Using a Digitization Index to measure the Economic and ... · Using a Digitization Index to measure...

44
Using a Digitization Index to measure the Economic and Social Impact of Digital Agendas (*) Euro CPR 2013 A Digital Agenda in search of evidence: Issues and Trends Brussels, Belgium March 21, 2013 Dr. Raúl L. Katz, Adjunct Professor, Division of Finance and Economics, and Director, Business Strategy Research, Columbia Institute of Tele- information (*) This presentation is based on a research jointly developed with Dr. Pantelis Koutroumpis.

Transcript of Using a Digitization Index to measure the Economic and ... · Using a Digitization Index to measure...

Using a Digitization Index to measure the Economic and Social Impact of Digital Agendas (*)

Euro CPR 2013 A Digital Agenda in search of evidence:

Issues and Trends Brussels, Belgium March 21, 2013

Dr. Raúl L. Katz, Adjunct Professor, Division of Finance and Economics, and Director, Business Strategy Research, Columbia Institute of Tele-information

(*) This presentation is based on a research jointly developed with Dr. Pantelis Koutroumpis.

Agenda

● Measuring digitization

● Assessing digitization’s economic impact

● Digitization in Europe

● Elements of a future agenda

2

3

Digitization is the capacity to use digital technologies to generate, process, share and transact information

●  At  the  most  basic  level,  digi2za2on  is  the  process  of  conver2ng  analog  informa2on  into  a  digital  format  

●  In  a  broader  context,  digi2za2on  is  defined  as  the  social  transforma2on  triggered  by  the  massive  adop2on  of  digital  technologies  to  generate,  process,  share  and  transact  informa2on  

●  Digi2za2on  builds  on  the  evolu2on  of  network  access  technologies,  semiconductor  technologies,  and  so@ware  engineering  

●  Leverages  the  spillover  effects  resul2ng  from  their  use  (common  plaDorms  for  applica2on  development,  e-­‐government  services,  e-­‐commerce,  social  networks,  and  availability  of  online  informa2on)  

 

To achieve a significant impact, digitization has to be widely diffused within the economic and social fabric of a given nation

●  Adopted  at  three  levels  –  U2lized  by  individuals,  economic  enterprises  and  socie2es  –  Embedded  in  processes  of  delivery  of  goods  and  services  –  Relied  upon  to  deliver  public  services    

●  For  this  condi2on  to  occur,  digi2za2on  has  to  fulfill  several  condi2ons  –  Affordable  to  allow  scalable  impact  –  Ubiquitous  reaching  most  popula2on  of  a  na2onal  territory  –  Accessible  by  mul2ple  fixed  and  mobile  voice  and  data  devices  –  Reliable,  providing  sufficient  capacity  to  deliver  vast  amounts  of  

informa2on  at  speeds  that  do  not  hinder  their  effec2ve  use  

4

5

The purpose of this research was to develop a digitization index and measure its contribution to economic and social development

●  Star2ng  premise:  most  research  measuring  social  and  economic  impact  of  ICT  focuses  on  discrete  technology  plaDorms,  such  as  mobile  penetra2on,  access  to  the  Internet  and  broadband  adop2on  

 ●  Holis2c  adop2on  and  usage  of  ICT  results  in  enhanced  effects  that  go  

beyond  the  contribu2on  of  specific  plaDorms  

●  The  transi2on  to  digitally-­‐intensive  socie2es  should  be  assessed  across  a  mul2ple  set  of  metrics,  capturing  not  only  penetra'on,  but  also  usage  of  these  technologies  in  order  to  capture  the  full  impact  of  digi2za2on

6

A composite index comprising 23 indicators was developed to measure the level of digitization of a given country

DIGITIZATION INDEX

Data to develop the index was compiled from multiple sources

7

NAME  OF  INDICATOR   SOURCE  

Residential fixed line tariff adjusted for GDP per capita ITU Residential fixed line connection fee adjusted for GDP per capita ITU Mobile cellular prepaid tariff adjusted for GDP/capita ITU Mobile cellular prepaid connection fee adjusted for GDP per capita ITU Fixed broadband Internet access tariff adjusted for GDP per capita ITU Investment per telecom subscriber (mobile, broadband and fixed) World Bank Fixed Broadband penetration ITU Mobile Phone penetration (2010) ITU Population covered by mobile cellular network ITU Percentage of population using a PC (2010) ITU 3G Penetration (2Q 11) Wireless Intelligence International Internet bandwidth (bits/second/internet user) ITU Broadband speeds (% above 2 Mbps) Akamai Internet retail (Retail internet as percentage of total retail) Euromonitor E-government Web measure index UN Percentage of individuals (users) using the internet (2010)` ITU Data as a percentage of wireless ARPU (4Q10) Wireless Intelligence Dominant Social Network Unique Visitors per month Per Capita Internet World Stats SMS Usage (Average SMS sent by consumers) Wireless Intelligence Engineers (Engineers as a percentage of total population) World Bank Skilled Labor (Labor force with more than a secondary education as a percentage of the total labor force) World Bank

8

The index was constructed following a typical methodology for composite index validity assessment

●  Define  the  theore2cal  framework  of  the  index  and  select  variables  ●  Each  sub-­‐index  was  normalized,  by  using  the  mean  and  one  standard  

devia2on,  and  cropping  the  extremes  –  Affordability  sub-­‐index:  the  inverse  of  the  maximum  is  used  to  cap  

it  –  Mobile  penetra2on:  was  capped  at  a  maximum  of  100%  to  prevent  

over-­‐weigh2ng  ●  For  each  of  the  six  components  a  minimum  of  subcomponents  is  

assigned  depending  on  the  scarcity  of  the  available  informa2on  ●  For  the  Index  calcula2on,  a  minimum  of  four  components  is  required  ●  Correla2ons  were  ini2ally  run  between  the  digi2za2on  index  and  other  

technology  indices  to  test  its  ranking  value  –  Network  Readiness  Index  (WEF)  –  ICT  Opportunity  Index  (ITU)  –  Digital  Opportunity  Index  (ITU)  

Finally, the Kaiser-Meyer-Olkin measure of sampling adequacy was ran, indicating that the index is statistically sound

9

Scale reliability coefficient: 0.9416Number of items in the scale: 23Average interitem covariance: 234.5332

Reversed items: a1 a2 a3 a4 a5Test scale = mean(unstandardized items)

Overall 0.8202 human 0.8311 usage 0.8394 capacity 0.8154 networkacc~s 0.7530 infrastruc~e 0.8741 affordabil~y 0.8854 Variable kmo

Kaiser-Meyer-Olkin measure of sampling adequacyScale reliability coefficient: 0.8640Number of items in the scale: 6Average interitem covariance: 198.3664

Reversed item: affordabilityTest scale = mean(unstandardized items)

The  es2mated  sta2s2cs  derive  from  factor  analysis:    •  KMO  measures  how  dis2nct  the  factors  (components  of  the  

index)  are  so  that  they  do  not  over-­‐iden2fy  latent  phenomena.  All  factor  es2mates  need  to  be  higher  than  0.60  and  the  overall  KMO>0.8  

•  The  Cronbach  coefficient  alpha  is  the  most  common  es2mate  of  internal  consistency  of  items  in  a  model  or  survey.  It  assesses  how  well  a  set  of  sub-­‐indicators  measures  a  single  one-­‐dimensional  object.  Reliable  A-­‐threshold  >0.8  

•  With  KMO=0.82  and  Alpha23=0.94  Alpha6=0.86  the  Digi2za2on  Index  is  sta2s2cally  sound  

10

The 2011 Digitization Index was calculated for 150 countries, revealing four developmental stages

●  Index  computed  for  150  countries  and  the  period  2004-­‐2011  

●  Four  clusters  iden2fied:  

–  Advanced  (Index>50)  –  Transi'onal  (Index>35&<50)  –  Emerging  (Index>20&<35)  –  Constrained  (Index<20)    

•  Barbados((•  Georgia((•  Philippines(•  Macao(•  Seychelles(•  Bosnia(and(Herzegovina(

•  Ecuador(•  Azerbaijan(•  Venezuela(•  Botswana(•  Peru(•  South(Africa(•  Algeria(•  China(•  Lebanon(•  Sudan((

•  Somalia(•  Uganda((•  Afghanistan((•  Burkina(Faso((•  Zimbabwe((•  Ethiopia((•  Mozambique((•  Solomon(Islands((

•  Madagascar((•  Comoros((•  D.((R.(Congo(•  Malawi((•  Lesotho((•  S(Tome(&(Principe((

•  C.A.(Republic((•  Mali((•  Burundi((•  Niger((

0

10

20

30

40

50

60

70

TRANSITIONAL(

•  Aruba •  Malta •  St. Lucia •  Greece •  Serbia •  Poland •  Bulgaria •  Saudi

Arabia •  Cyprus •  Slovak

Republic •  Kuwait •  Latvia •  Qatar •  Taiwan •  Chile ((

•  Liechtenstein •  Estonia •  Puerto Rico •  Malaysia •  Panama •  Hungary •  St. Vincent

and the Grenadines

•  Mauritius •  Mongolia •  Uruguay •  Croatia •  Argentina •  Kazakhstan •  Ukraine •  Turkey

ADVANCED(

•  Norway(•  Denmark(•  Switzerland(•  United(Kingdom(

•  Luxembourg(•  Iceland(•  Finland(•  Sweden(•  Korea(•  Hong(Kong((•  United(States(•  Canada(•  Belgium(•  France(•  Singapore(•  Australia(•  Japan((

EMERGING(CONSTRAINED(

•  Portugal(•  Austria(•  Germany(•  Israel(•  Ireland(•  New(Zealand(

•  Netherlands(•  Slovenia(•  Czech(Republic(

•  Spain(•  Romania(•  Italy(•  Belarus(•  Lithuania(•  UAE(•  Russia((

•  Dominican(Republic((

•  Indonesia((•  Maldives((•  Tunisia((•  El(Salvador((•  Gabon((•  Jordan((•  Trinidad(&(Tobago((

•  Albania((•  Armenia((•  Paraguay((•  Thailand((•  Jamaica((•  Egypt((•  Libya((•  Vietnam((

•  Kyrgyz(Republic((

•  Cuba((•  Bolivia((•  Tonga((((•  Iraq((•  Bangladesh(•  Cote(d'Ivoire(•  Swaziland(•  Nicaragua(•  Cambodia(•  Kenya((•  DjibouU((•  Yemen((•  Nigeria((•  Ghana(•  Tajikistan(•  Benin(•  Tuvalu((•  Angola(

•  Mauritania((•  TimorWLeste((•  Guinea((•  Lao((•  Cameroon((•  Zambia((•  Nepal((•  Turkmenistan((•  Eritrea((•  Rwanda((•  Chad((•  The(Gambia((•  Papua(New(Guinea((

•  Togo((•  Tanzania((•  Senegal((•  Vanuatu(•  Myanmar(•  Congo(

•  India((•  Morocco((•  Sri(Lanka((•  Cape(Verde((•  Bhutan((•  Uzbekistan((•  Pakistan((•  Syrian(Arab(Republic((

•  Honduras((•  Guyana((•  Belize((•  New(Caledonia((

•  Namibia((•  Fiji((•  Guatemala((•  Virgin(Islands((

•  Bermuda •  Moldova •  Brunei •  San Marino •  Oman •  Colombia •  Iran •  Antigua and

Barbuda •  Bahrain •  Costa Rica •  Montenegro •  Mexico •  Andorra •  Brazil •  Macedonia

When ranking countries for each of the six sub-components, we determined that access and affordability are less of a world problem

11

•  For  all  countries,  normalized  usage  sub-­‐index  never  matches  the  levels  of  access  sub-­‐indices,  which  indicate  a  big  challenge  across  the  world  

•  For  all  OECD  and  middle  income  countries,  the  sub-­‐indices  affordability  and  network  access  tend  to  be  consistently  above  the  digi2za2on  index  indica2ng  that  countries  have  addressed  the  access  challenge  

•  The  affordability  and  capacity  sub-­‐indices  tend  to  rapidly  drop  at  low  GDP  levels,  indica2ng  a  big  gap  between  mature  and  low  income  countries  

0

20

40

60

80

100

120No

rway

Luxembo

urg

United

France

Spain

Greece

Macao,  China

Mauritius

Croatia

Antigua  an

d

Mexico

Serbia

T.F.Y.R.

Ecuado

r

Mald

ives

Gabo

n

Mon

golia

Pakis

tan

Thailand

Kyrgyzsta

n

Cambo

dia

Zambia

Tanzania

Cuba

Cameroo

n

Eritrea

Ethiop

ia

AFFORDABILITY INFRASTRUCTURE  RELIABILITY NETWORK  ACCESS

CAPACITY USAGE HUMAN  CAPITAL

TOTAL

In addition to assessing digitization development paths, we tracked its evolution over time for selected countries

●  Constructed  a  2me  series  of  digi2za2on  for  18  countries  between  1995  and  2011  

●  Assessed  the  evolu2on  of  the  index  akemp2ng  to  determine  idiosyncra2c  country  paths  to  digi2za2on  –  Emerging  countries  undergo  quantum  leap  changes  in  digi2za2on  

triggered  by  key  policy  ini2a2ves  –  Mature  countries  exhibit  a  consistent,  yet  gradual,  change  in  digi2za2on  

performance  

●  Analyzed  changes  in  the  index  in  an  akempt  to  iden2fy  specific  events  or  policies  that  have  triggered  a  change  at  a  specific  point  in  2me  

12

0

10

20

30

40

50

60

70

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

U.S.

13

Most industrialized countries have consistently increased their digitization level over the past fifteen years, albeit at different rates

13 13

USA leading in 1997(pre dot com bubble) and now gradually making again its

way to the top

DIGITIZATION INDEX (1995-2011)

0

10

20

30

40

50

60

70

80

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

U.S. Norway Sweden

14

Scandinavian countries have significantly changed their digitization development path

14 14

Norway and Sweden leapfrogging through

systematic adoption and timely launch of network

access, skills and services

DIGITIZATION INDEX (1995-2011)

0

10

20

30

40

50

60

70

80

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Japan S. Korea U.S. Norway Sweden

15

In Asia, Korea achieved the same leapfrogging move in the late nineties

15 15

S. Korea government intervention for 100% broadband coverage

DIGITIZATION INDEX (1995-2011)

0

10

20

30

40

50

60

70

80

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Japan Singapore S. Korea Australia U.S. U.K. Germany Norway Sweden

16

Most industrialized countries have been converging over the past fifteen years

16 16

DIGITIZATION INDEX (1995-2011)

0

5

10

15

20

25

30

35

40

1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011

Brazil Egypt China India

The digitization development path of emerging countries depends on specific public policies

India very slow after 2007-08 rise, now trailing most emerging countries

China consistently improving based on

infrastructure deployment plans

17 17 17

Brazil leapfrogs after privatization/

liberalization

Brazil rebounds based on income

redistribution policies

DIGITIZATION INDEX (1995-2011)

In sum, data indicates different development paths towards digitization

●  Mature  countries  follow  a  gradual  progression  towards  digi2za2on  –  Ac2ve  government  interven2on  accelerates  development  (Korea,  

Norway)  

●  Some  emerging  countries  undergo  quantum  leap  changes  (25  points  in  five  years)  in  digi2za2on  triggered  by  specific  policy  ini2a2ves  –  Telecom  market  liberaliza2on  with  spill-­‐over  impact  on  the  ICT  eco-­‐

system  –  A  combina2on  of  ac2ve  government  involvement  and  private  sector  

par2cipa2on  –  Centralized  state  planning  

18

Agenda

● Measuring digitization

● Assessing digitization’s economic impact

● Digitization in Europe

● Elements of a future agenda

19

To understand the economic impact of digitization, a correlational view of the index and individual income was first developed

20

Albania

Albania

AlbaniaAngola

Angola

AngolaArgentina

Argentina

ArgentinaAustralia

Australia

AustraliaAustria

Austria

AustriaAzerbaijan

Azerbaijan

AzerbaijanBangladesh

Bangladesh

BangladeshBarbados

Barbados

BarbadosBelarus

Belarus

BelarusBelgium

Belgium

BelgiumBelize

Belize

BelizeBenin

Benin

BeninBhutan

Bhutan

BhutanBolivia

Bolivia

BoliviaBosnia and Herzegovina

Bosnia and Herzegovina

Bosnia and HerzegovinaBrazil

Brazil

BrazilBrunei Darussalam

Brunei Darussalam

Brunei DarussalamBulgaria

Bulgaria

BulgariaBurkina Faso

Burkina Faso

Burkina FasoCameroon

Cameroon

CameroonCanada

Canada

CanadaChina

China

ChinaColombia

Colombia

ColombiaComoros

Comoros

ComorosCosta Rica

Costa Rica

Costa RicaCote d'Ivoire

Cote d'Ivoire

Cote d'IvoireCyprus

Cyprus

CyprusCzech Republic

Czech Republic

Czech RepublicDenmark

Denmark

DenmarkDjibouti

Djibouti

DjiboutiDominican Rep.

Dominican Rep.

Dominican Rep.Ecuador

Ecuador

EcuadorEgypt

Egypt

EgyptEl Salvador

El Salvador

El SalvadorEstonia

Estonia

EstoniaGermany

Germany

GermanyGreece

Greece

GreeceGuatemala

Guatemala

GuatemalaGuyana

Guyana

GuyanaHong Kong, China

Hong Kong, China

Hong Kong, ChinaHungary

Hungary

HungaryIndia

India

IndiaIndonesia

Indonesia

IndonesiaIran (Islamic Rep. of)

Iran (Islamic Rep. of)

Iran (Islamic Rep. of)Italy

Italy

ItalyJapan

Japan

JapanJordan

Jordan

JordanKazakhstan

Kazakhstan

KazakhstanKenya

Kenya

KenyaKorea (Rep. of)

Korea (Rep. of)

Korea (Rep. of)Kyrgyzstan

Kyrgyzstan

KyrgyzstanLao P.D.R.

Lao P.D.R.

Lao P.D.R.Latvia

Latvia

LatviaLebanon

Lebanon

LebanonLesotho

Lesotho

LesothoLithuania

Lithuania

LithuaniaLuxembourg

Luxembourg

LuxembourgMacao, China

Macao, China

Macao, ChinaMadagascar

Madagascar

MadagascarMalaysia

Malaysia

MalaysiaMali

Mali

MaliMalta

Malta

MaltaMauritania

Mauritania

MauritaniaMexico

Mexico

MexicoMoldova

Moldova

MoldovaMongolia

Mongolia

MongoliaMontenegro

Montenegro

MontenegroMorocco

Morocco

MoroccoMozambique

Mozambique

MozambiqueNamibia

Namibia

NamibiaNepal

Nepal

NepalNew Zealand

New Zealand

New ZealandNiger

Niger

NigerNorway

Norway

NorwayOman

Oman

OmanPakistan

Pakistan

PakistanPanama

Panama

PanamaParaguay

Paraguay

ParaguayPeru

Peru

PeruPhilippines

Philippines

PhilippinesPoland

Poland

PolandQatar

Qatar

QatarRomania

Romania

RomaniaRussia

Russia

RussiaRwanda

Rwanda

RwandaSaudi Arabia

Saudi Arabia

Saudi ArabiaSenegal

Senegal

SenegalSerbia

Serbia

SerbiaSeychelles

Seychelles

SeychellesSlovak Republic

Slovak Republic

Slovak RepublicSlovenia

Slovenia

SloveniaSpain

Spain

SpainSri Lanka

Sri Lanka

Sri LankaSudan

Sudan

SudanSweden

Sweden

SwedenSwitzerland

Switzerland

SwitzerlandSyria

Syria

SyriaT.F.Y.R. Macedonia

T.F.Y.R. Macedonia

T.F.Y.R. MacedoniaThailand

Thailand

ThailandTogo

Togo

TogoTunisia

Tunisia

TunisiaTurkey

Turkey

TurkeyUganda

Uganda

UgandaUkraine

Ukraine

UkraineUnited States

United States

United StatesUruguay

Uruguay

UruguayUzbekistan

Uzbekistan

UzbekistanVenezuela

Venezuela

VenezuelaYemen

Yemen

Yemen00

02020

204040

406060

60Digitization IndexDi

gitiza

tion

Index

Digitization Index20000

20000

2000040000

40000

4000080000

80000

80000log of GDP per Capita (USD, 2000 constant)

log of GDP per Capita (USD, 2000 constant)

log of GDP per Capita (USD, 2000 constant)

21

Extending this hypothesis, an econometric model was built to test the contribution of GDP to economic growth

●  Simple  Cobb-­‐Douglas  form:  

Y=A(t)K1+BL1      where:    ● A(t)  represents  the  level  of  technology  progress  (in  our  case  the  Digi2za2on  Index)  

●  K  corresponds  to  the  fixed  capital  forma2on  

●  L  to  the  labor  force  

log(GDPit)=a1log(Kit)+a2log(Lit)+a3log(Dit)+εit

Digitization is found to have a positive and significant effect on economic output

●  The  model  states  that  10  point  increase  on  digi2za2on  increases  GDP  per  capita  by  0.76%  

●  The  index  is  a  weighted  average  of  different  indicators  that  might  be  endongenous  to  GDP,  like  broadband  and  wireless  penetra2on;  however,  given  their  overall  impact  of  these  on  the  metric  is  insignificant  (5%)  

●  Furthermore,  by  controlling  for  country  and  year  fixed  effects,  some  poten2al  endogeneity  problems  are  mi2gated  

●  Capital  forma2on  and  labor  contribu2on  are  posi2ve  and  significant,  while  digi2za2on  is  found  to  have  a  posi2ve  and  significant  effect  at  the  1%  level,  indica2ng  a  strong  effect  on  economic  output,  confirming  the  correla2onal  view  

●  Annual  Growth  Rate  (CAGR)  akributed  to  digi2za2on  is    derived  from  the  following  formula:  

 

●  A  ten  point  increase  in  the  Digi2za2on  Index  has  approximately  a  2.84%  impact  on  GDP  for  the  period  2004-­‐2011  resul2ng  in  an  annualized  effect  of  0.40%    

22

23

Furthermore, digitization has a larger contribution to GDP than stand-alone technologies

23

0.23

0.14

0.08

0.26 0.24

0.16

0.08 0.09

0.17

0.11

0.20

0.76

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Contribu'on  to  GDP  

growth  of  10  point  increase  in  variable  

Broadband  Studies  

10  POINT  INCREASE  IN  DIGITIZATION  YIELDS  0.76%  

INCREASE  IN  GDP  PER  CAPITA  

DIGITIZATION AND ECONOMIC DEVELOPMENT

§  This  s2pulates  that  full  economic  impact  ICT  is  achieved  through  the  cumula2ve  adop2on  of  all  technologies,  in  addi2on  to  the  assimila2on  and  usage  in  the  produc2on  and  social  fabric  

§  Achieving  broadband  penetra2on  is  only  one  aspect  of  required  policies;  maximiza2on  of  economic  impact  can  only  be  achieved  through  a  holis2c  set  of  policies  ranging  from  telecoms  to  compu2ng  to  adop2on  of  internet  and  eCommerce  

Mobile  Studies  

Additionally, digitization exhibits increasing returns to scale, where returns increase after an index score of 30

24

0

0

020000

2000

0

2000040000

4000

0

4000060000

6000

0

6000080000

8000

0

800000

0

020000

2000

0

2000040000

4000

0

4000060000

6000

0

6000080000

8000

0

800000

0

020000

2000

0

2000040000

4000

0

4000060000

6000

0

6000080000

8000

0

800000

0

020

20

2040

40

4060

60

600

0

020

20

2040

40

4060

60

600

0

020

20

2040

40

4060

60

602004

2004

20042005

2005

20052006

2006

20062007

2007

20072008

2008

20082009

2009

20092010

2010

2010Digitization Index

Digitization Index

Digitization IndexGDP per capita

GDP

per c

apita

GDP per capitaGraphs by Year

Graphs by Year

Graphs by Year

Countries  with  lower  scores  are  o@en  the  ones  that  lack  basic  access,  skills  and  usage  that  would  prevent  them  from  experiencing  important  effects  on  their  economies      

25

The increasing returns to scale hypothesis was also proven econometrically at lower levels of digitization (*)

25

•  Advanced:  2.26%  compound  impact  on  GDP  for  the  period  2004-­‐2011  resul2ng  on  an  annualized  effect  of  0.32%  

•  Transi2onal:  2.59%  compound  impact  on  GDP  for  the  period  2004-­‐2011  resul2ng  on  an  annualized  effect  of  0.37%  

•  Constrained:  2.66%  compound  impact  on  GDP  for  the  period  2004-­‐2011  resul2ng  on  an  annualized  effect  of  0.37%  

•  Emerging:  2.44%  compound  impact  on  GDP  for  the  period  2004-­‐2011  resul2ng  on  an  annualized  effect  of  0.34%  

(*) The sample is broken into four different equally populated clusters. Four dummy variables are created (high, medium, low and very low) that take the value of 1 if the country is within the Digitization scores of interest or 0 if not. For the advanced cluster the threshold is 50, for the transitional 35-50, for the emerging 20-35 and for the constrained 0-20.

26

Beyond its contribution to economic output, digitization also has a positive impact on employment creation

●  Simple  model  links  unemployment  rates  with  exis2ng  infrastructure,  income,  educa2on  levels,  total  exports  as  a  percent  of  GDP  and  the  credit  performance:  

 Uit=b1Dit+b2Kit+b3Eduit+b4GDPCit

+b5Expit+b6Crit+εit    ● Model  controls  for  country  and  year  fixed  effects  

The econometric impact model on job creation is also quite robust

● All  control    variables  have  the  right  sign  ● All  canonical  variables  are  sta2s2cally  significant  ●  The  model  is  very  stable,  meaning  that  signs  and  sta2s2cal  significance  of  each  of  the  coefficients  are  stable  even  when  the  specifica2ons  are  changed  

●  By  including  GDP  per  capita  in  the  variables,  we  negate  the  possibility  that  the  index  is  working  as  a  proxy  for  level  of  development  

● As  in  the  case  of  GDP  growth,  digi2za2on  has  a  higher  impact  on  job  crea2on  than  broadband  –  Full  deployment  and  assimila2on  of  ICT  has  a  much  larger  impact  on  employment  because  it  contributes  to  more  jobs  in  the  ICT  sector  (so@ware  development,  Business  Process  Outsourcing,  equipment  manufacturing  and  parts  supplies)  

–  In  addi2on,  the  impact  of  assimila2on  of  ICT  through  enhanced  usage  has  spill-­‐over  impact  on  other  sectors  of  the  economy  (in  par2cular,  trade,  financial  services,  health  care)  

27

One of the most interesting and yet unexplored parameters of Digitization is the link to overall societal welfare

●  This  suggests  a  hypothesis  that  Digi2za2on  has  a  direct  effect  on  the  overall  happiness  and  life  sa2sfac2on  that  people  earn  from  the  capaci2es  and  capabili2es  of  engaging  in  digital  technology  plaDorms  

●  A  classic  counter-­‐argument  stems  from  the  causal  link  between  the  life  sa2sfac2on  and  Digi2za2on,  manifes2ng  that  people  might  self-­‐select  to  be  in  a  country  or  regional  context  with  higher  provisions  of  digital  services  rather  than  being  the  subjects  of  various  offerings  

●  Nevertheless  for  the  vast  majority  of  popula2on,  one  would  infer  that  people  would  not  migrate  for  an  abundance  of  Digi2za2on  services  and  technologies  

●  For  this  purpose  we  choose  not  to  model  this  rela2onship  in  a  strict  quan2ta2ve  manner  but  prefer  to  highlight  it  in  a  correla2ve  approach  

28

Digitization appears to be correlated with life satisfaction and well being

10

10

1020

20

2030

30

3040

40

4050

50

5010

10

1020

20

2030

30

3040

40

4050

50

502

2

22.5

2.5

2.53

3

33.5

3.5

3.52

2

22.5

2.5

2.53

3

33.5

3.5

3.52

2

22.5

2.5

2.53

3

33.5

3.5

3.52005

2005

20052006

2006

20062007

2007

20072008

2008

20082009

2009

20092010

2010

2010Digitization Index

Digitiz

ation

Inde

x

Digitization IndexLife Satisfaction

Life Satisfaction

Life SatisfactionGraphs by Year

Graphs by Year

Graphs by Year •  Digi2za2on  &  Life  sa2sfac2on:  limited  effects  from  20  to  40,  consistent  improvement  from  40  to  60  

•  At  lower  levels  of  development  is  more  related  to  the  sa2sfac2on  of  basic  needs  (such  as  food  and  shelter  in  the  Maslow  Scale),  while  at  higher  levels  of  development,  once  these  needs  are  addressed,  digi2za2on  becomes  more  relevant      

29 29 29

Agenda

● Measuring digitization

● Assessing digitization’s economic impact

● Digitization in Europe

● Elements of a future agenda

30

The Europe 27 countries are, in aggregate, at an advanced digitization level, only below North America and Western Europe

31

62 59

55

45

33 31 35

12

42

25 32

0

10

20

30

40

50

60

70

DIGITIZATION BY REGIONS (2011)

Source: Katz, Koutroumpis, Callorda (2012) * Europe 27 countries, without Malta because lack of information

Furthermore, these countries have made progress in some areas of their digitization index in the last eight years

32

0  

10  

20  

30  

40  

50  

60  

70  

80  

90  

100  

2004   2005   2006   2007   2008   2009   2010   2011  

Index   Affordability   Infrastructure   Network   Capacity   Usage   Human  Capital  

Index CAGR (%)

Digitization 5.16%

Affordability 0.80%

Infrastructure 0.38%

Network Access 6.44%

Capacity 16.84%

Usage 4.30%

Human Capital 1.93%

EUROPE 27: DIGITIZATION INDEX(2004-2011)

When disaggregated, half of the countries in Europe 27 are at an advanced stage, while the remainder is in a transitional situation

33

68 66 65 64 63 60 59 58 58 57 56 55 55 55 54 54 54 54 53

48 48 48 48 47 46 45 44

0

10

20

30

40

50

60

70

80

Denm

ark

Unite

d Ki

ngdo

m

Luxe

mbo

urg

Finla

nd

Swed

en

Belgi

um

Fran

ce

Portu

gal

Austr

ia G

erm

any

Irelan

d Ne

ther

lands

Eu

rope

27

Mea

n Sl

oven

ia Cz

ech

Repu

blic

Rom

ania

Spain

Ita

ly Lit

huan

ia G

reec

e Po

land

Bulga

ria

Cypr

us

Slov

ak R

ep.

Latvi

a Es

tonia

Hu

ngar

y

EUROPE 27: DIGITIZATION INDEX (2011)

Source: Katz, Koutroumpis, Callorda (2012)

Each group of countries exhibit different sub-indices and challenges

34

95

26

71

88

56

32

0

20

40

60

80

100

Affordability Infrastructure Network Access

Capacity Usage Human Capital

90

16

60 63

39 32

Affordability Infrastructure Network Access

Capacity Usage Human Capital

EUROPE 27: COMPONENTES OF THE DIGITIZATION INDEX (2011) ADVANCED COUNTRIES (*) TRANSITIONAL

COUNTRIES (*)

(*) Denmark, United Kingdom, Luxembourg, Finland, Sweden, Belgium, France, Portugal, Austria, Germany, Ireland, Netherlands

(*) Slovenia, Czech Republic, Romania, Spain, Italy, Lithuania, Greece, Poland, Bulgaria, Cyprus, Slovak Republic, Latvia, Estonia, Hungary Source: Katz, Koutroumpis, Callorda (2012)

Digitization in the Europe 27 countries has generated USD 343 B in new GDP over the last 8 years

35

2004 2005 2006 2007 2008 2009 2010 2011 Total Digitization Index 38.6 39.7 41.7 43.8 46.7 49.0 52.6 54.9 -

GDP Impact (´000 000 000 USD) - $24.03 $46.53 $53.21 $67.21 $50.23 $61.48 $40.87 $343.57

Employment Impact (job/years) (´000) - 265 512 544 655 541 760 480 3,759

EUROPE 27: ECONOMIC IMPACT OF THE DIGITIZATION (2004-2011)

Europe 27 countries have made progress on affordability, network and capacity components, but still have limitations in infrastructure and usage

36

0

20

40

60

80

100 Affordability

Infrastructure

Network

Capacity

Usage

Human Capital

EUROPE 27: COMPONENTS OF DIGITIZATION (2011)

Source: Katz, Koutroumpis, Callorda (2012)

When compared with Norway, the country with the highest index, the challenges for the Europe 27 countries are highlighted

37

0

20

40

60

80

100 Affordability

Infrastructure

Network

Capacity

Usage

Human Capital

COMPARATIVE ANALYSIS OF THE COMPONENTS OF THE DIGITIZATION (2011) EUROPE 27 = 54.9 NORWAY = 72.3

Source: Katz, Koutroumpis, Callorda (2012)

0

20

40

60

80

100 Affordability

Infrastructure

Network

Capacity

Usage

Human Capital

The usage, infrastructure and capacity metrics help assessing the gap specifics

38

Component Indicator Definition Europe 27 Norway U.K. Greece

Usage   Internet  Retail   Retail  internet  as  percentage  of  total  retail   2.90  %   2.13  %   7.74  %   0.69  %  

E-­‐Government   Web  measure  index   0.69   0.86   0.97   0.58  Internet  usage   Percentage  of  individuals  

(users)  using  the  internet   72.10  %   93.97  %   82.00  %   53.00  %  

Spend  in  data   Data  as  a  percentage  of  wireless  ARPU   31  %   44  %   45  %   26  %  

Access  to  social  networks  

Dominant  Social  Network  Unique  Visitors  per  month  per  capita  

25.34  %   49.80  %   44.60  %   14.68  %  

SMS  usage   Average  SMS  sent  by  consumers   212   312   572   94  

Infrastructure   Investment   Investment  per  telecom  subscriber   103.16   549.97   151.14   109.48  

Capacity   Interna2onal  Internet  Bandwidth  

Interna2onal  Internet  Bandwidth  (kbps/user)   86,420   151,257   166,073   26,008  

If the Europe 27 countries were to reach the benchmark index (Norway), it would yield significant economic impact

39

Index 2011 Actual (Europe 27) 2011 Norway Scenario Digitization 54.9 72.3

Affordability * 92.0 96.0

Infrastructure 20.2 100.0

Network 65 67.0

Capacity 74.6 91.6

Usage 47.0 62.0

Human Capital 32.0 32.0

GDP Impact (2011) (´000 000 000 USD) 40.87 355.84

* With the same level of human capital index of Europe 27

Agenda

● Measuring digitization

● Assessing digitization’s economic impact

● Digitization in Europe

● Elements of a future agenda

40

The component analysis allows identifying the challenges facing Europe’s Digital Agenda

●  Europe  has  improved  5.16%  annually  in  terms  of  its  digi2za2on  in  the  past  eight  years    

●  While  affordability  has  barely  improved  (0.80%),  Europe  was  at  the  highest  level  of  all  industrialized  countries  

●  The  principal  improvement  areas  have  been  capacity  (interna2onal  connec2vity  and  broadband  connec2ons  >2  mbps)  (16.84%)  and  network  access  (broadband,  mobile  and  PC  penetra2on)  (6.44%)    

●  However,  three  areas  remain  with  significant  challenges:  infrastructure  (investment)  (0.38%),  usage  (e-­‐Commerce,  e-­‐Government,  Data  ARPU,  social  network  usage)  (4.30%),  and  human  capital  (1.93%)  

●  In  sum,  public  policy  focus  should  be  promo2on  of  investment,  s2mulate  demand  through  applica2ons  development,  and  grow  human  capital  

41 41 41

As a starting point, Europe might need to address inefficiencies in the supply and demand side of digitization

42

Demand for digital goods

Digital Technology Firms

DEMAND SIDE INEFFICIENCIES

Production Factors

SUPPLY SIDE INEFFICIENCIES

•  Consumers •  Enterprises •  Government

• Demand production factors (capital, labor, technology)

• Supply digital goods

•  Capital •  Human capital •  Technology

infrastructure

FRAMEWORK FOR ADDRESSING THE

DEMAND GAP

FRAMEWORK FOR ADDRESSING THE

SUPPLY GAP

Conclusion

●  Digi2za2on  is    a  global  concept,  with  significant  heterogeneity  within  and  across  regions  

●  Digi2za2on  index  builds  on  six  pillars:  affordability,  reliability,  capacity,  access,  usage  and  skills  

●  Index  linked  to  higher  growth,  and  employment  with  increasing  returns  to  scale  ●  Significant  finding  since  it  s2pulates  that  full  economic  impact  of  ICT  

is  achieved  through  the  cumula2ve  adop2on  of  all  technologies,  in  addi2on  to  the  assimila2on  and  usage  in  the  produc2on  and  social  fabric  

●  The  policy  implica2on  is  that  achieving  broadband  penetra2on  is  only  one  aspect  of  required  policies;  maximiza2on  of  economic  impact  can  only  be  achieved  through  a  holis2c  set  of  policies  ranging  from  telecoms  to  compu2ng  to  adop2on  of  internet  and  Ecommerce  

●  At  higher  levels  of  its  development,  digi2za2on  contributes  to  welfare,  thereby  improving  human  development  

43 43 43

44