🚀 go-pugleaf

	This post is a long one and for that I apologize.  However, I think
any readers may find it interesting in some ways.  First, a bit of explanation
to give some context.  

	This part of the thread began in the midst of a discussion about how 
to combine neoclassical analysis with exogenous institutional factors. 
I discussed how some strongly institutional research can go astray when it
ignores well established neoclassical explanations (not that there aren't 
many examples of the reverse) and I went on to give an example of just such
failings in Francis Fukuyama's latest book, _Trust:  The Social Virtues and
the Creation of Prosperity_.  My comments were based primarily on a review 
of the book by Robert Solow confirmed by my own study of the book.  After 
making my specific points about some of the book's shortcomings, I 
descibed it as something like a "waste of perfectly good paper."  

	Chris Swenson of RAND was evidently quite wounded by my remark and 
replied with some passionate invective (although without substantively 
challenging my remarks or Solow's review).  From his several posts in reply to
me, I figured that Mr. Swenson had picked up the book and found it not only
well worth the purchase price but perhaps one of the great analyses he'd ever
read of current affairs and economics.  I began to wonder if I'd been too 
hasty in my examination of Fukuyama his work and my acceptence of Solow's 
analysis so I gave _Trust_ another look today.  

	I really didn't know anything about Fukuyama beyond that he'd written 
_The End of History_ so I turned to the preface to get some feeling for what 
his background was.  I was considerably surprised to see that Fukuyama is some
sort of senior fellow at RAND.  My surprise came from Mr. Swenson's failure to
mention this in any of his posts, which is doubly odd because Mr. Swenson has
in the past been in no way reluctant to name drop from the list of notables at
RAND.  But now Mr. Swenson's passion for the book seemed somewhat clearer, in
defending Fukuyama, he was defending RAND and maybe Fukuyama was even an
acquaintance of his.  I admire loyalty and, not having any axe to grind 
against Mr. Fukuyama, I was about to admit grudging respect to Mr. Swenson
for wading into sci.econ in an attempt to defend RAND and Mr. Fukuyama.  Then
I read a little further....

	Fukuyama writes, "Much of the material in this book would never 
have come to my attention but for the hard work of my research assistants, 
Denise Quigley, Tenzing Donyo, 
                              and especially Chris Swenson

who was of invaluable assistance throughout all phases of this study."
(Emphasis added.)

	So rather than being an objective reader who was drawn to the book by
its merits, Chris Swenson was the lead research assistant on it.  Rather than
coming as a disinterested party who hoped to alert sci.econ to the unexpected 
pleasures of this book, he is instead someone who has never shelled out the
$25 in the hope of finding some serious modern economic research only to 
experience chilling disappointment.  Why in his many posts has Mr. 
Swenson failed to mention this?  My mind races with competing hypotheses.
Here are the four most likely.
	1)  He's a shill.
	2)  He's embarrassed by his part in the charade of selling this book 
under the guise of being about economics.
	3)  He knew he had no good answer to the criticisms and thought that
to respond as best he could while admitting to be the head reseach assistant
on the project (rather than just being a reader and fan) would cost the book 
any remaining credibility it might have had.  
	4)  He meant to but got distracted by some work he was doing on the
level of trust in American society.

	I will come to the point.  In this book, Francis Fukuyama attempts to 
show how specific institutional factors improve our understanding of the
preformance of the economies of various nations.  I must agree with Solow and
probably most, if not all, the readers on this group that this is a worthy
goal.  However, Fukuyama fails to specify how conventional economic factors
(such as captial accumulation or labor force change) fail in this and how his
suggested areas of study improve the fit of any model.  As such, his book is
a model and empirical work free zone.  (It recalls the old (and unfair) joke 
about what sociologists call the plural of anecdote.  "Data.")  That Fukuyama
falls into this trap is a shame since good data exists in abundance for the 
standard economic variables for many of these nations and he could have shown
how his various trust and social capital stories explain why some nations 
under or over-perform what is predicted from standard economic analysis.  
If he had been successful at this, it would have been an eye-opening piece of
work, instead of the opposite.  

	Lest Mr. Fukuyama's RA attempt to repeat the claim that I am
misinterpreting the book, let me quote some of it's economic claims.
Do note that the title of the book is, again, _Trust:  The Social Virtues
and the Creation of Prosperity_.  As such, Fukuyama claims to explain how
nations achieve prosperity.  Prosperity is usually defined as "economic 
well-being and Fukuyama offers no alternative definition.  In the preface, 
Fukuyama claims that it was only natural that he would follow up his last book
"with a book about economics."  Further, "[I]t sees to me that the emphasis 
on economics is almost inevitable."  "But veritually all political questions
today revolve around economic ones..."  In many places, he claims to explain
what standard economics does not (as is amplified below by his RA).  
What is the thesis of his book?  I think it's best stated by this paragraph 
from page seven.

		"Thus economic activity represents a crucial part 
	of social life and is knit together by a wide variety of 
	norms, rules, moral obligations, and other habits that 
	together shape the society.  As this book will show, one 
	of the most important lessons we can learn from an examination 
	of economic life is that a nation's well-being, as well as its 
	ability to compete, is conditioned by a single, pervasive 
	cultural characteristic:  the level of trust inherent in 
	the society."  (p. 7)  

	Fukuyama fills his book with statements that his "trust measures" 
are important economic variables on par with standard neoclassical ones. 

		"Social capital is critical to prosperity and to what
	has come to be called competitiveness, but its more important
	consequences may not be felt in the economy so much as in 
	social and political life."  (p. 355)

		"The social capital needed to create this kind of 
	moral community cannot be acquired, as in the case of other
	forms of human capital, through a rational investment 
	decision.  That is, an individual can decide to "invest"
	in conventional human capital like a college education, or
	in training to become a machinist or computer programmer,
	simply by going to the appropriate school.  Acquisition of 
	social capital to the moral norms of a community and, in its
	context, the acquistion of virtues like....  The proclivity
	for sociability is much harder to acquire than other forms
	of human capital, but because it is based on ethical habit, 
	it is also much harder to modify or destroy."  (p. 26-27)

Nowhere does Fukuyama back up the comparability of his favored variables
with the usual economic ones.  He does venture the following odd statement.

	"We can think of neoclassical economics as being, say, eighty 
	percent correct:  it has uncovered important truths about the 
	nature of money and markets because its fundamental model
	of rational, self-interested human behavior is correct about 
	eighty percent of the time.  But there is a missing twenty 
	percent of human behavior about which neoclassical economics
	can give only a poor account."

Do models that include Fukuyama's suggested added variable perform 20% better
than do standard neoclassical models?  He never gives any evidence that they
do.  (Where does he get the 80% figure anyway?  Did his RA give it to him?)

	If the stories Fukuyama tells about the importance of variables like
trust, social capital, and spontaneous sociability are right, then the nations
he identifies as having high levels of these things should have unusually
high returns on their other measured economic factors and nations with low
scores for these trust variables should have low returns on their labor and
capital.  Fukuyama makes great claims for the power of these variables but 
to back them up he offers only stories.  This is a shame because empirical 
tests would not be that tough, but institutionalist that he is, no neoclassical
analysis is done (in spite of the above lip-service quote).  However, 
given the empirical papers cited and explained in Solow's review, it's 
unlikely that the variables described by Mr. Fukuyama would have any
measurable impact.  

	Perhaps I am still in considerable confusion over Mr. Fukuyama's
intent and accomplishments with his book.  I confess to only having an
arm-chair interest in this matter (literally having spent two afternoons in an
arm-chair with the book).  If someone with a professional stake in the work, 
like Mr. Fukuyama's RA, would like to clarify things, that would be great.
He could start with a rebuttal of Robert Solow's very specific criticisms from
his review.  Although I've referred to it a few times, Mr. Fukuyama's RA has 
never gotten around to answering it.  Just to be up front with everything, 
I'll repeat the reference.


Solow, Robert M.  "But Verify", _The New Republic_, September 11, 1995, pp.
36-39.


Additionally, Mr. Fukuyama's RA could try to re-direct his replies to my
comments to their substance.  Not that a touch of flame, now and then, 
isn't highly enjoyable.  What follows is, IMHO, a nice mix.

In article <46od1c$c1h@rand.org>, Chris Swenson  <swenson@rand.org> wrote:
>
>Before I begin with my comments, I would like to express my extreme 
>disappointment.  Instead of an intelligent discussion over Francis 
>Fukuyama's latest contribution to social science (*Trust*, Free Press, 
>1995), what we have here is a mere exchange of insults.  This is too 
>bad, because a reasoned debate over the merits of Fukuyama's book and 
>the growing field of socioeconomics could have been extremely edifying.
>Is there a more temperate soul out there who has some thoughts about 
>Fukuyama's ideas on socioeconomics from an economist's standpoint? 

Hmm, fishing for complements perhaps?  Would you define "an intelligent 
discussion" as one confined to "merits" of this "contribution to social
science?"  Would a "temperate soul" be someone else with a vested interest
in the book?

>
>In article <4647rj$ch4@news.acns.nwu.edu>, @casbah.acns.nwu.edu (Mark 
>Witte) wrote:
>
>
>>My reading of the book was that Fukuyama is trying to explain the 
>>levels and growth rates of economic output across nations...
>>
>>He does give lip service to neoclassical determinants of these but then 
>>he very oddly ignores their explanatory power.
>
>
>As I have said before, you obviously haven't read his book (You even 
>admitted as much in a previous posting, saying that you only glanced at 
>it while you were at the bookstore.  Browsing does not an expert make). 

Well, I guess you are unfamiliar with the policies of Barnes and Noble's where
comfortable chairs and reading tables fill large areas of the store.  My
exposure to the book came after I'd read the review and I had a day to kill
while traveling and waiting for a friend to get out of work.  Having read
Solow's review, I was a little curious about the book and had hoped that it
would still be interesting and worth buying.  I found that Solow was pretty
accurate and that the book was not worth the money (which I spent instead on
darn good munchies from the Starbuck's inside the bookstore.  I ended up
spending over $100 on books that day but not on _Trust_.  The protracted length 
of this discussion (and our library's slow pace at getting it on the shelves)
has led me to buy it.  If any readers out there have an interest in the book 
and can't get it from the library, try to find a bookstore with a liberal
return policy so that you won't be stuck with this thing on your shelves.  
Of course, this may reduce any residuals that may be coming to Mr. Fukuyama 
or his RA but that's life.

> You simply read one book review by Robert Solow and think you know 
>exactly what Fukuyama is saying.  Well, you do not.

I may not, but then I've only read the book and a review.  Maybe you have a
better insight to what was really meant as opposed to what the book says.

>
>For example, you go on and on and on again about growth.  Although 
>Fukuyama makes some references to growth, the main point of his book is 
>not to show how neoclassical models of growth are inadequate and how 
>adding cultural factors will improve the accuracy of its models.  Still
>unclear about that?  It's not about models or their predictive ability.
>Fukuyama readily admits that these models to a pretty good job already.

	Well, I go on and on about growth and levels.  Fukuyama says 
that "a nation's well being, as well as its ability to compete, is 
conditioned by a single, pervasive cultural characteristic:  the level 
of trust inherent in the society."  In his title, he links trust to 
prosperity.  He repeatedly compares his trust variables to standard 
economic variables and uses them to explain aspects of international 
trade.  We can see national outputs and some factor inputs.  Where is
effect, as express in international comparisons, of these trust variables
on output (the importance of which he alludes to repeatedly) beyond the
contributions of standard economic factors?  If Alwyn Young were to 
claim that trust and national industrial structure don't matter, national 
prosperity is completely explained by twenty-five years of quarterly data on 
measured human and physical capital variables, how would you answer him?  
Would you tell him a story about An Wang?

>
>But neoclassical economics and its rational-actor component does not do 
>a good job explaining *all* economic behavior. 

	True.  There is something that does attempt to explain all behavior.  
It's called religion.

>
>On this point, Fukuyama says that economists might want to pay a little 
>more attention to what is going on in the field of socioeconomics, 
>because some of the theories emerging from this new field have 
>explanatory power in their own right (Developing some impressive 
>theories himself, Fukuyama contributes a great deal to this growing 
>field). 

And if you were to disagree with this last statement, how would you pay your
next month's rent?  Fukuyama is one amazing guy, he does all this with so
little help.

>See the relatively recent scholarship by James Coleman, or the
>work of Mark Granovetter on embeddedness if you doubt me on this point. 
>I say "relatively" because the field of socioeconomics is not new. Max 
>Weber's *The Protestant Ethic and the Spirit of Capitalism*, written 
>many, many years ago is considered to be the pioneering work in the 
>field.

Yes, Weber did some great stuff.  However, empirical testing has come a long
way since then.  You might want to see the recent scholarship by anybody 
who does econometrics.

>
>What kind of questions can socioeconomics help us explain?  For 
>starters, why are Chinese firms, on average, smaller than their Japanese 
>or American counterparts?  Neoclassical economics doesn't do a very good 
>job of explaining this. 

Really?  You've tested models of firm size based on the age of the
firm?  (It would seem reasonable with the economic change in China, that 
most of its firms would be newer than the really big firms in Japan or the
US.)  Have you looked at factors like capital market constraints and risk
levels?  (Firms in high risk countries like China often have a hard time 
raising external funds for expansion relative to firms in countries like the
US and Japan.)  Have to tested other institutional factors like Chinese
interpretations of Marxist thought that until recently limited many firms 
to fewer than eight employees?  

>Socioeconomics, on the other hand, does.  
>Fukuyama says that Chinese firms are smaller than others because of 
>Confucian values that prevent outsiders from assuming control of the 
>family business. 

And as I asked last time, what about Confucian Taiwan?  It was the effects 
of activist government that make the difference there?  Is China really 
out of step with other nations at similar states of economic liberalization 
in terms of the size of its privately developed firms?  Comparing China 
now with Japan and the US now is like comparing an acorn with oaks.

>As an example of this, Fukuyama shows how An Wang, the 
>CEO of Wang Laboratories, in traditional Chinese fashion turned the 
>massive corporation over to his incompetent son.  Thus, Fukuyama is 
>arguing that in trying to explain Chinese industrial composition (lots 
>of small family firms, very few large ones), we might have to look 
>outside of neoclassical economics.

Ah, the An Wang thing, yet again.  What is this, the third time?  Fourth?  
You seem very proud of it.  Did your personal "hard work" cause this little 
socioeconomic factoid to "come to [Fukuyama's] attention?"  Did he single 
out his RA for "especially" lavish praise for this?

>
>As you can see, much of Fukuyama's discussion is at the "micro" level, 
>removed from discussions concerning the intricacies of GNP growth rates.
>But this does not mean that a "micro" discussion has little relevance in
>the current debates over policy.  In our discussions of the trade 
>deficit with Japan, for example, wouldn't it help to understand the 

"Our trade deficit with Japan."  This phrase alone speaks volumes about the 
level of economic sophistication and understanding that went into this book.
It's a real shame that this comes out of RAND, home of Charles Wolf who 
writes with such care on macroeconomic trade issues.  (Of course, he's out on
the West Coast.)

What about our trade deficit with Holland?  Our trade surplus with Canada?  
Is it because "Canadian businesses were less vigorous than their American
counterparts?"  (Here the citation is from Engels, known for his careful 
study of data.)  Are you saying that your work consistently predicts trade
surpluses and deficits?  Or is your work on the level of, 

	"Well, the US has a bi-lateral trade deficit with Japan.  
	Japan has Keiretsus.  That must be the cause.  Well, let's 
	test the theory.  How do the countries with Keiretsus do
	in trade with the US?  Whoa!  100% of the countries with 
	Keiretsu's have trade surpluses with the US.  I knew it!"

Why do you think bi-lateral trade balances matter?  How do you know that 
Keiretsu' matter?  Have you read the work of Fumio Hayashi or Larry 
Christiano (Federal Reserve Bank of Minnesota Quarterly Review from a few
years back, you're an RA, you find it)?  Have you considered demographic
determinants of the Japanese savings rate?  And ours?  Levels of government
deficits relative to domestic saving?  If so, do you really find Japan
anomalous?  Do check the work of Charles Wolf, he's a damn clear writer
and marshalls the data convincingly.  From the framework in his models, you
could see if nations with Keiretus or chaebols really behave any differently
given other recognized factors.

>Japanese Keiretsu?  (the large networks composed of numerous firms 
>spanning several Japanese industrial sectors).  It seems as if the very 
>existence of a Keiretsu is a "non-tariff barrier" to free-trade in 
>itself.  Again, neo-classical economics can tell us very little about 
>this.

Certainly the neoclassical work in these areas is little cited in your book.
Start with the work of Paul Krugman (although one of his chatty opinion pieces
does merit a reference in _Trust_.  I'll bet everyone can guess which one
given the recent discussions on this group.)

>>Having had a chance with both the book and the review, I am in some way 
>>worse off than someone who seems utterly unfamiliar with the work of 
>>Robert Solow?
>
>I am not unfamiliar with the work of Solow.  Furthermore, I never 
>referred to one of Solow's books as a "waste of perfectly good paper".

Me either, but that's because they're pretty good.  Have you ever read them?
Or any of his articles?

>You on the otherhand, made such a comment with regard to Fukuyama's 
>book.  Furthermore, you made this comment after having only read Solow's 
>review and not the book itself (although now you are telling us that you
>browsed through it in the book store).  But as your comments in this 
>thread show, you have no idea what Fukuyama actually said in his book.

	I've been up-front all along about my contact with Fukuyama's 
book.  I'm afraid that it was Mr. Fukuyama's RA who distorted matters 
there.  As a result of this tread, I've had to deepen my study of the book
and I am sorry to say that my original assessment still stands.  When I
originally raised _Trust_ as an example, I cited Solow's criticisms because
I wanted to give him the credit for the work he did.  If I was wrong to cite
that review, please explain why.  A review of a work that claims to be
scholarship isn't like a "thumbs-down" movie review that can be shrugged
off.  If Solow's review is inaccurate or unfair, please spell out why.  
For that matter, explain why someone couldn't learn more about the roots
of national prosperity by reading Solow's discussion of the papers he 
cites in his article than from plowing through 420 pages of _Trust_.

>>>>And how *you* could continue to deny that the most basic facts about 
>>>>a society (such as its religion or its culture) might have even a 
>>>>small effect upon a nation's economic performance is beyond me.
>>
>>Whew! I am glad you ephamsized "you" there...
>
>You accuse *me* of denying something, I accuse *you* of not reading his 
>book and unfairly insulting Fukuyama -- ya see how that works?  Wow, I 
>didn't think that *emphasis* was such a *hard* concept to *understand*.

	Actually, I was just laughing at your profligate use of the asterix
key.  I guess some people think that emphasis is a substitute for content.  

	Anyhow, I have never denied that institutional factors are important. 
I have made the point that very poor work can be done when other economic
factors are ignored.  To support this statement I will make up an example and
cite one actually published piece of work.  

	The example:  There are two cities near the one where I live.  Many of
Skokie's citizens are Jewish.  Many of Kenilworth's are Protestant.  The
residents of Kenilworth tend to spend much more per capita than do the
residents of Skokie.  You could explain with with religious variables.  You 
could explain it by noting that the residents of Kenilworth are much richer. 
Both explanations might be right (or not) but without controlling for the 
obvious economic variable the religious explanation has no value.  

	The citation:  Francis Fukuyama's _Trust:  The Social Virtues & the
Creation of Prosperity_  (Who didn't see that one coming?)

>>let me repeat that I have never suggested rejecting institutional
>>explanations for growth out-of-hand. Call me unimaginitive, but
>>I think we should begin with well understood economic factors, 
>>before we look elsewhere.
>

As you can see, I've consistently made this point.  You've consistently 
missed it.

>
>Again, the main point of his book is not to explain GNP growth rates. 
>And again, neoclassical economics can't explain the Japanese Keiretsu, 
>the Chinese family firm, nor the Korean chaebol.  Historical, Political,
>and Cultural factors go a long way in explaining these phenomena.  

Maybe they do.  I can believe that.  Buy you've only asserted it, you've 
shown nothing.  Hence my "waste of paper" comment.  If you can show how
your explanation beat the neoclassical ones (and they might, if you'd 
try) then I'd be interested and it would be a good piece of work.

>However, this is not to say that cultural factors have no bearing on 
>economic growth.  You want an exact measure of their contribution (nay, 
>you want an exact measure of the *other* factors before you measure the 
>cultural and political ones), and you think this is where scholars 
>should start.
>
>I disagree.

Fukuyama claims to be able to describe nations as having high or low value of
his various trust variables.  Within these groups, do factor driven growth
models perform differently?  A test of your theory is that easy.  Fukuyama
has already described all the exactness you need.

>
>The first step lies in convincing the discipline of economics to realize
>that *some* aspects of economic life might be better explained with 
>socioeconomic theories.  Given your insults and hostility, it looks like 
>we have a ways to go.

Is that the royal "we?"  Given your understanding of empirical work, 
modeling, and intermediate (I'm giving you some credit here) economics, 
it's no wonder books like _Trust_ or the work of guys like Paul Kennedy and
James Fallows get taken seriously.  (Not to mention some of the people we see
elected or supported for high office.)

>>You should probably read or re-read the sections where he does attempt 
>>to classify, in an ordinal sense, countries according to these 
>>variables.  It would seem to me that he feels he can quantify these 
>>factors.
>
>First of all, the discussion of the "social capital" in various 
>countries makes up about 80% of Fukuyama's book.  This comment only 
>reinforces the notion that you are thoroughly unfamiliar with his work.

Uh, read what I wrote then read the book.  I really don't find that he spends
80% of the book attempting "to classifly, in an ordinal sense, countries
according to" there relative levels of trust as defined by things like
"social capital."  The point is that he does classify these countries (e.g.
Germany high, France low) and that's all you would need to test whether
productivity residuals behave differently in different countries.

Forgive me if I have trouble taking you seriously when you go throwing 
around the "80%" figure again.  

	"How much does neoclassical economics explain?"
	"Oh, I'd say, 80%."
	"And how much of my book do I devote to describing social
	capital?" 
	"Well, 80% is my best estimate."
	"Hmm, it's getting cloudy, do you think...oh, never mind."

>
>And as your statistics teacher failed to teach you, ordinal data can't
>be plugged into macroeconomic models as you suggest.  They merely imply 
>the notion of "greater" or "lesser".  True to the concept of "ordinal 
>data" Fukuyama categorized countries into two groups: those that had 
>little or no Trust and those that had a great deal of Trust in their 
>economic behavior.  The differences between the Chinese, who 
>traditionally don't allow outsiders to run the family business, and the 
>Japanese, who not only allow outsiders to run the business but adopt 
>outsiders into their own bloodline, fits well into this category.

Your misunderstanding of how to handle qualitative variables is noted.  
I'm sure I speak for many on this group when I ask, "Exactly what are the 
qualifications that RAND looks for in hiring RAs?"  

>
>>As for pity, I hope your interest in Fukuyama is purely personal. I
>>would feel awful if someone paid you for this quality of analysis.

Oops, I guess, given what I learned today (and I'm completely honest about
this, this was not a set-up), this must have hurt.  I'd say I was sorry but
I'm trying to stay with the honest approach.

>
>I'll leave the newsgroup with the above selection to remind them of how 
>this thread started -- with one of Mark's insults.  

Well, this isn't quite how the thread started.  I hope the tread doesn't end 
before Mr. Fukuyama's RA attempts to reply to the substance of my comments 
and the points raised in Mr. Solow's review of the book (which was how this
tread actually started).  

>Instead of responding to his immature insults, 

Yes, it's always a problem, how to best respond to immature insults?  I find
that the "I'm rubber and you're glue..." gambit is generally quite effective.  

>I once again ask: Is there anyone 
>out there you would care to have a reasoned, mature discussion on the 
>emerging field of socioeocnomics and in particular, Fukuyama's book?

Any sycophants in need of a little practice?  

>
>  --------------------------------------------------------
>  | Chris Swenson                                        |
>  | Resident Consultant                                  |
>  | The RAND Corporation                                 |
>  | 2100 M St., N.W., Washington, D.C. 20037-1270        |
>  | Tel: (202) 296-5000  Fax: (202) 296-7960             |
>  | E-Mail: swenson@rand.org   URL: http://www.rand.org/ |
>  --------------------------------------------------------
>  | RAND is a nonprofit institution chartered to improve |
>  | public policy through research and analysis.         |
>  --------------------------------------------------------
>  | DISCLAIMER: Views or conclusions expressed herein    |
>  | are the author's own, and do not necessarily         |
>  | represent the opinions of RAND or its sponsors.      |
>  --------------------------------------------------------