Bryan Caplan  

Naming the Puppy: Firing Aversion and the Labor Market

PRINT
Corporatism... French versus American Parenti...
In fiction (and "reality" television), firing workers almost seems fun.  How many times has Mr. Burns gleefully hissed, "Fire than man, Smithers!"?  In the real world, though, bosses dislike being the bearer of bad news.  They feel guilty when they let someone go.  So guilty, in fact, that some hire consultants to help them fire people.  To coin a behavioral econ phrase, most employers feel "firing aversion."

How does firing aversion play out in the real world?  For starters:

1. Firms often fail to hire workers who would be profitable in the short-run.  Why?  Because in the medium- or long-run, conditions might change.  Robots would respond by coldly discarding superfluous workers.  But many human bosses won't.  A human boss might feel guilty enough to continue paying the worker more than he's worth.  Or he might fire the worker and feel like a jerk.  Either way, it's not like returning a pair of pants to CostCo.  Upshot: If a job candidate is a close call, a boss who foresees his own psychological reaction will say, "No thanks."

2. Signaling matters more.  Suppose bosses can learn workers' abilities in one of two ways: (a) Examine their educational credentials, or (b) give them a chance to prove themselves on the job.  For robots, (b) might be an attractive option.  For human bosses, however, (a) has more appeal.  If you hire based on credentials, you never even have to meet most of the subpar candidates.  If you hire based on trial-and-error, in contrast, you get to know a lot of people, then dash their dreams.  Once again, a boss who foresees his own psychological reaction tailors his strategy accordingly.

3. Outsourcing looks better.  For most people, firing a visible human being hurts a lot more than firing a company.  Firing a gardening firm doesn't feel so bad.  Firing a gardener does.  As a result, firms will outsource more than narrow profit-maximization recommends.

Big picture: When you hire someone, you "name the puppy."  You accept them into your tribe.  Yes, their initial status is relatively low.  But if you're psychologically normal, the lowest member of your tribe still counts in your eyes.  You'll exile a person from your tribe if they're a massive burden or a traitor.  But if they're merely a moderate disappointment, you'll probably show mercy and lend them a hand. 

This may sound good.  But it's a mixed bag at best.  Yes, firing aversion makes life more secure for the employed.  But it also makes it harder to find a job in the first place.  From a social point of view, the labor market would work better if employers were as free of firing aversion as robots.

P.S. Two requests for the comments:

1. Can you name additional plausible real-world effects of firing aversion?

2. Do you have any real-world examples to share (with names changed to protect the guilty)?
 


COMMENTS (25 to date)
Joel writes:

The primary reason companies are averse to firing people is that they're worried about being sued, and the secondary reason is that (in certain cases) they're worried about the morale impact on the people not being fired. (Although firing really terrible employees can have a positive morale impact.)

I have never worked at a company where "the firing manager might find it unpleasant" was a serious obstacle to firing anyone.

Doug writes:

Most firms have "designated firers." In large firms this is often a dedicated HR person. In smaller firms it tends to be the most willing senior employee or partner. All things considered the returns to being a hard-ass increase substantially in a world filled with squeamish pansies. At many firms the designated firer is a mediocre worker who earns outsized compensation or status for "doing the dirty work."

Daniel Klein writes:

Super great post. I look forward to the comments.

The firing scenes in Moneyball were great.

Thomas writes:

In my experience (as a senior manager at a 500-employee company) the possibility of being sued weighed on us, but if an employee's performance wasn't acceptable we simply made sure that we were on solid legal ground before firing the person. That meant going through the warning, probation, documentation process before pulling the trigger -- except in cases of blatant misbehavior or insubordination, for which immediate firing was the usual response.

Our fastidiousness didn't make the task of firing someone any more pleasant, but it meant that a manager could fire someone with greater confidence that the company would not suffer as a result.

Did we ever fail to fire someone for fear of being sued? No. But we did give some people a lot of rope before firing them because they were in "protected" categories or had been with the company a long time. (Firing a long-time employee can be bad for morale -- unless the employee is widely disliked or disrespected -- and it invites a charge of age discrimination.)

As for the effect of "firing aversion" on hiring, here's my take: The risk of facing an EEOC claim or being sued for not hiring someone was almost as great as the risk associated with firing someone. For that reason, we were careful to document the basis for rejecting applicants, with special attention to those in "protected" categories.

On the whole, our hiring decisions were not affected by the prospect of having to fire someone. But that may have been due to the nature of our company: a not-profit with relatively stable funding. For-profit managers in highly competitive markets might well suffer "hiring aversion."

Thomas writes:

At my company, we had a couple of employees who were responsible for opening payments received from customers. Our customers don't send us many payments anymore--they wire the funds instead of sending a check. But no one wanted to fire the people we didn't need, so they stayed in place, doing less and less. And then the recession hit, and fat had to be cut, so they were out of a job, and there weren't any other jobs available.

One other real world effect of firing aversion is that it can be overcome at times that are particularly inopportune for the fired employees. In the case of my company, they'd likely have been better off if they'd been fired two years earlier.

nazgulnarsil writes:

what Thomas says indicates to me that firing aversion makes downturns worse. If firms are relatively similar in their reactions to exogenous shocks they will choose similar times to cut the fat, thus dumping excess workers on the market all at once.

nzgsw writes:

In my experience, firing aversion leads to batches of mass layoffs like Thomas mentions. I've come to think of it as the labor market equivalent of big bath accounting.

Thomas writes:

I am the first Thomas. The second Thomas and nzgsw are right about the batch effect. In our company, we took advantage of budget crunches to fire workers whose performance was sub-par but not bad enough to warrant immediate action. Our aversion, in such cases, was associated with the employees' length of service. We had little compunction about firing employees who had been with the company only a few years or less. But, again, we were slow to fire long-term employees, for reasons of morale and concerns about EEOC claims. We would do it without a budget crunch, but the crunch made the task easier.

For those readers who have never dealt with the EEOC, know this: The burden of proof is on the employer to rebut a fired employee's claim, not the other way around.


Mark Little writes:

Bryan,

You ask for comments, so here are mine.

As a manager in the software industry, I've had multiple personal examples (especially in earlier years, when my hiring mistakes were more frequent).

From my experience your post is mostly accurate, at least up to points 1 and 2. For point 3, not so much, at least as concerns core business functions.

As for you last point, that the labor market would work better if there were less firing aversion, I think this is simply wrong, at least for highly skilled positions. Arnold Kling in several posts relates hiring to organizational capital, referring to the "Garett Jones economy". I've not read Garett Jones, but I agree with Arnold on this. I'll refer you to him rather than elaborate on why more firing is not optimal.

Another motive for firing aversion that you don't mention is that the need to let someone go often reflects a failure of management in not hiring a good fit for the position. In addition to a natural reluctance to face up to such a serious mistake, as a hiring (and firing) manager one feels guilty inflicting such a serious consequence on someone, when the situation is as much or more your fault as theirs. (After all, we expect applicants to do what they can to get the job; it's the hiring manager's job to judge whether that is likely to work out or not.)

@Joel: Not in my experience.

@Doug: Not in my company. (You have to work through HR, but I've always had to make and take responsibility for the decision.)

@Thomas: Sounds like we work in different industries, but your experience sounds similar to mine.

@nazgulnarsil: Your point would seem to support Tyler Cowen's ZMP workers theory.

Noah Yetter writes:

We had an individual hired onto our team (of software developers) a number of years ago, we'll call him "Jimmy". He was hired over my objection, but it was a difficult position to hire for and we were tired of having it open, so I did my best to work with this guy.

Within a few months, it was clear we had made a mistake. He was an hour late his first day. His skill was less than advertised. His ability to collaborate with other team members was utterly lacking. He actively subverted our development processes. He invited female coworkers to swinger parties (seriously not making that up). His net productivity was negative at almost all times.

But I knew our boss wouldn't fire him, for all the reasons Bryan outlined above. I bet a coworker $5 he would last past some date, and won. Ultimately he lasted 13 months on the job before finally being fired. Even today some of his code is still around, causing problems.

My theory as to how he finally got himself fired dovetails neatly with Bryan's theory. I think that once he crossed a line of losing our boss' respect, to the point where there was no longer that feeling of empathy/sympathy, and firing Jimmy therefore became vastly less costly.

Jusitn writes:

Good post.

My take on firing aversion is that firms use recessions as an excuse to do it. I believe in animal spirits, loss aversion, and sticky wages, but I also believe that firing people during recessions gives cover for making necessary (and often overdue) changes that improve the efficiency of the firm.

Tom West writes:

I've seen a number of examples of firing aversion in smaller companies, where it required fairly desperate economic hardship before bosses were willing to fire sub-par workers.

However, the trade-off was loyalty. The companies also had many employees who could have earned many times more elsewhere (a fact that the boss brought up more than once) because they realized the company stood behind them wherever it was possible. Those that did leave made every effort to find replacements that were as competent as they were.

As always, loyalty swings both ways. More to the point, while not maximizing economic efficiency, both bosses and employees were, for the most part, happier. Thus form the human standpoint, I'd consider the 'policy' successful.

Anonymous writes:

Is it possible that firing aversion also derives from concern about morale of long-term staff?

Here is my first-hand example of firing aversion, stated very generically to protect the guilty. A business case was produced at the company I work for which indicated a preference for outsourcing the work so that it would be easier to terminate the project if it did not work out. (Exactly as you describe above.) In practice, even though the arrangement failed to produce the desired outcome, the vendor has not been terminated.

PrometheeFeu writes:

@Thomas:

"but if an employee's performance wasn't acceptable we simply made sure that we were on solid legal ground before firing the person. That meant going through the warning, probation, documentation process before pulling the trigger"

Isn't that an onerous process that would change the definition of "acceptable performance"? How much work are you willing to do to fire someone who is teetering on the edge of insufficient performance?

@Bryan Caplan:

How about the cost of hiring? Hiring someone means HR spending time sifting through resumes, managers and others spending time in interviews with candidates, meetings to talk about candidates etc... Some people might choose not to fire simply because they don't want to incur the hiring costs again.

There may also be some departmental politics. It's probably easy to sell maintaining your headcount due to status-quo bias. However, once someone has been fired, it might be open-season on the budget that was spent on that person and the manager will have to once-again spend political capital getting that budget.

Greg writes:

clearly caplan has never worked in a non-ivory tower job.

the MAIN reason for not firing people straight away is that it is not easy to replace the layed off staff. even a waiter is not all that easy to replace because it takes several weeks to really train a waiter (learning where the spare forks are is not as easy as it seems). when i worked in a factory it took me several weeks to get to grips with the speed of the production process and get to a point where i did not have to constantly ask the line manager for help or explanations of what to do if something is not right.

the key thing is not to fire or hire, it is to make sure that i properly train and manage the workers to be as productive as possible.

apart from IO economists, most economists think that a worker if hired is straight away as good as the worker fired.

i am not sure how it is in USA but in UK i have never seen a boss being afraid to fire anyone. if the worker cannot make himself useful the worker is fired straight away, especially if the worker has little human capital (knowing where the forks are kept for example). it is a very unemotional calculation. i have never seen any examples of 'loyalty' or emotional attachment. the boss and the worker do not have any illusions about their relationship.

Bryan Willman writes:

In addition to all of the above (sometimes including two conflicting views) I've seen one other choice bit.

Management time and attention and energy is, like everybody else, finite.

So when things are going well, firing Jo Loser may not be the highest return thing - meaning highest return on management attention. When the economy sucks, it becomes high return.

And ANY employee requires some part of the management bandwidth of the organization. So one less person is one less paycheck to cover, one less head to manage, one less thing to "worry about". Machines mostly don't have this problem.

So you avoid hiring because it's costly and may not be the biggest win. And you avoid firing because it's painful, costly, and isn't the biggest win until a crises or the employee REALLY screws up.

And so it will be, because people rather than robots own and run firms.

(Does the phrase "transaction costs" apply to this? "Management attention budget"?)

Mark V Anderson writes:

Tom West --

You make some really good points. Making money isn't everything. Enjoying the workplace is important to the boss too, and firing someone is always a downer. And a strong case could be made that the loyalty received by holding back on firing inefficient workers could be made up by benefits of loyalty from the employees.

On the other hand, in my experience firing aversion doesn't match up very well to this theoretical worker paradise of loyalty flowing both ways. It is the deadwood worker who usually benefits from this loyalty, and the dedicated hard working employee who pays for it. And quite often the hardworking employee isn't real pleased by the whole situation. Quite often there is a lot of underground distress going on in these firms, which doesn't lend itself to economic efficiency nor workplace happiness.

My vote is that firing aversion is mostly a bad thing.

Daniel Klein writes:

Firing aversion is an idea that fits Smith's enumeration of "the Order in which Individuals are recommended by Nature to our care and attention," or what is called by some his social-distance theory of sympathy, and by Levy and Peart the sympathetic gradient:

"Among well–disposed people, the necessity or conveniency of mutual accommodation, very frequently produces a friendship not unlike that which takes place among those who are born to live in the same family. Colleagues in office, partners in trade, call one another brothers; and frequently feel towards one another as if they really were so. Their good agreement is an advantage to all; and, if they are tolerably reasonable people, they are naturally disposed to agree. We expect that they should do so; and their disagreement is a sort of a small scandal. The Romans expressed this sort of attachment by the word necessitudo, which, from the etymology, seems to denote that it was imposed by the necessity of the situation."

That is from TMS pp. 223-24.

Jim Glass writes:

In the NFL the head coach could certainly enjoy cutting players from the team by delivering the message, "You're gone! Hand in your playbook", himself personally, if he wants to.

But invariably the job is delegated to a low-level coach, known as "the Turk", to perform.

It seems the Head Coach and other top coaches are sufficiently averse to the job to use their rank to make some low-level guy do it, in the typical case.

And this is where "firing" players is an absolutely basic part of the job -- in every training camp the number of players starts high but has to be cut to 53 -- so it reflects no unpleasant admission on the part of the coach as to "hiring the wrong people", or the business under-performing, or any failure like that. Apparently, they just don't like to do it.

Alan writes:

One major effect I saw of firing aversion was managers moving people to other less-critical positions. Results? Convincing others managment was clueless or spineless, having people do work that wasn't needed (burning $$$), or someone (in or out of the company) better suited not getting the less critical position. It is true that I generally used normal restructuring to clean those situations up as I progressed in responsibility, but I think it was an aversion to confrontation more than human kindness that led my predecessors to shuffle instead of clean.

Dave writes:

I worked in management at a Fortune 500 company whose name you all would recognize.

At the lower-management level, termination was psychologically difficult for a manager because our company had a well-cultivated reputation as a "family." So when an incompetent/incongruous employee came up for termination, he'd have that thought on his face as you told him it was time to go. You felt like a complete hypocrite, disowning your kin.

At the upper-management level, we all learned that the "family" stuff was just a PR ruse. It always amused me, in a dark way, to see it dawn on a manager that not only had he been deluded all his career, he was now expected to act as if that delusion had never occurred, and go out and treat subordinates (and peers) as if he worked in any other cutthroat business environment.

Harun writes:

When you fire enough people you learn its not so bad, I suspect. (In my company, my partner does this, and I think she almost hires and fires too much, but maybe testing employees is a good idea. Note this is occurring in Asia, where there is less legal stuff to worry about.)

We also had situations where our employees came and told us to fire a new hire - we wanted to give her more time, but they said she was hopeless. That was an easier fire to do than most.

Then you have the employee who managed to lose a large customer and cause other havoc - no feelings of guilt with that one!

JMW writes:

Interesting points. But really none have been applicable in my personal experience (of note, managing a global unit with 350 staff).

Most managers are not keen on firing employees. But the reasons are normally different than you noted. And naming the puppy has never been a factor.

1. Cost. Cost of replacing employees is high. Recruiting, training, etc., is high in time and money for new staff. Severance for terminated employee. The fear factor is more in trying to make the best hiring decision in the first place. Then doing your best to put the new employee in the best situation to succeed.

2. Lawsuits. The potential costs of firing staff is always a concern.

3. Employee morale. Other employees often become worried for their own jobs when another staff member is fired. That can impair overall performance in a company or unit.

4. Better the devil you know. The cost and time required to replace, potential suits, overall corporate happiness may lead managers to conclude that keeping an underperforming employee is better than the unknown future. And there are no guarantees that the replacement will be an improvement.

5. Having to fire someone reflects poorly on the manager - leadership, training, hiring practices, etc.

Those are concerns most managers think about.

Now that said, there are equal arguments in the other direction.

Bad employees cost the company, especially in the long run. If a change has to be made, better to do it early rather than create even more issues and costs.

Also, if one person underperforms, that leads to unhappiness from others who must pick up the slack. In the longer term, unhappy competent staff either resent their extra work and leave for greener pastures. Or their performance also tapers off as they see the offenders go unpunished

In my experience, we want our managers to work to hire the best staff possible. Education is not high on the list. Experience and actual testing of competence are more important. Also, assessing personality - fit into organization, drive, enthusiasm, etc. - is crucial in predicting success.

If we have underperformers, we want to work with them to try and improve. But if it is determined they cannot be brought up to speed, they need to be fired. The long term cost to the company, affect on other employees, etc., is too high not to make the change. And we want to be clear to other employees that this is a one-time event and not the prelude to other changes.

For additional wealth and business commentary, please visit www.personalwm.com.

leesus writes:

I deal with this on a daily basis. We are specifically instructed to look at a candidates growth path for at least the next two levels. So, unless i can see your growth to team lead and beyond.... "NO HIRE!"

And as a result, we pass on many of the most amazing developers i have ever met. Our "up or out" policy means that hiring the best developers will never happen.

in the end, we don't hire developers... we hire future leads and managers. Unfortunately these are not overlapping skill sets, and invariably leads to hiring subpar developers.

Glen S. McGhee writes:

Sometimes, employers find a way to ease their guilt feelings -- by progressively increasing salaries, until they are ridiculously high, and then they fire. This was done when a consultant was brought in to implement new software and training of line staff. Then they were let go.

The other piece was "mobbing" (wiki is good on this) -- thwarting the individual and frustrating them in order to produce unacceptable behaviors, for which the worker was held responsible and then fired. This is often called cubicling.

An especially egregious example in Florida's community colleges is "Max and Me" by Marion Brady.
http://freepages.family.rootsweb.ancestry.com/~bradytrilogy/marion-brady/max/max2.html

Comments for this entry have been closed
Return to top