Risk, Ambiguity and Daniel Ellsberg
The death of Daniel Ellsberg on Friday reminded me of his contribution to economics and his influence on my own thinking. In 1987, I was at Cornell, beginning an abortive PhD candidacy. In one of my courses there was an assigned reading on decision theory by Leonard Savage. One of the footnotes referred to an article by “Daniel Ellsberg” and I naturally wondered if it was the same Daniel Ellsberg of Pentagon Papers fame. “Risk, Ambiguity and the Savage Axioms” was indeed by the same Daniel Ellsberg. It also happened that a copy of his PhD dissertation was available at the Catherwood ILR library, which I read eagerly.
Ellsbergian ambiguity is very similar to Keynesian uncertainty with the important distinction that Ellsberg performed experiments, enlisting the leading decision theorists of the day, to demonstrate that these cheerleaders for rational decision making readily abandoned rational choice when faced with an ambiguous choice. I will only mention the beauty and sheer brilliance of his demonstration, if you want to experience it you will have to read it yourself. I can’t do it justice in describing it. When you read Ellsberg’s article and his dissertation, it becomes clear that his writing of the Pentagon Papers, as well as his public disclosure of it were deeply theoretically grounded. Ellsberg documented the persistent and pernicious irrationality of U.S. government officials and institutions when confronted with an ambiguous problem.
My own thought and research has been deeply influenced by Ellsberg’s work on ambiguity, which makes me somewhat of an anomaly. In the past few years I have focused on Karl Marx’s Grundrisse with a rather unconventional interpretation. A large part of that unconventionality is that I perceive ambiguity and ambivalence where Marx’s followers — and perhaps Marx himself — sought certainty.
Very interesting. I hope you post further ruminations on this topic.
One of the challenges of teaching first year medical students was to get them to appreciate the need to deal with ambiguous and incomplete information in the clinic. What more do I need to know and when do I know enough? These are students who excelled at multiple choice exams, where the challenge is to pick the single correct answer from a universe of five possible answers. Life isn’t like that.
Many academic disciplines, including economics, suffer from physics envy, convinced that there are laws to be discovered that are always and everywhere true. Reflexively citing laws is for religions. In real life, the best we can do as a guide to action is the weight of the evidence, which can change over time.
https://sites.socsci.uci.edu/~bskyrms/bio/readings/ellsberg.pdf
1957-1960
RISK, AMBIGUITY, AND THE SAVAGE AXIOMS
By DANIEL ELLSBERG
I. Are there uncertainties that are not risks?
II. Uncertainties that are not risks.
III. Why are some uncertainties not risks?
I. ARE THERE UNCERTAINTIES THAT ARE NOT RISKS?
There has always been a good deal of skepticism about the behavioral significance of Frank Knight’s distinction between “measurable uncertainty” or “risk,” which may be represented by numerical probabilities, and “unmeasurable uncertainty” which cannot. Knight maintained that the latter “uncertainty” prevailed – and hence that numerical probabilities were inapplicable – in situations when the decision-maker was ignorant of the statistical frequencies of events relevant to his decision; or when a priori calculations were impossible; or when the relevant events were in some sense unique; or when an important, once-and-for-all decision was concerned…
“Ellsbergian ambiguity is very similar to Keynesian uncertainty with the important distinction that Ellsberg performed experiments, enlisting the leading decision theorists of the day, to demonstrate that these cheerleaders for rational decision making readily abandoned rational choice when faced with an ambiguous choice….”
Really, really important right now and in parallel with what Jeff Sachs and James Galbraith are now finding.
Well, Keynes performed self-funded experiments, too. He made, lost and then re-made a fortune investing in the stock market. Not the same experiments as Ellsberg, but real-world experiments nevertheless. Keynes put his money where his mouth was.
“Ellsbergian ambiguity is very similar to Keynesian uncertainty with the important distinction that Ellsberg performed experiments, enlisting the leading decision theorists of the day, to demonstrate that these cheerleaders for rational decision making readily abandoned rational choice when faced with an ambiguous choice….”
[ Please set down any specific examples that you have in mind. Even a single example will be most helpful.
I have finished carefully reading the Ellsberg essay. ]
The examples I have in mind are rather complex. The ambiguity follows from the complexity. This is one of the inherent problems of examples. They are simplifications that typically abstract from the ambiguities.
Take Ukraine, for example, on the one side there is the simplification of Putin, the evil invader, versus the heroic Ukrainian people and their charismatic leader, Volodymyr Zelenskyy. On the other side is the long history of aggressive U.S. political and military machinations to encircle and “contain” Russia to the extent of denying it its own sphere of influence. To pick a side, one must suppress part of the complexity and ambiguity. Even those nutshell descriptions shy away from ambiguity.
In the case of Vietnam, the deciding factor was often how something would play politically, with Democrats afraid of being labelled “soft on communism” and Republicans eager to prove their anti-communist machismo. Thus strategic decisions were made in bad faith for political advantage. Even that, of course, is a simplification. A lot of people believed their own rationalizations.