VOLATILITY OF CORN FUTURES PRICES

Get Complete Project Material File(s) Now! »

CHAPTER 3 VOLATILITY OF CORN FUTURES PRICES

INTRODUCTION

Chapter 2 concluded with a discussion on the market variables determining the futures price level at which commodities trade. Although a number of variables can be identified, the supply and demand of the asset underlying the futures contract remains the single most important factor determining prices. The fundamental purpose of a futures market is to enable users to minimise the impact of changing market conditions on price fluctuations, but exaggerated price movement remains a major obstacle in fulfilling this objective.
While futures market participants can discount the direction of market movement through a thorough analysis of the relevant fundamental and technical variables, it is impossible to accurately allow for the magnitude of price movement. Fundamentally, commodity price behaviour over time is a combination of systematic intra- and inter-year fluctuations as well as randomness, but supply and demand shocks can allow for systematic shifts in price patterns and exemplified price movements. This increase in price uncertainty is a major threat to the social welfare of institutions with either a direct or indirect exposure to futures prices.
This chapter will define price volatility and analyse the magnitude of market movement on the CBOT corn futures contract. The reasons for the increase in price volatility will be discussed and evidence will be provided on the theory that corn is entering a new era of exceptional high price volatility. This is necessary since Chapter 4 will focus on the effect of price volatility on market participants.

DEFINITION OF VOLATILITY

Volatility indicators provide traders with an estimate on the magnitude of price movement that can be expected. A price series or an economic indicator whose value fluctuates a lot is said to be volatile. Hull (2002) defines volatility as a measure of the uncertainty of the price of an asset. Since volatility measures variability or dispersion around a central tendency, it is a simple measure of the degree of price movement. Volatility can be explained as the standard deviation of the percentage change in the asset price over a specified period. In other words, it is the change in the price of a futures contract over a given period. Given that volatility is calculated by means of the standard deviation of the asset price, it will trade within one standard deviation from the mean (average price) two-thirds of the time. Large values of volatility imply returns which fluctuate in a wide range, and therefore increased risk (Kotze 2005). If the day-to-day variation is low, the value of volatility will be low as well.
Volatility is not a transitory phenomenon, but a lasting element of world markets. According to Nivens et al. (2002), price volatility is the least manageable factor threatening market participants. This agrees with an article by Henriques (2008) who argues that the impact of derivative mechanisms on price risk management is minimised by volatility levels which are currently trading well in excess of the average of the last quarter-century.
While some analysts argue that volatility is a result of the random arrival of new information that has an impact on expected asset prices, others believe that price volatility is largely a result of trading itself. Fama (1965) and French (1980) tested the causes of volatility empirically. After collecting the closing prices of assets over a period of time, these two authors calculated the variance of asset prices on successive trading days in addition to the variance of asset prices over trading days interrupted by weekends. The fact that the variance of asset prices over weekends was only between 19% and 22% bigger than those on successive trading days, indicates that volatility is much higher when the exchange is open and therefore it is claimed that volatility is a result of trading itself.
While volatility is an essential tool in the pricing of assets and the risk management of market price movement, the majority of previous research has focused either on equities or on currencies with little reference being made to commodity volatility. This is of particular interest since commodities, and especially corn, differs from equities as the commodity is grown and harvested in a seasonal nature.

CALCULATING VOLATILITY

Unlike most other market parameters, volatility cannot be directly observed and needs to be estimated. This provides a challenge in itself, as it is not necessarily stochastic and does not conform to any mathematical model. Even though the evolution of volatility remains uncertain, it is crucial to estimate volatility levels accurately for the purpose of risk measurement and management. Market participants make use of alternative methods of volatility estimates based on different objectives. These estimations are based on historical-, implied-, actual- and seasonal volatility.

Historical volatility

Historical volatility provides the user with an indication on how volatile the price of an asset has been in the past. This is of particular interest in the evaluation of trading strategies on historical futures prices. According to Skerrit (2002), historical volatility is analysed with the objective of forecasting future levels of volatility. Modern risk management strategies are concerned with probability distributions of expected future price ranges based on historical data. Should historical volatility be applied in future risk management strategies, it is important to include as many observations as possible in order to ensure statistical significance.

 Implied volatility

Implied volatility is often used by professionals and is the view from market participants on where volatility will be in the future. It provides a clear indication on market uncertainty and asset price risk. Kotze (2005:5) defines implied volatility as “… the volatility of the underlying asset price that is implicit in the market price of an option according to a particular model”.
The Black-Scholes model is an example of an appropriate mechanism to determine the implied volatility of the price of an asset. By substituting all the appropriate variables in the equation (as discussed in section 2.7.2) with the relevant values, the implied volatility can be calculated.

Actual volatility

Actual volatility is only known at option expiration. As a result, researchers conducting empirical research will only make use of actual volatility in its application after expiry of the option contract.

Seasonal volatility

Anderson (1985) investigated the determinants of futures price volatility through an empirical study. His findings suggested that the dispersion of futures prices in agricultural markets is not constant, but it exhibits changes in price over time in a systematic manner. The corn crop is grown and harvested in the same calendar year, and given the varying stock levels throughout the course of the year it is necessary to examine the impact of seasonality on price volatility.
Commodity prices are traditionally extremely sensitive to circumstances, which may influence the supply of the commodity. This results in inconsistent volatility levels over the specified period and is particularly evident in the corn futures market throughout planting and pollination (Erickson 2005). According to Skerrit (2002), the seasonal volatility trends are not exclusive to commodity markets, but may also be found in financial assets based on certain year-end phenomena.
Figure 3.1 highlights the correlation in the upward trend between corn futures prices and market volatility when approaching planting (April to May) and pollination (July). This can be attributed to weather uncertainties impacting on the acreage to be planted and the expected crop yield.
Total production equals acreage planted multiplied by yield. Good precipitation in the period before planting and up to April will allow for an increase in planted acreage, while drier conditions will result in acreage being lost to commodities with a longer window period for planting. In addition to this, pollination over the month of July plays a significant role in the eventual crop yield. Erratic temperatures and major deviations from the average rainfall for that period may result in lower yields. This uncertainty gives way to an increase in volatility, as the size of the crop cannot be determined accurately. After pollination in July, the acreage and yield are no longer susceptible to weather conditions and assumptions and projections can be made regarding the expected total production. As uncertainty gives way to actual crop expectations, prices tend to decline along with market volatility.
Samuelson (1965) argued that the variance of futures prices tends to increase as the delivery date of the underlying commodity approaches. A limitation of his finding is the fact that the uncertainty structure of the model is summarised in an exogenously given stationary stochastic process for spot prices. This view was partially supported by Rutledge (1976) whose empirical study provided limited support for Samuelson’s proposition.
In contrast to the statements suggesting higher volatility levels as the delivery date draws nearer, Goodwin and Schnepf (2000) confirmed the strong presence of seasonality in price volatility. In agricultural markets, volatility was found to peak during the summer months. Seeley (2009) examined the seasonal patterns evident in the volatility of corn futures prices by using high-frequency, intra-day price data to determine the effect of growing seasons on both continuous and discontinuous portions of volatility. He applied a test introduced by Barndorff-Nielsen and Shephard (2004), which compares the realised variance of an asset price against the bipower variation of an asset price. The results indicated that both the realised variance and the bipower variation increase linearly from a low in February to a peak in July before decreasing towards the delivery date. This is consistent with the conclusion derived from Figure 3.1, where volatility is depicted through the months of February to July. The following section will highlight the substantial increase in price volatility experienced on the CBOT corn futures contract.

VOLATILITY OF CBOT CORN FUTURES PRICES

‘Volatility’ is the word which best describes agricultural markets. Agricultural commodity markets are notoriously volatile and over the last few years, these markets have become even more vulnerable to world events as agriculture has become more global in nature. Ray, Richardson, De la Torre Ugarte and Tiller (1998) presented the findings from their large-scale simulation model as motivation for their projections that corn prices may be up to 82% more volatile in the future than over the decade preceding their study. This is of major concern given the already significant price variability inherent in the corn market during the 1990s.
The increase in price volatility subsequently resulted in a heightened interest in the price risk management mechanisms brought about by derivative instruments (Wilson & Dahl 2009). A study by MacDonald, Perry, Ahern, Banker, Chambers, Dimitri, Key, Nelson and Southard (2004) highlighted the staggering correlation between the increased use of derivative instruments as hedging mechanism and market volatility.
Figure 3.2 presents the average annual price volatility from 2000 to 2009. It is evident that the average annual volatility has increased substantially from 2005 until 2008, after which it made a small correction in 2009 due to the majority of the economic shocks of the US recession already being absorbed in commodity prices. According to research, the overall trend in price volatility remains higher though (Irwin & Good 2009).
In addition to the volatility of corn futures prices, the actual magnitude of price movement is a major concern to all stakeholders in the CBOT corn futures market. Price limits represent the maximum price movement allowed throughout the course of a single trading day. The current price limit for corn futures contracts as traded on the CBOT is $0.30 per bushel per day. Should two or more corn futures contracts within the first five expiration dates trade at limit levels, the limit is expanded to $0.45 per bushel for the following trading day. This level can be expanded even further to a maximum of $0.70 per bushel on day three of a limit movement. Limit movements prevent further trading as the daily limit has been reached and the futures market consequently loses its objective of price discovery.
The increased uncertainty regarding futures prices and an explosion in the magnitude of price movement (see Figure 3.3) has resulted in numerous trading disruptions as a result of limit price movements. From 2000 to 2005, the annual trading range of corn futures contracts stabilised around $1 per bushel, but as a consequence of increased volatility, recent annual market movements have been as high as $5.06 per bushel. This may lead to exchanges adjusting the daily allowable price limit, which in turn will increase the risk of all traders with active positions in the futures market. It has become especially risky to all market participants over the last few years to have exposure to price movement, as was evident in 2008 when prices increased by 64% over the first six months to an all-time high of $7.88 per bushel, only to fall by 54% over the second half of the year to a level of $3.60 per bushel (see Figure 3.4).
It is evident that price deviations from the mean are increasing at an alarming pace, resulting in higher levels of volatility. It is important to consider the reasons for this disconcerting trend.

REASONS FOR INCREASED VOLATILITY

It is important to consider the reasons for the exaggerated price movements in the agricultural futures market. Researchers are unanimous about the factors contributing to the heightened volatility levels. These can be summarised as follows:

Supply and demand

Refer to section 2.7.1.1.
Trostle (2008), Irwin and Good (2009), as well as Wilson and Dahl (2009) cite the supply and demand of the commodity underlying the futures contract as the dominant reason behind increased levels of price volatility. This is in correlation with research by Goodwin and Schnepf (1999), which attributes spikes in volatility to lower stock levels as a result of adverse weather conditions.
Sufficient stock levels tend to make prices less sensitive to new market information, and will therefore lead to low levels of volatility coupled with a narrow trading range. In contrast to this, prices are extremely sensitive to new market information during periods of low stock levels. This will result in high volatility with big trading ranges. The US carry-over corn stocks and stocks to usage ratio decreased rapidly from 2005 as ethanol production exploded on the back of government subsidies and laws favouring the production of biofuel as a complementary and alternative method of energy. The lower stocks gave way to a sharp increase in annual price movement and consequently higher levels of volatility. According to Figure 3.5, annual market movement increased from a low of $0.40/bushel to a high of just under $5. In a highly correlated manner, the average annual volatility increased from a low of around 20% to an annual high in excess of 40%

Government policies

Explicit government policy variables, such as acreage constraints and price support programmes, have a surprising positive correlation with price volatility. Mills (2003) cites uncontrollable events such as government policies as a major contributor to price volatility. His view is supported by Wescott (1998) who found that government policies are inherent in all of the supply and demand variables, but manifest themselves most directly in the acreage, export and stocks to usage variables.
This is consistent with the conclusions derived from research by Goodwin and Schnepf (1999) who concluded that federal income support activities such as the deficiency payment variable and loan rate variable destabilised corn prices. Although these activities enhanced and stabilised producer incomes, it resulted in more volatile market prices due to the response from producers to distorted incentives provided by the programs.

Mutual fund trading

Mutual fund trading, also known as speculative trading, has received increasing attention from researchers. In the majority of published research reports, mutual fund trading is seen as one of the main culprits enhancing an already volatile commodity futures market (Cai, Cheung & Wong 2001; Wilson & Dahl 2009).
An investigation by the United States Senate (2006) into the collapse of a $9-billion commodity hedge fund concluded with the finding that excessive speculative actions from mutual funds result in significant price distortions and increased price volatility. When prices are continuously bid higher or offered lower by funds, the probability that excess returns will continue in subsequent periods decreases along with the length of excess returns. Therefore, the longer the speculative trend continuous, the more likely a reversal will occur.
Emekter, Jirasakuldech and Went (2011) investigated the existence of speculative bubbles in commodity markets, including corn futures contracts. A speculative bubble occurs when investors realise that the commodity traded is overvalued but refrain from liquidating the position in expectation that higher future returns will compensate for the increased risk of the bubble deflating. In their research, McQueen and Thorley (1994) applied their non-parametric duration test. The subsequent empirical findings suggested sufficient evidence of rational bubbles caused by speculation in commodity markets. The eventual correction in the value of the relevant commodity therefore increased volatility.
It is also noted that speculative pressure affects price levels in the long run, and is to blame for price reversals in the short run. Therefore, it can be concluded that speculators and speculative transactions enhance price volatility.

Outside markets

Outside market forces remain one of the biggest contributors to volatility in corn prices.
These include equities, credit markets and oil prices.

Equities and credit markets

The impact of credit markets on corn price volatility is highlighted by the pressure brought about by the global problems in credit markets since mid-2008 (Irwin & Good 2009). Once it became apparent that an economic recession in the US was inevitable, mutual funds and speculators liquidated their positions, resulting in prices falling from a peak of $7.88/bushel on 26 June 2008 to a low of $2.94/bushel on 5 December 2008.

Oil prices

During the mid-1980s the largest category of corn use was animal feed totalling in excess of 104 million tons of corn. In comparison, the industrial use of corn was a miniscule 29 million tons, even less than the volume of direct exports. Over the next 25 years, every category of corn use had double-digit percentage increases. In addition, the industrial use of corn has now overtaken animal feed as the single largest category of corn usage in the US. This phenomenal growth can be attributed to the use of corn as raw material in the fermentation of fuel ethanol, which is blended with regular gasoline (Finnegan 2011.)
Figure 3.6 illustrates the sharp increase in the volume of corn used for the production of ethanol, which is projected to amount to 42% of the total corn production by the end of 2011. This equates to just shy of 150 million tons of corn.
Given the astonishing growth in the usage of corn for ethanol fermentation, its value as an agricultural commodity is increasingly being reflected through its value as a blending component for fuel. Therefore, the price of corn is affected by movements in the price of oil. This is illustrated by Figure 3.7, where the close and continuous correlation between the prices of oil and corn can be observed.
Given the current unstable worldwide economic conditions, it is important and necessary to determine if the current high levels of price volatility are sustainable and whether volatility can even increase further in the future.

CONTENTS
CONTENTS
LIST OF FIGURES
LIST OF TABLES
SUMMARY
CHAPTER 1: INTRODUCTION
1.1 Background
1.2 Problem statement
1.3 Research objectives
1.4 The importance and benefits of the study
1.5 Research structure
1.6 Chapter layout
CHAPTER 2: THE FUTURES MARKET
2.1 Introduction
2.2 History of the futures market
2.3 Derivative instruments
2.4 Margining of derivative instruments
2.5 Commodity trading
2.6 Variables determining prices of derivative instruments
2.7 Summary and conclusions
CHAPTER 3: VOLATILITY OF CORN FUTURES PRICES
3.1 Introduction
3.2 Definition of volatility
3.3 Calculating volatility
3.4 Volatility of CBOT corn futures prices
3.5 Reasons for increased volatility
3.6 Expectations from future corn price volatility
3.7 Summary and conclusions
CHAPTER 4: PRICE RISK MANAGEMENT PERFORMANCE OF MARKET PARTICIPANTS
4.1 Introduction
4.2 Volatility transmission to market participants
4.3 Price risk management evaluation
4.4 Reasons for price risk management failures
4.5 Summary and conclusions
CHAPTER 5: THE EFFICIENT MARKET HYPOTHESIS
5.1 Introduction
5.2 Definition
5.3 Development and functioning of efficient market hypothesis
5.4 Price forecasting
5.5 Summary and conclusions
CHAPTER 6: TECHNICAL ANALYSIS AS PRICE FORECASTER
6.1 Introduction
6.2 Definition and contrast to fundamental analysis
6.3 Price forecasting ability of technical analysts
6.4 Technical oscillators and moving averages
6.5 Shortcoming of trading exclusively through technical analysis
6.6 Summary and conclusions
CHAPTER 7: PRICE AND VOLATILITY TRENDS
7.1 Introduction
7.2 Definition and classification
7.3 The validity of market trends
7.4 Duration and magnitude of trends
7.5 Exploring trends in historical data
7.6 Summary and conclusions
CHAPTER 8: RESEARCH DESIGN AND METHODOLOGY
8.1 Introduction
8.2 The business research process
8.3 Research objectives
8.4 Research design
8.5 The research sample
8.6 Gathering the data
8.7 Processing and data analysis
8.8 Statistical analysis results
8.9 Conclusions and reporting on research findings
8.10 Summary and conclusions
CHAPTER 9: EMPIRICAL RESEARCH FINDINGS AND PERFORMANCE MEASUREMENT
9.1 Introduction
9.2 Importance of conclusions derived from earlier chapters
9.3 Proposed price risk management methodology
9.4 Performance measurement by means of benchmarking
9.5 Returns achieved versus benchmark 2000–2009
9.6 Application of methodology on random data sets
9.7 Summary and conclusions
CHAPTER 10: SUMMARY, CONCLUSIONS AND RECOMMENDATIONS
10.1 Introduction
10.2 Summary of research
10.3 Summary of findings from research questionnaire
10.4 Summary of findings from application of methodology
10.5 Conclusions
10.6 Recommendations for further research
GET THE COMPLETE PROJECT