New paper on the Indian IPO market

Radha Gopalan and Todd A. Gormley have a new paper The Going Public Decision and the Role of Public Equity Markets in Emerging Economies which has the abstract:

We highlight the role of public equity markets in firm financing by studying firms' going public decision in India. The availability of detailed firm-level data for a large sample of private and public firms combined with the liberalization of the IPO market in 1992 and its subsequent collapse in 1997 make India an ideal setting to analyze the role of public equity markets. We find that firms going public after liberalization are smaller, younger, from industries with greater investment opportunities, and less likely to be affiliated with a business group then other private firms. The firms also exhibit significant increases in capital expenditure and sales around the time of their IPO. Highlighting a lack of alternate sources of capital, the equity market collapse in 1997 adversely affects the performance of private firms with characteristics similar to pre-1997 IPO firms. The firms that go public during 1992-1996 also exhibit sharp declines in sales growth rates, profitability and increases in leverage and bankruptcy rates beginning in 1997. Reflecting a lack of continued equity financing, the fall in sales growth is more severe for firms from industries with greater dependence on external finance and for firms with lower asset tangibility. Overall, the evidence suggests that public equity markets play a significant role in expanding access to finance for small, young firms in emerging markets.

Look at this search on google scholar for India IPO.

The holding company structure for financial firms

An `RBI Discussion Paper on Holding Companies in Banking Groups' has brought questions about the structure of financial firms to prominence.

I think that competition policy is a core principle, and that government should not prevent a firm that makes trucks from embarking upon making cars. In my opinion, achieving efficient and world-class financial firms critically requires breaking down the walls that have been artificially created within Indian finance. Such efforts in financial sector reforms would help achieve multi-product firms where competition, innovation and efficiency considerations determine what a financial firm does. For the forseeable future, in India, we're going to have atleast four financial regulators covering banking, securities, insurance and pensions. The clean holding company structure involves an unregulated mother company - which might typically be listed - which works as a corporate headquarters which has four subsidiaries. Each subsidiary might deal with one regulator, and each subsidiary would operate like a division of a multiproduct firm.

Government agencies might point out that, given their present organisation, this configuration makes life difficult for the regulator. However, it is only fair to propose that government must be restructured to fit the needs of an efficient and competitive industry, and not the other way around.

On this subject, I would like to point you to Section 2.1, titled `The holding company structure' in Chatper 11 `Reforming financial regime governance' of the MIFC report.

The wonderful world of algorithmic trading

A pair of articles by Will Acworth and Bennett Voyles in Futures Industry Magazine take you to the edge of the wonderful new world of algorithmic trading. A few remarkable things that I found:

• CME gives you co-location facilities for $6,000 per connection (not counting co-location facility charges), so that you can place your servers closer to the exchange and reduce latency.
• Eurex's co-location gets you a round-trip of below 10 milliseconds, when compared with 29 milliseconds for a connection in London and 128 milliseconds for a connection from Chicago.
• ICE is worrying about a new notion of quality of service: no more than 0.1% of the messages should get processed in worse than 50 milliseconds, once a new trading engine is in place.
• Acworth says: In the old days, a trader might be watching four screens at the same time, commented one speaker. Now its two rows of four, with one row displaying the markets and the other row displaying network conditions.
• People like the fact that in C++ you get to control when garbage collection is done, so as to avoid unpredictable glitches in performance. (Ugh!)
• The average order size on the E-mini has dropped to 2 contracts.
• The biggest systems are nearing a million messages a second.
• Some firms have started moving functions from software to FPGA in the quest for speed.

A million messages a second is a lot! I remember in the late 1990s, when I worked on the PRISM system (which does the realtime VaR at NSE), our goal was to be able to handle 100 trades/s, which is 200 VaR calculations a second. We were thrilled because the system evolved into exceeding 1000 trades a second.

For those interested in algorithmic trading, you may like to see Chapter 5 of the MIFC report.

Combating the costs of fund management

At a wholesale level, fund management can be very cheap, with customers being charged as little as one basis point out of assets under management a year. At a fee of one basis point, a fund manager who runs Rs.100,000 of your money earns Rs.10 a year out of it. This is the price achieved by bulk fund management abroad and in India.

These low prices are just not visible to the end customer in India. EPFO is expensive. Most mutual funds are very expensive. The fund management products sold by insurance companies are outrageously expensive. The key problem with mutual funds and insurance companies appears to be the distributor-centric business model. Achieving lower fees requires (a) Modifying the implementation of that model or (b) Starting over with a new business model.

SEBI has recently floated a new proposal to remove loads on direct applicants for mutual funds, who would then have a choice to achieve a lower cost channel by doing it on their own, when compared with the high costs associated with a distributor. See reports by Rachna Monga in Mint and George Mathew and Sandeep Singh in Indian Express. For particularly informed commentary on this proposal, see Ajit Dayal in Indian Express, and Gautam Chikermane, also in Indian Express.

I am sympathetic to this line of thought. What SEBI is proposing is a step forward. However, I worry that it represents a small move which does not challenge the fundamental agent-driven business model. What the agents could lose in this might easily be made up in other ways.

These recent events have brought a fresh focus upon ways to achieve low cost fund management. Four elements come to mind:

Avoid fund management products sold by insurance companies

The worst excesses in terms of fees and expenses seem to be emanating from insurance companies. On this subject, see this piece by Deepti Bhaskaran in Express Money. When your friendly neighbourhood accountant comes to you, suggesting that you sign up for an insurance product so as to save taxes, it would help to think thrice about it.

Quantum Mutual Fund

Ajit Dayal's outfit is a rare mutual fund which is trying to break with the distributors. If active management is your thing, then this is a good way to go. But unlike (say) Vanguard, quantum is not single-mindedly fighting to bring down fees and expenses - a thought process that inevitably takes you to index funds.

Index funds and ETFs

The cheapest fund management products in India today are the index funds, particularly the ETFs. Benchmark's Nifty fund charges an AUM fee of 45 basis points. As an aside, Benchmark staff informally says that if there is a 10x increase in the AUM, this price can come down to 20 bps and if there is a 100x increase in the AUM, the price can come down to 7 bps. A key insight of the ETF path is that the customer gets his work done through the extremely competitive, transparent and efficient stock broking industry. A customer who goes to his stock broker seeking to buy an ETF gets a low price with no hidden catches - unlike the customer who deals with an agent and tries to buy a conventional mutual fund product or insurance product.

The New Pension System (NPS)

The best eye-popping numbers for saving money in fund management come from the New Pension System. These are early days and it is not yet clear how small these numbers will be. Stay tuned, for the NPS reflects a full-blown thought process of fully redesigning a fund management mechanism so as to achieve a good deal for the customer.

Update: Business Standard has an editorial on these issues:

A simple idea that is well appreciated worldwide is the importance of fighting fees and expenses in fund management. Good returns in fund management are a possibility, but the fees and expenses paid are a certainty. The international evidence suggests that paying more for fund management generates reduced returns. A remarkable twist to this tale in India lies in the “distributors”, or agents, who sell fund management products. Fund managers have resorted to making very large payments to these agents, to the point where the profits of the agent sometimes exceed the profits of the financial firm. The chief executives of financial firms privately gripe about this state of affairs, but have not been willing to break with the existing system. As a consequence, customers of mutual funds, and particularly those of insurance companies, are getting a poor deal. All too often, customers of insurance companies are blindly placing money in schemes with the intention of saving taxes, and ending up fuelling the profits of the agent who is selling them the scheme. The non-transparency and high charges that are in place with most mutual funds, and particularly with insurance companies, fuel the discomfort of critics of modern finance, who see “professional fund management” as a way to channel the money of ordinary citizens to financial fat cats.

Exchange traded funds (ETFs) are a promising way out, because the customer is able to go to a stock broker to get invested, and stock broking is a transparent, competitive and efficient industry with very low charges. The lowest-cost fund management accessible to an individual in India is the ETF. Only one active management vehicle — Quantum Mutual Fund — has broken with the agent system, and has so far experienced limited success.

The Securities and Exchange Board of India (Sebi) has now proposed that customers be given a zero-cost alternative whereby they are able to use the Internet and get their money to the mutual fund while bypassing the agent. This is a sound initiative. The insurance regulator, by way of contrast, has offered no positive proposal for addressing the egregious practices of insurance companies. However, the essence of the problem lies not in tweaking the agent system but fundamentally re-shaping the fund management industry.

The New Pension System (NPS) represents an effort at using public policy in order to address the market failures of the fund management industry. In the NPS, the selection of fund managers is done on the basis of an auction where the sum total of fees and expenses is stated up front by the fund manager. This ensures that customers get low prices. Centralisation of record-keeping at the “Central Recordkeeping Agency” (CRA) and an unbundling of customer front-ending to Points of Presence (POPs) help to reduce dramatically the costs of bulk fund management. In a recent auction conducted by the Coal Miners Provident Fund, the winning bid was as low as one basis point of assets per annum.

Many details remain to be worked out — how the CRA will operate, what the charges of the CRA and the PFMs will be, how POPs will be incentivised in a way that avoids the flaws of mutual funds and insurance companies, and how every detail of operation will be structured so as to constantly put pressure in favour of obtaining low fees and expenses. The onus is upon the ministry of finance and the pension regulator to translate the promise of the NPS into reality.

Day one in the implementation of the New Pension System

On 10 April this year, I wrote a blog entry: Day zero of implementation of the New Pension System. From 10 April till 24 August, SEBI tried to prevent NSDL from becoming the Central Recordkeeping Agency for the New Pension System (NPS). This explicit ban ended yesterday. So, 136 days later, we are now at day one of implementation of the New Pension System, though encumbered by strictures imposed on NSDL by SEBI.

Okay, 135 days later, now we get going on building the New Pension System. I think it's going to be a very exciting time from here on.

In parallel, it looks like there may be some movement on new recruits into PSU banks being placed into the NPS.

Competition policy perspective on banking

There is a news item in FE saying that the Competition Commission of India has started waking up to the anti-competitive policies in banking [link]. The MIFC report has written extensively about the importance of a competition policy perspective on Indian finance.

Reforming the role of credit rating agencies

Business Standard has an editorial on rating agencies:

Each crisis leads to incremental improvement of the institutional structure of the global financial system. The “sub-prime crisis” may generate important reforms in the role and functioning of credit rating agencies, which have come under attack for overly generous ratings that misled the market. On the one hand, it is always easy for investors to complain about losing trades. However, genuine mistakes do appear to have been made by credit rating agencies, who under-estimated the correlations between defaults by households, under-estimated the effect of higher interest rates on defaults, and under-estimated the consequences of default upon home prices and thus further defaults.

Such genuine mistakes can be understood when they come from an impartial analyst. But they look particularly shocking in the context of the close nexus implicit in the rating agencies working jointly with financial firms in structuring products, and getting paid handsomely for this work. When the sub-prime market grew from $120 billion in 2001 to $600 billion in 2006, the rating agencies profited handsomely. The behaviour of a paid analyst who is a party to the launch of a product is inherently different from the behaviour of an impartial analyst who is a bystander.

The European Union says it plans to examine the agencies’ roles, and investigate possible conflicts of interest between the agencies and the issuers of mortgage bonds. In Washington, Representative Barney Frank, the chairman of the House financial services committee, is planning hearings in October to examine the same issue.

The first key reform that is now being discussed is to roll back to the pre-1970 situation, when credit rating agencies performed the job of being independent research companies who sold subscription services based on their true merits, and never took payments from the firms that they rated. The second key reform that is being proposed is the removal of credit rating agencies from the regulatory treatment of institutional investors.

The appeal of these proposals is that they would convert credit rating agencies into genuine information and research companies, working at a healthy distance from financial transactions. A rating agency would rate a bond in the hope that investors would like to pay for the subscription service. Institutional investors would judge credit rating agencies alongside other information and research vendors, all of which offer comparable information and research services. In such a world, credit rating agencies would have to pass the market test, instead of being government-supported gatekeepers.

These reforms are particularly pertinent in India, where it is now practically impossible to do a primary issue of a bond without the involvement of a credit rating agency. Fees to credit rating agencies have become akin to a tax. Infosys has zero debt, and its first Rs 1,000 crore bond issue should surely face a good market in a rational world without any credit rating. It is better to trust the processes of the competitive and speculative market, rather than trying to install a set of government-supported profit-making gatekeepers.

Recent Sebi proposals involve mission-creep, from bond issuance to equity issuance. It is proposed that initial public offers (IPOs) should be “graded”. All these criticisms apply equally here. If a credit rating agency (or any research firm) has something useful to say about an upcoming IPO, it should pass the market test of being able to sell a subscription service to investors.

While on this subject, you may like to see this blog entry; We need a better way to judge risk in FT by Charles Calomiris and Joseph Mason; Overrated in portfolio by Jesse Eisinger. Update: The CEO of Standard & Poors resigned, possibly as a consequence of difficulties with ratings of securitisation products. I would always argue that what matters is not individuals but incentives. Witchhunts don't help; our goal should be to reshape the incentives surrounding the bright and well-motivated people who work at credit rating agencies.

Update: Business Standard carries a debate between Susan Thomas and Roopa Kudva on a related theme:

Susan Thomas: Concerns about credit rating agencies (CRAs) have been gathering weight in the world of finance for a while now. Starting from the problems of highly rated companies like Enron which proved to be insolvent, to countries like Argentina, to the more recent problems of the default of sub-prime loans, the reputation of CRAs have been taking a beating over the last two decades. In response, new ways of thinking about credit risk have proliferated. One of the most important alternatives for thinking about the default risk of a corporate bond is the “KMV model” which utilises the stock price to produce continuously updated measurement of the default probability of a firm.

Financial practitioners have already de-emphasised the role of CRAs. In a competitive market, CRAs would have faded away, or reinvented themselves.

The distortion in the role of the CRA comes from the responsibilities vested by the government — when regulators of different financial intermediaries mandated that credit ratings be obtained before these intermediaries could invest in the bonds of firms. This paved the path for firms to pay CRAs to get their bonds rated, in order to get investments from banks or insurance companies. But once the firms started paying for their ratings, the incentive structure for the CRA to do a trustworthy job became bent towards more favourable ratings. This perverse incentive reached a pinnacle with the role of the CRA in helping to structure securitised products such as mortgage loans or sub-prime credit products. Given that there would tend to be a larger fraction of poor credit to good credit in any economy, sub-prime loans was a volumes business that earned the CRAs large profits.

By definition, the sub-prime loans business was also greater risk, which showed up as large losses with the increase in high market volatility of the last six months or so. CRAs are facing the responsibility of what they reaped such rich profits from: as they reaped, so they are sowing. In a competitive market, there would not have been the regulatory mandate the CRAs have benefitted from. In a competitive market, existing CRAs would lose market share to more sophisticated alternatives. India is specially worrisome when it comes to the regulatory responsibility of credit ratings. Listed firms have to mandatorily get credit ratings for bonds; non-government pension funds and gratuity funds can invest only in investment grade bonds with two credit ratings; commercial banks can invest only in rated non-SLR bonds; and the latest peculiarity, a Sebi mandate that IPOs must be “graded”. If these regulatory responsibilities persist in the role of the Indian CRA, then we will indeed have to get serious about “rating the raters”. Instead, if the credit information business can be fundamentally re-engineered so that CRAs pass the market test and compete to obtain subscription revenues from investors, without payments from firms for credit information sent by government agencies, we won’t have to worry about it.

Roopa Kudva: Absolutely yes. And indeed rating agencies are constantly subject to scrutiny, evaluation and questioning by investors, media and regulators. Since ratings are opinions, it is important that markets are convinced about their robustness before acting on them. Rating agencies therefore publish extensive data on rating default and transition statistics, and metrics on predictive capability of ratings vis-à-vis macro-economic and corporate performance. CRISIL regularly holds investor discussion forums where its methodologies and views are hotly debated by about 100 analysts present at each such event.

How should raters be evaluated? Firstly, through a long-term record demonstrating that higher ratings are consistently more stable and have a lower probability of default than lower ones. Adherence to clearly stated and widely disseminated criteria is another parameter. The governance and business practices are critical in evaluating their independence. Some of these include: multi-layer decision-making processes, dedicated criteria and quality assurance teams, prohibition of analysts’ involvement in fee decisions, and analyst compensation not linked to assigned ratings. India’s rating industry scores well on these counts. There is a two-decade history of credible and robust default statistics. Since credibility is its bedrock, the rating industry has proactively and suo motu provided its report card to the market.

The issuer-pays model is much debated, but would investors paying work? When a rating is assigned, the investor is generally not known. Issuers have to commission ratings and provide them to a range of potential investors who then decide whether to invest. And if investors paid, only paying investors would get access to ratings. Markets should not underestimate the huge benefits of the public disclosure of ratings — today all ratings and rating changes are available to the entire market free of charge, as they are widely disseminated.

Rating agencies are the most independent providers of credit opinions compared to other potential opinion providers — borrowers, lenders, investment bankers or brokers. Ratings have a vital role to play in India’s credit markets. The Indian rating agencies proactively alerted the market to the impending risks in the NBFC sector in the 1990s, well before the crisis precipitated. The contingent liabilities of state government guarantees as a risk factor were first highlighted by CRISIL. The potential problem in the collective investment plantation schemes was nipped in the bud by rating agency alerts. The recent actions by CRISIL highlighting the heightened risks associated with leverage in corporate balance sheets to fund their global expansion were keenly appreciated by the market. Not surprisingly, Indian investors’ acceptance of ratings preceded regulatory recognition. Even today, investors demand ratings in areas which go well beyond those where regulations make them mandatory.