The dependence of the world economy on oil represents a crucial issue that has been widely explored in the extensive literature on energy economics. As oil price volatility has risen, there has been an increasing interest both by academics and practitioners in understanding the behaviorof oil prices. Higher demand for inventories and reduced supply will determine higher oil prices. In this regard, it emerges that oil pricevariations are partly deterministic (oil reserves, oil supplyand demand, OPEC policy-decisions) and partly random and unpredictable (oil commodity as a financial asset exchanged in the stock market). This peculiarity makes the oil commodity difficult to predict both for industrial policy purposes and from a financial investment point of view. Nonetheless, several specifications have been proposed in the economic literature forstudying and forecasting oil price volatility, in particular examining boththe oil price term structure and ability of futures prices in predicting oilspot prices, as well as the volatility of spot and futures prices themselves. One of the causes of this unpredictability and the consequent instability ofoil prices has to be analyzed through the "oil price dichotomy” (Fattouh2010), i.e., oil seen both as a commodity traded over the physical market where the equilibrium price emerges from the interaction of supply anddemand, and on the other side, as a financial asset where traders exchange their assets in order to cover their risky positions and speculative actions.
Several empirical techniques such as ARCH and GARCH models (Bollerslev et al. (1986), and Engle, R. (1982)), principal components analysis (PCA), jump-diffusion processes and the one, twoand three factor models have been implemented to model oil price volatility¹. As discussed in Bollerslev et al. (1992) Figlewski (1997) and Poon and Granger (2003), there are extensive reasons why volatility is a key variable to be investigated in commodity market literature. As underlined in Brook et al. (2004), the volatility factor of oil prices has to be assessed in considering the persisting character of this phenomenon. Looking at the time series data of the oil shock in the seventiesand the nineties, the different nature of such shocks prevents researchersfrom recognizing common factors and thus leaving the future oil price behavior unpredictable. Trying to identify drivers responsible for oil price volatility, Herce et al. (2006) claim that the great part of oilprice volatility is due to short term transitory factors that tend to lose theirinfluence in the long run. Observing weekly oil futures prices traded on NYMEX from September 1989 to May 2006, they look at what portion ofoil price oscillations is due to short-run factors such as volatility, and howmuch has to be attributed to long-run factors, such as demand and supplystructural changes, and at the same time, what part is due to the factorcorrelations. They show that longer maturity futures contracts incorporate a small component of short-term transitory volatility. Huang et al. (2009) analyze both the short-term dynamics and arbitrage possibilities between futures and spot prices of crude oil. Following the Pindyck’s hypothesis, (2001), they observe the decade from 1990 to 2001, findingan inverse relation ship between volatility (caused by external shock) and inventory level. Pindyck (2001) analyzes short-run commodity pricedynamics addressing oscillations, production and inventory levels. He disaggregates the principal oil market drivers, focusing on several key variables and their effects on market volatility. He asserts that producersdecide oil production levels considering their expected inventory, and thesedecisions are then taken considering two different prices (a spot price for saleof the commodity itself and a price for storage). Cash markets depend onother variables such as weather conditions, aggregate income, capital stocks, random shocks and technological changes. Following the Krichene (2005) let us identify the oil demand and supply as:
O^D=B₁x+B₂w+B₃z+ε
O^S=B₁x+B₂w^e+B₃g+B₄d+ε
And the oil demand (O^D) in function of:
x=Oil production (millions of barrels per day)
w=Oil price (per barrel)
z=GDP index for the G7 countries
ε=constant
And the oil supply (O^S) in function of:
x=Oil production (millions of barrels per day)
w^e = Expected oil price (per barrel)
g=Natural Gas production (billions of m³)
d= Dummy variable for the large oil price swings
ε=constant
In this mechanism, it is observable a component characterized by “fundamental” factors (supply and demand equilibrium) and another one led by “noise trading” speculative drivers (randomshifts). Pindyck (2001) theorizes that the fundamental variables can explain a largepart of the short-run price dynamics, but they are inefficient to entirely capture the whole dynamic behind price oscillations. Empirically, what he hasshown is that when real oil prices were approximately US$ 20 per barrel in 2000 dollars, the U.S. productivity growth in manufacturing was between 1.18% and 1.99% per annum. Instead, with the oil prices averaged over US$43 per barrel (in real price adjusted terms), productivity growth was only0.31% per annum. He thereby concludes that both short and long-term oilprice volatility is still a large and unsolved problem, but that a risk sharingmechanism and a larger buffer stock of oil prices can be an effective instrumentfor volatility reduction. Theoil price oscillations characterizing the oil price path during last decade can be observed in a series of reverse oscillations: the WTI oil price traded at US$ 50 p/b at the beginning of 2007, US$100 p/b in March of 2008, US$145 p/b in August 2008, US$33 p/b in Febraury 2009 and newly at US$113 p/b in April 2011 (figure 1).
FIGURE 1 WTI PRICE 2001-2011

These abrupt price variations, in such a short time, have irreversibly impacted on the investment decisions of producers, consumers and policy makers. Even if this turmoil can be interpreted as an oil price characteristic, it still exists a margin to limit the volatility and to reduce the frequency with the aim to contain the negative impacts on the aggregate output.
By the demand side, the policy makers initiatives might be focused on the strength of mechanisms able to provide a prompt answer to market dynamics, including price oscillations in international markets. Specific insurance schemes could be implemented as protection mechanisms for people principally affected by the volatility. From the supply side, stable and planned investments with fiscal aids by country, can reduce the market turbulences offering a stable and less uncertain view of the oil production over the medium and long term. Public-private agreements could take advantage of synergies and specific competences of any actor in the field. A higher market transparence level and a greater supervision from the regulatory point of view could offer an important aid to maximize the benefits coming from the financialization, and, at the same time, it could decrease the potential risks linked to the monetary stability. A higher frequency and a greater disaggregation of financial flows and detailed info on derivative contracts might clarify the dynamics of factors influencing the price composition process.
Overall we can highlight that, in order to avoid extreme price variations, producers, financial investors and market regulators have to contribute in their specific roles to improve both the financial market regulations (in matter of commodity financial trading) and the real mechanisms in the international oil markets. Concluding, we can say that the short-term price volatility is a key feature in the oil market with crucial effects on investments. In the short run,as underlined by Brook (2004), the low price elasticities of world demand and non-OPEC oil supply make oil prices highly reactive to supplyand demand variations. Price volatility oscillations, increased by geopolitical frictions and disequilibria, raise uncertainty about underlying pricetrends, tending to depress oil exploration.
________________________
¹ Arch (autoregressive conditional heteroskedasticity). Garch (General autoregressive conditional heteroskedasticity).

Bibliography
BOLLERSLEV T., Generalized Autoregressive Conditional Heteroskedasticity, Journal of Econometrics 31, (307-327) 1986.
BROOK A., PRICE R., SUTHERLAND D. S., D, WESTERLUND N. and ANDRE C., Oil price developments: Drivers, Econmomic Consequences and Policy Responses. Economics Department Working Papers, No. 412, 2004 .
ENGLE R., Autoregressive Conditional Heteroskedasticity with Estimates of Variance of United Kingdom Inflation, Econometrica 50, (987-1008) 1982
FATTOUH B., Oil market dynamics through the lens of the 2002-2009 price cycle. Oxford Institute for Energy Studies, 2010.
FIGLEWSKY S., Forecasting Volatility, Financial Markets, Institutions and Instruments, 6 (1-88), 1997.
HERCE M., PARSONS J. E. and READY R. C., Using futures prices to filter short-term volatility and recover a latent, long-term prices series for oil. Center for Energy and Environmental Policy Research (CEEPR) 06-005 WP, 2006
HUANG B. N., YANG C. W. and HWANG M. J., The dynamics of a nonlinear relationship between crude oil spot and futures prices: a multivariate threshold regression approach. Energy Economics, 31(91-98), 2009.
KRICHENE N., World Crude Oil and Natural Gas: a demand and supply model, Energy Economics 24, (557-576) 2002.
KRICHENE N., A simultaneous equations model for world crude oil and natural gas markets. IMF Working Paper, pages (1-24) WP/05/32, 2005.
PINDYCK R. S., The dynamics of commodity spot and futures markets: A primer. The Energy Journal, 3 (1-22), 2001.
POON S. H. and C.W.J. GRANGER C. W. J., Forecasting volatility in financial markets: a review. Journal of Economic Literature, 2003.