[tt] Meme 119: Donald McCloskey and Arjo Klamer: One quarter of GDP is persuasion

[Premise Checker]

Meme 119: Donald McCloskey and Arjo Klamer: One quarter of GDP is 
persuasion The American Economic Review 85(2) (1995.5): 191-195 sent 
7.11.7

[Deirdre told me after she gave the inaugural James M. Buchanan Lecture, 
"The Hobbes Problem: From Machiavelli to Buchanan," on 2006.4.7 at George 
Mason that she thinks persuasion now amounts to 30% of American GDP.]

* Department of Economics. University of Iowa, Iowa City, IA 52242, and 
K.CW, Erasmus University, 3000 DR, Rotterdam, Netherlands, respectively. 
An earlier version of this paper was presented at two meetings of the 
Southern Economic Association, in Washington in 1992 and in Orlando in 
1994; parts have appeared in earlier versions in McCloskey's Knowledge and 
Persuasion in Economics (Cambridge University Press, 1994). We thank 
Gerald Bosch, VaNee Van Vleck, Wade Shilts, Allan Wurtz, Santhi Hejeehu, 
David Levenstam, Stephen Ziliak, Louis Soloff, and Mark Stemeu for recent 
comments.

Economists view talk as cheap and culture as insignificant. Yet humans are 
talking animals, talking in their markets. The talk probably matters: why 
else would the human animals bother doing it? The usual economic view of 
the talk is that it issues orders and conveys information. Workers at GM 
are ordered to report for work tomorrow; credit ratings are conveyed. 
Economic analysis takes these parts of the talk into account without fuss. 
Production theory can be viewed as the theory of one mind issuing orders. 
Much of game theory is concerned in one way or another with information 
(though game theory, as Joseph Farrell (1995) and others have found, 
requires more than bits of information). But issuing the orders and 
conveying the information does not account for all of the talk. The third 
part of the economic talk is persuasion.

I. Knowledge Is Information Plus Judgment, from Persuasion

Many distinguished economists have focused on knowing and ignorance: John 
Maynard Keynes, Frank Knight, Friedrich Hayek, G. L. S. Shackle, Ronald 
Coase, Leonid Hurwicz, Kenneth Arrow. The knowledge problem can of course 
also be stated as the uncertainty problem: "It's hard to predict," said 
Yogi Berra, "especially about the future." The colleagues of many 
distinguished economists have reacted with puzzlement, and with a 
simplification. "Aha," they have replied, as in George Stigler's (1961) 
classic paper on the economics of information, "I see: you mean that 
economic actors do not have perfect information. All right, introduce 
information as a mined or manufactured commodity, and we will then be back 
to economics as usual, uncertainty reduced to risk." The usual model takes 
knowledge to be analogous to a calculation on a computer, an analogy 
conspicuous also in linguistics and psychology. Pure information is the 
input. Information is the part merely conveyed, like a telephone number, 
acquired by a mechanical procedure of search.

But the "conveying" is socially and economically determined. The computer 
uses a program and depends on a human purpose served. Without social 
programming you do not know what to pay attention to in reading the phone 
book (for example, in what order to take the numbers, or what the numbers 
mean, or to what uses the number 911 can be put, or what Aunt Hattie's 
number means to you personally). Information is not simply a natural 
property. Humans must judge the information relevant or accurate or 
interesting for it to be "information," selected from the blooming, 
buzzing confusion of the world. Information, to put it another way, is 
only part of knowledge. A wise philosopher said, "what we may be said to 
know will be found to be conjunctions of what is called 'information' and 
what I shall call "judgment'" (Michael Oakeshott, 1989 p. 51). Knowledge 
is information plus judgment. An economics of information alone is going 
to miss the judgment part and is not going to be a complete economics of 
knowledge.

One can see this even in the realm of sheer information. In looking up a 
telephone number one transforms oneself for the time being into a special, 
gullible, judgment-free audience. One ordinarily reads with a certain 
skepticism. But it would be foolish to be skeptical about the judgment of 
veracity conveyed with the USWest phone directory (though not foolish to 
doubt a similar collection of phone numbers in, say, Moscow). For the time 
being, one judges it wise to become machine-like, a mere receiver, a 
facilitating device, a black box. The purpose in such cases is to do a 
machine's job well, such as the job of getting from wanting to call Aunt 
Hattie to actually speaking to her sweet old self. The machine models of 
information, and the ordinary economics of information, apply to the parts 
of knowledge where one has on good grounds decided not to be human.

The other, human, part is the judgment part, the matter of persuasion. 
Persuasion and judgment are related as persuading speaker and judging 
audience. Note that the human frequently judges--gets persuaded-- that it 
is wise to become machinelike and non-judgmental, which is to say that 
judgment and persuasion are involved even in mere information-gathering. 
Only in an extended sense of the word is a machine "persuaded" to take a 
certain bit of information or command at its face value. Information and 
command are machine languages, which the human reads by making himself, 
for convenience, into a temporary machine.1

1 The argument is an adaption of what Harry Collins (1990) said recently 
in his book about computers and expert systems.

Persuasion will ordinarily be involved, therefore, in completing one's 
knowledge. Collecting information is nice but ordinarily insufficient. One 
of us as a little boy decided to collect social statistics in the streets 
of Amsterdam, and for this purpose he recorded in neat columns of a 
notebook one Saturday all the car license numbers he could see. Examining 
his hard-earned evidence at home in the evening it dawned on him, as 
Immanuel Kant said, that facts without concepts are blind. The economy, 
then, is a field of persuasion and judgment, speakers and audiences, in 
the pitch to the loan officer at the bank, in the comforting talk of the 
secretary shepherding a document through the bureaucracy, in the 
supercharged and openly "rhetorical" presentation by the CEO to the crisis 
meeting in the factory's cafeteria.

II. Persuasion Bulks Large

Two economic historians, John Wallis and Douglass North, have argued that 
transactions costs, that is, expenditures to negotiate and enforce 
contracts, rose from a quarter of national income in 1870 to over half of 
national income in 1970 (Wallis and North, 1986 Table 3.13). Their 
measurement is suggestive of the importance of talk, but for our purposes 
it is too broad. Transactions costs include, for example, silent 
protection of property, such as prison walls and door locks. And in the 
talk it includes information and commands, the mechanical parts of 
knowledge. For present purposes we want to narrow the calculation to the 
persuasive and judgmental part of transactions costs, that is, to sweet 
talk.

One way to measure sweet talk is as follows. Take the categories of 
employment of the 119.3 million person-years worked in the United States 
in 1993 and make an educated guess as to the percentage of the time or 
marginal product in each occupation spent on persuasion-- not information 
providing or command giving, but sweet persuasion. The 100-percent 
occupations are lawyers and judges (0.815 million in 1993, according to 
the 1994 Statistical Abstract of the United States [U.S. Bureau of the 
Census, 1994, series 637, pp. 407-9]), public relations specialists (0.155 
million), actors and directors (0.096 million), and social, recreational, 
and religious workers (1.10 million). (Observe the modest size of the 
100-percent occupations, which attract so much disdain for attempting to 
persuade us.) The bulk of the employment involving persuasion occurs in 
the 75-percent occupations: counselors (who also shift paper as 
information, 0.224 million), editors and reporters (0.266 million), 
supervisors of four sorts (clerical, 0.778 million; construction trades, 
0.735 million; protective services, 0.185 milliom; and precision 
production, 0.220 million), and the big battalions of teachers including 
professors (5.17 million). salespeople except cashiers (11.7 million), and 
executives, administrators, and managers (15.4 million). Occupations whose 
marginal product might be viewed as 50 percent earned by persuasion 
include police and detectives (0.923 million, who persuade with more than 
commands backed by violence), adjusters and investigators (1.37 million, 
who seek more than information mechanically reported), teachers' aides 
(0.508 million), authors and technical writers (0.202 million), social 
scientists and urban planners (0.399 million, among which 0.116 million 
economists), and workers in health assessment and treating (such as 
nurses, who exercise persuasion minute-by-minute, 2.60 million), liven the 
two 25-percent occupations, natural scientists (0.531 million) and legal 
assistants (0.254 million), wish to persuade. Adam Smith, in the notes 
that became Lectures on Jurisprudence (1978 pp. vi, 56), puts it this way: 
'"Every one is practicing oratory on others thro the whole of his life." 
Perhaps not the whole of life, but a good part of it. In 1993 in the 
United States, then, weighting the employment figures as suggested, it 
comes to about 26 percent of the person-years employed (and it has been 
rising: in 1991 the same categories were just over 25 percent of 
employment, in 1988 just under 25 percent, and in 1983 23 percent).2

2 The calculation could be improved. For instance, in accord with 
marginal- productivity theory the persuasion workers could be weighted by 
salaries (which would raise their weight in the economy, since many earn 
above average salaries). The educated guesses about the share of 
persuasion could be improved, by studying foi example the Occupational 
Outlook Handbook (U.S. Department of Labor, 1994), which gives detailed 
descriptions of what is entailed in 91 percent of the nation's 
occupations, from Able Seamen to Zoologists. And these could be relined by 
close study of behavior on the job. Operationally, "persuasion" is 
non-machine-like communication, according to a Turing test: would a 
machine do just as well?

The same point can be made from the product side of the national accounts. 
The more obviously "talkie" parts of production are a large part of 
production for final consumption, and much of it is persuasion rather than 
information or command. Out of an American domestic product of $5,720 
billion in 1991 (U.S. Bureau of the Census, 1994 [series 685] pp. 447, 
783, and 1276) the sum of wholesale trade ($375 billion), retail trade 
($532 billion), paper and allied products ($45 billion, producing 
memoranda for the circular file), printing and publishing ($73 billion), 
legal services ($81.9 billion), educational services ($42.5 billion), 
social services ($64.8 billion), general government (state. Federal, and 
local, $629 billion), finance, insurance, and real estate ($1,040 
billion), hotels ($52 billion), and air transport ($42 billion, filled 
with salespeople on the way to persuade) amounted to fully $3,300 billion, 
or about 58 percent of domestic product. It would not be hard to see 
within this a figure of about a quarter devoted to persuasion.

It will be larger in the future. The silent labor required to make a 
radio, a window pane, or an automobile is disappearing. True, the 
technology associated with persuasion has improved since classical Greece, 
as in printing, telegraphs, railway signals, telephones, color 
advertising, xerox, e-mail, and cheap transport of persuaders. Such 
developments in modern times have made central planning seem plausible.

The machine-like part can get better and better, yet leave the human part 
still requiring persuasion. Persuasion is a rank-order tournament, similar 
to queuing (though in most cases not a social waste like queuing). The 
technology of queuing has no effect on the effective length of queues. 
Giving people numbers in bakeries merely cuts down on fights among 
customers; it does nothing to the marginal amount of inconvenience that 
must be suffered by the customer who wants a loaf of underpriced bread. 
Persuasion is similar. Improvements in the technology of persuasion merely 
arm both sides better. A more educated audience requires more complex 
commercials. Savvier inventors mean that bankers and patent officials have 
to be savvier, to distinguish the best from the worst inventions. (There 
are of course net gains, some of them large. In democracies we are 
accustomed to arguing that the net gains from "useless" competition for, 
say, political office are large.)

Is the persuasive talk then empty, mere comforting chatter with no further 
economic significance? If that were all it was, then the economy would be 
engaging in an expensive activity to no purpose. A quarter of national 
income is a lot to pay for economically functionless warm and fuzzies. The 
fact would not square with economics. By shutting up we could pick up a 
$20 bill (or more exactly a $1,500,000,000,000 bill).

III. Examples: Trust, Entrepreneurship, and the Stock Market

Trust is a familiar part of an economics of talk (see Diego Gambetta, 
1988). The persuasion that establishes trust is of course necessary for 
doing much business, which is why the Rotary Club is not very selective. 
The Old Believers in Russia during the 18th and 19th centuries refused to 
adopt the late-17th-century reforms in the Russian church and were in 
other ways far from progressive. Yet because of their peculiarity they 
were able to establish a speech community within the larger society. For 
example, in the early 18th century, Old Believers on the northern River 
Vyg were able to become major grain merchants to the new St. Petersburg 
"by utilizing their connections with the other Old Believers' communities 
in the southern parts of the country" (Alexander Gerschenkron, 1970 p. 
19). Sir William Petty observed at the time that "trade is not fixed to 
any species of religion as such, but rather to the heterodox part of the 
whole" (quoted in Gerschenkron [1970 p. 45]). Any distinction will 
suffice. Thus in the 18th-century the Quakers were great merchants in 
England, as were the Mennonites in Holland. In recent times the overseas 
Chinese, segregated from the rest of the population and therefore able to 
talk inexpensively with each other about breaches of contract among their 
own, are more successful in trade than their cousins at home. Yet in the 
silent world of economic theory since Bentham the economy is supposed to 
work without such talk. It does not.

The entrepreneur, as argued recently in Metin Cosgel and Klamer (1990), is 
above all a persuader, in the classical word a "rhetor," exercising the 
characteristic faculty of human nature for pay. The egregious Donald Trump 
is an example. Trump had the power of persuasion to close deals, the art 
of felicitous speech acts. As he puts it, "You have to convince the other 
guy it's in his interest to make the deal" (Trump, 1987 p. 53). Persuasion 
was the main way he transformed the Commodore Hotel into the Grand Hotel: 
"First, I had to keep [the owners of the hotel] believing [such and 
such].... At the same time, I had to convince an experienced hotel 
operator to [do so and so).... I also had to persuade city officials [thus 
and such].... That [persuasion] ... would make it far easier to prove to 
the banks that [so and such]" (Trump, 1987 p. 122).

The chatter in the stock market (that ideal of a marketplace) is another 
example of persuasion in the economy. Portfolio managers talk full-time to 
decide on buying or selling. Stockbrokers talk to clients and to each 
other. Technical elves spend their days researching the thoughts the 
brokers ought to have. Journalists spend their careers reporting the talk 
on Wall Street, elvish or human. Their reports nourish in turn the talk 
among stockbrokers, between stockbrokers and their clients, and among the 
clients themselves. Wall Street buzzes with chatter and is littered with 
paper reporting the chatter. Efficient markets convey through prices all 
the information that a trader can expect to get publicly. No need to talk, 
since any informational advantage is reflected in price changes. Such 
efficiency would provide few rewards to talking if the talkless model were 
the whole story. The best a loquacious trader could hope for would be the 
quick exploitation of minor information advantages, or a turn of luck. The 
conventional story conjures up a silent film of people throwing darts or 
staring at computer screens, and typing (silently) their orders. Turn on 
the computers and retire to Rye. Something is missing, namely, the 
judgment part of knowledge, persuasion, as Robert Shiller (1995) has shown 
in detail.

IV. An Economics that Acknowledged Persuasion

If the economy depends on the faculty of speech, then the economy will 
require verbal interpretation (cf. Cosgel, 1992; Richard Ebeling, 1990). 
Economic institutions will look to some degree like religious ceremonies 
or social gatherings. They will need to be read in terms of human 
intentions and beliefs. An economy that depends on speech is one that can 
be listened to and read, like a text.

The average academic economist is going to view such a conclusion with 
alarm. He has been raised to believe that his tools are epistemologieally 
superior to those of the departments of English or anthropology. But the 
economist who can adjust will have an additional set of scientific tools, 
those of interpretation. One cannot ignore a quarter of national income, 
the human as against the mechanical part of knowledge. The conclusion is 
not that the present tools are worthless and should be discarded. They are 
worth a lot and should be kept, for their present uses and for bringing 
measurement into an interpretive economics. But if the economy needs 
sometimes to be drilled rather than hammered, or planed rather than sawn, 
the economist had better have a drill and a plane.

REFERENCES

Collins, Harry. Artificial experts: Social knowledge and intelligent 
machines. Cambridge, MA: MIT Press. 1990.

Cosgel, Metin. "Rhetoric in the Economy: Consumption and Audience." 
Journal of Socio-Economics, Winter 1992, 2/(4), pp. 363-77.

Cosgel, Metin and Klamer, Arjo. "Entrepreneurship as Discourse." Mimeo, 
University of Connecticut, 1990.

Ebeling, Richard M. "What Is Price?" in Don C. Lavoie, ed Economics and 
hermeneutics. London: Routledge, 1990, pp. 177-94.

Farrell, Joseph. "Talk Is Cheap." American Economic Review, May 1995 
(Papers and Proceedings), 85(2), pp. 186-90.

Gambetta, Diego, ed. Trust: Making and breaking cooperative relations. 
Oxford: Blackwell, 1988.

Gerschenkron, Alexander. Europe in the Russian mirror. Cambridge: 
Cambridge University Press, 1970.

Oakeshott, Michael. "Learning and Teaching," in Timothy Fuller, ed., The 
voice of liberal learning: Michael Oakeshott on education. New Haven, CT: 
Yale University Press, 1989, pp. 43-62.

Shiller, Robert. "Conversation, Information, and Herd Behavior." American 
Economic Review, May 1995 (Papers and Proceedings), 85(2), pp. 181-85.

Smith, Adam. Lectures on jurisprudence. Oxford: Oxford University Press, 
1978.

Stigler, George. "The Economics of Information." Journal of Political 
Economy, June 1961,69(3), pp. 213-25.

Trump, Donald, with Tony Schwartz. Trump: The art of the deal. New York: 
Warner, 1987.

U.S. Bureau of the Census. Statistical abstract of the United States. 
Washington, DC: U.S. Government Printing Office, 1994.

U.S. Department of Labor, Bureau of Labor Statistics. Occupational outlook 
handbook. Washington, DC: U.S. Government Printing Office, 1994.

Wallis, John Joseph and North, Douglass. "Measuring the Transaction Sector 
in the American Economy, 1870-1970," in S. L. Engerman and R. E. Gallman, 
eds., Long-term factors in American economic growth. Chicago: University 
of Chicago Press, 1986, pp. 95-161.

[I am sending forth these memes, not because I agree wholeheartedly with 
all of them, but to impregnate females of both sexes. Ponder them and
spread them.]