OIL - emerging BUBBLE

Last Thursday Brat Stupak chairman of Congressional energy panel said that oil products markets were being “manipulated” by the biggest trading houses in the futures market, through he said a probe hasn’t uncover illegal activity.

I must say that I don’t understand what congressmen Stupak wanted to communicate by his comment.

Earlier this week Gerorge Soros testified before US congressional committee on manipulation in the energy markets . He said the record oil prices weighing on the economy are the result of a ``bubble'' caused by speculation from index funds and a tight balance between supply and demand. Few days earlier in the interview for Daily Telegraph Mr Soros highlighted 4 main factors behind soaring oil prices

1) Peak oil – Meaning increasing costs of developing new oil reserves and aging of the existing oil fields

2) The backward sloping oil supply curve –the higher the oil price goes the less incentive oil producing countries has to convert their oil reserves into depreciating currencies (mainly US dollars). In other words the OPEC grew richer and less in need of current income, while negative real interest rates made it more attractive to keep oil in the ground than cash it at bank

3) The mayor of growing areas of growing demand subsidizes the energy prices so the higher oil prices does not affect oil consumption in those areas.

4) Trend following speculation which possibly will end up with the bubble.

I think his analysis is correct and the 4th factor is increasingly affecting the oil prices. Let’s start with the definition of the bubble

The term “bubble” is widely used but rarely clearly defined. Following investopedia “bubble” is defined as

“A spike in asset values within a particular industry, commodity, or asset class. A speculative bubble is usually caused by exaggerated expectations of future growth, price appreciation, or other events that could cause an increase in asset values. This drives trading volumes higher, and as more investors rally around the heightened expectation, buyers outnumber sellers, pushing prices beyond what an objective analysis of intrinsic value would suggest.”

This definition as all others contain flaw. Indeed defining bubble is very tricky because acceptance or rejection of the bubble is depends on the model of fundamentals one uses. The concept of a fundamental price reference does not necessarily exist, nor is it necessarily unique. A major problem is that apparent evidence for bubbles can be reinterpreted in terms of market fundamentals that are unobserved by the researcher A key problem is that bubbles are usually defined as exponentially explosive growth phases, but how can one then distinguish them from the growth of a fundamental valuation process which is also generically expected to follow an exponential growth path?

This problem can be solved by defining a bubble as a faster-than-exponential accelerating price, which may be refed to as “superexponential.” Being faster than exponential, i.e., the growth rate is itself growing; it is necessarily unsustainable if we assume a standard geometric growth for the global economy. Why then the prices start to grow at unsustainable “superexponential” rate? Soros explain it by his reflexivity theory which says that market participants are trying to discount a future that is itself shaped by market expectations. This is self reinforcing “rational” process which is a seed for rational bubble.

In recent years prof. Didier Sornette developed a new methodology to test fingerprints of speculative bubbles called Log Periodic Power Law (LPPL). In a series of works, financial bubbles have been defined as regimes in which the market exhibits an unsustainable super-exponential growth, that can be characterized quantitatively as a genuine critical phenomenon with specific log-periodic power-law signatures. The main ingredient of the theory is the existence of positive feedbacks, that is, self-reinforcements, which leads to cooperative herding and imitation between investors. As a consequence, the actions of investors tend to produce waves of contrariant and imitative behaviors leading to self-reinforcing bullish or bearish phases, themselves decorated by the opposite regimes in a hierarchical fashion. Positive feedbacks, when unchecked, can produce runaways until the deviation from the fundamental price is so large that other effects can be abruptly triggered and lead to rupture or crashes. Alternatively, it can lead to prolonged depressive bearish markets

Mathematically, the LPPL model is represented by the equation giving the anticipated expected trajectory of the log-price I(t) ≡ ln p(t) of a given asset as a function of time, expressed in terms of the distance τ = tc − t (respectively τ = t − tc) to the critical time tc for bubbles:

I (t) = A + Bτ^α + Cτ^α cos [ω ln (τ )]

A is the log-price at tc (since τ = 0 at t = tc, all the other terms are vanishing at t = tc), B (respectively) controls the amplitude of the power law acceleration (respectively the log-periodic component) of the log-price. The exponent α encodes the structural shape of the acceleration. It is usually found between 0 and 1, which ensures a finite price at tc together with an asymptotic infinite rate of change close to tc. The parameter ω is the log-periodic angular frequency of the log-periodic oscillations. It should be stressed that ω is not the inverse of a time scale, but rather it is proportional to the inverse of the logarithm of a scale factor λ, where λ is roughly speaking the ratio of the distances between successive peaks of the log periodic oscillations.

LPPL formula (chart) when fitted to the oil data shows 1) very low alfas which confirms acceleration of the underlying trend 2) superexponential growth is accompanied with stable log periodic oscillations which confirms an existence of positive speculative price to price feedback. This is an initial confirmation that oil price incresingly depends on speculation

The initial estimation suggest that oil price within next two weeks will spike to around 150 dollars per bbl.