At the macro level, yes, the size of the company's payroll is a cost. But very few members of large companies have any incentive to bear that cost. The CEO, the Board, and that's about it. Almost nobody below the C-level has any direct incentive to keep payroll costs manageable.
At the management level, the manager -- who bears no personal cost for managing N+1 workers instead of just N workers, will almost always choose N+1 if given the option. He doesn't pay their salaries, after all, and to him, the increased headcount is a status symbol. It's also something he will convince himself he actually needs. (You never hear middle managers complaining that their divisons are overstaffed, but the opposite complaint is almost universal).
This is what's known as an "agency problem." Many (most?) of the agents of the greater whole (the company) have personal incentives that work at odds with the company's greater incentives. This clash of incentives leads to waste, bloat, inefficiency, and so forth, because almost nobody is personally on the hook for the company's total health in the long run. (Sure, they're indirectly on the hook. If the company starts doing poorly, they could risk losing their jobs. But people tend to externalize failure, and don't hold themselves personally responsible).
> At the management level, the manager -- who bears no personal cost for managing N+1 workers instead of just N workers, will almost always choose N+1 if given the option.
I do not know if you are familiar with how big companies work, but usually when you reach a large enough size, such companies start to track "productivity indexes" between their departments and against competition, when comparison is available. SUch an index would look like = sales of the department / headcount of that department, which basically gives you an "average value" of an employee in that department. Then in order to prove that you need additional headcounts, you need to have a high productivity index in the first place to justify it. So that's why big companies don't just keep growing forever: they start to become more efficiency-sensitive, and consider carefully the cost of an employee versus the actual benefit to have more.
That is how it should work in theory, but often that's not how it really works. Obtaining the actual value of a department or a single employee can be extremely hard to quantify. It can also be hard to figure out when a project needs the plug pulled.
Personal anecdote. A company I worked for fired the more costly tech support staff right before an important partner product launch which left a bunch of undertrained customer service reps supporting the new product which gave a bad experience to customers. It also irritated the partner because the training the reps received essentially told them that almost any problem needed to be referred to the partner, swamping them with customers wanting them to fix a problem that wasn't theirs. They then had to scramble & rehire a tech support staff.
Take Intel's Itanium project which has struggled on for 16 years, yet Intel still has resources devoted to it. Sometimes you cannot just kill a very unsuccessful project & it can take many many years to wind it down.
You could also look at any need for layoffs as being a miscalculation by the company. When Yahoo announces 2000 layoffs, does that mean they're being efficient by cutting staff or does it mean that they've had 2000 employees on staff that shouldn't have been there in the first place & for how long? Why didn't the indexes and metrics in place tell them to not hire these people?
Also the government is not immune to layoffs. They have actually been one of the top organizations laying people off over the last few years.
At the management level, the manager -- who bears no personal cost for managing N+1 workers instead of just N workers, will almost always choose N+1 if given the option. He doesn't pay their salaries, after all, and to him, the increased headcount is a status symbol. It's also something he will convince himself he actually needs. (You never hear middle managers complaining that their divisons are overstaffed, but the opposite complaint is almost universal).
This is what's known as an "agency problem." Many (most?) of the agents of the greater whole (the company) have personal incentives that work at odds with the company's greater incentives. This clash of incentives leads to waste, bloat, inefficiency, and so forth, because almost nobody is personally on the hook for the company's total health in the long run. (Sure, they're indirectly on the hook. If the company starts doing poorly, they could risk losing their jobs. But people tend to externalize failure, and don't hold themselves personally responsible).