Passive Investors and Managed Money in Commodity Futures. Part 2: Liquidity. Prepared for: The CME Group. Prepared by:

Transcription

1 Passive Investors and Managed Money in Commodity Futures Part 2: Liquidity Prepared for: The CME Group Prepared by: October, 2008

2 Table of Contents Section Slide Number Objectives and Approach 3 Findings Corn 4-12 Soybeans Chicago Wheat KC Wheat MN Wheat Cotton Natural Gas Crude Oil Summary

3 Objectives and Approach! The objective of this section is to examine the association, if it exists, between the market presence of passive and trend-following traders and liquidity in the studied markets.! The liquidity measure is plotted in time-series fashion to identify recent trends. Growth in volume and open interest is also presented.! A series of scatter diagrams are used to illustrate the relationship between market presence of each trader group and average liquidity.! For the purposes of this work, liquidity is measured as trading volume as a percentage of open interest during the last 200 trading days. 3

4 Findings - Corn! Overall, there has been at least a moderate increase in average liquidity for Corn contracts over the study period.! In 2005 and 2006, liquidity generally averaged between 13% and 19%.! From the beginning of 2007 onward, average liquidity per contract consistently ranges between 17% and 23%.! Liquidity is seasonal, being lowest for the September contract and highest for the May contract. 4

5 Findings - Corn 24% VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 22% 20% PERCENT (%) 18% 16% 14% 12% 10% MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL CONTRACT - YEAR 5

6 650% 550% 450% 350% 250% 150% 50% 650% 550% 450% 350% 250% 150% 50% 6 Avg. Daily Volume As Pct of Open Interest as Pct of 2002 Volume and Open Interest Growth - Corn Avg Daily Volume Avg Open Interest

7 Findings - Corn! The correlation between the presence of Non- Commercials and liquidity in the studied corn contracts is the strongest and it is positive.! Correlations between the presence of Commercials and Indexers and liquidity were much weaker and negative.! Interestingly, liquidity appears to decline whenever small traders make up a larger percentage of the market. 7

8 Findings - Corn 0.30 Commercials, Corn 0.28 y = -0.11x R 2 = N7 Z7 Z5 N8 H7 K5 Z6 N5 K8 H6 N6 K6 K7 H8 U H5 U6 U % 35% 40% 45% 50% 55% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 8

9 Findings - Corn 0.24 Non-Commercials, Corn 0.22 y = 0.33x R 2 = 0.34 N8 K7 N7 K H K5 Z6 N5 Z5H6 Z7 N6 K6 U7 H U5 H5 U % 15% 20% 25% 30% 35% 9

10 Findings - Corn Indexers, Corn y = -0.30x R 2 = 0.12 K7 N7 K8 N N6 H8 H7 U7 K6 Z7 H6 Z6 N5 K5 Z U6 H5 U % 2% 4% 6% 8% 10% 12% 14% 16% 18% 10

11 Findings - Corn Money Managers, Corn 0.24 K8 y = 0.13x R 2 = 0.04 N K7 N K6 H8 N6 H6 U7 H7 N5 Z5 K5 Z6 Z U5 U6 H % 7% 12% 17% 22% 11

12 Findings - Corn Small Traders, Corn N8 K8 K7 N7 y = -0.23x R 2 = H H8 U7 K5 Z6 Z5 N5 Z7 K6 H6 N U5 H5 U % 17% 19% 21% 23% 25% 27% 29% 31% 33% 35% 12

13 Findings - Soybeans! There doesn t appear to be a consistent, overall trend in average liquidity for Soybeans contracts over the study period.! Average liquidity per contract seems to have declined from the early to mid-2005 through mid and stabilized through 2007 before showing signs of an increase in 2008.! Overall, though, average liquidity has bounced between 22% and 36% per contract over the study period. 13

14 Findings - Soybeans VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 40% 35% 30% PERCENT (% 25% 20% 15% 10% JAN MAR MAY JUL AUG SEP NOV JAN MAR MAY JUL AUG SEP NOV JAN MAR MAY JUL AUG SEP NOV JAN MAR MAY JUL CONTRACT - YEAR 14

15 400% 350% 300% 250% 200% 150% 100% 50% 400% 350% 300% 250% 200% 150% 100% 50% 15 Avg. Daily Volume As Pct of Open Interest as Pct of 2002 Volume and Open Interest Growth - Soybeans Avg Daily Volume Avg Open Interest

16 Findings - Soybeans! Correlation between the presence of any of the large trader groups and average liquidity was very weak for soybean futures.! The data do not indicate a strong correlation between the presence of either Indexers or Money Managers and liquidity.! The presence of small traders is negatively associated with liquidity. 16

17 Findings - Soybeans Commercials, Soybeans y = 0.12x R 2 = 0.05 X6 F7 N8 H5 X7 H7 F6 X5 Q7 F8 N7 N6 Q6 K5 N5 H8 U5 H6 U7 K7 Q5 K6 K U % 30% 35% 40% 45% 50% 55% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 17

18 Findings - Soybeans N8 H5K5 Non-Commercials, Soybeans Q5 K8 H8 y = -0.12x R 2 = X5 F7 F6 X6 H6 H7 N5 X7 K7 K6 F8 N7 N6 U5 Q6 U7 Q U % 15% 20% 25% 30% 35% 18

19 Findings - Soybeans Indexers, Soybeans Q5 U5 Q7 U7 Q6 H7 N8 K8 H8 F7 H6 F6 K7 N7 X7 K6 N6 F8 X6 y = 0.27x R 2 = 0.06 H5 X5 N5 K U % 2% 4% 6% 8% 10% 12% 19

20 Findings - Soybeans Money Managers, Soybeans 0.35 N Q5 K8 H8 K5 H7 F7 X5 U5 H6F6 N5 U7Q7 K7 F8 Q6 K6 N6 H5 X6 X7 y = 0.16x R 2 = 0.10 N U % 7% 12% 17% 22% 27% 32% 20

21 Findings - Soybeans 0.35 N8 Small Traders, Soybeans K8 H8 K5 Q5 H5 y = -0.23x R 2 = N7 U7 N5 K6 U5 K7 X7 H6 Q7 F8 N6 X5 Q6 H7 F7 F6 X U % 15% 20% 25% 30% 35% 40% 45% 21

22 Findings Chicago Wheat! Average liquidity has increased over time in Chicago Wheat futures.! Average liquidity per contract moved between 14% and 22% during 2005 and 2006.! From early 2007 on, average liquidity has consistently been above 20% and peaked as high as 28%. 22

23 Findings Chicago Wheat VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 30% 28% 26% 24% PERCENT (%) 22% 20% 18% 16% 14% 12% 10% MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL CONTRACT - YEAR 23

24 Volume and Open Interest Growth Chicago Wheat 500% 450% Avg Daily Volume Avg Open Interest 500% 450% 400% 400% 350% 350% 300% 300% 250% 250% 200% 200% 150% 150% 100% 100% 50% 50% 24 Avg. Daily Volume As Pct of Open Interest as Pct of 2002

25 Findings Chicago Wheat! The presence of each of the major trading groups exhibits an extremely weak correlation to average liquidity for Chicago Wheat futures.! There is not enough evidence in the charts on the following pages to suggest that the market presence of any trading group has even a modest impact on liquidity. 25

26 Findings Chicago Wheat y = 0.06x R 2 = 0.01 Commercials, Chicago Wheat K8 H U H5 Z6 K5 N5 K7 H7 K6 N6 N8 U5 N7 Z U6 Z5 H % 25% 30% 35% 40% 45% 50% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 26

27 Findings Chicago Wheat y = -0.09x R 2 = 0.02 Non-Commercials, Chicago Wheat K8 H U N8 N7 Z6H5 Z7 N6 N5 H7 K5 K7 U5 K H6 Z5 U % 10% 15% 20% 25% 30% 27

28 Findings Chicago Wheat Indexers, Chicago Wheat y = -0.14x R 2 = 0.05 K H K7 N7 N8 N6 H7 K6 U7 Z7 U5 N5 H5 Z6 K H6 U6 Z % 10% 15% 20% 25% 30% 35% 28

29 Findings Chicago Wheat Money Managers, Chicago Wheat 0.28 y = 0.27x K R 2 = 0.09 H U U5 K6 Z6 K7 K5 N5 N6 H7 Z7 N7 H5 N U6 H6 Z % 7% 9% 11% 13% 15% 17% 19% 21% 23% 25% 29

30 Findings Chicago Wheat Small Traders, Chicago Wheat 0.28 K8 y = 0.04x R 2 = 0.00 H U Z7 K5 U5 N8N5 K6 N7 K7 H5 H7 Z6 N Z5 H6 U % 12% 14% 16% 18% 20% 22% 24% 26% 28% 30% 30

31 Findings Kansas City Wheat! For Kansas City Wheat, average liquidity per contract has shown a rather strong down-trend over the study period.! In 2005 and 2006, average liquidity was consistently 15% or higher, peaking around 20%.! That steadily declined over time, and liquidity for recent Kansas City Wheat contracts has consistently been below 10%.! Recently, volume has declined faster than open interest. 31

32 Findings Kansas City Wheat VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 25% 20% PERCENT (%) 15% 10% 5% 0% MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL CONTRACT - YEAR 32

33 Volume and Open Interest Growth Kansas City Wheat 300% 300% Avg Daily Volume Avg Open Interest 250% 250% 200% 150% 100% 200% 150% 100% 50% 50% 0% 0% 33 Avg. Daily Volume As Pct of Open Interest as Pct of 2002

34 Findings Kansas City Wheat! Although the correlation is not very strong (Rsquared = 0.27), the data suggests that the presence of Indexers may have a positive impact on liquidity in the Kansas City Wheat futures market.! Interestingly, the presence of Money Managers may have a very modest but negative impact on liquidity.! There are no strong or definitive patterns between market presence and liquidity for the Commercial, Non-Commercial and Small Trader groups. 34

35 Findings Kansas City Wheat y = 0.11x R 2 = 0.04 Commercials, Kansas City Wheat U5 K5 N K7 U6 H7 Z6 H5 N6 Z5 H6 K N7 U7 K Z7 H N8 35% 40% 45% 50% 55% 60% 65% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 35

36 Findings Kansas City Wheat y = 0.07x R 2 = 0.01 Non-Commercials, Kansas City Wheat U5 K5 N H5 K6 N6 Z H6 Z6 H7 K7 U N7 K8 U H8 Z N8 5% 10% 15% 20% 25% 30% 36

37 Findings Kansas City Wheat Indexers, Kansas City Wheat y = 0.86x R 2 = 0.27 U5 N5 K Z6 N6 U6 H6 Z5 K6 K7 H7 H U7 N7 K Z7 H N8 2% 4% 6% 8% 10% 12% 14% 37

38 Findings Kansas City Wheat Money Managers, Kansas City Wheat U5 y = -0.44x K5 N R 2 = 0.23 K6 Z5 H6 U6 N6 H7 K7 H5 Z U7 K8 N H8 Z7 N % 7% 9% 11% 13% 15% 17% 19% 21% 23% 38

39 Findings Kansas City Wheat 0.20 Small Traders, Kansas City Wheat 0.18 y = -0.23x R 2 = 0.06 K5 U5 N H5 K6 Z5 N H6 U6 Z6 H7 K U7 K8 N Z7 H8 N % 17% 19% 21% 23% 25% 27% 29% 31% 33% 35% 39

40 Findings Minneapolis Wheat! Average liquidity has varied widely over the study period for the Minneapolis Wheat futures market, ranging anywhere from 11% to 20%.! Although liquidity has varied more in recent months, the general trend has been toward modestly lower liquidity.! Significant seasonality is present, with liquidity routinely peaking in the May contract.! Recently volume has fallen faster than open interest, causing the liquidity metric to decline. 40

41 Findings Minneapolis Wheat VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 20% 19% 18% 17% PERCENT (%) 16% 15% 14% 13% 12% 11% 10% MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL SEP DEC MAR MAY JUL CONTRACT - YEAR 41

42 Volume and Open Interest Growth Minneapolis Wheat 300% 300% Avg Daily Volume Avg Open Interest 250% 250% 200% 150% 100% 200% 150% 100% 50% 50% 0% 0% 42 Avg. Daily Volume As Pct of Open Interest as Pct of 2002

43 Findings Minneapolis Wheat! The following charts indicate that any correlation between the market presence of each trading group and liquidity in Minneapolis Wheat futures is extremely weak.! Indexers are conspicuously absent in this futures market. 43

44 Findings Minneapolis Wheat y = 0.04x R 2 = 0.02 K8 Commercials, Minneapolis Wheat K N8 N7 K6 U6 N6 U7 Z6 H7 H8 H5 H6 K5 N5 Z5 U Z % 45% 50% 55% 60% 65% 70% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 44

45 Findings Minneapolis Wheat 0.20 Non-Commercials, Minneapolis Wheat y = 0.03x R 2 = 0.00 K8 K K5 N5 Z5 H5 U5 U6 H7 H6 N7 N8 N6 U7 K Z6 H Z % 4% 6% 8% 10% 12% 14% 16% 18% 45

46 Findings Minneapolis Wheat 0.20 Indexers, Minneapolis Wheat K7 K8 y = -2.52x R 2 = K5 N5 H5 U5 N7 Z5 K6 U6 H7 N6 H6 U7 Z6 H8 N Z % 0.2% 0.4% 0.6% 0.8% 1.0% 1.2% 1.4% 46

47 Findings Minneapolis Wheat K8 K7 Money Managers, Minneapolis Wheat y = -0.03x R 2 = K6 N8 H8 H7 U5 K5 N7 H6 Z6 Z5 U7 U6 N5 N6 H Z % 4% 6% 8% 10% 12% 14% 16% 18% 20% 47

48 Findings Minneapolis Wheat 0.20 Small Traders, Minneapolis Wheat 0.19 y = -0.04x R 2 = 0.01 K K H5 U5 N5 Z5 H6 K5 N6 U7 N7 U6 H7 K6 N H8 Z Z % 30% 35% 40% 45% 50% 48

49 Findings - Cotton! Liquidity was consistently above 10% and as much as 15% or higher during 2005.! For cotton, average liquidity per contract appears to be declining over time but this is heavily influenced by a substantial decline in liquidity during early 2008.! The sharp drop in the liquidity measure during early 2008 may have been driven by reduced future volumes as price limits became binding more frequently and volume appeared to shift toward the options market. 49

50 Findings - Cotton VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION 25% 20% PERCENT (%) 15% 10% 5% 0% MAR MAY JUL OCT DEC MAR MAY JUL OCT DEC MAR MAY JUL OCT DEC MAR MAY JUL CONTRACT - YEAR 50

51 500% 450% 400% 350% 300% 250% 200% 150% 100% 50% 0% 500% 450% 400% 350% 300% 250% 200% 150% 100% 50% 0% 51 Avg. Daily Volume As Pct of Open Interest as Pct of 2002 Volume and Open Interest Growth - Cotton Avg Daily Volume Avg Open Interest

52 Findings - Cotton! The presence of Commercials exhibits a very modest positive correlation to liquidity.! A very modest negative correlation may exist between the presence of Non-Commercials and liquidity in this market.! The presence of Indexers and Money Managers displays a much weaker and positive correlation to liquidity. 52

53 Findings - Cotton 0.18 Commercials, Cotton y = 0.21x R 2 = 0.19 K7 H5 K5 V6 H7 N8 K8 N7 Z5 Z6 N5 V5 K6 N6 H V7 Z7 H % 40% 45% 50% 55% 60% 65% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 53

54 Findings - Cotton y = -0.27x R 2 = 0.22 N5 H5 Non-Commercials, Cotton K N6 H6 K6 Z5 K Z6 V5 H7 N7 V6 N8 K Z7 V7 H % 10% 15% 20% 25% 30% 35% 54

55 Findings - Cotton Indexers, Cotton y = 0.18x R 2 = 0.04 N6 K6 H6 N5 K7 K5 Z5 H V6 V5 K8 Z6 H7 N7 N V7 Z7 H % 2% 4% 6% 8% 10% 12% 14% 16% 55

56 Findings - Cotton Money Managers, Cotton y = 0.28x R 2 = 0.11 N5 K5 H V6 V5 K8 K6 N6 N8 H6 K7 N7 Z6 H7 Z V7 Z7 H % 4% 6% 8% 10% 12% 14% 16% 18% 20% 22% 56

57 Findings - Cotton 0.18 Small Traders, Cotton K5 y = -0.27x N5H5 R 2 = 0.19 K6 H6 Z5 N8 K8 H7 N6 K7 Z6 N7 V5 V H8 Z7 V % 10% 15% 20% 25% 30% 35% 40% 57

58 Findings Natural Gas! For Natural Gas contracts, liquidity was steady possibly even declining slightly from early 2005 through early 2007.! From there, average liquidity increased sharply and peaked in mid-2007 only to retreat to previous levels before rebounding again into and through 2008.! While liquidity once averaged between 15% and 25%, it has regularly been above 25%, and even up to 40%, since the middle of last year. 58

59 Findings Natural Gas VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION JAN2005 FEB2005 MAR2005 APR2005 MAY2005 JUN2005 JUL2005 AUG2005 SEP2005 OCT2005 NOV2005 DEC2005 JAN2006 FEB2006 MAR2006 APR2006 MAY2006 JUN2006 JUL2006 AUG2006 SEP2006 OCT2006 NOV2006 DEC2006 JAN2007 FEB2007 MAR2007 APR2007 MAY2007 JUN2007 JUL2007 AUG2007 SEP2007 OCT2007 NOV2007 DEC2007 JAN2008 FEB2008 MAR2008 APR2008 MAY2008 JUN2008 JUL2008 AUG % 40% 35% 30% 25% 20% 15% 10% CONTRACT - YEAR 59 PERCENT (%)

60 300% 250% 200% 150% 100% 50% 0% 300% 250% 200% 150% 100% 50% 0% 60 Avg. Daily Volume As Pct of Open Interest as Pct of 2002 Volume and Open Interest Growth Natural Gas Avg Daily Volume Avg Open Interest

61 Findings Natural Gas! There appears to be little or no correlation between the market presence of the trader groups and average liquidity in this market. 61

62 Findings Natural Gas 0.30 Commercials, Natural Gas 0.25 y = 0.04x R 2 = 0.00 J5 M7 Q J7 J8 H5 J6V7 H8 H6 M8 V5 Z7N5 X7 V6 G5 U5 X6 K5 N6 M5 G8 Z5 N8 Q7 N7 M6 K8 U7 Z6 F8 F6 Q6 K7 K6 F7 G7 U6 X5 G6 Q H % 20% 25% 30% 35% 40% 45% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 62

63 Findings Natural Gas 0.30 Non-Commercials, Natural Gas 0.25 y = -0.24x R 2 = 0.03 Q8 M7 J K7 J7 N7 Q7 K8 K5 U7 F8 G7 J6 M8 N8 Z7 X6 Q5G8 H8 J8M5 N5 X5 U6 X7 M6 H5 N6 U5 G5 Q6 F6 G6 Z6 F7 V6 V7 V5 Z5 K6 H H % 6% 7% 8% 9% 10% 11% 12% 13% 14% 15% 63

64 Findings Natural Gas Indexers, Natural Gas y = 0.23x R 2 = 0.05 J5 M7 Q V5 G6 V6 Z7 Z5 K7 Q5 M5 Z6 Q6N6 G5 U6 F8 U5 K6 H6 F7 Q7 J8 M6 J6 V7K8 X5 G8 F6 N8 X7 U7 H8 N7 M8 X6 K5 N5 H G7 J H % 11% 13% 15% 17% 19% 21% 23% 25% 64

65 Findings Natural Gas y = -0.12x R 2 = 0.07 Money Managers, Natural Gas M7 Q8 J N7 M6 Q5 X5 N8 Q7 M5 H5 K5 U7Z6 U6 N5G5 F6Q6 G6 K6 F7 G8 Z5 X7 X6 U5 N6 M8 K8 F8 V6 H6 V7 V5 K7 J6 Z7 H8 J G7 J H % 25% 30% 35% 40% 45% 50% 55% 65

66 Findings Natural Gas 0.28 Small Traders, Natural Gas y = 0.27x R 2 = 0.11 Q8 M7 J J8 K8V7 H8 K7 F8 V6 Z7 H6 K6 J6 V5 M8 X7 Z6 X6 Z5 F6 F7 G6 N8N7 Q7 K5 U7 N5 U6 X5 G8 U5 Q6 N6 M6 H5 Q5 M5 G J7 G H % 7% 9% 11% 13% 15% 17% 19% 21% 23% 25% 66

67 Findings Crude Oil! Average liquidity was consistently around 40% during 2005 and 2006.! After a sudden upward shift, liquidity averaged mostly between 50% and 60% from early 2007 into 2008.! Liquidity for the July and August 2008 contracts soared to more than 70%, however.! It s safe to say that Crude Oil futures have seen a marked increase in liquidity over the study period. 67

68 Findings Crude Oil VOLUME / OPEN INTEREST FROM 200 DAYS TO EXPIRATION JAN2005 FEB2005 MAR2005 APR2005 MAY2005 JUN2005 JUL2005 AUG2005 SEP2005 OCT2005 NOV2005 DEC2005 JAN2006 FEB2006 MAR2006 APR2006 MAY2006 JUN2006 JUL2006 AUG2006 SEP2006 OCT2006 NOV2006 DEC2006 JAN2007 FEB2007 MAR2007 APR2007 MAY2007 JUN2007 JUL2007 AUG2007 SEP2007 OCT2007 NOV2007 DEC2007 JAN2008 FEB2008 MAR2008 APR2008 MAY2008 JUN2008 JUL2008 AUG % 80% 70% 60% 50% 40% 30% 20% 10% CONTRACT - YEAR 68 PERCENT (%)

69 400% 350% 300% 250% 200% 150% 100% 50% 0% 400% 350% 300% 250% 200% 150% 100% 50% 0% 69 Avg. Daily Volume As Pct of Open Interest as Pct of 2002 Volume and Open Interest Growth Crude Oil Avg Daily Volume Avg Open Interest

70 Findings Crude Oil! As the following charts suggest, the presence of Indexers, as a trading group, does have some positive correlation with liquidity in Crude Oil futures.! The presence of the other major trading groups exhibits much weaker correlation with liquidity. 70

71 Findings Crude Oil y = -0.34x R 2 = 0.06 Commercials, Crude Oil Q8 N Q K8G8 K7J7 N7 H7 G7 M7 M6 V7 M8 U7 H6 Z7 X7 J8 H8 J6 M5 K5 K6 Q5 N5 G5 H5 U5 V5 J5 Q6 X6 X5 Z5 F7 V6 U6 F6 G6 F8 N Z % 35% 40% 45% 50% 55% Regression Equation: Liquidity = a + b x R 2 = the correlation coefficient squared. 71

72 Findings Crude Oil 0.60 Non-Commercials, Crude Oil Q y = -0.09x R 2 = 0.00 N Q X7 G5 H5 J5 Z7 K5 X5 J7 V7 H8 U7 F8 N5 G6 G7M8 Q5 M5 V5 X6N6 Z5 K8 K7 N7 J6 M7 U5 Q6 M6 U6 F6 J8 H6 V6 H7 F7 G8 K Z % 10% 12% 14% 16% 18% 20% 22% 24% 72

73 Findings Crude Oil y = 0.57x R 2 = 0.35 Indexers, Crude Oil N8 Q Q Z5 G6 M5 Q5 K5N5 G5 K6 V5X5 J5U5 N6 Z7 F7 F6 J8 G8 H5 Q6 H6 M6 U6 V6 J6 M7 F8 X6 H7M8 H8 K8 V7 K7 N7 G7 U7 J7 X Z % 15% 20% 25% 30% 35% 40% 45% 73

74 Findings Crude Oil 0.60 Q8 Money Managers, Crude Oil 0.55 y = -0.50x R 2 = 0.13 N Q X7 N7 J6 U7 F8 X6N6 Q6 V7 K7 F7 G7 K6 G5 U6 V6 J8 F6 K8 J7 N5 H5 H7 M8 H8 G8 K5 M5 M7 Q5 U5 J5X5 V5 H6 G6 Z5 M6 Z Z % 5% 10% 15% 20% 25% 30% 74

75 Findings Crude Oil 0.60 Q8 Small Traders, Crude Oil 0.55 N8 y = -0.74x R 2 = Q X7 K8G8 J7 V7 J8 M8 N7 H8 F8 M7 X6 F7 Z7 V6 U6 K7 G7 H7 U7 J6 K6 X5V5 N6 Q6 H6 G6 M6 F6 Q5 N5 U5 Z5 K5 M5 J5 H5 G Z % 4% 6% 8% 10% 12% 14% 16% 18% 75

76 Summary! Overall, it appears that liquidity increased over the study period (2005 through mid-2008) for Corn, Chicago Wheat, Natural Gas, and Crude Oil.! The most pronounced increase in liquidity was in Crude Oil futures.! Liquidity has declined over time in Kansas City Wheat, Minneapolis Wheat and Cotton.! Liquidity has been relatively stable in Soybean futures with little change over time. 76

77 Summary! The strongest observed positive associations were the Non-Commercials trading in the corn futures market and the Indexers trading Kansas City wheat and crude oil futures.! In all three instances, however, we observed R- square statistics that were no higher than 0.35, suggesting a relatively weak correlation by most guidelines for statistical analysis. 77

78 Summary! Liquidity tends to be seasonal in many contracts, with certain months favored over others. In corn, Dec liquidity is almost always higher than in Sep.! This seasonality may mask trends in liquidity to some degree.! If there were enough data we would have preferred to isolate by contract month (e.g., only compare Dec contracts with other Dec contracts). Unfortunately, there were only three years of data available. 78

79 Summary! Overall, there is little evidence to suggest that any one particular trader group has a strong impact on liquidity either positive or negative.! It is more likely that, in markets where liquidity gains have occurred, all of the trader groups contributed in some manner. The factors that fostered increased trading for one group, likely did so for all of them.! We must remember that correlation does not imply causation. However, the lack of correlation makes a strong case for the absence of causation. 79