Overtime and Productivity

While we still don’t know if 40 hours is the most optimal amount to work, we know that sometimes we are working more than that. Studies have been done on working overtime in industries similar to the creative digital industry and how that affects productivity.

In my experience, this can be an industry that performs to deadlines, and work can sometimes be pilled up at the end, which can result in working harder, or working more near the end of a project, or sometimes referred to as ‘crunch mode’.

In Evan Robinson’s study, ‘Why Crunch Mode Doesn't Work’, he states that crunch mode actually slows development and causes more bugs than a standard 40 hour work week. (Robinson, 2005)

As Sydney Chapman stated in his 1909 paper on Hours of Labour, as industrial processes became more intense, workers would need more time off to recover so the optimal length of the working day would decline as work processes demanded more mental concentration. (Chapman, 1909)

Sydney Chapman’s Productivity and Factory Work

In the above graph, OX stands for the hours worked in day and OY is the value created and P represents the productivity of the workers.

An example of this theory applied to today’s knowledge workers; a developer produces more code and fewer bugs when fresh and alert. The first hour is usually the worker warming up, or getting into tasks and then the next few hours tend to be the most productive ones. (Robinson, 2005)

Later in the day, as they get tired, they get less done per hour, it could take a much longer time to code a simple task or they may make more mistakes and cause bugs that they would not have earlier in the day.(Robinson, 2005)

The combination of working slower and producing mistakes causes productivity to dip, where it eventually reaches zero or negative productivity (point b in the graph). (Chapman, 1909)

One of the assumptions of Chapman’s theory is that the length of the working day is static over a period of time. This way it includes both simple and accumulated fatigue of the worker. The first declines seen in the output per hour reflect the simple fatigue of working 1 or 2 hours extra a day, but overtime, working more exponentially causes the worker’s productivity to decrease. Or better stated:

“Eventually daily fatigue is compounded by cumulative fatigue. That is, any additional output produced during extended hours today will be more than offset by a decline in hourly productivity tomorrow and subsequent days.” Robinson, 20045

Essentially working overtime today will affect the workers productivity not only that day but also tomorrow. As long as overtime is continued, the loss in productivity compounds and increases.

So should we never work overtime?

In 1980, a study was done on the construction industry on scheduled overtime and productivity. They found that when working 60 hour weeks compared to 40 hour weeks, they initially receive a boost in productivity, but it quickly drops. They found that there was an average 10–15% loss of productivity when working more than the average 40 hours a week. (Business Roundtable 1980)

Holding to Chapman’s theory, they found that productivity loss compounded over time. Within two months on 60 hour work weeks, the productivity loss is so great that the teams would actually be producing more if they had been working 40 hours the entire time. (Business Roundtable 1980)

The study was repeated in 1997, to prove the previous study’s findings. In this study overtime was only worked for 3 weeks or less, instead of 12, like 1980 study. They found that most crews followed the same loss rate as the BRT study curve, and established that “the BRT curve is a reasonable estimate in the calculation in the loss of productivity and overtime.” (Thomas and Raynor, 2005)

Scheduled Overtime and Labour Productivity

A study on overtime and an industry which is a bit more similar to the creative digital industry is “Transcending Socialization:
A Nine-Year Ethnography of the Body’s Role in Organizational Control and
Knowledge Worker Transformation.” The USC professor, Alexandra Michel, studied workers for 9 years in one of the the highest stress, fastest paced environments — Wall Street, and the effect it had on their health and work performance.

The bankers habitually worked 80 hour weeks, sometimes even up to 120. “Bankers saw time sheets as a “game,” trying to “beat their own best,”(Michel, 2011) In the first 1–3 years of starting at a bank, the workers neglected hobbies, families and their own health to work. (Michel, 2011)

On these workers performance, Michel found that “The banks’ yearly performance reviews of these workers (1–3 years) showed the most exhibited high technical and judgment performance. Creativity was distributed normally; up to twenty-seven percent displayed high creativity.” (Michel,)

By year 4, the bankers performance began to decrease. They also began giving in the addictions and compulsions. Many developed anger issues and body tics, all affecting their job performance.

By year 9, 60% of workers had a ‘breakdown’, either of mind or body. This caused some workers to pull back on the gruelling hours, and the bank benefit from this, as these workers experienced an increase in their creativity, technical skills and judgement, based on the bank’s annual reviews, shown above. (Michel, 2011)

From these examples we can see that working over 40 hours a week for extended periods of time has proven to decrease productivity dramatically. Are hours always this closely intertwined with productivity? What about when looking at a somewhat “normal” work week?

How close is the relationship between hours and productivity in 40 hour work weeks? >>