the lectern banner

By Steve Kelman

Blog archive

The devil, the details and contract terms

I promised in my blog earlier this week, the one about research on cooperation among colleagues, to write about another interesting presentation we heard from a candidate for an assistant professor job in management: Eileen Chou of Kellogg Business School at Northwestern.

Chou's presentation -- with the cutesy title "The devil is in the details" -- presents some experiments challenging the common view that contracts with more specific contract terms are better than those with less specific ones. Some of the experiments were done in the lab, and others involved recruiting people for a job on the website mTurk. (The site is usually used to recruit real people for jobs they do at home, and has become an increasingly popular tool for academics who wish to study psychological phenomena.)
Chou looks at the effect the specificity of contract terms has on the performance of the person who has signed the contract. The contracts she considers are employment contracts, and she looks at them from the perspective of the employee. In the experiments, she manipulates some of the terms of the contracts to make them more or less specific.

So, for example, in one experiment the two versions are "we will check the responses of 25 percent of the employees" vs. "we might check some employee responses." In another, the alternatives are a group that meets “two days a week (on weekdays)" vs. "a bi-weekly focus group." In a third, participants will speak "for two to three minutes" vs. "for a couple of minutes."
What she found is that the less specific contracts increased employee persistence, creativity and organizational commitment compared with the more specific ones. Chou argues that this occurred because the less specific contracts increased employee intrinsic motivation, which research has shown is related to a person's feelings of competence and autonomy. She suggests that lack of detail may be interpreted as a vote of confidence in the employee's competence, and that fewer constraints, and incentives that may not be well specified, promotes a feeling of autonomy.
These experiments touch on age-old issues in government contracting and in how to promote good contractor performance. In some sense, these results are more consistent with the "trust/cooperation" view of how to organize contracting than with the "adversarial/monitoring" view.

Of course, it's a real stretch to go from these results to definite conclusions about government contracting. First, it should be noted that all the specific terms involved either inputs or monitoring -- the experiments did not include performance demands ("do your best" vs. "keep the system up and running 99.5 percent of the time"). Second, the experiments looked at effects of specificity on the behavior of individual employees in one-on-one contracts, and it is unclear whether or to what extent these findings would apply to employees if the contract is between a government agency and a service provider.
Nonetheless, it is an interesting contribution to debates on government contracting from an unlikely source.

Posted on Jan 20, 2012 at 12:09 PM

The Fed 100

Read the profiles of all this year's winners.


  • Then-presidential candidate Donald Trump at a 2016 campaign event. Image: Shutterstock

    'Buy American' order puts procurement in the spotlight

    Some IT contractors are worried that the "buy American" executive order from President Trump could squeeze key innovators out of the market.

  • OMB chief Mick Mulvaney, shown here in as a member of Congress in 2013. (Photo credit Gage Skidmore/Flickr)

    White House taps old policies for new government makeover

    New guidance from OMB advises agencies to use shared services, GWACs and federal schedules for acquisition, and to leverage IT wherever possible in restructuring plans.

  • Shutterstock image (by Everett Historical): aerial of the Pentagon.

    What DOD's next CIO will have to deal with

    It could be months before the Defense Department has a new CIO, and he or she will face a host of organizational and operational challenges from Day One

  • USAF Gen. John Hyten

    General: Cyber Command needs new platform before NSA split

    U.S. Cyber Command should be elevated to a full combatant command as soon as possible, the head of Strategic Command told Congress, but it cannot be separated from the NSA until it has its own cyber platform.

  • Image from Shutterstock.

    DLA goes virtual

    The Defense Logistics Agency is in the midst of an ambitious campaign to eliminate its IT infrastructure and transition to using exclusively shared, hosted and virtual services.

  • Fed 100 logo

    The 2017 Federal 100

    The women and men who make up this year's Fed 100 are proof positive of what one person can make possibile in federal IT. Read on to learn more about each and every winner's accomplishments.

Reader comments

Tue, Feb 7, 2012 Tim

I agree with the sentiment here. If you need a system and try to write a work statement that identifies that the system shall do x and perform y and look like z, then yes, it will be a very painful process and the end product is not likely to look like what you identified up front. There are 2 reasons for this: 1. Requirements and design is likely to be part of the process and to include the requirements and design in the work statement pre-supposes that part. 2. Time. Things happen, life happens, and what we need today will very likely be different from what we need tomorrow. My recommendation, be very very specific, but be specific about the right thing. You are not buying a system in that contract. You are buying services. Write the work statement to correctly account for what it is you are buying. Don't waste any energy trying to identify the "What" but rather focus on the "How". For an application, how many requirements sessions will you need? What does a model requirements gathering session look like? What are the artifacts of that process? This is how I structure my work statements, and I have been on-time, on-budget, on-scope, on-quality and most importantly, on-satisfaction with my projects since adopting this approach.

Wed, Jan 25, 2012 Gorgonz.

Herr Professor: You should run a column on the Booz Allen layoffs and what they mean for the contracting industry. Just a thought.

Tue, Jan 24, 2012 Steve Kelman

Thanks all, for these interesting and thoughful comments.

Mon, Jan 23, 2012 Vern Edwards

In 2010 Ms Chou co-authored an article entitled, "The Hidden Costs of Contracts on Relationships and Performance." It's available here: Here is the abstract: "Rather than acting as a safeguard, contracts may actually damage relationships. Using field and lab studies, we assessed the effects of contracts on contract formation, implementation, and ultimate outcomes. Studies 1 and 2 showed that the presence of a contract led people to anticipate more contentious interactions, which then led to a reduction in their actual cooperative behavior. Studies 3 and 4 showed that contracts that included sanctions led to poor team performance." Fascinating paper, and instructive.

Mon, Jan 23, 2012 Vern Edwards

The findings as described by Steve, while not about government contracts specifically, seem to be consistent with the findings of Stewart Macaulay in the 1960s. See "Non-Contractual Relations in Business: A Preliminary Study," American Sociological Review, Vol. 28, No. 1, February 1963, available at See also the paper that Ralph Nash and I wrote for the Procurement Round Table, "A Proposal for a New Approach to Performance-Based Services Acquisition," Defense Acquisition Review Journal, Vol. 14, No. 2, September 2007, available at Any attempt to write a detailed statement of work for a long term (one year or longer) and complex (multi-function or multi-task) service contract will fail, because "the devil is in the details" and the details cannot be known in advance. Trust, cooperation, teamwork, and ad hoc decisionmaking, not contract language, are the keys to success.

Show All Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group