The subprime mortgage crisis of '08 made everyone interested in finance and economics know two English abbreviations overnight: CDO and CDS.

Countless literary and artistic works have blamed these two similar-looking financial instruments as the culprits of the greatest economic crisis of the new century, but in reality, there is nothing wrong with these two instruments themselves, the problem is that the greed of the capitalist group has rendered all regulation and censorship useless.

In layman's terms, CDO is to separate and package various bonds of different ratings and integrate them into a new bond, and this diversification strategy does have the possibility of reducing bond risks in principle.

But when CDOs continue to evolve, the situation changes, and each financial institution will issue different CDOs, and the buyers of the CDOs themselves will package the different CDO bonds they buy into new CDOs. Eventually, for every dollar of debt, there may be a twenty-dollar CDO betting that it will repay the loan as agreed.

As for CDS, it is essentially an insurance that pays for the risk of default by an unrelated third party when a debt defaults. In the field of insurance, the policyholder must be a related person, and only one insurance policy can be taken out for a subject matter, and the total amount of compensation will not exceed the actual value of the subject matter. But CDS is different, anyone can insure a subject matter, and as many copies as they want.

For example, if you own a house, you can take out property insurance for your house. I have nothing to do with your house, but I also have 10 insurances for your house, and if your house catches fire or is smashed, I can get 10 insurances to pay.

These two instruments did not make high profits for Wall Street as soon as they were introduced, and the popularity of any financial instrument must be in line with the historical course and economic conditions of the time.

Just as the most lucrative product on Wall Street throughout the eighties was junk bonds, in the first decade of the New World, the CDO was the cornerstone of Wall Street's ability to sustain profiteering operations.

It has been more than a decade since the advent of the CDO, but it has always maintained a tepid state, on the one hand, because the external environment is not mature, the Glasssteagall Act limits the scale of investment banks, and the mortgage policy has not been liberalized, and on the other hand, because there is no suitable mathematical model to provide theoretical support for the CDO, resulting in the risk "uncertainty" of the CDO is too high, and it is difficult to obtain a high credit rating from the rating agency.

Wall Street is not risk-averse, high risk also represents high returns, and what Wall Street really hates is "the risk of uncertainty." ”

Standard and poor's, Moody's, and Fitch are the three major international rating agencies and the most authoritative bond rating agencies in the world. Any kind of bond issuance can only be sold in the market if it has been rated by these three rating agencies.

Risk credit ratings represent everything in the derivatives space, and the main buyers of derivatives are not retail investors, but fund groups such as the Minnesota Pension Fund. These fund groups with semi-official backgrounds have clear rules that only allow the purchase of bonds with the highest creditworthiness, that is, AAA credit rating. Without the highest rating, it is impossible for the CDO to gain access to this largest market.

The uncertainty of the traditional CDO risk makes it difficult for CDO to become the darling of Wall Street, and every financial institution will be cautious about the issuance of CDO.

In 2000, Chinese mathematician Li Xianglin published this paper entitled "Default Correlation Research: Gaussian Connection Codependence Function" in the famous financial SSCI journal "Fixed Income Journal", and Wall Street keenly discovered the great significance of this model for financial derivatives, especially CDO.

The reason why it is difficult for CDOs to price and measure risks is that the data is too large and complex, and it has become an almost impossible problem to calculate the interconnection between different financial products. However, the Gaussian connection dependency function adopts a very clever approach: it directly calls the market data of the CDS instead of referring to the historical default data to construct the default correlation model.

In layman's terms, the Gaussian connection dependency function is like the difference between the evolution from a handicraft workshop to full industrial automation compared with the traditional risk calculation with reference to historical default data.

It is precisely because of the Gaussian connection dependency function tool that the annual sales of the CDO have expanded rapidly from more than $10 billion per year at the end of the nineties to more than $500 billion in the seventeenth year.

In the global financial derivatives trade, a total of 8 trillion US dollars have used this functional model, and in the United States alone, the total number of CDOs exceeded 4.7 trillion US dollars in 06 years.

Zhang Chen is not worried that Jamie Dimon will find anything wrong with this function, because this function is a completely correct formula theorem, as long as the conditions for the use of the function can be met and the original data is real, then the Gaussian connection dependence function cannot be invalidated.

But is it possible?

"Jamie, after some calculations, this functional model is valid. Prince, chief analyst of Salomon Smith Barney's securities sales department, pushed his glasses, "If this model is used to reprice the CDO currently launched by Salomon Smith Barney, there will be about three percent increase." In addition, according to the rating rules of S&P's AAA rated bonds, more BBB or B-rated securities can be tolerated, and the expected return can continue to be raised by about a point. ”

Jamie Dimon frowned, "Are you sure?"

Prince grinned, "Jamie, are you questioning my professionalism? Let's put it this way, we have launched a CDO, because we don't have a reasonable risk calculation model, in order to be able to get a higher rating from S&P, which contains more than 80 percent AAA rated bonds." However, according to the calculation of this model, we only need to have a minimum of 65% AAA-rated bonds plus 20% BBB-rated bonds plus 10% BB or B-rated bonds to meet S&P's rating requirements for AAA-rated bonds. ”

"You know, the lower the rating, the higher the yield, and when packaged, we've got 25 percent more high-yield bonds than we did before, and still get a AAA rating from S&P or Moody's. I think that if this model is accepted by Wall Street, it will be the biggest revolution since Ronald Reagan opened up the financial markets in '82. ”

Prince was full of praise: "The person who came up with this model is a genius, the algorithm of the Gaussian connection dependency function is not complicated, and a college graduate who has studied advanced mathematics can complete it, but the model is extremely accurate." After eliminating erroneous data, the accuracy can reach 100%. ”

"Revolutionary, genius?" Jamie Dimon touched his chin and turned to Bulvo Nelson of Salomon Smith Barney's Risk and Credit Control Department, "What about risk?

Bulvo Nelson spread his hands: "Logically speaking, I don't see anything wrong with this formula. I need more time to do the risk assessment, but from the results so far, the risk is close to zero. ”

Jamie Dimon nodded thoughtfully and picked up the phone, "Sandy, there's something I want to report to you... ”

(End of chapter)