Computer Models: The Bane of Modern Society

In a hearing before Rep. Henry Waxman’s House Committee on Oversight and Government Reform, Alan Greenspan, former chairman of the US Federal Reserve Board (a position touted as one of the most powerful unelected offices in the world), said he got it wrong in answer to questions about his role in the recent financial meltdown. His extremely mobile face deadpanned that his economic models, which he had relied on for 40 years, were wrong. He did not apologize; it was merely a statement of fact that portrayed no irrational exuberance. He gave no hint of concern about the massive damage his reliance on the models had done. Huge losses of money among those who exploited the situation his models allowed garnered no sympathy. However, the dashing of hope at the bottom of the economic pyramid, the disaster of losing one’s home or job, the stress created by worrying about losing either, and myriad other such stories in the US and across the world appeared to be dismissed with a wave of the academic and intellectual hand.

There were warnings. In October 2005, Stephen King wrote in the British newspaper The Independent,

Despite Mr Greenspan’s colossal reputation, I have my doubts that his approach will survive his departure: by giving the impression that all risks can be contained through his own wizardry, Mr Greenspan may have encouraged excessive risk-taking, most obviously with the equity bubble in the late 1990s and, more recently, with the emergence of a housing bubble.

The wizardry was his models. Greenspan likely believes he absolved himself from any blame or responsibility by his statement that it was the model’s fault – I am not responsible or accountable. As the Naked Capitalism blog puts it,

Being an objectivist means never having to take responsibility for your actions. Greenspan has now decided to pin the financial market crisis on models.

The cliché about models is garbage in garbage out (GIGO); but who put the garbage in, decided what happened to the garbage while it was in the model, and then decided how the garbage was used once it was out?

Ironically, to a certain extent, Greenspan is correct. The models are the problem. Models are useful tools as long as they are used for simple readily measured situations. However, even there they can be wrong. Consider the failures that occur with constructions. The bridge in Minneapolis is a good example. When complexity increases, particularly through interactions between various segments, their use becomes extremely questionable. They depend upon the amount and accuracy of the data on which they are built, and in most cases this is less than adequate. They assume an ability to quantify every variable, but this is not possible. Even the largest computers cannot include all variables. Which ones do you leave out? How would you quantify Greenspan’s irrational exuberance? Indeed, how do you quantify human behavior?

Greenspan failed to quantify human reaction to policies he formulated based on his model output and in his testimony he admitted he did not anticipate what happened. This is a typical academic response and why the phrase “it is purely academic” means it is irrelevant to the real world. What is remarkable is his naivete, his belief in his model, and his lack of understanding of human nature. Unfortunately, he is not alone in an implicit belief in models and their ability to simulate the complexity of real world conditions. He is not alone in the application of the model outputs as the basis for major public policies. They are the bane of society wherever they are used.

Models range in form from hardware models (which are simply scale reductions, such as a model airplane) to purely abstract models that replace individual components with symbols. These, in the simplest model, are usually letters of the alphabet to represent a variable. Everyone is familiar with Einstein’s famous model, symbolized in the mathematical formula E = MC². E represents energy, M is for mass, and C is for the speed of light. Almost all models in science or social science are mathematical.

Modeling became dominant in every discipline with the advent of the computer. This allowed for inclusion of vast amounts of data on which complex calculations could be performed. Unfortunately, this gave them a credibility that they didn’t deserve. As Pierre Gallois said,

If you put tomfoolery into a computer, nothing comes out of it but tomfoolery. But this tomfoolery, having passed through a very expensive machine, is somehow ennobled and no one dares criticize it.

Working with models in a laboratory or academic environment only requires logic, rigorous method, and adherence to scientific standards. Too often today even these are not being met, but they only do damage within academia. However, once you use the output of your models for policy, then a social and political responsibility is required – and as we see more and more, often it is not being met. Greenspan and his model are a disastrous example.

Economics is a discipline within the general area of the social sciences. The term implies that somehow you can apply the scientific method to individual and group behavior within a society. However, there is a fundamental difference between science and social science – and that is the ability to predict. A simple definition of science is the ability to predict. The scientific prediction does not trigger a response or a change, but remains measurable. Social science predictions inevitably produce a response and triggers change that jeopardizes the prediction. For example, if an economist studies a community and produces a predictive report, leaders and innovators in the community react by changing their behavior and thus that of the community. This results in invalidating the original predictive report. Obviously, this is what happened when Greenspan applied his predictive model output to the US economy.

Greenspan’s model was the basis for US financial policy for the entire time he was Chairman of the Reserve Board after his appointment in 1987. It was also the basis for world economies, as the reverberations of the collapse demonstrate. However, it is not the only flawed model influencing global policy and driving it in the wrong direction. The Intergovernmental Panel on Climate Change (IPCC) climate model is the sole source of evidence that human CO2 is causing climate change; yet it is being used to create completely unnecessary taxes, policies, and hardships.

The IPCC models are also the source of predictions about threatening future climates. This despite their own warning in their first Report (Climate Change 1992) that…

Scenarios are not predictions of the future and should not be used as such.

…while the Special Report on Emissions Scenarios says,

Scenarios are images of the future or alternative futures. They are neither predictions nor forecasts.

By Climate Change 2001 they were saying,

The possibility that any single in emissions path will occur as described in this scenario is highly uncertain.

They later say,

No judgment is offered in this report as to the preference for any of the scenarios and they are not assigned probabilities of recurrence, neither must they be interpreted as policy recommendations.

The hypocrisy of these words is provided by the Summary for Policymakers (SPM) they produce.

Some argue climate models are better than economic models because they are based on physics. If this was true, then their predictions would be accurate; but they are not. It’s not surprising, because they are not validated. This is a standard test in which a model attempts to recreate previous known conditions. Everyone is aware they cannot provide accurate weather forecasts beyond 5 days, so it is unreasonable to claim they make accurate climate forecast for 50 and 100 years. The argument that weather forecasts are different than climate forecasts is not upheld because climate is an average of the weather. They are only as accurate as our knowledge of the weather and its mechanisms. At a recent conference on climate modeling in Reading, England, Tim Palmer, a leading climate modeler at the European Centre for Medium-Range Weather Forecasts, said:

I don’t want to undermine the IPCC, but the forecasts, especially for regional climate change, are immensely uncertain.

A paper by Demetris Koutsoyiannis et al argues that climate models have no predictive value.

The failure of the IPCC models is not surprising. They are built on the theory of warming/climate change, which uses the fundamental assumption that an increase in CO2 will cause an increase in temperature. In every record of any duration for any time period in the Earth’s history, temperature increases before CO2. However, a major problem is the models’ focus on human causes, as their mandate dictates. As Roy Spencer said in his testimony before the US Senate EPW Committee:

And given that virtually no research into possible natural explanations for global warming has been performed, it is time for scientific objectivity and integrity to be restored to the field of global warming research.

The IPCC and their totally inadequate and incomplete climate models exploit people’s fears and lack of understanding while driving politicians to completely wrong policy. They present scenarios and warn against using them as predictions, yet produce a Summary for Policymakers. Individual IPCC members actively encourage policies.

Greenspan’s bland and unapologetic statement that his model failed is frightening. It is even more frightening that the solutions do not deal with the fundamental flaws that allowed it to exist. Spending more than you earn is a problem from the individual through to government. He encouraged credit and then chastised the irrational exuberance with which it was adopted. Now, those who provided and often exploited the credit shock him.

The same is true of climate models. They are grossly flawed, being built on at least one critically false assumption, on inadequate data, and omit major mechanisms while consistently making inaccurate predictions. The damage of energy and environmental policies based on their output is already extensive and will get worse as politicians plan massive CO2 reductions. The question is, how did Greenspan get away with it? Why wasn’t he challenged? How are the IPCC getting away with their deceptions and failed models? Bartholomew and Goode provide the answer succinctly in this paragraph on mass hysteria:

Many factors contribute to the formation and spread of collective delusions and hysterical illness: the mass media; rumors; extraordinary anxiety or excitement; cultural beliefs and stereotypes; the social and political context; and reinforcing actions by authorities such as politicians, or institutions of social control such as the police or military. Episodes are also distinguishable by the redefinition of mundane objects, events, and circumstances and reflect a rapidly spreading folk belief which contributes to an emerging definition of the situation.

They should add academia as a reinforcing authority.

The “redefinition of mundane objects” applies to weather events and climate change. These natural events have been redefined as unnatural and therefore problematic. They are then wrapped in the larger environmental hysteria. It also appears George Orwell was correct when he wrote,

In a time of universal deceit, telling the truth becomes a revolutionary act.

Wax I’d. Will, when way to now generic viagra online because quality is try aesthetician’s it goes.

You may also like...

2 Responses

  1. May 13, 2011

    […] SOURCE […]

  2. June 17, 2013

    […] the inadequacy of the computer models. It is covered in many articles on this web site including a general concern about their application in society. In one article I quote Pierre Gallois’ comment […]