We love company matches. It may be the closest thing in financial services to a free lunch—but only if you’re the employee, not the employer. For the employer, it’s a very real cost. So before you decide you need to offer a match (or before you decide not to offer a 401(k) because you think you can’t afford to match), let’s clarify how you should think about this decision. You might be quite surprised.

Primary reasons people offer a match

Given that there is a very real cost of matching (on average it’s about 3% of your payroll if you were to adopt one of the common matching formulas and get typical enrollment rates), why do companies offer them? It usually comes down to three reasons:

  1. You want to offer the richest, most competitive and attractive benefits package you can.
  2. It’s necessary to increase participation to avoid being “top-heavy,” or otherwise fail nondiscrimination testing.
  3. It’s required because the company implemented a “Safe Harbor Plan” that automatically passes nondiscrimination tests.

Those are the primary reasons, but should you offer a match? While everyone’s circumstances are different, here are some general guidelines.

Yes, if . . .

. . . you want to offer the best, most competitive benefits package you can. If you’re a large company, you pretty much need to in order to be competitive. About 90% of large companies, meaning 1,000 employees or more, offer a match according to the American Benefits Institute. And a Fidelity study indicated that workers were more likely to accept a position that had a company match. In fact only 13% said that it was likely that they would accept a job without a match. I suspect that this is a case where what people say and what people actually do may be a little different, but it’s pretty clear that a match is important to prospective employees.

As a smaller company, it’s still a good idea if you want to be as competitive as possible. You might be surprised to learn that because these matching contributions are tax-deductible, the true cost is lower than you might think. (In a separate article, we’ll talk about the most common matching formulas and their actual impact on the bottom line.)

But bear in mind that most small companies don’t even offer a 401(k) to begin with, so there is less of an expectation.

No if . . .

. . . you’re a small company and you’re worried about running out of cash in the near future. If that’s you, you’ve got many more things to worry about than a 3% increase in payroll.

Also no if . . .

. . . you’re doing a match – or are planning to – just to increase participation.

There is a much more efficient way to increase participation, and it’s called automatic enrollment.

This is rapidly gaining traction because of just how damn effective it is. We routinely see enrollment rates of 90% AND high employee engagement AND satisfaction – the 401(k) trifecta! When done right, auto-enrollment becomes a super effective way to boost participation rates dramatically.

But, you ask, doesn’t a match also encourage people to join the plan? Yes, but it is not the most effective way. A survey by the International Society of Certified Employee Benefit Specialists says only 35% of surveyed employers believe getting the match is the “primary reason” employees join the plan. (The bigger reason is “personal desire to save for retirement.”)

Maybe, if . . .

. . . you’re concerned about failing your non-discrimination testing, then you may want to set up a Safe Harbor plan. With a Safe Harbor plan, you have no choice but to offer a match.

But in many cases, employers think they need a Safe Harbor plan, when all they need is a solid plan with strong enrollment. We think it makes sense to check with an expert to make sure you really need that Safe Harbor design. That’s because regular plans with automatic enrollment are so effective at increasing participation, many employers are finding that they can pass their non-discrimination testing where they might have otherwise failed.

There certainly are cases where a Safe Harbor plan makes sense, but it’s better to test assumptions first: You can start without a Safe Harbor plan, find out how you fare with the nondiscrimination tests, then switch if you need to.

Or a short discussion with an expert can probably help you sort this whole question out.

We love company matches and so will your employees, but they don’t make sense for everyone. Know yourself, know your goals, and the decision to match will become clearer.