develop (slowly) into a huge epidemic if the forgetting rate is low enough.
Narratives also differ in their recovery rate or forgetting rate. Narratives with
high recovery rates often are isolated, not part of a constellation. Narratives with
low recovery rates include those with constant reminders. For example, when we
see homeless people and beggars on the streets, we remember narratives about
massive unemployment during a depression. Longer-term narratives are more
likely to have an impact on one’s view of the world or one’s sense of the
meaning of life.
As the mathematical model in the appendix shows, a high contagion
parameter and a low recovery rate mean that almost the whole population
eventually hears the narrative, sometimes very quickly. But the same narrative
can reach most of the population rather slowly if the contagion parameter is low
but the recovery rate is even lower. The following example is illustrative.
I conducted a questionnaire survey in the United States right after the October
19, 1987, stock market crash, which was the biggest one-day drop in US history.
I asked a random sample of US high-income individuals exactly when they first
heard about the crash. Of the respondents, 97% said they heard of it on the day
of the drop. The average answer was 1:56 p.m. Eastern Time / 10:56 a.m. Pacific
Time.1 Most of the respondents did not hear about this drop via the morning
newspapers or the evening television news. They heard it by direct word of
mouth as the event was happening.

Proposition 2: Important Economic Narratives May Comprise a
Very Small Percentage of Popular Talk
In trying to judge the importance of economic narrative epidemics, we should
not base our conclusions on the assumption that the most economically
important narratives are those that are constantly talked about. Very significant
epidemics may generate very little talk. In addition, because people are always
talking, some kind of narrative is always spreading. In studying economic
narratives, we must not be distracted by the small talk that is not useful in
explaining economic changes.
In 1932, near the height of the Great Depression, Franklin Roosevelt
challenged incumbent Herbert Hoover in the US presidential election. Writing
for the New York Times, Pulitzer Prize–winning journalist Arthur Krock tried to
summarize what ordinary people were saying about the economic situation. He
listened to people talking, “avoiding prompting as much as possible”:2
By train, motor car, airplane and on foot I have wandered 10,000 miles. I
have talked with, observed and listened to many hundreds of people on trains,
in restaurants, on the streets, in speakeasies, in hotel lobbies, in clubs and in
their own houses.
He visited twenty US cities over the course of a month and wrote down casual
conversations he’d had, or overheard, word for word, that seemed to exemplify
what people were saying. He was a little surprised that almost all of the talk was
banal:
Little did I hear of books or plays. Not one new joke was told by a drummer
in my hearing. Not a word of personal enthusiasm for any candidate for office
did I hear.
Krock’s article stands as a warning not to be complacent about narratives that are
contagious only in certain venues, and that are not talked about except at certain
times. Economic theories are not the topic of casual conversations, even though
the news media discuss economic ideas frequently, and people must be thinking
about them.
Krock found that people wanted to talk incessantly about the effects and
terrors of the Great Depression. For example, he records the words he heard in