The following is a letter sent by Mr. Djordjevich to The Bond Buyer and H. Russell Fraser, chairman of Fitch Investors Service, in response to a Fitch report on bond insurance - entitled "Bond Insurers' Bonanza." The Bond Buyer published a story covering the report April 23.
No doubt, the recent Fitch report, Bond Insurers' Bonanza, started out with good intentions, but it has failed as an analytical contribution to the understanding of the industry.
Fundamentally, when all is said and done, there are five quintessential characteristics of a bond insurer that would reveal it current status and future prospects. These five "Qs" are: * Quality of book of business written (risks assumed). * Quality of invested assets (paid in capital, retained earnings, and unearned premium reserves). * Quality of operations (management, staff, productivity - operating expenses).
Depending on and resulting from the aforementioned key factors, we have two additional aspects to consider: * Quality of earnings. * And quality of protection margins for the bondholder.
Given a good quality of book, invested assets, and productivity, the [return on equity] becomes a function of how adequately the book of business has been priced, what investment yield is obtained on invested assets, and how well costs of operations are controlled.
I will briefly comment on these five determinants of the corporate performance of a bond insurer. The purpose of the exercise is to indicate how poorly Fitch's analysis has been done, and how much it misses the mark, insofar as Capital Guaranty Insurance Co. is concerned in particular, and the whole study in general.
One cannot assess "capital adequacy" and the amount of exposure without also scrutinizing credit quality or the other risk characteristics of a company's insured portfolio, such as geographical distribution, bond-type concentration, and average size of risk.
In the long run, there is nothing more important to a company's survival than the quality of its book of business. Fitch's report fails to deal with this axiom. By stating that "the amount of exposure does not distinguish between credit quality comparisons or other risk characteristics of a company's insured portfolio," Fitch's analysts inexplicably ignored or dismissed this most important criterion for measuring or ranking bond insurance companies.
Admittedly, this is not an easy task, but Fitch's analysts should have made an effort to develop an appropriate approach before publishing an important industry-wide analysis and ranking.
The industry is familiar with one pertinent measure: the Standard & Poor's Corp. "risk-weighted capital charge" (which is an aggregate proxy for the degree of risk of the insured book of business). Using this comparison against MBIA - the number one ranked insurer, according to Fitch - you can see that Capital Guaranty, both in trends and in actual numbers, has recorded better results [see chart one].
Fitch's report focuses on net investment yield rather than quality of investments; that is, the higher the yield, the better the ranking. Is Fitch suggesting that investing in junk bonds with the higher "net investment yield" would enable a company to attain a higher ranking in Fitch's analytical methodology?
For [long-term] stability and survivability, the quality of invested assets, not their yield, is absolutely most important. In times of stress and difficulties, that is the resource pool that a bond insurer would draw from to play losses.
The following comparisons of Capital Guaranty's Dec. 31, 1991, investment portfolio with the indsustry's as a whole, based on an Association of Financial Guaranty Insurors report of Sept. 30, 1991, demonstrates the quality of Capital Guaranty's investment portfolio [see chart two].
Assessing productivity (cost factors) is always difficult. I must point out, however, that using raw figures of expenses to par does not reveal the whole picture, particularly when comparing the 20-year companies of MBIA and AMBAC with start-up companies such as Capital Markets Assurance Corp. and Capital Guaranty.
For instance, our concern, among other things in this respect, is the trend in comparing salaries to the increased volume of business (an indication of productivity). As related to gross par insured, we averaged 0.21% over the last four years, with a declining trend. Also, start-up, or "early-in-life-cycle" companies, will always have special expenses, often one-time or caused by unexpected requirements.
The quality of earnings is primarily a function of the three previously mentioned characteristics - book of business, invested assets, and operating efficiency. In the long run, these three factors, plus adequate pricing, will determine more than anything else the viability of a bond insurer. Let me focus for a moment on pricing.
I have fought long and vigorously against - and lamented over - the industry's practice of inadequate pricing. Also, the record is clear that Capital Guaranty has tried, under the most difficult circumstances for a start-up company, to price our products properly. Here again, let the record speak for itself.
Relating premiums charged to total debt service comparisons between the industry's average and Capital Guaranty's attests to my assertion [see chart three].
Another important flaw should be highlighted. An analyst of more understanding would have realized the need to explain why young companies which have not employed a significant portion of their capital, are penalized when they are compared to fully leveraged companies. In our case, approximately 40% of our capital is unemployed, earning only an investment rate of return, while the employed 60% of our capital has been earning a 15%-17% return.
And going higher. I am quite confident that our return on equity, within a year or two, will be in the same range and, thus superior to, MBIA's.
In the "quality of protection" margins for the bondholder we have, essentially, at least two components to focus on - leverage ratio and reinsurance.
In general, assigning "ranks" to companies due to the "trend" in the leverage ratio is meaningless, particularly in regard to companies that are overcapitalized and are early in their life cycle.
Of course, the leverage ratio has to be observed in conjunction with the quality of the book of business and invested assets. But yet another problem the Fitch study encounters is comparing individual insurers such as CapMac and Financial Security Assurance with monoline municipal bond insurers, such as Capital Guaranty or AMBAC. Here we have the proverbial comparison between apples and oranges.
Any analysts worth his or her salt would not consider CapMap's leverage ratio number as being the best without some qualifications, or that FSA's ratio of 79.5 as being "better" than Capital Guaranty's 91.9.
As for reinsurance, about 25% of Capital Guaranty's book of business is protected by a proprietary first-loss coverage with a triple-A reinsurance company. This coverage is substantial per single risk, as well as in the aggregate. But no consideration was given to this. Overall, the report has not covered the reinsurance aspect at all.
In addition to the analytical shortcomings, there is also a matter of style. My staff is very disturbed that they were not contacted [for] additional information that certainly would have presented a more accurate picture of Capital Guaranty.
I would hope that Fitch would find some way to correct or amend this report. It is certainly not an analytical "bonanza". I am personally not too disturbed about its conclusions as regards Capital Guaranty because, in fact, it tells more about Fitch than Capital Guaranty.
Rating agencies play a decisive role in shaping the investment community's opinions and understandings of our industry. Hence, reports of this magnitude and intent must be analytically thorough and impeccable. Mr. Djordjevich is president and chief executive officer of Capital Guaranty Corp. Note: Capital Guaranty is rated triple-A by Standard & Poor's Corp., and does not have a claims-paying rating from Fitch or Moody's Investors Service.
