CGIAR Research Program on Policies, Institutions, and Markets Workshop on Rural Transformation in the 21st Century (Vancouver, BC – 28 July 2018, 30th International Conference of Agricultural Economists). Presentation by Insa Flachsbarth (University of Göttingen, Germany), Nina Grassnick (University of Göttingen, Germany), Amjad Masood (Khwaja Fareed University of Engineering and Information Technology, Pakistan), Bernhard Brümmer (University of Göttingen, Germany).

1. The Uneven Spread of Private Food Quality
Standards over Time and Space
Insa Flachsbarth, Nina Grassnick,
Amjad Masood, Bernhard Br¨ummer
CGIAR Policies, Institutions and Markets
Pre-Conference Workshop:
Rural Transformation in the 21st Century: The Challenges of Low-Income,
Late-Transforming Countries
28th July 2018

2. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
2 / 18

3. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
2 / 18

4. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
2 / 18

5. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
2 / 18

6. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
2 / 18

7. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
+
Supermarket revolution in
developing countries
(GlobalGAP membership
of 15 non-European retail
chains)
2 / 18

8. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
+
Supermarket revolution in
developing countries
(GlobalGAP membership
of 15 non-European retail
chains)
⇓
Demand for certified
produce intensifies across
the world in oligopsony
market structures
2 / 18

9. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
+
Supermarket revolution in
developing countries
(GlobalGAP membership
of 15 non-European retail
chains)
⇓
Demand for certified
produce intensifies across
the world in oligopsony
market structures
⇒ Compliance by farmers becomes de facto mandatory for market participation,
2 / 18

10. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
+
Supermarket revolution in
developing countries
(GlobalGAP membership
of 15 non-European retail
chains)
⇓
Demand for certified
produce intensifies across
the world in oligopsony
market structures
⇒ Compliance by farmers becomes de facto mandatory for market participation,
yet certification rates unequal around the world
2 / 18

11. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Motivation
Discerning consumers
demand safe high-quality,
sustainably and ‘ethically’
produced food
+
Strict public standards in
major markets
⇓
In global agricultural
supply chains, standards
must be fulfilled
independent of origin of
produce
=⇒
Modernisation
of procurement
systems of
agri-food
companies and
retail chains
=⇒
Food quality standards
ubiquitous in western
markets (GlobalGAP
membership of 30
European retail chains)
+
Supermarket revolution in
developing countries
(GlobalGAP membership
of 15 non-European retail
chains)
⇓
Demand for certified
produce intensifies across
the world in oligopsony
market structures
⇒ Compliance by farmers becomes de facto mandatory for market participation,
yet certification rates unequal around the world
Why?
2 / 18

12. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
GlobalGAP
Absolute change of F&V GlobalGAP certified producers (2008-2015)
Absolute change (2008-2015)
no certification
≤ -50
> -50 - (-1)
no change
> 0 - 6000
> 6000 - 12000
> 12000
Legend
Source: Authors’ elaborations based on GlobalGAP Secretary
Sample: 183 crop producing countries, 7 years, 111 countries certified in 2015
⇒ 24 new countries certified in 2015 compared to 2008 and increase in certified areas
in most countries
3 / 18

13. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
4 / 18

14. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
4 / 18

15. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
4 / 18

16. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
• Lack of dynamic component of analysis due to missing panel
data
4 / 18

17. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
• Lack of dynamic component of analysis due to missing panel
data
• Lack of theoretical foundation
4 / 18

18. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
• Lack of dynamic component of analysis due to missing panel
data
• Lack of theoretical foundation
=⇒ Crucial to understand mechanisms!
Research questions
4 / 18

19. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
• Lack of dynamic component of analysis due to missing panel
data
• Lack of theoretical foundation
=⇒ Crucial to understand mechanisms!
Research questions
• Why does certification start to emerge in some countries and
in others not? (Extensive margin)
4 / 18

20. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Contribution to the literature and research questions
Contribution to the literature
• Lack of macro-studies for general conclusions
• Lack of dynamic component of analysis due to missing panel
data
• Lack of theoretical foundation
=⇒ Crucial to understand mechanisms!
Research questions
• Why does certification start to emerge in some countries and
in others not? (Extensive margin)
• What determines higher levels of certification and the fast
spread in some regions? (Intensive margin)
4 / 18

21. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Conceptual framework
Spread of standards
(Macro-level)
Extensive margin
Adoption decision process over time
(Micro-level)
Macro-level indicators
shaping micro-level behaviour
Empirical model results
Information
phase
Networks and information flow
 Information on existence of
GlobalGAP
 Knowledge spillovers by peers
Persuasion
phase
Confirmation
phase
Implementation
phase
Sustained exceeding of benefits
beyond costs of re-certification
Perceived direct costs and benefits
of certification and risk attitude
Macroeconomic environment
and sectoral policies
 Transaction costs
Rogers’ phases
of innovation
adoption:
Intensive margin
Dynamics
5 / 18

22. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Zero inflated negative binomial model
Estimation equations - First stage: Logit model
6 / 18

23. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Zero inflated negative binomial model
Estimation equations - First stage: Logit model
P r(yi,t = 0|xi,t) =exp(β0 + β1GGAP-Neighbouri,t + β2Urbani,t + β3Colonyi + β4FVareai,t
+ β5Inputsi,t + β6ExShareCropsi,t−1 + β7Age1534i,t + β8lnPopi,t
+ β9Ii,t + β10lnERi,t + β11DBi,t + β12lnDFAgi,t + β13MRLi,t
+ β14ISOi,t + β15MGDi,t + β16Roadi,t + β17wwwi,t + β18lnGDPpci,t
+ β19lnGDPpc
2
i,t + β20AgLandi,t + ψt + i,t)
(1)
where i = country, t = time period, ψt = year fixed effects and i,t = error term clustered at the country-level.
GGAP − Neighbouri,t: Average number of
neighbouring countries’ certified crop producers
Urbani,t: Share of urban population
Colonyi: Colonial history with one of the EU6 countries
F V areai,t: Fruits and vegetables area harvested (’000 ha)
Inputsi,t: Agricultural inputs indicator
ExShareCropsi,t−1: One-year lag crop-export share of
total exports
Age1534i,t: Share of population in age cohort 15-34 years
lnP opi,t: Log number of inhabitants per country
Ii,t: Inflation rate
lnERi,t: Log exchange rate
DBi,t: Doing-Business indicator
MRLi,t: Strict Maximum Residue limits dummy
ISOi,t: ISO membership dummy
lnDF Agi,t: Log development flows to agriculture
(constant 2015 USD)
MGDi,t: Share of modern grocery sales in total food retail
and wholesale market
Roadi,t: Road density per km2
wwwi,t: Share of population with fixed broadband
subscription
lnGDP pci,t: Log GDP per capita (current USD)
AgLandi,t: Total arable land (’000 ha)
6 / 18

24. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Zero inflated negative binomial model
Estimation equations - Second stage: Negative binomial model
7 / 18

25. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Zero inflated negative binomial model
Estimation equations - Second stage: Negative binomial model
E(yi,t|xi,t, yi,t > 0) =exp(β0 + β1GGAPi,t−1 + β2GGAP-Neighbouri,t + β3Urbani,t
+ β4ExShareEUi,t−1 + β5Colonyi + β6FVareai,t + β7Inputsi,t
+ β8ExShareCropsi,t−1 + β9Age1534i,t + β10lnPopi,t + β11Ii,t
+ β12lnERi,t + β13DBi,t + β14lnDFAgi,t + β15MRLi,t + β16∆TFPi,t
+ β17MGDi,t + β18Roadi,t + β19wwwi,t + β20Auditi
+ β21lnGDPpci,t + β22lnGDPpc
2
i,t + β23AgLandi,t + ψt + i,t)
(2)
GGAPi,t−1: One-year lag number of certified producers
GGAP − Neighbouri,t: Average number of
neighbouring countries’ certified crop producers
Urbani,t: Share of urban population
ExShareEUi,t−1: EU crop export share
Colonyi: Colonial history with one of the EU6 countries
F V areai,t: Fruits and vegetables area harvested (’000 ha)
Inputsi,t: Agricultural inputs indicator
ExShareCropsi,t−1: One-year lag crop-export share of
total exports
Age1534i,t: Share of population in age cohort 15-34 years
lnP opi,t: Log number of inhabitants per country
Ii,t: Inflation rate
lnERi,t: Log exchange rate
DBi,t: Doing-Business indicator
MRLi,t: Strict Maximum Residue limits dummy
∆TFPi,t: Total factor productivity growth
lnDF Agi,t: Log development flows to agriculture
(constant 2015 USD)
MGDi,t: Share of modern grocery sales in total food retail
and wholesale market
Roadi,t: Road density per km2
wwwi,t: Share of population with fixed broadband
subscription
lnGDP pci,t: Log GDP per capita (current USD)
AgLandi,t: Total arable land (’000 ha)
7 / 18

26. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Summary of main findings: Extensive margin
Why does certification start to emerge in some countries and in others not?
Variable Effect Rational
Information Communication
Technologies
+
• Fast information flow
• Lower costs of documentation and traceability.
Urbanisation +
• Agglomeration economies
• Home market effect due to ‘western diets’
Modern grocery distribution in
both developed and developing
countries
+
• More access to standard-specific information
• Increase in buyer power
• Home market effect - less export dependency
Development flows to agriculture +
• Financial support: certification costs
• Financial and technical support: initial investments
Strict public standards (Maximum
residue limit)
+
• Additional required private investments lower
Note: A positive sign means that the variable increases the probability of entering the certification market.
The horizontal lines separate Roger’s phases: (1) Informatio phase, (2) Persusasion phase, (3) Implementa-
tion phase
8 / 18

27. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Summary of main findings: Intensive margin
Why do some countries show high certification rates and why does it spread fast?
Variable Effect Rational
Existing trade networks with EU +
• Network effects with core market
• Vertical integration: multinationals – farmers
• Export dependency and oligopsony market structures
Domestic auditor + • Reduced certification costs
Input indicator + • High total factor productivity due to access to inputs
(labour, land, fertiliser, technology)
Size of home population + • Home market effect
Inflation rate – • Lower propensity to invest
Doing business indicator +
Reduced transaction costs, specifically:
• Access to credit
• Effective land-markets (secure property rights)
• Functioning contract enforcement
• Efficient trading across borders
Strict public standards
(Maximum residue limit)
+
• Overlap of public and private standard
• Farmers’ investments imperative to serve domestic mar-
ket → Part of initial investments needed to comply to
GlobalGAP already made
Transportation infrastructure +
• Fast information flow
• Lower trade costs to high-value markets
Note: A positive sign means an increasing effect on the spread of GlobalGAP.
The horizontal lines separate Roger’s phases: (1) Informatio phase, (2) Persusasion phase, (3) Implementation phase
9 / 18

28. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Some implications and major trends
10 / 18

29. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Some implications and major trends
We find that...
... the domestic presence of modern retailers in more urban
societies increases a country’s probability of GlobalGAP
market entry.
10 / 18

30. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Some implications and major trends
We find that...
... the domestic presence of modern retailers in more urban
societies increases a country’s probability of GlobalGAP
market entry.
Thus...
...ongoing urbanisation and the expanding supermarket
revolution (spreading also into rural areas) will open the door
for some of the poorer countries in the future.
10 / 18

31. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Some implications and major trends
We find that...
... the domestic presence of modern retailers in more urban
societies increases a country’s probability of GlobalGAP
market entry.
Thus...
...ongoing urbanisation and the expanding supermarket
revolution (spreading also into rural areas) will open the door
for some of the poorer countries in the future.
We also find that...
...the macroeconomic environment, sector policies, and
infrastructure investments can support and intensify this
process.
10 / 18

32. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
11 / 18

33. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
11 / 18

34. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
11 / 18

35. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
11 / 18

36. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
11 / 18

37. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
• Central governments should create a stable and stimulating
macroeconomic environment:
11 / 18

38. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
• Central governments should create a stable and stimulating
macroeconomic environment:
• Trade agreements with core GlobalGAP markets
11 / 18

39. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
• Central governments should create a stable and stimulating
macroeconomic environment:
• Trade agreements with core GlobalGAP markets
• Guarantee functioning executive authorities to enforce
(GlobalGAP) contracts and to secure land tenure rights.
11 / 18

40. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
• Central governments should create a stable and stimulating
macroeconomic environment:
• Trade agreements with core GlobalGAP markets
• Guarantee functioning executive authorities to enforce
(GlobalGAP) contracts and to secure land tenure rights.
• Infrastructure investments to reduce costs
11 / 18

41. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Policy conclusions
• Agricultural specific bilateral development flows can help to
overcome initial market access constraints.
• Financial or technical support by regional development banks
can reinforce agricultural sector development.
• Non Governmental Organisation carry part of the high
certification costs.
• Agricultural sector policies and standard setters can facilitate
access to inputs via input subsidies and extension services.
• Central governments should create a stable and stimulating
macroeconomic environment:
• Trade agreements with core GlobalGAP markets
• Guarantee functioning executive authorities to enforce
(GlobalGAP) contracts and to secure land tenure rights.
• Infrastructure investments to reduce costs
• The development of domestic public food safety standards by
agricultural ministries obliges farmers to invest. ⇒ Initial
investment costs to comply with GlobalGAP now lower.
11 / 18

42. Thank you for your attention!
Contact:
Mail: Insa.Flachsbarth@agr.uni-goettingen.de
Web: GlobalFood: Transformation of Global Agri-Food Systems

43. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Properties of dependent variable: Number of certified
producers per country
Specification tests of overdispersion
• min=0, max=32550, mean = 602, variance = 6690634
⇒ variance-mean ratio = 11117
02.0e-044.0e-046.0e-048.0e-04
Density
0 10000 20000 30000
GlobalGAP number of producers (scope crops)
⇒ Dependent variable highly
skewed to the right
⇒ DGP leads to zero-inflated count data models with
overdispersion
14 / 18

44. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Data I
Descriptive statistics
Certified Non-certified Sources
mean sd min max mean sd min max
GGAP producers (no.) 1130 3460 1.00 32550 GlobalGAP
Information phase
GGAP producers neighb.
countries (mean no.)
1081 3040 0.00 32550 345 973 0.00 6894 GlobalGAP
Share urban population 62.10 20.93 13.89 97.82 48.83 23.60 8.55 100.00 Worldbank
Export share crops
EU/world
44.91 29.92 0.00 99.10 30.94 32.27 0.00 100.00 UN Comtrade
Colonial history with EU6 0.35 0.48 0.00 1.00 0.36 0.48 0.00 1.00 Cepii
F&V (area harv. in hec.) 997 3857 0.00 37966 185 510 0.00 5070 FAOSTAT
Persuasion phase
Inputs (princ. component) 0.31 2.71 -0.59 19.76 -0.73 1.37 -4.27 3.3 USDA
Export share crops/total 7.51 10.17 0.00 70.38 6.90 14.13 0.00 90.80 UN Comtrade
Age cohort 15-34 years
(share pop)
33.07 6.43 19.12 46.67 37.24 4.51 23.90 47.90 Worldbank
Population (million) 60.45 186.27 0.02 1364.27 12.38 27.50 0.01 176.46 Worldbank
Implementation phase
Inflation rate 5.20 5.45 -8.28 44.39 6.54 7.69 -35.84 62.17 Worldbank
Exchange rate (LCU/USD) 659 2701 0.28 25941 627 2248 0.21 19068 Worldbank
Doing Business (princ.
component)
0.57 1.30 -3.72 3.19 -0.73 1.37 -4.27 3.30 Worldbank
Strict MRL (1=yes) 0.44 0.50 0.00 1.00 0.04 0.20 0.00 1.00 Homologa
ISO membership (1=yes) 0.80 0.40 0.00 1.00 0.33 0.47 0.00 1.00 ISO
15 / 18

45. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Data II
Certified Non-certified
mean sd min max mean sd min max Sources
Confirmation phase
TFP growth rate 0.02 0.08 -0.47 0.67 0.01 0.09 -0.37 0.55 USDA
Variables relevant in more than one of Rogers’ phases
MGD 35.01 29.97 0.00 96.44 8.18 17.80 0.00 95.41 PlanetRetail
Road density (per km2) 0.83 1.27 0.01 9.68 0.69 1.52 0.01 11.27 CIA, Worldbank
WWW subsc. (pop. share) 12.74 14.14 0.00 104.52 4.83 8.49 0.00 36.55 Worldbank
Domestic auditor (1 = yes) 0.39 0.49 0.00 1.00 0.00 0.00 0.00 0.00 GlobalGAP
GDP per capita (‘000 cur-
rent USD)
16.40 19.77 0.33 103.06 10.06 18.27 0.20 119.23 Worldbank
Development flows to agr.
(Mio. USD, constant
2015)
48.99 88.36 -0.22 629.69 18.14 33 0 325.77 FAOSTAT
Controls
Arable land (‘000 hec.) 11360 26743 1.00 161812 3525 12426 0.14 122240 FAOSTAT
Developing country
(1=yes)
0.64 0.48 0.00 1.00 0.81 0.39 0.00 1.00 Worldbank
Note: MGD = Modern gocery distribution; F&V = Fruits and vegetables; MRL = Maximum residual limit;
TFP = Total factor productivity
16 / 18

46. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Regression results I
Table 2: Determinants of the spread of GlobalGAP certified farms
VARIABLES NB - Stage two Logit - Stage one
Information phase
Lag no. GlobalGAP certified producers 0.001*** (0.000)
Average no. certified neighbours 0.000 (0.000) 0.001** (0.000)
Urban population 0.001 (0.019) -0.107** (0.048)
EU crop export share 0.022*** (0.007)
Colony with EU6 1.493*** (0.489) -0.959 (1.483)
Fruits and vegetables area -0.001*** (0.000) -0.003 (0.002)
Persuasion phase
Agricultural Inputs 0.729** (0.315) 0.797 (3.101)
Lag crop-export share of tot. Exports 0.055* (0.031) -0.015 (0.023)
Age cohort 15-34 years 0.074 (0.051) 0.292** (0.142)
ln Population 0.549*** (0.192) 0.392 (0.638)
Implementation phase
Inflation -0.064*** (0.024) -0.166 (0.111)
ln Exchange rate -0.011 (0.077) 0.552** (0.238)
Doing-Business 0.662*** (0.233) 0.149 (0.655)
Maximum Residual Limits 1.770*** (0.463) -18.75*** (3.093)
ISO membership 0.757 (1.049)
Confirmation phase
Total factor productivity growth 0.300 (0.737)
17 / 18

47. Motivation Research questions Conceptual framework Empirical model and data Results Conclusions Appendix
Regression results II
Continued: Determinants of the spread of GlobalGAP certified farms
Variables relevant in more than one of Rogers’ phases
ln Development flows to agriculture 0.074 (0.188) -1.593*** (0.542)
Modern grocery*Developing 0.002 (0.008) -0.160*** (0.054)
Modern grocery*Developed -0.036*** (0.012) -0.638** (0.298)
Road density 0.342* (0.186) -2.295 (1.510)
Fixed broadband subscription -0.042 (0.027) -0.471** (0.191)
Audit 1.882*** (0.390)
ln GDP per capita -0.788 (0.672) -0.312 (0.891)
ln GDP per capita squared 0.334*** (0.129) 1.299*** (0.498)
Control
Arable land 0.000*** (0.000) 0.000 (0.001)
Constant -9.174** (4.037) -6.379 (11.068)
year fixed effects yes yes
alpha 0.764*** (0.111)
Observations 895 895
AIC 7420 7420
Chi-square test 324 324
Note: Robust standard errors are clustered by country and given in parentheses;
*** p<0.01, ** p<0.05, * p<0.1;
18 / 18