COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	%politi[w90,jmc]		Politics for engineers and scientists
C00010 ENDMK
C⊗;
%politi[w90,jmc]		Politics for engineers and scientists

Remarks:

One can read political journals for years and never realize that
the advances in human welfare over the last few hundred years
are almost entirely consequences of advances in technology and
science.

An engineer looking at the failures of government will ask why
we can't have social engineering with the same standards of
objectivity, basis in science and fairness in controversy as exists
in engineering.  Here's why we can't---at least for now.

	There are two main kinds of politics---democratic and dictatorial.
Dictatorial politics, at least in its ideal, more resembles the
decision processes of engineering.  Someone has authority and makes
the decisions, getting such help and advice as he finds useful.
This system doesn't work as well as democratic politics in spite of
the disadvantages of the latter, which we will get around to recounting.
Here's why dictatorial politics doesn't work well.
[must I distinguish at this point the economic and political systems].
For the time being, I'll ignore the distinction between political and
economic policy.

There are two reasons---ignorance and corruption.

	Ignorance is the reason most emphasized by the critics of
socialism---whether it be the all-embracing socialism of the communist
countries or the government programs of the industrial democracies.
The pointy-headed bureaucrats don't know enough to make the correct
decisions that are made by the market.  This is subject to the counter-argument
that mathematics and computers and automatic data collection might
fix this.  After all prices are just a vector in the dual space
of the production vector.  Optimizing production will optimize
prices as well.  Perhaps the socialists have just not had good
enough computers, good enough mathematicians and good enough ways
of collecting the data.  My opinion is that this argument has some
merit---improving these matters could result in better decisions---
maybe even good enough if it weren't for the corruption.

However, there are some other information problems.  First comes the
problem of judging welfare of individuals.  Thinkers of a socialist
inclination come up with measures of welfare.

The counter argument is given by the caricatured dialog going back
at least to the 1930s and probably much earlier.

Communist: Comes the revolution, we will all have strawberries and
cream.

Citizen: But I don't like strawberries and cream.

Communist: Comes the revolution, you will have strawberries and
cream and like it.

We put the argument more generally with the slogan:

{\it Each man is the best judge of his own welfare.}

This means that people should be paid money for their work and
have as wide as possible a choice in how to spend it.  Exceptions
should be very limited.

The country will survive requiring everyone to pay a share of the
costs of insurance against individual economic disaster.

Note this tricky formulation designed to evade libertarian arguments.
The libertarian will say, ``Let the individual buy insurance if
he wants it.  If he doesn't buy it and disaster strikes, let him
depend on such voluntary charity as his fellows offer.''
I don't say that the libertarian is wrong, but I do say that society
is rugged enough to afford some deviation from libertarian principles.
This isn't the main cause of social inefficiency.

Not only is each individual the best judge of his own welfare,
but he is also the best judge of what work he can do given the
available opportunities for a person of his talents.  This has
even deeper anti-socialist implications.  Capitalist society
has developed an enormous variety of social roles.  Consider
the fact that in America, long distance truck driving is divided
between drivers who own their own trucks---costing as much as
\$200,000---and drivers who are employees.  The former is more
recent, but neither form of truck driving seems likely to displace
the other.  Consider the recent development of chicken farmers
who don't own the chickens or buy the chicken feed, but board the
chickens for \$.0025 per chicken day.  One can now be a competent
chicken farmer without having to be a competent speculator in
grain and chicken prices.  Neither of these options was the
invention of an economics research institute.  Contrast this with
the Soviet situation in which the graduating class of a university
is divided among the institutions that employ graduates, paying
some attention to the graduates' preferences.

As a researcher in artificial intelligence, I'm inclined to believe
that some day computer programs might be able to give good estimates
of individuals' welfare and talents, but this is unforseeably far
in the future.  In the meantime, economic calculations using
Leontiev type production functions are worthwhile but far from the
whole story.

However, ignorance isn't the main problem that damages socialist
economies; it's corruption.