Feedback – rapidly building noise can lead to madness of the crowds

Electronic communications media can suffer from something akin to the noise you get when the microphone is too close to the speaker. Some messages are picked up, amplified, and repeated—only to be picked up and amplified over again. With noise it’s annoying, but with ideas it can be dangerous.

Feedback can be dangerous

Our electronic media are designed to take a message and make it easy to transmit to a large number of people. This, of itself, can be a good thing—like the microphone, amp and speaker that let us hear the best mans speech at a wedding. The difference comes when things go wrong. If the best man stands too close to the speaker one note gets picked up and repeated until it builds to an unpleasant noise. If a message that prompts people to act in a particular way goes wrong we can get economic or social chaos.

Electronic share trading systems contributed to investor panic to create the global stock market crash known as Black Monday in 1987. By that time many stock-broking firms were using then ‘state of the art’ trading programs which would automatically buy or sell stocks based on price thresholds. When markets stared to fall some computer started to reach their ‘sell’ thresholds and dumped stock. This caused prices to fall triggering more selling, by both humans and machines. The feedback cycle did not break until after a number of trading halts for individual stocks on the New York Stock Exchange (NYSE) and both the Chicago Board Options Exchange (CBOE) and Chicago Mercantile Exchange (CME) stopped trading.

While there are systems in place to try to prevent this happening again the ‘Flash Crash’ (May 2010) shows that there is an ever present danger of feedback in trading systems. Again it appears that a break in trading stopped the feedback effect.

But the economic dangers of ideas feedback go way beyond trading systems. All asset price bubbles are symptoms of idea feedback—where too many people share the same information we get the madness of the crowds. 

In Ireland we are suffering the effects of a property price crash after an exceptional price bubble. The bubble in prices in Ireland was in part driven by information from here and across the globe showing that property prices were rising. Once this ‘good news’ story took hold it created its own feedback loop: Banks were keen to lend money on property because asset values was rising. Sellers were keen to sell because they were making more money. Buyers were keen to buy because they were afraid to be left off the ladder. News media were keen to promote the story because they were making more advertising revenue. Everyone in the property business was keen to keep telling the same story because they were making more money. Even the Government wanted this to go on because it was taking in more taxes, and the economy was booming. 

All of those messages were being amplified by electronic media. Anyone interested in buying or selling property was bombarded by advertising messages by all of those in the property business: Banks lending money, sellers, developers, property supplements in the newspapers, specialist websites, even government ministers saying this is a good thing.

You can’t hear the noise

This type of feedback is more difficult to react to because, unlike a howling loudspeaker, it can be difficult to detect. We can’t tell the difference between the real story and the feedback. And in this case, as in many price bubbles, the feedback actually changed the story.  People did not just believe that it was wise to buy property at higher prices, they went out and bid against each other to drive the prices up.  The alternative view was out there but was drowned out by the noise.  In most cases friends and family would tell you what they had read, heard and seen was that prices were going to continue to go up.

This is where the real danger in electronic media feedback comes in. The more we communicate the more we risk the madness of the crowds. James Surowiecki, author of The Wisdom of Crowds, suggests that for group intelligence to work the group needs to be diverse, preferably using some different ideas and information to come to a conclusion.  The more we share information the more likely we are to create the conditions where we, as groups, act irrationally.

I think one safe bet is that we will see more, and more frequent, boom and bust cycles before we get to grips with the feedback phenomenon. 

It’s not just financial ideas that get picked up and amplified, cultural ones do too. For example the idea that children can be damaged by compliments, but should be praised for effort, as described in NurtureShock by Po Bronson & Ashley Merryman has taken hold widely. So we now have children’s TV programmes, like Special Agent Oso, where the audience are shown a medal and told that this is a reward for all their hard work—presumably shouting at the TV when prompted.

We are taking decisions in a world of electronic media using minds which evolved in an unplugged world. Before the spread of mass communications we made decisions based on tried and tested ideas passed from generation to generation; on our weighing of the advice from a few trusted sources; and what we see around us. Now we have gone electronic, and we have access to information from a myriad of electronically amplified sources across the world. Figuring out how to balance those sources is hard enough. The trouble is that any amplified system is prone to feedback. So a lot of what we are hearing is the same message repeated over getting more amplified all the time.

It is said that Joseph Kennedy decided to get his money out of stocks and shares prior to the 1929 crash when even shoeshine boys were giving stock tips.  Recognising that feedback happens, and being on the look out for its signs is one way we can protect ourselves from feedback cycles.

Even more import we need to remember that, if we are the ones holding the microphone, don’t point it at the loudspeaker.