Labor Day, the 47-hour work week, and the problem with polls

How many hours do you work per week?

It seems like a straightforward question, and yet it’s less clear than it seems. For instance, which week? This week? Many Americans have today, Monday, off, so this isn’t a “typical” week. But what is a typical week?

I was thinking about this question when I read the results of a recent Gallup Poll. According to the LA Times write-up:

“Just 42% of full-time employees work 40 hours a week, the traditional total based on five 9 a.m.-to-5 p.m. workdays, Gallup said of findings it released ahead of the Labor Day weekend. Nearly the same percentage — 39% — say they work at least 50 hours a week. And almost one in five Americans, or 18%, said their workweek stretched 60 hours or more.”

Gallup’s usual model is to contact people and ask them things. But when it comes to estimating work hours, this methodology doesn’t work as well as asking whether you approve of Pres. Obama’s performance. The truth is, many of us don’t know how many hours we work, and we’re also prone to systematic bias based on things that may not be true.

So let’s start with the first problem. Forty hours is not the definition of full time. Usually 35 hours is. That still may mean 9 to 5, but few workers work straight through. When I was working at a fast food restaurant, an 8-hour shift involved a 30-minute break for lunch, and 2 15-minute “smoke” breaks. Most office workers take breaks too. It’s unclear from the LA Times write-up how many people chose 35 hours as an option, but it’s not uncommon.

Second, we have some reasonable data about the average work week already that we can compare this poll to for accuracy. According to the BLS’s employment situation summary, it’s 34.5 hours per week. Now, granted, Gallup is asking about full-time workers. But if we look at the American Time Use Survey, which relies on time diaries rather than asking people to estimate, we find that 71.9 percent of FT workers worked on an “average” day, and they worked 8.09 hours on that day. If you multiply a 7-day week times 0.719 you get almost exactly 5 days (it’s 5.03). So when you multiply that by 8.09 you get 40.7, or very close to the 40-hour ideal. If a survey gets 47 hours as the average, there’s clearly something different going on.

The difference is hinted at by the finding that 39 percent work more than 50 hours a week and 18 percent work more than 60. Generally, as people work more hours, they earn more. Consequently, when you’re looking at very high earners, you’d expect to see longer work weeks. But when I recently looked at the time diaries of women making 6-figures (for Mosaic, my next book), I found that fewer than 6 percent worked more than 60 hours a week.  

The truth is, unless people get paid by the hour, they may not know how many hours they work. And in an environment where people like to talk about how busy and stressed they are, where no one wants to appear like a slacker compared to the guy in the next cubicle, it’s easy to succumb to work week inflation. A “60-hour week” sounds like the kind of long week you’re working, so that’s what you tell Gallup.

But that doesn’t make it true. The only 60-hour workweek I’ve ever managed was when my kids were gone for a week and it felt like I was working around the clock to finish my book manuscript. I didn’t have a commute or anything else to interrupt me. I pretty much went from my bed to my desk. I worked all 7 days, and still just barely topped 60 hours.

Talking about how overworked everyone is may be one way to celebrate Labor Day, but I tend to think taking an honest look at one’s workweeks is the best way to start using them better. That’s how to start turning one’s labor into something to celebrate.

10 thoughts on “Labor Day, the 47-hour work week, and the problem with polls

  1. “Generally, as people work more hours, they earn more. Consequently, when you’re looking at very high earners, you’d expect to see longer work weeks ”
    This statement is completely ignoring the wage rate (and, for that matter, salary vs. hourly earning). Because of income and substitution effects going in opposite directions, it could go either way. People with higher hourly wages could work more because the opportunity cost of their time is higher or they could work less because they’re making their target income with fewer hours.
    The world of six figure female earners with kids says very little about most US workers.
    Also, as an aside, the ATUS separates commute time and work breaks (such as lunch) from work time, IIRC. It isn’t clear that that is the right thing to do if you’re talking about the burden of work.

    1. Totally agree about your last point. On my time audit, I track commuting and my billable hours separately, but relative to the 168 hours overall, they comprise work. It’s time that I’m NOT doing all of the other things I want to be doing…

  2. When I was working full-time on a salaried basis, and keeping a time diary, I didn’t work straight through the days, and one week I only “worked” 28 hours. But it still felt like 50 and had the same effect on my home life. Why? Because I didn’t have control over the time. Much of the time in between bouts of real work was wasted because it was broken up/interrupted. Or, I was waiting for information or instruction from others. Often I didn’t get this information until the end of the day, when my boss finally had time for it, but by then I needed to rush to go home, get dinner ready, and/or meet a child care deadline. So even though I’d spent quite a bit of the day not really working, I still had 30 minutes of stressed-out rushing around between 4:45 and 5:15 pm, and still didn’t even start making dinner until after 6 pm. I think there’s more to it than estimation bias or inflation on the part of the workers. A sense of control, or lack thereof, over your own schedule plays a big role too.

    1. Don’t forget commuting time, or time spent doing daycare dropoffs and pickups. Those aren’t technically “work” but both can utterly destroy a person’s day.

  3. I guess it depends, I normally work 40 hours at my regular job as an instructor. Then countless hours working on other projects, many to keep up with my profession, others because of kids and grand-kids etc.

  4. I think in the end, the only thing that counts is that you get planned things done. Therefore, if you are not able to pull yourself together and take maximum out of your working hours, then you need to stretch the limit a bit further. But that’s an excellent point that people are even proud and happy to admit that they are overworked and busy all the time.

  5. The commute + any pickup/dropoff time. That time is not YOURS and must be spent to do that job, therefore it belongs to work. Despite what the American Time Use Survey calls it, it is work.

Leave a Reply

Your email address will not be published. Required fields are marked *