Managerial implications and limitations

From our study we can draw a number of management implications for Mexican avocado producers and buyers. First, increasing information exchange, in particular by buyers providing information to the producers on the quality and quantity of final demand, leads to higher performance. Thus, producers are advised to seek buyers that have a reputation of providing information to their suppliers. Second, establishing sustainable trading relationships between buyers and sellers will benefit both of them. Producers will benefit from secure sales, while packers will benefit not only from secure deliveries but also from producer commitment to comply with buyer demands.

In our study we have limited ourselves to two relationship characteristics that may affect behavioural uncertainty, viz., commitment and expectation of continuity. Other relationship characteristics mentioned in the literature on efficient inter-organizational relations are trust and reputation. Further research may explicitly include those other relationship characteristics, and see how they complement or substitute for commitment and/or the expectation of a continued relationship.

of the Mexican avocado industry APPENDIX 4.1 Constructs of variables

Factor analysis is used to calculate the loadings for constructs information exchange, buyer commitment, expectation of continuity, asset specificity, behavioural uncertainty and environmental uncertainty. Under this method, the loadings indicate how much of variance in each independent item is accounted for by the latent construct (Lattin et al., 2003). Thus, to determine whether constructs such as asset specificity, behavioural uncertainty, environmental uncertainty, buyer commitment, expectation of continuity and information exchange can be used in the model, each construct is evaluated in terms of individual item reliability, internal consistency (composite reliability and cronbach’s alpha ‘α’) and discriminant validity (average variance extracted and interconstruct correlations) (Fornell and Larcker, 1981).

Individual item reliability was determined by examining the loadings of measures on their corresponding constructs (see Table 4.6). Except for constructs behavioural uncertainty and environmental uncertainty, the rest of constructs present loadings greater than or close to 0.7 indicating a high degree of individual item reliability.

Internal consistency was assessed using two measures: composite reliability and cronbach’alpha. Regarding composite reliability, an internal consistency of 0.7 or greater is reasonable for exploratory research. All the constructs exceed 0.70 (see Table 4.7) indicating a good internal consistency. In terms of cronbach’alpha, a minimum reliability of 0.7 is required. Except for the constructs behavioural uncertainty and environmental uncertainty, the rest of constructs present values ‘α’ closer to or higher than 0.7 (see Table 4.6)

The discriminant validity was carried out in two ways. First, the square root of the variance extracted (the numbers on the diagonal in Table 4.7) should be greater than all construct correlations (the numbers on the off-diagonal in Table 4.7), as is the case. Second, the test involves assessing how each item is related to the latent constructs. Table 4.8 reports the item loadings and cross-loadings on the constructs. For all constructs, no item loaded higher on the other constructs than it did on its associated construct. Both criteria indicate that the discriminant validity of the constructs used in the model is satisfactory.

Based on these three tests (individual item reliability, internal consistency, and discriminant validity) we can confidently rely on the constructs asset specificity17_{, buyer commitment,} expectation of continuity, and information exchange. On the contrary, because the constructs behavioural and environmental uncertainties were not valid, we individually incorporated their items in the model.

17 _{For asset specificity, we eliminated the item related to losing part of the physical investment when the} producer switches to another buyer (Physical asset specificity). The new construct for asset specificity passed the tests for reliability, internal consistency, and discriminant validity.

Table 4.6 Constructs and items used in the model for the producer sample

Constructs Loadings

Information exchange (= 0.85, eigenvalue= 2.58) Information exchange

planning

We receive information to help us plan according to his needs

0.74 Information exchange

product requirements We are frequently informed of his product requirements 0.87 Information exchange

forecasting

We are provided with long-range forecasts of supply requirements

0.68 Information exchange

preference and requirements

We are informed in advance of impeding changes

in preferences and requirements 0.80

Buyer commitment (= 0.85, eigenvalue= 1.53) Buyer commitment

helping

Our main buyer tries to help us when we incur problems

0.70 Buyer commitment

sharing

Our main buyer shares in the problems that arise in the course of dealing

0.74 Buyer commitment

improving Our main buyer is committed to improvements that benefit our relationship 0.68 Buyer commitment

assistance

Our main buyer has supported us with technical assistance and inputs

0.84

Expectation of continuity (= 0.67, eigenvalue= 2.66) Expectation of continuity

a long time We expect our relationship to continue a long time 0.85 Expectation of continuity

renewal

Renewal of the relationship is virtually automatic 0.73

Asset specificity (= 0.74, eigenvalue = 2.94) Human asset specificity We have made significant investments in training

of workers and in equipment specific for our main buyer

0.90

Dedicated assets We have made significant investments in fulfilment of production requirements for our main buyer

0.94

Behavioural uncertainty (= 0.58, eigenvalue = 1.95) Payment uncertainty We are uncertain whether our buyer will stick to

the payment agreement

0.99 Damage uncertainty We are uncertain whether our buyer will damage

the orchard 0.43

Environmental uncertainty (= 0.47, eigenvalue = 1.08) Price uncertainty The price for my product varies significantly over

the seasons

0.92 Demand uncertainty There are always many buyers for my product 0.65

of the Mexican avocado industry

Table 4.7 Description of the Constructs [Mean (M), Standard Deviation (SD), Composite Reliability (CR), Average Variance Extracted and Intercorrelations of the Constructs]

Sample n = 122

Constructs M SD CR 1 2 3

1. Asset Specificity 4.51 1.90 0.92 0.84

2. Behavioural Uncertainty 3.08 1.93 0.72 -0.07 0.60

3. Environmental Uncertainty 5.06 1.53 0.77 -0.13 0.31 0.63

[Mean (M), Standard Deviation (SD), Composite Reliability (CR), Average Variance Extracted and Intercorrelations of the Constructs]

Sample n = 122

Constructs M SD CR 1 2 3

1. Buyer commitment 1.74 1.72 0.90 0.69

2. Expectation of continuity 1.45 1.46 0.91 0.47 0.74

3. Information exchange 2.50 2.14 0.85 0.50 -0.23 0.71

Note: The boldface numbers on the diagonal are the square root of the variance shared between the constructs and their measures (square root of Average Variance Extracted). Off-diagonal elements are correlations among constructs.

Table 4.8 Construct to Measure Item, Loadings and Cross-Loadings

Items Information

exchange Commitment Continuity

Information exchange planning 0.745 0.222 0.254

Information exchange product

requirements 0.873 0.117 0.211

Information exchange forecasting 0.684 0.326 0.386

Information exchange preference and

requirements 0.804 0.226 0.149

Buyer commitment helping 0.426 0.700 0.248

Buyer commitment sharing 0.298 0.736 0.303

Buyer commitment improving 0.235 0.681 0.563

Buyer commitment assistance 0.076 0.837 -0.072

Expectation of continuity a long time 0.200 0.079 0.847

Expectation of continuity renewal 0.358 0.184 0.729

Items Asset specificity Behavioural _Uncertainty Environmental _Uncertainty Asset specificity in terms of training

of workers 0.900 -0.055 -0.065

Asset specificity in terms of

production requirements 0.937 -0.068 -0.161

Payment uncertainty -0.067 0.990 0.313

Damage uncertainty 0.082 0.437 0.004

Price uncertainty -0.145 0.311 0.919

Interfirm coordination in the Mexican avocado