It is a great honour to be invited to address this gathering and also it is a very fine development
that different professions which together are needed in collaboration in order to face up to the many problems
that modern technology and modern society has presented us with.
So, I am glad for two reasons, the second being to be allowed to participate in inter-disciplinary activity.
I am worried that our universities are not designed and organised to make it easy to find these collaborations
but outside universities there are other such opportunities and in particular the “Tagungen” in Lindau have that character.
Now my topic then is on long-range projections of alternative energy futures
which is one particular study in the field of energy modelling itself, the development of the last ten to twelve years.
So please tell me when I have only five minutes left.)
The development of energy modelling took place in United States, Europe, Mexico, India and other countries.
The aim of that technique is to visualise alternative energy futures.
The nature of the problem makes one need to draw on various fields of knowledge and actually some speculation.
Technological knowledge to draw on physics, chemistry, biology, engineering, behavioural, to draw on economics
in regard to the behaviour of consumers and of producers faced with the market and the resource availability problems
in which we need to turn to geologists and mining engineers.
There is in this type of work a speculative element and that is inevitable in models that look far ahead into the future.
For that reason the conclusions reached have the following logical form: If such and such, then so and so.
And the ‘ifs’ must be emphasised.
If you find that someone reports on a model study and those were my conclusions and that person only gives the ‘thens’
and not the ‘ifs’ then that is unsatisfactory.
Now, I would like then to use one recent modelling study in which I was involved
that illustrates the type of work along these lines.
There was in the United States a very substantial and widely ramifying study of energy futures and energy presence
called by the committee on nuclear and alternative energy systems.
A committee set up by a combination of the national academy of sciences and the national academy of engineering.
There was an overview committee that was responsible for the ultimate publication,
a very substantial book that has come out of that work.
There were also panels to provide analytical work into the mill so to say of the deliberation of the overview committee.
And I was chairman of a panel called Modelling Resource Group.
Now the word ‘resource’ in that sentence is human resources, experts, resources of expertise.
The assignment of that group was to compare the answers given to the same set of questions by three different models
that were already available before the work was started.
My emphasis in this report is on the methods and on the kinds of questions that can be answered.
Therefore the ‘ifs’ that we did use and we had several alternative ‘ifs’ side by side should be revised as time goes on.
And any results if I may use that word of the particular study would have to be revised as the ‘ifs’ are revised.
Now, (can we dim the top light a bit, enough now for the, I have to switch this on.)
This diagram is in a way a layout of the study.
We did consider three groups of variables, the driving variables, the realisation variables, excuse me.
The word ‘realisation’ was used for something that is not subject to policy decisions.
It is there already.
Or it is caused by circumstances and decisions that the energy related policies do not have an effect on, or a sizeable effect on.
Now we had there the GNP growth rate as one, you may also say exogenous variable.
And that was estimated on the basis of population extension, labour force participation and such factors at 3.2 per annum
from the beginning year of the study which was 1975, up to 2010.
Then another realisation value was cost levels of energy technologies.
At that point we did not have the subsequent experience and so we had the following figures in terms of dollars of 1975,
the figures are for capital cost of electricity generation.
For coal fire generation we had $520 of 1975 per kilowatt electric.
Light water reactor the number goes up to $650.
Advance converter reactor $715.
Fast breather reactor $810, again dollars of 1975 per kilowatt electric.
And solar central station $1,730 in the same unit.
Now, these numbers have all gone up since then so I report on the study as made at that time
without trying to bring up to date any numbers.
We also had resource doc availability numbers on there for oil and gas, it was in the United States a quantity of 1,720 quads
where one quad is 10^15 BTU and this was oil and gas at a cost of extraction up to $2 of 1975, per million BTU.
For uranium it was 3.7 x 10^6 tons of U308 at a cost of extraction not exceeding $30 in that same unit.
Now then it says there on the right, demand elasticity with regard to price and with regard to income.
It turned out towards the end of the study that was an extremely important parameter.
And I will come back later to its precise definition.
It is a measure of the response of demand to a given increase of price and a given increase of income.
I have a slide later that will indicate both the numbers used and the definitions of the concepts.
In any case the three important models that we used differed in their price elasticities
and that was in fact helpful because it indicated the importance of that parameter.
Now the models that were used, we had six models actually in the study but the only ones long enough looking into the future
were DESOM, ETA and Nordhaus and I will refer to those in more detail later on.
Then we had policy variables, first of all base case which was going on pretty much with the developments
as they are presently called for.
That is called the base case.
And then we had other cases that were obtained from the base case by policies that were in there
because they were much under discussion at the time, the nuclear moratoria was not then considered as a policy.
Since then something in that direction has taken place, not really as a policy but as a result of mishaps and fears.
So it is desirable to have this in the study regardless of whether or how such a curtailment might come about.
It was a defined nuclear moratorium in the case of applied to all nuclear reactors.
That was one case, another case was to apply it not to the light water reactors already in service for quite some time.
And it was, so these two cases were distinguished.
Also, limits on the use of coal and oil shale coal because of the acid rain and oil shale as a result of water use
and water deterioration as a result of use.
And in both cases, in the case of all fossil fuels there is also a long range concern with the CO2 content of the atmosphere.
So that’s the reason why that is in there also.
These limits were defined, we won’t give the figures,
by drawing some curves that would level off to an asymptote in the case of coal and another curve in the case of shale oil
and the limits would be that at no point would the annual rate of production of coal or shale oil exceed that curve.
Now then finally we had a third category of variables that we called a blend variable,
which are called blend variables because they combine the properties of realisation variables and of policies.
The discount rates as we have found to our dismay in the United States are also subject to policy.
but they are also in the absence of specific determinant policy still are a reflection of the behaviour of parties
in the capital market of the economy.
And so that is a blend variable but we ended up using 13% for pre-tax discount rates applied to investment and pricing decisions
of the energy producing business, forms, industry.
And 6% post-tax applied to the consumers’ relative weight to future benefit,
given to future benefits as compared with present benefits.
For that we had 6% discount rate.
Then there was a ceiling on quantity and price of imports of fuels that I will not dwell on.
And an estimate of the commercial availability if wanted of advanced converter reactors,
fast breather reactors or solar electricity generation all assumed to be available at and from the year 2000
if that were to be aimed for.
Then, the technique, or the ideas on the economic side were that optimisation,
can be looked at as a simulation of the behaviour of a competitive market system.
I like to put it this way, there is in a world neither prefect competition nor perfect planning.
But if there were, then they would be equivalent.
The perfect planning would be guided by prices similar to those produced by perfect competition or producible
by perfect competition as well.
So that we have this fictitious image of efficiency in the use of resources and it doesn’t matter
how it is obtained assuming it can be obtained.
It was meant to be actually an approximation of how our not quite perfect system of markets operates.
So you could say we used Adam Smith’s invisible hand but with foresight.
Now then the price and income illustrated is as of demand for energy, I'm now ready to say some more about that.
Here are the numbers that we in fact used and now I also want to define the price elasticity,
that was the most important of the two.
Let X be the demand and P the price, the demand for energy and the price of energy in terms of some aggregates
over the various forms of energy, just one number.
Then, the elasticity as defined as the derivative of the log of quantity demanded with respect to the log of price in the market.
And that is as you see from the use of logs a dimensionless quantity.
Now there were, if policies that constrain energy supply are strong,
are applied strongly then a high price increase is needed to constrain demand accordingly.
And if the absolute, yes I should have said each, normally each price eleasticity of demand is a negative number.
Because as price goes up quantity demanded goes down.
Now I go back to my sentence, a high price increase is then needed to constrain demand accordingly
and if in absolute value the elasticity is low and then less remains to spend on other goods.
If the absolute value of the elasticity is high then only a smaller price increase results
and the amounts spent on other goods are less affected in the total of GNP, or gross national product is less affected.
This simple straight forward reasoning indicates that either the price elasticity of demand, that negative number,
is a very critical parameter.
And I'd like to mention also, I will first describe the elasticities that have been used in the study.
Here DESOM was a model developed at Brookhaven laboratory and for reasons that were connected with the purpose of that model,
it did not have consumers’ response dependent on price.
It just projected the curve of consumers’ demand as a function of time into the future up to 2000.
And for that reason the price elasticity of demand was really very small.
The only sensitivity or response to price was still in the choice of the particular energy technology, conversion,
transport or what a certain sensitivity to price came out before the energy reached the consumer.
Then the other two models were quite similar in structure.
The energy technology assessment model of Alan Manne at Stanford University distinguished electric and non-electric demands
and it came to an estimate of -0.25, so absolute value of the elasticity of 1/4.
The Nordhaus model had more subdivisions –
Residential and commercial users,
Industrial
Transportation
Specific electric services that could only be done by electric power.
And he came at an ETA of -0.4.
There was a study done subsequently by the energy modelling forum, an organisation based on Stanford University
but operating nationwide, with a nationwide following and participants of the elasticity values
that had been produced by various studies, by econometric methods, statistical methods as against judgemental estimates.
Now the modelling forum had found a range of -0.4 ~ -0.7 of those obtained by statistical and econometric methods.
The -0.25 in the Manne-model was labelled as a judgemental estimate.
But at the request of other members in the group, Alan Manne, who was himself a member of the committee,
of the modelling resource group and so was Nordhaus and so were two people from Brookhaven,
Kenneth Hoffman and William Marcuse of the group, the estimate that, since the estimate was,
though we didn’t use that term at that time had some judgemental aspects to it.
We asked Manne and he willingly did add a second -0.50.
Then I will go in to the results that came out of the, here we have to look carefully at the definitions of what is on the axis.
I think I first must list the policies here, the policies were those that I had already enumerated, namely,
the base case and then the moratorium only on advanced converter and breather reactors and then moratorium on all reactors.
And then the coal and shale limits.
Those were the first assumptions and then certain combinations,
the moratorium on both the moratoria and the coal and shale limits.
And of those the following diagrams indicate what was found and let’s now read carefully what is on this axis.
Here we have E, the ratio of aggregate energy consumption in 2010 trajected for policy I,
that is the one of that list of five that refers.
And that ratio is set off here, so the more drastic the policy the further the point corresponding to
it will be down on this axis.
The other is the ratio of cumulative discounted GNP for 1975 to 2010, projected again for the same policy and then which model,
that is indicated by the particular symbol that is there to indicate the dot.
Here are the three DESOM responses to policy and that is the only one until we get there
where there is a direct response to some constraining measures with regards to consumption used.
We note then, that all the other points somehow … (inaudible, 26:54) the vertical line.
And that indicates that where the elasticities, that are no longer there on the screen, are moderately high from -0.4 on,
the points remain and the effect on GNP of the constraints on energy use are not severe.
However, in this point here this is the ETA point at -0.25 price elasticity I constrained in quantity
that has an effect this much.
Then also the GNP is constrained.
So here you see it very clearly before us from these measurements that it depends on the price elasticity of demand
in the model that is produced by the model.
Now let me then just, I miscalculated my time somewhat, but let me just then indicate my summary
of what was learnt from this study.
First of all, the smallness on the effect on GNP as long as you don’t get below le’s say -0.5 or -0.4 in the price elasticity
and it is a matter of econometric work to improve the assurance that we can have in reading these estimates.
But I do read maybe a little ahead of myself that the principle conclusion that I draw is
that there is some time left to overcome the problems of widespread concern with the safety of nuclear reactors
including the breather reactors.
And this is important because if we mostly rely on fossil fuels we have to deal also with their side effects,
the acid rain or the CO2 in the atmosphere.
The acid rain mostly from coal but the CO2 problem as much from oil and gas if those are to be the mainstay.
So I read out of this study, first of all it’s a first study of its kind, therefore provisional.
And not to be dogmatic about but second also it indicates that there is enough time to try out alternative methods
of energy generation or rather the mix thereof, we are not under the sword of Damocles.
Thank you.
