Get documents about "cftc
Document Sample
THE SPECULATIVE AND HEDGING STRUCTURE OF FINANCIAL FUTURES CONTRACTS
Robert T. Daigler
Florida International University
Miami, Florida
Visiting Scholar, 1990-91
Graduate School of Business
Stanford University
Paper for the Southwestern Finance meetings
Houston, Texas
March 1991
1
THE SPECULATIVE AND HEDGING STRUCTURE OF FINANCIAL FUTURES CONTRACTS
I. Introduction
One of the most persistent controversies regarding futures markets is
whether these markets promote speculation or if they are hedging markets. While
famous situations concerning onions, potatoes and silver have fueled the
speculative controversy of futures, trading in stock index futures has created
the most disagreement in the financial community concerning the motives of the
participants in futures markets. This paper investigates various hypothesis
regarding the speculative and hedging nature of financial futures by employing
the CFTC's Commitment of Traders data. Stock index and interest rate futures
contracts are examined by using various measures and indices of speculation and
hedging activity.
Determining whether financial futures markets are speculative or hedging
in nature is important for regulatory policy, hedgers choices, and the attitude
of the financial community concerning futures. The CFTC employs hedging
effectiveness as a primary criteria in deciding whether a new futures contract
should be traded. Hedgers desire a market that has depth and stability, i.e. one
that will not be unduly influenced by "speculator's whims". The attitude of the
financial community determines the extent that futures markets are used and the
community's influence over the regulation of futures. 1
The following sections of this paper discuss the hypotheses concerning
speculative and hedging positions in financial futures markets, the data and
results, and the conclusions. The results show that financial futures are
primarily hedging markets.
1
One aspect of the jurisdictional dispute between the CFTC
and the SEC over the control of stock index futures concerns
margins. The SEC would like to increase margins on stock index
futures, which is a direct result of the critics contention that
futures cause the cash stock market to be more volatile. This
contention is directly related to the "speculative influence" of
stock index futures.
2
II. Hypotheses and Methodology
A. Hypotheses for Hedging and Speculative Activity
The existence of monthly open interest broken into speculative and hedging
positions provides the opportunity to examine several hypotheses concerning the
structure of futures markets. Alternative hypotheses are not explicitly stated
unless needed for clarification. Many of these hypotheses tested relate to a
"successful" futures contract; successful is defined here in terms of the size
of the open interest of the contract. 2
Hypothesis 1: Financial futures markets are primarily hedging markets.
Hypothesis 1A: Financial futures markets are primarily speculative markets.
The CFTC criteria for trading a new futures contracts lists hedging as one of the
two primary characteristics needed for a futures contract. 3 More importantly,
a key controversy concerning futures markets is whether they are speculative or
hedging vehicles.
Hypothesis 2: A high degree of hedging activity is a necessary condition
for a successful futures contract.
Hypothesis 2A: A high degree of hedging activity is a sufficient condition
for a successful futures contract.
The characteristics of a successful futures contract are of vital interest to the
exchanges as well as to market participants. Market wisdom postulates that
contract specifications are developed in order to encourage hedging activity.
For example, the demise of various versions of the GNMA futures are blamed on
changing industry needs for hedging. Other studies on contract success have
attempted to explain why some contracts become popular while others fail; these
studies have been largely unsuccessful in determining the key characteristics of
2
While trading volume is another candidate to measure
success, volume is affected by scalper activity in the pits, and
therefore is more of an artificial value. For example, when the
CBT, CMEX and NYFE all had Certificate of Deposit contracts trading
simultaneously, the scalpers on CBT generated a large volume trades
among themselves to create the impression that the market was
active.
3
The other characteristic is price discovery.
3
success.
Hypothesis 3: Net hedging of zero is needed for a successful futures
contract.
Hypothesis 4: Speculative activity is needed to have a successful futures
contract.
While speculation may create uncertainty and volatility of prices due to the
short time horizon of speculators, 4 speculation is needed to span the timing
differences between short and long hedgers. 5 A balance is sought between
excessive speculation and "insufficient" speculation in order to create an
orderly market. However, in the longer-run the effect of speculation can be
minimized by having short and long hedging balance out, which is the motivation
for hypothesis 3.
The concentration of open interest among a few traders is an indication of
a narrow market, while a broad market would have a low concentration of open
interest.
Hypothesis 5: Concentration of trading power will cause a contract to be
unsuccessful.
The degree of concentration of market power in a few traders hands is viewed as
detrimental to a futures contract, since it signals the possibility of a squeeze
or similar control of the market. At the very least, adverse price anomalies may
occur with a concentration of trading power, especially during delivery periods.
Hypotheses concerning the relative degree of hedging or speculation across
markets also may be posited. In particular, stock index futures have been
criticized for being more speculative than other futures markets, which in turn
would justify higher margins and additional regulations.
Hypothesis 6: Stock index futures are more speculative than interest rate
4
The average holding time for a speculator is estimated to be
two to three weeks.
5
Scalpers and day traders in the pit provide very short term
liquidity between buyers and sellers. However, scalpers and day
traders do not take positions overnight. Therefore, longer term
speculators are needed.
4
futures markets.
B. Methodology
The importance of hedging and speculative activity for a given futures
market can be measured in several ways. The simplest method is to determine the
proportions of the total open interest positions held by hedgers and speculators.
Examining the net hedging balance between short and long hedgers is a measure of
the long-term hedging stability of the market. In other words, speculators may
provide a bridge between temporary long and short hedging imbalances, but the
matching of long and short hedging positions is needed for long-term stability.
Speculative and hedging ratios and the speculative index provide additional
measures of the balance in the market. The objective of these measures is to
determine the extent of any excess speculation in the market in relation to the
amount of hedging. The speculative ratio is defined as:
SR = SL/HS if HS>HL
or = SS/HL if HL>HS (1)
The hedging ratio is determined by:
HR = HL/HS if HS>HL
or = HS/HL if HL>HS
Where:
SR = the speculative ratio
HR = the hedging ratio
SL = the amount of long speculation
SS = the amount of short speculation
HL = the amount of long hedging
HS = the amount of short hedging.
The speculative and hedging ratios are measures used to determine how the
larger component of the long/short hedging category is offset. For example, when
short hedging is greater than long hedging then the speculative ratio examines
the ratio of long speculation to short hedging, while the hedging ratio examines
the ratio of long hedging to short hedging. When short and long hedging are not
5
balanced then speculation must create the needed balance. Thus, when HS>HL then
long speculation is the balancing category; to the extent that excess long
speculation exists then additional short speculation must occur to create an
overall balance between the long and short open interest figures. 6 Consequently,
the speculative ratio examines the relationship between the dominant speculative
and hedging long/short categories. Similarly, the hedging ratio measures the
extent of the balance between the short and long hedging open interest.
The speculative index is an additional measure of the relationship between
speculation and hedging. The speculative index, as first stated by Working
(1960), is defined as:
SI = 1 + SS/(HL + HS) if HS>HL
or = 1 + SL/(HS + HL) if HL>HS (3)
Where:
SI = the speculative index.
The speculative index concentrates on the proportion of speculation that exists
which is not needed to balance net hedging. Thus, when short hedging dominates
then short speculation is not needed to make the market function or to create a
net hedging balance. Such short speculation necessitates additional long
speculation to balance the market. Thus, the speculative index provides a
measure of the amount of excess speculation in percentage terms. Large values
indicate a speculative market and small values a balanced market.
III. Data and Results
A. Data
The CFTC Commitments to Traders (1983-88) is employed to obtain month end
open interest totals for speculators (noncommercial traders), hedgers (commercial
6
The balancing open interest may be either in the large
trader or the non-reporting categories. This distinction is
discussed in connection with the data sources.
6
traders), spreaders, and non-reporting traders. 7 The CFTC data is for "reporting
traders" only, i.e. it is restricted to traders with positions above a certain
minimum reporting position. Therefore, the analysis conducted here may either
be considered as an examination of large trader positions or an analysis with the
assumption that the non-reporting traders have the same hedging/speculative
ratios as the reporting traders. The data also includes information on the
concentration of total open interest within the eight largest short and long
traders in each futures contract.
The contracts examined in this study are stock index futures, long-term
interest rate futures, and short-term interest rate futures. The stock index
futures are the S&P 500, NYSE, MMI Maxi and Mini, Value Line, and S&P 100
contracts. The long-term interest rate contracts are the T-bond, Ten Year bond,
GNMA, and Municipal bond contracts. The short-term interest rate contracts are
the Eurodollar, T-bill, and Certificate of Deposit contracts. 8
B. Results
Table 1 presents the stock index, short-term and long-term interest rate
futures contracts ranked by average monthly open interest. The open interest
figures are then broken into the percentage holdings for the speculative,
hedging, and non-reporting categories. Table 2 presents the values of the
measures of hedging and speculation which were described in Part II. Table 3
shows the concentration of market activity for the eight largest short and long
traders in each market. Results from these tables are used to test the six
7
Spreading positions are small and therefore are not
considered in the analysis. The commercial category is defined as
traders holding offsetting cash positions. Thus, such traders
could be hedgers, arbitrageurs, or dealers. The generic name of
hedgers is given here to describe the commercial category.
Individual small customers of brokerage houses, FCMs, etc. are
considered to be individual non-reporting traders.
8
Since both the long and the short open interest categories
equal the total open interest, the proportions employed in the
analysis add to 200%.
7
hypotheses stated previously.
Hypotheses 1 and 1A states that futures markets are primarily hedging or
speculative markets, respectively. Comparing the commercial and speculative
categories in table 1 shows that these markets are primarily hedging markets,
since all the contracts except for the NYFE, Value Line, and S&P 100 futures have
a larger hedging percentage than speculative percentage; in fact, the hedging
percentage is typically significantly larger than the speculative percentage.
Moreover, even if the entire nonreporting category is added to the speculative
category, only the S&P 500 and MMI Mini contracts would be added to the
speculative majority list, and these two contracts would only be marginally
speculative.
To examine whether a high degree of hedging activity is a necessary and/or
sufficient condition for a successful futures contract (hypotheses 2 and 2A),
Spearman rank correlations are calculated between the ranks of the values of the
various measures of hedging activity and the ranks of the size of the open
interest. 9 The Spearman rank correlation between open interest and commercial
activity is .76, while the rank correlation between open interest and the hedging
ratio in table 2 is .82. Also, the speculative index rank correlation with open
interest is .72, where a low speculative index (high hedging index) receives a
rank consistent with a large open interest. Thus, all of these correlations are
consistent with the hypothesis that hedging is a necessary condition for having
a successful futures contract. However, a high degree of hedging is not a
sufficient condition for a successful contract, since the rank correlations are
not equal to one and the failed GNMA and CD contracts have a high degree of
hedging.
Hypothesis 3 states that net hedging of zero is needed for a successful
futures contract. The rank correlation in table 2 for the net hedging results
of .55 shows that there is a significant correlation between a low net hedging
value and open interest, although this is the smallest rank correlation for all
9
Correlations between the absolute size of the open interest values and the
various measures is misleading because of the outlier effects of the very large
and very small open interest contracts.
8
of the categories examined.
Hypothesis 4 states that speculative activity is needed for a successful
contract. The rank correlations between open interest and the speculative
proportions in table 1, the speculative ratio in table 2, and the speculative
index in table 2 clearly show that larger speculative activity is strongly
negatively correlated with open interest. 10 Thus, speculation is detrimental to
a successful futures contract.
The fifth hypothesis states that a concentration of trading power is
detrimental to a futures contract becoming successful. Table 3 shows a -.67 rank
correlation between open interest and the percentage of open interest controlled
by the eight largest long and short traders. This shows that a large
concentration of trading power is detrimental to a successful futures contract.
These results imply that smaller traders do not want to be involved in a market
dominated by a few large players. Table 3 also shows that the Certificate of
Deposit, MMI Mini, Municipal bond, T-bond and Ten Year T-bond were out of balance
between the long and short large traders by at least 10%. 11
The sixth hypothesis states that stock index futures are more speculative
than other financial futures contracts. Ranking of the non-commercial
proportions in table 1, the speculative ratio in table 2, and the speculative
index in table 2 substantiates this hypothesis, since the stock index futures
rankings represent the six most speculative contracts by the first two measures,
and six of the seven largest speculative index values. While the most
speculative stock index contracts either no longer trade or trade with very low
volume, even the S&P 500 futures is more speculative than any of the interest
rate futures contacts.
10
The speculative index is ranked from the smallest value to the largest
value, i.e. from the lowest speculative activity to the largest. Thus, the
positive rank correlation showed that large open interest values are correlated
with small levels of the speculative index.
11
In fact, the large long CD futures traders had over 30% more of the open
interest than the shorts. This is interesting since it was claimed that large
commercial banks were dominating this market with significant short positions.
9
IV. Conclusions
CFTC Commitments of Traders data shows that financial futures are primarily
hedging markets and that the degree of success of a financial futures contract
is directly related to hedging measures and inversely related to speculative
measures. Thus, hedging is a necessary condition for a successful market;
however, hedging is not a sufficient condition for success since futures that
were primarily hedging contracts have failed. These results are supported by
various measures of hedging activity and rank correlation results. In addition,
a contract with a high concentration of trading power among a few traders tends
to be less successful, and stock index futures are more speculative contracts
than interest rate futures. These results suggest that regulators and the
financial community need not be concerned about speculative influences for the
successful financial futures contracts. Moreover, unsuccessful contracts have
the depth and participant stability wanted by hedgers. Finally, additional
restrictions or margin increases for stock index futures do not seem warranted
for the contracts currently trading.
10
BIBLIOGRAPHY
Anne E. Peck (1980a), "The Influence of Hedging on Futures Markets Activity: Some
Further Evidence," International Futures Trading Seminar: Proceedings,
Vol. VII, May 1980, pp. 1-23.
Anne E. Peck (1980b), "Measures and Price Effects of Changes in Speculation on
the Wheat, Corn, and Soybean Futures Markets," Research on Speculation:
Seminar Report, The Chicago Board of Trade, November 1980, pp. 138-149.
Commodity Futures Trading Commission (1983-88), "Commitments of Traders in
Commodity Futures," monthly, 1983-88.
Holbrook Working (1960), "Speculation on Hedging Markets," Food Research
Institute Studies, Vol. I, No. 2.
11
TABLE 1
Proportions of Trades by Category
CONTRACT OPEN SPECULATIVE HEDGING NON-
INTEREST REPORTING
T-BOND 238722 0.157 1.127 0.562
EURODOLLAR 181909 0.105 1.290 0.523
S&P 500 75300 0.300 0.959 0.704
10 YEAR BOND 50400 0.165 1.465 0.314
TBILL 36933 0.244 1.078 0.615
GNMA 25266 0.119 1.229 0.520
CD 14649 0.247 1.142 0.485
MUNI BOND 11874 0.316 1.103 0.544
NYSE 10204 0.507 0.410 0.809
MMI MINI 9252 0.492 0.854 0.554
VALUE LINE 6114 0.436 0.345 1.147
MMI MAXI 5742 0.370 1.053 0.529
S&P 100 2152 1.046 0.212 0.726
SPEARMAN RANK -.824 .764
CORRELATION
12
TABLE 2
Hedging and Speculative Measures
CONTRACT NET HEDGING SPECULATIVE HEDGING SPECULATIVE
(ABSOLUTE) RATIO RATIO INDEX
T-BOND 0.099 0.165 0.839 1.051
EURODOLLAR 0.060 0.099 0.916 1.028
S&P 500 0.086 0.432 0.811 1.193
10 YR. BOND 0.100 0.141 0.873 1.039
TBILL 0.196 0.279 0.709 1.076
GNMA 0.096 0.115 0.856 1.037
CD 0.205 0.172 0.696 1.127
MUNI 0.349 0.128 0.532 1.059
NYSE 0.139 1.181 0.539 1.685
MMI MINI 0.216 0.671 0.582 1.710
VALUE LINE 0.195 2.987 0.329 3.346
MMI MAXI 0.195 0.437 0.708 1.121
S&P 100 0.133 4.426 0.225 4.423
SPEARMAN RANK 0.555 -.720 .819 .720
CORRELATION
13
TABLE 3
Concentration of Trading Power
CONTRACT EIGHT EIGHT EIGHT
LARGEST LARGEST LARGEST:
LONGS (%) SHORTS (%) AVERAGE (%)
T-BOND 37.7 27.1 32.42
EURODOLLAR 36.3 38.9 37.61
S&P 500 35.3 36.5 35.87
10 YR. BOND 63.2 49.0 56.10
TBILL 34.5 43.2 38.85
GNMA 58.5 54.6 56.53
CD 76.2 43.4 59.79
MUNI 45.2 59.0 52.08
NYSE 47.9 53.5 50.69
MMI MINI 67.2 63.5 65.35
VALUE LINE 29.3 46.0 37.67
MMI MAXI 64.8 65.4 65.08
S&P 100 61.5 53.9 57.68
SPEARMAN RANK -.670
CORRELATION
