I. DOWN THE RABBIT HOLE
“In another moment down went Alice after it, never once considering how in the world she was to get out again”
(Carroll 1974, p. 14).I fell into doing oral history much in the same way that Alice fell down her rabbit hole. Like Alice, I was in some ways totally unprepared for the experience that was to come. To a degree, I would need to live by my wits and learn quickly on the job if I was going to survive my own mad tea parties. The particular white rabbit I was chasing was named George Joseph Stigler. In pursuing him, I was trying to pin down an elusive ghost.Footnote 1
My journey started innocently enough. After initially gaining some insight into the essential Stigler by analyzing his published work, I was intending to use a much-anticipated sabbatical leave to work my way through the archival records held at the University of Chicago library. Interviewing those who knew the dead economist was an afterthought, a ‘why not talk directly to these people’ idea. The inspiration for doing so came from an interesting set of memorial papers published in the Journal of Political Economy (1993) and edited by Gary Becker. I found myself wanting to ask those authors some further questions about their fallen hero. This formed the starting point of what might have appeared to be a rather quixotic quest.
My interest in one of the giants of post-war economics centered on a series of articles in which George Stigler fiercely defended standard price theory against all and any rival explanation.Footnote 2 What struck me was his highly contentious approach. He would not grapple with an accurate depiction of an opposing theory head-on. Instead, Stigler preferred to create a highly flammable straw man, douse it liberally with kerosene, ignite it, and then dance triumphantly around the burning wreckage of the deposed theory. His combative strategy seemed more redolent of an adversarial courtroom battle where the sole objective is trouncing your opponent rather than analyzing the worth or validity of a contrary argument. My preliminary research seemed to indicate that all these key, hotly contested alternatives undercut the standard explanation behind distributive shares in any market economy. This distributional aspect harked back to the dissertation that Stigler had completed under the guidance of Frank Knight.Footnote 3 I hoped my archival work would be able to shed some light on this question. I was less optimistic about what, if anything, was to be gained by conducting interviews. As is often the case, my naivety in this instance complemented my ignorance.
II. MY SECOND VOYAGE
Well then, it seemed to me next, since I’d wearied of studying the things that are, that I must take care not to incur what happens to people who observe and examine the sun during an eclipse; some of them, you know, ruin their eyes, unless they examine its image in water or something of that sort. I had a similar thought: I was afraid I might be completely blinded in my soul, by looking at objects with my eyes and trying to lay hold of them with each of my senses. So I thought I should take refuge in theories, and study in them the truth of the things that are
(Plato Reference Plato and Gallop1975, p. 51).When setting out to interview key individuals who had known, or worked with or against George Stigler, my innocence knew no bounds. I was ignorant to the extent that I was largely unaware of how unprepared I really was. But while my preparation was lacking,Footnote 4 my instincts managed to come to my rescue. In retrospect, I know today that much has been written about how to do oral history.Footnote 5 But back in 1997, one didn’t automatically hit the Internet to ferret out otherwise hidden knowledge.Footnote 6 And as I indicated, I didn’t realize that any such preparation was necessary. The interviewing process seemed self-evident. Perhaps in the long run, this foolish approach was strategically beneficial. I was less self-conscious about dutifully following the right procedure and more willing to go with my instincts, which would, in any case, become more finely honed with practice.Footnote 7 Many years later, when I had the leisure to check over the prescribed dos and don’ts, it turned out that those mysterious instincts were broadly on target. And even now I doubt that any amount of preparation would have made me into the compulsively methodical individual such guides seem to prefer. Besides, I didn’t want so much to interview as to talk to the people I had targeted. These are two distinct activities. In a formal interview environment, individuals prepare themselves beforehand in order to provide the most informative answers possible. Interviewers often compose a precise list of questions, which they proceed to wade through dutifully, one item after the next. In what I like to think of as the extreme, or Japanese, approach, the questions are forwarded to the interview subject prior to the actual event. This I would label as being a rather low-risk, conservative approach. It lessens the chance for spontaneous episodes and at best provides deep but narrow responses. A prepared subject is more likely to simply repeat what he or she has said before, to switch on the automatic pilot. A skillful interviewer can use follow-up questions to tease out additional information of interest, but the subject will be able to fend off such attempts and be careful not to give away anything too controversial unless that is his or her deliberate intention. Removing preparation time also means risking memories that are distinctly faulty. Individuals have no recourse to fact checking prior to the conversation. Hazarding that risk is more likely to elicit offbeat or distinctly unfiltered impressions of quite relevant material. Pinpoint accuracy may not be too great a cost to pay if your main aim is focused on subjective evaluations.
To some extent the usual provisos attached to interviews did not really apply in my specific case. My search wasn’t so much for additional biographical material about George Stigler but was instead more sensitive to opinions and evaluations of Stigler’s work and work habits. These are quite comfortably categorized as impressions revealing the man himself. Certainly, a wealth of details and information accompanied these interviews with all the attendant problems that define selective and slanted memories.Footnote 8
Finally, at the level of personal images the analysis might again be useful to historians primarily because of what it reveals to be normal and predictable. If its claims about cognitive structuring are correct, one would fully expect economists to make contradictory self-presentations at different times and in different places, perhaps even within single utterances. This does not make their personal testimonies worthless as sources of information, but it does mean that those testimonies should seldom be expected to form single coherent narratives
(Reay Reference Reay2007, pp. 150–151).Information, however, does not form the core of these interviews, but rather comes as an additional bonus. This is clear in the very best of these conversations. Paul Samuelson provided a wealth of details, but, after reading his answers carefully, his view of the whole Chicago project and those who created it generates the more interesting insights. In particular, what Professor Samuelson seems to convey is a long-standing sense of frustration with people like Stigler, Friedman, and Wallis, and the tactics they employed.Footnote 9 Indeed, what people think they remember may at times be as equally revealing as what actually occurred. I scrupulously sought for something that approximated a balance of impressions by talking to some nineteen people, all of whom had very different relationships and reactions to George Stigler.Footnote 10
In contrast to the more formal interview, there is something much more serendipitous about conversations. What they lack in structure, they make up in surprise and unexpected insight. Even more than is the case with formal interviews, the empathy between the parties becomes essential.Footnote 11 If people are comfortable with one another and become interested in the conversation, they are more likely to be expansive and less likely to be defensive. Engaged and relaxed individuals involved in a serious discussion will be apt to allow candid opinions to slip out, as was made evident in my conversations with Paul Samuelson and a number of other individuals.
People in our profession have always been kind of scared of Milton Friedman as a polemicist. So, he gets away with a certain amount of murder. And when he’s safely dead and when they’ve salted his grave against any revival, the daggers will come out. I’m flogging the point. That doesn’t change his overall status…. One of the notable characteristics of George Stigler was his humor. There was an element of cruelty in that humor, and maybe that grew over the years. There are people who are still alive who carry the scars, and will to their grave (Conversation with Paul Samuelson, October 1997).
Part of establishing the ease with which one can elicit confidential remarks depends on an ability to securely park one’s ego at the door. My opinions and thoughts were largely immaterial. The conversation was never intended to be about me. Certainly, one of the keys to success was developing a sense of when to shut up. The views of those with whom I spoke were paramount. This statement implies that I was not there to argue a point and especially not to convert them to one of my preordained views. Antagonism effectively destroys empathy when the objective is to create a comfortable atmosphere that allows the individual to open up.Footnote 12 In fact, the hope was that those interviewed would come away without any definite idea or concern about my own rather evolving stance. I would put forward possible ideas and interpretations simply to get an interesting response. Whether they agreed or disagreed was largely immaterial. I was as happy to have any tentative ideas shot down as supported. (Though, perhaps, I was secretly pleased to have Gary Becker agree with a possible hypothesis: “But his views did become more consistent. I agree with you on that. Other people may not think so, but I think definitely that was true. He began to re-think some positions he had just inherited” [Conversation with Gary Becker, October 1997]).
In my case, I was even more fortunate since the conversations I held were not about the person being interviewed. Instead, they were about a third person, George Stigler. Most individuals will be more eager to protect themselves and their own reputations than that of a third person. (Though, clearly, what someone says about a third person is often quite revealing about him or herself.) I was particularly fortunate in talking about someone who, in his lifetime, had generated strong feelings, whether positive or negative. What was often expressed was an acute feeling of loss, as if someone larger than life had left the scene.
He was a great person to be around. It was fun to be around him because he made professional work fun. And he was a great gossip about the profession. You could always find out what was going on and what George’s opinions were about what was going on. Every dinner you had with him was entertaining. He was a man of great wit and great style and I miss him very much (Conversation with Harold Demsetz, October 1997).
Without thinking, then, without any attempt to weigh up alternative approaches, I chose the high-risk option when I went off with tape recorder in hand. I didn’t at the time expect to use these conversations for anything other than raw material. I was, in fact, surprised when so many of them turned out to be interesting in and of themselves. They all rambled, as could be expected, given the approach I employed. And those where I didn’t manage to establish any real empathy with the person were those that were least satisfactory. They were mostly lifeless, filled with what sounded like recycled material. But in many cases, in the course of a conversation, we managed to hit on ideas and topics that provided ground for fresh exploration. Of course, when I claim that I went into an interview situation not knowing what I was doing, I don’t mean to imply that I wasn’t thoroughly prepared. In each case I was fully knowledgeable of Stigler’s work and the work of the person being interviewed. I had also prepared a series of topics I wanted to discuss in the course of the conversation. If you end up reading all the interviews, it becomes clear that there were common themes brought up in all the conversations. My difficult job was to listen to each person’s answer while thinking about where I wanted to go next. Being a conversation, it was largely unstructured in that if a response was provocative or otherwise interesting, I was prepared to follow that up rather than march linearly from question to question. To emphasize the point once again, I viewed my objective as getting the individual to open up and reveal his or her thoughts. I served simply as a medium to facilitate an interesting flow of conversation. In such circumstances, my thoughts or opinions were insignificant. I wanted to understand what these people thought rather than foolishly trying to convey my own ideas. What I had to do to accomplish this would vary from individual to individual. For example, the very easiest conversation was the one conducted with Paul Samuelson. As a routine, I would often begin a session with a softball question to put the subject at ease and try to create an atmosphere conducive to reminiscing. Usually I asked how they came to know George Stigler.Footnote 13 Professor Samuelson spent some forty-five minutes answering that question, while I did a very good imitation of the slave boy in Plato’s Meno. At that point Paul Samuelson announced he had ‘shot his bolt’ but would be willing to take any other questions. We then spent another forty-five minutes covering additional territory. The point is that I was as active or passive as I deemed necessary to get the individual to open up. This was certainly a judgment call, which I did not always get right, but in general I think I managed to be more effective than I would have thought possible. However, what remained a core requirement was convincing these people, who were kindly giving up their time, that it was worth the sacrifice.Footnote 14
In contrast, what I think to be an equally invaluable conversation with Gary Becker demanded much more on my part. (In fact, following the eighty minutes spent with Professor Becker, I felt the need of a good lie down.) In looking at the interview, one aspect that stands out is the additional effort expended in suggesting possible interpretations, asking more questions, and following up on answers. In comparison, I had only to give Paul Samuelson his head. What doesn’t come through by simply reading the interviews are the pauses. While there are few with Samuelson, with Gary Becker there is often a delay while he carefully considered the question and all his possible responses. However, what was common in all these sessions was the intensity with which I was required to listen to what they were saying. I couldn’t just nod my head and proceed to the next stage of my mapped route.
Except for the first and last of these conversations, I managed to squeeze in the following interviews in October of 1997. They’re listed in the order in which they were conducted:
1. Milton Friedman, Rose Friedman, Aaron Director
2. Harold Demsetz
3. Arnold Harberger
4. Armen Alchian
5. Gary Becker
6. Lester TelserFootnote 15
7. Sam Peltzman
8. Clair Friedland
9. Ronald Coase
10. Stephen Stigler
11. Sherwin Rosen
12. Paul Samuelson
13. Robert Solow
14. John Kenneth Galbraith
15. James Kindahl
16. Paul Sweezy
17. Mark Blaug
The core of these individuals came, as I have mentioned, from the memorial JPE volume. Others were added as circumstances dictated. For instance, it was Milton Friedman who suggested that Rose Friedman and Aaron Director also be present.
AD: No, I didn’t get to know him then.
Milton Friedman: I came back to Chicago for a year in ’34 to ’35 and that’s when I first got to know George. Rose got to know George …
Rose Friedman: A little bit in his first year.
Milton Friedman: But then we really got to know him when I came back (Conversation with Milton Friedman, Rose Friedman, and Aaron Director August 1997).
At UCLA primarily to interview Harold Demsetz, I decided to include Armen Alchian as well, since he was part of the Chicago West contingent. It was Harold Demsetz, clearly pleased with our conversation, who subsequently decided to call Arnold Harberger on the phone and set up a quick interview.
I did not have a lot of day-to-day contact with him. His workshop used to meet over at the Law School for example, the other side of the Midway. He was a professor in the Business School as well as in the Department. But in any case I really feel that he is one of the most interesting and in some sense, most enigmatic characters in our profession in this last century (Conversation with Arnold Harberger, October 1997).
Besides the original core of people in Chicago, I sought out Stigler’s son Stephen. His permission had to be gained to view the Stigler archives and it seemed worthwhile to have his personal impressions while I was there.Footnote 16 And I couldn’t pass up the opportunity to question Ronald Coase about Stigler’s Coase Theorem since I was already there on campus.
I think he was one of the most difficult people to explain because there is no one like him. I’ve described how in an argument he jumps around.[Footnote 17] He puts in a bit of theory, a bit of statistics, a reference to the earlier economists. It’s like no one else’s form of argument that you can recall (Conversation with Ronald Coase, October 1997).
Lester Telser, a long-time colleague at Chicago, had himself written a short biographical sketch about George Stigler. But in particular I was puzzled by a curious statement made in Stigler’s autobiography that seemed to differ from Telser’s own published views (Telser Reference Telser1960). To be precise, what Stigler claims as an explanation of resale price maintenance (Stigler Reference Stigler1988, p. 163) seems to be more of a hypothesis, which, in the case study provided by General Electric, lacks evidential validity.Footnote 18 Then again this incongruity should not have been surprising. The two of them endured more than a few differences over the years.
Lester says to me [Claire Friedland], “That George is so stubborn. He is so unreasonable. I told him his research was wrong. I explained it to him. I was right. George was wrong. And he won’t listen to me.” Lester is just letting off steam. He goes out Stage Left; George arrives Stage Right and says, “That Lester is crazy. I don’t know why I pay any attention to him. But let’s try it his way” (Conversation with Claire Friedland, October 1997).
I then headed for Boston to hear what some of his opponents might have to say. Clearly, Paul Samuelson was a prime candidate for an interview as someone who had known him since the 1930s.
I probably saw George the most in my final senior year, which was the academic year 1934–1935. George and Allen Wallis had taken squatter rights possession of a storeroom in the basement of the social science research building where the economics department was.… The reason that I saw them [Stigler, Friedman, and Wallis], and a great deal of them, was that to make some money during the Great Depression I had what was called an NYAS (National Youth Administration) scholarship supplement.… I don’t know how much an hour I received, maybe twenty-five cents an hour, but they had to find some perfunctory work for me to do. And just as in Pinafore, I polished up the brass on the door. I was given the job of dusting off in that Department Records Room the pictures of the great economists, Bohm-Bawerk, John Stuart Mill, David Ricardo, Adam Smith, and maybe Knut Wicksell (Conversation with Paul Samuelson, October 1997).
Moreover, Professor Samuelson had been on the receiving end of Stigler’s wit as well as being cavalierly dismissed by Stigler as unlikely to achieve any eminence as an economist: “Paul Samuelson, preceding George on a panel, ended his remarks with ‘I know what George Stigler’s going to say and he’s all wrong.’ George stood up, said ‘2 + 2 = 4.’ And sat down” (Friedland Reference Friedland1993, p. 782).
It may merely be prejudice, but I’m inclined to write him [Paul Samuelson] off as an economist. Two of his recent jobs … were pure mathematical exposition, as is also his current Economica item (which, by the way, has already been done better by Wold), and his textbook suggests that he doesn’t know anything that hasn’t appeared in the Survey of Current Business (Hammond and Hammond Reference Hammond and Hammond2006, p. 97).
Solow became a last-minute addition when I discovered the close friendship that had developed between the two.Footnote 19
There was a sixty-fifth birthday party for George that the University of Chicago put on, and my wife and I flew out to be there. I remember that we were all put up in a University building, the Center for Continuing Education which had bedrooms. We came in one evening and the next morning we walked into the dining room where all the other guests at this party were having breakfast and a hush fell over the whole dining room. Finally a friend of mine, I think it was Si Rottenberg from the University of Massachusetts, came over and said, “We were all wondering, ‘what are you doing here?’” I had to explain that I was there because I adored George. And so we remained close friends until he died (Conversation with Robert Solow, October 1997).
I had already written a paper (Freedman Reference Freedman1998b) on the adversarial relationship between Galbraith and Stigler, and since I was in Cambridge already, thought to follow it up. That almost automatic sparring continued for more than four decades, with everything Galbraith wrote sure to receive a stinging review from George Stigler.
He saw me as somebody who was a menace to that pure supply and demand market system. At that AEA Conference in Philadelphia, there was a big crowd which attended the meeting where I was presenting a paper. I was on my way to my meeting there, and I overheard somebody say, “Well, it is time to go to the meeting to hear George Stigler kill off Galbraith.” And George made a systemic, orthodox and very plausible attack on the idea that one position of strength in the market tended to beget another, which was my case (Conversation with John Kenneth Galbraith, October 1997).Footnote 20
In contrast to Galbraith, Jim Kindahl had been his co-author of The Behavior of Industrial Prices, a landmark work published in 1970. He had also been one of my teachers in graduate school, one whom I greatly admired. Moreover, Amherst was only a short hop away from Cambridge.
I was teaching at Amherst College at the time. One day the phone rang and a voice said, ‘This is George Stigler.’ Now, only George Stigler could say, ‘This is George Stigler,’ end of sentence. He would expect that I would know who he was. And I did of course. Any Chicago PhD student knew who he was.… He said he would like me to come, to take a year off, and work with him. This is out of the blue. This is a man I’d never met in my life. He said Milton Friedman had recommended me and Milton’s recommendation was good enough for him. He didn’t need to know anything more (Conversation with James Kindahl, October 1997).Footnote 21
I had consulted Paul Sweezy when writing my original article on George Stigler and the Kinky Demand Curve (Freedman Reference Freedman1995).
It’s very simple as a model, in its form as you know. And it made sense. And I still think it’s pretty good. A pretty good point about the pricing of a monopoly product, but I certainly never thought it would ever be a great achievement or breakthrough or anything like that. It turned out to have a much more illustrious career than I expected it to have.… I didn’t pay much attention to Stigler in those days. I was probably in one of my ultra-left moods, or something like that (Conversation with Paul Sweezy, October 1997).
Lastly, when Mark Blaug had kindly contacted me concerning the same article, he had revealed himself as being Stigler’s first PhD student while studying at Columbia.Footnote 22
I was immediately and mainly attracted to George Stigler’s lecturing style. I don’t think he was a particularly good lecturer. But the lectures were like a lot of his articles. Lots of sarcastic jokes. And very arrogant. Quite cynical. This was much more the kind of lectures that I like. So, I attended the lectures very religiously. As a result, I decided I wanted to do a Ph.D. in this area. Eventually my dissertation became my first book on Ricardian economics, on what happened to the Ricardian system after Ricardo’s death (Conversation with Mark Blaug, November 1998).
So while what passes as a grand strategy was put together in a haphazard manner, I did end up interviewing an interesting core of key individuals. About the only principal I failed to snag in my net was Allen Wallis, who, with Stigler and Milton Friedman, formed the Three Musketeers of Stigler’s graduate years at Chicago.Footnote 23 I should have made a greater effort to contact him, as he died the next year in 1998.
III. TRAPPING THE QUARRY
I hope I’m not only speaking for myself. I do have this impression that I know what George thought.… Well, the only thing that I hope will save us from the trap of thinking that Claire knew what George thought is that you’ve interviewed other people. There’s that mountain of evidence argument again. If the others don’t agree with me, forget everything I said, especially since some of those others might have known George better. A few not only knew George better but also know more about economics than I do. So, if I’ve made some mistake, I know that they will correct me (Conversation with Claire Friedland, October 1997).
The difficulty is always in knowing how to frame a conversation so that it will be fruitful. Given my approach, this was always going to be chancy. Certainly, the opening question was simple. It was usually some variation of how the individual came to know George Stigler. Once, though, the conversation was started, how could it best be carried forward?
In a sense, if it was a real conversation, the discussion progressed naturally, given some guidance from me. However, it was important that I kept focused on my objective of shining a light on how George Stigler actually worked and thought. By definition, the responses would be subjective evaluations. My hope was that by asking the same or similar questions, or bringing up key points with each individual, I would indirectly achieve a sort of second-hand discussion among the participants while I remained a discreet midwife. The best way to convey what I gained by this method is to look at a few key areas.
A. A Priori Beliefs, Empirical Evidence, and Marketing
It was just that he was so enthusiastic about quantitative measures. He thought that he was going to change the world.… I was sitting with Aaron Director at the time when he gave his Presidential Address[Footnote 24] and we did look at one another at the time to try to see what each one thought about all of this (Conversation with Ronald Coase, October 1997).
The natural question to ask of someone who was almost messianic about using empirical evidence to test hypotheses was whether his own views and beliefs ever changed due to such testing. It is unfortunately possible to reduce empirical evidence to a mere rhetorical device that allows an economist to better market his or her prejudged opinions.Footnote 25 And it is undeniable that Stigler fully understood the importance of marketing and was adept at using it. He might be seen as carrying marketing to heights not previously considered. The Chicago counter-revolution emphasized style as well as content. As evidenced by his pioneering paper with Gary Becker (1977), convincing target audiences that the flow of services provided are closely congruent to their given preferences is vitally important.
Part of it is the persuasion. There’s no question. I mean George Stigler, I remember when I was a young person, wired and said ‘Selling is very important in your research. So write better. Work on writing because that is important. You’ve got to sell what you are doing. I think he’s exactly right. You’ve got to sell what you are doing (Conversation with Gary Becker, October 1997).
Because Stigler early on saw the importance of marketing his views, part of his technique was to state his research and his results in the strongest way possible. Certainly, this partially reflects his personality, as a number of discussants happily pointed out. But it was more than that. As George Stigler himself would have admitted, and as Claire Friedland points out, he liked to stir things up. He very much wanted to have a positive impact on the economics profession.Footnote 26 From that standpoint, the worst thing that can happen to one’s work is for it to be simply ignored. The cardinal rule then in the marketplace for ideas is to get noticed.
But it got people riled up and they jumped in. It’s not the result that we have today. But we have a whole literature as a result. So he was very much this kind of a person who would say, ‘I’ll state this result as strongly as I possibly can, even if it’s not completely justified by the evidence.’ I mean I think at some level he said to himself, ‘I’ll make the strongest case I can and then if that stirs people up.…’ In many ways he was using a bully pulpit that he had acquired from his stature in the profession. He was able to do that (Conversation with Sam Peltzman, October 1997).
What these series of conversations end up reinforcing is the inherent difficulty in coming to grips with exactly what George Stigler was marketing. Was it merely the results of his empirical investigations, the facts as he knew them, or was he trying to push some very deeply held beliefs by using his evidence as a vehicle for promoting a priori conclusions? Such a quandary is not easily resolved. This makes actually trying to discover the source of a person’s beliefs not unlike trying to tap dance on top of quicksand. To succeed, only the greatest of skill combined with the lightest of touch stands any chance whatsoever. When faced with choosing between empirically or ideologically grounded, perhaps it is best to dodge the issue by following traditional economic reasoning and accepting those core beliefs simply as given. “I don’t know where his views came from. That’s a hard question. Who knows about things like that” (Conversation with Sherwin Rosen, October 1997)?
Otherwise we are forced to become amateur psychologists, looking at genetic and environmental markers. It is quite easy to point out, as a number of people did, Stigler’s basic shyness, sensitivity, and even a sense of insecurity that was well camouflaged beneath a gruff exterior and an almost compulsive habit of distancing himself by the use of his corrosive wit.Footnote 27 Alternatively, we could focus attention on the influence of Viner and Knight in forming his bedrock conservatism and adversarial nature.Footnote 28 Undoubtedly, the role of the Cold War era in sharpening a sense of the importance of the work he undertook should not be dismissed. Such influences and determinants are easy sources of speculation, but almost impossible to pin down. The more interesting issue that can be tackled is what might have caused George Stigler to change his mind or, even more radically, whether any evidence would have possibly undermined his firm belief in efficient markets. Views provided during the course of my conversations are far from conclusive, but do provide interesting possibilities.
At the very least, taking empirical work seriously should imply a capability of being surprised by one’s results. If however, one accepts Melvin Reder’s conception that a belief in a Tight Prior EquilibriumFootnote 29 forms the core of the Chicago approach, the legitimacy of empirical research is undermined as evidence becomes the handmaiden to prior beliefs. A rather lengthy quote from Reder makes this abundantly clear:
Any appropriate inconsistency of empirical findings with implications of the theory, or report of behavior not implied by the theory, is interpreted as anomalous and requiring one of the following actions: (i) re-examination of the data to reverse the anomalous finding; (ii) redefinition and/or augmentation of the variable in the model, particularly the permissible objects of choice and the resource constraints; (iii) alteration of the theory to accommodate behavior inconsistent with the postulates of rationality (constrained optimization) by one or more decision makers (resource owners); (iv) placing the finding on the research agenda as a researchable anomaly.
TP [Tight Prior Equilibrium] implies shunning (iii): i.e, the subject matter of the tight prior is the adequacy of this approach to theory as an explanation of whatever behavior is considered as economic. It is not appropriate to characterize this prior as “dogmatic”, and to suggest that it implies placing no credence in the possibility that (iii) is appropriate. It suffices to say that TP adherents—with variation among individuals—focus attention upon (i) and (ii) and, failing a quick resolution of the anomaly move to (iv) but pay little attention to (iii)
(Reder Reference Reder1982, p. 13).The variety of identifiable and subjective views presented in regard to this issue is particularly interesting throughout the course of these extended conversations. All are based on observing the same individual and even the same set of actions. George Stigler undoubtedly was one of the great champions of empirical work. He conscientiously followed his own precepts in trying to make such evidence a key part of his own research. Ironically, the issue in interpreting the evidence of Stigler’s performance is essentially the task of evaluating what is equally available for all to observe and examine. The problem lies in teasing out the meaning. Unfortunately, researchers all too often persist in bringing to a subject their own preordained interpretation.
The interesting thing is he was a great enthusiast for quantitative methods. So, it doesn’t seem altogether consistent. But he certainly was. On the other hand, he knew what the answer was going to be. He just regarded it, as I say, as a way of persuading other people (Conversation with Ronald Coase, October 1997).
Clearly, George Stigler’s views did change. The most notable instance is his analysis of monopoly and anti-trust. The young George Stigler seems to have followed the teachings of his mentors at Chicago, particularly Henry Simons and Frank Knight. His views, as expressed in a basic review of monopoly theory (Stigler 1942), are quite conventional. In more popular articles (Stigler Reference Stigler1952), he unreservedly sees big business as the scourge of market competition and willingly testified to that belief when appearing before the Subcommittee of the House Judiciary Committee in its study of monopoly power in the steel industry (1950).
Certainly, it is my belief—speaking as an economist and not as a lawyer—that the basic trouble with the anti-trust laws … is their failure to recognize that when an industry is made up of, or dominated by, a few big firms, there is inherently a structure inconsistent with efficient competitive operation
(Stigler Reference Stigler1988, pp. 98–99)Stigler (Reference Stigler1988) has a chapter in his autobiography in which he elegantly describes his conversion from conventional thinking to a position where issues of monopoly or oligopoly are of minimal importance. Aaron Director, one of the few individuals who clearly influenced his thinking, must have played a key role here. Director, of course, a genuine iconoclast, had migrated to the Chicago Law School in the post-war period to fill the position vacated by the death of Henry Simons. “He was an iconoclast. But he didn’t develop new data with respect to industrial organisation. He didn’t develop and articulate new theories. He just said that the conventional belief wasn’t so” (Conversation with Paul Samuelson, October 1997). He taught a course along with Edward H. Levi on anti-trust law, which could rightly be claimed as the Petri dish that brought forth the culture of Law and Economics. This newly established approach would then influence thinking in this area for many decades to come. Certainly, by the time George Stigler (1963) came to write a classic paper on oligopoly, there had been a clear break with his earlier thinking.
I think he is maturing as a person. He’s getting away from Simons. He begins to understand that there must be good reasons why Eastman Kodak dominates the film industry. Obviously there must be market forces involved. Why wasn’t capital flowing into an industry with high basic returns? He is asking himself the kinds of questions that just didn’t occur to Simons. A guy like Simons would just say, ‘Well, they’re too big. Break them up! The text books tell you, the more firms, the better’. And that’s it. Simple. End of story. Advertising screws up people’s minds. Tax advertising and it’ll be fine. So, he is beginning to understand that market forces are deeper than simple textbook stories. That’s one of the lessons that I learned from him (Conversation with Sam Peltzman, October 1997).
His long-term research associate, Claire Friedland, makes this view clear. By embracing this approach, George Stigler became evidence driven. Nothing else but accumulated evidence would ultimately make up his mind.
George would never just write an article in which he said, “Yes anti-trust laws were mistaken, and a lot of the policies that they forbid were really efficient.” This is the result of the work that’s been done in Law and Economics over the last fifty years, since Henry Simons and especially due to Aaron Director. George would never say a thing like that without having an empirical test, because anybody can have a theory. George said, in his 1964 Presidential Address, you can find a theory to support any policy and the question is, “What is the evidence?” So I think of George as very, very empirically oriented. If he didn’t like something, like monopolistic competition and game theory, it was because he felt that empirically they didn’t hold up. He also felt that intellectually they didn’t hold up. But I think the empirical side of it was very strong. And not just empirical research, examining evidence, but whether it moved the profession ahead, whether it provided useful concepts to the profession (Conversation with Claire Friedland, October 1997).
Nonetheless, it is difficult to imagine the particular type of empirical evidence that would lead him to doubt market efficiency or to see regulation as advantageous. Again, this need not reflect a conscious choice. When faced with comparative mountains of evidence, choices would have to be made and results interpreted. Discovering exactly what we expected to find is a trap not easily avoided.
He would come across empirical work which was contradictory to other empirical work. Somehow it always seemed to him that the empirical work which favoured his side was done better than the empirical work which didn’t[Footnote 30] (Conversation with James Kindahl, October 1997).
A third alternative presents a possible means of escaping from these two clearly marked extremes. George Stigler had a passion for consistency and an affinity toward comprehensive explanation. Employing this approach implies that empirical evidence clearly would have caused Stigler to change his mind, but only insofar as it accorded with the more consistent and clear-cut view of economics that he had developed painstakingly over his long career. Namely, his view on monopoly and anti-trust did evolve and become more in line with what he accepted as a mountain of undeniable evidence. But his hard-found convictions also sat quite comfortably with his unshakable belief in competitive markets. Results that reinforced that pre-existing predilection would have been naturally more congenial to him.
I think he went to a more satisfactory position, absolutely. The earlier view, as you say, he picked up, that was the literature, he hadn’t really thought it through. I mean, you know, he hadn’t thought through everything at that point, and he hadn’t really thought it out. As he thought through more and more, I think he came to a more satisfactory thesis on the issue.… But his views did become more consistent.… Other people may not think so, but I think definitely that was true. He began to re-think some positions he had just inherited. Inherited you know, from his teachers and so on, or from the literature and he put it in a more consistent framework (Conversation with Gary Becker, October 1997).
B. The Ideological Honeypot
I have just finished reading your essay “An Introduction to Privacy in Economics and Politics.”[Footnote 31] If I had a postcard with a picture of a famous economist like Adam Smith, Karl Marx or George Stigler I would write on it. But since I don’t have a postcard, I will just say that it looks like ideology to me (Letter from Douglas North to George Stigler, 10 August 1979).
No one would question that George Stigler was clearly conservative in outlook. His non-research output makes this amply clear. No one joined the Mont Pelerin SocietyFootnote 32 out of a desire to advance causes favored by the radical left. As Claire Friedland remarked, “George was probably the last man in Chicago still wearing a fedora back in the sixties” (Conversation with Claire Friedland, October 1997). Why he became so receptive to conservative beliefs is an interesting question, but one perhaps better explored by employing techniques more familiar to other disciplines. Stigler saw himself as continuing in the tradition of the classical economists, especially as epitomized by the liberal tradition of John Stuart Mill. Like everyone else, he had a set of beliefs, an ideology, though possibly more defined than most. As pointed out before, his conservative beliefs did not lead him to consciously become an advocate for conservative causes through his work (specifically what we may term his “scientific output”).Footnote 33 In fact, as will be noted in the next section, he came to dismiss such advocacy as being essentially useless. But the real underlying issue is whether his very conservative beliefs, his belief in efficient markets and the efficacy of the price system, ever came to color his empirical and more technical research. Here opinions diverge widely.
What is uncontroversial was his own position on ideology expressed often and in many places. Ideology just did not have an effect on economic theory. It ran orthogonal to his insistence on rationality.Footnote 34
I don’t know how important ideology is, but think it is unimportant. You don’t know how important it is, but think it is important. My position is better because I try—feebly and so often unsuccessfully—to use a trusted theory of human behavior to explain social phenomena. Your position is worse because you try—with marvelous ease—to explain the mysteries by a deus ex machina (Letter from George Stigler to Milton Friedman, 29 March 1984).Footnote 35
Unfortunately, one’s stated intentions need not be consistent with what one actually accomplishes.Footnote 36 This may come down to the issue of interpretation and how one decides whether there is a link between one’s beliefs or ideology and the outcomes that research may produce. It does seem fair to say that Stigler strongly defended neoclassical price theory. He was skilled, as few others are ever able to be, at extending the reach of price theory and dispelling seeming anomalies.
He was interested, I would say primarily, in a particular sort of puzzle and it’s a typical Chicago puzzle. And I don’t mean that in any bad way. It’s the sort of puzzle that the Chicago School’s presuppositions require. Show me an apparent anomaly, something that does not seem to be explicable using the Smithian apparatus and the Marshallian apparatus and I will show you that it can be explained that way. That was exactly the sort of thing that George went looking for. And that’s not a bad thing. I’d have to say that it can actually be very good.… If you have a very clever man whose life is devoted to showing how you can explain apparent contradictions without altering the basic formulation, that’s what you tend to do with empirical work as well (Conversation with Robert Solow, October 1997).
For this reason, it has been strongly suggested that the core work coming out of the Chicago School during the days of Stigler and Friedman was highly predictable. You knew what to expect before you began and the only item of suspense was exactly how the particular researcher would achieve this predetermined conclusion. Again, even were such claims to be true, it need not invalidate the work itself.
It seems to me that when you get to his later work, say with Becker, you know what the conclusion is going to be before you start the argument. In a sense, you’re assembling arguments to support a conclusion. I mean, that may be unkind and untrue but it’s an impression. And, it’s even more so in the work of Richard Posner. Have you read any of that? It seems to me that the plot is always the same, and the characters stay fixed (Conversation with Ronald Coase, October 1997).
Some, such as Paul Samuelson, would take a diametrically opposed stance, not accepting that economic work is generally ideologically free. Furthermore, Samuelson proved willing to support his case with the sort of empirical evidence that George Stigler would automatically demand to see. (Whether Stigler would accept such evidence is, of course, an interesting question.)
Well, this is a popular Friedman view too. And it’s wrong. I say that flatly. But it’s interesting—I have somewhere a National Bureau Yellow Jacket manuscript of a research study by Victor Fuchs from Stanford University, Jim Poterba from this University and Allan Krueger of the Woodrow Wilson School at Princeton.[Footnote 37] They did an extensive sampling of economists in two areas of economics. And what they did was they gave a whole set of questions on what each person’s factual opinion was on that question. And so he had all these factual differences in the group. But they also asked questions about their value judgements. They asked these in whole different areas. The third less important area were their political affiliations, which I presume these days would be Republican or Democratic, but they might have gone further. Then they tried to see how you explained the differences in policy recommendations. And their finding is the opposite of Milton Friedman’s. There was very considerable degree of consensus on factual matters. There were some differences in the degree of confidence they had in their answers. The confidence intervals varied quite considerably. Now what they found was the difference in their policy recommendations were—I’m using your language, not their language—ideologically premised values. They were not fact driven. Now there are a few cases like the minimum wage, or Ricardian comparative advantage, where you can almost get certain unanimity, free of ideology. But these are exceptions in my opinion (Conversation with Paul Samuelson, October 1997).
Again you might argue that George Stigler, in his most important work, deliberately avoided any hint of preaching or any concerted attempt to push forward certain policy positions. Although this may not be overt, still, even in the core of his more empirically founded work, the possibility that results lined up all too easily with ideological presuppositions cannot be entirely ruled out. Again, this would not be unique to George Stigler. However, it would be particularly worth noting in such a careful and conscientious economist (and one of the best of a remarkable post-war cohort).
Oh, he was a true believer. He wouldn’t like that term. But put that in because he thought in that sense. He was absolutely convinced that he was right. It wasn’t a doctrinal battle. It was a battle of facts. I almost said good and evil. He was sure.… He was absolutely sure the economy was on his side and if research was properly done it would show this. He really believed that he understood how the world works. And the way the world works had been shown to him by the theory of price.… His [Gardiner Means’] facts were wrong. Wrong. Wrong. Wrong. And he knew it. How could they be anything but wrong if he thought the economy was not competitive? … My guess is that he came to his conclusion from an a priori position. But then my guess is that most economists make their judgements on a priori beliefs and not on empirical evidence (Conversation with James Kindahl, October 1997).
In an almost unconscious reprise of the conundrum facing George Stigler himself, each of those interviewed faced similar evidence concerning how this consummate economist conducted himself. But, as noted before, evidence does not speak for itself. What I was able to gather is some very sharp differences in subjective evaluations, which raise interesting questions without providing any definitive answers.
I gradually realised, I don’t know when I realised, that he was one of the most fascinating examples of how economists act even though they say that economics ought to be value-free (and it is sometimes value free). I find it extremely difficult to resist reading a lot of economic theory, politically. Now, you can see the way George Stigler had a wonderful nose for attacks on neo-classical economics that were dangerously capable of undermining the very foundations of orthodox economics. It’s no accident that he went for Gardiner Means’ Administrative Prices, Harvey Leibenstein’s X-Efficiency, Galbraith’s Affluent Society and his other thing, the Kinked Oligopoly Curve of Sweezy. These are all ideas that are very, very dangerous, subversive even of orthodoxy. It’s those ideas which roused his critical fire. Obviously he thought his views about the markets and so on were identical to Milton Friedman’s but they were much more carefully disguised. He never, (unlike Milton, came out with them openly, or wrote about them, as Milton did, in Capitalism and Freedom, where Milton sort of let go. You know, you could read right through all of George Stigler’s stuff, even on industrial organisation, and his implicit endorsement of markets, and the condemnation of all government intervention, which of course eventually inspired his work on The Theory of Regulation. This proved to be an example where ideology is productive. The trouble with ideology is that it can blind you, but it can also sometimes create a spurt. Who would have thought up The Theory of Regulation, unless you were already inclined to regard all government action with deep suspicion, always inclined to believe it does more harm than good (Conversation with Mark Blaug, October 1997).
C. The Paradox of Legitimacy
But what exactly he would allow as a proper realm for government? He got into this—I remembered his name for it after talking to you—‘paradox of legitimacy’ he called it, or sometimes he called it the ‘problem of legitimacy.’ At the time of his death, this was the problem he was working on. It was very much of concern to him.… But he was very much concerned about how you could call something inefficient if the political arena, where we do have a democracy, more or less, where we do have representative government, has allowed it to happen (Conversation with Claire Friedland, October 1997).
I was particularly struck by an article published posthumously by George Stigler (Reference Stigler1992), the last work he completed before he died. What made it interesting was his seeming defence of the indefensible. The article in essence starts off with a defence of US sugar subsidies. It seems to be a version of a market dictum attributed to Armen Alchian; namely that what is, is efficient. (Underlying this supposition is once again a sense of the test of time. Any inefficiency would create an incentive for individuals to change the existing status quo.)
In this latter view, every durable social institution or practice is efficient, or it would not persist over time. New and experimental institutions or practices will rise to challenge the existing systems. Often the new challenges will prove to be inefficient or even counterproductive, but occasionally they will succeed in replacing the older system. Tested institutions and practices found wanting will not survive in a world of rational people. To believe the opposite is to assume that the goals are not desirable: who would defend a costly practice that produces nothing? …
Consider the following example. The United States wastes (in ordinary language) perhaps $3 billion per year producing sugar and sugar substitutes at a price two to three times the cost of importing the sugar. Yet that is the tested way in which the domestic sugar-beet, cane, and high-fructose corn producers can increase their incomes by perhaps a quarter of the $3 billion—the other three quarters being deadweight loss. The deadweight loss is the margin by which the domestic costs of sugar production exceed import prices. Lacking a cheaper way of achieving this domestic subsidy, our sugar program is efficient. This program is more than fifty years old—it has met the test of time
(Stigler Reference Stigler1992, p. 459).According to Claire Friedland (Conversation, October 1997), this was the problem occupying George Stigler’s mind in the last years of his life. It is not, in fact, difficult to see why he was driven to it or why it generated controversy even among his closest colleagues. Essentially, this assigned a very minimal role to economists in policy matters, if any at all.
‘Consistency is the hobgoblin of small minds,’ somebody once said. He was not consistent. That’s clear. And this inconsistency has led to a lot of misunderstanding. There are people outside of Chicago who read him this way. With Becker it is even more powerful. It’s all part of Becker’s stuff about optimality and redistribution. Outsiders kind of read both of them as ‘This is kind of the senescence of the Chicago school. They have become toadies for big government, apologists for big government.’ And I could see why. It is a really subtle kind of distinction we are making here between the two. But look, if you’re going to regulate, conditional on wanting to redistribute income, I can’t tell you that this is wrong. So, if I don’t like it, if I tell you it’s wrong, it has to be because I don’t like the resulting redistribution (Conversation with Sam Peltzman, October 1997).
This controversial approach is really a matter of taking the ideas of consumer sovereignty and rationality seriously, which George Stigler undoubtedly did. If we remember Stigler’s drive for greater consistency and comprehensiveness, his grappling with this final issue becomes almost inevitable.
Something he was thinking about was how to reconcile consumer sovereignty, or voter sovereignty, with his previous notions of inefficient government. Can we say this is illegitimate if the public wants it? Is that consistent with our extreme position on consumer sovereignty, which is that no matter what horrible things the public wants, as free market economists we can never question it. That’s certainly one of the basic principles of neo-classical economics. Consumer sovereignty is both the end of the story and the beginning (Conversation with Claire Friedland, October 1997).
Starting from this premise, we must logically return to the tenet fabricated by Armen Alchian, someone “who is more Catholic than the Pope, who never went to University of Chicago but is a real Chicagoan” (Conversation with Paul Samuelson, October 1997). What is must be what is most efficient.Footnote 38 To emphasize this point once again, if markets meet the test of time, then an inefficient situation must eventually be replaced by an efficient one. There is clearly an incentive for some entrepreneur to attempt to more closely meet consumer preferences if they are not currently being met and met in an efficient manner. Seizing such opportunities is how economic profits are gained. This must extend to the political marketplace as well, again with a lag, but certainly over time. If the current government or set of existing institutions are not meeting voters’ preferences, then they will be replaced. The parallel logic tying together the two markets is compelling. Competitive markets of any type must ultimately align themselves to reflect individual preferences. It is, then, no more the role of an economist to question the objectives of those political voters than to evaluate the choices consumers make in the marketplace. Besides, as Stigler (Reference Stigler1971) has pointed out, legislation, especially in the case of regulation, reflects relevant interest groups. Accepting these imposed constraints, an economist’s role is automatically limited to understanding the political arena. Despite concentrated effort, this particular market mechanism will continue to run its course unimpeded by the policy prescription of any academic.Footnote 39
Milton Friedman: There’s no problem. It’s true, that George did want to change things.
Aaron Director: But he preferred to study them, not to change them.
Milton Friedman: He preferred to say that he preferred to study them.
Aaron Director: He preferred to study them. I should quit the argument.
Milton Friedman: It was partly a long-running difference between him and me.
Aaron Director: You’re right.
Milton Friedman: And he liked to stress, ‘I just want to understand the world and Milton wants to change it.’
Aaron Director: That’s right. And predominantly I think that is correct.
Rose Friedman: You would have to have them both psychoanalyzed (Conversation with Milton Friedman, Aaron Director, and Rose Friedman, August 1997).
It is perhaps these policy implications or lack of policy directions that make Stigler’s drive for consistency in his work and thought most controversial. If we are to evaluate the Chicago Revolution in economics, along with its connection to the Mont Pelerin program, then the dual aims seem to consist of undermining the post-war Keynesian orthodoxy, both in its theoretical approach and its policy imperatives. George Stigler’s direction toward the end of his career (one which had been growing for a number of decades) quite clearly puzzled many, especially those within his own camp.
I’m sure that if you ask him almost any question of economic policy he was right next door to Milton or me or any of a number of more or less free-market oriented people with a great deal of respect for market forces, etc. Yet, almost I would say not too long after Roofs or Ceilings?[Footnote 40], George began to pull away from advocacy and one of his favorite sayings was that over the course of the centuries, on the issue of free trade, the economists win all the debates in the halls of the universities but the protectionists win all the debates in the halls of Congress.… I think that the reason why you should know why people get dumb policies passed is to learn how better to fight against those dumb policies, rather than any other reason. I am a different style of missionary from Milton, but I’m definitely a missionary. I never thought of myself in any other way. I have strong views about what is good policy, bad policy, in thinking that economics has a lot to teach us, even when it is unpopular, even when governments don’t do it.… I don’t think much of this fatalistic view of these political economy guys that it is just determined by these underlying factors and what can we do about it? I think that would be a terrible vision. In fact, I don’t think George was there. But why was he as reticent as he became over the last decades of his life? … So, the question really is, how does the one George transfuse into the other, or how do these two Georges live side-by-side in the same head. That is a puzzle that I’ve always had (Conversation with Arnold Harberger, October 1997).
What comes across in these series of conversations is the impression that, despite his great respect for Milton Friedman, George Stigler went his own way according to his own lights. He examined the issues and evidence, and did not accept claims blindly. (Though it does seem the case that he was more than willing to defer to Milton Friedman in matters focused on macroeconomic issues where he refrained from claiming any great comparative advantage.)Footnote 41 The two large influences in his later years were Aaron Director and Gary Becker. It is only by understanding Stigler’s relationship to these two close friends and colleagues that we may begin to puzzle out his paradox of legitimacy.
On the whole, if I tell you what I think, I will tell you only what I think. But I will also tell you that this is almost an a priori statement. I think that George was more broad minded than Gary Becker, and so a collaboration between them would have been good for Gary and bad for George … but that’s a purely a priori statement. I really don’t know what it was like when they worked together (Conversation with Robert Solow, October 1997).
IV. A POSTMAN BIDS FAREWELL: WHAT WAS GAINED
‘Indifferent! Oh no—I never conceived you could become indifferent. Letters are no matters of indifference; they are generally a very positive curse.’
‘You are speaking of letters of business; mine are letters of friendship.’
‘I have often thought them the worst of the two,’ replied he, coolly. ‘Business you know may bring money, but friendship hardly ever does.’ (Austen Reference Austen1996, p. 98).
Going through the Stigler papers at the University of Chicago, I was struck by the sheer volume of correspondence.Footnote 42 These are the sort of letters almost no one receives today. Where twenty years ago I would purchase frequent aerograms to keep in touch with distant friends, such overseas post is almost never seen today. Personal letters have shrunk to fill an occasional postcard or birthday greeting. Instead, we have the Internet and text messaging. Certainly, almost all professional discussion and collaboration among academics come via email. For all the great advantages in speed, this bonus inevitably comes at a cost. The quality and character of communications seem to change as the underlying technology evolves. Today, messages and ideas are transmitted virtually in an instant, making it easier to transcribe pure stream of consciousness rather than deliberated thought. The transmissions are not only more ethereal in substance but also in pure longevity. For centuries, historians have pawed their way through piles of often revealing letters. In the future those files will be empty. Such communication is essentially vaporized soon after receipt.Footnote 43 Given this forbidding reality, interviews may gain even greater prominence in piecing together the past. In any case, the potential value of the interview should be self-evident. Who today wouldn’t like to be able to sit down to a tête-à-tête with Adam Smith or chat with his great friend David Hume about Smith’s thought and idiosyncrasies?
More than anything else, talking to the people who were there for an event, in this case the Chicago counter-revolution, gave me a distinct feel for life in the trenches. I understood more about the way in which the battles were fought and why. In particular, subjective feelings and viewpoints come across in interviews in a way they never can by using any other investigative technique. Being in Chicago and talking to so many people who were part of the battles Stigler and Friedman fought, I began to understand, if only imperfectly, the intensity with which they practiced their trade. And I could see how at least some of that intensity emanated from George Stigler, his conscientiousness, his unwavering puritan ethic, and an almost workaholic personality. When faced with people of such accomplishments, my reaction is usually to wonder when they ever found time to sleep.
The result of all these interviews was to shine a rather diffuse light on a complex character, perhaps, in doing so, highlighting the complexities involved in tackling first-rate economic work. “I really feel that he is one of the most interesting and in some sense, most enigmatic characters in our profession in this last century” (Conversation with Arnold Harberger, October 1997).
No one who spoke denied the care and conscientiousness that George Stigler lavished on his data. Few, then, would feel moved to quarrel with his son’s considered observations.
I could only quote my experiences with him in other instances. He was open to being convinced by contrary evidence. He did feel that if you have something that has proven itself through one sequence of tests, that the evidence to overturn it had to be conclusive. One did not overthrow theories casually (Conversation with Stephen Stigler, October 1997).
However, there is another sense in which a wide array of people also concluded that he heavily marketed his results until they squeaked. He championed his underlying conviction of market efficiency almost by force majeure. It was not sufficient for his theory to account for a large majority of all the available evidence; it needed to encompass 100% of the total.
I don’t think it was clear in his mind what the distinction was, between stating a powerful position that covers a lot of the cases, and what happens in a particular event. When confronted with a particular application he literally almost believed it. He wouldn’t back off and say ‘Well, look, that’s part of the remaining 10 or 20 percent, which is doing better than explaining just fifty percent—tossing coins—that’s just part of the noise.’ Methodologically he understood that but in his gut I don’t think he really understood that (Conversation with Sam Peltzman, October 1997).
Anyway, the essence of it was that the people there absolutely denied anything about price flexibility, hidden discounts or things like that. They put out a price book and that was that. Everybody paid the same price. There were quantity discounts that were clearly stated in this price book. Anyone who bought large quantities got the large quantity prices, and small quantities got small quantities prices. George Stigler was clearly disappointed in that. He came back and he told me, ‘Well, we’ll get to the bottom of this’ (Conversation with James Kindahl, October 1997).
Perhaps a need to sell his own results in the marketplace for ideas induced him to assume a cloak of unshakable confidence he truly did not feel.
You have to remember this kind of public persona he had. If something bothered him a lot I saw the side of him that said, ‘I don’t know what to do about this problem!’ But the rest of the public saw that other side. ‘Here’s what I’ve done about this problem, and isn’t it convincing.’ (Conversation with Claire Friedland, October 1997).
He was, however, reluctant to admit mistakes.
I told George about it.[Footnote 44] I asked him what to do about it. This was some twenty years later. George’s answer was that there was no point in making a big fuss about this mistake because it was twenty years ago and nobody cared.… But we have never done anything publicly to correct our mistake (Conversation with Claire Friedland, October 1997).
How much, if any, of this was ideological in essence and how much of it attached to which aspects of his work is difficult to determine solely from the various conversations undertaken. For all Stigler’s fair-mindedness, he saw the perceived threats to liberty and freedom in the fifties and sixties as dead serious.
The proof that there are dangers to the liberty and dignity of the individual in the present institutions must be that such liberties have already been impaired. If it can be shown that in important areas of economic life substantial and unnecessary invasions of personal freedom are already operative, the case for caution and restraint in invoking new political controls will acquire content and conviction (Stigler Reference Stigler1975, p. 18).
This led to a seeming contradiction that identified him clearly as a vocal critic of the campus excesses and the constraints posed to academic freedom in the sixties, yet recognized his willingness to be mute during the fifties reign of terror perpetuated by the advent of McCarthyism.
I really have absolutely no idea. But it wouldn’t surprise me that he had a schizophrenic attitude to attacks on intellectual leadership depending on which direction it came. A lot of right-wing people must have found McCarthy to be vulgar, brutal but basically less of a problem than the alternative. Basically on the right side of the issues. Not unlike what you read that people said about Hitler, in the early ’30s, when he was on the rise. Yes, he’s a gangster. Yes, he’s a hooligan but he’s anti-communist. He’s going to keep back the communists and control trade unions. These are benefits. And they didn’t realise that they were rearing a Frankenstein’s monster who would eventually chop off their own heads. A lot of people were sympathetic, who would have been horrified if they had foreseen what would happen (Conversation with Mark Blaug, October 1997).
What does appear true about all academic ideologues is the one-sided nature of their skepticism. While those on the left are deeply skeptical about market outcomes, they may hardly question the results of potential government intervention. On the right, the tinge of iconoclasm shifts in the opposite direction. Markets can do no wrong and government intervention is automatically deemed guilty unless proved innocent. While sounding absolutely assured is the key to good marketing, more modesty and a willingness to acknowledge what we do not know may get us into less trouble.
It’s hard to imagine an empirical observation that would convince most members of this department [University of Massachusetts] and the University of Chicago to change their minds. My personal view is that if someone holds a view it cannot be dislodged by any conceivable empirical data. Evidence from a data system doesn’t convince them. These people have made their decisions already. They’ve become true believers and no amount of empirical evidence will ever convince them by definition (Conversation with James Kindahl, October 1997).
From these conversations, the question of whether Stigler could be moved only by certain selective mountains of evidence remains unresolved. What does come across is his need to market his ideas, which acted as a curb on his otherwise healthy sense of skepticism. Successful marketing moves us away from the sort of skepticism and essential modesty that Knight claimed was a key component of a scholar’s personality: “Frank Knight … tells us that the ‘basic principle of science—truth or objectivity—is essentially a moral principle, in opposition to any form of self-interest. The pre-suppositions of objectivity are integrity, competence and humility’” (Coase Reference Coase1994b, p. 15).Footnote 45
If we take Stigler’s own work to heart, to succeed in the marketplace for ideas, you have to focus on meeting your listeners’ given preferences, convincing them that what you are selling meets their requirements. Unfortunately, this may entail a sacrifice. As Gary Becker points out, success in terms of marketing need not lead to a desirable outcome, one to which George Stigler, given his abiding respect and devotion to the economics profession, would happily subscribe. Whether Stigler fell into this seductive trap where a priori beliefs determine factual findings is a question posed by these conversations rather than resolved by them. An honorable man may be true to his methodological reasoning and techniques, but be possessed by a creed at times so powerful that it can blind him to uncongenial facts or at least leave him feeling that he needs must willfully ignore whatever contradictory evidence arises.
It may be that in the long run good ideas do surface but they surface faster, if written in a persuasive fashion. Moreover, bad ideas may be put persuasively. And they may gain the necessary threshold. However, taking that same analogy in competition among ideas, there is a presumption, although not a certainty, that in the longer run, the good ideas are going to compete out the bad ideas. But that may take a long time and may not even always operate. There’s nothing necessary about that. Nothing guaranteed about that (Conversation with Gary Becker, October 1997).
