Transparency matters

Market transparency is a central design feature of financial markets. One traditional motivation which motivated transparency was concerns about ethical standards. Transparent markets are innately less vulnerable to malpractice, and this has encouraged many regulators to favour transparency as a vehicle for reducing the costs of monitoring and enforcement. This is particularly important in a context like India, where high quality enforcement of laws is extremely costly in terms of monetary resources, political costs and the costs of time and focus of policy makers.

In recent decades, there are three new arguments which favour transparent markets: (a) the idea that enhanced market transparency leads to greater liquidity, (b) the idea that the anonymous electronic exchange makes quantitative measurement and control of transactions costs possible and (c) the idea that a transparent market makes risk management much easier by supporting marking to market and VaR estimation.

All these arguments notwithstanding, when we look at financial markets in the real world, high standards of transparency are not often found.

In India, the equity market has made enormous progress towards market transparency. When NSE built a stock market based on anonymous, electronic order matching, it became India's largest stock market within a period of one year, and all other stock markets in India followed suit. In recent weeks SEBI has taken the momentous decision to force all market participants (even institutions) to expose their orders to the discipline of the screen - i.e. private transactions conducted purely over telephone would be forbidden.

In contrast, India's fixed income and currency markets lag far behind in terms of market design. These markets are characterised by entry barriers in intermediation, domination of a few large players who engage in strategic games, private negotiation of deals, and a lack of public observability of prices and liquidity. When X buys from Y, external observers do not know anything about what transpired, and even if you would be happy to sell to X at a lower price than that offered by Y, there is no way for you to break into the transaction.

The most limited notion of market transparency is one where external observers get an opportunity to know one "reference rate" for the day. This is a very limited idea, since individual orders and trades are not reported. However, there is more transparency in a market with a reference rate as compared to one without.

On 15 June 1998, NSE commenced the dissemination of the MIBID/MIBOR reference rates for the overnight call money market. From 7 September 1998 onwards, NSE started calculating reference rates for 14--day, 30--day and 90--day interest rates, however, these new rates were not disseminated.

On 10 November 1998, dissemination of the 14--day rate began, and on 1 December 1998 dissemination of the 30--day and 90--day rates began. These events are valuable natural experiments where we can observe the impact of improved transparency on market liquidity. For days before dissemation, the reference rates observed by NSE (which were not disseminated) tell us about market quality without transparency.

In a recent paper A natural experiment in market transparency by Susan Thomas, Mandira Sarma and myself, we examined the empirical evidence to learn about the impact of these three natural experiments. At the simplest, the bid-ask spread - the best measure of market liquidity - dropped by 19.9 basis points (14--day), 11.8 (30--day) and 21.5 basis points (90--day) in the days after transparency as compared with the days before. After controlling for fluctuations in the spread owing to external factors, we find that the spread dropped by 22.8 basis points (14--day), 13.6 basis points (30--day) and 22 basis points (90--day) with extremely high levels of statistical significance. In all three cases, the spreads stood at 80 to 90 basis points before the reference rates became publicly available, which helps suggest that these reductions are quite substantial in economic terms.

This experiment is a valuable illustration of the impact upon market liquidity of the most limited notion of market transparency: the public reporting of "the market price" once a day. It suggests that even small gains in market transparency have a big impact.

NSE's Wholesale Debt Market (WDM) is often criticised for not having true anonymous trading like NSE's equity market. This research suggests that even if WDM plays a highly limited role of information dissemation (and not true anonymous order matching), this is a valuable function as compared with a 100% non-transparent market.

With this backdrop, we can think about RBI's announcements about interest rate swaps (IRS) and forward rate agreements (FRA). The vision for market design that is implied in these announcements is one that is rooted in market non--transparency.

These markets operate on a purely bilateral basis. There is no public reporting of transactions, there is no public visibility of unmatched orders or liquidity, and trading procedures lack anonymity. The only information which has come out to the public about IRS or FRA transactions is based on journalists interviewing dealers. I am not aware of whether the RBI knows the open interest of the market on a same--day basis; external observers certainly do not.

The management of systemic risk is innately harder with bilaterally negotiated contracts. I am not aware of the extent to which modern risk measurement is present in the IRS or FRA markets. Financial market regulators has frequenty voiced concerns about their inability to comprehend the risks present in bilateral derivatives transactions and to control the probability of systemic crises. In his book Derivatives : The Wild Beast of Finance, Alfred Steinherr has a beautiful analogy. He compares the exchange-traded derivatives market to a fishtank, and the universe of privately negotiated derivatives to the great wide sea. Regulators and policy makers can monitor the fishtank, and have a good sense of risk containment and fraud in the market: this is very hard to achieve in the sea.

In the case of the FRA, there is a mature and well--understood alternative: interest rate futures. Interest rate futures are synonymous to interest rate forwards, with the difference that interest rate futures use superior standards of transparency and risk containment. Interest rate futures would give 100% transparency as opposed to what is present with the FRA market.

In the case of the IRS also, it is possible to design superior market mechanisms in a variety of directions: it is possible to have realtime trade reporting; it is possible to employ risk management through novation at the clearing corporation; it is possible to define standardised IRS contracts which trade at an exchange with anonymity. However, the vision for market design that is implicit in the existing IRS market is one of 100% non-transparency.

In summary, market transparency should be a central goal of policy in the area of financial markets, for the wide spectrum of reasons: ethics, fairplay, enhanced liquidity, enhanced measurement and control of transactions costs, and improved risk management. The currency, commodity and fixed income markets in India could benefit enormously from making the transition into anonymous order matching on an exchange.

Back up to Ajay Shah's media page