Martin Colebourne

Unlimited vacation is a perk that has been making waves in tech and startup circles. At first glance it sounds great: if we are interested in results, rather than effort, why do we ask employees to work set office hours and only allow them a set amount of vacation? Why not assign them their tasks and let them decide how to achieve them, taking time off as they wish, providing they are still meeting their targets?

Whilst some people see this as a fantastic idea and it is being touted as a great new perk to attract top talent; others see it as a terrible attack on the rights of employees. How is it that the same policy can be seen as both hugely positive and hugely negative?

An illuminating way to think about unlimited vacation is as an example of the classic Prisoner’s Dilemma. The concept of the Prisoner’s Dilemma is commonly discussed in Game Theory and Economics. It is usually formulated along these lines:

Two people are apprehended following a crime, taken to the police station and interrogated separately. They are unable to communicate with each other and do not have time to agree a story.

• If they both remain silent, there will not be sufficient evidence to convict them of the main crime. They will be found guilty of a lesser crime and each receive 3 years in jail.
• If one of them ‘defects’ by giving evidence against the other, then they will receive a reduced sentence of 1 year, while their colleague will get the ‘sucker’s pay-off’ — a full conviction, with 7 years in prison.
• If they both give evidence, they will both receive around 5 years in jail.

The best overall outcome occurs if they both stay silent, but there is a clear incentive for them to defect, in the hope of receiving a lower sentence or just to avoid the risk of receiving the sucker’s payoff.

Rational decision making models tell us that defection is the only logical choice. Since there is no reason for them to trust each other, they have to accept a less then ideal outcome.

The problem for economics is that this is not what most people do when faced with a Prisoner’s dilemma (laboratory tests are usually done using sums of money, rather than actual incarceration). People are far more willing to try to co-operate than they should be.

This could be taken as a sign of how illogical people are. However, this would be a mistake, because their willingness to co-operate often leads to them getting a better outcome than theory predicts they should be able to. Clearly, something is wrong with the theory.

The trick is to realise that we don’t usually play such games only once. In life, we face situations similar to the Prisoner’s Dilemma all the time, but they are repeated time and again, often with the same people.

In an environment in which you play the game repeatedly, provided that you can afford to lose some of the time, you can build up knowledge of how other members of the group play the game — do they always co-operate, or are they always looking out for an opportunity to take advantage. You can also build up a picture of the culture of the group itself. If the group culture is cooperative, and cheats are actively discouraged then you can afford to rely on cooperation.

Why the long discussion about the Prisoner’s Dilemma? Because the offer of unlimited vacation fits the requirements for the game. Interestingly though there are two alternative ways to frame the situation.

For some commentators (and probably for many businesses), the concern with unlimited vacation is that employees will abuse it by taking too much time off. In this view, the game is formulated as follows:

The aim for employees is to take more time off.

• If everyone co-operates, then vacation time is shared equitably and everyone benefits through greater flexibility
• However, employees will be tempted to defect by taking an unreasonable amount of time off. When they do this, other employees get the sucker’s payoff by having to pick up the slack.
• The risk for the company is that everyone defects, destroying productivity.

This formulation might apply in situations where employees are low in motivation and autonomy. But then employees have always had the option of slacking (even if they don’t have the option of simply taking extra time off). In response, companies have developed methods for controlling poor behaviour — ultimate power always rests with the employer. In any case, unlimited vacation is far less likely to be offered as a perk in companies like these.

A better formulation is to assume that in companies where it might be offered, the aim for employees is to succeed and get promoted.

• If everyone cooperates, then everyone takes a reasonable amount of vacation and everyone benefits from the increased flexibility.
• However, there is a incentive for ambitious employees to defect by nottaking vacation time. If they do this they can get more done and prove their commitment to the company — leaving those who take holidays looking like slackers. In this formulation, the sucker’s payoff is getting fired because you took the most time off.
• This means there is a danger that everyone has to cut back on time off, ending up in a worse situation that with a standard vacation policy.

Seen like this, what happens when unlimited vacation policies are introduced is dependent on the culture of the company.

If the environment is co-operative and management truly believes in the value of employee welfare and down-time, then everyone can benefit.

Employees who defect will be pressured by colleagues and management to stop — the attempt to gain favour won’t work, because it will be seen in a negative light.

However, in a company with a demanding, high-pressure culture, in which the message is that only the best are hired, and those who aren’t up to scratch will be fired, then the opposite is true.

In a high-pressure culture, management is unlikely to disapprove of defectors who refuse to take vacation — after all, this fits with their work-hard cultural message, meaning that defection pays off. In such an environment, all employees will end up having to minimise their holiday time, since taking vacation looks like weakness and a lack of commitment to the job.

For this reason, an agreed vacation allowance can be seen as a form of protection — a minimum allocation of rest that employees can take without it being used against them. (Even when this is available, many people don’t take all of their vacation, because they feel that they are too busy, or that it will look bad.)

So is unlimited vacation a good thing or a bad thing?

There may well be some companies approaching the perk with the right mindset. However, many come from the high-pressure world of Silicon Valley. It seems unlikely that they are offering the perk because they think it will lead to a happier, healthier workforce. Instead it is seen as a useful recruiting tool, an aid to them hiring the brightest and best. An unlimited vacation policy can also help to reduce costs: by removing official vacation allowance, companies can avoid paying employees for unused vacation time when they leave.

These are clear risk factors and suggest that, in many cases, the result of an unlimited vacation policy will be negative for employees. The key question for anyone considering taking a job with this perk is: what is the culture of the company? This will determine if they are actually being offered more flexibility and vacation time, or less.