The limits of “unlimited” vacation
Obligatory Disclaimer: this post discusses unlimited vacation policies. My employer (Heroku) has one such policy. However, this post isn’t really specifically about Heroku’s policy; it’s about the concept in general, not any specific implementation. As is always the case, on this site I speak for myself, not my company or anyone else.
Many companies, especially tech companies, have adopted “unlimited” vacation policies — instead of a set number of days of paid time off (PTO), employees are told something along the lines “take as much time as you need”. For example, here’s what Netflix CEO Reed Hastings on vacation, from his famous and influential slide deck on Netflix’s employment culture:
(Don’t miss the next slide, which hints at some of problems around unlimited vacations, and their solutions…)
On the face of it, this policy sounds awesome: take vacation whenever! As much as you want! Never mess with cruddy time off request or PTO tracking systems! However, is it really that great? Surely “unlimited” doesn’t really mean totally unlimited. Right?
Beyond the question of where the limits actually lie, what especially concerns me about unlimited vacation policies is the potential for inequity and abuse. There’s already enough inequity around pay, and pay is fairly strictly defined compared to the wild-west of unlimited vacation. Is it possible that unlimited vacation might do more harm than good?
This post represents my attempt at thinking through this problem, enumerating the pros and cons of unlimited time off policies, and seeking to come to a conclusion about an equitable unlimited time off system.
Pros
There are, obviously, some very compelling reasons for an unlimited vacation policy:
As mentioned above, on the surface, it sounds pretty fantastic. If implemented fairly, unlimited time off is a pretty great perq, letting staff take as much time as they need to ensure good work-life balance.
Time off can be used for vacations, personal time, sick time – whatever. This is in stark contrast to the trend of lumping both sick time and time off into a single limited “PTO” bucket. Having an unlimited budget of time off means you don’t have to worry about “banking” time in case you get sick. If unseen opportunities or obligations come up – invited to speak abroad? Have a relative or friend get sick? Just get burnt out and need to regroup? – you can always take those opportunities.
It seems fair: everyone gets the same policy, eliminating unfairness and resentment. For example, I’ve always valued time off, so I’ve nearly always negotiated a lower salary in exchange for a bigger PTO pool. In the past this has been a sticking point; my colleagues would feel resentful of the extra time I took, and I’d feel guilty about taking it. An unlimited policy eliminates this resentment (maybe… see below).
An unlimited policy seems like it would trend towards more time off in general, leading more healthy Europe-like averages (4-6 weeks per year) than the pitiful 2 weeks or so that’s typical of most US tech companies. There’s actually no federally-mandated PTO in the US, so it’s probably not a surprise that companies tend to take advantage of the lack of regulation and offer unhealthily low amounts of time off.
It gives agency to the employees. This is a really big one. As Alfie Kohn argues in Punished by Rewards:
[T]o describe much of what is wrong with our workplaces is to enumerate the effects of restricting people’s sense of self-determination (p. 193).
Time off is something employers use to punish and reward employees, and thus comes down to essentially another controlling factor. Giving that power back to the employees, allowing that level of self-determination, is powerful:
[P]eople are most motivated when they are able to participate in making decisions about organizational goals (p. 194).
Keeping the agency around things like time off in the hands of individual employees goes a long way towards helping staff feel motivated, committed, and, yes, happy at their jobs.
Cons
Of course, there’s a dark side, too; there are dangers lurking in unlimited time off:
Accrued time off can be a forcing function: some people, left to their own devices, won’t actually take enough (or any!) time off. This can mean serious burnout. For some people, the forcing function of having time off “expire” at the end of the year can force a needed vacation that wouldn’t be taken otherwise.
Along those lines, a big trip, a long vacation, can feel easier to justify to your manager if you’ve been “saving up” for that vacation for a while.
The most overworked people, with the most responsibilities, will be the ones who feel least able to take time off. However, paradoxically, it’s usually these people who may need it most. Time off is a valuable antidote to burnout, so leaving it up to the people who feel least able to take some is dangerous.
“Unlimited” usually also means “untracked”, so it can be really nebulous trying to figure out how much time off is normal or appropriate. This can make asking for time off stressful – you never know when you hit the “too much” point – and it can lead to resentment (“How come Greg keeps taking these week-long vacations? I only took five days off last year!”).
Along the same lines, unlimited vacation can also lead other employees to feel guilty about taking what they feel to be more than their peers, and could thus end up over-working because they don’t know what’s normal. This dovetails especially badly with impostor syndrome: people with impostor syndrome may feel like they “deserve” less vacation than their peers. Since impostor syndrome in technology is especially prevalent among minorities (women, people of color, etc.), this can create self-imposed structural inequalities around vacation.
Most unlimited time off plans have some sort of manager approval step – “unlimited time off as long as it’s cleared with your manager” is pretty typical. However, different managers may have different systems for approving these requests, which can lead to inequity between teams.
Worse, the lack of a tracking system could cover up structural inequities. If a manager is approving time off at a higher rate for men than for women, how would anyone know? Remember: people are creatures of bias: we rank resumes with common male names higher than identical ones with female names [PDF], so this doesn’t even have to be a bigoted manager we’re talking about! In the absence of a centralized policy and tracking, how do we know we’re not falling victim to bias, be it conscious or not?
The interaction between unlimited time off and federal law around extended leave (i.e. the US Family Medical Leave Act) is a minefield. I’m not a lawyer, nor have I had to take advantage of extended leave laws, so I can’t speak much to this other than to say that based on everything I can find, the interaction between unlimited vacation and FMLA is a minefield.
What happens when you leave? With PTO policies, time off gets paid when you leave, but with unlimited vacation there’s no such pay-out. Does unlimited vacation actually end up amounting to wage theft?
An equitable unlimited time off system
So where does this leave us? It’s clear to me that there are a lot of positives about unlimited time off. To me, the matter of agency discussed above is critical: keeping employees as empowered in their own success as possible seems incredibly important, so that alone seems to me to make unlimited vacation a net positive.
However, how do we ensure that we get this positive aspect and avoid the issues articulated above? I think I have a few simple solutions that can make sure that unlimited vacation is an equitable system:
First, decouple “unlimited” from “untracked”. The lack of defined upper limit should not free the company from an obligation to track people’s time off. Tracking time off means that we can analyze it for trends and bias, and ensure that it’s being distributed and managed fairly.
Hold managers accountable for seeing that their staff take fair amounts of time off, both compared to the rest of the team and to the company at large. Don’t allow overworked and stressed employees to avoid time off because they’re “too busy”; in the absence of “use it or lose it”, managers need to replace those forcing functions. Regularly review time off for systemic bias, as part of a equal pay review. (You are reviewing compensation for equity, right?)
Set clear expectations and targets around how much time off is “normal” and “expected”, and hold employees accountable for keeping close to these targets. For example, it’d be great to see a company say something like “most employees take about 4 weeks of vacation, total, each year. We expect you to take on the order of 3-5 weeks, with at least one week-or-longer vacation.” This can also help bring back the forcing function around vacation (one of the biggest positives of a limited PTO system). I like what Evernote reportedly does:
Evernote, which also doesn’t limit employee vacation days, actually gives a $1,000 stipend to anyone who takes an entire week off in order to encourage vacation taking.
Finally, work with HR and legal advisors on a clear, fair, and legal answer to the FMLA question. I’m hand-waving here because I’m not a lawyer and I don’t know what the answer is, but it’s incredibly important that a workplace that wants to implement a fair unlimited time off policy answer this question. Without an answer, an unlimited time off policy simply can’t be equitable. It’s clearly not a coincidence that we see unlimited time off policies at tech companies, where the mostly-young-guy demographics mean that only a small minority of employees run into the need for extended family or medical leave. If we want our workplaces to become more diverse, we need to answer this question in a way that doesn’t make being older, or disabled, or pregnant into a complicated and stressful issue.
Other than the last one, which is a bit sticky and might require some serious work, these don’t seem particularly hard to implement, and they do seem as if they’d solve most of the problems I can identify with unlimited time off.
I’ve been thinking about this for over a year, and my thoughts have been shaped by a lot of people: Leigh Honeywell, Alex Gaynor, Matt Zimmerman, Sharon Schmidt, and Peter van Hardenberg. Thanks for the discussions, notes, feedback, links, and information, y’all!