Arnold Kling  

Some Observations About Firing

PRINT
Why They Haven't Been Fired... DeLong and Summers...

Bryan poses the issue of why bosses are reluctant to fire unproductive employees. My observations:

1. Firing involves a confrontation. You tend to want to avoid confrontations. In large organizations, the confrontation is indirect. The boss tells HR, and HR sends an angel of death to give the employee the news.

2. Much of the cost of a bad employee is borne by his co-workers. The boss may bear less of the cost. Remember that many workers are Garett Jones workers, operating in teams. Even if one worker does very little, the team may produce the desired increment of organizational capital--a new reporting system, a new marketing campaign, or what have you. Note that the relevant choice is "retain or fire" rather than "raise pay or lower pay." With Garett Jones workers, the value of a worker to a firm tends to be either well above the cost of compensation or close to zero. The concept of equating wage to marginal product does not apply.

3. Don't assume that co-workers know who is unproductive. I remember when employee X thought that I was unwisely retaining employee Y, who was weak technically, and that I failed to appreciate employee Z, who was technically brilliant. Then I took a different job and employee X was promoted to my job. What he learned was that employee Z was completely undisciplined and it was a good thing that employee Y was around to make sure projects were completed on time. Employee X told me that I had been right all along.

All this said, I think that firing low-quality workers is a tremendous way to boost morale in an organization. It sends a message that you care about what the team does and that you are paying attention to who is helping and who is not. It increases workers' sense that they are getting what they deserve.

I think that it is a mistake for an organization to cultivate a belief that bad workers can be rehabilitated. I am sure that sometimes rehabilitation is possible, but on average the costs of trying far exceed the benefits. Middle managers have much better things to do with their time than spend it attempting to rehabilitate a weak employee. If I were a high-level executive, I would try to create a culture of firing as opposed to a culture of rehabilitation. I would rather see someone occasionally fired who might have worked out if given a chance than see lots of effort going into retaining employees who are not going to help.


Comments and Sharing





COMMENTS (36 to date)
daryl jensen writes:

I agree with your comments that more unproductive employees should be fired. Sometimes that is what it takes to show them they need to shape up. I have been fired from a couple different jobs. I would like to think that I have learned from those mistakes, making me more productive in my current position. Each time it was for the better, because I was eventually able to land my job, with less stress, while making more money to boot. People need to realize that getting fired is not the end of the world. Dust yourself off, get back out there, and improve your skills.

ajb writes:

I think Bryan and the commenters don't often distinguish clearly enough between workers whose pay is above their marginal product vs. those who are "bad" (impose negative externalities on the others). The former might simply indicate quasi rents -- the result of implicit contracting that is compensation for loyalty or earlier periods in which pay was below that workers' marginal product. Not firing this group is often efficient. The latter "bad" workers are the real puzzle for theories of firing.

Randy writes:

I've worked in organizations that cultivate a belief in rehabilitation (aka development), and in organizations that have a culture of firing. There are problems with a culture of rehabilitation, but there are also problems with a culture of firing, and in my opinion the problems of the latter are the more severe, and often underestimated by management. Specifically, management often underestimates the negative impact on morale of those who are not fired, but who are witness to a steady stream of friends and co-workers escorted out of the building, often for seemingly random or minor infractions. In my experience, a culture of firing results in a culture of workers with negative attitudes, the best of whom will move on at the earliest possible opportunity.

Jeff writes:

I think getting rid of the least productive in the context of a general round of layoffs is the way to go. I was at an internet company that had 5 rounds of layoffs as the bubble was popping about 10-12 years ago... after every round I was like, wow, I like how the team looks now... the lurkers have been purged.

It softens the blow from the standpoint of being confrontational, or the sense of having people singled out, et.

Of course the company needs to be in a state where a round of layoffs makes sense.

Kayla writes:

I agree with Randy. Even if employees are very good workers, their morale may be lowered by a fear of being fired, because good workers may not trust their bosses to recognize their merit.

John Fembup writes:

"firing low-quality workers is a tremendous way to boost morale in an organization"

This is absolutely true. The higher rank the low-quality worker has, the greater the morale boost.

I would add that your term "low quality" is a rather more subjective (and understandable) test than whether an employee's true production is 25% below his cost of employment. It depends on a good manager's good judgement - which I think reinforces my comment above.

Zlati Petroff writes:

I'd again bring up the point that we need to define a measure of productivity.

A worker's median daily productivity may be very low, yet his long-term average productivity may be sky-high if every once in a while he cooks up a big idea that leads to a tidy profit.

This is very relevant to knowledge industries, like tech, scientific research, pharmaceuticals, and marketing.

On a daily basis we may observe low median productivity, yet a manager may choose to wait, hoping that in time the worker will come up with something big.

Paul writes:

"The boss tells HR, and HR sends an angel of death to give the employee the news." This has not been my experience at all. I have always fired people personally, although with an HR person there as a witness. And the time I was fired was just the same.

But when I was suddenly (at 10 minutes notice) made CEO of a company where I had previously worked, everyone thought same-old same-old until I fired the two vice-presidents who were really screwing things up. It totally changed morale and turned the company around.

Jim writes:

For teams with common goals (what other teams are there?) I generally disagree with #3. Co-workers in well functioning teams know exactly how their team mates contribute, and therefore should have acknowledged and formal input on who gets hired and fired from their team.

Providing that autonomy to a team has social, motivational and other benefits.

R Richard Schweitzer writes:

There are differences between managerial and proprietary relations in matters of employment.
Some are paternalistic, but there are others.

There is also the "Yale" syndrome at some management levels. Just as a school will not impugn its admissions process by "allowing" failure or inadequacies (and thus seek other means of dissociation), the employee selection process can also lead to a sense of "not making bad choices" on the part of the selector (more keenly in the proprietor, but also in the manager).


RGV writes:

"All this said, I think that firing low-quality workers is a tremendous way to boost morale in an organization. It sends a message that you care about what the team does and that you are paying attention to who is helping and who is not. It increases workers' sense that they are getting what they deserve."

Strange. While I agree with you, I've heard several directors explicitly claim that firing the worst workers lowers the morale of the team as a whole. That is why they prefer group layoffs to individual firings. What do you think is going on there?

RGV writes:

I should add that most newer internet companies (Google, Facebook etc.) are moving towards a peer based review. Flat hierarchies with most promotion decisions coming from peer inputs. If you are right on 3), then this model should fail spectacularly.

unproductive writes:

"Firing involves a confrontation. You tend to want to avoid confrontations. In large organizations, the confrontation is indirect. The boss tells HR, and HR sends an angel of death to give the employee the news"

I have never seen this. In my experience HR always insist the the boss does the firing.

HR will coach the boos how to do it, and HR will be in the room for support, as a witness, and in case things get nasty, or something unexpected happens, but the boss does the talking.

Shangwen writes:

It's hard to have a discussion of the psychological dynamics of firing (and discipline) without discussing how people get hired. Hiring is a process fraught with weak information. Unless you spend thousands per candidate on testing, evaluation, and background checks, you are taking on a lot of risk. You also have to start most people on the job with a lot of guidance and check-ins, otherwise they get lost and you miss early signals.

Making that initial investment is costly and often infeasible. So when it comes to your attention that the person isn't working out and can't be turned around, you feel like a dupe and just want to bury the problem. One of the reasons most managers hate firing people is because they see their own failures of selection and supervision in the person's failure as an employee.

Costard writes:

In a business with quotas it is easy to know who is unproductive. But in a company providing services, or one employing specialists (like software engineers) whose product it is difficult to evaluate without employing outside expertise, it may not be easy to sort productive from unproductive. Personality conflicts, on the other hand, are apparent. Firing someone who is unproductive benefits the company; but firing someone the boss dislikes, benefits the boss. Which is likely to be a stronger tendency?

The problem of mismanagement further complicates this. Lower management specifically seems vulnerable to the sort of intra-organizational hires that turn good workers into mediocre bosses, and expands cliques and prejudices up the ladder. To what extent should these folks have free reign to hire and fire, when there are fixed costs like training and severance that the company, not the boss, must pay?

IMO it is probably too expensive, and management's judgment is rarely good enough, to justify a culture of easy-come, easy-go.

yet another david writes:
I think that it is a mistake for an organization to cultivate a belief that bad workers can be rehabilitated. I am sure that sometimes rehabilitation is possible, but on average the costs of trying far exceed the benefits. Middle managers have much better things to do with their time than spend it attempting to rehabilitate a weak employee.

Strongly agree. In addition to the reasons given, it tends to create an egalitarian culture which drives the best away in disgust. One is left with mediocrity. Plus, there's the making-a-silk-purse-out-of-a-sow's-ear problem (it's not simply hard - it's usually impossible).

Much of the cost of a bad employee is borne by his co-workers. The boss may bear less of the cost.

So true but also so under-appreciated.

Rohan writes:

Wasn't there a major company (GE?) that used to have the policy that the bottom 10% were fired every year?

Wouldn't it be easier to look at the data about how that worked there and in companies with similar policies? You know, actual data rather than anecdotes.

bkj writes:

I most certainly agree that the costs of rehabilitating unproductive employees far outweigh the benefits. Not only is rehabilitation costly, but a weak employee may not improve after being rehabilitated or may even be resentful at the fact that rehabilitation is needed, therefore causing the individual in question to be less willing to comply with the process. I also agree that firing unproductive workers who have a noticeable disparity in work output compared to other productive workers may boost overall employee morale, although I do believe it depends on the specific worker in question. For instance, an affable unproductive worker who's part of a large social circle and has many friends and acquaintances within the organization may induce unhappiness upon being fired thereby reducing employee morale.

Paul writes:

These days everybody is in a "protected class" of one sort or another. The probability and expense of a discrimination lawsuit is higher. And senior management will blame the manager, regardless of how justified the firing was. That's why Jack Welch instituted the 10% firing rule at GE, so managers would be free to fire without incurring the wrath of their superiors.

Glen Smith writes:

RGV,

I think that is because what my boss may consider me a low-quality worker is not necessarily what my team thinks. Also, unless there is an obvious reason a person got fired, firing is a good indicator that the rest of the team needs to polish up their resumes because the company is not doing well.

Randy writes:

I notice that many here are talking about whether or not to fire or rehabilitate "unproductive" workers. But in a "culture of firing", being unproductive is not the only reason, and not the primary reason, that people are fired. Managers are people too, and a very great many of them are inexperienced or wrongly experienced, and frequently display all manner of generally unacceptable behaviors - that is, they are unproductive managers. In a "culture of firing" the primary reason that workers get fired is unproductive management.

Glen Smith writes:

I've also noticed that I am more likely to have sympathy for a hard-working but unproductive co-worker than a lazy but productive co-worker. As a software engineer, I think that is related to the fact that I always got praised by management for production rather than work.

Ben T writes:

"Rehabilitation, Retrain, avoid confrontations" are ways for executive to avoid facing their employees. I couldn't agree more. Referred to what Rohan say about GE during Jack Welch era, the company "rose 4000%" because he cut 1% of the bottom-line every year. If only company can apply some of Mr. Welch tactic then many company have loyal employees and higher ROI.

http://en.wikipedia.org/wiki/Jack_Welch

Norm writes:

I have often wondered if any company collects the low producing employees into one group/division and then sells the division, thus avoiding the lawsuits etc.
My first supervisory job was of a group which had been formed this way, not for selling, but because management thought, correctly for several years, that competence did not matter much in this function and firing was heavily discouraged. Then the business changed and this group was seriously limiting success. What a mess. Eventually I discovered that one of the bachelor level employees could do the work of four PhDs. I could only fire one of the PhDs though.

Flis Lister writes:

"I think that it is a mistake for an organization to cultivate a belief that bad workers can be rehabilitated. I am sure that sometimes rehabilitation is possible, but on average the costs of trying far exceed the benefits"

This reminds me of how my Dad's company(apparently) operates:

  • it is a manufacturing company, so there is a distinction between the hourlies (often unionised) and staff. The "best" hourlies are promoted to staff

  • once you become staff, you probably won't be laid off without good reasons. It has never laid off staff because of a downturn. Other companies in this industry, however, load up on people in the upturn and lay the worst ones off in the downturn [it is a very cyclical industry]

  • when a staff member is struggling, it is not due to a lack of work ethic (they made sure of that before making them staff); it is instead because of a lack of skill/experience/whatever. So, the company surrounds that person with competent people, so that the person can learn and improve.

  • The struggling worker sees this happening. After some time, the worker becomes more competent and also much more loyal- the company stuck with him, so he'll probably stick with the company

My Dad's company is the most respected companies in its industry.

So I would dispute your point: it really depends on why the worker is "bad".

Dan Hanson writes:

A clarification on GE policy: GE doesn't automatically fire the bottom 10%. I recognizes the bottom 10% as those who are not providing value to the company, and then gives them the opportunity to engage in a performance improvement plan. Only if they refuse a PIP are they terminated. And if they carry out the Pip satisfactorily, they must not land in the bottom 10% again in the next year. It seems like a good compromise.

As for why companies don't fire more ZMP workers - I think the answer is partially because it's really hard to tell who they are. Even the bottom 10% may not be ZMP - they're just the tail of the bell curve. Whether they are truly non-productive is another issue entirely.

How do you rate an employee who gets all his work done - but leans heavily on others? You know the type - the person who, when confronted with a problem instinctively goes to the person in the next cube or office for advice, or who asks some to just 'help for a second' and then winds up getting the other person to do half the work for him. Such people can be very hard to identify.

Then there are workers who don't get their own work done - but it's usually because they've spent so much time helping other people. They're great team players and amplify the value of the entire team, but their personal stats may be terrible.

There are also people who are non-productive because they are put in roles they are not suited for, but who thrive in different roles.

Then there are the group dynamics - put a person under a domineering co-worker or boss, and that person may be afraid to step out and take chances or be creative and just work to rule. Give that same person a little bit of autonomy and a supportive work group, and suddenly that person blossoms into a creative powerhouse.

There are also people who wind up having poor stats because they are the 'go-to' people when you have really difficult problems to solve. Difficult problems means high failure rates, but is that the fault of the employee?

In software engineering it can be very difficult to develop metrics to measure employees because of the disparity in difficulty of assignment - the guru who has been tasked with debugging 10,000 lines of microkernal code may scratch his head for weeks trying to do it, and may not even succeed. If he doesn't, management can't tell if he was tearing his hear out in deep thought or surfing the web.

In the meantime, the plodder with no creative ability is cranking out thousands of lines of mundane, low value code in the cube next to him. Which one is more productive?

There's also the general confusion of large firms - companies where employees are spread across multiple offices, with many managers in the mix, and it starts to become really difficult to figure out what's going wrong when teams start to lose effectiveness.

Finally, in my experience the real productivity gains in my field (software engineering) don't come from dumping the low-paid but under-performing coders at the bottom of the ladder. The real productivity gains come when the bad employees in management or high technical levels are fired or quit. They are the ones who derail multi-million dollar projects with stupid ideas, or who infect entire teams with paranoia or apathy because of their poor people skills.

They're the ones who turn 100 very good programmers into ZMP workers because they put them to the task of building software no one wants, with an architecture that will not survive first contact with a real customer. Unfortunately, they're also the ones who have managed to climb the ladder to relatively high station, and the people who put them there have a vested interest in making that look like a good decision.

Saturos writes:
If I were a high-level executive, I would try to create a culture of firing as opposed to a culture of rehabilitation.

You sound like a cliche evil capitalist. Your employees must have been terrified of you.

Essen writes:

I fired some and got fired as well...from the same Company.
Reminds you of the French Revolution?
Not sure who reaped the ROI: the shareholders or the stakeholders.
There is a thing called Karma as well.

http://www.beedictionary.com/blog/?s=catalyst

Anna Gurova writes:

Typically one in ten employees in a professional organization is the super performer / brilliant / key employee, who creates value for the employer in excess of 20-40 times compared to average peer coworkers. The other 9 employees play secondary and supportive roles. So that the key employee can delegate less important work onto his average-productive peers, while keeping 100% of his focus on the ultra high yield and mission critical work activities. Through such "resource leveling", this combination balances the work load in such a way that creates the ultimate best profit margins for the employer. Now the reason and the real "intristic value" of keeping the average and unproductive employees becomes fairly obvious. If those mediocre employees are let go, then part of their work load will be pushed onto those high performing individuals, slowing them down and distracting from high-value creating activities.

Skeptical writes:

"The boss tells HR, and HR sends an angel of death to give the employee the news." This has not been my experience at all. I have always fired people personally, although with an HR person there as a witness. And the time I was fired was just the same.

If were' talking about ZMP/deadwood/headache employees, the boss wants to get rid of them. In my experience, large corporations make the boss jump through months of HR hoops to actually fire someone.

If the boss has to fire otherwise adequate employees due to the company having financial trouble, yeah, sure they want HR to do it.

All this said, I think that firing low-quality workers is a tremendous way to boost morale in an organization. It sends a message that you care about what the team does and that you are paying attention to who is helping and who is not. It increases workers' sense that they are getting what they deserve.

Sometimes yes.

Or maybe what work was done by five people has to be done by four or three. The deadwood might have been half as productive, but at least it didn't get dumped on the other employees.

David Zetland writes:

Add laws to the psychological barriers and you get Italy, where those who try to change the laws get assassinated: http://www.economist.com/node/21551046

unproductive writes:

i have worked in lots of large companies.
I think people get fired for three reason, none of which is to do with productivity.

1 - a specific misconduct (theft, racism, sexual misconduct, lying, looking at porn etc)

2 - as part of a corporate RIF (reduction in force) which takes out 100s ie nothing personal, really. Some of the 100s will be the weakest, others are just swept up in it. Eg working in Elbonia when the Elbonian office closes

3 - because they are actively disruptive, so that at a personal level, in the end people just don't like having them around. NB nothing oto do with being unprodcuctive, though may well coincide, but I have seen plently of productive, but also unpleasant people fired, just because in the end no one wants to work with them

Productivity or effectiveness rarely enters it.

Chris writes:
I think that firing low-quality workers is a tremendous way to boost morale in an organization.

Low-quality workers are not necessarily bad people and are usually friends with other workers.

Obviously, it does not boost morale to see your friends fired.


Some guy writes:

I am the boss for the last few years (big corp). My employees on the whole are bright and productive. Some are mediocre. Some of the mediocre ones are prone to stir the pot. They will be the first to go if I ever have to do layoffs.

HR requires such a crazy process that it isn't worth pushing it.

Bryan Willman writes:

As I commented on Caplan's post - firing is costly in terms of management time and other resources - is firing Pat loser the best use of management resource this week/month/year?

Peter writes:

[Comment removed pending confirmation of email address and for crude language. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top