INCREASE SAVINGS January  2015       By:  Kristen  Berman  of  Irrational  Labs       . OVERVIEW................................................4   THE  PROBLEM………….…………………………….6   RULE  OF  THUMB…………................………....9   MENTAL  ACCOUNTING.………...................13   PRE-­â€COMMITMENT.………………………...…..16   OPPORTUNITY  COST  NEGLECT…….…….….17   FAST  FEEDBACK……………...………………..…..19   Table of Contents ACHIEVABLE  GOALS……………...……………...21   CHANGE  THE  MINDSET……………...…….…..24   PUT  FRICTION  IN  THE  RIGHT  PLACES……..26   CONCLUSION………………………………...……..28   WORKS  CITED…………………………….....……..30           2   .       METLIFE  FOUNDATION   Since   its   founding   in   1976,   MetLife   Foundation   has   provided  more  than  $530  million  in  grants  and  $100  million   in   program-­â€related   investments   to  nonprofit  organizations.     The   MetLife   Foundation   believes   that   affordable,   accessible   and   well-­â€designed   financial   services   can   transform  the  lives   of   those  in   need,  and  have   committed   $200   million   over   the   next   five   years   to   advancing   this   effort  around  the  world.  The  foundation   has   built  its   vision   for   global   financial   inclusion   on   three   powerful   pillars:   Access   and   Knowledge,   Access   to   Services   and   Access   to   Insights.   IRRATIONALLABS   IRRATIONALLABS   Led   by   famed   behavioral   economist   Dr.   Dr.   Dan   Ariely   and   Led   by   famed   behavioral   economist   Dan   Ariely   and   founder   Kristen   Berman,   Irrational   Labs   is   a   is   a  nonprofit  that   founder  Kristen   Berman,  Irrational   Labs   nonprofit   that   applies  behavioral  economics  findings  to  findings   marketing,   applies   behavioral   economics   product,   to   product,   and   marketing,   and   organizational   design   problems.   By   organizational   design   problems.   By   understanding   human   decision  making   and   motivations,   Irrational   Lmotivations,   understanding   human   decision   making   and   abs  help   companies   create   compelling   value   propositions,   grow   their   Irrational   Labs   help   companies   create   compelling   value   customer  base  and  grow  ttheir  ccustomer   engaged.   keep   those   propositions,   keep   hose   ustomers   base   and   customers  engaged.   3   . I. OVERVIEW…..….........… A  group  of  researchers  studied  the  effects  of  peer  pressure  and  self-­â€help  groups   on  savings  behavior  and  found  that  it  was  effective  at  helping  individuals  save   money.  1   The  experiments  were  conducted  in  Chile  with  low-­â€income  micro-­â€   entrepreneurs  who  earned  an  average  of  84,188  pesos  (175  USD)  per  month.     Sixty-­â€  eight  percent  of  participants  did  not  have  a  savings  account  prior  to  the   study  and  were  required  to  sign  up  for  an  account  based  on  the  savings  group   they  were  assigned  to:   1. Savings  group  1  -­â€  a  basic  savings  account  with  an  interest  rate  of  0.3%.     2. Savings  group  2  -­â€  a  basic  savings  account  with  an  interest  rate  of  0.3%.   The  participants  were  also  part  of  a  self-­â€help  peer  group,  where  they   could  voluntarily  announce  their  savings  goals  and  monitor  their   progress  on  a  weekly  basis.     3. Savings  group  3  -­â€  a  high  interest  rate  account  with  a  rate  of  5%  (the  best   available  rate  in  Chile).   The  study  found  that  participants  that  were  part  of  the  self  help  peer  group   (savings  group  2)  deposited  3.5  times  more  than  others  and  their  average   savings  balance  was  almost  double  of  those  who  held  a  basic  savings  account.   The  high  interest  rate  had  a  very  little  effect  on  most  participants.     To  further  understand  why  self-­â€help  peer  groups  work,  a  second  study  was  con-­â€   ducted  a  year  later.  The  participants  were  divided  into  two  groups.     One  group  received  text  messages  that  notified  participants  of  their  progress   and  the  progress  of  other  participants.  They  were  assigned  a  savings  buddy  with   whom  they  would  meet  on  a  regular  basis  and  who  would  hold  them   accountable  to  their  savings  goals.                                                                                                                             1 2  Mast,  Meier,  Pomeranz,  2012.  http://www.nber.org/papers/w18417.pdf    http://www.dartmouth.edu/~alusardi/Papers/American_Life_Panel.pdf   4   .

The  other  group  only  received  text  messages  that  notified  participants  of  their   progress  and  the  progress  of  other  participants.     The  results  of  the  second  experiment  found  that  having  a  savings  buddy  made   very  little  difference  and  that  receiving  text  messages  was  just  as  effective.     As  noted  by  the  researchers,  having  peer  groups  was  an  effective  commitment   device  to  achieving  savings  goals  but  meeting  in-­â€person  was  not  necessary.   Receiving  text  messages  indicating  their  progress  and  the  progress  of  the  peers   was  just  as  effective.         Small  changes,  big  effects   As  this  example  shows,  and  many  other  academics  and  banks  have  proven,   small  changes  to  the  design  of  a  savings  scheme  can  have  big  effects  on   participation  rates.    In  the  case  above,  a  simple  SMS  message  that  reported  and   reminded  them  how  they  were  doing  and  how  their  peers  were  doing  was   enough  to  increase  savings.     Below   we   summarize   some   of   the   main   principles   that   should   be   used   when   designing   a   savings   account,   and   in   particular,   many   of   our   examples   will   be   about  designing  an  emergency  savings  fund.  An  emergency  savings  fund  is  the   first  step  in  savings  and  something  many  low-­â€income  people  (and  people  with   low  financial  literacy)  are  lacking.     These   principles   and   the   approach   can   be   used   by   Fin-­â€tech   companies   to   encourage   good   financial   behaviors,   policy   makers   to   inform   policy   decisions   and  banks  to  increase  the  financial  well-­â€being  of  their  customers.           5   . II. THE PROBLEM.……… Financial   health   in   the   US   is   at   an   all-­â€time   low.   Almost   half   of   all   households   have   less   than   three   months   worth   of   savings,   and   less   than   30%   feel   confident   they   will   have   enough   money   to   cover   their   basic   living   expenses   (Helman  2014).       Unfortunately,   the   current   approach   to   solving   our   financial   health   problems   in   America   is   not   working.   In   order   to   address   this   massive   and   growing  problem,  we  need  to  take  a  different  approach,  a  behavioral  one.   Until   recently,   the   primary   approach   to   fix   financial   health   has   been   to   improve  financial  literacy.  Lusardi  and  Mitchell  (2007)2  and  Lusardi  and  Tufano   (2009),   found   incredibly   low   levels   of   financial   literacy   in   the   US   population   –   this  includes  the  inability  to  understand  basic  concepts  such  as  the  importance   of  retirement  savings,  and  poor  judgment  in  borrowing-­â€decisions.   Given   this   information,   solution   designers   hypothesized   that   if   we   could   only  teach  people  the  right  things  to  do  with  their  money,  they  would  be  more   responsible   with   it.   And,   thus   the   US   spent   $670   million   in   financial   education   training  programs  (Consumer  Financial  Protection  Bureau  2013).    However,  time   and  again,  academic  experiments  have  proven  that  information  alone  does  not   sufficiently  motivate  positive  decision-­â€making  (Fernandes  et  al  2014)3.       Most  Americans  know  that  they  should  quickly  pay  down  debt,  put  more  money   toward  emergency  and  retirement  savings,  and  spend  less  on  unnecessary   items.  However,  psychologists  have  demonstrated  that  cognitive  hurdles  such   as  a  lack  of  self-­â€control,  hyperbolic-­â€discounting  and  general  optimism-­â€bias,   prevent  this.  Instead,  if  we  have  money  readily  available  we  find  it  hard  to  make   decisions  that  positively  impact  our  future.  (Benhabib,  Bisin,  Schotter,  2009)4                                                                                                                           2  http://www.dartmouth.edu/~alusardi/Papers/American_Life_Panel.pdf    http://pubsonline.informs.org/doi/abs/10.1287/mnsc.2013.1849   4  http://www.nyu.edu/econ/user/bisina/GEB%20BBS.pdf 3   6   .       7   . A   BEHAVIORALLY   LED   APPROACH   to   increase   savings   will   fundamentally   look   at   ways  to  make  money  feel  less  available  by:   • Making   it   easier   to   save   by   intervening   at   the   point   of   decision   (e.g.,   automating  a  savings  deduction  from  paycheck)     • Increasing   general   motivation   to   save   through   reward   substitution   and   incentive   design   (e.g.,   creating   a   lottery   type   system   associated   with   saving  deposits)     • Changing   a   person’s   mindset   about   savings   and   what   it   means   to   them.   (E.g.,   personally   identifying   as   the   kind   of   mom   who   prepares   for   their   child’s  college  tuition)           8   . III. RULE OF THUMB…………   People   make   a   tremendous   number   of   decisions   that   involve   money   every  day.  We  decide  whether  we  should  buy  coffee,  take  public  transportation,   if  we  should  get  the  soup  or  the  steak  sandwich.   However,  the  reality  is  that  financial  decisions  are  incredibly  complex.     Consider,  for  example,  the  simple  case  of  a  buying  a  cup  of  coffee  for  $4.   To   carry   out   this   decision   properly,   we   need   to   think   about   not   only   the   pleasure  that  we  will  get  from  the  cup  of  coffee  but  also  the  opportunity  cost  —   what   we   will   give   up   for   the   pleasure   of   the   coffee.   Only   if   the   expected   pleasure  of  the  coffee  is  larger  than  the  opportunity-­â€cost  should  we  spend  the   money.   This   could   be   an   incredibly   hard   decision.   So   how   do   we   make   these   seemingly  simple,  but  ultimately  very  complex  decisions?   Sometimes  we  don’t.       For   example,   instead   of   imagining   all   the   ways   we   could   substitute   a   coffee  with  something  cheaper  or  the  ways  in  which  we  could  use  $4  that  would   provide  more  utility,  we  buy  the  coffee.     We   generally   don’t   consider   the   short-­â€term   utility   of   the   purchase,   relative  to  the  long-­â€term  utility  of  saving  the  money  or  spending  it  differently.   Instead,  we  simply  repeat  our  past  behavior.             9   . WHAT  SHOULD  WE  BE  DOING?     One   solution   is   to   educate   people   more   on   how   to   make   these   complex   tradeoffs  at  the  point  of  decision.    But  this  sounds  tiring.    If  we  weighed  the  cost   and  benefits  of  everything,  all  the  time,  our  lives  would  be  stressful  and  busy.     Besides   just   remembering   how   to   do   this   -­â€   which   is   difficult   in   itself   -­â€   imagine   the   poor   coffee   buyer   who   would   have   to   pull   out   a   notepad   before   purchasing   his  cup  of  coffee.     Now   imagine   this   process   for   a   much   more   complex   financial   decision,   like  how  much  to  save  for  retirement.    In  order  to  calculate  an  optimal  personal   savings   rate,   we   have   to   make   assumptions   on   very   uncertain   outcomes,   like   future  rates  of  return,  income  flows,  retirement  plans,  and  health  care  needs.     People  do  not  do  this.  In  a  survey  of  faculty  and  staff  at  the  University  of   Southern   California,   58   percent   spent   less   than   one   hour   determining   their   contribution   rate   and   investment   elections   in   a   retirement   account   (Benartzi   and  Thaler  1999)5.         So  how  do  we  make  these  financial  decisions?  The  results  from  multiple   experiments   show   that   people   rely   on   simplified   rules   of   thumb   —   heuristics   —   to  help  us  solve  these  complex  problems.                                                                                                                           5   http://wolfweb.unr.edu/homepage/pingle/Teaching/BADM%20791/Week%209%20Behavioral%20Microeconomic s/Benartzi-­â€Thaler%20Biased%20Savings%20Behavior.pdf   10   .   For  example,  Ideas  42  ran  a  test  on  teaching  a  financial  curriculum  to  two   groups  of  people.  One  was  a  standard  literacy  curriculum.    And  one  was  based   on  strong  rules-­â€of-­â€thumb.  The  rules-­â€of-­â€thumb  training  increased  the  fraction  of   people  keeping  accounts  and  those  separating  household  and  business  finances   by  11  percentage  points  each.  It  also  raised  sales  during  bad  weeks  by  18.5%   relative  to  the  comparison  group  (Ideas42  n.d.).6       Below  are  four  proven  rules  of  thumb  we  should  consider  as  we  design  a   retirement  savings  program,  application  or  experience  (Benartzi  Thaler  2007).       • SAVING   THE   MAX:   When   there   was   a   max   set   at   16%   contribution,   21   percent  of  new  hires  deferred  16  percent  of  their  income.    After  moving   the  max  to  100%,  5  percent  deferred  16  percent  and  7  percent  deferred                                                                                                                           6  http://www.ideas42.org/financial-­â€heuristics/   11   . more   than   16   percent.   Thus,   the   share   of   employees   saving   at   least   16   percent  decreased  from  21  to  12  percent  (Benartzi  Thaler  2007).     • SAVING   THE   MIN   TO   GET   THE   MATCH:   Common   rule   of   thumb   is   to   contribute   to   a   retirement   account   the   minimum   necessary   to   get   the   full   employer   match.   For   example,   if   the   employer   matches   employees’   contributions  up  to  6  percent  of  pay,  then  many  employees  contribute  6   percent  (Benartzi  Thaler  2007).     • SAVING   IN   MULTIPLES   OF   FIVE:   Hewitt   Associates   found   that   the   distribution  of  contribution  rates  spikes  at  multiples  of  five  percent,  even   though   there   were   no   specific   thresholds   of   either   five   or   10   (Benartzi   Thaler  2007).     • SAVINGS  USING  THE  1/N  RULE:  We  divide  evenly,  when  possible.    Imagine   having  a  sum  ($100)  and  you  can  put  however  much  money  you  want  into   N  number  of  buckets.    If  you  have  four  buckets  (n=4),  65%  of  the  people   use  the  1/nth  rule.    But,  if  you  have  three  (n=3)  to  divide  $100,  1/n  rule  is   only  used  by  18  percent  of  the  participants  (Benartzi  Thaler  2007).       12   .   IV. MENTAL ACCOUNTING…. We   mentally   assign   money   to   different   categories   of   spending.   And   while   this  is  just  a  mental  exercise  and  nothing  is  actually  different  about  the  money,   we   spend   money   according   to   these   categories.     In   addition,   money   that   we   haven’t  assigned,  feels  different  than  money  we  have  assigned  (Thaler  1985).         If   you   were   to   get   a   refund   or   ‘money   back’   from   a   clothing   merchant,   would  you  use  this  money  for  rent  or  would  you  use  this  money  to  spend  in  the   same   category   (clothing)   where   the   money   was   returned   from?   Most   people   would  be  more  willing  to  use  this  money  towards  clothing.       Within   banking,   how   we   manage   our   money   between   checking,   savings,   loan  and  bill  accounts  displays  our  affinity  for  mental  accounting.    Why  do  we   leave  money  in  a  checking  account  and  also  have  high  interest  loans?    This  is  not   the  optimal  or  rational  decision  for  us.  However,  doing  something  different,  like   paying  off  all  high  interest  loans  and  having  very  little  money  in  checking,  may   feel  more  irresponsible.   In  both  of  these  cases,  people  are  relating  to  money  like  it  is  not  fungible.   Once   it   is   tagged   for   something,   we   view   it   as   difficult   to   use   it   for   another   purpose.   And,   if   it   is   not   tagged   for   something,   we   consider   it   open   hunting   season  and  are  more  willing  to  spend  it  freely.     There   are   two   different   solutions   we   could   use   to   deal   with   our   tendencies  around  mental  accounting.    First,  we  could  try  to  create  something   with   perfect   fungibility.   In   this   case,   we   would   ask   people   to   think   about   all   the   ways   they   could   use   the   money   before   they   spent   it,   so   as   to   not   pigeon-­â€hole   themselves   into   one   category.     The   other   solution   is   to   harness   our   tendency   to   do   mental   accounting   and   optimize   decision-­â€making   within   this   tendency.   The   behavioral  economics  approach  leans  toward  the  latter  solution.       13   .   HOW  WOULD  WE  DO  THIS?   To   increase   general   financial   outcomes   using   mental   accounting,   a   bank   would   want   to   create   multiple   spending   and   multiple   savings   accounts   with   different   names.   As   soon   as   someone   opened   a   checking   or   savings   accounts,   they   may   opt   into   having   additional   ‘fake-­â€accounts’   opened   for   them.     These   would   have   names   ranging   from   “Rainy   Day   Fund’   to   ‘Retirement   Fund’   to   ‘Kids   College’.     It   may   not   even   matter   if   these   were   actual   new   banking   accounts.   The   main   criteria   is   that   the   user   thinks   they   are   different   buckets.   The   visual   display   would   represent   this   mental   accounting   and   each   new   account   would   have  running  balances  and  goals.     SmartyPig  does  something  similar  to  this  with  their  goal  interface,  below.       WHAT  WOULD  THIS  LOOK  LIKE  AT  A  BANK?     Imagine   that   each   bank   creates   an   “Emergency   Fund”   account   within   a   user’s   checking  or  savings  account.    Anything  that  is  put  into  the  “Emergency  Fund”  is   visibly  hidden  from  checking  account  balance.    This  emergency  fund  is  treated   like  a  new  account  in  the  user’s  eyes.         14   . To   view   what   an   end   to   end   experience   could   look   like   for   a   bank’s   interface,   please  download  the  full  visual  mock  ups  on  irrationallabs.org.     15   . V. PRE-COMMITMENT……. Getting  people  to  pre-­â€commit  to  a  goal  can  help  with  self-­â€control.   In   one   study,   students   were   given   the   opportunity   to   create   their   own   deadlines  for  coursework.  The  students  could  pre-­â€commit  to  deadlines:  “I  want   to  turn  my  papers  in  at  different  points  during  the  semester”.  Or,  students  could   choose   to   turn   in   all   of   their   coursework   at   the   end   of   the   semester.   The   students   who   choose   to   set   deadlines   for   themselves   received   higher   grades.     Why?   It   seems   they   recognized   they   might   procrastinate.   By   making   this   pre-­â€ commitment   at   the   start,   they   were   able   to   spread   work   out   and   avoid   the   impulsive   (and   tempting)   distractions   that   may   have   plagued   the   other   students.  (Ariely,  Wertenbroch,  2002)7   Pre-­â€commitment   is   a   tool   to   help   people   follow   through   on   decisions.   One   does  this  by  making  a  decision  on  behalf  of  their  future  self.  Instead  of  relying   on  ourselves  to  be  great  people,  we  make  it  harder  for  our  future  self  to  mess   up.     In   this   case,   one   could   pre-­â€commit   to   re-­â€occurring   transfers,   increasing   transfer   amounts   over   time   or   answering   questions   before   withdrawing   from   their  savings.  These  commitments  would  be  made  at  the  start  and  be  thoughtful   and   rational   decisions.     Like   the   students   who   set   their   own   deadlines,   this   may   help  people  avoid  temptation  when  it  hits.                                                                                                                               7  http://dl1.cuni.cz/pluginfile.php/95343/mod_resource/content/0/Ariely_2002_procrastination.pdf   16   . VI. OPPORTUNITY COST NEGLECT…… “Thinking   of   money   in   the   right   way   is   thinking   of   opportunity   costs   in   the   right   way,   which   is   impossible,”   Dan   Ariely   said.   Instead   Ariely   said   humans   think  of  money  in  relative  terms.    Given  this,  how  could  we  present  the  decision   to  improve  your  financial  situation  in  relative  terms?  (Ariely,  2007)8   Let’s  just  look  at  the  actual  moment  when  someone  decides  to  save  or  not.   We  have  three  choices.     • We  can  do  ‘opt  out’  and  default  them  into  the  plan.  The  default  is  to  take   no  action  and  join  the  plan.         • We  can  do  ‘opt  in’  and  let  them  decide.  The  default  is  to  take  no  action   and  not  join  the  plan.       • We  can  present  the  choice  as  an  ‘active  choice’  and  they  must  choose  one   or  the  other.  There  is  no  set  default.     In   the   context   of   savings   or   other   big   decisions,   many   times   it   is   not   possible   to   design   an   ‘opt   out’   experience   due   to   legal   reasons,   even   though   research   proves   it   will   result   in   the   highest   amount   of   savings.   In   these   cases,   we   can   present   the   opportunity   cost   of   not   saving   by   using   the   ‘active   choice’   approach.  We  can  ask  the  users  to  view  the  decisions  in  a  relative  manner  and   actively  consider  the  alternatives.     For   example,   in   one   study   workers   at   a   certain   company   were   asked   to   make   an   active   decision   whether   or   not   to   join   their   savings   plan.     Employees   were   forced   to   check   a   “yes”   or   a   “no”   box   for   participation   in   this   plan.     In   this   study,   participation   rates   increased   by   about   28   percentage   points   compared   to                                                                                                                           8  http://danariely.com/the-­â€books/excerpted-­â€from-­â€chapter-­â€1-­â€%E2%80%93-­â€the-­â€truth-­â€about-­â€relativity-­â€2/   17   . standard   opt-­â€in   enrollment   procedures.   The   ability   to   understand   and   use   numbers  is  key  to  making  good  financial  decisions  (Choi  et  al  2005).       BELOW   ARE   THREE   IDEAS   FOR   ‘CALLS   TO   ACTION’   WITHIN   AN   APPLICATION   EXPERIENCE.     Bank  interface     “Yes,  activate  my  emergency  fund”  or  “No,  I  don’t  want  to  claim  my   Emergency  Fund  at  this  time”       Insurance  companies   “Insure  against  emergencies”  or  “Take  my  chances”           18   . VII. FAST FEEDBACK………….. If   you   binge   on   pizza   and   soda   and   then   1   week   later   gain   a   pound,   it   would   be   very   difficult   to   track   the   pound   back   to   your   pizza   and   soda   binge.     This   is   the   case   for   most   behaviors.   If   we   do   a   new   behavior,   like   junk   food   binging,  and  the  reward  (or  in  this  case  punishment)  for  the  behavior  is  delayed   at  all,  it  is  very  difficult  to  learn  from  your  actions.       Imagine   a   world   where   you   took   all   of   your   tests   throughout   the   school   year,  but  only  got  the  grades  for  them  at  the  end.  How  likely  is  it  that  your  study   methods   would   improve?   Unlikely.   There   would   be   no   feedback   after   a   test   indicating  that  your  studying  methods  worked  or  did  not  work.     How  should  we  apply  this  insight  to  savings?     If  we  only  reward  or  acknowledge  a  user  when  they  hit  a  large  milestone,   we  are  delaying  feedback  in  a  way  that  makes  it  difficult  to  learn.     So   while   our   savers   could   easily   look   down   at   their   bank   accounts   to   monitor  progress,  our  strategy  must  frequently  acknowledge  the  user’s  progress   after  they  do  something  positive  and  create  optimal  reinforcement.     Below   is   an   example   of   how   a   behaviorist   would   approach   showing   progress  on  an  emergency  savings  goal.     19   .     Another   example   of   “fast   feedback”   can   be   found   from   Quicken   Loans.     Every  time  a  user  does  something  (uploads  or  signs  documents)  they  get  a  super   cheerful   email   telling   them   that   they   are   great   for   doing   it,   and   hooraying   the   fact  that  they  are  one  step  closer  to  closing  on  the  loan.         20   . VIII. ACHIEVABLE GOALS ……..   IF  WE  SET  TOO  HIGH  OF  GOALS  THERE  IS  THE  CHANCE  THAT  PEOPLE  WILL  OPT   OUT   David  Liabson  showed  that  peer  information  can  generate  an  oppositional   reaction  than  expected.    He  asked  people  to  enroll  or  increase  their   contribution  to  their  401k  savings  plan  by  showing  them  what  their  peers   were  doing.     Interestingly,  this  did  little  to  change  behavior.    The  hypothesis?  When  a   savings  goal  feels  too  out  of  reach,  low-­â€saving  individuals  may  shift  away   from  the  peer-­â€norm  and  decrease  their  savings  relative  to  a  control  group   (Beshears  et  al  2011).    Translation:  When  peer  behavior  is  outside  of  an   attainable  realm  individuals  can  become  discouraged  and  less  likely  to  follow   the  norm  –  i.e.  increase  their  savings  rates.       IF  WE  SET  TOO  LOW  OF  GOALS  OR  DEFAULTS,  PEOPLE  MAY  ANCHOR  TO  THEM   WITHOUT  EVER  INCREASING   Madrian  and  Shea  (2001)  found  that  many  employees  who  were  set  at   default  rates  of  savings  of  2%  actually  continue  saving  at  the  default  rate  of  2%.   This  rate  is  obviously  far  too  low  to  provide  sufficient  funding  for  retirement.   However,  our  affinity  for  the  status  quo,  once  default  is  set,  is  very  hard  to   change.    This  means  that  savings  programs  may  actually  have  the  opposite   affect  than  intended.           21   . EXAMPLE:  APPLYING  THIS  TO  EMERGENCY  FUND  SET  UP   When  a  product  or  application  is  helping  a  user  set  up  and  maintain  an   emergency  fund,  how  should  the  contribution  amount  be  determined?  Below   are  three  things  to  consider  when  setting  savings  anchors:       • CHOICE  ARCHITECTURE:  Contribution  rates  should  be  different  for   different  customers,  depending  the  current  amount  the  customer  has  in   the  bank,  their  ability  and  desire  to  grow  a  rainy  day  fund  quickly  or  over   time.  The  customer  must  not  choose  this  amount,  it  should  be   recommended  to  them.       • PERIODICITY:  How  frequently  should  a  saver  transfer  money  to  savings?   The  question  of  weekly  vs  monthly  should  be  driven  by  when  the   customer  has  an  inflow  of  money.  If  they  have  an  inflow  of  money  (get   paid)  on  a  weekly  basis,  we  can  create  a  weekly  pull  into  the  emergency   fund.  If  it  is  monthly,  we  should  recommend  only  monthly  pulls  into  the   fund.       • COMMIT:  Customers  should  actively  commit  and  accept  the  savings   decision.  If  they  do  not,  they  may  not  internalize  the  decision  and  be   more  willing  to  make  changes  when  short  term  needs  arise.     22   .       WHAT  DO  WE  DO  WITH  THIS  INFORMATION?     Everything  is  testable.       Developing  A/B  tests  will  help  financial  companies  and  financial  institutions   help  their  clients  become  better  savers,  better  borrowers  –  and  still  be  good   customers  for  the  financial  institution.  “Anchors”  can  be  set  for  an  A  group,  and   differently  for  a  B  group  –  then  the  institution  watches  what  happens.    Do  you   prompt  someone  to  save  5%,  10%,  15%  of  their  monthly  income?     How  do  you  respond  to  good  behavior?  An  application  interface,  bank  teller   (or  an  ATM  machine)  may  be  prompted  to  congratulate  a  certain  group  of   clients  on  maintaining  their  commitment  to  savings;  while  another  group  does   not  receive  this  encouragement.     23   . IX. CHANGE THE MINDSET…. The  Duke  Center  for  Advanced  Hindsight,  led  by  Dan  Ariely  and  in   partnership  with  the  Gates  foundation,  ran  a  study  this  year  in  Kumor,  South   Africa  to  get  people  to  save.  (Akbas,  Ariely,  Robalino,  Weber,  2014)9   The  study  was  designed  for  people  in  extreme  poverty.  The  team  tested   many  different  hypotheses  on  what  would  increase  general  savings  within   families.     The  experiment  conditions  included:   • MONETARY  INCENTIVES.  It  provided  small  (10%)  and  large  (40%)  savings   matches  in  hopes  that  people  would  be  motivated  by  the  savings  matches   enough  to  become  active  savers.       • TRIGGERS.  SMS  reminders  were  sent  on  a  weekly  basis.  The  idea  with   reminders  is  that  it  is  not  that  people  don’t  want  to  save  or  cannot  save,  it   is  that  they  forget  to  save  on  a  regular  basis.       • ACCOUNTABILITY.  SMS  messages  were  sent  from  the  saver’s  children.   This  personalized  message  was  aiming  to  provide  a  deep  sense  of   accountability.       • IDENTITY.  The  Saver  was  given  a  cheap  (fake)  gold-­â€colored  coin.  The   people  in  this  condition  were  instructed  to  put  this  cheap  (fake)  gold   colored  coin  somewhere  special,  like  a  jewelry  box.  They  were  also  asked   to  scratch  off  a  tic  mark  (which  was  on  the  coin)  every  time  they  saved.   The  coin  had  a  set  number  of  tic  marks  on  it  that  the  Saver  could  easily   scratch  off.                                                                                                                               9  http://sites.duke.edu/merveakbas/files/2014/08/How-­â€to-­â€help-­â€the-­â€poor-­â€.pdf   24   . The  winning  intervention  creating  the  biggest  savers  was  a  surprise,  even  to   the  team  of  researchers.     The  gold-­â€colored  coin  beat  all  the  other  conditions.     How  did  this  happen?  How  did  a  fake  coin  beat  out  a  savings  match  to   increase  savings  rates?   At  this  point,  the  researchers  have  multiple  hypotheses  as  to  the  root  cause.   A  very  plausible  one,  which  we  put  forth  here,  is  that  the  coin  created  a  mindset   change  in  the  family.  This  physical  object  was  a  conversation  starter  around   savings.  Savings  now  had  a  constant  place  within  the  family  and  instilled  a   saving  identity.  The  researchers  believe  that  the  coin  helped  people  identify  as   people  who  save  money  and  consequently,  they  think,  increase  their  overall   savings.     In  the  US  banking  world  we  can  seek  to  achieve  a  similar  mindset  shift.  Some   ideas  to  encourage  this  mindset  shift:   • Create  a  Banking  PLUS  membership.  This  is  an  elite  membership  to  the   bank  that  everyone  is  given  upon  signing  up,  but  to  maintain  must   contribute  a  minimum  amount  to  their  savings  every  month.  Instead  of   monetary  benefits  to  membership,  the  user’s  debit  cards  are  branded  and   their  ATM  screen  is  branded  with  PLUS.       • In  the  progress  meter  online,  give  people  ‘badges’  who  have  activated  an   emergency  fund.  These  badges  will  help  create  an  identity  of  being  a   saver.  If  someone  has  not  activated  this  emergency  fund,  the  progress  bar   will  be  obviously  incomplete  and  the  badge  will  be  noticeably  missing.  On   the  flip-­â€side,  if  someone  does  activate  the  emergency  fund,  they  will   become  a  saver  and  the  progress  meter  and  badges  will  be  constant  in   their  account  view.             25   . X. PUT FRICTION IN THE RIGHT PLACES…………...   WHY  DO  PEOPLE  SAVE?     One  may  think  that  the  savers  have  a  strong  preference  for  the  future,  or   that  they  are  more  conservative  with  risk.  This  may  be  true  sometimes.   However,  experiments  have  found  that  a  better  indictor  for  likelihood  to  save  is   the  friction  required  to  complete  a  savings  action.       One  extreme  example  of  reluctance  to  join  an  attractive  retirement  plan   comes  from  the  United  Kingdom.  (Benartzi,  Thaler  2007)  Some  defined  benefit   plans  in  the  UK  do  not  require  any  employee  contributions  and  are  fully  paid  for   by  the  employer.  However,  to  join,  an  employee  must  take  some  action.  So  how   many  join  these  seemingly  magical  plans?  Only  half.  Half  of  employees  signed   up,  even  though  there  was  nothing  but  upside  for  them.  Why?  One  can  only   guess  it  takes  too  much  work.     In  another  case,  when  automatic  enrollment  was  adopted,  enrollment  of   new  employees  jumped  to  90  percent  immediately  –  from  a  low  20  percent.   Automatic  enrollment  has  two  effects:  participants  join  sooner,  and  more   participants  join  eventually.   Given  these  examples,  we  want  to  make  it  frictionless  to  create  a  savings   plan.  However  in  the  current  paradigm,  all  products  and  applications  have  the   concept  of  ‘creating  an  account’.    In  the  banking  world,  you  are  asked  to  ‘open   an  account.’   To  the  person  who  doesn’t  like  to  put  in  ANY  effort,  ‘opening  an  account’   feels  like  a  lot  of  effort!    For  products  and  services  that  are  trying  to  get  people   to  set  up  accounts,  we  recommend  avoiding  the  concept  of  ‘opening’  an   26   . account.    Instead,  we  can  use  the  word,  ‘Activate’  to  signal  that  the  person   already  has  the  account  and  they  just  need  to  (easily)  turn  it  on.       But,  on  the  reverse  side,  adding  friction  is  a  good  thing  when  it  prevents  a   behavior  that  we  don’t  want.     For  example,  how  do  you  prevent  people  from  taking  money  out  of  their   savings  plan?    In  this  case,  we  do  want  to  add  friction.  See  below  for  the  type  of   questions  we  can  add  before  the  user  is  able  to  take  out  their  savings.  This   increased  the  amount  of  effort  needed  to  perform  the  desired  action  and   thereby  may  decrease  the  number  of  people  who  take  this  action  out  of   impulse.         27   .   XI. DISCUSSION.....…………... These  are  complex  decisions.    Should  we  just  let  users  decide  how  much   to  save  by  themselves  and  not  interfere  with  the  decision  making  process  at  all?           Benartzi  and  Thaler  tested  this  question  in  a  clever  manner.  They  decided   to  study  the  very  people  who  actively  chose  to  opt  out  of  a  professionally-­â€made   investment  plan  in  favor  of  creating  their  own  custom  investment  plan.       To  study  these  people,  they  asked  them  to  rate  the  returns  of  both  plans  -­â€   theirs  and  the  professionally-­â€designed  plan.    The  question:  After  seeing  the   returns,  which  one  do  you  like  more?     Self-­â€constructed  portfolios  received  the  lowest  average  rating,  2.75,  and   the  professionally-­â€managed  portfolios  received  the  significantly  higher  mean-­â€ rating  of  3.50.   This  gives  us  an  indication  that  even  the  people  who  have  the  initial   desire  to  control  their  retirement  liked  their  portfolio  less  than  the   professionally  managed  portfolio  plan  and  they  agree  that  they  can  benefit   from  additional  help.       This  brings  out  a  bigger  point  to  our  discussion.     If  we  let  people  do  as  they  please  with  their  finances,  we  should  not  be   surprised  if  they  fail.  The  idea  of  letting  people  make  decisions  without  any  help   is  proven  to  be  misguided.  The  evidence,  time  and  again,  proves  that  this  is   stressful  for  the  consumer  and  is  very  likely  to  put  them  in  an  objectively  worse   financial  situation.         28   . TO  DRIVE  THE  SUCCESS  OF  A  SAVINGS  PLATFORM,  WE  ENCOURAGE  PRODUCT   DESIGNERS  AND  FINANCIAL  LEADERS  TO  CREATE:   1) 2) 3) 4) 5) 6) 7) 8)   Rules  of  thumb       Mental  Accounting     Pre  –  Commitment     Clear  Opportunity  Costs     Fast  Feedback     Achievable  Goals     Mindset  Changes     Friction  in  the  right  places     29   . XII. WORKS CITED...…......... Ariely  (2007)  Predictably  Irrational.     Akbas,  Ariely,  Robalino,  Weber.  (2014).  How  to  help  the  poor  to  save  a  bit:     Evidence  from  a  field  experiment  in  Kenya.  Department  of  Economics,   Duke  University.     Ariely,  Wertenbroch.  (2002)  Procrastination,  Deadlines,  Performance:  Self     Control  by  Pre-­â€commitment.  Psychological  Science.  Vol  13,  No.  3.     Benartzi,  S.,  &  Thaler,  R.  H.  (1999).  Risk  Aversion  or  Myopia?  Choices  in   Repeated  Gambles  and  Retirement  Investments.  Management  Science,   45(3),  364-­â€381.   Benartzi,  S.,  &  Thaler,  R.  H.  (2007).  Heuristics  and  Biases  in  Retirement  Savings   Behavior.  Journal  Of  Economic  Perspectives,  21(3),  81-­â€104.   Beshears,  J.,  Choi,  J.  J.,  Laibson,  D.,  Madrian,  B.  C.,  &  Milkman,  K.  L.  (2011).  The   Effect  of  Providing  Peer  Information  on  Retirement  Savings  Decisions.   Choi,  J.  J.,  Laibson,  D.,  Madrian,  B.,  &  Metrick,  A.  (2005).  Optimal  Defaults  and   Active  Decisions.   Consumer  Financial  Protection  Bureau.  (2013).  Navigating  the  Market.   Consumer  Financial  Protection  Bureau.  Retrieved  from   http://files.consumerfinance.gov/f/201311_cfpb_navigating-­â€the-­â€market-­â€ final.pdf   Fernandes,  D.,  Lynch  Jr,  J.  G.,  &  Netemeyer,  R.  G.  (2014).  Financial  Literacy,   Financial  Education,  and  Downstream  Financial  Behaviors.  Management   Science.   Helman,  R.  (2014).  The  2014  Retirement  Confidence  Survey:  Confidence   Rebounds—for  Those  With  Retirement  Plans.  EBRI  Issue  Brief,  (397).   30   . Ideas42.  Financial  Heuristics.  Ideas42.org.  Retrieved  from   http://www.ideas42.org/financial-­â€heuristics/   Kast,  Meier,  Pomeranz  (2012).  Under-­â€savers  anonymous:  Evidence  on  self  help   groups  and  peer  pressure  as  a  savings  commitment  device.  National  Bureau  of   Economics  research.       Lusardi,  A.,  &  Mitchell,  O.  S.  (2007).  Baby  boomer  retirement  security:  The  roles   of  planning,  financial  literacy,  and  housing  wealth.  Journal  of  monetary   Economics,  54(1),  205-­â€224.   Lusardi,  A.,  &  Tufano,  P.  (2009).  Debt  literacy,  financial  experiences,  and   overindebtedness  (No.  w14808).  National  Bureau  of  Economic  Research.   Madrian,  B.  C.,  &  Shea,  D.  F.  (2001).  The  Power  of  Suggestion:  Inertia  in  401(k)   Participation  and  Savings  Behavior.  Quarterly  Journal  Of  Economics,   116(4),  1149-­â€1187.   Thaler,  R.  (1985).  Mental  accounting  and  consumer  choice.  Marketing  science,   4(3),  199-­â€214.       31   .