Experimental Evidence on Virality
Experimental evidence shows that the success of individual creative works
depends on how people assess the reactions of others who are observing the
work. In one experiment,23 sociologist Matthew J. Salganik and his colleagues
set up an “artificial music market” online. The market included an array of songs
that customers could listen to, rate, and, if they chose, download. Unknown
bands performed all the songs, and none of the listeners had ever heard any of
the songs before taking part in the experiment.
This artificial market simulated real online markets in that subjects never
communicated with one another except that they could observe the popularity of
songs. This popularity ranking was the only “spark.” The subjects were
randomly assigned to two conditions: independent and shared. Those in the
independent condition had to choose songs entirely independently, never seeing
others’ choices. Those in the shared condition were divided into eight worlds and
saw others’ downloads in their own world only. In the extreme shared condition,
the computer screen always showed the songs in rank order in terms of
popularity measured by downloads. The first subject-customer to buy in each
shared-condition world saw no information about others’ choices, the second
customer saw the first customer’s first choice, the third customer saw the first
two customers’ choices, and so on.
The researchers found that each of the eight worlds developed its own set of
hits, only imperfectly correlated across worlds, and that the inequality of success
across worlds was uniformly higher than in the independent world where
customers never saw information about others’ choices. It seems logical to
conclude that something about the random initial choices in the shared worlds
got amplified as time went on. In the real world, the effect is likely even stronger
because real-world marketers attempt to play up the audience size as much as
possible. This research may be taken as experimental confirmation that random
small beginnings can lead to big epidemics.
The lesson is that history, including economic history, is not the logically
ordered sequence of events that is presented by subsequent narratives that try to
make sense of it or try to achieve public consensus. Major things happen because
of seemingly irrelevant mutations in narratives that have slightly higher
contagion rates, slightly lower forgetting rates, or first-mover effects that give

one set of competing narratives a head start. These random events can feed back
into bigger and more pervasive narrative constellations, as we will see in the
next chapter, which examines the narrative constellations associated with the
famous (or infamous) Laffer curve.

Chapter 5
The Laffer Curve and Rubik’s Cube Go
Viral
One of the toughest challenges in the study of narratives is predicting the all-
important contagion rates and recovery rates. Despite all the work by
epidemiologists and other scholars, we can’t precisely observe the mental and
social processes that create contagion, and so we have trouble understanding
how they play themselves out.1
To take an example from popular culture, predicting the success of motion
pictures before their release is widely known to be all but impossible.2 Jack
Valenti, former president of the Motion Picture Association of America, said:
With all the experience, with all the creative instincts of the wisest people in
our business, no one, absolutely no one can tell you what a movie is going to
do in the marketplace.… Not until the film opens in a darkened theater and
sparks fly up between the screen and the audience can you say this film is
right.3
Screenwriter William Goldman had a similar thought, in the opening lines of his
book:
Nobody knows anything. Not one person in the entire motion picture field
knows for a certainty what’s going to work. Every time out it’s a guess and, if
you’re lucky, an educated one.4
In fact, many films and songs by one-hit wonders5 attest to the difficulty of
going viral. The same person who’s had a hit often can’t do it again. Also, hits
from past years never seem to become real hits again, at least not without
significant modification.
Economics has its own one-hit wonders, including the now-infamous Laffer
curve. Examining how this economic narrative went viral provides further
insight into how economic narratives lead to real-world results.