The market forces that devastated the workers’ compensation market have wreaked even greater havoc in the subprime mortgage market. Brian O’Sullivan asks why those lessons weren’t learned.

During the latter part of the 1990s, the workers’ compensation industry lost nearly $30bn. A disproportionate share of that loss was borne by the reinsurance and retrocession markets, in particular the life and health companies that wrote workers’ compensation carve-out reinsurance.

While the combined ratios for insurers who wrote workers’ compensation insurance during that period were typically between 130% and 170%, meaning they paid out $1.30 to $1.70 for every dollar in premium they wrote, it was not unusual for the reinsurers to have loss ratios of 500% or sometimes substantially higher.

Indeed, CNA recently announced a $250m settlement with John Hancock Life Insurance of an arbitration involving several retrocessional contracts reinsuring carve-out business on which CNA received approximately $2.5m in premium (ie a loss ratio of approximately 10,000%).

Accident waiting to happen

Predictably, losses of this magnitude spawned significant litigation and arbitration, as reinsurers and retrocessionaires alleged they had been defrauded into entering into contracts reinsuring this business.

A popular argument was that the carve-out market that existed during the late 1990s was a “fraudulent market” in which managing general underwriters (MGUs) wrote business that they knew would lose substantial amounts of money, aided by reinsurance intermediaries that placed business they knew would be gross loss making.

These arguments, however, ignore the fact that a convergence of market forces created a “perfect storm” which made it almost inevitable that reinsurers and retrocessionaires would lose substantial amounts of money.

Some of those same market forces recently converged to cause even more damage in the subprime mortgage market.

The excesses that characterised both markets had their genesis in their prior successes. The debacle in the workers’ compensation/carve-out market came on the heels of a record boom cycle. During the mid 1990s, workers’ compensation carriers were reporting record profits and it appeared that the cost of claims would continue to decrease as a result of legislative reforms and the implementation of managed care strategies.

“A convergence of market forces created a 'perfect storm' that made it almost inevitable that reinsurers and retrocessionaries would lose substantial amounts of money

This expectation of lower claims came during one of the longest economic booms in history, creating an expectation that companies would continue to earn substantial returns on their investment portfolios. Together, these factors created a widespread assumption that workers’ compensation would continue to be an extremely profitable line of business. As a result, many insurers and reinsurers sought to enter the market or increase their market share, creating extreme competition at the insurance, reinsurance and retrocessional levels.

Similarly, in the years immediately preceding the current subprime mortgage crisis, the mortgage industry achieved record profits. Between 2000 and 2003, as interest rates dropped to historic lows, a record number of people either bought or refinanced homes. This resulted in a nearly fourfold increase in the total amount of mortgages originated annually, and a flood of new entrants into the mortgage lending market.

Pass the parcel

Companies that had historically retained most of the risk on the business they wrote were now ceding much of the risk into the secondary markets. Prior to the mid-1990s, insurers had historically purchased reinsurance for only the largest claims, retaining liability for the vast majority of workers’ compensation claims. Thus, the workers’ compensation carriers alone bore the financial consequences of poor underwriting. This changed in the mid-to-late 1990s as reinsurers and retrocessionaires began to offer protections that attached at lower and lower layers at lower and lower rates. By the end of the 1990s, workers’ compensation carriers were buying substantial amounts of low layer reinsurance at ridiculously low rates (usually from life and health companies).

Similarly, in the early 2000s, lenders began passing on more of the risk they had historically retained by selling an increasing number of the mortgages they issued in the form of mortgage-backed securities. Between 2003 and 2006, the total dollar value of mortgage-backed securities purchased by private investors doubled, with much of the increase being attributable to the securitisation of subprime mortgages.

Much of the underwriting was done by entities that were compensated based upon the amount of business they wrote. Much of the low layer reinsurance and retrocessional contracts that were written in the late 1990s were written by MGUs writing on behalf of life and health companies.

Practically all MGUs earn most or all of their compensation based upon the volume of premium they write, regardless of the profitability of that business. Similarly, many of the subprime mortgages were “underwritten” by mortgage brokers who were likewise compensated based upon the number (and amount) of mortgages sold. In both instances, the entities doing the underwriting had a strong financial interest to write even the most risky business.

Cash flow underwriting

There was a general relaxation (if not abandonment) of underwriting discipline in both cases. The law of market cycles teaches that, as competition increases and rates begin to fall, if companies maintain underwriting discipline, they will write less business and rates will level off.

“Reinsurers and retrocessionaries began to offer protections that attached at lower and lower layers at lower and lower rates

That law was ignored in the 1990s as insurers and reinsurers kept writing substantial volumes of workers compensation/carve-out business. Some did so in order to maintain or increase market share, believing that they would be able to make up their losses in the long term. Others did so with the expectation that the investment income they would earn would be sufficient to offset their underwriting losses.

Still others did so because they believed they would be able to cede most of their losses to their reinsurers and retrocessionaires. And, finally, some simply did not understand the business they were writing.

In the mortgage market, the principle worked slightly differently but with the same catastrophic results. In 2004, interest rates began to rise, resulting in a sharp decrease in the number of prime borrowers. Rather than reduce the number of mortgages they were writing, mortgage brokers and lenders began to more aggressively target subprime borrowers.

To encourage people to borrow as much as possible, companies made greater use of “affordability” products that were designed to reduce the cost of mortgages in the short term. These included “2/28” or “3/27” hybrid loans, where the borrower pays a “teaser” rate for the first two or three years of the loan, interest-only loans, where the borrower initially pays only interest, and negative amortisation loans, where the borrower pays back less than the full amount of interest with the difference being added to the mortgage principal. Lenders also began routinely issuing mortgages based on little or no documentation to the so-called “NINJA” borrowers: No Income, No Job, No Assets.

The secondary markets subsidised the irresponsible behaviour at the primary level. In retrospect, it is clear that workers’ compensation carriers and subprime mortgage lenders engaged in conduct that can only be characterised as extremely irresponsible. In both cases, they were able to do so because the secondary markets were pouring substantial amounts of money into the market with little understanding of the risks they were reinsuring or securitising.

Neither the life and health companies, nor the MGUs that underwrote the workers’ compensation carve-out business on their behalf, understood the risks associated with the low layer coverages they were writing. In fact, many of those companies and MGUs wrote reinsurance and retrocessional contracts based largely upon the assumption that the cost of workers’ compensation claims would continue to go down and companies that wrote that business would continue to make money.

In fact, because of the cumulative effect of the underpricing, together with the significant brokerage and commissions paid, companies writing workers compensation carve-out business often received pennies of premium for every dollar of exposure they reinsured. As a result, these companies bore a disproportionate share of workers’ compensation losses.

Many investors did not underwrite the mortgage-backed securities they purchased, relying instead on the ratings assigned to those securities. The rating agencies, in turn, rated these securities based on how similar securities had performed in the past, rather than the actual risks associated with the underlying mortgages. Both investors and rating agencies were oblivious to the fact that the new “affordability” products substantially increased the risk that borrowers would default once they were required to pay the true cost of the loan, especially if the housing market cooled off.

So what is to be learned from the workers’ compensation/carve-out and subprime mortgage debacles? That nothing is new under the sun and those who fail to heed the lessons of history are doomed to repeat them. Stay tuned in a few years for the next time the confluence of market forces create the right conditions for another perfect storm.

Brian O'Sullivan is a partner in the New York office of Cadwalader, Wickersham & Taft LLP.