Impact of innovation platforms on marketing relationships the case of volta basin integrated crop livestock value chains in ghana

A Partner of

Impact of Innovation Platforms on Marketing Relationships: The Case of Volta Basin Integrated

Crop-Livestock Value Chains in Ghana

Volta Basin Development ChallengeFinal Scientific Workshop

Ouagadougou, Burkina Faso17-19 September, 2013

Zewdie A., J. Cadilhon and C. Werthmann

http://www.snvworld.org/en

http://www.google.co.ke/url?sa=i&rct=j&q=&esrc=s&frm=1&source=images&cd=&cad=rja&docid=LGTUDeVexgLaNM&tbnid=NkiMlyRyRRnatM:&ved=0CAUQjRw&url=http://en.wikipedia.org/wiki/Council_for_Scientific_and_Industrial_Research_%E2%80%93_Ghana&ei=ZqwVUtnTE5GT0QXJyICoDg&bvm=bv.51156542,d.d2k&psig=AFQjCNH36zjpKZ1soU13hSAbZMTDYzsKcg&ust=1377238458340628

Andes • Ganges • Limpopo • Mekong • Nile • Volta

Content1. Main message

2. Background and Purpose

3. Methodology

4. Results

5. Lessons

6. Recommendations

2


Main messages

The IPs have created an additional option for value chain actors

The interaction through the platforms contributes to reduction in transaction costs and improvement in access to markets

No sizable market niche has been created although there is a potential for the future

3


1. Background and purpose

The changing nature of Agricultural ResearchAgricultural innovations as multi-dimensional and

co-evolutionary processes technological, organizational and institutional

innovations

creating synergies “convergence of agricultural sciences” (Huis et al.

2007)

Growing use of Innovation Platform (IP) approaches

4


What is an Innovation Platform?

Source: PAEPARD, October 2012

IP meeting in Lawra, June 27, 2013

5

http://paepard.blogspot.com/2012/10/learning-and-innovation-platforms-for.html

http://www.fara-africa.org/media/uploads/Image/NSFs/NSF2/DONATA/IPTA-complete-s.png


The Volta2 Innovation Platforms

Ghana Volta2 IPs were established in July 2011

Main goals of the Volta2 IP project:

to facilitate communication and collaboration among

actors

to support value chains development

to serve as spaces for Participatory Action Research

(PAR)

6


Purpose of the study

to assess the interrelationships between various

actors

to investigate the impact of communication and

information sharing on market access

test a new conceptual framework for evaluation of

IPs

7


Why to use a new framework?

Limitations with conventional methods

difficulty to check cause-effect relationships

only few econometric methods

E.g. DID and IV methods, but not applied to IPs (and use control

groups)

Statistics and record on agricultural data in LDCs is poor

Perception of respondents based on Likert-scale as an

alternative?

8


2. Methodology

Data Four communities (two IPs)

Lawra: Upper West region Orbilli and Naburinye

Tolon-Kumbungu: Northern region Digu and Golinga

43 IP members 34 farmers, 6 traders, 3 processors

9 key respondentsSource: Diamenu and Nyaku 1998

9

http://www.google.co.ke/url?sa=i&source=images&cd=&cad=rja&docid=43NBE6POjBGMvM&tbnid=Nw3m2AvfynkRcM:&ved=&url=http://www.sciencedirect.com/science/article/pii/S0277953698000288&ei=7wAJUvXLPI7Oswb5ioG4BQ&psig=AFQjCNEeXQOzNu4Oo1JQFtx97saGOACP9w&ust=1376408176040406


Data cont…

Socio-economic information and 5 point Likert-scale data from IP members (1 = strongly disagree, 2 = disagree, 3 = undecided, 4 = agree, 5 = strongly agree)

Assumption: 5-4 = 4-3 = 3-2 = 2-1

10

Simplifying the Likert-scale, to ensure better response quality


Method of analysis

Mixed methods using a new framework

Qualitative

Quantitative response summaries and averages on various

statements factor analysis on the Likert-scale variables

Principal components factor

regression Ordinary Least Squares (OLS)

11


The conceptual framework

Based on a mix of concepts from various disciplines

Industrial Economics

Structure-Conduct-Performance (SCP) hypothesis

New Institutional Economics of Markets

transaction costs

governance structures Concepts from the marketing literature

value chain relationship12


“A priori linear relationship between Structure, Conduct and Performance”

The SCP hypothesis

13


The regression model

Semi-logarithmic multiple regression

14


Explanation of terms marketaccessj = the jth dependent variable of market access IP = a dummy variable that assumes 1 for Tolon-Kumbungu and 0

for Lawra to account for possible differences between the two IPs gender = 1 for male and 0 for female age = age of the IP member lnnbhous = natural logarithm of household size Incestm = annual income of the participant communicationi = the ith variable that represents the level of

communication and information sharing of by an IP member , , ….. are partial effects of each respective explanatory variables

on market access. is an intercept term is the error term or model residual i and j depend on the outcome of the factor analysis

15


4. Results

16


17

More descriptive


Three factors* on communication and information sharing

Four factors* on market access

Results of the factor analysis

*Based on the Kaiser criteria is used

18


Results for regression equation 1Regression Equation

Dependent Variable

Explanatory variables

Coefficient Beta t P>|t|

1

factor11 IP 0.0638 (0.617) 0.032 0.10 0.918 gender 0.2767 (0.349) 0.139 0.79

0.436

lnnbhous -0.0182 (0.542) -0.007 -0.03

0.973

age -0.0014 (0.011) -0.020 -0.13

0.896

incestm -0.0003 (0.000) -0.240 -1.19

0.247

focq50i 0.5782 (0.255)

0.365** 2.26

0.032

factor1 0.1543 (0.256) 0.156 0.60

0.553

factor2 -0.0642 (0.231) -0.069 -0.28

0.783

factor3 -0.1543 (0.619) -0.157 -0.95

0.349

constant -2.1627 (1.472)

. -1.47

0.154

- Robust) standard errors are shown in brackets and betas are standardised coefficients. - * and ** represent statistical significance of betas at 1% and 5% levels of significance - focq50i = individual statement about improvement in communication in the past 2 years

19


Results for regression equation 2


Regression Equation

Dependent Variable



2

factor12

IP 0.6432 (0.443) 0.324 1.45

0.159

gender -0.2612 (0.296) -0.131 -0.88

0.387

lnnbhous 0.1248 (0.242) 0.054 0.51

0.611

age -0.0122 (0.012) -0.169 -0.96

0.344

incestm -0.0001 (0.000) -0.026 -0.17

0.867

focq50i -0.1968 (0.258) -0.124 -0.76

0.453

factor1 0.2535 (0.2252) 0.257 1.13

0.271

factor2 0.3339 (0 .117)

0.359* 2.84 0.009

factor3 -0.0460 ( 0.142) -0.047 -0.32

0.749

constant 1.068 (1.3416) . 0.80

0.433

20




Regression Equation

Dependent Variable



3

factor13 IP -0.3810 (0.553) -0.192 -0.69

0.497

gender -0.8305 (0 .381)

-0.418**

-2.18 0.039

lnnbhous 0.0440 (0.422) -0.418 0.10

0.918

age -0.0228 (0.014) 0.019 -1.55

0.132

incestm 0.0002 (0.000) -0.314 0.71

0.486

focq50i 0.3342 (0.322) 0.122 1.04

0.308

factor1 0.3007 (0.272) 0.305 1.10

0.279

factor2 0.0222 (0.162) 0.023 0.14

0.892

factor3 -0.1405 (0.223) -0.143 -0.63

0.534

constant 0.01651 (1.864) . 0.01

0.993

21




Regression Equation

Dependent Variable



4 factor14 IP 1.8330 (0.402)

0.923* 4.55

0.000

gender -0.2039 (0.297) -0.102 -0.69

0.499

lnnbhous -1.0078 (0.429)

-0.438**

-2.35

0.027

age 0.0123 (0.011) 0.170 1.14

0.265

incestm 0.0006 (0.000)

0.449* 3.02

0.006

focq50i -0.0157 (0.228) -0.009 -0.07

0.946

factor1 -0.1235 (0.165) -0.125 -0.75

0.463

factor2 -0.3224 (0.137)

-0.347**

-2.35

0.027

factor3 -0.0318 (0.118) -0.032 -0.27

0.790

constant 0.4784 (1.324) . 0.36

0.721

22


Regression results

Improvement in access to input and output markets is related to improvements in overall communication or interaction

Women have better access to market than men

Participants in Tolon-Kumbungu IP have better market access than those in Lawra: same information from baseline survey?

The higher the annual income the higher the level of market access

(log) household size (local indicator of wealth) is negatively related to market access

Use of the media reduces the likelihood of bypassing intermediaries although it improves access to market information and hence to markets

23


Results from qualitative information

Improved knowledge because of trainings and meetings

trainings by urban traders and processors to their rural counterparts and to farmers

value chain training by a marketing expert from ILRI

Knowledge on price standardization, commercialization and use of weighing scales

New, but limited additional market options because of the IP

members met potential trade partners due to the IP meetings

farmers and traders check prices before engaging in transactions

24


Results from qualitative information cont…

Improved knowledge on crop and livestock production

Improved post-harvest management and timing of sale

Knowledge on cooperatives

Shelter for small ruminants, constructed after a training under the IP project : at Digu village

25


5. Lessons

The IPs played a ‘role’ in improving communication and

information sharing and opened new options

Proximity to market centers and level of income of the

members seems to be key determinants of access to

market

So, were the IPs helpful to the communities?

It is not time to judge whether the IPs had significant

impact

26


6. Recommendations

IPs should not be left alone at this stage

further education on commercial production, marketing,

record keeping, cost benefit analysis and business plan

development

more effort to link the value chain actors to engage in

commercial transaction and hence creating new markets

further efforts on cooperative formation

linking them to other development partners

Evaluating the overall impact on livelihoods is also

required

27


Annex

28


29

Name of factor

Statements contributing to the variances in the respective factors representing communication and information sharing

Remark (assigning name to the factors)

Factor1 I exchange information with my value chain partners about my on-going activities

Information sharing?

My value chain partners exchange information about their on-going activities with me

Factor2 I listen to weekly radio announcements to get market information

Using media to acquire information/better communication?I am satisfied with the quality of communication I

was having with my business partners in the last two years

Factor3 I am satisfied with the communication frequency I had with value chain actors in recent business relationships

Frequent communication to obtain market information?I ask relatives and friends in the village for market

information

Underlying factors for communication and information sharing


30

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

5 5 5 5 5

4

5 5

4 4 4

5

4 4

5 5 5 5 5

4

5 5

4

5

3

4 4

1

2

1

5

3

5

3

4 4

3

4

3

4 4

3

4

5

4

5 5 5

4

4

5

4

5

4

5

5

4

5 5 5 5 5

4

5

4

4

4

4

1

4

3

1 2

4

3

5

3

4 4

3

3

3

4 4

3

4

Relationship between responses on statements 28_a (I exchange in-formation with my VC partners about my ongoing activities) and 28_b (my VC partners exchange information about their ongoing activities

with me) which together constitute Factor1 (information

28_a 28_b


31

Name of factor

Statements contributing to the variances in the respective factors representing market access

Remark (assigning name to the factors)

Factor11 The number of marketing companies buying products from the villagers has increased in the past two years

Improved access to input and output markets?My access to input markets has improved in the past two

years My access to output market has improved in the past two years

Factor12 Information on the market is easily accessible to value chain actors

Better access to market information?Farmers in the IP negotiate with buyers as a group

Factor13 I can now better negotiate market prices than two years ago Improved negotiation for better price?I am satisfied by the prices I get from my customers for my

products

Factor14 I sell my output directly to processers or consumers Bypassing market intermediaries?There is a ready market for farm produce during harvesting

seasons in my area

Underlying factors for market access


32

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43

5

4

5 5

4 4

5

4

5

4 4

5 5

4

3

4 4 4 4 4 4

3

5 5 5 5

4

3

5

4

5

3

4

2

4 4 4 4 4

5

4 4 4

5

3

5 5

5

4

5

4

5

5 5

5

4

4

3

4

2

4

1

4 4 5

4

5 5 5

5

4

4

4

4

4

4

4

4

2

4 4 4

5

3

5

4

Relationship between responses FocQG_55m (my access to input market has improved in the past two years) and FocQG_55n (my

access to output market has improved in the past two years)

FocQG_55m FocQG_55n


33

Factor analysis

Kaiser-Meyer-Olkin (KMO) Measure of Sampling Adequacy

Bartlett’s test of sphericity

Cronbach’s Alpha

Chi-square p-value

Conduct 0.748 142.887* 0.000 0.81Performance 0.641 93.161* 0.000 0.72 H0: variables are

not intercorrelated

• * implies that the test rejects the null hypothesis at the 1% level of significance

• KMO > 0.6; Cronbach’s alpha > 0.7 and P-value < 0.05 for Bartlett’s tests all suggest that conducting factor analysis is appropriate both for Conduct and Performance indicators

Tests of factorability and reliability


34

Variable VIF Tolerance = 1/VIF Variable VIF Tolerance = 1/VIF

IP 2.84 0.352489 lnnbhous 1.24 0.808407

factor1 2.60 0.385266 age 1.22 0.822854

factor2 1.40 0.712389 incestm2 1.21 0.823299

gender 1.32 0.757049 factor3 1.13 0.888709

focq50i 1.29 0.776226 Mean VIF = 1.58

VIF < 5 implies absence of serious multicollinearity problem

Multicollinearity test for explanatory variables using VIF


35

Shapiro-Wilk W test for normality Breusch-Pagan / Cook-Weisberg test Variable W V Z P>Z Variable chi2(1) P>chi2Resid1 0.957

1.794

1.235 0.108 fitted values of factor11

0.38 0.539

Resid2 0.946

2.256

1.719**

0.042 fitted values of factor12

3.99** 0.045

Resid3 0.941 2.434 1.880** 0.030 fitted values of factor13

1.32 0.249

Resid4 0.964

1.464

0.805 0.210 fitted values of factor14

0.16 0.687

Ho: error term is normally distributed Ho: dependent variable has constant variance

- ** implies that the test rejects the null hypothesis at the 5% level of significance.- Resid refers to the residuals of the corresponding regression equations- No serious deviation from normality when the 1% level of significance is

considered

Tests of variance equality and residual normality in each of the four equations


36

Regression Equation no.

Dependent Variable

F-value

Prob > F R-squared

1 factor11 0.35 0.7920 0.33242 factor12 0.54 0.6584 0.50783 factor13 0.43 0.7315 0.27924 factor14 0.42 0.7396 0.5264 Ho: model has no omitted variables

All the four models are well specified, no serious problem of omitted variables

R-square is quite low particularly in equation 3

Ramsey regression equation error specification test (RESET) and overall fit of the models


Pictures from focus group discussions

37


Thank you!

Also credit to the contributors, CPWF, CSIR-ARI, SNV, ILRI and others

38

Technology

Impact of innovation platforms on marketing relationships the case of volta basin integrated crop livestock value chains in ghana