TCS Daily


Surfing for a Living?

By Stephen Bainbridge - June 2, 2006 12:00 AM

In a recent TCS column, Glenn Reynolds inveighed against workplace policies by which employers limit employee web-surfing, block instant messaging, and so on:

"Sell your stock in companies with policies like this one. The management is obviously stupid, and the only employees likely to stay, long-term, in the face of this kind of a policy are those who can't get a job someplace else, someplace where the management is brighter than a bag of hammers."

I wonder whether Professor Reynolds' self-interest as a blogger might have lead him astray this time. As former Delaware Chancellor William Allen aptly noted, albeit in a rather different context, "human nature may incline even one acting in subjective good faith to rationalize as right that which is merely personally beneficial."

Indeed, as one of his fellow bloggers, I share Professor Reynolds' frustration with such policies. An unscientific review of my referral logs suggests that a substantial percentage of my site traffic comes from corporate and government organizations, from which it seems reasonable to infer that most of my readers are visiting the blog during working hours from their office computers. As more employers block employees from web-surfing, our readership likely will decline, reducing the pecuniary and psychic benefits of blogging. Yet, what is in Reynolds' and my self-interest as bloggers may not be in the best interest of employers and the economy as a whole.

Professor Reynolds makes three basic claims:

  1. The continuing development of technologies offering pervasive web access renders such policies unenforceable.
  2. Web-blocking policies are bad corporate policy: "the more intrusive the policies seem, the harder it will be to attract bright, creative employees who are marketable elsewhere: Just the kind of people that companies ought to want to hire and to keep."
  3. Instead of focusing on inputs, such as how employees spend their time, managers can and should focus on measuring productivity: "measuring the work done, rather than just whether employees manage to look busy, is going to be the management trend of the future."

The first may well be right, but the second is debatable, and the third is almost certainly wrong.

To be sure, the kids who work at places like Google probably would rebel if their Web access were restricted. I suspect it's precisely that sort of folks whom Reynolds has in mind. Yet, not everybody is -- or wants to be -- a Googler.

In my scholarship on employment and corporate workplaces, I've spent a lot of time reading the academic literature on the purportedly evolving nature of work and employee personalities. Based on that experience, I've concluded that there must be some sort of projection bias at work, because academics routinely overstate the extent to which creativity matters. In fact, however, as I've documented elsewhere, lots of economically vital jobs remain dull with little demand for individual creativity and lots of workers are perfectly happy in such jobs.

In arguing that web blocking policies are bad corporate policy, moreover, Professor Reynolds overlooks the costs associated with broad employee web access:

  • Exposing the corporate network to viruses and spyware
  • Use of corporate resources by employees running a side business out of the office
  • Waste of bandwidth
  • Display of inappropriate images on computer screens could expose the firm to liability for sexual or racial harassment, as can offensive emails and chat
  • Lost productivity: US business employers supposedly lost half-a-billion dollars on the single day in 1999 when Congress posted the Starr Report on the web. By some estimates, over 13 million employees read the Starr Report during working hours on office computers.

Reynolds is probably right that a blanket prohibition of Internet access is unwise. Establishing and enforcing acceptable use policies, however, probably is a very good idea. After all, by one estimate, employee use of the Internet for personal purposes costs "American corporations more than $178 billion annually in lost productivity."

As for Reynolds' claim that managers can and should focus on monitoring outputs rather than inputs, most students of corporate organization likely would disagree. Measuring outputs was easy when most employers worked manufacturing jobs on an assembly line. If one worker slacked off, the cause of the backup could be spotted readily, as famously illustrated in the classic "I Love Lucy" episode in which Lucy and Ethel fought a losing battle with the assembly line at the candy factory.

Much modern work takes place in teams, which are often more-or-less self-directed. Such teams create a severe monitoring problem: Even if the team's joint productivity can be measured on an output basis, it is likely to be difficult to measure the contributions of each individual to the total. Economists Armen Alchian and Harold Demsetz offered the classic example of two workers who jointly lift heavy boxes into a truck. The marginal productivity of each worker is very difficult to measure and their joint output cannot be easily separated into individual components. In such situations, obtaining output-based information about a team member's productivity and appropriately rewarding each team member are very difficult and costly. The problem is even worse in modern service jobs, where team productivity is essentially non-separable.

Given the nearly intractable monitoring problems inherent in the modern workplace, it is hardly surprising that many employers resort to prophylactic rules, such as banning Web surfing. Indeed, even in the academy, where both Professor Reynolds and I make our living, a growing number of schools are adopting restrictions on Web-surfing in class room settings, for many of the same reasons that firms are doing so in the workplace. Hence, I suspect it is not "measuring the work done," but rather devising workable and cost-effective acceptable use policies that will be "the management trend of the future."

Stephen Bainbridge writes a weekly column for TCS and teaches law at UCLA.

Categories:

9 Comments

Well, then...
No one will ever visit TCS...

Surfing while working SWW
I would like to offer a possible solution to the surfing while working. I prefer the first writers proposal of freedom to surf. At the same time, I believe that restrictions should be on unproductive time-wasting. Block games and porn sites. Leave the productive areas of the web available. I work at a school and surf while I work. It is very productive and educational to do so. I read news (not celebrity news), blogs like this one, health and military websites. I am a much more valuable employee as a result of my education and staff development at no cost to the school/workplace because I get my work done. It would be a disaster to block productive web use.

You're both wrong.
Unrestricted web browsing is a problem for business for many of the reasons you listed. Viruses, inappropriate materials, bandwidth usage... but lost productivity is a hard sell.

The studies you're talking about (unless I missed something) basically count up the number of hours spent surfing, multiply that by the pay of the worker, and voila we have lost productivity. I'm sorry, I don't buy it.

It's been shown that small breaks throughout the day increase overall productivity. There are down times when a worker is waiting for somebody else to get things done so they can continue (that is lost productivity, but not caused by the web, here the web use is caused by lost productivity).

You then go on to use service jobs as the example of how individual work cannot be separated from the whole. How often are service workers sitting in a chair if front of a computer? And in any team, the team knows who is pulling their weight and who is not. If that is not being communicated to management there is a management style issue.

Limiting use is a necessary thing, but broad bans can be just as harmful as open web browsing.

Exposure
I'll go back to what I said to Glenn's piece. One simple tool, that already exists I might add, makes this whole argument moot. Post everyones web logs on the corporate intranet for all to see. This would clean up almost all bad behavior and make any problems obvious. Most people don't realize that at most companies this information is already accessible to the system admins, software developers, and computer support folks. We might as well spread the knowledge out to all employees rather than limit it to the chuckling few.

Simple Solution
Set up a dedicated Server that is reserved for internet use only.

Workers must use Terminal Services (a remote access utility built into Windows) to tap into this server to browse the web. The enables protection against viruses and company files going in and out via undetected 3rd party email.

Workers are limited to 30 minutes total a day.

Employees need to be able to check their bank accounts, check weather, school web sites and so forth during the 8-hour work day. They can easily get those jobs done through out the day with 30 minutes total. God forbid they used their 30 minutes during their lunch break.

In general, office discipline is a joke in many places. The companies who do not keep strict guidelines in place for non-work related activity while ON THE CLOCK do so at their on peril.

No simple solution
I spent some time today finding a service manual for a lab type power supply I picked up on Ebay. I'll use it at work. How do you allocate the time I spent. Was it surfing, work or what.

Oh yes, The manual was a 19 Mbyte PDF file. Downloading it over a 56k line took about 1.5 hours, (started before lunch, finished after lunch was over.) How do you account for this if you are going to talk about bandwidth. (The actual cost here was one local phone call.)

My job is to keep equipment running. When all the stuff is running, and I don't have parts to repair, it can get very boring. I read electronics mags, some blogs, TCS, shop for test equipment on ebay, etc. It's a great alternative to looking at a large room full of equipment, hoping for something to break.

You need to find a way to measure outputs
When Prof. Reynolds posted his piece, his item #2 reminded me of a dinner discussion I had a couple weeks ago with (pardon the blatant name dropping that follows) occasional TCS columnist Ted Balaker. We were talking telecommuting and I was picking his brain about the conditions needed for employers to agree to it. We agreed that a management focus on results and not on process was key. One thing I suggested as working against telecommuting is that automated management tools seem to focus more on process than results. Think key loggers, time clocks, etc.

I've been thinking that one more thing is important here, and that's how we compensate people. Most workers are paid hourly, not by output. Most managers are paid salaries, with expectations for the length of their work days and place of operation. This is cemented in both law and the national psyche, with "minimum wage" laws and "annual income" lines on credit applications. In managing transactional processes, such as technical support or order taking, we often think in terms of average cost per transaction (to be optimized down over time by managing) rather than price we are willing to pay per transaction and service level we wish to achieve (to be optimized by letting a market of workers take over).

The question our managers have to face is whether they are trying to do something great or whether they're trying to be the best babysitters they can. As Reynolds points out, workers will start to gravitate to the style that best fits them.

Hmmmm
Seems to me that the author of this article spends too much time in the classroom and not enough in the boardroom.

To reiterate some points already made by other posters:

- empirical evidence demonstrates that small breaks every hour make for more productive employees (in my office this includes fussball and web surfing).

- the author couldn't be more wrong when he declares it near impossible to measure team activity. Any company which wants to stay in business has a review process and means of communication to address non-performers.

- Glenn had a great point. For white collar workers (which is really who we are talking about) who are "in demand", they will choose to NOT work for a company that restricts such trivial priviliges as surfing the web. Such a restriction is a good indicator that other such silly disrespectful policies exist.

- as an executive who manages teams and is known to read blogs at work (Instapundit, TCS and LGF come to mind), I give me team members goals. If they accomplish these goals while only working 5 minute workdays, good for them. I care about results and like any good manager, I have established ways to measure those results.

Real world
The private school where I work has web blocking software. The Tech Team has periodic discussions about what options to activate. Our discussions usually focus on what we are trying to accomplish in providing Internet access. Our goal is to teach students to learn about technology and use it responsibly.

About a month ago we turned on an option to block "student time wasting sites". The next day I had a teacher tell me that a student was blocked from a site he needed to access to finish a research project, so I turned the option off.

What do we block? Basically, just porn. MySpace got blocked before it got in the news because a parent checked out a "friend's" links and none of our teachers could think of an "educational purpose" in blogging.

We recently rewrote our acceptable use policy and removed some prohibitions, like the one against "shopping", since it did not make sense when the Junior High computer teacher was giving assignments simulating shopping and the prohibitions did not make sense with our goal to teach students to learn about technology and use it responsibly.

One summer project is to automatically send teachers a listing of where students went at the end of each class period. That is more in line with our goals that playing control games.

No web blocking software is perfect and can lead to funny false positives being blocked like the time our lunch coordinator got blocked from the Chick-Fil-A (who caters student lunches on Wednesdays) web site, because they had "big" drinks and chicken "breast" sandwiches. Of course, Chick-Fil-A is now on our exemption list.

TCS Daily Archives