Many companies that come out on top offer some pretty fancy, even creepy, benefits

LAST WEEK, Fortunemagazine published its latest survey of the 100 best companies to work for in the US.

As I have never worked at any of them I can't offer first-hand corroboration of the rankings – though in 1981 I did try to get a job at Boston Consulting Group (rated number two in the Fortunelist), but was rejected after letting slip in an interview that I had no idea what a learning curve was.

In an announcement posted on the BCG website, the consultancy explains how it comes to be the second-best company to work for in the whole country. “An emphasis on employee development, a collaborative culture, extensive training, high-impact client work, progressive benefits, and a commitment to social-responsibility initiatives contributed to the strong performance,” it says.

This doesn’t sound quite right. I’m not sure what “high-impact client work” is, so don’t understand why employee happiness stems from it. I am sure what “extensive training” is – and am equally sure that it is quite low down the list of what thrusting professionals are really after. As for “social-responsibility initiatives”, I’ve never met a strategy consultant who could honestly say that such things were what kept them in the job.

The magazine states that its ranking is “the most extensive employee survey in corporate America”. Indeed, to compile it, an awful lot of questionnaires appear to have been sent out, requiring the employees to complete a “trust index survey” and employers to conduct a “culture audit”.

Many of the companies that come out top of this exercise turn out to offer some pretty fancy benefits. One company has a clinic that gives staff Botox in their lunch break. Another teaches its employees Zumba, “a Latin-style dance exercise”. A third offers on-site pet care facilities and a fourth makes its corporate jet available to employees when there is a death in the family. A fifth gives members of staff having trouble getting pregnant $25,000 towards IVF treatment.

Some of these benefits are more alluring than others. If a relative of mine died, I’m not sure that I’d be in the mood to enjoy the corporate jet. And I’m so allergic to the thought of a Latin dance exercise, I’d pay my employer to avoid having to take part.

But even if these perks were all things I wanted, they would make no difference to my assessment of whether my employer was a good one. Such benefits, even the nice ones, are all pretty creepy. There ought to exist a boundary between employer and employee – between home and work – and Botox and IVF are a long way on the wrong side of it.

The problem with Fortune's rankings – and with all similar ones – is not that the exercise is a daft one. Actually, it is quite worthwhile. It is helpful both to prospective employees and to managers. The problem is that it makes an extremely simple thing seem fantastically complicated.

We all know what distinguishes a good employer from a bad one. A good one provides four basic things. First, it makes sure that everyone has a proper job to do. Second, it pays them fairly. Third, it makes employees feel that their efforts are recognised. And fourth, it gives them nice people to work with. That’s all: there is nothing else.

Fortunately, there is an easy way to measure whether a company is succeeding at these things. It doesn’t involve answering tiresome questions on long feedback forms. It does not require any examination of benefits or of corporate social responsibility policies. There is nothing subjective about the test at all.

It simply measures how long people stay with a company. This is the only consideration that matters. Anyone who is not happy with their job will eventually go somewhere else. If most people stay put for a long time, the company automatically proves itself to be a good place to work.

I suggest two modifications to this test. The first is to ignore departures in the first two years of service – sometimes people are hired who should not have been, and some people have ants in their pants and so move on quickly, however decent their employer.

The other modification is to allow only for voluntary departures. Any company that responds to a dip in its business by firing its workers is not a good place to work. It therefore does not belong in any ranking – no matter how much Botox or Zumba it supplies to employees who are still on the payroll.