Welfare and scientific ethics

[Keith Wilde <kwilde@magi.com>]

I am responding first to Alan Zundel re "sorting out the issues" not
because I think his questions or comments are more important than Michael's
or Shann's, but simply because they are easier for me: 

 ALAN SAYS: "General welfare" is manifestly an ethical concept, and an
ambiguous one into  which a wide variety of more specific meanings can be
read. If this is the  term economists use, then they are either acting as
social ethicists or using  an ethical term misleadingly to mean something
more prosaic such as the  macroeconomic indicators you list (i.e.,
employment, growth, interest rates,  etc.) I say it is misleading because
it ties these indicators to an ethical  term in a way that could influence
people to believe that these actually  comprise true social well-being,
which is a controversial claim as you note.  Are you saying that you wanted
a discussion here among economists as to what  "general welfare" should mean? 

KEITH REPLIES:
To start with your question: I don’t think that was quite what I had in
mind, but perhaps it is inevitable.   
There is no doubt that “general welfare” is an ethical concept, and quite
self-consciously so.  Economics after all was set on its feet by Adam Smith
as a moral science.  When economists address issues of public policy we are
inevitably working in the domain of ethics.  The corner of economistic
thinking called “welfare economics” attempted to make a science out of
public policy evaluation–a science of moral judgments.  The few principles
I recited in response to your “sorting of the issues” represent nearly the
whole outcome of this exercise.  Nevertheless, if ethics is distinguished
from morals by meaning studied, rational evaluation in contrast to mass
convention, then economics is unmistakeably a branch of ethics.  This, in
my view, is how it is distinguished from subject matter of business schools
(how to make money) on the one hand and from political science (how to
manipulate opinion and exert social power) on the other.  Either of these
ancillary domains is more straightforward conceptually, and hence more
amenable to the engineering approach, than is the ethical domain of
economic policy-–as I view the three.

 [Alan posits that "conventional economic theory sees broadened capital
ownership as bad or neutral.  Keith says not necessarily so.] 

ALAN: I was going on my memory of the "Paul Krugman" point made early on; I
thought  the idea was that PK might say broadened ownership is not such a
hot idea and  the aim here was to come up with good arguments to convince
him otherwise.  Maybe I didn't quite catch it. But it seems to me that the
conventional  economist would be most likely to say that either (a) the
distribution of  ownership does not really matter, what matters is the
efficient allocation of  resources, ....
KEITH:  That's about what I had in mind.  It isn't who owns the capital
that counts; it is what is done with the assets.  But what does it mean to
"count"?  Our metaphorical "PK" says that in the absence of any better
criterion, it would have to be some evidence of aggregate growth.  If the
total pie is bigger, there is some chance that everyone could be better
off.  (This is my interpretation of Michael's early comment about
macroeconomic indicators, which also invoked the PK metaphor.)  By
contrast, we might know absolutely that some group is better off (by virtue
of an ESOP for example) yet still be unable to make the judgment that there
has been a gain in general welfare. To do that we must be able to point to
some aggregate measurement which suggests that there is more to go around.
And even if we had such a measurement, we would not be able to say with
certainty that general welfare had been improved. Economists are the butt
of much derisory comment about the misleading nature of indicators like
GDP. The critics seem unaware that the economists who developed national
accounting conventions published these caveats along with them.

ALAN CONTINUES:  or (b) broadened ownership is not a good idea because you
need  people with resources beyond their consumer needs in order to invest,
spreading the wealth means spreading income and the new capitalists will
spend their new income on consumer goods rather than investing it. The
Kelso argument against (b) relies on the idea that investment can come from
credit rather than financial savings.
KEITH:  This is not the one I was thinking of, but yes, it is an objection
that might be raised by some economists.  In my view, the Kelso argument is
an effective counter, but do we have the empirical goods to make a
conclusive case?  That is one of the research questions that we will want
to identify in reporting out the product of our discussions.

ALAN:  I don't...know of a macro-economic argument against (a):
KEITH:  It seems that an effective "argument" has to be empirical.  That is
my understanding of Michael, and is my intent in the paragraph above.
ALAN:  Why does "redistributive impact" trip warning bells...[?]
KEITH:  Because our moral standard is "Don't rob Peter to pay Paul."
ALAN: ... and a bias towards miminalizing it? Transfer costs and distortion
of incentives, difficulty of predicting how politicians will use
redistributive tools, or...? 
KEITH:  In principle (abstracting from individual predilections) there is
no bias toward minimizing redistribution.  There is simply a recognition
that it is very difficult, if not impossible, to make a "scientific"
judgment that the world is a better place after Robin Hood has transferred
a pot of gold from the sheriff's treasury to some homeless peasants in the
forest. 
ALAN: I thought Keynesian economics had made redistribution a tool that
could be considered  as a route to smoothing the economic cycle. It sounds
like it is regarded as something of a dirty word; I'd like to understand why.
KEITH:  The triumph of Keynes was to solve the distributive problem without
direct redistribution.  This was accomplished by making the pie bigger,
through policy instruments designed to stimulate investment, employment and
thereby jobs and incomes. The Keynesian solution was therefore entirely
consistent with the economists' posture that the general welfare can only
be improved unequivocally if there is an increase in aggregate wealth
(income).  But don't forget the big caveat, that even with such an
increase, "growth in GDP does not equate to a higher level of
welfare--whatever that means".
 ALAN:  ...[I]f Ellerman or any of you [economists] endorse any particular
conception of "general welfare" you are in the same situation, making
ethical judgments. 
KEITH:  True. And that is why we are so cautious.  As economists, living by
our methodological code of conduct, it is next to impossible for us to give
an absolutely postive evaluation of any policy.  The evidence is just too
hard to obtain.  We can be much more certain of negative judgments--which
is no doubt why we are so universally loved! As individuals we can have our
enthusiasms about policy instruments and goals, the same as anyone else,
but we cannot expect very often to win the universal assent of our
disciplinary peers that our preference represents the way things absolutely
ought to be. 
Such is the ethical "purity" (I'm not about to let Norm Kurland beat me on
purity!) of economics as a moral science.  You may easily feel a bit dizzy
over this, for economists do other kinds of things that have a much more
familiar appearance of engineering.  Look carefully at these, however, and
you will find them always harking back to this principle of redistributive
purity.
 ALAN:  I agree with you [that the "truly free market" is rhetorical
bullshit].
KEITH: I understood that; otherwise I would not have been so vehement. (And
I don't mean just as the term is used by Kelsonians.)  ALAN: A lot of
Kelso's theory is built on [the notion that in a truly free market] income
shares  would reflect productiveness,  but I have yet to get  an answer as
to*why*. Kelso dismisses actual income shares (skewed towards labor, not
capital, and thus not reflecting relative "productiveness") as a product of
a variety of political interferences (laws stemming from labor union power,
etc.).  I...haven't received a good answer from Bob Ashford or Norm K.
about why,  in their theoretical "truly free market", income shares would
be expected to reflect  productiveness. If they cannot answer this, a lot
of Kelso's structure (the stuff that most antagonizes economists) falls down. 
KEITH:  This is obviously a very important point, and I wish you would
develop it into a short statement, appropriately titled, to start a
separate "thread" in our discussion.
 ALAN:  .... current ownership rules give unmerited wealth to current
owners. It may be that aspects of financing capital formation such as the
provision of credit by the gov't are simply not a matter of merit; no one
(or perhaps everyone) has earned the new wealth and the fruits of it....
Your economist's training shows here in referring to moral judgments as
"preferences" or "biases." I think they are much more than that, and are a
proper subject of our reasoning, discussion and criticism, especially when
our professional work is on subjects that impact public policy. Even a
"preference" for efficiency is an ethical position, after all.
 KEITH:  I  agree with all of this.  Why should access to credit not be
democratized, the same as the ballot box? That is my personal preference,
my moral judgment if you prefer, and I am not constrained by the fact of
being an economist from expressing it.  What I cannot do is demonstrate to
the satisfaction of my professional peers that democratic capitalism is an
improvement to the general welfare--not unless I have the macroeconomic
indicators that Michael alluded to at an early stage of this discussion.
And even then it would not be "scientifically" conclusive.  There are other
standards than those of economists which can be invoked in a rigorous
discussion of ethics, and I welcome such interaction. The only caution I
raise is that when we do go into that broader territory we will be losing
our "purity" as economists--which could go against us in promoting COG
objectives as "economically sound".  We will be talking about economic
policy, for sure, but will not be conducting it in terms of economics. That
would not prevent us from coming to agreement on a policy preference at
some point, and noting that economistic reasoning had been one of our
avenues of approach.

Keith Wilde
Canada Pension Plan
Ottawa
kwilde@magi.com
613 990-8125 (office)
613 747-6847 (res)