Today’s column continues with my critiques of Open Oil’s financial modeling exercise of the Guyana 2016 Production Sharing Agreement, PSA. As indicated earlier, the critiques, which are raised here are in no particular order of priority; save and except that, the first of these, represents in many ways, the most fundamental, in a theoretical and logical sense. That is so because this critique rests on both the intrinsic limitations of financial/economic modeling per se and the highly likely failure of modelers to portray their results, with an appropriate measure of caution. Indeed, rather than such caution, there is an unacceptable tendency to “amplify” their results, whether to attract attention or otherwise. Today’s column will concentrate on rounding out the first critique, for the benefit of my local readership.
I should warn readers though; that passionate as I might appear to be when highlighting the limitations of the predictive power of financial/economic models, I remain a firm believer of their usefulness. I therefore hereby readily acknowledge that, the development of robust solutions to economic modeling of petroleum projects and their assessment of associated project risks, are absolutely essential for Guyana’s progress, in its coming time of oil and gas production and export circa 2020.
Put another way, financial and economic models which seek 1) to assess the economic and/or financial viability of petroleum projects 2) to simulate the contributions to Government and/or Contractor take, prior to the project’s development/ production phases (for example Liza 1) 3) to support project strategy and decision making by the resource Owner and/or Contractor, during the project cycle. Such information will clearly aid the optimization of risk management and thereby further support the financing of projects, investments, and the provision of returns/benefits from these.
Limitations of Modeling
Assuming that the significance of the above statements is taken fully on board, I would go further and urge readers to bear in mind that, ultimately, financial and economic modeling of petroleum projects is always founded on simplifying the petroleum project’s contextual reality, as well as these values modelers input into the project simulation. This circumstance puts an enormous amount of responsibility on modelers to reveal fully, the interpretation that they put on the uncertainties attached to their inputted data.
In this regard, I urge my readers to consider that the modeler of Guyana’s 2016 PSA estimates the project life of Liza 1 to be over four decades long: 1999 – 2040. And further, that at the time the modeling exercise of the project was executed, there was still about two (2) years to go before the start-up of production! Obviously, the complexity of detail looking backwards is enormous. And, for looking-forward, this remains equally uncertain and therefore risky. Yet in light of experiencing about half the project’s life, little, if anything significant has been highlighted about such basic realities as the project’s geo-strategic, political (expectations), environmental, institutional and governance challenges. Indeed the implicit, as distinct from explicitly stated stance, which the model takes is that Guyana’s societal outcomes continue to trend on the past trajectory, despite the many known-unknowns and exogenous factors at play.
At this juncture, I refer readers to Wikipedia’s definition of an economic model; which is: “a theoretical construct representing economic processes by a set of variables and a set of logical and/or quantitative relationships between them”. Such a definition makes an economic model basically: “a simplified often mathematical, framework designed to illustrate complex processes”. Readers should note the use of the verb illustrate. Models do not re-create an alternate reality, as some are wont to believe and indeed go as far as to represent this in their writings. No, instead they illustrate reality.
Wikipedia goes on to state that “simplification is particularly important … given the enormous complexity of economic processes”. And that, 1) “economic models rest on … assumptions that are not entirely realistic 2) “omit issues that are important” and 3) as a result do not reveal the extent to which “the results are compromised by inaccurate assumptions”. (Wikipedia, Economic Models)
Similarly, according to Wikipedia financial modeling consists of “building an abstract representation … of a real world financial situation. This is a mathematical model designed to represent (a simplified version of) … a project”. While there has been some debate in the industry as to the nature of financial modeling, particularly as to whether financial modeling is a “trade craft … or a science the task of financial modeling has been gaining acceptance and rigor over the years”. (Wikipedia, Financial Modeling)
Financial Modeling Excesses
A wide swathe of expert analysts, academics, and political pundits attribute, at least, the proximate cause of the 2007/08 financial crisis, to the collapse of the subprime global mortgage market, triggered by excesses led by financial modelers, or quants, as they are called. These operate as major players in global financial markets. This 2007/8 crisis, as readers may be aware, led to the Great Depression, the effects of which, still linger in today’s global economy. As a famous “quant,” Paul Wilmot stated back in 2000:
“Unfortunately, as the mathematics of finance reaches higher levels so is the levels of commonsense seems to drop”. He went on to warn about what could eventually happen; “it is clear that a major rethink is desperately required if the world is to avoid a mathematically-led market meltdown”. As we now know, that happened as the decade progressed.
While personally, I do not attribute the global financial meltdown to the behaviour of the mathematician-led financial operators in the global marketplace. I am heartened though, by the efforts of those in that profession to introduce social responsibility to financial modeling. And, it is clear from my comments I am grateful for such actions.
Indeed, the financial modeling profession has gone one important step further. It has produced the Financial Modelers Manifesto, modeled on Marx’s Communist Manifesto; as well as requests that all financial modelers subscribe to a Modelers’ Hippocratic Oath, to do no harm.
The aim of both these efforts is to impose on modelers the same basic responsibility of not simplifying their results and overstating their predictive power, especially for constituencies easily bedazzled by headlines with numerical allusions, like “outlier low” and ‘costing Guyana billions”.
Next week I turn to some of the other critiques.