Automatic Synthesis of Tri-Diagonal Markov Transition Matrices: A Failure of the Inverse Ferron-Frobenius Algorithm


Sylvain Raynes experimented with a method developed by Goya and Boyarski (1993) to standardize the synthesis of conditional Markov transition matrices for deal entry in our automated re-rating system, ABSTRAK®. In the analytical literature, the reverse-engineering of a Markov matrix from its spectral-radius eigenvector is referred to as the Inverse Perron-Frobenius Problem. Our analysts do this synthesis manually, so a successful outcome based on a numerical method would add considerable efficiency and precision to our process. Unfortunately, the method produced forward roll-rates too small for real-world delinquency modeling.

See more: Blog_Perron_Frobenius_Failure 2015 0730

R&R Consulting Copyright 2015 (c) All Rights Reserved

Kaisa Group Holdings Ltd (佳兆业集团- HK:1638) Default

Kaisa Group Holdings Ltd., a HK-listed real estate developer with large scale mixed-use real estate developments in over 30 cities in China was the first Asian credit to default in 2015. English language coverage of the story has been provided by Bloomberg, the FTReuters and the South China Morning Post. Chinese language coverage has been provided by Sina and Epoch Times. From the website, Kaisa has commercial real estate assets in the Pearl River Delta, Shanghai Delta, Central China, Southwest and North/Northeast; however, the 2013 annual report states that 30% or more of sales came from Shenzhen assets. According to the Sina account, its Shenzhen government connections  have been an asset and, most recently, a liability.

Here is a summary of events leading to the default:

  • Kaisa came to the attention of analysts on December 3, when the company disclosed it was in talks with the Shenzhen government to lift a ban on property sales blocking the sale of some unsold property units.
  • On December 4, Haitong International Securities put out a sell recommendation on Kaisa.
  • In mid-December members of the founding family trust (Chairman Kwok Ying Shing and his two brothers) sold 11% of the company’s issued stock to Sino Life Insurance Company for HKD 1.7 ($219 MM) to create interim liquidity while the ban was in effect.
  • The December sale of the shares lowered the public float to 20.8%, which is less than the required minimum 25% on the Hong Kong exchange.
  • Meanwhile, the resignation of Chairman Kwok triggered an early payment clause under a HK 400 MM loan by HSBC to the company and causing a default.
  • The CFO, Cheung Hung Kwong and Vice Chairman Tam Lai Ling resigned the next day.

According to Chinese language sources, Kaisa has seven credit facilities valued at $2.48 BN, of which 78.6% are US dollar denominated. Default on the HSBC loan could potentially trigger cross-default clauses on other loans and bonds, including an $800 MM US-denominated bond facility due 2018, which allows accelerated repayment if 25% of the investors vote to accelerate. The original yield on those bonds was 9% but jumped to in excess of 29% on January 2(Epoch Times) and exceeded 45% on January 4 (Sina). Also as of January 4, the yield on another $500 MM US dollar facility due 2020 is about 40%.

In the market for its shares, Kaisa (HK:1680) dropped from a two-year high of HKD 3.2 on 9/24/2014 to a low of HKD 1.59 when trading ended at the end of 2014. Since January 1, trading has been suspended indefinitely since the Group received HSBC’s default notice.


Thank (or Blame) Adam Hartung for this blog

Adam Hartung wrote a great column in Forbes on thinking prospectively.  So good, I took it on board to frame a conversation about R&R.

  1. Stop waxing eloquently about what you did last year or quarter. Yesterday has come and gone. Let’s talk about the future.

In 2015, R&R Consulting will focus more on selling.

  1. Tell me about important trends that are going to impact your business. Is it demographics, aging population, the ecology movement, digitization, regulatory change, organic foods, mobility, mobile payments, nanotechnology, biotechnology…?  What are the critical trends that will impact your business going forward?

Our business is essentially protected, and protectionism is a formidable adversary when you are an advocate of change. The most critical trend facing the ratings business is need to evolve from ordinal rankings to bona fide risk measures of risk and uncertainty. This is a very dense assertion that needs several steps to unpack.

Taylor Swift & Art Finance Experiments

What is a song worth? Who creates the value: the songwriter, the singer, or the “star-maker machinery behind the popular song”? Perhaps all three, but how to divvy up the spoils equitably?

Taylor Swift’s decision to withdraw from Spotify has become a flashpoint for such questions. In her Yahoo interview last week, Swift is quoted as saying “… everything new, like Spotify, all feels to me a bit like a grand experiment. And I’m not willing to contribute my life’s work to an experiment that I don’t feel fairly compensates the writers, producers, artists, and creators of this music.”

The question of fair compensation to creatives is not new, but some new tools have become available to talk about it. Not surprisingly (since this is an R&R blog,) securitization is one of them. Rather than launching into a factual defense of securitization or explain here how it was sabotaged, this short blog presents a succinct case for using securitization in arts finance by attacking the standard interpretation of the balance sheet.

The balance sheet is a brilliant construct that simultaneously displays the resources of the world (assets) and the claims on it (liabilities and equity). However, the narrow scope of interpretation imposed on it is, arguably, ruinous. That is because the only people who really look at balance sheets are financiers and accountants; and, in the for profit world, the rules on constructing balance sheets prohibit the listing of most intangibles. There is no place on the asset side of Apple Inc. for the colors and sleek design that draw customers into an Apple store en masse, nor that of Dream Fluff Donuts (Berkeley) for the tantalizing first bite of their glazed buttermilk bar, nor that of MCG Jazz for the music, history and sense of belonging it has given jazz artists in America for the past 35 years.

No place, except a hint of the value in the time series of their receivables (which is disallowed in standard accounting). How much are people willing to pay for a good or service, and how stable is the demand? As an artist, you don’t have to produce Taylor Swift-like rates of return to earn out a stable revenue stream. You just have to develop a stable following. The information on payment stability translates into a measure of the professionalism of the artist and the loyalty of her or his client base.

Twenty-five years ago, securitization was invented to bridge a disconnect between valuing assets and valuing liabilities. The intangible value of the assets was quantifiable, but only when the receivables were analyzed using private data that are not part of the accounting disclosure package. A firm that was a BB to the financial world and so paid more to borrow money, might actually have AA asset quality and in principle could borrow funds more cheaply using an expanded information set.

The source of the disconnect is an incomplete information set that is biased in favor of lenders at the expense of creatives. In the music world, the same bias exists. It favors the studio, which takes the equity value created by musicians and gives them a salary in return. Unquestionably, studios deserve some of that equity for helping musicians commercialize their art. But the question is, where to draw the line between a fair rate of return and expropriation?

Securitization techniques go a long way to quantify intangible value. Using it to build incentives for creatives to create and giving them a path to build their business on and share equitably in the value they create may be an experiment worth investing in.


Chang on Changing Economic Dialogue

We can try to change the system, or we can try to change human psychology and behavior. For my money, attempting to change the system by substituting one vision for another is a better way to change the dialogue.

Ha-Joon Chang is a Cambridge economics professor, Guardian columnist and influential critic of capitalism, on a mission to change the dialogue on economics from a”god-given” puritanical framework that rewards status-quo wealth to a social system that can be improved, for the benefit of nearly everyone. His recent critique of the positive spin on austerity in The Guardian captured my attention:

Can funding for renewable energy ever be AAA risk?

Twenty years ago, rating agency analysts disagreed on whether non-AAA corporations could issue AAA-rated structured securities. Aircraft finance was the poster child of the debate. “Impossible. Too volatile,” corporate analysts reasoned. Structured analysts would retort: “Not even $1 of securities can be funded at AAA?” And, if $1 can be funded at AAA levels, what about $2? $3? $1 MM?

This spring, R&R got close to the emerging solar securitization market through committee work coordinated through the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL), which is working to promote solar energy and preparing for the imminent (2016) solar tax credit phase-out. In December 2013, NREL’s Travis Lowder and Michael Mendelsohn published a white paper, The Potential of Securitization in Solar PV Finance, which argues for securitization as replacement funding. R&R discovered that rating agency bias exists against solar PV lease securitization, similar to corporate. The agencies willing to issue ratings (some are not) are capping the rating at low Investment Grade risk levels.

Really? Not $1 of securities backed by energy that continuously self-generates is capable of being rated AAA?

Thomas Piketty: Le Monstre Sacré

Over the last few weeks, much ink has been spilled in an attempt to refute, undermine and marginalize the book “Capital in the 21st Century” by Thomas Piketty, a French economist. In some, unsurprisingly British quarters it has even reached the level of “the lady doth protest too much, methinks.” You cannot convince people of your viewpoint by just silencing the opposition. It’s a safe bet that, had a British economist said the same things or reached similar conclusions, no British institution would have attacked him so publicly. But who knows, perhaps “Le Monde” would have done so with gusto and joie de vivre! It’s clear that the Financial Times is not a proper forum to argue about economic time-series with any seriousness. Although one can find fault with the Mona Lisa, doing so does not take anything away from its value as a work of art.

Alibaba’s business model

Two days ago, Business Insider published a good summary of little known facts about Alibaba.

Several facts highlight how the mega trade-web dwarfs competitor e-commerce platforms in the U.S. and in China on several critical measures, including employees, registered users, web visits, sales throughput (on track to handle $1 TN in transactions) and revenues ($1.8 BN in 3Q2013).

Another competitive advantage of Alibaba is that it blocks search engine spiders, thereby retaining the control and benefit of most of the information published on its site. The March 23 2013 Economist article on Alibaba discusses the spider tactic in more detail and offers a deeper discussion of Alibaba’s information advantage generally.

But the most interesting aspect of Alibaba is one that has received almost no attention. That is Alibaba’s role in transforming and developing China’s SME financial services market.

Alibaba, together with its financial backer and deal sponsor Orient Securities Asset Management, is the micro-loan securitization pioneer in China. Approval was granted by the CSRC in June 2013 for the creation of Alibaba Separate-Account Asset Management, a master trust structure. It will issue 10 series, each backed by RMB 100-200 MM, with a 1-2 year maturity. The first series, backed by RMB 500 MM, priced with three tranches (A: 75%, B: 15% and C:10%) of a 15-month maturity. The bonds were listed on the Shenzhen Stock Exchange on September 25, 2013.

With this foray into self-financing, I believe it is well worth reconsidering Alibaba’s business model. Is it really a giant, diversified trade web? Or is it (as I believe) a brand new banking model for the 21st Century?

When an article on subprime goes bad

Floyd Norris, chief financial correspondent for the New York Times, gives a disturbingly uninformative  account of RMBS and CDO markets in his article, When a Deal Goes Bad, Blame the Ratings, November 14, 2013. Not having done the research to really understand the causes of the credit bubble, he fails to see the irony of his words: Did you make an incredibly bad decision during the great credit bubble? Don’t worry. Join the crowd denying responsibility. 



Smiley face

Latest Post:

Automatic Synthesis of Tri-Diagonal Markov Transition Matrices: A Failure of the Inverse Ferron-Frobenius AlgorithmABSTRACT: Sylvain Raynes experimented with a method developed by Goya and Boyarski (1993) to standardize the synthesis of conditional Markov transition matrices for deal entry in our automated re-rating system, ABSTRAK®. In the analytical literature, the reverse-engineering of a Markov matrix from its spectral-radius eigenvector is referred to as the Inverse Perron-Frobenius Problem. Our analysts do this synthesis manually, so a successful outcome […]

Click for all posts

Ann Rutledge on China

Smiley face

Latest Post:


Click for all posts