Peer effects in companies

"Corporations are people" said Mitt Romney. He was more right than he knew, because two new NBER papers by Christopher Parsons and colleagues show that companies are like people, in the sense that they are prone to peer effects:

We find that a firm's tendency to engage in financial misconduct increases with the misconduct rates of neighboring firms. This appears to be caused by peer effects, rather than exogenous shocks like regional variation in enforcement. Effects are stronger among firms of comparable size, and among CEOs of similar age.

In this sense, firms are just like ordinary people in that they are more likely to commit crimes if their peers do. This might be one reason why so many banks were involved in fixing Libor and in mis-selling PPI.

Secondly (pdf):

We find that a firm's investment is highly sensitive to the investments of other firms headquartered nearby, even those in very different industries...
We, of course, cannot rule out the possibility that area co-movements arise because of irrational herding, which would be the case if managers put too much weight on the beliefs of their neighbors.

Again, there's a direct analogy here with people;there are peer effects in individuals' spending, perhaps in part because their expectations are infectious.

There is, I suspect, an important political point here. There's a tendency to defer to "business leaders" as if they have superior insight into the economy; how often are they uncritically interviewed on the Today programme, for example? Such deference is misplaced. Rosewell and Ormerod have shown that "bosses have very limited capacities to acquire knowledge about the true impact of their strategies." And Charles Lee and Salman Arif have shown that higher capital spending leads to disappointing profits, consistent with it being driven by sentiment rather than a rational assessment of profitable projects. We should read Parsons' research in this tradition. It suggests that corporate bosses are not wise and rational leaders but are instead as easily led as impressionable teenagers.

So, why are we paying them so much? 

Comments

[Calum]

There is also the simple point that employees move between institutions. Regularly, in finance.

[Jamie]

I agree with some of this but not all of it.

There is an observable phenomenon in human activity which involves copying ideas.

When Apple invents the iPhone, other firms attempt to copy Apple in the specifics of the design & features of its products, and also in its management practices. Apple tries to protect itself by taking out patents but patents just slow down the copying process. The first mover takes the majority of the risk so expects the majority of the reward. The copiers try to keep up with the first mover without themselves incurring much risk. Both strategies can work given the right circumstances.

This type of behaviour occurs in all walks of life including science, sport and the arts. It also occurs on a personal level via “keeping up with the Jones’s” behaviour.

While this phenomenon is mostly positive, it can also have negative repercussions.

If one trader in a bank succeeds because he takes more risk than other traders then the other traders will copy this behaviour. In order to re-establish superiority the first trader then has to take greater risk. This escalates until it fails, often spectacularly.

Business leaders are no different from anyone else in this regard. The behaviours that allow Steve Jobs to create the biggest company in the world are the same ones which can lead to disaster. Again, this is no different from other walks of life. The only people who die in mountaineering accidents are those who place themselves at risk on mountains in the first place.

What conclusions should we draw?

First, this type of copying behaviour is the antithesis of the “representative agent” who makes decisions independently of anyone else.

Second, someone has to be the first mover, and the first mover faces the greatest risk of failure. Hence, the first mover expects the greatest reward. If risk taking is not rewarded then why would anyone innovate? Often success and failure are separated only by a few (mis)judgements and/or slices of (bad) luck. If everyone adopts the copying strategy then no innovation occurs.

Third, I do agree that we shouldn’t defer to business leaders (or anyone else). A large business is far too complex an entity for anyone to have a fully rounded understanding of the whole system. Successful large businesses thrive because they use the many different perspectives of the people in the business to create something greater than the sum of the parts. Unsuccessful ones often rely too much on one or two perspectives.

Fourth, the same thing is true when we study any large system. When economists claim to understand the entire economy based on one specific perspective which isn’t consistent with observation then we should suspect that they are deluded. When they refuse to acknowledge the different perspectives of other economists, or the perspectives of the participants in the economy, then we should know that they are deluded.

Ultimately, life divides into participants and observers. The participants who take the risks create all of the innovations and make all of the mistakes. The observers don’t. Keynes understood the importance of risk & uncertainty but I think that this has been lost by many modern economists and other observers. An observer may see one or two things that are not seen by a participant, but the participant sees many more things that are not seen by the observer.

[neil]

Because they're the Mob of our age. As you just outlined.

[Stevenclarkesblog.wordpress.com]

Plato thought "the city was the soul writ large" - why should not the corporation also be the soul writ large?

The corporation being composed of individuals will necessarily share some of the virtues and vices of those individuals (I'm sure this applies just as much to the shop floor worker as the CEO - and I'm sure much of their behaviour is subject to peer effects. I think it was Tim Harford who talked of the checkout assistants at a supermarket who worked as fast as the coworker they were facing.)

Surely the key is to educate people, especially CEOs, to be aware of their liability to peer effects and train them to discount it when making decisions (if possible).

This post suggests that economics needs to incorporate social psychology.

[theOnlySanePersonOnPlanetEarth]

"While this phenomenon is mostly positive, it can also have negative repercussions."

Assertion alert, assertion alert!

"life divides into participants and observers."

What is the scientific basis of this astonishing new bit of insight into human development? What process is at work here? Do animals follow the same pattern? Is it an evolutionary law? Has your theory been peer reviewed. Would love to learn more about this latest theory of human development. I checked wikipedia and couldn't find anything.

[An Alien Visitor]

"The corporation being composed of individuals"

But some are more individual than others!

(To paraphrase the old observation about the USSR - everyone is equal but some are more equal than others)

"Surely the key is to educate people, especially CEOs"

especially the CEO's being the point I suspect!

I guess this is the problem with peer pressure theory, it assumes an unconscious criminality, it assumes people were not aware of exactly what they were doing and a bit of re-education will solve the problem. And while it says peers are more likely to become criminals it doesn't sat makes someone a peer in the first place.

I guess for criminals there is not just the element of education but the much bigger element of punishment.

I favour punishment over education in this regard.

"I think it was Tim Harford who talked of the checkout assistants at a supermarket who worked as fast as the coworker they were facing"

I have never been in a supermarket where the checkout staff faced each other. They are always side by side, and we the customer go out the same door.

[Stevenclarkesblog.wordpress.com]

@An Alien Visitor

They face the back of another checkout assistant - enough to see the pace they work at.

Some experiments were done, where the pace of scanning was measured. Checkout assistants viewing fast workers worked fast, those facing slackers also slacked.

[An Alien Visitor]

Oh experiments!

How long was that experiment? A day, a week, a year? Do we take the result of the experiment as being the final, absolute truth? Forever to be used in an argument as proof?

In reality, outside the lab, they are too busy to see what the peers are doing, and there are usually shoppers obscuring the view anyway!

[Stevenclarkesblog.wordpress.com]

@An Alien Visitor

It took part in a real supermarket for 2 years.

And I got the conclusion wrong - it was BEING LOOKED AT by a fast worker that sped you up, not LOOKING AT a fast worker.

The article is here:

http://www.ft.com/cms/s/0/226ef8f0-8a78-11db-8940-0000779e2340.html#axzz39dTTfA3F

[An Alien Visitor]

So a totally different experiment then!

So this experiment, that lasted for 2 years, involved 'fast' workers looking at 'slow' workers and someone was writing down if the effect of being looked at, to see how fast you worked, increased how fast you worked!

I have never seen this in practice in any supermarket. I wonder why?

Maybe in the real world the 'slow' workers are more sociable with the customer than the 'fast' worker. Maybe that sociability is important in retaining customers and fast workers give a 'bad impression'. But who knows?

One thing is for sure, you have proved nothing.

[Dave Timoney]

The Mas and Moretti experiment is essentially a reconfirmation of "soldiering", which is a form of peer pressure, with top notes of environmental surveillance a la Foucault's Panopticon metaphor.

This is all well and good, but it isn't of much relevance to the issue of peer pressure among CEOs and other execs, and how this translates into suboptimal business decisions.

A good study of how politics, (neoliberal) ideology, social norms and culture all combine to create institutionalised criminality and herd stupidity is Fintan O'Toole's 'Ship of Fools'. Long story short: people do evil when they think they can get away with it; and when everyone else is bent, you're disadvantaged if you're not.

[Stevenclarkesblog.wordpress.com]

I wasn't trying to prove anything. I put in a throwaway remark about an article I half remembered and it seems to have consumed the comment thread. Whoops!

Thank you @FATE for the recommendation.

[An Alien Visitor]

I think it was worth the probing, if you had said this was a throwaway remark about some fundamentally flawed experiment that you had only half remembered in your first comment we could have avoided all this.