QUES­TIONS FOR

The World Bank is now us­ing be­havioural in­sights to tackle wicked prob­lems.

Rotman Management Magazine - - CONTENT - Varun Gauri

The World Bank’s Global In­sights Ini­tia­tive (GINI) was formed to put the key find­ings of your World De­vel­op­ment Re­port into prac­tice. Please sum­ma­rize those find­ings.

The main mes­sage that came out of that re­port was that de­vel­op­ment pol­icy is due for a re­design, based on a more re­al­is­tic un­der­stand­ing of how peo­ple think, be­have and make de­ci­sions.

In con­trast to the ra­tio­nal model — whereby peo­ple have un­lim­ited com­pu­ta­tional power to make ra­tio­nal de­ci­sions on the ba­sis of con­sis­tent and self-in­ter­ested pref­er­ences and ex­plicit in­for­ma­tion — re­al­is­tic think­ing is very dif­fer­ent. It re­lies on three prin­ci­ples: We usu­ally think au­to­mat­i­cally and fast; we are highly in­flu­enced by our en­vi­ron­ment and by so­cial norms; and ex­plicit and im­plicit in­for­ma­tion de­rived from our worldviews (i.e. men­tal mod­els) af­fect how we in­ter­pret in­for­ma­tion and make de­ci­sions. In the re­port, we em­pha­sized that every­one thinks this way — not only end users and con­sumers, but pol­icy mak­ers, heads of com­pa­nies, and even heads of state.

Put sim­ply, de­ci­sion mak­ing is highly con­tex­tual. You might typ­i­cally as­sume that if you raise the price of

some­thing by 10 per cent, peo­ple will de­mand less of it and sup­ply will in­crease; but when you make some­thing 10 per cent more salient in an in­ter­ac­tion, de­pend­ing on the en­vi­ron­ment, a wide va­ri­ety of things can hap­pen. That’s why it is so im­por­tant to un­der­stand the de­ci­sion-mak­ing con­text.

Why are in­sights from be­havioural sci­ence so well suited to the pol­icy realm?

Be­cause ev­ery pol­icy re­lies on ex­plicit and im­plicit as­sump­tions about how peo­ple make de­ci­sions and why they do what they do. As in­di­cated, most of those as­sump­tions rely on the ra­tio­nal, ide­al­ized model of how peo­ple think. By em­brac­ing be­havioural in­sights, we can re­design pub­lic poli­cies us­ing new tools, in­clud­ing choice ar­chi­tec­ture; so­cial re­wards and recog­ni­tion; the way in­for­ma­tion is pre­sented; and ac­ti­vat­ing new so­cial norms.

Your goal is to in­cor­po­rate be­havioural in­sights into some of the World Bank’s work. How are you go­ing about this?

We have three main cri­te­ria for se­lect­ing projects. The first is, do they ad­dress an im­por­tant de­vel­op­men­tal chal­lenge? Ar­eas that ad­dress ba­sic hu­man rights are given high pri­or­ity. Sec­ond, is the ini­tia­tive ‘owned’ by the World Bank, or a de­vel­op­ment part­ner? That is re­ally im­por­tant for sus­tain­abil­ity and en­sur­ing a long-term im­pact. Then fi­nally, can the in­ter­ven­tion be eval­u­ated? At this stage of our work, it’s re­ally im­por­tant to show that we’re hav­ing an im­pact, so we look for projects where we can pro­vide some form of eval­u­a­tion.

So far, our main ar­eas of work are in tax com­pli­ance, dig­i­tal fi­nance and sav­ings, ed­u­ca­tion, gen­der norms and or­ga­ni­za­tional ef­fec­tive­ness.

Why is this ‘ex­tra work’ nec­es­sary?

Be­cause time and again, we find that peo­ple have a very hard time fol­low­ing through on their com­mit­ments. We might want to save more, lose weight, or stop smok­ing, but we just can’t fol­low through be­cause we over­value the present and the willpower ex­er­tion is too large. Gov­ern­ments of­ten of­fer pro­grams — from con­di­tional cash transfers to health and ed­u­ca­tion pro­grams — and are shocked when peo­ple don’t take them up on it. This hap­pens a lot, par­tic­u­larly in the poor­est en­vi­ron­ments. The ques­tion is, why? One rea­son is that avail­ing your­self of such pro­grams re­quires ef­fort and takes time. Even though in the long run, it makes sense to take ad­van­tage of the pro­gram, in the short run, peo­ple may not have the band­width to do it.

As a re­sult, all peo­ple — rich and poor alike — some­times make choices that don’t pro­mote their own well-be­ing. Peo­ple may also get stuck in habits, suc­cumb to in­er­tia, and re­peat­edly pro­cras­ti­nate de­spite in­ten­tions to do oth­er­wise.

There was a study in Morocco in which the gov­ern­ment was pro­vid­ing a credit sub­sidy for a wa­ter hookup. Wa­ter avail­abil­ity is a big prob­lem there: peo­ple spend seven to nine hours each week col­lect­ing wa­ter, and it is of­ten the women do­ing the work. Yet, peo­ple weren’t sign­ing up for the pro­gram. The re­searchers found that the prob­lem was the ‘has­sle fac­tor’ — mak­ing time to sign up for the pro­gram. So, they de­cided to go di­rectly to peo­ples’ homes and get the nec­es­sary doc­u­men­ta­tion, and they brought it to the mu­nic­i­pal­ity. The re­sult: the sign up rate went up from a base­line of 10 per cent to 69 per cent! The im­pli­ca­tion is that mak­ing things eas­ier for peo­ple can have a huge im­pact.

Should gov­ern­ments be get­ting in­volved in shap­ing in­di­vid­ual choices?

I be­lieve they should, for three rea­sons. First, do­ing so helps peo­ple ob­tain their own goals. Re­minders to save money or take medicine help peo­ple achieve ob­jec­tives that they them­selves have set. Re­search shows that com­mit­ment con­tracts, which mar­kets un­der-pro­vide, can re­in­force de­ci­sions to adopt ben­e­fi­cial be­hav­iours. Match­ing the

De­vel­op­ment pol­icy is due for a re­design, based on a more re­al­is­tic un­der­stand­ing of how peo­ple think.

tim­ing of so­cial transfers to the tim­ing of charges for school en­roll­ment, or mak­ing it eas­ier to buy fer­til­izer at har­vest time when cash is at hand, helps to over­come in­ten­tion-to-ac­tion di­vides for peo­ple who may be for­get­ful or pos­sess in­suf­fi­cient willpower (that is to say, most of us). Many de­vel­op­ment poli­cies that op­er­ate at the bound­ary of Eco­nomics and Psy­chol­ogy can be un­der­stood in these terms.

Sec­ond, be­cause de­ci­sion mak­ing is of­ten based on only the most ac­ces­si­ble and salient in­for­ma­tion — and is also in­flu­enced by sub­tle so­cial pres­sures and ex­ist­ing men­tal mod­els — peo­ples’ pref­er­ences and im­me­di­ate aims do not al­ways ad­vance their own in­ter­ests. In­di­vid­u­als might choose dif­fer­ently, in ways more con­sis­tent with their high­est as­pi­ra­tions, if they had more time and scope for re­flec­tion. The as­sump­tion that we al­ways make choices that pro­mote our own in­ter­ests — of­ten a fun­da­men­tal bench­mark for pol­icy anal­y­sis—is mis­guided. But if de­ci­sion mak­ers do at times re­quire as­sis­tance, what guide­lines are to be used for the pol­icy in­ter­ven­tions aimed at shap­ing choice? We be­lieve that gov­ern­ments should fo­cus on the most im­por­tant free­doms, and in the de­vel­op­ment con­text, those in­clude free­dom from poverty, dis­ease, and op­pres­sion.

Third, so­cially-re­in­forced prac­tices can block choices that pro­mote well-be­ing and pre­vent in­di­vid­u­als from even con­ceiv­ing of cer­tain cour­ses of ac­tion. Our Re­port ar­gues that so­cial in­ter­de­pen­dence and shared men­tal mod­els af­fect choices, some­times cre­at­ing traps for com­mu­ni­ties and in­di­vid­u­als, in­clud­ing low trust, eth­nic prej­u­dice, and gen­der dis­crim­i­na­tion.

You be­lieve that be­haviourally-in­formed ap­proaches and tra­di­tional eco­nomic in­ter­ven­tions can be com­ple­men­tary— and that some of the big­gest pay­offs emerge from adapt­ing tra­di­tional in­ter­ven­tions. Please walk us through this.

Tra­di­tional in­ter­ven­tions usu­ally take two forms: in­cen­tives and ed­u­ca­tion/in­for­ma­tion. In­cen­tives usu­ally take the form of chang­ing prices in some way, or chang­ing penal­ties as­so­ci­ated with rules. The key idea is that peo­ple don’t re­spond to in­cen­tives plainly: they re­spond to in­cen­tives as they are rep­re­sented in their own minds. Some­times, they for­get about the con­se­quences of a cer­tain be­hav­iour and pay lit­tle at­ten­tion to it; or they might mis­cal­cu­late the prob­a­bil­i­ties as­so­ci­ated with a penalty. This pro­vides scope for chang­ing in­ter­ven­tions that high­light penal­ties. For in­stance, in the tax com­pli­ance do­main, peo­ple know that there’s a penalty for non-pay­ment, but if you high­light this fact, it might change how they be­have. If you think about it, vir­tu­ally all in­for­ma­tion can be made more salient, more timely and eas­ier to un­der­stand.

How can lead­ers know if be­havioural in­sights ap­ply to a par­tic­u­lar project?

They would need to think through a few things. First, are in­di­vid­u­als aware of what they need to do, but un­able to ac­com­plish it, or does a de­sired be­hav­iour need to be ac­ti­vated? Sec­ond, are in­di­vid­u­als mo­ti­vated enough to nudge them­selves? Third, will more de­lib­er­a­tion lead peo­ple to un­der­take the de­sired be­hav­iour, or are they al­ready suf­fer­ing cog­ni­tive over­load? And fourth, is the de­sired ac­tion not be­ing ac­com­plished be­cause of a com­pet­ing ac­tion, or due to in­er­tia? Con­se­quently, should you aim to dis­cour­age a com­pet­ing ac­tion or en­cour­age a tar­get ac­tion?

GINI has de­vel­oped a five-step process to us­ing be­havioural in­sights (see page 116). Which step have you found to be the most chal­leng­ing?

The most chal­leng­ing is prob­a­bly learn­ing from pi­lot­ing an in­ter­ven­tion (the fifth and last step), be­cause time is al­ways short for pol­i­cy­mak­ers and they tend to want things done very fast. Carv­ing out time for a pi­lot, es­pe­cially one that has learn­ing em­bed­ded into it, re­quires a mind­set change. We have found it chal­leng­ing to make the case to pol­i­cy­mak­ers about the im­por­tance of this step.

Peo­ple don’t re­spond to in­cen­tives plainly: We re­spond to in­cen­tives as they are rep­re­sented in our own minds.

The prob­lem is, pol­i­cy­mak­ers are of­ten look­ing for offthe-shelf solutions, in the way that econ­o­mists might ad­vise them to ‘raise the price of some­thing by five per cent’. Done. They don’t quite re­al­ize that if you are go­ing to walk down this road of believ­ing that de­ci­sion mak­ing is highly con­tex­tual, you need to take the time to make sure you get your in­ter­ven­tion right. Maybe some­day we’ll be able to give a quick pre­scrip­tion like, ‘Make this in­for­ma­tion 10 per cent more salient’; but we’re not there yet.

Also, the tar­get be­hav­iour (step 2) needs to be very pre­cise. It should not be de­com­pos­able into other con­stituent be­hav­iours. For ex­am­ple, ‘in­creas­ing re­tire­ment sav­ings of a house­hold’ is gen­eral and could be de­com­posed into pre­cise be­hav­iours such as ‘open­ing a new re­tire­ment ac­count’, ‘set­ting aside more money each month’, ‘re­duc­ing spend­ing on dis­cre­tionary items’, etc. The be­hav­iour should also be eas­ily ob­serv­able.

De­scribe some of the com­plex is­sues you are cur­rently tack­ling.

We are find­ing that a richer un­der­stand­ing of why peo­ple save, use pre­ven­tive health care, work hard, learn and con­serve en­ergy can pro­vide a ba­sis for in­no­va­tive and of­ten very cost-ef­fec­tive poli­cies and de­vel­op­ment in­ter­ven­tions.

At the mo­ment, we have a few projects on in­creas­ing women’s par­tic­i­pa­tion in eco­nomic and so­cial life. In Jor­dan, we’re do­ing a very long pi­lot. Rather than go­ing right to an in­ter­ven­tion, we’re try­ing to un­der­stand how so­cial norms in­flu­ence women’s de­ci­sions about whether to join the labour force, par­tic­i­pate in pub­lic life, and open a bank ac­count.

We started out with a sur­vey that asks house­holds — men and women — about how they per­ceive so­cial norms in this area, what they think other peo­ple do and what they think other peo­ple think they should do. We’re also look­ing at im­plicit bi­ases, to see whether these af­fect de­ci­sion mak­ing on the part of women as they de­cide whether or not to join the work­force. We’re tak­ing a long time to do this, be­cause we want to re­ally un­der­stand whether our in­ter­ven­tions should tar­get so­cial norms, the sto­ries peo­ple tell about them­selves or im­plicit bi­ases.

Look­ing ahead, how do you see the GINI ini­tia­tive evolv­ing?

Our aim is to start us­ing be­havioural in­sights to ad­dress some of the world’s big­gest prob­lems. What we need to do, there­fore, is re­main rel­e­vant to pol­i­cy­mak­ers by work­ing with them to en­able ‘quick wins’. For ex­am­ple, we can help them col­lect more rev­enues through taxes rel­a­tively quickly, or re­duce en­ergy con­sump­tion with changes in mes­sag­ing. We are truly un­der­stand­ing more about in­flu­enc­ing peo­ples’ be­hav­iour ev­ery day.

Varun Gauri is Head of the Global In­sights Ini­tia­tive (GINI) and Se­nior Econ­o­mist in the De­vel­op­ment Eco­nomics Vice Pres­i­dency of the World Bank. He serves on the ed­i­to­rial boards of the jour­nals Be­havioural

Pub­lic Pol­icy and Health and Hu­man Rights, the World Eco­nomic Fo­rum Coun­cil on Be­hav­iour and the Ad­vi­sory Board of Aca­demics Stand Against Poverty.

Nina Mažar is an As­so­ciate Pro­fes­sor of Mar­ket­ing at the Rot­man School of Man­age­ment and Co-di­rec­tor of Be­havioural Eco­nomics in Ac­tion @ Rot­man (BEAR). She is on leave from the School un­til Novem­ber of 2017 to serve as a Se­nior Be­havioural Sci­en­tist for the Global In­sights Ini­tia­tive (GINI) at the World Bank.

Newspapers in English

Newspapers from Canada

© PressReader. All rights reserved.