out of a trade as their capital market counterparts did. Once these companies bought catastrophe coverage, they intended, for the most part, to keep that position for its duration. At the end of the third quarter of 1997, there were a total of 494 trades for an average of 25 call spreads per trade. We had hoped to have tens of thousands of trades. It was an uphill battle to build exchange volume.

Although Kidder, Peabody welcomed my efforts, the firm was after all an investment bank, and I needed an insurance company to champion the idea. General Electric's insurance company was similarly uninterested in pioneering these new products.

In 1992, a friend called me after reading an article about the CBOT's plan to launch catastrophe futures contracts. He was founding a reinsurance company. I told him about the business opportunities in the catastrophe space because of the recent resurgence of natural calamities, changes in demographic trends, and the increase in property values in high-risk coastal areas. It was a good time to enter the business, and a recent spate of catastrophes had driven reinsurance rates to very high levels. I ratified his business plans, and his enthusiasm added to my own. I wanted to get more involved in the industry and wondered how I could participate.

My friend referred me to Michael Palm. Michael was one of the brightest people I had met in the business world, and immediately understood my logic about structural changes and standardization occurring in the insurance markets. In fact, he himself was in the middle of his own financial innovation at the time. Michael had, along with a partner, cofounded Centre Reinsurance, a subsidiary of Zurich Insurance that specialized in a new product called finite risk reinsurance, a way to spread the premiums and risks over a particular time period. The losses paid were capped, or in other words, finite.

Michael and I tried to figure out how to work together. Initially, we thought about having me become an employee at Centre Re. Negotiating and renegotiating the specific terms of my employment was an arduous process, and I began to get frustrated. I called my friend to ask for his advice, and received a response that profoundly changed my life. He said, “Why are you negotiating as if you're applying for a job? You're thinking of yourself as an employee when you really want to be an entrepreneur.” He was dead right.

I decided to put up all the working capital, negotiate the right to use Centre's name and balance sheet, and give them a share of profits. As soon as I changed my approach, we quickly came to a mutually satisfactory agreement.

On March 1, 1993, I resigned from Kidder and founded Centre Financial Products (CFP). My dreams of becoming an entrepreneur had finally become a reality. I was both nervous and giddy with excitement. For the first time in 25 years, I was no longer part of a large organization.

SO₂ Emissions Trading at Centre Financial Products

CFP's core business was developing products that focused on the intersection between the capital markets and the insurance industry. There was a robust connection between insurance and the environmental space. In the event of a climate change–induced catastrophe, insurance companies would be one of the first to take the brunt. What was happening in the environment therefore mattered a great deal to insurance and reinsurance companies.

For example, one possible effect of climate change was a reduction of snowfall. This could cause a dramatic change in the landscape of the country. It could also threaten the Swiss way of life and the country's vibrant tourism industry. A reduction in snowfall would negatively impact the income of ski resorts, which was an indispensable part of Switzerland's GDP. These concerns were what drove the country's growing involvement in climate change initiatives—in particular, through Swiss Re, Switzerland's predominant insurance provider.

Anticipating the likely convergence of insurance and climate change in the future, I continued to pursue emissions trading as part of our activities at CFP. We started with SO₂ allowances, anticipating CO₂ allowances. I went to different power companies, soliciting their SO₂ allowance business. Thanks to the large size of our balance sheet, we were able to serve as a principal and not just a broker.

SO₂ allowances could also be used as a financing instrument. Priced at $50 million or more, scrubbers were expensive and had to be financed. Utilities could borrow the money required to install a scrubber and pay for the scrubber using proceeds from a future sale of emission allowances. The first trade registered by the EPA was actually a financing that allowed the utility to bypass the assumption of debt.

In 1993, Henderson Municipal Power & Light in Kentucky sold 150,000 tons of sulfur dioxide pollution allowances to CentreFinancial Products for $26.8 million. The sale of allowances represented the third largest SO₂ trade at the time.²² This was a unique sale, as the revenue from the sale was used to finance and install scrubbers in the Station Two plant of Henderson Municipal Power, planned in April 1993. The scrubbers were estimated to have cost $41 million, yet by installing these scrubbers, Henderson Municipal was able to decrease its sulfur emissions by 95 percent.²³ At the market price, the proceeds from the sale of allowances were enough to finance the scrubbers.

In September 1993, CFP in turn sold the original 150,000 allowances to Carolina Power and Light (CP&L). CP&L used very low sulfur coal in its generators. Thus, the marginal cost of removing remaining sulfur by scrubbers in their plants was higher than the industry average. In 1993, CP&L estimated this cost to be approximately $500 per ton of SO₂. Comparatively, in March 1993, the EPA auctioned 150,010 allowances at the CBOT at an average price of $143 per ton. CP&L had little interest in installing scrubbing equipment at the time due to the high cost of scrubbing and the still-evolving scrubbing technology.²⁴

Nonetheless, CP&L expected to exceed its EPA sulfur allotment even while introducing fuel switching and demand side management to reduce emissions. Between 2000 and 2009, the EPA allocated 143,968 allowances to CP&L per year. By CP&L estimates, if it failed to make any changes in the way it operated its system, it would emit approximately 230,000 tons of sulfur dioxide in the year 2000, creating a deficit of 86,000 tons. As a result, CP&L would have to either reduce its sulfur dioxide emissions or purchase additional allowances.

CP&L first purchased 85,103 allowance credits at the 1993 EPA auction for $11,490,000 at an average price of $135. They required additional credits and entered into a formal agreement with CFP in 2000 for the purchase of 150,000 SO₂ emission allowances at $47,250,000.²⁵ Using an 8 percent market interest rate over eight years, the present discounted value of that payment in 1993 was $27,560,000.

CP&L needed to raise additional capital or borrow money to pay for the SO₂ allowances, which posed an additional problem for CFP. Not only did we have to guarantee the interest rate for CP&L's capital raise, we also had to guarantee the price and quantity of the allowances. To solve the first problem, we agreed to lend money to CP&L vis-à-vis Zurich Insurance, which CFP had direct access to under the terms of our contract with Centre Re. Solving the second problem was more challenging. Since the EPA registry wasn't operational at the time that the deal was consummated, CFP wrote the contract such that it closed when the registry was inaugurated.

Although the price and quantity of allowances were fixed through a forward purchase agreement with another utility, CFP still had to hedge against the interest rate risk. As there was no futures market on corporate bonds, we had to use the CBOT Treasury bond futures contract, which created basis risk. To mitigate this risk, CFP decided to buy puts on the Treasury bond futures contract as well.

The former risk manager at Kidder joined me at CFP. He had come from Salomon Brothers, and his expertise in futures markets proved to be invaluable in managing the hedge at CFP. This was critical because we bought and sold the allowances at the same price. The profit came from the price at which we bought the debt from CP&L and then sold it to an insurance company. Interest rates fell during the time we signed the contract and when we closed the deal. The structured transaction turned out to be very profitable, in spite of the fact that we made little money from the purchase and sale of the allowances.

In the end, CP&L secured their future allowances at favorable prices and financed the transaction at attractive interest rates. The seller of the allowances also fared well. The sale and purchase was done at a higher price than the OTC bid, and at a lower price than the OTC offer.

While this seemed like a simple transaction, a closer examination revealed that the allowance market not only enabled a low-cost compliance tool, but a financing vehicle as well. I later said to the trader that managed the hedge, “I never realized that interest rate futures would enable the first SO₂ allowance trade. Talk about unintended consequences.”

Partly due to the failure of the CBOT to list a futures contract on SO₂, the SO₂ market gradually became a brokered OTC market. This diminished profit opportunities for principals like us, so we decided to turn our full attention to insurance and CO₂ emissions.

Creating New Insurance Products

At CFP, we wrote reinsurance policies on catastrophic risks that were contract for differences, which could then be hedged in the futures markets. For certain risks without a corresponding futures market, we hedged indirectly in a related futures market. We did this with crop damage risks using agricultural futures contracts. Other risks, such as airline crashes and oil spills, could be standardized but couldn't be hedged directly or indirectly with existing futures markets. There was no way to hedge these risks except through diversification. We built a quantitative model based on portfolio theory for these risks.

To assume catastrophic risk directly, Centre Re formed a reinsurance company in Bermuda, a major reinsurance center, shortly after we joined. This left little appetite to underwrite more catastrophic risk. CFP therefore had to find more industry warranty contracts to hedge. However, because the futures markets for ILWs were already so transparent, profit opportunities on arbitrage in this were limited. Paradoxically, the transparency in futures market fostered by the CBOT contracts actually hindered the ability to write insurance policies that exploited the differences between reinsurance prices and futures prices.

In addition, there were other barriers that limited the use of futures by reinsurance companies. Key among them was the fact that futures and options positions were never officially recognized as reinsurance, and therefore couldn't be treated as capital. This by default increased transaction costs. Reinsurance companies justifiably viewed insurance futures as a competitive threat. Even if insurance companies didn't use the CBOT, the transparency was putting some pressures on reinsurance rates. Reinsurance companies were not amused by this.

Left with no choice, we sought out reinsurance companies to turn these futures contracts into reinsurance policies that would be recognized by the insurance regulators. We traveled to London to find companies that were willing to hedge by buying insurance policies based on ILWs. The natural place to start was with members and brokers of Lloyd's of London. We had very limited success.

We also tried to find reinsurance companies willing to hedge their exposures in noncatastrophic risks using insurance policies based on contracts for differences, which were termed index-linked or event-based insurance policies. Index-linked insurance, as demonstrated by the CBOT ISO and PCS indexed options and futures, was a financial product in which the payout, or the level of payout, depended on prespecified levels of the index as opposed to the actual losses. Theoretically, this should reduce the moral hazard of having increased losses be in the financial interests of insured. I worked to persuade some brokers in London to originate this type of risk.

Most brokers wanted to stick to the traditional way of conducting business and found no reason to switch from their existing model, especially because it was profitable. We did, however, find two smaller brokerage companies that were enthusiastic about standardizing the process: Willis Faber and Benfield. I got to know Neil Eckert, a partner and broker at Benfield, who immediately understood the concept of index-linked insurance policies. Neil was creative and a great salesman.

The two of us flew to Germany to discuss the concept with Klaus Gierstner, an entrepreneur who didn't fit the mold of a stereotypical German reinsurer. He was an out-of-the-box thinker, as evidenced by his ponytail. We met at a quaint restaurant in a suburb of Berlin and discussed how index-linked insurance could diversify his risk at Francona, where Klaus was the underwriter. Neil suggested that for the purpose of standardization, we could divide risks into different buckets: hurricanes, typhoons, European windstorms, earthquakes in the United States, earthquakes in Japan, aviation insurance, environmental risks, and so on. We could then write index-linked reinsurance policies based on these risk buckets. Three hours later, we had a deal in principle. Neil and I followed up by writing the index-linked reinsurance policies for a pilot program. Instead of insuring the damage from a specific oil spill, we defined the event in terms of the number of barrels spilled. We also defined the damage to hulls from airline crashes based on the manufacturer and type of aircraft. As we began to model the associated risks, we recognized that the risks varied between airlines run by developed and developing countries. If the same aircraft were under the U.S. and European banner, there would be a specific dollar loss specified in the insurance policy. If the same aircraft flew out of Russia, where maintenance procedures were different, the claims paid would be lower for the same type of aircraft.

I built the team at CFP with former colleagues from Drexel and Kidder. We also hired a number of inexperienced youths from diverse backgrounds ranging from naval officers and to economists who knew nothing about insurance to a nuclear engineer. We continued to build a portfolio of market index-linked reinsurance contracts, which included a policy indexed to the medical component of the Consumer Price Index (CPI).²⁶

Crop Insurance

We also entered the crop reinsurance business. The business was very cyclical. Major European reinsurance companies entered and exited the business routinely, because of the significant amount of volatility and the limited profitability of the business. CFP wanted to position itself competitively as a stable, long-term provider of crop reinsurance.

We started to write crop insurance, and hedged in the corn, wheat, and soybeans futures markets. Prices and yields were becoming uncoupled, and therefore net farm income couldn't be hedged without a futures market in yield. We advocated a yield futures contract. It was listed on the CBOT, but died due to the lack of a champion.

In addition to the difficulties in my new business, I experienced a devastating personal loss in 1996.

Charlie O. Finley's Death

The last time I saw Charlie was sometime in 1995 when he visited me in New York. We went to the Jockey Club, where he showed me a glow-in-the-dark football he had patented for use in small towns that wanted to be able to play at night without stadium lights. It was wonderful to see him.

Charlie died on February 19, 1996. He was 77 years old. Ellen heard about it on the evening news when I was still in London. Penya told Ellen not to wake me in the middle of the night with the bad news, so Ellen called me early in the morning, London time. I canceled the rest of my appointments and flew back to the United States immediately. Charlie's funeral service was held three days later in Merrillville, Indiana. I gave the eulogy alongside Charlie's old baseball players, Catfish Hunter and Reggie Jackson.

Catfish shared with us the story of his earliest encounter with Charlie. Back when Catfish had yet to join the Oakland A's, Charlie had sent his scout to Catfish's home in North Carolina in a dogged attempt to recruit him. Charlie called the house, and Al Hunter, Sr. picked up the phone: “Do you want to speak to your scout?” he asked. Charlie replied, “Not unless he signed your son.” Al Hunter, Jr. was then ushered to the phone to hear him out. Charlie offered the baseball player a $65,000 salary. The young man bartered it up to $70,000. Charlie agreed to the raise. Feeling empowered, Al Hunter, Jr. decided to push his luck and suggested that Charlie also throw in a red Corvette. Charlie retorted, “I gave you a damn raise. Buy it yourself.”

It didn't take long for Charlie to address the next important topic: “Son, do you have a nickname or a hobby?” “No,” said Al Hunter, Jr. “Then listen carefully,” said Charlie. “When you were a kid you used to catch catfish with a passion, and once landed a whole lot of them. You and your friends were caught by a truant officer. Ever since then you've been called Catfish. Now repeat the story after me.” And so Catfish was born. At the eulogy, he reflected to us, “If it hadn't been for Charlie, I'd still be Al Hunter, Jr. Nobody would remember that name, but everyone remembers the Catfish.”

Reggie Jackson was the last to speak. He recalled his feelings before joining the New York Yankees. While Reggie made it in Oakland, there were widespread press reports about how tough it was to make it in the Big Apple. Reggie said that he was undaunted by New York because of the high standards set by Charlie. Charlie had always insisted his players to live up to their god-given talents, and Reggie went on to become a superstar in New York. Discipline, training, and determination complemented his natural abilities. I would have to use that same discipline and determination in the next couple of years and beyond.

I thought about my mother after Charlie's funeral. She died a year after my dad did, in a hospital while being cared for by my brother. I think she lost her will to live after my father had passed. I flew in to see her the day before she died. She was in bad form. She said sadly, “Richard, look at what a mess I am.” I replied, “No, Mom. You look great.” I felt her pain for not being the woman she wanted to be. Nonetheless, I left the hospital after about an hour of very pleasant stories and laughs.

I remember her visiting us in Chicago after my dad died. She stayed with us in Lake Point Tower, and I invited Charlie to come up to meet her. Charlie's mom was strong-willed, too, and he recognized that trait in my own mother. It was funny to see him, the famous Charlie O. Finley, so cautious and tentative around my mother. I was sorry my father couldn't be there, as it would have been a perfect moment.

Exiting the Insurance Space

Sometime after the mid-1990s, the incidence of catastrophes diminished, causing a decline in reinsurance rates and a demand for hedging in the index-linked markets. Premiums had fallen as a result, and we were in what insurance professionals called a soft market. We were also seeing a limited demand for our other index-linked reinsurance policies. Zurich Insurance was trying to make the most out of the declining demand for catastrophic risk, while reducing their involvement in the business.

It became clear that we would have to start our own insurance company. We hired an investment banker to draw up the prospectus. The chairman and CEO of Zurich Insurance had been a big supporter of mine from the outset and agreed to be a seed investor. He said, “This is the perfect solution to our dilemma of avoiding taking more catastrophic risk on our balance sheet, yet still betting on insurance derivatives.” He committed to invest $50 million in a new insurance company called Hedge Re.

I thought we would be able to raise $200 million for Hedge Re, but soon realized that I was wrong. There were no catastrophes happening, and reinsurance rates had dropped dramatically.

The bloom had left the rose on the capital raise for Centre Financial Products (CFP), and I was forced to sell the company. Fortunately, I had met the CEO of CNA, an insurance company based in Chicago, during the research phase for the relaunch of PCS-based catastrophe futures, and asked if he was interested in buying CFP or its assets. We finally reached an agreement. There was one hitch. CNA wanted me to become the executive chairman of Hedge Re, the renamed CFP. It was a turning point in my life. I enjoyed being an entrepreneur and did not want to become a full-time employee. It was a tough decision. On one hand, the security of being an executive at a prestigious company appealed to me. I was interested in insurance derivatives, and believed that I could expand the research laboratory of Hedge Re to include new capital markets products. On the other hand, I would lose my independence and would not be able to devote significant effort to emissions markets, which was my real passion. However, its chances at commercial success were limited at that time. We found a compromise that worked for both of us. My colleagues would run the business at Hedge Re, while I agreed to serve as nonexecutive chairman of Hedge Re. I pursued the environmental business separately.

Meanwhile, open interest in catastrophe insurance options at the CBOT continued to grow in 1997 and reached over 18,000 contracts by the end of the year. Hedge Re continued to focus on crop insurance during the following year, but I was unfortunately unable to persuade the CBOT to put more resources into a crop insurance contract. CNA decided to exit the business and closed Hedge Re in the second quarter of 1999.

An analysis of the failure of the CBOT catastrophe insurance options contract yielded some interesting implications for financial innovation. The original hypothesis that structural changes—in this case, the trend of U.S. migration to coastal regions and the increasing frequency and severity of natural calamities—require new capital to be channeled into the insurance markets still rings valid. However, this new capital would come from an influx of Bermuda-based reinsurance companies because of the ease of entry and favorable tax treatments for reinsurance companies, rather than from futures and options. In the years following Hurricane Andrew, nearly $3 billion in new capital flooded into the market and eight different reinsurance companies were formed. Furthermore, Lloyd's of London permitted corporate membership to attract additional capital into their business.

The other source of new capital came from catastrophe-based bonds, also known as cat bonds. A cat bond is an exchange of principal for periodic coupon payments wherein the payment of the coupon or the return of the principal of the bond or both is linked to the occurrence of a specified catastrophic event. Before an insurance company can issue a cat bond, it must establish an offshore special purpose vehicle (SPV) reinsurer from which it will buy a reinsurance contract. The risk premium is therefore passed onto the SPV, who in turn passes this on to the investors in the form of a coupon. Investors will post the notional amount of the bond in a trust and all the funds in the trust are then invested in U.S. Treasuries. This will collateralize the risk of the investors and eliminate credit risk.

Cat bonds were considered more attractive than PCS options because of their inherent flexibility. In a cat bond, a reinsurance company can customize its hedge to be indexed on its own losses, as is done in traditional reinsurance, or it can be indexed on PCS. Moreover, they can be structured to resemble a traditional excess-of-loss reinsurance contract or a quota-share contract, whereby investors share proportionately in the gains and losses of the reinsurer. Cat bonds and the SPV structure also provide the issuing insurance company with access to a broader set of investors than PCS options. Some investors, such as pension funds and mutual funds, are restricted from transacting in derivatives such as PCS options, but are allowed to invest in securities, such as bonds or notes. The ability to offer principal-protected tranches of a note increases the investor base even further because there are some investors who can invest only in AAA-rated securities. This larger set of potential investors may be especially important for companies seeking to transfer large amounts of risk to the capital markets.

In 1996, the United Services Automobile Association (USAA) issued its first cat bond of $477 million, with limited success. It was priced at 575 basis points over the London Interbank Offered Rate (Libor) and had coverage of 2.5:1. This was followed by Goldman Sachs, who successfully marketed a $68.5 million reinsurance securitization deal for St. Paul Re in 1997. From 1996 to 1998, there were a total of 18 separate issues. By the end of 2007, there were 116 separate issuances at an average of $188 million per deal. This totaled approximately $25 billion in new capital, or $2.5 billion per year. Between the new Bermudian reinsurance companies and cat bonds, the shortfall in capital in the insurance market was filled.

Although the CBOT continued to market insurance futures, there was a very big difference between marketing and selling. For products to be truly successful, one had to sell a product by assessing the customer's specific needs and finding solutions tailored to meet those needs, as opposed to more general marketing through conferences and presentations.

On the regulatory front, catastrophe insurance futures and option positions were never qualified as reinsurance, which by default increased the transaction costs. Since there were no internal champions for the product, there was no budget to try to change the rules. Furthermore, futures markets require a constant flow of information in order to prompt hedging and speculating. This was difficult to achieve with earthquakes, which were often hard to predict, limiting the price information available. On the other hand, the method used to forecast hurricanes was more reliable, therefore supplying enough data to warrant trading. Even so, a quiet hurricane season presented little change in risk and therefore a limited need to hedge.

Successful futures markets benefit largely from the participation of primary dealers who need to transfer interest risk, and commodity market participants like grain merchants who need to hedge their price risk. The structure of the insurance industry differs significantly from those of the grain and bond industries. Insurance and reinsurance brokers act as agents who control the flow of business. In fact, the market capitalizations of these brokerage entities were very large compared to the amount of risk capital in the insurance business. In 1994, insurance services comprised 55 percent of revenue for Marsh & McLennan, making the market cap attributable to insurance services $3.2 billion versus its total market cap of $5.8 billion.²⁷ It seemed uncanny that brokers had more capital than the insurance professionals who were underwriting their risk, especially given that the insurance products being sold were practically homogenous. Selling insurance was akin to selling white paint, and should not have required such a large brokerage force to back it up. The fact that most insurance sales are beginning to be done online now through automated systems suggests the obsolescence of the old model.

It was a tradition for those who bought reinsurance to be compensated by premiums rebated to them in the case that there was no catastrophic event. This practice was abused by brokers who attempted to capture some of the reduction themselves. It didn't help when the Marsh & McLennan brokerage was almost ruined as a result of these payments.²⁸ The lack of volatility in the catastrophe insurance futures didn't inspire speculation either. From the point of view of minimizing the bid-offer spreads, there were permits sold at the CBOT but no dedicated market makers.

The design of the insurance futures contract was imperfect for three separate reasons. First, the ILWs traded in the reinsurance market were based on a fixed level of losses as the trigger, the event that resulted in payment, that is, an earthquake with a Richter scale reading of 6.0 at specified data stations. The futures contract only hedged a layer of losses, that is, $5 billion to $10 billion. This implied a less-than-perfect hedge and thereby increased transaction costs. Second, the presence of basis risk was a big issue among the insurance industry, as it was possible that the insurer or reinsurer's losses would not correlate perfectly with the total industry losses, as well as with the development time for the losses. Basis risk in financial markets was well known, and generally well understood by commodity or financial futures market participants who could easily adjust their hedging strategy to account for such risk. The issue here was that it was much more difficult to quantify basis risk in insurance markets.

The lack of an internal champion for the product ultimately resulted in the demise of catastrophic insurance futures. Although we were active as principals at CNA, the exchange would have had to actively market to FCMs and hedge funds to bring in speculative capital. Furthermore, the exchange never put resources into having the insurance regulators treat hedges like reinsurance. Since hedges on the exchange were not treated as reinsurance, they fell inferior to reinsurance, according to calculations based on regulatory capital.

In the 1990s, the leadership of the CBOT devoted a significant amount of time and energy toward building a new trading floor. Les and I fought against it. We thought that building a single purpose building for $200 million had no economic logic. Les jested bitterly, “I think it will become the largest McDonald's in the country.” Phil Johnson predicted that it would become a food court.²⁹ The building was completed just as electronic trading was about to become a reality in the United States. It was a perfect example of Parkinson's Law—“work expands so as to fill the time available for its completion.”³⁰

Postmortem

At the end of the day, insurance futures failed at the CBOT, only to be reborn at the Insurance Futures Exchange (IFEX), an entity affiliated with the Chicago Climate Futures Exchange (CCFE). Insurance futures had a very modest beginning in 2007. In 2005, Hurricane Katrina struck New Orleans, with insured damages estimated at $70 to $150 billion.³¹ The earthquakes in New Zealand and Japan have indicated that the industry will continue to need substantial capital. Perhaps it will end up as a centrally cleared product on some exchange. Counterparty risk is ever-present in the reinsurance sector.

There are reasons to be alarmed. In effect, many insurance and reinsurance companies believe that the state or the federal government will bail out companies, when in fact, I am not sure that states like Florida will have the money to do so. The insurance business essentially enjoys a “free put,” whereby the government will always bail them out in the event of default. I also worry about the contingent liabilities on the U.S. government.

Insurance derivatives are sleeping, or in a coma. It's still too early to pronounce their death as there was still an immense amount of risk relative to capital in the industry. The jury is still out.

Despite the apparent failure of insurance futures, the experience enabled me to develop key relationships within the insurance sector. In light of my later endeavors in the environmental market, these relationships became critical.