Contents

1 Basics 5

Q1.1 What are derivatives? . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

Q1.2 What is a forward contract? . . . . . . . . . . . . . . . . . . . . . . . 5

Q1.3 Why is forward contracting useful? . . . . . . . . . . . . . . . . . . . . 5

Q1.4 What are the problems of forward markets? . . . . . . . . . . . . . . 5

Q1.5 What is a futures contract? . . . . . . . . . . . . . . . . . . . . . . . . 5

Q1.6 Why is the cash market in India said to have futures-style settlement? 6

Q1.7 What is an option? . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6

Q1.8 What are \exotic" derivatives? . . . . . . . . . . . . . . . . . . . . . . 6

Q1.9 How are derivatives di�erent from badla? . . . . . . . . . . . . . . . . 6

Q1.10 Why are derivatives useful? . . . . . . . . . . . . . . . . . . . . . . . 6

Q1.11 What are the instruments traded in the derivatives industry, and what

are their relative sizes? . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Q1.12 Worldwide, what kinds of derivatives are seen on the equity market? 7

Q1.13 At the security level, are futures or options better? . . . . . . . . . . 7

Q1.14 Why have index derivatives proved to be more important than securityderivatives? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7

Q1.15 Who uses index derivatives to reduce risk? . . . . . . . . . . . . . . 8

Q1.16 How will retail investors bene�t from index derivatives? . . . . . . . 9

Q1.17 What derivatives exist in India (today) in the interest-rates area? . . 9

Q1.18 What derivatives exist in India (today) in the foreign exchange area? 9

Q1.19 What is the status in India in the area of commodity derivatives? . . 9

Q1.20 What is the present status of derivatives in the equity market? . . . 10

Q1.21 Why do people talk about \starting derivatives in India" if some

derivatives already exist? . . . . . . . . . . . . . . . . . . . . . . . . . 10

Q1.22 What should the time to expiration of these contracts be? . . . . . . 10

2 Market Microstructure 11

Q2.1 How do derivatives trade? . . . . . . . . . . . . . . . . . . . . . . . . . 11

Q2.2 If a contract is just a relationship between long and short, how do we

ensure \contract performance"? . . . . . . . . . . . . . . . . . . . . . . 11

Q2.3 What is the role of arbitrage in the derivatives area? . . . . . . . . . . 11

Q2.4 What happens if there are only a few arbitrageurs ready to function in

the early days of the market? . . . . . . . . . . . . . . . . . . . . . . . 11

Q2.5 Isn't India's cash market much too ine�cient to support concepts like

derivatives? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

1

2 CONTENTS

Q2.6 What is the role of liquidity in enabling good derivatives markets? . . 12

Q2.7 What should a market index be? . . . . . . . . . . . . . . . . . . . . . 12

Q2.8 How does liquidity matter for market indexes? . . . . . . . . . . . . . 12

Q2.9 What is special about Nifty for use in index derivatives? . . . . . . . . 13

Q2.10 What is the impact cost seen in trading Nifty? . . . . . . . . . . . . 13

Q2.11 How does this low impact cost matter? . . . . . . . . . . . . . . . . . 13

Q2.12 Is the liquidity in India adequate to support well-functioning deriva-

tives markets? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

Q2.13 What kind of liquidity is expected on index derivatives markets? . . 13

Q2.14 How does spot-futures arbitrage a�ect the cash market? . . . . . . . 14

Q2.15 Going beyond spot-futures arbitrage, how do derivatives in uence liq-

uidity on the underlying market? . . . . . . . . . . . . . . . . . . . . . 14

Q2.16 What is the international experience in terms of how the underlying

market is changed once derivatives start trading? . . . . . . . . . . . . 14

Q2.17 Program trading in the US is often accused of generating di�culties.

What does that mean for us? . . . . . . . . . . . . . . . . . . . . . . . 14

Q2.18 Will derivatives destabilise the stock market? Could this happen in

extreme events? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

Q2.19 Is there more or less of a \natural monopoly" in derivatives trading,as compared with the spot market? . . . . . . . . . . . . . . . . . . . . 15

Q2.20 What are the policy implications of this lack of a natural monopoly? 15

Q2.21 At the operational level, how do security contracts compare versus

index-based contracts? . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

3 Derivatives Disasters 17

Q3.1 Why do we keep reading about disasters involving derivatives? . . . . 17

Q3.2 Why have we seen more disasters in the recent years? . . . . . . . . . 17

Q3.3 How much money has been lost in these derivatives disasters? . . . . 17

Q3.4 What happened in Barings? . . . . . . . . . . . . . . . . . . . . . . . 17

Q3.5 What should be done to minimise disasters with derivatives? . . . . . 17

4 Policy Issues 19

Q4.1 What emerging markets have already created derivatives markets? . . 19

Q4.2 What was China's experience in this area? . . . . . . . . . . . . . . . 19

Q4.3 What �nancial markets in India are ready for derivatives today? . . . 19

Q4.4 Are derivatives in interest rates viable in India? . . . . . . . . . . . . 19

Q4.5 Why are commodity futures markets important? . . . . . . . . . . . . 19

Q4.6 What are the issues in the creation of commodity derivatives markets? 20

Q4.7 What can be done in derivatives on real estate? . . . . . . . . . . . . 21

Q4.8 Should foreigners be restricted in India's derivatives markets as a mat-

ter of policy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

Q4.9 What would access to derivatives do to FII and FDI investment? . . . 21

Q4.10 Is India ready for derivatives today? . . . . . . . . . . . . . . . . . . 21

Q4.11 What are the costs and bene�ts of delaying the onset of exchange-

traded �nancial derivatives in India? . . . . . . . . . . . . . . . . . . . 22

Q4.12 What international derivatives exchanges are working towards launch-

ing products o� underlyings in emerging markets? . . . . . . . . . . . 23

CONTENTS 3

Q4.13 What derivatives on Indian underlyings are currently trading abroad? 23

Q4.14 Would foreign derivatives markets be interested in launching trading

on Indian underlyings? . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Q4.15 Is this a real threat? . . . . . . . . . . . . . . . . . . . . . . . . . . . 23

Q4.16 What are the implications of derivatives on Indian underlyings trading

abroad? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

Q4.17 But in the GDR market experience, the foreign market has hardly

a�ected order ow into India { why are derivatives di�erent? . . . . . 24

5 Regulatory Issues 25

Q5.1 What are the objectives of regulation? . . . . . . . . . . . . . . . . . . 25

Q5.2 What kinds of competition are possible in the �nancial market scenario? 25

Q5.3 How does deviation from perfect competition (or situations of market

power) arise on �nancial markets? . . . . . . . . . . . . . . . . . . . . 25

Q5.4 How should regulation of exchanges work? . . . . . . . . . . . . . . . 25

Q5.5 What can regulation do to encourage competitiveness? . . . . . . . . 25

Q5.6 What should entry or eligibility requirements be for derivatives trading? 26

Q5.7 What is fraud? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

Q5.8 How can regulation diminish the extent of fraud? . . . . . . . . . . . . 26

Q5.9 Should there be regulatory control over contract de�nition? . . . . . . 26

Q5.10 Should the securities that are used in an index be required by regula-tion to be in depository mode? . . . . . . . . . . . . . . . . . . . . . . 27

Q5.11 Should regulation require that derivatives trading be organised at an

exchange which is distinct from the spot market? . . . . . . . . . . . . 27

Q5.12 At an operational level, is it better to have the spot and futures market

under one roof? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

5.1 Risk Management at Clearing . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Q5.13 Why is the clearinghouse central to derivatives markets? . . . . . . . 28

Q5.14 How should margining for derivatives work? . . . . . . . . . . . . . . 28

Q5.15 How is initial margin (i.e. \exposure limit") calculated? . . . . . . . 28

Q5.16 How is daily mark to market (MTM) margin calculated? . . . . . . . 29

Q5.17 How does the slow payments system change these calculations? . . . 29

Q5.18 What are prospects for improvements of the banking system? . . . . 29

Q5.19 How does options margining work? . . . . . . . . . . . . . . . . . . . 30

Q5.20 What are the special di�culties of margining options? . . . . . . . . 30

Q5.21 What constraints should regulation impose upon the time to expira-

tion of these contracts? . . . . . . . . . . . . . . . . . . . . . . . . . . 30

Q5.22 How does the margining system change the way people trade? . . . . 30

Q5.23 What are the policy issues in clearing corporation failure? . . . . . . 30

5.2 Other Aspects of the Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Q5.24 What are the risks that derive from usage of derivatives? . . . . . . . 31

Q5.25 What are the policy issues in individual risk? . . . . . . . . . . . . . 31

Q5.26 What about systemic risk, or risks to the economy? . . . . . . . . . . 32

Q5.27 How should the use of derivatives by mutual funds be regulated? . . 32

5.3 Market manipulation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

Q5.28 What are the issues in terms of manipulation in the context of deriva-

tives markets in general? . . . . . . . . . . . . . . . . . . . . . . . . . . 32

4 CONTENTS

Q5.29 What are the issues on short squeezes? . . . . . . . . . . . . . . . . . 32

Q5.30 What kinds of manipulation are found with index derivatives? . . . . 32

Q5.31 How do manipulators attack an index? . . . . . . . . . . . . . . . . . 33

Q5.32 What kinds of manipulation can take place on security options? . . . 33

Q5.33 Are individual securities in India liquid enough to support securityoptions? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Q5.34 What would concerns about manipulation imply for the sequencing of

index derivatives vs. security options? . . . . . . . . . . . . . . . . . . 34

Q5.35 Would a slow launch of security options harm the economy? . . . . . 34

Q5.36 What spot market should supply prices which are used for calculating

payo�s with cash{settled security options? . . . . . . . . . . . . . . . 34

Q5.37 To what extent are these issues a regulatory issue? . . . . . . . . . . 34

5.4 In Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35

Q5.38 What does this discussion translate to in terms of speci�c regulations

in the derivatives area? . . . . . . . . . . . . . . . . . . . . . . . . . . 35

6 Derivatives and the Economy 37

Q6.1 What are the bene�ts of derivatives to India? . . . . . . . . . . . . . . 37

Q6.2 How do index derivatives change the overall level of equity investment

in the economy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Q6.3 How will index derivatives assist capital formation and growth in the

economy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

Q6.4 What is the evidence about derivatives and market liquidity and e�-

ciency? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

Q6.5 What is the evidence about market quality and economic growth? . . 38

Q6.6 How do derivatives alter the exposure of di�erent people in the economy? 38

Q6.7 How do derivatives alter the informational structure of the economy? 38

Chapter 1

Basics

Q1.1: What are derivatives?A: Derivatives, such as options or futures, are �-nancial contracts which derive their value o� aspot price time-series, which is called \the under-lying". For examples, wheat farmers may wishto contract to sell their harvest at a future dateto eliminate the risk of a change in prices by thatdate. Such a transaction would take place througha forward or futures market. This market is the\derivative market", and the prices on this mar-ket would be driven by the spot market price ofwheat which is the \underlying". The terms \con-tracts" or \products" are often applied to denotethe speci�c traded instrument.

The world over, derivatives are a key part of the�nancial system. The most important contract-types are futures and options, and the most im-portant underlying markets are equity, treasurybills, commodities, foreign exchange and real es-tate. ��

Q1.2: What is a forward contract?A: In a forward contract, two parties agree to doa trade at some future date, at a stated price andquantity. No money changes hands at the timethe deal is signed. ��

Q1.3: Why is forward contracting useful?A: Forward contracting is very valuable in hedgingand speculation.

The classic hedging application would be thatof a wheat farmer forward-selling his harvest at aknown price in order to eliminate price risk. Con-versely, a bread factory may want to buy breadforward in order to assist production planningwithout the risk of price uctuations.

If a speculator has information or analysiswhich forecasts an upturn in a price, then she can

go long on the forward market instead of the cashmarket. The speculator would go long on the for-ward, wait for the price to rise, and then take areversing transaction. The use of forward marketshere supplies leverage to the speculator. ��

Q1.4: What are the problems of forward mar-kets?A: Forward markets worldwide are a�icted by sev-eral problems: (a) lack of centralisation of trading,(b) illiquidity, and (c) counterparty risk.

In the �rst two of these, the basic problem isthat of too much exibility and generality. Theforward market is like the real estate market inthat any two consenting adults can form contractsagainst each other. This often makes them de-sign terms of the deal which are very convenientin that speci�c situation, but makes the contractsnon-tradeable. Also the \phone market" here isunlike the centralisation of price discovery that isobtained on an exchange.

Counterparty risk in forward markets is a sim-ple idea: when one of the two sides of the trans-action chooses to declare bankruptcy, the othersu�ers. Forward markets have one basic property:the larger the time period over which the forwardcontract is open, the larger are the potential pricemovements, and hence the larger is the counter-party risk.

Even when forward markets trade standardisedcontracts, and hence avoid the problem of illiq-uidity, the counterparty risk remains a very realproblem. A classic example of this was the famousfailure on the Tin forward market at LME. ��

Q1.5: What is a futures contract?A: Futures markets were designed to solve all thethree problems (a, b and c listed in Question 1.4)of forward markets. Futures markets are exactly

5

6 CHAPTER 1. BASICS

like forward markets in terms of basic economics.However, contracts are standardised and trad-

ing is centralised, so that futures markets arehighly liquid. There is no counterparty risk(thanks to the institution of a clearinghouse whichbecomes counterparty to both sides of each trans-action and guarantees the trade). In futures mar-kets, unlike in forward markets, increasing thetime to expiration does not increase the counter-party risk. ��

Also see: ?.

Q1.6: Why is the cash market in India saidto have futures-style settlement?A: In a true cash market, when a trade takes placetoday, delivery and payment would also take placetoday (or a short time later). Settlement proce-dures like T+3 would qualify as \cash markets"in this sense, and of the equity markets in thecountry, only OTCEI is a cash market by this def-inition.For the rest, markets like the BSE or the NSE

are classic futures market in operation. NSE's eq-uity market, for example, is a weekly futures mar-ket with tuesday expiration. When a person goeslong on thursday, he is not obligated to do deliv-ery and payment right away, and this long positioncan be reversed on friday thus leaving no net obli-gations with the clearinghouse (this would not bepossible in a T+3 market). Like all futures mar-kets, trading at the NSE is centralised, the futuresmarkets are quite liquid, and there is no counter-party risk. ��

Q1.7: What is an option?A: An option is the right, but not the obligation,to buy or sell something at a stated date at astated price. A \call option" gives one the rightto buy, a \put option" gives one the right to sell.Options come in two varieties { european vs.

american. In a european option, the holder of theoption can only exercise his right (if he should sodesire) on the expiration date. In an americanoption, he can exercise this right anytime betweenpurchase date and the expiration date. ��

Also see: ?, ?.

Q1.8: What are \exotic" derivatives?A: Options and futures are the mainstreamworkhorses of derivatives markets worldwide.

However, more complex contracts, often called ex-otics, are used in more custom situations. For ex-ample, a computer hardware company may wanta contract that pays them when the rupee has de-preciated or when computer memory chip priceshave risen. Such contracts are \custom-built"for a client by a large �nancial house in whatis known as the \over the counter" derivativesmarket. These contracts are not exchange-traded.This area is also called the \OTC Derivatives In-dustry".

An essential feature of derivatives exchanges iscontract standardisation. All kinds of wheat arenot tradeable through a futures market, only cer-tain de�ned grades are. This is a constraint for afarmer who grows a somewhat di�erent grade ofwheat. The OTC derivatives industry is an inter-mediary which sells the farmer insurance which iscustomised to his needs; the intermediary wouldin turn use exchange-traded derivatives to strip o�as much of his risk as possible. ��

Q1.9: How are derivatives di�erent frombadla?A: Badla is closer to being a facility for borrowingand lending of shares and funds. Borrowing andlending of shares is a functionality which is partof the cash market. The borrower of shares pays afee for the borrowing. When badla works withouta strong marginning system, it generates counter-party risk, the evidence of which is the numerouspayments crises which were seen in India.

Options are obviously not at all like badla. Fu-tures, in contrast, may seem to be like badla tosome. Some of the key di�erences may be sum-marised here. Futures markets avoid variabilityof badla �nancing charges. Futures markets tradedistinctly from the cash market so that each fu-tures prices and cash prices are di�erent things (incontrast with badla, where the cash market and allfutures prices are mixed up in one price). Futuresmarkets lack counterparty risk through the insti-tution of the clearinghouse which guarantees thetrade coupled with marginning, and this elimina-tion of risk eliminates the \risk premium" that isembedded inside badla �nancing charges, thus re-ducing the �nancing cost implicit inside a futuresprice. ��

Q1.10: Why are derivatives useful?A: The key motivation for such instruments isthat they are useful in reallocating risk either

7

Badla Futures

� Expiration date unclear � Expiration date known� Spot market and di�erent expiration dates are

mixed up� Spot market and di�erent expiration dates alltrade distinct from each other.

� Identity of counterparty often known � Clearing corpn. is counterparty� Counterparty risk present � No counterparty risk

� Badla �nancing is additional source of risk � No additional risk.� Badla �nancing contains default-risk premia � Financing cost at close to riskless thanks to

counterparty guarantee� Asymmetry between long and short � Long and short are symmetric� Position can breakdown if borrowing/lending

proves infeasible� You can hold till expiration date for sure, if youwant to

Table 1.1: A Comparison of Futures and Badla

across time or among individuals with di�erentrisk-bearing preferences.

One kind of passing-on of risk is mutual insur-ance between two parties who face the oppositekind of risk. For example, in the context of cur-rency uctuations, exporters face losses if the ru-pee appreciates and importers face losses if therupee depreciates. By forward contracting in thedollar-rupee forward market, they supply insur-ance to each other and reduce risk. This sort ofthing also takes place in speculative position tak-ing { the person who thinks the price will go upis long a futures and the person who thinks theprice will go down is short the futures.

Another style of functioning works by a risk-averse person buying insurance, and a risk-tolerant person selling insurance. An example ofthis may be found on the options market : an in-vestor who tries to protect himself against a dropin the index buys put options on the index, and arisk-taker sells him these options. Obviously, peo-ple would be very suspicious about entering intosuch trades without the institution of the clearing-house which is a legal counterparty to both sidesof the trade.

In these ways, derivatives supply a method forpeople to do hedging and reduce their risks. Ascompared with an economy lacking these facilities,it is a considerable gain.

The ultimate importance of a derivatives mar-ket hence hinges upon the extent to which it helpsinvestors to reduce the risks that they face. Someof the largest derivatives markets in the world areon treasury bills (to help control interest rate risk),the market index (to help control risk that is asso-ciated with uctuations in the stock market) andon exchange rates (to cope with currency risk).

Derivatives are also very convenient in terms ofinternational investment. For example, Japaneseinsurance companies fund housing loans in the USby buying into derivatives on real estate in the US.Such funding patterns would be harder withoutderivatives. ��

Q1.11: What are the instruments traded inthe derivatives industry, and what are theirrelative sizes?A: This information is summarised in Tables 1.2and 1.3 which are drawn from ?. ��

Q1.12: Worldwide, what kinds of derivativesare seen on the equity market?A: Worldwide, the most successful equity deriva-tives contracts are index futures, followed by indexoptions, followed by security options. ��

Q1.13: At the security level, are futures oroptions better?A: The international experience is that at the se-curity level, options markets are almost alwaysmore successful than futures markets. ��

Q1.14: Why have index derivatives proved tobe more important than security derivatives?A: Security options are of limited interest becausethe pool of people who would be interested (say) inoptions on ACC is limited. In contrast, every sin-

gle person in the �nancial area is a�ected by index uctuations. Hence risk-management using indexderivatives is of far more importance than risk-management using individual security options.

This goes back to a basic principle of �nan-

8 CHAPTER 1. BASICS

1986 1990 1993 1994

Exchange Traded 583 2292 7839 8838Interest rate futures 370 1454 4960 5757Interest rate options 146 600 2362 2623Currency futures 10 16 30 33Currency options 39 56 81 55Stock Index futures 15 70 119 128Stock Index options 3 96 286 242

Some of the OTC Industry 500 3450 7777 11200Interest rate swaps 400 2312 6177 8815Currency swaps 100 578 900 915Caps, collars, oors, swaptions - 561 700 1470

Total 1083 5742 16616 20038

Table 1.2: The Global Derivatives Industry (Outstanding Contracts, $ billion)

cial economics. Portfolio risk is dominated by themarket index, regardless of the composition of theportfolio. In other words, all portfolios of aroundten stocks or more have a pattern of risk where80% or more of their volatility is index-related. Insuch a world, investors would be more interestedin using index{based derivative products ratherthan security-based derivative products. The ac-tual experience of derivatives markets worldwideis completely in line with this expectation. ��

Also see: ?.

Q1.15: Who uses index derivatives to reducerisk?A: There are two important types of people whomay not want to \bear the risk" of index uctua-tions:

� A person who thinks Index uctuations are

peripheral to his activity

For example, a person who works in primarymarket underwriting e�ectively has index ex-posure { if the index does badly, then theIPO could fail { but this exposure has noth-ing to do with his core competence and in-terests (which are in the IPO market). Sucha person would routinely use measure his in-dex exposure on a day-to-day basis, and in-dex derivatives to strip o� that risk. If full- edged bookbuilding becomes important inIndia, then there is a very important role forindex derivatives in the \price stabilisation"

that the underwriter does in the bookbuild-ing process (see ? for an exposition aboutbookbuilding).

Similarly, a person who takes positions in in-dividual stocks implicitly su�ers index expo-sure. A person who is long ITC is e�ectivelylong ITC and long Index. If the index doesbadly, then his \long ITC" position su�ers.A person like this, who is focussed on ITCand is not interested in taking a view on theIndex would routinely measure the index ex-posure that is hidden inside his ITC expo-sure, and use index derivatives to eliminatethis risk. The NYSE specialist is a prime ex-ample of intensive use of index derivatives insuch an application.

� A person who thinks Index uctuations are

painful

An investor who buys stocks may like thepeace of mind of capping his downside loss.Put options on the index are the ideal formof insurance here. Regardless of the composi-tion of a person's portfolio, index put optionswill protect him from exposure to a fall in theindex. To make this concrete, consider a per-son who has a portfolio worth Rs.1 million,and suppose Nifty is at 1000. Suppose theperson decides that he wants to never su�era loss of worse than 10%. Then he can buyhimself Nifty puts worth Rs.1 million withthe strike price set to 900. If Nifty drops be-low 900 then his put options reimburse himfor his full loss. In this fashion, \portfolio in-

9

1874 Commodity Futures1972 Foreign currency futures1973 Equity options1975 T-bond futures1981 Currency swaps1982 Interest rate swaps; T-note futures; Eurodollar futures; Equity index

futures; Options on T-bond futures; Exchange{listed currency options1983 Options on equity index; Options on T-note futures; Options on currency

futures; Options on equity index futures; Interest rates caps and oors1985 Eurodollar options; Swaptions1987 OTC compound options; OTC average options1989 Futures on interest rate swaps; Quanto options1990 Equity index swaps1991 Di�erential swaps1993 Captions; Exchange-listed FLEX options1994 Credit default options

Table 1.3: The Global Derivatives Industry: Chronology of Instruments

surance" through index options will greatlyreduce the fear of equity investment in thecountry.

More generally, anytime an investor or a fundmanager becomes uncomfortable, and doesnot want to bear index uctuations in thecoming weeks, he can use index futures orindex options to reduce (or even eliminate)his index exposure. This is far more con-venient than distress selling of the underly-ing equity in the portfolio. Conversely, any-time investors or fund managers become op-timistic about the index, or feel more com-fortable and are willing to bear index uctu-ations, they can increase their equity expo-sure using index derivatives. This is simplerand cheaper than buying underlying equity.In these ways, the underlying equity portfo-lio can be something that is \slowly traded",and index derivatives are used to implementday-to-day changes in equity exposure.

��

Q1.16: How will retail investors bene�t fromindex derivatives?A: The answer to this �ts under \People who �ndIndex uctuations painful" category in Question1.15. Every retail investor in the economy who isin pain owing to a downturn in the market indexis potentially a happy user of index derivatives.One key requirement from the viewpoint of the

retail user is contract size. If the minimum in-

vestible lot on the index derivatives market is Rs.1million or so, then it will not be useful for retailinvestors. ��

Q1.17: What derivatives exist in India (to-day) in the interest-rates area?A: There are no derivatives based on interest ratesin India today. ��

Q1.18: What derivatives exist in India (to-day) in the foreign exchange area?A: India has a strong dollar-rupee forward marketwith contracts being traded for one, two, .. sixmonth expiration. Daily trading volume on thisforward market is around $500 million a day. In-dian users of hedging services are also allowed tobuy derivatives involving other currencies on for-eign markets. ��

Q1.19: What is the status in India in the areaof commodity derivatives?A: Futures markets exist on six commodities (cas-tor seed, hessian, gur, potatoes, turmeric and pep-per). The pepper exchange, which is at Cochin, isbeing upgraded to the status of an \internationalpepper futures market", which will accept ordersfrom all over the world. The Forward MarketsCommission (FMC) oversees these markets.A high level of interest exists on futures mar-

kets for other commodities. In September 1994,the Kabra Committee recommended that fu-tures trading should additionally be permitted

10 CHAPTER 1. BASICS

in 17 commodities. These are (a) basmati rice,(b) cotton, (c) kapas, (d) raw jute and jutegoods, (e) groundnut, its oil and cake, (f) rape-seed/mustardseed, its oil and cake, (g) cottonseed, its oil and cake, (h) sesame seed, its oil andcake (i) sun ower, its oil and cake, (j) sa�ower, itsoil and cake, (k) copra, coconut oil and its oilcake,(l) soyabean, its oil and cake, (m) ricebran oil, (n)castor oil and its oilcake, (o) linseed, (p) silver,and (q) onions. On 4 December 1996, the Cof-fee Board decided to recommend that a domesticfutures market for co�ee should be setup.On 28 February 1997, the �nance minister an-

nounced that futures markets would be setup incotton and jute, and that an international futuresmarket would be created in castorseed and castoroil. ��

Q1.20: What is the present status of deriva-tives in the equity market?A: As mentioned in Question 1.6, trading on the\spot market" for equity has actually always beena futures market with weekly or fortnightly settle-ment (this is true of every market in the countryother than OTCEI). These futures markets fea-ture the risks and di�culties of futures markets,without the gains in price discovery and hedgingservices that come with a separation of the spotmarket from the futures market.India's primary market has experience with

derivatives of two kinds: convertible bonds andwarrants (a slight variant of call options). Sincethese warrants are listed and traded, options mar-kets of a limited sort already exist. However, thetrading on these instruments is very limited. Therecent ICICI bond issue bundles a twelve-year ex-piration BSE Sensex warrant with the bond. Ifthis warrant is detached and traded, it would bean exchange-traded index derivative.A variety of interesting derivatives markets ex-

ist in the informal sector. These markets tradecontracts like bhav-bhav, teji-mandi, etc. For ex-ample, the bhav-bhav is a bundle of one in-the-money call option and one in-the-money put op-tion. These informal markets stand outside themainstream institutions of India's �nancial systemand enjoy limited participation. ��

Q1.21: Why do people talk about \startingderivatives in India" if some derivatives al-ready exist?A: It is useful to note here that there are

no exchange-traded �nancial derivatives in In-dia today. Neither the dollar-rupee forward con-tract (Question 1.18) nor the option-like contracts(Question 1.20) are exchange-traded. These mar-kets hence lack centralisation of price discoveryand can su�er from counterparty risk. The nextstep in these areas is institutionalisation, and abroad-basing of access. ��

Q1.22: What should the time to expirationof these contracts be?A: The time to expiration of these contracts\should" be whatever the market wants it to be{ if four-year contracts attract high trading vol-ume, then four-year contracts should exist. Theinternational experience is that most of the trad-ing volume in index futures is concentrated in con-tracts which expire one, two, three and four quar-ters away. Limited interest is seen in contractswhich go upto two and three years out.There is a widespread intuition in India, shaped

by decades of experience with clearinghouses thatdo not guarantee trades, that longer time to expi-ration is associated with higher counterparty risk.However, when daily mark-to-market margins areapplied, the link between length of contract life and

counterparty risk is broken. A brand-new positiontoday is no di�erent from an old position (regard-less of the history) as long as the person has paidup his loss in full as of today. This is exactly whatthe mark-to-market margin does. ��

Chapter 2

Market Microstructure

Q2.1: How do derivatives trade?

A: In the cash market, the basic dynamic is thatthe issuer puts out paper, and people trade thispaper. In contrast, in derivatives, there is no is-suer. The net supply of all derivatives contractsis 0. For each long, there is an equal and oppositeshort. A contract is born when a long and a shortmeet on the market.

There would be a clear \contract cycle" whichthe exchange de�nes. For example, using quar-terly contracts, we would have something like this:On Jan 1, four contracts start trading. The near-est contract expires on 31 Mar. On 31 Mar, this�rst contract ceases to exist, and the next (30June) contract starts trading.

In the case of options, the exchange additionallyde�nes the strike prices of the options which areallowed to trade. ��

Q2.2: If a contract is just a relationshipbetween long and short, how do we ensure\contract performance"?

A: The key innovation of derivatives markets isthe notion of the clearinghouse that guaranteesthe trade. Here, when A buys from B, (at a legallevel) the clearinghouse buys from B and sells toA. This way, if either A or B fail on their obliga-tions, the clearinghouse �lls in the gap and ensuresthat payments go through without a hitch.

The clearinghouse, in turn, cannot create sucha guarantee out of thin air. It uses a system ofinitial margin and daily mark-to-market margins,coupled with sophisticated risk containment, toensure that it is not bankrupted in the process ofsupplying this guarantee. ��

Q2.3: What is the role of arbitrage in thederivatives area?A: All pricing of derivatives is done by arbitrage,and by arbitrage alone.In other words, basic economics dictates a re-

lationship between the price of the spot and theprice of a futures. If this relationship is violated,then an arbitrage opportunity is available, andwhen people exploit this opportunity, the price re-verts back to its economic value.In this sense, arbitrage is basic to pricing of

derivatives. Without arbitrage, there would be nomarket e�ciency in the derivatives market: priceswould stray away from fair value all the time. In-deed, a basic fact about derivatives is that the

market e�ciency of the derivatives market is in-

versely proportional to the transactions costs faced

by arbitrageurs in that market. When arbitrage is uent and e�ective, market e�ciency is obtained,which improves the attractiveness of the deriva-tives from the viewpoint of users such as hedgersor speculators. ��

Q2.4: What happens if there are only a fewarbitrageurs ready to function in the earlydays of the market?A: In most countries, there are bigger arbitrageopportunities in the early days of the futures mar-ket. As larger resources and greater skills getbrought into the arbitrage business, these oppor-tunities tend to vanish.India is better placed in terms of skills in ar-

bitrage, as compared with many other countries,thanks to years of experience with \line opera-tors" who are used to doing arbitrage betweenexchanges. These kinds of traders would be eas-ily able to redirect their skills into this new mar-ket. These \line operators" are uent with a host

11

12 CHAPTER 2. MARKET MICROSTRUCTURE

of real-world di�culties, such as di�erent expira-tion dates on di�erent exchanges, bad paper, etc.Their skills are well-suited to index arbitrage. ��

Also see: ?, ?, ?, ?.

Q2.5: Isn't India's cash market much too in-e�cient to support concepts like derivatives?A: There is no evidence to suggest that marketine�ciencies on the cash market make it di�-cult to sustain derivatives markets. Many emerg-ing markets that have derivatives markets aremore primitive than India on the key determi-nants of market e�ciency, i.e. (a) high informa-tion availability, (b) high skills in keeping accountsand reading accounting reports, (c) high popula-tion of speculative traders and (d) low transac-tions costs. Derivatives markets are successful ifpeople face risks that they wish to hedge them-selves against; market ine�ciency on the underly-ing market probably serves to increase the demandfor these hedging services.

? is a Ph.D. thesis which is devoted to an ex-amination of BSE returns data from 1990 to 1995.This evidence supports the notion that the mar-kets are quite informationally e�cient, given the(high) level of transactions costs that has prevailedin the past. One widely prevalant practise thatserves to interlink market prices and corporatenews is insider trading. Insider trading is unfairand detrimental to market liquidity in a subtlefashion, but it does serve to rapidly bring marketprices in line with corporate information.

This research is carried further in ? which ex-amines the impact of automation and competitionupon the functioning of the BSE. Here the evi-dence suggests that transactions costs have comedown with automation, and exactly as predictedby economic theory, market e�ciency has im-proved as a consequence.

One interesting piece of work in this area is?, where the publication of a research study wasfollowed by a swift elimination of the market in-e�ciency which this research study documented.This is an example of how market e�ciency any-where in the world works: pro�t-maximising spec-ulators detect mispricings on the market, andwhen they trade in exploiting these mispricings,the ine�ciency goes away.

The �nal litmus test of market e�ciency is mu-tual fund performance. If India's markets wereine�cient, it would be possible for professional

fund managers to obtain excess returns throughinformed trading. The available evidence (?) sug-gests that three-quarters of Indian funds under-perform the index, after adjusting for the level ofsystematic risk adopted. This fraction is almostexactly the same as that seen in the US. Thismakes it di�cult to support the hypothesis thatIndia's markets are much less e�cient than thoseseen in OECD countries, after controlling for the

levels of transactions costs. ��

Also see: ?, ?, ?, ?.

Q2.6: What is the role of liquidity in enablinggood derivatives markets?A: The role of liquidity (which is de�ned as lowtransactions costs) is in making arbitrage cheapand convenient. If transactions costs are low, thenthe smallest mispricings on the derivatives marketwill be removed by arbitrageurs, which will makethe derivatives market more e�cient. ��

Q2.7: What should a market index be?A: A market index is a large, well-diversi�ed port-folio which is an approximation to returns ob-tained in owning \the overall economy". Portfoliodiversi�cation is a powerful means of stripping out�rm- and industry-e�ects, so that the returns onthe well-diversi�ed portfolio re ect only economy-wide e�ects, and are relatively insensitive to thespeci�c companies or industries in the index port-folio. Market index returns time-series are cen-tral to modern �nancial economics, and have enor-mous value for a variety of real-world applications.A good market index should be highly liquid tosupport products in the real world, it should havea high hedging e�ectiveness against a huge varietyof real-world portfolios, and it should be hard tomanipulate. ��

Q2.8: How does liquidity matter for marketindexes?A: At one level a market index is used as a pureeconomic time-series. Liquidity a�ects this appli-cation via the problem of non-trading. If somesecurities in an index fail to trade today, then thelevel of the market index obtained re ects the val-uation of the macroeconomy today (via securitieswhich traded today), but is contaminated with thevaluation of the macroeconomy yesterday (via se-curities which traded yesterday). This is the prob-lem of stale prices. By this reasoning, securities

13

with a high trading intensity are best-suited forinclusion into a market index.As we go closer to applications of market in-

dexes in the indexation industry (such as indexfunds, or sector-level active management, or indexderivatives), the market index is not just an eco-nomic time-series, but a portfolio which is traded.The key di�culty faced here is again liquidity, orthe transactions costs faced in buying or sellingthe entire index as a portfolio. ��

Q2.9: What is special about Nifty for use inindex derivatives?A: The methodology created for the NSE-50 indexexplicitly isolates a set of securities for which themarket impact cost is minimised when buying orselling the entire index portfolio. This makes Niftywell-suited to applications such as index funds, in-dex derivatives, etc. Nifty has a explicit method-ology for regular maintenance of the index set. Itis successful at expressing the market risk inherentin a wide variety of portfolios in the country. ��

Also see: ?.

Q2.10: What is the impact cost seen in trad-ing Nifty?A: In calendar 1996, on average, the impact costfaced in buying Rs.5 million of the Nifty portfoliowas 0.25% or so. This means that if the indexlevel is 1000, then a buy order of Rs.5 million isexecuted at 1002.5 and a sell order is executed at997.5. This is the lowest level of transactions costsseen in market indexes in India.An example of the impact cost analysis of Nifty

is shown in Figure 2.1, which uses data for 5 June1996, and shows how the impact cost in trades onNifty varies as the transaction size is increased.

��

Q2.11: How does this low impact cost mat-ter?A: As is the case in all areas of �nance, in the con-text of index derivatives, there is a direct mappingbetween transactions costs and market e�ciency.Index futures and options based on Nifty will ben-e�t from a high degree of market e�ciency becausearbitrageurs will face low transactions costs whenthey eliminate mispricings. This high degree ofmarket e�ciency on the index derivatives marketwill make it more attractive to pure users of thederivatives, such as hedgers, speculators and in-

0 5 10 15 20 25

Rs. Million

0

20

40

60

Bas

is P

oint

s

12 Noon1 PM2 PM

05 June ’96

Figure 2.1: Impact cost for Nifty for Various

Transaction Sizes

vestors. High liquidity also immediately impliesthat the index is hard to manipulate, which helpsengender public con�dence. ��

Q2.12: Is the liquidity in India adequate tosupport well-functioning derivatives markets?A: The one-way market impact cost faced by ar-bitrageurs working the NSE-50 is around 0.25%.This is similar to that seen by arbitrageurs work-ing the S&P 500. This suggests that market liq-

uidity by itself will not be a serious constraint inthe face of an index derivatives market in India.

It should be noted that market impact cost isnot the only component of transactions costs thatarbitrageurs face. It is true that post-trade costsare higher in India (thanks to the small role thatthe book-entry trading plays (as of today)). How-ever, market liquidity is not a constraint in index-based products based on Nifty. ��

Q2.13: What kind of liquidity is expected onindex derivatives markets?A: Impact cost on index derivatives markets islikely to be much smaller than that seen on thespot index. One thumb rule which is commonlyused internationally is that the round{trip cost(i.e. twice the impact cost plus brokerage) oftrades on index futures of around $0.5 million arearound 0.01%, i.e. the index futures are around

14 CHAPTER 2. MARKET MICROSTRUCTURE

20 times more liquid than the spot index.1 Highliquidity is the essential appeal of index deriva-tives. If trading on the spot market were cheap,then many portfolio modi�cations would get donethere itself. However, because transactions costson the cash market are high, using derivatives isan appealing alternative. ��

Q2.14: How does spot-futures arbitrage af-fect the cash market?A: Spot-futures arbitrage increases the ow ofmarket orders to the cash market. This increasesthe revenues obtained by day traders who placelimit orders, and induces an increased supply oflimit orders. Limit orders are the ultimate sourceof liquidity on the market (indeed, low impact costis synonymous with a thick limit order book whichis highly populated with limit orders). Hence theintroduction of spot-futures arbitrage will improvethe liquidity on the cash market. ��

Q2.15: Going beyond spot-futures arbitrage,how do derivatives in uence liquidity on theunderlying market?A: There are also less direct channels of in uencefrom derivatives to enhanced liquidity on the un-derlying market.

Day traders in individual stocks, who supplyliquidity in these stocks, will be able to use indexfutures to o�set their index exposure, and hencebe able to function at lower levels of risk. Forexample, the NYSE specialistmakes phone calls toChicago almost every half an hour (while tradingis going on) adjusting his index futures position asa function of his inventory. Everytime a day traderis long security he will simultaneously be shortindex futures (to strip out his index exposure),and vice versa.

Another aspect is rooted in security optionsmarkets. When security options markets exist,speculators on individual securities tend to gotrade on the options market, and the focus of pricediscovery moves away from the cash market to theoptions market. More informed traders tend tocluster on the options market, and less informedorders tend to go to the cash market. This reducesthe risk of trading against an informed speculator

1For example, in the US, the S&P 500 futures con-tract (on the CME) has spreads of around $100 on aminimum tradeable lot of around $400,000; i.e., theone{way impact cost is around 0.0125%.

on the cash market. This reduces impact cost (i.e.increases liquidity) on the cash market. ��

Q2.16: What is the international experiencein terms of how the underlying market ischanged once derivatives start trading?A: The international experience is that marketquality on the underlying market improves oncederivatives come to exist. Liquidity and markete�ciency of the underlying market are increasedonce derivatives come to exist. ��

Q2.17: Program trading in the US is oftenaccused of generating di�culties. What doesthat mean for us?A: Many post-mortems of the October 1987 crashconcluded that \program trading was related tothe crash". Some observers distorted this to \pro-gram trading caused the crash".A more accurate depiction of the sequence of

events in October 1987 may be expressed as fol-lows:

� A market drop commenced on overseas mar-kets (before NYSE trading time) and on thefutures market (which always shows marketmovements before the spot market),

� As is always the case, this led to a surgeof program trading orders as arbitrageursrushed in to exploit the slight mispricings.

� The communications system to the marketmakers overloaded and could not cope withthe orders.2

� This led to con�rmations of many trades tak-ing over an hour.

� This led to a panic selling on the part oftraders across the world, which produced amajor crash.

Hence, it is correct to say that \program trad-ing had something to do with the October 1987crash", but it is incorrect to blame the crash uponprogram trading. The blame, if any, falls on thecomputer networking which links up the world to

2It should be noted that the biggest 50 to 100stocks in the S&P 500 were present in arbitrage trans-actions in quantities larger than 2100 shares, so thatprogram trading could not be done for these. Tradingin these stocks involved a human runner carrying theorder to the specialist post.

15

the market maker, and on the basic methods offunctioning at the NYSE.

In India, in any case, because the major marketsuse no market makers, the entire method of ordermatching is quite di�erent { it consists of comput-ers directly talking to the central order-matchingcomputer. In this sense \program trading" (i.e.,trading by using computers) is routine in India.The NYSE started out as a labour-intensive mar-ket, and computerised communications with themarket maker was put in3 as a sideshow to themain processes of the market. In contrast, mar-kets in India are purely computer-driven, and theircomputer networking is less fragile. ��

Also see: ?, ?, ?, ?.

Q2.18: Will derivatives destabilise the stockmarket? Could this happen in extremeevents?A: The available international evidence says thatmarket quality on the underlying market improves

once derivatives come to exist. Derivatives im-prove liquidity on the underlying market (for avariety of reasons outlined in Questions 2.14 and2.15), and a more liquid market is one that is ableto absorb larger shocks for a smaller change inprices. This would be the most useful in extremeevents { it is in an extreme event that the liquidityof a market is taxed the most, and at such times ahealthier cash market would be the most valued.��

Q2.19: Is there more or less of a \naturalmonopoly" in derivatives trading, as com-pared with the spot market?A: In the spot market, the ability for exchanges todi�erentiate their products is limited by the factthat they are all trading the same paper. Thisreduces the avenues for product di�erentiation byexchanges.

In contrast, in the derivatives area, there are nu-merous avenues for product di�erentiation. Eachexchange trading index options has to go throughthe following major decisions:

3The \Designated Order Turnaround" system,which allows computers to communicate orders tomarket makers, was setup at NYSE in 1976 and up-graded to \SuperDOT" in 1984. As of October 1987,orders of smaller than 2100 shares could be sent elec-tronically.

1. Choice of index

2. Choice of contract size (i.e. multiplier)

3. Choice of expiration dates

4. Choice of american vs. european options

5. Choice of rules governing strike prices

6. Choice of trading mechanism (whether mar-ket makers, or order-matching market, etc)

7. Choice of time of day when market opens andcloses

In the derivatives area, it is easier for exchangesto di�erentiate themselves, and �nd subsets of theuser population which require di�erent features inthe product. In the US, the experience of futuresmarkets is that between 1921 and 1983, 180 dif-ferent futures contracts had been launched, and afull 40% of these failed to survive four years. Sucha steady process of entry and exit is extremelyhealthy in terms of the basic economics of compe-tition.In this sense, the derivatives area is less of a

natural monopoly than the cash market. ��

Also see: ?.

Q2.20: What are the policy implications ofthis lack of a natural monopoly?A: To the extent that a marketplace is competi-tive, with a steady pace of entry and exit, the self-interest of exchanges will drive them to do thingswhich their investors like. It will not be necessaryto force them to do these things via regulation.The role for regulation in such competitive mar-kets is limited to the classic regulatory functions,i.e. \health{safety{environment" style regulation.��

Q2.21: At the operational level, how do se-curity contracts compare versus index-basedcontracts?A: The basic fact is that index-based contractsattract a much more substantial order- ow, whichhelps them have tighter spreads (i.e. greater liq-uidity). At a more basic economic level, we saythat there is less asymmetric information in theindex (as opposed to securities, where insiders typ-ically know more than others), which helps index-based trading have better liquidity.At settlement, in the case of security-options,

there is the possibility of delivery, and in that

16 CHAPTER 2. MARKET MICROSTRUCTURE

case arises the question of depository vs. physi-cal delivery. Both alternatives are quite feasible.However, in index-based contracts, that questiondoes not arise since all index-based contracts arecash-settled.The index has much less volatility than individ-

ual securities. That helps index options have lowerprices, and index futures can work with lower mar-gins.The most important di�erence between the in-

dex and individual securities concerns manipula-tion. Given that an index is carefully built withliquidity considerations in mind, it is much harderto manipulate the index as compared with the dif-�culty of manipulating individual securities. ��

Chapter 3

Derivatives Disasters

Q3.1: Why do we keep reading about disas-ters involving derivatives?A: Disasters involving derivatives make for goodreporting. In an multi{trillion dollar worldwideindustry, some disasters are inevitable. ��

Q3.2: Why have we seen more disasters inthe recent years?A: As the derivatives industry grows, more dis-asters would be observed. This is perhaps likethe airline industry: when more planes y, moreplanes will crash (see Table 1.2 for the growth ofthe global derivatives industry). ��

Q3.3: How much money has been lost inthese derivatives disasters?A: The cumulative losses from 1987 to 1995 addup to $16.7 billion. This is a tiny fraction of theoutstanding positions of the industry, which werearound $50 trillion as of 1995.Derivatives account for a small fraction of the

overall picture of �nancial disasters. Over thissame period, i.e. from 1987 to 1995, the �nan-cial industry has seen other large disasters:

1. Malaysia's Central Bank lost $3 billion in1992 and $2 billion in 1993 in taking posi-tions on the UK pound.

2. In December 1993, the Bank of Spain tookover Spain's �fth biggest bank, which had$4.7 billion in hidden losses.

3. In 1994, Credit Lyonnais (the biggest state{owned bank in France) was kept a oat usinga $10 billion subsidy from the government.

4. In the 1980s, the \savings and loans" indus-try of the US lost $150 billion.

5. Japan's �nancial institutions are said tobe sitting on $500 billion of nonperformingloans.

��

Also see: ?.

Q3.4: What happened in Barings?A: Mr. Nick Leeson, a trader for Barings Futuresin Singapore, had positions on the Japanese Nikkei225 index worth $7 billion. In addition, he hadother positions on options and bond markets. Mr.Leeson was able to dodge internal corporate con-trols and adopt these large positions unchecked.This was assisted by weak enforcement at the ex-changes in Singapore and Osaka, who did not gen-erate alerts to his large positions. ��

Q3.5: What should be done to minimise dis-asters with derivatives?A: At the level of exchanges, position limits andsurveillance procedures should be sound.At the level of clearinghouses, margin require-

ments should be stringently enforced, even whendealing with a large institution like Barings.At the level of individual companies with po-

sitions on the market, modern risk measurementsystems should be established alongside the cre-ation of capabilities in trading in derivatives. Thebasic idea which should be steadfastly used whenthinking about returns is that risk also meritsmeasurement. ��

Also see: ?.

17

18 CHAPTER 3. DERIVATIVES DISASTERS

Chapter 4

Policy Issues

Q4.1: What emerging markets have alreadycreated derivatives markets?A: The status is summarised in Table 4.1, whichshows emerging markets that have derivativesmarkets today, and Table 4.2 which shows emerg-ing markets which are in the process of buildingderivatives markets. ��

Also see: ?.

Q4.2: What was China's experience in thisarea?A: China had a mushrooming of derivatives ex-changes in the early 1990s. Many of these werepoorly run, and experienced signi�cant episodesof market manipulation and counterparty risk. In1994, the 50 exchanges were consolidated into 15.In 1995, China's futures markets did a trading vol-ume of around $1.2 trillion (for a comparison, In-dia's equity markets do an annual trading volumeof roughly $180 billion).

Many observers have cited China's experiencewith 50 exchanges as an example of how poorly{regulated and hasty growth of derivatives mar-kets may be problematic. However, the other sideof the picture is now clear: the experience withthese 50 exchanges got the Chinese markets o� theground, and generated the necessary know-howamongst exchange sta�, regulators and users. Inthe end, China's derivatives exchanges has stolena march on their rivals: they now have signi�canttrading volumes on a world scale. ��

Q4.3: What �nancial markets in India areready for derivatives today?A: In India, two areas are ripe for derivatives: theequity market and foreign exchange.

In the case of the dollar-rupee exchange rate, aforward market already exists; it is just a matter offormally institutionalising it at an exchange, andturning it into a modern futures market. ��

Q4.4: Are derivatives in interest rates viablein India?A: In the case of interest-rate risk, derivatives inIndia are hindered by the poor liquidity on the�xed-income market.However, a few approaches towards designing

interest-rate derivatives could commence. An ex-ample of this would be a futures contracts on trea-sury bills, which would give people the ability tobuy or sell treasury bills in the future. The lack ofa liquid and transparent market for treasury bills,and constraints such as the inability to short-selltreasury bills, would hurt the ability to do arbi-trage on this market. Hence, the market e�ciencyof the interest-rate futures market would be lim-ited.However, in an environment where economic

agents are exposed to interest-rate risk and haveno alternative risk management facility, such con-tracts could still prove to be viable. If interest-ratefutures came about, they would generate greaterorder ow and improve market quality on the�xed-income market. ��

Also see: ?.

Q4.5: Why are commodity futures marketsimportant?A: India's farmers, and downstream industrialusers of agricultural output, are exposed to ex-tremely high risks. The creation of commodityderivatives markets will provide them with thechoice of obtaining insurance against price uctua-tions. It will improve liquidity and price discovery

19

20 CHAPTER 4. POLICY ISSUES

Table 4.1: Derivatives Exchanges in Emerging Markets

� Brazil (BM & F) � China (SSE, SME, SHME, SCCFE)

� Guatemala (BDP) � Hungary (BCE & BSE)

� Korea (KSE) � Malaysia (KLOFFE, KLCE, & MME)

� Philippines (MIFE) � Portugal (PSE)

� Russia (MICEX & MCE) � Slovak Republic (Bratislava)

� Slovenia (Ljubljana) � South Africa (SAFEX)

Table 4.2: Emerging Markets Working Towards Derivatives

Argentina Bulgaria ChileColombia Costa Rica Czech Republic

Greece India Indonesia

Mexico Poland Taiwan

Thailand Turkey

in the underlying spot markets. Once futures mar-kets exist, the private sector will maintain bu�erstocks which will reduce spot price volatility, andthe private sector will do this far more e�cientlythan government-sponsored e�orts at maintainingbu�er stocks. In addition, the creation of thesemarkets is consistent with the growth of skills inIndia's �nancial industry in the area of derivatives.��

Q4.6: What are the issues in the creation ofcommodity derivatives markets?A: Like most traditional �nancial markets in In-dia, the commodity futures markets are weak interms of modern skills in how exchanges shouldbe run. These markets are weak on a varietyof issues: the use of modern market mechanisms(such as the electronic limit order book market),enforcement of contract standardisation, dealingwith hetereogenous grades, the counterparty guar-antee of the clearinghouse, calculation and en-forcement of margins, and checks against marketmanipulation.

The commodity markets which are now in thespotlight are: the international pepper market,the proposed markets in cotton and jute, and theproposed international castorseed market. Ideally,the management of these exchanges will be able tofunction to international standards. In this case,

the value of having commodity futures marketswould become apparent, and the stage will be setfor further expansion of commodity futures mar-kets in India. If these markets experience a visibleepisode of manipulation, or if they experience apayments crisis, then it will be harder to estab-lish a consensus about the future of commodityfutures markets, and the development of India's�nancial system will be slowed.

One serious weakness in India lies in the wayindividual commodity futures markets are an out-growth of trading on individual spot markets. Thecotton trading community will create a cottonfutures market, the jute trading community willcreate a jute futures market, etc. This is inef-�cient insofar as it does not foster the growthof specialised skills which are common to all fu-tures markets and not speci�c to one commodity.For a well-functioning derivatives exchange, spe-cialised skills are required on the part of exchangeand clearinghouse sta� and on the part of tradingmembers. These skills are primarily in the deriva-tives area, and they are easily transferable fromone commodity to another.

Ideally, a derivatives exchange should have a fo-cus on futures, options, and other derivatives re-gardless of what the underlying is, and each fu-tures exchange should trade dozens of commodityfutures contracts. This is similar to markets like

21

CME and CBOT which trade derivatives on hun-dreds of commodities, and on a host of other un-derlyings such as stock market indexes, treasurybills, foreign exchange, etc.

One useful alternative here is to involve main-stream �nancial markets into the commoditiesarea. Exchanges which have clearing corporations(and can hence supply the counterparty guaran-tee) can easily introduce cash-settled derivativeson commodities. ��

Q4.7: What can be done in derivatives onreal estate?

A: In the case of real estate, derivatives canonly follow clear asset securitisation. Governmentshould work towards removing hurdles in the faceof real estate asset securitisation, which wouldthen enable derivatives on real estate to take place.Mutual funds which invest in real estate are an-other, easier, stepping stone towards derivativeson real estate: markets could easily trade units ofsuch funds, and options on such units. This maybe a shorter route towards obtaining derivativeson real estate. ��

Q4.8: Should foreigners be restricted in In-dia's derivatives markets as a matter of pol-icy?

A: As in other areas of industry and commerce,the key objective for policy in India should be toobtain the best quality of products and services forIndia's economy, ensure conditions of intense com-petition on the domestic market, and to employIndian labour and Indian capital at the highestpossible levels of productivity. Given these objec-tives, the nationality of the ultimate owners of a�rm operating on Indian soil has little importance.

In the case of all derivatives, foreigners shouldbe allowed free access to trading and brokerageon India's derivatives markets. This will help im-prove the quality of India's derivatives markets,and help the dissemination of knowledge about therisk management capabilities that these marketssupply to the community of foreign investors.

Hurdles to this level of liberalisation make itdi�cult for us as a country to realise the full po-tential of the investment that can be attracted intoIndia given the level of development of our �nan-cial system. ��

Q4.9: What would access to derivatives doto FII and FDI investment?A: Access to derivatives would increase the owof FII and FDI investment. The two importantkinds of risk that foreign investors are exposed toare currency risk and country risk. The �rst wouldbe manageable using dollar{rupee futures and op-tions, and the second would be manageable usingindex futures and options.The details of usage would be subject to the

individual requirements. For example, some FIIsmight choose to completely eliminate their dollar{rupee exposure, coupled with \portfolio insur-ance" to cap their downside exposure at no worsethan x%. Other FIIs might choose to use indexderivatives as a liquid way to increase their equityexposure. Similarly, investors in India throughFDI would be able to use dollar{rupee futures tocontrol their risk of a currency devaluation, anduse index futures to proxy for the overall successof India's economy.These methods of controlling risk are quite rou-

tine in the international �nancial community.1 IfIndia had derivatives markets, then it would be aless risky environment and would better be ableto attract foreign investment.While India lacks index derivatives as of to-

day, there is a direct opportunity to make progresson these issues via the dollar{rupee forward mar-ket. The constraints that are placed in the way ofFIIs on using the dollar-rupee forward market arecounterproductive. If the FII is allowed to obtaininsurance using this market, he will bring moremoney into India. ��

Q4.10: Is India ready for derivatives today?A: There are four key aspects to this question.

� Market Size

Derivatives markets need to work o� a largefoundation of asset value that is traded onan underlying market. India's debt markethas a total market capitalisation of aroundRs.3 trillion, and India's equity market has amarket capitalisation of around Rs.5 trillion.

1See, for example, the article Investors Return to

Latin America by Margaret Elliott, Derivatives Strat-egy, May 1997, where Hari Hariharan of SantanderInvestment is quoted as saying that 60{70% of FIIinvestment into Latin America involves some use ofderivatives.

22 CHAPTER 4. POLICY ISSUES

India's foreign exchange market also has aconsiderable underlying market size.

International experience and the success ofderivatives in many countries of much smallermarket size (for example, the Johannesburgfutures market, and the Brazilian futuresmarkets) shows that in India, each of thethree markets mentioned above is ready forderivatives.

� Liquidity

In India, NSE proposes to launch futures andoptions contracts on the NSE-50 index. Themarket impact cost seen with Nifty is com-parable to some of the most liquid marketindexes in the world (e.g. the S&P 500 ofthe US). India's foreign exchange market alsopossesses the low transactions costs to sup-port a healthy derivatives market. India'sdebt market might currently not ready fortrading derivatives since most of the key in-struments traded here are quite illiquid.

Thus from the point of view of the liquidity ofthe market, two of India's markets are readyfor derivatives trading.

� Clearing Corporation

For derivativesmarkets to support large-scaleuse, it is important to have a clearing corpo-ration which guarantees the trade.

From July 1996 onwards, with NSCC guar-anteeing trades on the NSE, this prerequisitefor a derivatives market now exists in India.The human capabilities that go into creatinga clearing corporation can also be easily re-deployed to new markets, such as the foreignexchange market which has such a deep needfor trading derivatives.

� Sophistication of traders

Derivatives are complex. The payo�s thatbuyer and seller face, the risks that buyerand seller face, and the economic theory thatis used for pricing derivatives: all these areconsiderably more di�cult than that seen onthe equity or the debt market.

India's �nancial industry already has experi-ence with many kinds of derivatives. As com-pared with many emerging markets wherederivatives exist, India's �nancial industrypossesses very strong human skills. It ap-pears that the foundations of human capital

that are required for derivatives might wellbe in place in India.

In the context of the four points above, deriva-tives on the debt market would be highly attrac-tive to investors who face interest rate risk, butthe debt market su�ers from an illiquid under-lying. The foreign exchange market is also anarea where derivatives are clearly valuable; theconstraints faced there concern the sequencing ofliberalisation of this market. The Indian equitymarket satis�es all the four criterion laid out andis the logical candidate to carry the �rst exchange-traded derivatives in India, with the index as theunderlying.2 ��

Q4.11: What are the costs and bene�ts ofdelaying the onset of exchange-traded �nan-cial derivatives in India?A: The costs are on two directions:

1. The most important cost is opportunity cost.

� India's investors will bene�t from be-ing able to access derivatives. Everyinvestor who experiences pain owing toindex uctuations could be happier ifindex derivatives existed.

� India's markets will become more liq-uid and e�cient once derivatives arepresent.

� India's �nancial industry will growskills and capabilities through workingwith derivatives which will help it comeup to international standards.

Each of these three developments is put o�further into the future when derivatives inIndia are delayed.

2. The second cost is the threat of foreign ex-changes creating derivatives markets on In-dian underlyings. If this took place, it wouldmake it harder for derivatives exchanges inIndia to succeed.

The apparent bene�t of delay is the opportu-nity to create a concerted e�ort in training andimproving skills as a preparation for the launch ofa market. This bene�t is illusory, for two reasons:(a) a concerted e�ort to pickup skills will not take

2See Barclay (1996) for issues connected with In-dia's derivatives markets.

23

place unless the market launch is imminent, and(b) the best form of training in derivatives is thatobtained when actually using the instruments. SeeQuestion 5.25 for a further treatment of this issue.��

Q4.12: What international derivatives ex-changes are working towards launching prod-ucts o� underlyings in emerging markets?A: The Chicago Mercantile Exchange (CME),Chicago Board Of Trade (CBOT), Chicago BoardOptions Exchange (CBOE), American Stock Ex-change, Sydney Futures Exchange, Hong KongFutures Exchange and Singapore InternationalMonetary Exchange (SIMEX) have all launchedemerging market initiatives, whereby they aim totrade derivatives o� underlyings from emergingmarkets. ��

Q4.13: What derivatives on Indian underly-ings are currently trading abroad?A: Examples of contracts that exist abroad as oftoday fall into the following categories:

� Many GDR issues are bundled with war-rants (option-like instruments), which arethen traded separately. For example, theHindalco issue done on 2 Nov 1995 bundledevery two shares with one warrant. Similarissues have taken place on India's primarymarket { the di�erence here is that the war-rants are listed and traded, in contrast withIndia's secondary market where the warrantsare not traded.

� Warrants on mutual fund paper such as theLazard Birla India and Fleming Indian arelisted in London.

� Custom built (OTC) derivatives { speci�-cally, options and swaps { on Indian marketindexes and baskets of Indian GDRs alreadyexist on the internationalmarket. Essentially100% of the OTC derivatives industry on In-dian underlyings lies abroad.

� Restrictions upon access to the dollar-rupeeforward market in India has led to the de-velopment of the \non-deliverable forward"(i.e., cash settled) dollar-rupee market o�-shore.

��

Q4.14: Would foreign derivatives markets beinterested in launching trading on Indian un-derlyings?A: Internationally, derivatives trading is a �ercelycompetitive area where exchanges are constantlytrying to �nd interesting new contracts based onwhich trading volume can be attracted. Hundredsof new contracts have been attempted in the lastfew years, only around half survive more than afew years. In such an atmosphere of hectic innova-tion in contract design, Indian currency and stockindex products constitute a glaring opportunity.��

Q4.15: Is this a real threat?A: If India does not progress towards derivativesswiftly enough, contracts based on Indian under-lyings will start trading on markets elsewhere inthe world. The NSE-50 time series or the dollar-rupee exchange rate are available on internationalinformation services such as Reuters or Knight-Ridder, and nothing prevents a foreign marketfrom launching contracts on these.3

Such a scenario is not without precedent. Someexamples can be cited which show the forces in-volved:

� In 1989, regulatory errors in Japan led tothe market for derivatives based on Japan'sNikkei 225 index moving o� to Singapore andpartly to Chicago. Today, Japan's marketsfeature a better set of regulations, but themarket has not yet moved back to Japan.

� Similarly, a large part of Sweden's �nancialmarkets moved to London after the Govern-ment imposed a tax on trading volume in1989. After the transactions tax was revoked,most of this volume did not return.

� After prolonged delays in creating a deriva-tives market in Taiwan, derivatives on theTaiwanese market index4 started trading inChicago and Singapre on 9 January 1997 (?),

3Foreign markets would prefer to have the bene�tof cooperation from the Indian owner of an index be-fore using it for derivatives. But as the example of theNikkei-225 illustrates, this is not a binding constraint.

4Taiwan's equity market remains relatively closedto foreign investment and has short trading hours.Taiwan's SEC tried unsuccessfully to prevent CMEand SIMEX from introducing Taiwanese index futures,but interminable delays a�icted e�orts to get a lo-

24 CHAPTER 4. POLICY ISSUES

� India's GDR market experience is a directexample of markets moving o� o�shore if therequired facilities don't come about in Indiafast enough (see Question 4.17, however).

These examples suggest that the movement ofmarkets indulging in \regulatory arbitrage" is notjust a abstract possibility but a real alternative forinvestors seeking to meet their objectives. ��

Q4.16: What are the implications of deriva-tives on Indian underlyings trading abroad?A: As the Nikkei-225 experience suggests, oncea contract gets well-established at a market, itdoesn't easily move, even if the alternative des-tination proposed is the home country of the un-derlying. In this sense, India's �nancial industrycould then face an uphill struggle for order owif foreign markets are successful on establishingderivatives markets �rst. This would have tworami�cations:

1. As long as India lacks capital account con-vertibility, India's citizens would be disad-vantaged by not being able to access deriva-tives while foreigners would be able to. Thiswould generate increased incentives for In-dian citizens to use illegal channels throughwhich positions would be adopted on foreignmarkets.

2. Such an event would be a setback for the de-velopment of skills and businesses in India's�nancial industry, and for the potential ofBombay as a world �nancial centre.

��

Q4.17: But in the GDR market experience,the foreign market has hardly a�ected order ow into India { why are derivatives di�er-ent?A: In the case of the GDR market, two things weredi�erent.

1. The GDR market is self-liquidating in thesense that GDRs gradually convert into un-derlying shares. Hence the size of the GDR

cal derivatives exchange o� the ground and CME andSIMEX chose to not wait for the local market to comeabout. Taiwan's SEC initially prohibited local bro-kerage �rms from trading on these contracts o�shore.Later, the SEC announced that they \may" allow localorders to go to derivatives exchanges o�shore.

market inherently diminishes over time. Thisis not the case with derivatives.

2. In India's GDR episode, GDRs were an alter-native in the face of a thriving Indian spotmarket. It is always di�cult to take awayorder ow from an existing market. In thecase of derivatives, if foreign markets get es-tablished �rst, this will not be the case.

��

Chapter 5

Regulatory Issues

Q5.1: What are the objectives of regulation?A: There are three basic objectives of regulation:

� to protect market integrity,

� to ensure �erce levels of competition, and

� to prevent fraud.

��

Also see: ?, ?.

Q5.2: What kinds of competition are possi-ble in the �nancial market scenario?A: There are many dimensions of competition:

1. The market should have a large number oftraders.

2. There should be easy entry of new tradersand investors.

3. No individual trader should be too large ascompared with the size of the overall market,i.e., no single individual trader or coalitionsof traders, should have market power.

��

Q5.3: How does deviation from perfect com-petition (or situations of market power) ariseon �nancial markets?A: One form of market power that is commonlyobserved in the world arises with an exchangewhich limits the supply of seats so as to increasebrokerage rates. This behaviour re ects itself inthe price of a seat on the exchange, or the \seatprice". In an ideal economy, the seat price (de-void of any real estate or other facilities) shouldbe close to 0. A high seat price implies bid-ask

spreads and brokerage fees above the level that isfound in perfect competition.This \implicit elevation" can sometimes even

become overt: prior to 1974, NYSE speci�ed a (el-evated) brokerage commission schedule, and mem-bers were required to not o�er prices better thanthe de�ned schedule.In addition to this form, every economy has

some unusually large traders. This is another av-enue through which deviations from perfect com-petition are observed. ��

Q5.4: How should regulation of exchangeswork?A: The most important intuition in regulation ofexchanges is to view the exchange as a manufac-turer of liquidity services. If exchanges do thiswell, they will get satis�ed customers. Exchangesthat fail to do this well will fail to get businessand go bankrupt. In India we have seen numerousindustries and services where competition and thesteady process of entry and exit have proved to bea great success in producing high quality and lowprice. The area of trading services is no exception.The key role for public policy is to keep en-

try barriers low and therefore keep the competi-

tive pressure upon the incumbents high. It shouldbe easy to start new exchanges; even for businesshouses to start exchanges. It should be easy (say)for CBOT to come to India and start an exchange.That will serve to keep up competitive pressureand steadily improve the services and costs thatend-users, the investors, face. ��

Also see: ?.

Q5.5: What can regulation do to encouragecompetitiveness?A: Brokerage fees are elevated as long as a re-

25

26 CHAPTER 5. REGULATORY ISSUES

stricted supply of exchange seats is used by ex-changes in India. Hence regulators should pay at-tention to seat prices, and require exchanges toincrease the supply of seats when seat prices riseto signi�cant levels.

There is an intuitive urge to set very high capi-tal adequacy requirements to ensure that the riskof counterparty failure is reduced. But a funda-mental fact of the counterparty guarantee of aclearing corporation is that it eliminates creditrisk, regardless of the size of the company that istrading. Hence the intuitive urge to set very highcapital adequacy requirements should be checked,since one of the less attractive outcomes of settinghigh capital adequacy requirements is low compet-itiveness of the industry.

Position limits have been proposed as a wayof preventing the damage that a large trader cancause. Position limits are particularly commonin the area of commodity futures, where a shortsqueeze is the constant danger (with cash-settledcontracts, this is less of an issue). However, posi-tion limits have not been very successful in thepast, because a manipulator can always spreadhis position amongst several entities and avoid theposition limit. Famous episodes of manipulation,like the Hunt brothers in silver, were done in theface of strong position limits. Thus the regulatorshould be wary of using position limits in the hopeof preventing abuses of market power. ��

Q5.6: What should entry or eligibility re-quirements be for derivatives trading?A: The thrust of economic policy in India todayis to encourage the competitive forces of the mar-ketplace to di�erentiate winners from losers. A�rm that unwittingly goes into derivatives tradingwithout understanding the business is no di�erent(say) from a �rm which unwittingly goes into anyother high technology area (like computer softwareor banking or oriculture). If the �rm is unable tocope with the complexities of this area, it would gobankrupt. Thanks to the system of margins andcounterparty guarantee, such bankruptcies wouldhave no impact upon the rest of the market.

The danger with eligibility criteria is that theye�ectively become entry barriers. All too often,entry barriers are used by incumbents to reducethe degree of competition in an area. The basicfocus of economic policy should always be to max-imise the degree of competition in any industry.The brokerage industry is no exception. ��

Q5.7: What is fraud?A: Fraud consists of market participants mak-ing commitments which are not later upheld. Amore tenuous situation is if a market participant is\opaque", which then means that there are largecosts to be be paid (a) in establishing antecedentsand (b) in con�rming that the promise will ac-tually be upheld, before inter-party transactionstake place.

The importance of \trust" and \reputation" inthe world is a re ection of relationships which areable to avoid fear of fraud. Unscupulous compa-nies have a way of going bankrupt over time, sothat companies with a longstanding reputation areless likely to indulge in fraud.

This ties in with the idea of restrictions to en-try mentioned in Question 5.6. If many marketparticipants require a great deal of trust and along-standing relationship before they do business,this is e�ectively an entry barrier which limitsthe competitiveness of the industry and elevatesprices. A well functioning market economy is onewhere strangers can trade with each other; theneed to establish trust and long-standing relation-ships should be as limited as possible.

In this context, the counterparty guarantee ofthe clearinghouse is a crucial device which elim-inates part of the need for trust, and hence in-creases the contestability of the market. The rolefor regulation is to steadily reduce the role fortrust and relationships in the market, so as tofoster free entry and increase competition on themarket. ��

Q5.8: How can regulation diminish the ex-tent of fraud?A: There are two key methods through which pub-lic policy can reduce the extent of fraud: throughimproved disclosure and by ensuring swift andcredible legal redress, in cases of fraud. ��

Q5.9: Should there be regulatory controlover contract de�nition?A: Some countries require that the regulator havea say in contract design. It is hard to support theclaim that this has played a useful role.

When a contract is poorly designed, it will diea natural death under conditions of low tradingvolume. For example, ? analyses the life cycle ofthe �rst index contract to be traded in the U.S.,

27

the Kansas Value Line index contract, which wasvery complex to price and, therefore, complex totrade. These contracts had hence su�ered poorliquidity since conception, and have been eventu-ally removed from trading.

Worldwide, the derivatives industry features ahectic pace of contract design. Every year, newcontracts are launched and old contracts die. Itis a di�cult proposition for regulation to second-guess the success of a contract.

In the case of index derivatives, one possiblearea of concern is the extent to which an indexcan be manipulated. However, the marketplaceis typically a very e�ective check here. If futuresstart trading on an index which is easy to manip-ulate, then these contracts will �nd it di�cult toattract trading volume. This is a more e�ectiveway to control poorly thought out contracts thanto use regulation.

On the negative side, regulatory control overcontract de�nition has often been a vehicle forpolitical lobbying by pressure groups against fu-tures markets. For example, large and entrenchedonion traders have successfully lobbied with theCFTC in the US to prevent CBOT from allowingthe trading of onion futures, even though it is al-most certain that onion futures will be a highlyuseful contract.

Hence the worldwide experience in regulationof contract de�nition is ambiguous, and does notshow the regulator as adding any improvementover market forces in de�ning good vs. bad con-tracts. ��

Also see: ?, ?.

Q5.10: Should the securities that are used inan index be required by regulation to be indepository mode?A: This is an example of the usefulness of the prin-ciple about regulatory non-involvement in con-tract de�nition, shown in Question 5.9.

If the securities are not in depository mode,the transactions costs of trading the index willbe higher. This means that market e�ciency ofthe index will be reduced, and trading volume onthe derivatives market will be reduced. Hence, theself-interest of the exchanges points towards usingsecurities in the depository. This is a question thatthe derivatives exchange should address. There isno role for regulation here. ��

Q5.11: Should regulation require that deriva-tives trading be organised at an exchangewhich is distinct from the spot market?A: This is similar to the question should the en-

gine factory of Mahindra & Mahindra be located at

the same place as their paint shop?. To be moreprecise, this is like the question should Mahindra

& Mahindra buy their gearbox from a third party

or should they make it themselves?.How much vertical integration is optimal is a

question for exchanges to address. If there arebene�ts of one solution over the other, then cer-tain exchanges will succeed in obtaining order ow(see Question 5.4), and the exchange industry willendup using one kind of technological solution.Once again, there is little role for a regulator onthis issue. ��

Q5.12: At an operational level, is it better tohave the spot and futures market under oneroof?A: At the operational level, it makes sense to haveboth index futures and the spot index being tradedon the same exchange in the same time of day, forthree reasons:

1. Margining can be done correctly if both legsof the transaction go through the same ex-change; if a person is long index and shortthe spot then he should ideally be chargedless initial margin.

2. A coordinated e�ort at manipulation is easierto detect. If the same person is long the Gu-jarat Cotex spot and long in call options onGujarat Cotex, then it can send alarm bellsgoing. In contrast, genuine inter-exchangecoordination in margining and in surveillanceis much more di�cult to create { the Baringsexample shows that SIMEX and Osaka don'treally talk to each other.

3. Arbitrage is made easier since both legs of thetransaction are on the same exchange, whichreduces the possibility that the spot and thefutures markets can deviate from fair valueseven for a short interval of time.

At a mundane level, the distinctions betweenspot or derivatives (or the distinction betweentrading commodities vs. trading equity vs. trad-ing foreign exchange) that are seen in other coun-

28 CHAPTER 5. REGULATORY ISSUES

tries are typically historical accidents that re- ect (a) peculiarities of regulation, and (b) pre-technological implementations of markets withoutusing computers. We in India have the advantageof building a �nancial system in an era of mod-ern technology, with a smaller set of entrenchedinterests defending the status quo. ��

5.1 Risk Management at

Clearing

Q5.13: Why is the clearinghouse central toderivatives markets?A: The key factor enabling exchange-tradedderivatives is the credit guarantee supplied by theclearing corporation. If derivatives involved obli-gations between individual market participants,then large positions between two random individ-uals would be less feasible thanks to counterpartyrisk (as mentioned earlier in Question 5.7).

With the clearinghouse counterparty guaranteeaccounting for counterparty risk, small individu-als and big individuals can form positions againsteach other without any special risk factor favour-ing any one side. With the counterparty guaran-tee, derivatives can exist with both sides being freeof the worry that the other will default. ��

Q5.14: How should margining for derivativeswork?A: Today, margining systems are accepted as be-ing the foundation through which the clearing-house guarantees the trade on a futures market,all over the world. This has become extremelyimportant in India where futures-style settlementis used in \the cash market" at all stock exchangesother than OTCEI. At futures markets worldwide,margining works in two steps:

1. An initial margin is charged, which dependsupon the position taken.

In India, unlike in other countries worldwide,the banking system is unable to move fundsswiftly. However, a situation where initialmargin is paid after the position is adoptedis unsafe for the clearinghouse. Hence thesolution, which has been widely utilised inIndia, is the exposure limit. This can be in-terpreted as an advance payment of initial

margin and members are constrained to nottake a position larger than that supported bythe funds deposited in advance. For example,if an exposure limit of 33 times base capitalis in place, then it means that the exchangerequires a 3.33% initial margin.

2. The net pro�t or loss on a position is paid outto or in by the member on the very same day,in the form of daily mark-to-market (MTM)margins.

A large loss, when accumulated over severaldays, generates a temptation to default atsettlement. To prevent this from happening,the loss of each day is paid up on that dayitself. The member will not default on theMTM payments as long as the the one-dayloss is smaller than the initial margin (whichthe exchange forfeits if the member defaults).

��

Also see: ?, ?.

Q5.15: How is initial margin (i.e. \exposurelimit") calculated?A: Earlier, we remarked that \the member willnot default when the one-day loss is smaller thanthe initial margin". Assuming that MTM marginis fully and correctly computed and charged (i.e.,net losses are taken from members on the sameday, net pro�ts are paid to members on the sameday), the \correct" level of initial margin is thatwhich is larger than what can be expected for aone-day loss to the position, with a comfortablesafety margin.Intuitively, the \correct" level has to be sensi-

tive to the composition of the position taken. If aperson has a position with 100% of the exposure inApollo Tyres, then this is a highly risky position.The level of initial margin required here would bequite large. If, instead, a person has a well-spreadout position with positions spread over numeroussecurities, then the risk is lower because he is di-versi�ed. In this case, the \worst one-day loss"scenario becomes less volatile, and therefore, thelevel of initial margin required is lower.We quantify risk in terms of the standard devi-

ation of returns of the portfolio or �. Typically,a very high level of safety could be obtained bycharging initial margin of four times the �. Incase a person has a 100% position in Apollo Tyres,the � is around 8%. So a safe initial margin for

5.1. RISK MANAGEMENT AT CLEARING 29

the person with the 100% Apollo Tyres positionis around 30%. In other words, his exposure limitshould be around three times of base capital.

At the other end, a person having exposure ina fully diversi�ed portfolio (i.e., he has bought allthe 50 stocks in their correct proportions in theNSE-50 index) has a � of just 1.3. The NSE-50index, being highly diversi�ed, is much less riskyas compared with Apollo Tyres. In this case, theidentical level of safety (i.e., four times the �)is obtained by charging him an initial margin of5.2%, i.e., an exposure limit of 20 times base cap-ital.

These two examples illustrate a key idea: de-

pending upon the portfolio composition of expo-

sure, the same level of safety can be obtained by

capping exposure of 3 times or 20 times the capital

deposited with the clearinghouse. The correct levelof initial margin varies strongly with the portfoliocomposition of the exposure, whereas simple ruleslike \33 times base capital" or \10 times base cap-ital" will not work correctly: they charge too littleinitial margin for risky positions and too much ini-tial margin for relatively safe positions.

One more idea that ows from this logic is thatgross exposure is an incorrect measure of risk. Weneed to focus on the � of his full portfolio expo-sure.

A nuance here concerns long vs. short posi-tions. A position which is long Reliance and longSBI has a certain �. A position which is long Re-liance and short SBI has a smaller �. This is be-cause when market index uctuations take place,then the short position is a hedge against the longposition. In this sense, if a person has Relianceexposure, he can actually reduce his risk by in-creasing his gross exposure (i.e. by shorting SBI).This is an example of how gross exposure is a poormeasure of risk. Good portfolio margining wouldcorrectly integrate long vs. short positions intothe initial margin calculation.

These ideas are standard procedure at futuresmarkets all over the world. Well-established soft-ware systems named SPAN or TIMS are availableto calculate margins, and less-well-established al-ternatives are available which do more sophisti-cated calculations of the true � of the portfolio.This is the direction which should be adopted inIndia's markets also. ��

Q5.16: How is daily mark to market (MTM)margin calculated?

A: The calculation of daily MTM margin is eas-ily done as the net loss associated with a posi-tion. This is paid up each evening after tradinghas ended. Two nuances are of interest here:

1. The correct computation of MTM margin isto focus on the net loss across all di�erentsecurities on which positions are held by themember.

2. On futures markets all over the world, pro�tsare paid by the clearinghouse to members ona daily basis, just like losses are paid in tothe clearinghouse by members. The marginsre ect the symmetry in taking positions onthe futures markets { the losses made by oneside of the contract are the pro�ts made bythe other side.

��

Q5.17: How does the slow payments systemchange these calculations?A: Suppose daily MTM payments cannot be con-�rmed on the same day. In this case, the clearing-house takes a risk of a multi-day loss instead of asingle-day loss. This is easily handled using thepT formula: T day exposure has a standard de-

viation which ispT times larger than one-day ex-

posure. Hence, if we think that the typical initialmargin has to be 4�, and if the payments systemintroduces a three day delay, then the appropriatelevel of initial margin is 7�, where 7 is roughly4p3.

Many exchanges abroad have the capability ofsuspending trading at 11:30 in the morning ondays of exceptional market index volatility, anddoing a MTM margin call. This obviously de-mands a strong banking system which can movefunds within �ve minutes. This capability allowsthe exchange to further reduce the size of initialmargin required. If the exchange has this capabil-ity, of stopping trading for �ve minutes halfway inthe day on exceptionally volatile days and charg-ing MTM margin on the spot, then the appropri-ate level of initial margin becomes 4p

2�, or 2:8�.

In this way, infrastructure in the form of a fastpayments system reduces the working capital re-quired in the �nancial industry. ��

Q5.18: What are prospects for improvementsof the banking system?

30 CHAPTER 5. REGULATORY ISSUES

A: Some banks are already much faster than oth-ers on movement of funds. As of today, the Na-tional Securities Clearing Corporation has con-tracted with HDFC Bank as a clearing bank;HDFC Bank has Electronic Funds Transfer (EFT)and o�ers same-day con�rmation of funds. GlobalTrust Bank also o�ers 30-minute con�rmation offunds.

Canara Bank (also a clearing bank with NSCC)is in the process of setting up EFT. The ReserveBank of India has a major initiative to establisha nationwide infrastructure for electronic fundsmovement. ��

Q5.19: How does options margining work?

A: In the case of futures, both short and long arecharged initial margin, and after this, both sidespay daily mark-to-market margin. This is not howoptions work. In the options market, the long paysup the full price of the option on the same day,and the short puts up initial margin. After this,the long is relieved of all responsibilities to hisposition, and the short pays daily mark-to-marketmargin.

The initial margin of the option short is thelargest loss that he can su�er with a one-day pricechange that goes against him. This is calculatedusing theoretical option-pricing formulas. ��

Q5.20: What are the special di�culties ofmargining options?

A: For options series that have a strike price that isfar away from the current spot price, the optionsmarket is often quite illiquid. For these optionsseries, mark to market margins (which are chargedto the option short) is hard to calculate { eitherbecause the illiquidity of the options market makesthe market's option price less reliable, or becausethe market fails to trade the option at all on agiven date. In such a situation, theoretical modelsare used to impute the fair price of the option, andmark-to-market is done using this notional price.

In keeping with our argument of Question 5.15above, the initial margin calculation is always con-cerned with calcualating the largest loss which aposition can su�er. This becomes quite complexwhen options are a part of the portfolio, given thenonlinear payo�s of options. ��

Q5.21: What constraints should regulationimpose upon the time to expiration of thesecontracts?A: As discussed earlier, there is a common in-tuition in India where we know that \forwardcontracting becomes more dangerous as the timeto expiration increases". This intuition is outof touch with the functioning of futures marketswhich have daily mark-to-market margins. Aslong as daily mark-to-market margins are chargedcorrectly, it is asif daily settlement is in force.Daily mark-to-market margins break the link be-tween time to expiration and default risk.

Hence the time to expiration seen in the marketis a question that the exchanges should address.As long as daily MTM margin is being chargedcorrectly, it is not a regulatory concern. ��

Q5.22: How does the margining systemchange the way people trade?A: One of the subtle and valuable things about agood margining system is the way it changes thebehaviour of people who trade. People will alwaysadjust their behaviour to minimise the marginsthat they have to pay up. If margins are calcu-lated correctly using portfolio reasoning, then wewill start seeing the phenomenon of \undiversi�edrisk" diminishing. As in the Apollo Tyres case,the level of initial margin charged there would bevery steep (a limit of three times base capital) andpeople would start avoiding such risks.

Such understanding, and wisdom in safe specu-lation, will be good for exchanges and good for thecountry. As long as we charge initial margin in theform of �xed rules like \10 times base capital", wedo not give people incentives for improving theirskills in diversi�cation and hedging. ��

Q5.23: What are the policy issues in clearingcorporation failure?A:While famous events like Barings have capturedthe media attention, it should be noted that thesedid not interrupt the smooth functioning of theclearing corporation, which is the center of focusof the regulator.

However, it is to be expected that once everyfew decades, market uctuations will take placewhich are large enough to bankrupt clearinghouse.For such infrequent events, it makes a lot of sensefor the central bank to supply a line of credit for a

5.2. OTHER ASPECTS OF THE RISKS 31

few days to a clearinghouse at such a time to tideover the exigency.1

It is all too easy for such a guarantee to be coun-terproductive. Once such a guarantee is given,there is, what is called, a moral hazard problem.If the clearinghouse thinks that the central bank isthere to take care of di�culties, then the qualityof attention that the clearinghouse puts into itswork of ensuring clearinghouse strengths throughsafe margining systems would be reduced.

Hence if the central bank wishes to be a lenderof last resort to the clearing corporation, then itshould set standards to con�rm that the risk con-tainment procedures and margin calculations arestrong enough to support a failure of no more thanonce in 25 years. This implies that a stress-testingof the clearing corporation should �nd that theprobability of failure of the clearing corporationis below 0.015 or so. The econometrics profes-sion knows a fair amount about measuring theseprobabilities, using models of the data generatingprocess underlying the returns. ��

Also see: ?, ?.

5.2 Other Aspects of the

Risks

Q5.24: What are the risks that derive fromusage of derivatives?A: ? observe that because all derivative instru-ments are equivalent to combinations of existingsecurities, they cannot introduce any new or fun-damentally di�erent risks into the �nancial sys-tem. What derivatives do accomplish, however, isa facility for transferring these risks, and concen-trating their risk management into a few entities.A common classi�cation of the risks in the

derivatives area uses three areas (a) risks to indi-vidual users owing to mistakes in their positions,(b) risks to the clearinghouse owing to large mar-ket uctuations and (c) risks to the economy froma breakdown of all the markets in the country.��

Also see: ?.

1The terms \credit risk", \default risk", and \coun-terparty risk" all pertain to the same thing. \Settle-ment risk" and \Herstatt risk" are default risk on thedate of settlement. These terms do not hence implynew kinds of risk.

Q5.25: What are the policy issues in individ-ual risk?A: Examples like Barings fall into this �rst cate-gory of risks.

Individual users of markets will make mistakesfrom time to time. But this is not unusual { ona larger plane, companies make mistakes in man-ufacturing, or marketing, or in the relationshipswith banks, etc. This is one more kind of mistakethat companies can make. In the modern world, acertain collection of skills are required in order todo certain things, and people who lack those skillswill experience di�culties. There are umpteen ex-amples in India of people who have made losses inshort-selling. There is no direct role for regula-tion here; the role of regulation would be to as-sure that the market itself does not breakdown ina payments crisis.

At the level of an individual company, episodeslike Barings re ect \agency con icts", where em-ployees of the �rm fail to act in the interests ofthe owners (shareholders) of the company. Agencycon icts are a problem well-known to economists(?) and are rampant in the functioning of allsorts of companies. In all these situations, theeconomic challenge is one of creating an organi-sational structure which encourages employees toact in the interests of the organisation.

In the speci�c situation of Barings, several solu-tions (that should have been built into the bank'sbusiness plans) would have ensured that the in-dividual risk was reduced. One is the separationof trading and backo�ce functions (which was notdone at Barings) which would allow managementto get untainted information about the trading ac-tivity. Another is by setting position limits fortraders limits the size of the bets, and the worstpossible damage, that they can do. Compensat-ing traders based on long-term performance, in amanner which is sensitive to the risk adopted, issafer than paying bonuses for short-term tradinggains without regard for risk.

At a policy level, the most pragmatic approachis to create the markets, and then the skills willcome. There is no real incentive for �rms to grap-ple with these risks until derivatives markets arein India. For example, companies in India did notinvest in obtaining computer skills until comput-ers became available in India.

Once the markets arrive, top management willstart looking for consultants in risk management,

32 CHAPTER 5. REGULATORY ISSUES

J. P. Morgan's \Riskmetrics" software, notionslike \Value at Risk", etc. This entire process oflearning will begin once India has derivatives mar-kets. ��

Also see: ?.

Q5.26: What about systemic risk, or risks tothe economy?A: Systemic risk manifests itself when there is alarge and complex organisation of �nancial posi-tions in the economy. \Systemic risk" is said toarise when the failure of one big player or of oneclearing corporation somehow puts all other clear-ing corporations in the economy at risk.

At the simplest, suppose that an index arbi-trageur is long the index on one exchange andshort the futures on another exchange. Such a po-sition generates a mechanism for transmission offailure { the failure of one of the exchanges couldpossibly in uence the other.

Systemic risk also appears when very large po-sitions are taken on the OTC derivatives marketby any one player.

Neither of these scenarios are in the o�ng inIndia. Hence it is hard to visualise how exchange-traded derivatives could generate systemic risk inIndia. ��

Q5.27: How should the use of derivatives bymutual funds be regulated?A: Mutual funds are just one special case of \in-dividual risk" described in Question 5.25 above.Mutual funds could make mistakes in the securi-ties that they invest in, they could make mistakesin the way they interact with their investors, etc.Mistakes in derivatives trading is just one morekind of mistake that they can make.

Given free entry into the mutual fund industry,funds which are unable to cope with such complex-ity will go bankrupt, and funds which are good at�guring out this world will succeed. There is norole for some agency to protect mutual funds frommaking mistakes.There is, however, a useful role for disclosure.

Mutual funds should clearly show investors (in theprospectus) their planned policies about how theywould use derivatives. This will enable investorsto use the fund with knowledge. An analogy herewould be that a car manufacturer should be sup-plying complete information to the potential buyerof the car of the internal details of how the car

would work. ��

5.3 Market manipulation

Q5.28: What are the issues in terms of ma-nipulation in the context of derivatives mar-kets in general?A: In all areas, a basic fact about derivatives isthat they magnify the pro�t rate from manipu-lating the underlying. In other words, if there arepro�ts to be made from manipulating Gujarat Co-tex, then the gains to the manipulator would bemagni�ed if he had purchased call options on Gu-jarat Cotex in advance. ��

Q5.29: What are the issues on shortsqueezes?A: In the commodity markets area, a major con-cern is the \short squeeze", where a manipulatorknows the amount of physical goods which canpossibly show up for delivery and buys futurescontracts worth more than this oating stock (heoften tries to also buy the physical goods to reducethe oating stock. This is another sense in whichIndia's \cash market" for equity is actually a fu-tures market { we have seen short squeezes takingplace on the weekly futures market.

However, with cash settled derivatives (e.g.index-based contracts), this style of manipulationis not a threat. This is a very important di�erencebetween the traditional reasoning employed in thecontext of futures markets. Between 1874 (whenthe CBOT �rst started supplying the counterpartyguarantee) and the early 1980s (when cash settle-ment �rst appeared), the history of futures mar-kets has been pockmarked with short squeezes. Itis very important to observe that as India movesinto futures markets with the most modern styleof product (using cash settlement), this is a verydi�erent environment than that which has charac-terised a century of experience worldwide. ��

Also see: ?.

Q5.30: What kinds of manipulation arefound with index derivatives?A: Index derivatives are cash settled, hence theshort-squeeze style of manipulation is infeasible.

Typically, the index derivatives are far more

5.3. MARKET MANIPULATION 33

liquid than the underlying stocks. Manipulationwould hence work by

1. Adopting a position on the derivatives mar-ket and then

2. Trying to move the index in order to makethat position yield a pro�t.

At the policy level, dealing with this style ofmanipulation has two implications:

1. Understanding manipulation in the contextof index derivatives is synonymous with un-derstanding manipulation of the underlyingindex, and

2. The exchange which supplies spot priceswhich are used in an index calculation isthe place where attempts at manipulationwould take place, and it is the liquidityand surveillance procedures on this exchangewhich should be seen as a check against ma-nipuilation. In general, there is no reasonwhy the spot and the derivatives have totrade at the same location.

��

Q5.31: How do manipulators attack an in-dex?A: The reasoning of a pro�t-maximisingmanipula-tor leads him to focus on stocks which have a high

weight in the index but have poor liquidity. Thiswould obtain the maximum change in the indexper unit of capital deployed into manipulation.To cite an example, if a manipulator has Rs.1

million of capital, it makes no sense for him tospend that on trying to a�ect the price of StateBank (a highly liquid stock, where a purchase ofRs.500 million would typically move the price byless than 1%). Instead, that money is much betterspend on Hindustan Lever (a less liquid stock).The best stocks to target would be those whereliquidity is low and weight in the index is high.A manipulator would choose those index stocks

where the number w � i is the highest, where w =weight of the security in the in index (in percent),and i = impact cost (in percent). If impact cost ishard to measure, then stocks with large values ofw=s would be used, where s = bid-ask spread (inpercent). These formulas isolate vulnerabilities inthe form of largecap stocks which are illiquid. Ob-viously, these formulas would use data for liquidity(such as impact cost or the bid-ask spread) from

the exchange which supplies the prices which areused in calculating the index under question.

The basic index construction methodology ofan index like Nifty works via the impact cost seenin actual index purchases or sales of Rs.5 million.Hence this methodology e�ectively requires thatstocks should have liquidity in proportion to theirmarket capitalisation. This ensures that there areno unusually weak points for attack by a manipu-lator. ��

Q5.32: What kinds of manipulation can takeplace on security options?A: In options markets, manipulationwould consistof �rst adopting a position in the options market,and then trying to manipulate the underlying so asto obtain a good payo� from the options position.This is akin to the increased activity that takesplace to a�ect the 3:20pm Friday price in Calcuttain the context of their teji-mandi market.

Another style of manipulation involves optionswhich use physical delivery, and it is basically avariant of the short squeeze. The manipulator be-comes long on call options to the tune of moreshares than can be obtained for physical delivery.This would lead to a skyrocketing of the price ofthe underlying, and hence of the call option price.

A useful policy for derivatives exchanges wouldbe something like this: security options marketsshould only be launched for securities which meeta rule such as \the security price should move byless than 0.5% upon purchases of Rs.0.5 crore".2

It should be noted that the liquidity of the under-lying stock would improve once security optionscome about, so that a security which meets sucha criterion would be likely to seen see the pricemovement upon purchases of Rs.0.5 crore drop toless than 0.5%. ��

Q5.33: Are individual securities in India liq-uid enough to support security options?A: The most liquid stocks in the country are. Theliquidity of State Bank and Reliance in one snap-shot of NSE's order book, taken from June 1996,are displayed in Figure 5.1. Here we see aroundRs.50 crore of shares of State Bank being sourcedwhile impact cost stays under 1%; it is the high-est liquidity available in India's equity market asof today. Many other securities which are in the

2This is basically the levels of liquidity seen inNifty.

34 CHAPTER 5. REGULATORY ISSUES

-400 -200 0 200 400

Quantity (Million Rs.)

-100

-50

0

50

100

Impa

ct C

ost

(bas

is p

oint

s)

RelianceState Bank

Figure 5.1: Impact cost on SBI and RIL at Various Transaction Sizes

NSE-50 index are also highly liquid. ��

Q5.34: What would concerns about manip-ulation imply for the sequencing of indexderivatives vs. security options?A: Given the concerns about market manipulationin India, the safer sequencing is to �rst have index-based contracts. ��

Q5.35: Would a slow launch of security op-tions harm the economy?A: The vast majority of trading volume in eq-uity derivatives worldwide lies in index deriva-tives. This suggests that the economy really �ndsindex -based contracts very valuable; usage of in-dex derivatives is very widespread, while usage ofsecurity options is restricted to a smaller set ofpeople.

Hence a a slow start for security options wouldhave smaller deleterious economic consequencesthan delays in availability of index derivatives.��

Q5.36: What spot market should supplyprices which are used for calculating payo�swith cash{settled security options?A: An important principle here is that regard-

less of where the options trade, the payo�s fromthe option should be calculated using the marketwhere the underlying is the most liquid. If we haveoptions on Reliance trading on an exchange whereReliance is illiquid, and if the payo�s from the op-tions are calculated using the cash market priceson that same exchange, then it would encouragemarket manipulation on that exchange.

Hence, at the level of individual securities, op-tions markets anywhere should only calculate pay-o�s using closing prices from an exchange whichmeets two conditions: (a) strong surveillance pro-cedures, and (b) it should have the highest liquid-ity in the country (i.e. it should have the lowestimpact cost at transactions of Rs.0.5 crore or so).

A facility for borrowing and lending of shareswill also greatly help reduce the risk of a shortsqueeze in the security options market. ��

Q5.37: To what extent are these issues aregulatory issue?A: Given the basic competitive market structureof derivatives markets, there are strong incentivesfor themarket to be careful about issues surround-ing manipulation. If investors su�er manipulationon one derivatives market, they will move theirorder ow out to other derivatives markets or toalternative avenues of investment.

5.4. IN SUMMARY 35

Policy analysis should adopt the frameworkthat the exchange is an entity that tries to at-tract order ow and maximise volumes. As longas this is the case, the aims of the exchange andthe needs of the investor are the same. Exchangesthat fail to cater to the interests of investors willlose order ow.The policy suggested in Question 5.36 above,

i.e. that \cash{settled options payo�s should becalculated with respect to the prices seen on themost liquid exchange" is an example of this prin-ciple. If (say) the Poona exchange tries to tradeoptions on State Bank, then there are two choices:to use the State Bank price from PSE or fromthe most liquid exchange (NSE). If the former isadopted, there is a greater risk of manipulation.This fear would serve to widen the spreads onthe PSE options market, and hence diminish theorder ow to that market. This would sponta-neously generate a strong pressure for PSE to re-de�ne their option contract de�nition to calculatepayo�s di�erently. There is little need for regula-tion to enter the picture.This is perhaps like the question about the

Maruti 800 being an unsafe car. To the extentthat the market for cars lacks entry barriers, thesafety of the Maruti 800 is not a regulatory con-cern: if consumers feel unsafe with the Maruti 800,they can always buy another car. ��

5.4 In Summary

Q5.38: What does this discussion translateto in terms of speci�c regulations in thederivatives area?A: Translating these abstract ideas into speci�cpolicy avenues:

1. SEBI should not allow any exchange to func-tion without a clearinghouse that guaran-tees the trade, and it should conduct inspec-tions to con�rm that margin payments arebeing calculated as claimed, and actually be-ing charged to members.

2. The clearinghouse must charge initial marginusing a portfolio approach to measuring risk.If the clearinghouse can move funds swiftlyenough, then it can be a true \initial mar-gin", alternatively it should be a \exposurelimit". If the payments system is slow, then

the initial margin should be correspondinglylarger.

3. SEBI should require exchanges to open uptheir entry criteria to the extent required sothat the pure seat price (devoid of physicalinfrastructure) drops to low levels.

4. SEBI should require that exchanges disclosecopious information about the trading (in-cluding things like open interest, the stan-dard deviation of member-positions, etc) onthe exchange. This information should befreely available in newspapers and on the In-ternet.

5. The surveillance department at SEBI shouldrequire, and possibly do an investigative fol-lowup on, reports of positions and tradingactivity of \large" players on the market.\Large" could possibly be de�ned as an openposition above Rs.100 crore, or a one-daytrading volume above Rs.50 crore. Theseare the traders who might command marketpower and possibly manipulate the market.

6. SEBI should be accessible to individual usersof the market who would be able to complainabout manipulative episodes where they havebeen hurt.

��

36 CHAPTER 5. REGULATORY ISSUES

Chapter 6

Derivatives and the Economy

Q6.1: What are the bene�ts of derivatives toIndia?A: India's �nancial market system will stronglybene�t from smoothly functioning index deriva-tives markets.

� Internationally, the launch of derivatives hasbeen associated with substantial improve-ments in market quality on the underlyingequity market. Liquidity and market e�-ciency on India's equity market will improveonce the derivatives commence trading.

� Many risks in the �nancial markets can beeliminated by diversi�cation. Index deriva-tives are special insofar as they can be usedby investors to protect themselves from theone risk in the equity market that cannot bediversi�ed away, i.e., a fall in the market in-dex. Once investors use index derivatives,they will su�er less when uctuations in themarket index take place.

� Foreign investors coming into India would bemore comfortable if the hedging vehicles rou-tinely used by them worldwide are availableto them. See Question 4.9 for more detailshere.

� The launch of derivatives is a logical nextstep in the development of human capitalin India. Skills in the �nancial sector havegrown tremendously in the last few years,thanks to the structural changes in the mar-ket, and the economy is now ripe for deriva-tives as the next area for addition of skills.

Once India has skills in the core derivativesmarkets, capabilities in derivatives can be easily

applied into unexpected areas. The world over, in-novative contracts such as pollution permits, elec-tricity prices, garment quotas, etc, are being usedto solve economic problems. Each of these mar-kets is small when taken in isolation, but has atremendous impact upon the speci�c area. Butprogress in these directions �rst requires a corecapability on the part of exchanges and traders inmainstream �nancial derivatives. ��

Q6.2: How do index derivatives change theoverall level of equity investment in the econ-omy?A: We can think of the overall economy mak-ing choices about debt vs. equity, based on deci-sions by households and �rms. In this big picture,derivatives have no role. Recall that all derivativesare in net zero supply: when one person leveragesby going long on index futures, there is an equaland opposite counterparty who is deleveraging bygoing short index futures. If either of a short or along are unavailable, no trade takes place.

In this sense, when aggregating at the level

of the full economy, derivatives are unimportant.Where derivatives do help, however, is in allowingthe repackaging and movement of risk from peo-ple who do not want to bear it to the people whoare willing to bear it. Derivatives allow many eco-nomic agents in the economy to sell insurance toothers, and the availability of this insurance en-ables many economic activities without which therisks would be too high. ��

Q6.3: How will index derivatives assist capi-tal formation and growth in the economy?

A: At the larger level of the economy, well-functioning derivatives markets will improve themarket e�ciency of the underlying cash market.

37

38 CHAPTER 6. DERIVATIVES AND THE ECONOMY

It will improve the market's ability to carefully di-rect resources towards the projects and industrieswhere the rate of return is highest; this will im-prove the allocative e�ciency of the market. Byimproving the allocative e�ciency, a given stockof investible funds will be better used in procuringthe highest possible GDP growth for the country.

Hence the real linkages go (a) from derivativesto market liquidity and market e�ciency, and (b)from market e�ciency to GDP growth. ��

Q6.4: What is the evidence about derivativesand market liquidity and e�ciency?

A: There is strong empirical evidence from othercountries that after derivatives markets have comeabout, the liquidity and market e�ciency of theunderlying market have improved. ��

Q6.5: What is the evidence about marketquality and economic growth?

A: An important study of the relationship be-tween stock market development and long-termeconomic growth has been recently conducted byRoss Levine and Sara Zervos (1996). They cre-ate a measure of stock market development whichcombines three dimensions of market quality. Bythis measure, a country is said to have a well-developed stock market when (a) the assets inter-mediated by the stock market are large when com-pared with GDP (b) the stock market is highly liq-uid, and (c) the stock market is highly integratedinto world markets.

Their empirical analysis controls for the inde-pendent contribution of seven kinds of other fac-tors. After taking into account the contribution ofall these factors, they �nd that stock market de-velopment is highly signi�cant statistically in fore-casting future growth of per capita GDP. Theirregressions imply that stock market developmentis also highly economically signi�cant. For exam-ple, their regressions forecast that if Mexico orBrazil were to obtain stock markets as advancedas Malaysia, then they might obtain an additionalper capita GDP growth per year of 1.6 percent-age points. Even allowing for the imprecision andhazards of such extrapolation, this is an extremelylarge e�ect. ��

Also see: ?, ?.

Q6.6: How do derivatives alter the exposureof di�erent people in the economy?A: Derivatives allow a shifting of risk from a per-son who does not want to bear the risk to a personwho wants to bear the risk. Without derivatives,people su�er risk without much choice. The onlyinvestment decision that can be made is whetherto be in a certain area of business or not. For ex-ample, if a garment exporter dislikes currency risk,the only choice that he faces (in a world beforederivatives) is whether to be in garment exportor not. With derivatives, he has the ability andchoice to insure against currency exposure. Andhe is able to do this by trading this exposure withothers in the economy who are equipped to dealwith it. ��

Q6.7: How do derivatives alter the informa-tional structure of the economy?A: Both futures and options markets have a sig-ni�cant impact upon the informational e�ciencyof �nancial markets.In the case of futures:

1. The simplest and most direct e�ect is thatthe launch of a derivatives market is cor-related with improvements in market e�-ciency in the underlying market. This im-proved market e�ciency means that the mar-ket prices of individual securities are moreinformative.

2. Once futures markets appear, a certaindelinking of roles in the two markets is ob-served. The cash market caters to relativelynon-speculative orders, and the futures mar-kets takes over the major brunt of price dis-covery. The futures market is better suitedfor this role, because of high liquidity andleverage. Whenever news strikes, it �rst ap-pears as a shock in the futures market prices,which arbitrage then carries into the cashmarket.

3. Another unique feature applies for the mar-

ket index. In todays economy, speculationon the level of the index is di�cult, becausea tradeable index does not exist. Hence in-formed speculators might try to take posi-tions on individual securities in order to im-plement views about the index, but this isdi�cult because of higher transactions costs.

39

Index futures will hence improve the infor-mational quality of the market index.

In the case of options,

1. Options are important to the market e�-ciency of the underlying in much the sameway that futures are important,

2. In addition, options play one unique role ofrevealing the market's perception of volatil-

ity. High-quality volatility forecasts have se-rious rami�cations for decisions in portfoliooptimisation, production planning, physicalinvestment decisions, etc.

By using the option price in the market, itis possible to infer the market's consensusview about volatility through a simple for-mula. This is a completely unique role thatoptions play, that neither the cash marketnor the futures market can possibly play.

This is a very important reason why securityoptions are important. If options on TISCOexisted, the entire market would be able toobserve the price of options on the market,and infer a very good forecast about volatilityon TISCO in the coming weeks and months.

��

Also see: ?.