The Downside of Transparent Decision Making
Skip to content
Organizations Leadership Jan 4, 2018

The Down­side of Trans­par­ent Deci­sion Making

Why you’ll get a bet­ter rec­om­men­da­tion from a com­mit­tee that delib­er­ates behind closed doors.

How transparent should decision making be?

Michael Meier

Based on the research of

Ronen Gradwohl

Timothy Feddersen

You’re an exec­u­tive with a big deci­sion to make. Per­haps there is a key C-suite posi­tion to fill, or a ques­tion about whether to enter a new mar­ket. So you turn to an advi­so­ry com­mit­tee, appoint­ing peo­ple who you believe will offer wise coun­sel because they col­lec­tive­ly have more infor­ma­tion about the issue than you do.

Add Insight
to your inbox.

We’ll send you one email a week with content you actually want to read, curated by the Insight team.

It would seem log­i­cal that you would want that com­mit­tee to be ful­ly trans­par­ent with you in their delib­er­a­tions, right? After all, you would get more infor­ma­tion if you knew what went on behind closed doors. 

But accord­ing to a recent analy­sis by a pair of Kel­logg School researchers, requir­ing trans­paren­cy may actu­al­ly yield less infor­ma­tion than allow­ing delib­er­a­tions to go on in private. 

At first blush, the intu­ition that trans­paren­cy is best in these advi­so­ry set­tings makes per­fect sense, explains Ronen Grad­wohl, an assis­tant pro­fes­sor of man­age­r­i­al eco­nom­ics and deci­sion sciences. 

If all you receive is a committee’s con­clu­sion, you are left with just a high­er-lev­el sum­ma­ry of the issue, he says. Where­as if you have trans­paren­cy, then you should get more infor­ma­tion about the prob­lem you’re try­ing to solve. And most peo­ple would think more infor­ma­tion is better.” 

But Grad­wohl, who teamed up with Tim­o­thy Fed­der­sen, a pro­fes­sor in the same depart­ment, points to the key flaw in this intu­ition: trans­paren­cy means you get to see what every­body said, not nec­es­sar­i­ly every­thing they know. 

Think twice before imple­ment­ing any­thing like rad­i­cal transparency.” 

What infor­ma­tion you see depends on what the com­mit­tee mem­bers are will­ing to dis­close — but trans­paren­cy tends to sup­press the will­ing­ness to dis­close infor­ma­tion, both to the oth­er peo­ple on the advi­so­ry com­mit­tee and to the final decision-maker. 

If the com­mit­tee knows that it’s going to be trans­par­ent,” Grad­wohl says, the com­mit­tee mem­bers will manip­u­late the infor­ma­tion they share or share less infor­ma­tion than they would otherwise.”

When Unan­i­mous Is Mis­lead­ing

You may have expe­ri­enced an inkling of this phe­nom­e­non. Meet­ing min­utes, for exam­ple, some­times say that a par­tic­u­lar vote was unanimous. 

Now, was it real­ly unan­i­mous?” Grad­wohl asks with skep­ti­cism. It could be that peo­ple went into the meet­ing with dif­fer­ing opin­ions, but they want­ed it to look unan­i­mous because they know that who­ev­er makes the deci­sion is going to try to read into not just what the rec­om­men­da­tion was, but also how many peo­ple were in favor, and how com­pelling was the evi­dence that this was the right decision.” 

For exam­ple, a group that wish­es to open a new branch of a retail chain might con­sist of four peo­ple who believe the new loca­tion will be very prof­itable and one who is unsure. After learn­ing how cer­tain the oth­ers are, that unsure per­son is like­ly to assume that her infor­ma­tion was sim­ply infe­ri­or to every­one else’s and become con­vinced that the new loca­tion makes sense. 

But if the deci­sion-mak­er knew that one per­son start­ed out uncer­tain about the new loca­tion, he might scut­tle the project. So to make the most com­pelling case to the chain’s own­er, it is in the best inter­est of the group to claim that all five peo­ple strong­ly believe the new loca­tion will be lucrative. 

In oth­er words, know­ing that the deci­sion-mak­er will have full trans­paren­cy into the rec­om­men­da­tion process can actu­al­ly change what peo­ple rec­om­mend. More­over, know­ing this, the deci­sion-mak­er might be sus­pi­cious of a unan­i­mous vote — and right­ly so. 

Thus, in this case, trans­paren­cy fails to achieve its goal of reveal­ing bet­ter, more accu­rate information. 

A Bab­bling Committee

The researchers turned to game the­o­ry to mod­el these sorts of sce­nar­ios. Their mod­el is pred­i­cat­ed on the idea that the com­mit­tee and deci­sion-mak­er have dif­fer­ent incen­tives: the deci­sion-mak­er might be more con­ser­v­a­tive, for instance, because he has more skin in the game,” while the com­mit­tee may be more open to tak­ing smart risks. Their mod­el also assumes that the only rea­son deci­sion-mak­ers turn to a com­mit­tee is because they do not have all the rel­e­vant infor­ma­tion themselves. 

In the opaque ver­sion of the game — when the only infor­ma­tion the deci­sion-mak­er is giv­en is the committee’s final rec­om­men­da­tion — the mod­el shows that the committee’s deci­sion ends up accu­rate­ly rep­re­sent­ing the group’s aggre­gate opin­ion; the ratio­nal deci­sion-mak­er might then find the committee’s rec­om­men­da­tion con­vinc­ing and take action on it. In this case, opac­i­ty works well, accord­ing to the researchers’ mod­el. Any dif­fer­ences between the com­mit­tee mem­bers and the deci­sion-mak­er in terms of moti­va­tions or tol­er­ance for risk is neutralized. 

Then we show the oppo­site case,” Grad­wohl says, refer­ring to the sce­nario of full trans­paren­cy. In this case, the com­mit­tee mem­bers don’t say any­thing meaningful.” 

This com­plete break­down of com­mu­ni­ca­tion comes about grad­u­al­ly, as each play­er sec­ond-guess­es the others. 

For exam­ple, con­sid­er the five com­mit­tee mem­bers who want to con­vince the own­er of the retail chain to open a new branch. Sup­pose that the com­mit­tee mem­bers are in favor of open­ing a branch when­ev­er a major­i­ty of them have con­vinc­ing pos­i­tive infor­ma­tion, but that the more con­ser­v­a­tive own­er is only will­ing to do so if all five have con­vinc­ing information. 

Here is what each member’s indi­vid­ual rea­son­ing on the com­mit­tee might look like, Grad­wohl explains: 

If oth­er peo­ple lack con­vinc­ing infor­ma­tion, then what I say doesn’t mat­ter, since we will be unable to con­vince the own­er regard­less of how con­vinc­ing my infor­ma­tion is.

The only sce­nario in which what I say does mat­ter is if all four of the oth­er com­mit­tee mem­bers have con­vinc­ing infor­ma­tion. In this sce­nario I am also in favor of open­ing a branch, because their infor­ma­tion suf­fices to con­vince me. Thus, I should tell the own­er that I am in favor.

But, the own­er will real­ize that I claim to be in favor whether or not I have con­vinc­ing infor­ma­tion, so he will essen­tial­ly dis­re­gard what I say.

At the same time, oth­er com­mit­tee mem­bers are engaged in this same rea­son­ing, Grad­wohl points out. 

So at the end of the day, the own­er gets five peo­ple claim­ing to be in favor but actu­al­ly learns noth­ing from this because they will make that same claim regard­less of their infor­ma­tion. At the end of the day, every­one babbles.” 

The com­mit­tee mem­bers share so lit­tle use­ful infor­ma­tion that they might as well be speak­ing gobbledygook. 

In real life, the out­comes won’t be as dras­tic as in the game, Grad­wohl acknowledges. 

Peo­ple aren’t per­fect­ly ratio­nal, so trans­paren­cy might not lead to zero infor­ma­tion, but it’s def­i­nite­ly there as a road­block to infor­ma­tion aggregation.” 

Mis­aligned Incen­tives

Such sce­nar­ios — where deci­sion-mak­ers decide whether to demand full trans­paren­cy when they seek rec­om­men­da­tions from par­ties with dif­fer­ing incen­tives — are com­mon in oth­er con­texts as well. 

Con­sid­er the case of a man­ag­er debat­ing whether to launch a risky new prod­uct. To help her decide, she turns to her employ­ees, each of whom has infor­ma­tion rel­e­vant to the deci­sion. Is she like­ly to get bet­ter infor­ma­tion by ask­ing employ­ees for their opin­ion one by one (anal­o­gous to trans­paren­cy) or by let­ting the employ­ees con­fer in pri­vate to come up with a joint rec­om­men­da­tion (the opaque scenario)? 

Our result sug­gests that ask­ing each employ­ee sep­a­rate­ly would lead to less infor­ma­tive advice and would be infe­ri­or to let­ting the employ­ees come up with a joint rec­om­men­da­tion,” Grad­wohl says. 

For exam­ple, employ­ees might be more eager than the man­ag­er to launch the new prod­uct, with less to lose if the prod­uct flops. So when the man­ag­er approach­es employ­ees indi­vid­u­al­ly, each employ­ee has an incen­tive to exag­ger­ate the new product’s odds of success. 

But once again, the dif­fer­ence in incen­tives between the man­ag­er and the employ­ees does not pose as much of a prob­lem in the opaque sce­nario. If employ­ees can have a pri­vate group dis­cus­sion, they can freely share their infor­ma­tion with one anoth­er before the group makes their joint recommendation. 

For exam­ple, if all employ­ees think the prob­a­bil­i­ty of suc­cess is low, the joint rec­om­men­da­tion will be to not launch the new prod­uct — and the man­ag­er has less rea­son to doubt that the rec­om­men­da­tion tru­ly reflects everyone’s infor­ma­tion. The same is true if all employ­ees think the prob­a­bil­i­ty of suc­cess is high: they will rec­om­mend the launch, and the man­ag­er will accept the recommendation. 

Lessons for Deci­sion-Mak­ers

Gradwohl’s main take­away from the research is straight­for­ward: Think twice before imple­ment­ing any­thing like rad­i­cal transparency.” 

Con­trary to con­ven­tion­al wis­dom, back­room dis­cus­sions and oth­er pri­vate con­ver­sa­tions can be more help­ful in get­ting peo­ple to share information.

These sorts of offline con­ver­sa­tions might actu­al­ly be ben­e­fi­cial to every­body — not just to the com­mit­tee, which is obvi­ous, but also to the even­tu­al deci­sion-mak­er,” he says. 

Featured Faculty

Ronen Gradwohl

Assistant Professor of Managerial Economics & Decision Sciences

Timothy Feddersen

Wendell Hobbs Professor of Managerial Politics, Professor of Managerial Economics & Decision Sciences

About the Writer

Marina Krakovsky, author of The Middleman Economy (Palgrave Macmillan), writes and speaks about ideas in the social sciences.

About the Research

Gradwohl, Ronen, and Timothy Feddersen. “Persuasion and Transparency.” Journal of Politics. Forthcoming.

Read the original

Suggested For You

Most Popular

Organizations

How Are Black – White Bira­cial Peo­ple Per­ceived in Terms of Race?

Under­stand­ing the answer — and why black and white Amer­i­cans’ respons­es may dif­fer — is increas­ing­ly impor­tant in a mul­tira­cial society.

Leadership

Why Warmth Is the Under­ap­pre­ci­at­ed Skill Lead­ers Need

The case for demon­strat­ing more than just competence.

Most Popular Podcasts

Careers

Pod­cast: Our Most Pop­u­lar Advice on Improv­ing Rela­tion­ships with Colleagues

Cowork­ers can make us crazy. Here’s how to han­dle tough situations.

Social Impact

Pod­cast: How You and Your Com­pa­ny Can Lend Exper­tise to a Non­prof­it in Need

Plus: Four ques­tions to con­sid­er before becom­ing a social-impact entrepreneur.

Careers

Pod­cast: Attract Rock­star Employ­ees — or Devel­op Your Own

Find­ing and nur­tur­ing high per­form­ers isn’t easy, but it pays off.

Marketing

Pod­cast: How Music Can Change Our Mood

A Broad­way song­writer and a mar­ket­ing pro­fes­sor dis­cuss the con­nec­tion between our favorite tunes and how they make us feel.