Rating agencies want to diversify

Although its business had yet to get off the ground, Reliance Power had a grade of 4 from CRISIL, which indicated that it had above-average fundamentals. Between the listing date of February 11, 2008, and September 12 this year, shareholders lost 83 per cent on the scrip, while the Sensex declined by just 5.5 per cent on a pointto-point basis. On the other hand, Bhagwati Banquets & Hotels, to which CRISIL assigned a grade of 1 for poor fundamentals, delivered 106 per cent returns after listing in May 2007, in stark contrast to a 19.6 per cent rise in the Sensex. One more example: Parabolic Drugs, which got a Grade 2 from CARE Ratings and Grade 3 from Brickwork Ratings, is now down 60 per cent since it listed in July 2010, compared to a 6.1 per cent drop in the Sensex.

These are not isolated cases. Many say this sort of thing is happening because the core business of credit rating agencies in India is becoming a smaller part of the work they do.

On the surface, credit rating agencies are in good health. CRISIL and ICRA, the two listed rating agencies among the six that operate in India, have seen revenues rise by more than 200 per cent in the last five years. CRISIL's revenues expanded 287 per cent, from Rs 156.8 crore in December 2006 to Rs 606.8 crore in December 2010. ICRA's grew more than three times from Rs 46.8 crore in March 2007 to Rs 141.8 crore in March 2011. Today, CRISIL calls itself a global analytical company which provides rating, research, and risk and policy advisory services. Five years ago, its India operations accounted for more than 80 per cent of its revenue. That contribution is now 50 per cent, and ratings account for just a third of revenue. "We have continuously pursued diversification and innovation as a strategy," says Roopa Kudva, CRISIL's Managing Director and Chief Executive.

CRISIL's Roopa Kudva in BT's Most  Powerful Women in Business list

Rating agencies say they must diversify, because the lack of a vibrant corporate debt market in India leaves them with little choice. They have a point. According to the Bank for International Settlements, based in Basel, Switzerland, the United States had the lion's share ($2.9 trillion) of total corporate debt securities of $6.7 trillion outstanding at the end of December 2010, while India had just $25 billion.

Some in the industry say that listed rating agencies are under greater pressure to diversify. CRISIL, with global ratings major Standard & Poor's, or S&P, as its 52.43 per cent shareholder, has a market capitalisation of Rs 5,900 crore. And ICRA, in which Moody's has a 28.51 per cent holding, has a market cap of Rs 918 crore. Like any listed company, a rating agency must deliver consistent revenue growth and keep its shareholders happy. Many in the industry argue that, given the lack of a sizeable debt market, agencies look beyond their core business and end up grading business schools , microfinance institutions, mutual fund schemes, initial public offerings or IPOs, real estate development projects, and even hospitals and maritime training institutes. Demand from various industry regulators is also partly responsible.

"The Director General of Shipping wants to ensure that maritime training institutes have the wherewithal for training," says D.R. Dogra, Managing Director and Chief Executive, CARE Ratings. ICRA's Managing Director, Naresh Takkar, says the stock market regulator, the Securities and Exchange Board of India, or SEBI, requires IPOs to be graded on the fundamental strengths of the companies.

The pressure on listed agencies to grow, coupled with competition, spurs them to undercut each other, and this could hurt assessment quality. Agencies deny this, but many market players differ. The chief executive of a rating agency says: "We lose a lot of business because our competitors charge nearly half of what we do."

A Harvard Business School working paper on how competition affects credit ratings, written by finance professors Bo Becker and Todd Milbourn and published a year ago, notes that competition made things worse in an industry historically dominated by Moody's and S&P. "Increased competition from Fitch coincides with lower quality ratings from the incumbents," say the authors.

"Rating levels went up, the correlation between ratings and market-implied yields fell, and the ability of ratings to predict default deteriorated."

In India, rating agencies deny that competition leads to 'rating inflation'. D. Ravishankar, Founder Director of Brickwork Ratings, an agency he co-founded in 2008 after a decade at CRISIL, argues that, while issuers (companies that are being rated) do not like low ratings, lenders and investors do not appreciate higher ratings as they usually mean lower yields.

But ratings industry insiders concur with Becker and Milbourn, adding that a new agency, in the initial years, will lure clients with better ratings, and that its competitors will follow suit. They add that large issuers are repeat customers, so keeping them happy is important. CRISIL's Kudva says diversification is the way out. "You are not dependent on issuers when your business model is diversified," she says. Diversification means getting to know new businesses and roping in relevant expertise.

Diversification challenges
Giving the example of business schools - CRISIL has rated 40 since January - Kudva says: "We spent 19 months developing the methodology, and did shadow grading of 150 business schools before launching the service." Similarly, for real estate projects, she says, CRISIL uses city-specific ratings, and has hired people with appropriate legal and technical skills.

Rating agencies also assess loan seekers. A recent trend is to rate small and medium enterprises, or SMEs. CRISIL, which had rated more than 20,000 SMEs up to April this year, maintains that this is a natural extension of its core business. The trend of grading IPOs is perhaps more of a cause for concern.

While some market experts say agencies lack the relevant skills, ICRA's Takkar differs. "The fundamental assessment is similar to corporate ratings," he says. Rating agencies grade IPOs on fundamental strengths, and SEBI does not expect them to comment on issue pricing.

MUST READ: Firms seek ratings for raising funds

But some argue that grading on just fundamentals at a given point of time is inappropriate. "Equity is a high-risk, long-term instrument, and requires constant review," says Prithvi Haldea, Chairman and Managing Director of Prime Database, which tracks the primary market. "A grade that doesn't factor in the offer price gives investors a misleading message."

A former member of SEBI's Primary Market Adv i s o ry Committee, Haldea points out that India is the only country where credit rating agencies grade IPOs. "The anomaly is that assessment is based on the past, while the instrument looks to the future," he says. "Worse, grading is absolute, not relative," he adds, alluding to the fact that agencies do not compare issuers with sectoral peers.

Not everyone agrees. "What should be questioned is pricing, not grading," says Atul Joshi, Managing Director and Chief Executive of Fitch Ratings India. "Should investors pay a premium for an IPO graded as having poor fundamentals?" But data shows that IPO grading is not reliable (see They Got It Wrong). "Experience shows little relationship between grades and returns," says Haldea. While an IPO's value can only be known in the medium to long term, an IPO grade, as CARE's Dogra points out, does not have such a long shelf life.

IPO grading is aimed at retail investors, not institutional ones. "Once you go for a grading system, you risk driving these investors to a number, and away from reading the offer document and making an informed decision," says Haldea.

Glossing over complexity
An extension of IPO grading is the equity research that some rating agencies do. Research reports are far removed from the core business, but CRISIL's Kudva argues in favour of such products. Of 5,000 listed companies in India, she says, research reports are available for just over 150. "And these come from sell-side firms," she adds, referring to brokerages.

Dogra points out that rating agencies have no conflict of interest, as they do not buy or sell the stocks they research. But the chief executive of a domestic broking firm points out that research is usually done at the company's request, which affects the output, mirroring the traditional 'issuer pays' business model of the ratings business.

Another diversification is the rating of microfinance institutions, or MFIs. Chennai-based Ramesh S. Arunachalam, researcher and author of the book The Journey of Indian Microfinance: Lessons for The Future, says rating MFIs requires specialised skills, since the business involves large numbers of small loans, often in remote locations. He says: "Rating agencies have a few days in which they are guided to scrutinise an MFI's operations and systems."

CRISIL's Kudva concedes that guided tours have shortcomings, and, unlike the audit process, the scrutiny lacks the element of surprise. "But we spend time with the management and in the field to do a fair job," she says. The business model of microfinance is to take wholesale loans from one or more lenders, and then lend the same money to small borrowers.

"All the players collectively underplay risk," says Amit Tandon, former Managing Director of Fitch Ratings India. He cites a November 2010 Fitch report on Indian microfinance securitisation which highlights the complexity of unmitigated risk. "At the end of the day, ratings are relative," he says.

The core business of rating companies and instruments has its own set of problems. There is the 'issuer pays' model, discussed above, with its potential bias, a concern that rating agencies say is unfounded.

ICRA's Takkar says when issuers pay, rating information becomes available to more investors. If investors paid, circulation would be restricted. Dogra of CARE says an 'investor pays' model could reduce cooperation by the issuer. Brickwork's Ravishankar adds that if investors paid, they would push for lower ratings so they could command higher yields. Takkar says that if the motive was to please issuers, his agency's ratings would have averaged higher than BB or BBB.

Shopping for ratings
Shopping for ratings is another concern. As a ratings industry insider puts it, issuers know that in today's market, a rating of AA- or higher gets their instrument placed with investors or lenders with the least hassle. So every issuer aims for at least an AA- rating, whatever the instrument. Merchant bankers make money only if the instrument is placed, so they favour higher ratings, too, he says. But investors have just one top concern. "They want higher yields," he says. They would be happy to hold AAA paper (low risk) which gives the yield of an AA one (higher risk). So rating agencies fall in line with these expectations, he concludes. This is why internal safeguards are crucial. "This is where you need to look at rating withdrawals," says Tandon.

A case in point is Tech Mahindra, which got a AAA rating with a stable outlook from Fitch on November 10, 2008. On March 20, 2009, Fitch withdrew the rating, saying Tech Mahindra had submitted an expression of interest to buy Satyam Computer Services, which could have a financial impact.

Interestingly, the next day, CARE assigned Tech Mahindra a AAA rating. Business Today asked CARE whether ratings could be assigned overnight. CARE's Dogra said Tech Mahindra's liability in Satyam was limited to the proportion of the equity it had bought. "It is possible that Fitch's comfort level with Tech Mahindra, and its view of the transaction, differed from ours," says Dogra. The concern was that if Tech Mahindra bought Satyam, its financials would change substantially. After Tech Mahindra entered the bidding process, Fitch could have moved its rating to 'negative watch'. "But Tech Mahindra was not comfortable," says a source privy to these developments.

Tech Mahindra did not comment despite many requests by BT. The story does not end there. On April 13, 2009, Satyam announced that Venturbay Consultants, a subsidiary controlled by Tech Mahindra, was the highest bidder. On September 29, 2010 - a full 17 months after Fitch's withdrawal - Mahindra Satyam (Satyam's new avatar) published its financials for the period from April 1, 2002 to March 31, 2010. From the rating agency's perspective, this had been a huge information void for a company whose promoters were implicated in financial irregularities.

In its defence, CARE wrote to BT that it gave Tech Mahindra a AAA rating because of its experienced management, track record, domain expertise in information technology for telecommunications, comfortable gearing ratios and liquidity position, and healthy order book. Although Tech Mahindra's bid for Satyam was public, CARE's March 21, 2009, release said: "The company has planned 'some acquisitions' .... The legal and financial implications ... could impact the rating." CARE points out that the rating had factored in the company's existing debt, and that its release said an increase would trigger a review. It says: "The review was done post this acquisition and the rating was reaffirmed, as the debt raised was within the amount projected to be raised while assigning the rating."

With CARE's AAA rating, Tech Mahindra tapped the market with Rs 600-crore non-convertible debentures (NCDs) and a Rs 275-crore commercial paper issue, which were lapped up by mutual funds, banks and insurance companies. On April 16, 2009, Business Line noted that, given falling interest rates and easy liquidity, the 10.25 per cent interest on NCDs of four and five years' maturity and 8.5 per cent on commercial paper of one year's maturity, plus the fact that Tech Mahindra was an almost zero-debt company, proved to be "the clincher" for the IT company. The paper quoted a senior bank official as saying: "The yield on 10-year government securities is around 6.4 per cent. So by comparison, the return offered by Tech Mahindra's issue is excellent."

Generally, higher rated companies bargain for lower interest rates, but Tech Mahindra had to do the reverse. One merchant banker points out that even with a AAA rating, the company paid nearly four percentage points more than government securities, which suggested that ratings were irrelevant, and that investors got paid for the risk they took.

Managing conflicts
Sometimes, an agency assigns a higher or lower rating than its rivals. For instance, in early August, when S&P downgraded the US sovereign credit rating from AAA to AA+, Moody's and Fitch took no action, keeping their AAA ratings unchanged.

And CARE's Dogra, who says credit rating is not a completely objective and mathematical process, points out that Moody's rated India BBB- for four consecutive years, but S&P did not lower its rating from BB+ to BBB-. That one notch was the difference between investment and speculative grade. "A difference of a notch in credit ratings is accepted," says Dogra. But experts say the logic that applies to sovereign debt does not transfer over to corporate ratings.

"You are paid to rate corporates," says a financial sector veteran. Sometimes the difference is more than a notch. Take the case of Coastal Gujarat Power Ltd, or CGPL, held by Tata Power Company. CGPL is implementing a 4,000-megawatt power project in Mundra, Gujarat. On February 9, 2009, Fitch assigned a BBB rating and a stable outlook to its project bank loans, which stayed in effect until April 27, 2011. Then Fitch moved the rating to the 'non-monitored' category, citing a lack of adequate information from the company. But CRISIL, which had assigned CGPL an A+ stable rating on April 6, 2010, upped the outlook on May 5, 2011 - after Fitch's move - from stable to positive.

CRISIL's Kudva explains that the agency factored in the support of Tata Power, which had an AA+ rating, and raised the standalone rating of CGPL. "When we rate Tata Power, too, we take the CGPL liability into account," says Kudva. She does not think the difference between BBB and A+ is stark. "When you're making a judgment about the quality of ratings, you must look at the agency's default and transition rates." The default rate refers to the number of defaults among rated entities during a specified period, expressed as a percentage of the total number of entities whose ratings were outstanding during that time. The transition rate indicates the probability of a change in the credit rating over a specified period.

Kudva says that conflicts of interest exist in one form or another across the financial services spectrum. "The challenge is how well you manage them," she says. CRISIL and Fitch say they follow practices such as ratings being assigned by a team, rather than an individual, delinking analysts' remuneration from the number or type of ratings given, and separating business development and ratings.

The ratings industry's challenge is to be on top of analysis, with upto-date surveillance that keeps ratings relevant. Brickwork's Ravishankar says: "Ours is the only business in which people are paid to tell customers how bad they are."

Edited by Uma Asher