Hello Steve,
Sorry to be so late replying but I have been away. This will be a little
long because it appears that many do not know how public analysis is and can
be done. In the following I will try to explain my position more clearly.
I hope to be at least somewhat successful. IMHO is applied
implicitly/explicitly to everything below. Forgive me if I think that I am
right! :-)
It is *not* necessary for the analyst of public issues to determine, or find
some method to reflect, private information such as the individual value set
or preference function.
The problem with, for example, analysis of water resource projects is to
identify the problem(s) and the conflict set related to those problems.
With that information, the analyst looks at the several alternative
proposals to determine how they each will impact the particular conflict
sets. When provided with the information about alternatives and their
impacts, the individuals separately and collectively apply their values by
supporting or opposing any or all alternatives proposed. Since public
projects result from political decisions, the separate public positions are
reconciled in the political arena. That is why it is said that "politics is
the allocation of values."
Take a real world problem: the management of flow in the lower Missouri
River which has been newsworthy in the US recently. For years that river
has been dammed, controlled for, among other purposes, flood control and
navigation. Consideration is being given to changes in the rules by which
the dam operators control the flow from the upstream reservoirs.
Two problems to begin with: 1) Upstream states, South Dakota, etc. want
the reservoirs operated to maintain water levels during the lucrative
recreation season. 2) Navigation interests downstream want them operated
to preserve navigable river levels. A couple of associated problems: 3)
farmers downstream operating in the flood plane would be flooded by some
scenarios during farm season. 4) different scenarios have different
effects, some positive and some negative on riverine habitat and thus
wildlife.
The analysts hold public meetings throughout the region to give the public
information technical aspects and solicit public input on defining the
several problems (including those not readily seen by analysts) and
interests. Those meeting are held at intervals throughout the study and at
the draft stage. Based on the information expressed relevant to the
particular problem, individuals can draw their own conclusions and pressure
their representatives to support particular choices. The market oriented
b/c is important but greatly insufficient for public political choice.
Ultimately, the politicians will decide. No doubt, in some (many?) cases,
the politician will incorporate information/biases not immediately relevant
to the particular project under study. That's politics! The old days when
a simple ratio of b/c was all that was necessary made life very simple for
politicians. All they need was a favorable ratio and if the project was
important politically for them they would apply pressure to get one. Now
they should and must be required to take a serious look at the different
interests - representing different values - and more truly weigh conflicting
interests and values. IMHO.
At no time is it necessary to define individual preference functions. In a
democratic society (ideally) we give the individual the information they say
they need and they excercise their own individual preferences whether
revealed or not.
This particular project has some small but interesting political
implications nationally and even perhaps internationally. Senator Daschel,
Democratic majority leader of the US Senate is from South Dakota which wants
one type of regualtion; Senator Lott, Republican and minority leader of the
Senate represents a downstream state which tends to support another.
President Bush, Republican, supports a Republican for the South Dakota
Senate seat to be voted on this fall. Daschel, of course, supports the
Democrat contender. Well, there appear to be many political axes in play
for this mid-term election in 2002 and for the Presidential election in
2004.
I am long familiar with the literature to which you refer. In my view and
experience it is not applicable to the issue represented above.
Ray
-------------------
You wrote in part:
> My criticism actually was quite similar to yours. You mention that
> people have differnt value sets. Fine. I agree with you. What I am
> saying is going even beyond that. That people have different value
> sets and that they are unobservable. Therefore it is not possible to
> simply observe how much people value a given project. Lets suppose
> we have a project, call it Project Ray (in your honor). Now, we have
> three people (for simplicity here).
>
> Typically with benefict/cost (b/c) analysis we would look at the
> value of the project over time and discount it. What if the three
> people all have different values? Which one do we use? Do we add
> them up? Take an average? In any event, how do we determine exactly
> how much they value the project? Ask them? Well they could lie.
> Those who want it badly might overstate their value and those who
> don't might understate it. Also, if the project is going to be
> financed through tax dollars this could have an impact on people's
> responses.
>
> Getting people to reveal their true valuation of the project is quite
> problematic.
>
> Here is something from Jean-Jaques Laffont's text, _Fundamentals of
> Public Economics_.
>
> "A fundamental characteristic of economic systems is decentralizing
> of information. When a collective decision must be taken, whether it
> be the choice of the level of a tax that internalizes an externality,
> the choice of the level of a public good, the choice of the level of
> output for a firm exhibiting increasing returns, or even the choice
> of the social welfar function, the social decision maker must collect
> information that by its very nature is decentralized since each agent
> posses aspects of this information that he alone knows. A problem of
> strategic behavior arises with respect to this private information."
>
> Steve here: In short, what Laffont is saying is that I know
> something and when you ask me I am going to act in a way that
> benefits me the most. That is I am going be behave strategically
> when giving my answer.
>
> Laffont continues:
>
> "In other words, can a game be found for which the strategic
> equilibria lead to the desired outcome?"
>
> Steve again: That is using the structure of game theory, can we
> resolve the problem noted above. The answer is in some regards yes.
>
> The Revelation Principle (Gibbard 1973): Let (g,M) be a mechanism
> that implements the social choice function f(.) for the dominant
> equilibrium concept. There there exists a direct mechanism that
> implements by revelation f(.) in dominant equilibria.
>
> So far so good. The problem is that these problems are very, very
> complex. Even simple examples quickly become quite cumbersome.
> Further, many people might have problems with revealing certain types
> of information.
>
> Now lets jump over to the chapter on Cost-Benefit Analysis.
>
> This chapter starts off with the following:
>
> "First we consider an economy with two goods and we assume that the
> consumer has the following quasi-linear, quasi-concave, increasing
> utility function:
>
> u(x,y) = x + v(y).
>
> There is no income effect on the damand for good y; furthermore, we
> choose a normalization of the utility function yeilding a constant
> marginal utility equal to 1 for good x."
>
> Steve here: The last assumption can be thought of as treating good x
> as an aggregate good according to the Hicks-Leontief theroem.
> Anyhow, the problem, I was commenting on is that that is a pretty
> damn stringent restriction on the utility function above. It does
> not give you lots of room to incorporate lots of different "value
> sets". We are stuck with that specification for the utility
> function. Further, what do we know about v(.)?
>
> Now the general case works with the indirect utility function, V(p,R)
> (where R is income, and P the vector of prices). However, Laffont
> closes the section for the general case as follows:
>
> "In concluding this section, we assert that the concepts of
> compensating or equivalent income variation have a precise meaning
> for INDIVIDUAL demand functions. The economist who possesses the
> APPROPRIATE INFORMATION CONCERNING INDIVIDUAL DEMAND FUNCTIONS can
> provide an approximation to these variations. However, information
> IS NOT AVAILABLE IF ECONOMETRIC ANALYSIS FURNISHES THE APPLIED
> ECONOMIST WITH ONLY AGGREGATE DEMAND FUNCTIONS." (caps for emphasis)
>
> So what we see here is a need for massive amounts of data and
> micro-analsysis (at the individual level so you could really call
> this micro-micro analysis). Even still the results could be wrong.
> (We are talking estimates only here). So the idea that we can
> somehow arrive at the "Truth" using cost-benefit analysis is doubtful
> at best.
>
> I suspect similar criticisms can be levelled at the Multiple
> Objective Planning type of analysis as well. I do admit I don't know
> much at all about that methodology. I have only looked at a few
> things the other night on-line. However, from what little I have
> read it seems that it is an attmept to go beyond cost-benefit
> analysis and include other factors ("non-economic" factors such as
> social and environmental factors). While this is a good thing, it
> still, from what I have read, does not seem to solve the problems
> noted above, of course, since I have only just started reading that
> literature I could be wrong.
>
> Does this mean that cost-benefit analysis is just a simple waste of
> time? That it should never be done? No, I think that is an extreme
> conclusion.
>
>
> > Maybe Stevie boy, you should stay *under* the porch since your
> > "pearls of
> > wisdom" merely regurgitate the received wisdom and you seem not
> > competent
> > enough to get outside the box.
>
> Would you stop it with this "outside the box" stuff. You are
> starting to sound like those know-nothing consultants that come
> through say this and virtually nothing else. Besides, "thinking
> outside the box" is a waste of time IF the solution to your problem
> is inside the box.
>
>
>
> > Tell me boy, how do you define "environmental ethics"? Or do you
> > even have
> > a clue?
> >
> > It seems you cannot discuss controversial issues without insulting
> > the
> > other. Maybe you should examine your ways of operating in an
> > interpersonal
> > way with those who do not share your *simplistic* world view. Or
> > do you not
> > understand that others see themselves within the world in much more
> > complex
> > ways than do you.
> >
> > You have alot to learn boy. And do not deceive yourself that your
> > infantile
> > maunderings will lead me to fall back in my views. I have dealt
> > with folks
> > of your persuasion who have much, much more moxie than you have.
> > And I
> > learned to respect them. But not you boy. As I said before: "grow
> > up".
> >
> > And, as I said before, WHERE ARE THE 2 LIST OWNERS?? WHY DO YOU
> > CONDONE THE
> > INFANTILE, OBNOXIOUS, INSULTING posts from this infantile, wannabe
> > intellectual??
>
> Pot, kettle black. Look at yourself Ray. How many times did you use
> the diminutive term 'boy' in that last two posts?
>
> Steve
>
> __________________________________________________
> Do You Yahoo!?
> LAUNCH - Your Yahoo! Music Experience
> http://launch.yahoo.com
|
