1. University of Dar es Salaam Business School; Molde University (Norway)
2017 REPOA ANNUAL WORKSHOP
Buyer-supplier integration and logistics performance in health care facilities in Tanzania: the moderating effect of centralized decision control
Salema, G & Buvik, A
University of Dar es Salaam Business School; Molde University (Norway)
REPOA MARCH 2017

2. Paper Layout Practical problem of interest Research problem
Study hypothesis and the research model
Methodology
Data quality
Findings
Discussions and implications
Limitations and further research
REPOA MARCH 2017

3. Practical problem
Access to essential medicines is still a critical problem in Tanzania’s public health facilities…intensifies as one moves down the referral system
According to the annual health assessment report 2014, only 30% of public facilities had an entire package of essential medicines (URT, 2016)
This paper addressed the availability dimension and posit that it is depends on the supplier’s (MSD) delivery performance
There is existing evidence to support that MSD’s has unreliable supply system (EHG, 2007; JAHSR, 2006; MoHSW, 2008; MoHSW, 2008a; URT, 2016 ). Therefore this paper tries to examine antecedents for MSD’s logistics performance.
REPOA MARCH 2017

4. !!!!!!!!!!!!!!!! There is still value in OT !!!!!!!!!!!!!
Research problem
Is the research applications of the organizational design theory waned (Koufterous et al., 2014)?
This paper supports Vora’s (1992) arguments that integrating constructs and reasoning from basic organizational design theory (OT) and other fields of supply chain management research makes OT useful
!!!!!!!!!!!!!!!! There is still value in OT !!!!!!!!!!!!!
REPOA MARCH 2017

5. Backround and research problem
This paper:
contributes to the existing debate on the effects of supply chain integration on logistics performance (Fabbes-Costes and Jahre, 2008).
examines the link between buyer-supplier integration (external integration : interaction and joint action between the buyer and supplier) and supplier logistics performance in a health care setting (Public health facility – medical supplies department “MSD”).
Missing evidence on how organisation structure and integration are related
Little work has been done in testing the effects of organization structure on external relationships with suppliers/customers (Koufteros et al., 2014)
REPOA MARCH 2017

6. Background and research problem
Based on the resource based view (RBV) theory, buyer-supplier integration is a unique dyadic resource which synchronizes the core logistics competencies and capabilities of all actors to jointly achieve improved logistics performance.
majority of researchers support investing more in buyer-supplier integration enhances logistic performance (Danese and Romano, 2013; Gimenez and Ventura, 2005; Stank et al., 1999a), while others like Das et al., (2006) are skeptical and suggest that this effect is rather modest.
The observed inconsistencies in current research suggest that there are other factors influencing this association
REPOA MARCH 2017

7. Background and research problem
According to the literature context ( uncertainty, complexity an organization structure) influences the link.
evidence shows that structure of an organization determines its performance (Ruekert et al., 1985; Stanley, 1993).
However little attention has been paid on the performance impact of purchasing structure (Glock and Hochrein, 2011).
In the purchasing organization literature purchasing structure is considered a contextual factor (Glock and Hochrein, 2011) which plays a significant role improving supplier effectiveness in its logistical functions
However little work has been done in testing the effects of organization structure on external relationships with suppliers and customers (Koufteros et al., 2014)
REPOA MARCH 2017

8. Backround and research problem
This paper extends Koufterous et al., (2014) contributions, who suggest that there is a link between centralization and external integration
The main objective of this paper was to examine the combined effect of purchasing centralization and buyer-supplier integration on supplier logistics performance
REPOA MARCH 2017

9. Hypotheses
Based on the organisation design theory and RBV theory......
H1: There is a positive association between the extent of buyer-supplier integration and supplier logistics performance
H2: The association between buyer-supplier integration and supplier logistics performance decreases as the extent of buying firm’s purchasing centralization increases.
REPOA MARCH 2017

10. Research model
SUPPLPERF=b0 + b1 UNCERT + b2 SUPPLMON + b3 PROX + b4 GEO + b5 CENTRAL +b6 SUPPLINT + b7CENTRALxSUPPLINT + e
REPOA MARCH 2017

11. Key constructs Buyer-supplier integration
the extent of inter-firm coordination/interaction and collaboration between a public health facility and its focal supplier of essential medicines
Buyer-supplier integration
how well the buying firm perceives the performance of their zonal medical supplier with respect to on time delivery; order filling capacity, lead time and accuracy in order delivery.
Supplier logistics performance
the extent of autonomy in purchasing decision making residing with the purchasing managers in the public health facilities (the degree of influence of district and regional medical office)
Purchasing centralization
level of unpredictability in the exchange environment of the public health facility
Environmental uncertainty
Extent of verification of the delivery of essential medicines based on product quality, delivery timelines, and order accuracy
Supplier output monitoring
REPOA MARCH 2017

12. Methodology
Empirical setting: the public essential medicines supply system in Tanzania
unit of analysis: exchange relationship between the public health facility and its focal supplier (MSD)
stratified random sampling was used to sample 215 public health facilities (hospitals, health centers and dispensaries).
A questionnaire in a seven point Likert scale was used to collect data from purchasing managers of health facilities.
The average experience of key informants was approx. 6 years: indicate satisfactory familiarity with purchasing management and procurement issues.
Data analysis: Hierachical Moderated Multiple Regression method estimated the model at n = 164
REPOA MARCH 2017

13. Table 2: Correlation Matrix, average variance extracted (AVE) and Descriptive Statistics
Note: Values above the diagonal are the shared variances between constructs, while those below the diagonal are the inter-construct correlation estimates.
* Correlation is significant at the 0.05 level (2-tailed). a mean centered variables
All factor-items loadings were above 0.4, cronbach alpha was above 0.7 for each of the constructs, therefore internal consistency and data reliability was supported
Confirmatory factor analysis using AMOS: resulted into satisfactory model fit indices were; χ2 =352.2 df 262, p<0.01, IFI =0.951; TLI =0.95; CFI =0.95, and all indices fell within the cutoff point of 0.9 (Byrne, 2010), and the RMSEA value (0.046) was within the 0.05-limit as proposed by Byrne (2010).
REPOA MARCH 2017

14. Convergent and discriminant validity
Convergent validity demonstrated: All factor loadings were significant and greater than 0.5 with t values > 2.00 (Droge et al., 2004).
Discriminant validity demonstrated: The estimated average variance extracted (AVE) was greater than the percentage of variance shared by each construct (confer Table 2)
- except for buyer-supplier integration (SUPPLINT) and environmental uncertainty (UNCERT) which had AVE-scores below 0.5, i.e.0.46 and 0.44 respectively. However, the validity assessments revealed that SUPPLINT and UNCERT satisfied the other criteria above and support satisfactory discriminant for the research model.
REPOA MARCH 2017

15. REGRESSION MODEL
REPOA MARCH 2017

16. First test of the significance of interaction effects
The results demonstrated that when the interaction term (CENTRALxSUPPLINT) was introduced into the regression model, the model's explanatory power increased by 0.02 (2%). This corresponds to an F-value of 4.55 and exceeds the critical F-value of 3.84 for F (1,156) at a significance level of p<0.05, and demonstrates that the interaction term contributes significantly to the explanatory power of our model.
REPOA MARCH 2017

17. Second test of significance of interaction effect
Based on, Aiken and West (1991) and Schoonhoven (1981) approach (Estimated effects of buyer-supplier integration on supplier logistics performance at different levels of purchasing centralization (+/-2 scale units, s.u) were estimated as shown in table 2 below “Recall that CENTRAL and SUPPLINT are mean-centered to mitigate multicollinearity problems)
(1) SUPPLPERF/SUPPLINT = b6 + b7 CENTRAL
By inserting the data from the regression outputs, we get:
(2) SUPPLPERF/SUPPLINT = *CENTRAL
 Table 2:
Value of centralization (CENTRAL): Low (-2 s.u) Mean High (+2 s.u)
 SUPLPERF/ SUPPLINT:
Significance: t-values .
Implications: at low and medium levels of centralization (CENTRAL), the effect of buyer-supplier integration (SUPPLINT) on supplier logistics performance (SUPPLPERF) is significant and high (the slope is 0.53, p<0.01 and 0.33, p<0.01 for these levels). This effect vanishes when centralization increases to a "high" level where the slope is 0.13 (p>0.10), and demonstrates how substantial centralization of decision control erodes the effect of supplier integration on supplier logistics performance.

18. Discussions and conclusions
Buyer-supplier integration significantly improves supplier logistics performance “ as information exchange and collaboration (joint action in solving problems) between a public health facility and MSD increased, MSD’s delivery performance increases. Similar to Devaraj et al., (2007); Sillanpää et al., (2014)
REPOA MARCH 2017

19. Discussions and conclusions
Substantial centralization of purchasing decisions (deprivation of autonomy/decision making authority) in public health facilities weakens the effects of buyer-supplier integration on supplier logistics performance.
As medical and regional medical offices increase their influence in medicines procurement decision making, the role of bilateral governance mechanisms such as buyer-supplier integration in improving MSD’s delivery performance weakens
REPOA MARCH 2017

20. Discussions and conclusions
extensive buyer-supplier integration should not be paired with substantial centralization of buying firms' decision control(Koufterous et al., 2014)
Support Galbraith (1974) who argues that substantial centralization of decision control erodes organization communications and information flows.
REPOA MARCH 2017

21. Theoretical implications
(i) the combined effect shows that there is a still substantial value in organizational design theory (Koufterous et al., (2014)
(ii) On the Resource Based View theory (RBV): It contributes on the existing debate on the effects of supply chain integration on logistics performance (Fabbes-Costes and Jahre, 2008).
It acknowledges buyer-supplier integration as an important resource/capability (Fantazy et al., 2011) in improving logistics performance.
further shows that the effectiveness of a resource in improving performance is influenced by contextual factors such as organizational structure. It should be understood that resources do not exist in isolation, they can bundle with other factors the extent that their effect on performance may be enhanced or weakened. .
REPOA MARCH 2017

22. In order to improve MSD’s delivery performance:
the government and its policy makers in the health sector should reduce the bureaucracy/decision control of DMOs/RMOs in the decision making process of purchasing essential medicines ( too many layers); as it weakens information exchange (inter-coordination) and collaboration between the public health facility and MSD ( it intensifies information asymmetry; and chain invisibility).
“This may be critical to dispensaries and health facilities which are highly controlled by the DMOs office”.
Respective purchasing managers should be given more autonomy on tactical and operational decision making in order to support improvement of MSD’s logistics operations and eventually logistics performance.
REPOA MARCH 2017

23. Managerial Implications
The government, and practitioners should provide a more supportive environment to strengthen information exchange and collaboration between public health facilities and MSD (e.g. information systems)
REPOA MARCH 2017

24. Limitations & future research
Limited external validity- a highly regulated sector
Cross sectional study
This study used only one dimension of organisation structure “centralisation”, future research may incorporate other dimensions like formalisation, standardisation.
In addition this paper collected data from only one side “the buyer”; supplier perspectives may also be important.
REPOA MARCH 2017

25. Thank you
REPOA MARCH 2017