Experiential knowledge and export expansion

From the analysis of the preliminary case studies, it was revealed that experiential knowledge in the domain of alliance management and foreign marketing may give rise to capabilities in these areas. The capabilities of alliance management (ALLMGMTCAP) and the marketing planning and implementation capability (MKTCAP) were linked to export expansion. This model consisted of four struc-tural paths: from the ALLMGMTCAP to SPEED, ALLMGMTCAP to SUC-CESS, MKTCAP to SPEED and MKTCAP to SUCCESS. The structural path model for hypotheses 1a and 1b and 2a and 2b is shown in Figure 18.

Figure 18. Structural path model for hypotheses 1 and 2 (a and b).

** show significance at < 0.05 level; *** show significance at < 0.01 level

Speed was measured in terms of how satisfied firms were in the timely execution of the following activities: being ahead of key competitors, capturing the key ex-port market, introducing the right product to customers when they needed it and entering into a market at a time when profit margin opportunities were still avail-able. In this section, the structural model related to experiential knowledge and the success of export expansion is presented. Success was measured with SUC-CESS of export expansion in terms of financial indicators as well as the firm’s satisfaction with product performance, managing the timeframe for the overall

Alliance management capability R²=0.12

Marketing planning &

implementation capability

R2=0.16

Speed

R2=0.24 0..28 ***

0..31****

Success

R2=0.21 0..35 ***

0.14

project, relationships with alliances in the foreign market and the performance of the firm.

The measurement model was tested by examining a) internal consistency b) con-vergent validity and c) discriminant validity. For the reliability assessment of this model the internal consistency of individual items as well as constructs was ana-lyzed first. The individual item loadings and the composite reliability of the re-flective latent constructs are shown in Table 22.

Table 22. Loadings, cross loadings and the composite reliability of the structural relationship

The items RPLACMNT and TARSAVOL were dropped from the scale because of negative loadings. All other items for both ALLMGMTCAP and MKTCAP indicated ≥ 0.5 loadings. The composite reliability for all the reflective latent

constructs was≥ 0.7. Thus, all the reflective constructs showed good internal con-sistency of measures.

The values for average variance extracted demonstrated that both the latent inde-pendent variables captured 59 % and 66 % variance from their indicators respec-tively, whereas the latent dependent variables of SPEED and SUCCESS captured 57% and 43 % respectively. Further, the cross loadings of the latent variables indicated that all the indicators except one loaded higher on their respective latent variables than the other blocks in the model. KYEXPMKT from the construct of SPEED loaded high on the construct of SUCCESS indicating issues with dis-criminant validity. This is shown in brackets and bold in Table 22. The AVEs and the cross loadings are shown in Table 22.

The discriminant validity of the model showed that all the square roots of the AVEs of the latent variables were larger than the correlations among the latent variables. This is shown in Table 23. The correlation between speed and success was only slightly lower than the square root of the AVE of the construct of SUC-CESS. This could have been fixed if the item KYEXPMKT was removed. This item from SPEED loaded higher on SUCCESS as mentioned above. However, as the correlation value for SPEED was still lower than the square root of the AVE of the construct of SUCCESS, and as there are no clear guidelines in literature as to how much greater the AVE squared along the diagonal should be, it was ac-cepted as thus. Therefore, the structural model demonstrated internal consistency at the standard levels of indicator reliability and construct reliability, convergent validity and discriminant validity.

Table 23. Inter-construct correlations among the reflective constructs and the AVE squared along the diagonal

Latent variables ALLMGMTCAP MKTCAP SPEED SUCCESS

ALLMGMTCAP 0.81

MKTCAP 0.070 0.76

SPEED 0.263 0.299 0.75

SUCCESS 0.347 0.118 0.613 0.655

The analysis of R²of the latent dependent variables SPEED and SUCCESS dem-onstrated moderate explanatory power of the model. Further, the redundancy in-dexes Q² indicated that both the latent dependent variables demonstrated predic-tive relevance of the model. The effect size of the model indicated that ALLMGMTCAP at the structural level had the largest effect on the latent

de-pendent variable of SUCCESS and that MKTCAP had the largest effect on SPEED. R², Q²and f²are presented in Table 24.

Table 24. R², Q², and f ²for hypotheses 1 and 2 (a and b)

Latent variables R² Q²

f 2

SPEED

f2

SUCCESS

SPEED 0.24 0.035 - -

SUCCESS 0.21 0.030 - -

ALLMGMTCAP 0.12 0.28 0.35

MKTCAP 0.16 0.31 0.14

The significance of the structural relationships was assessed on the basis of the statistical significance of path coefficients using t-tests. The observed probabili-ties of the t-values were compared to the standard significance levels of probabil-ity. Furthermore, the confidence intervals and t-values for individual path coeffi-cients were assessed. Path coefficoeffi-cients for the sample and resample in this case were created by the bootstrap.

The first structural relationship between ALLMGMTCAP and SPEED was highly significant (β=0.28; t=2.59; p<0.05 and >0.01). The observed probability was closer to 0.01 significance level of p=0.01. The first structural relationship was captured in terms of hypothesis 1a, which stated that the alliance management capability of an exporting firm will have a positive relationship with the speed of export expansion. Thus, hypothesis 1a was accepted at the significance level of 99

%. The second relationship between ALLMGMTCAP and SUCCESS was also highly significant (β=0.35; t=2.06; p<0.05). The second structural relationship was captured in terms of hypothesis 1b, which stated that the alliance manage-ment capability of an exporting firm will have a positive relationship with the success of export expansion. Thus, hypothesis 1b was accepted at an observed significance level of 95 %.

An analysis of each of the t-values for ALLMGMTCAP with both the dependent variables revealed that commitment from the firms in jointly sharing the problems and helping the alliance relationships significantly impacted on the speed of ex-port expansion. Maintaining a long-term relationship with alliances was consid-ered important in gaining a better understanding of the market. An analysis of t-values for individual indicators suggested that a firm’s ability to share in the prob-lems that arise in the course of dealing with major distributors contributed more

on the success of export expansion. Moreover, the commitment to improvement that could benefit the relationship with a major distributor as a whole had a greater impact on success than simply maintaining a long-term relationship with a foreign distributor.

The third structural relationship between MKTCAP and SPEED of export expan-sion (β= 0.31; t = 3.34; p = 0.001) was also highly significant. This was captured in terms of hypothesis 2a and stated that the marketing planning and implementa-tion capability of an exporting firm will have a positive relaimplementa-tionship to the speed of export expansion. Thus, hypothesis 2a was accepted at an observed signifi-cance level of 99 %.

The fourth structural relationship between MKTCAP and SUCCESS of export expansion (β= 0.14; t= 0.62; p> 0.05) was positive but statistically insignificant.

The path coefficient was also lower than the standardized beta coefficient of 0.20.

This can be interpreted as marketing planning and implementation capability that did not affect the success of export expansion, even though the ability of the firm to effectively translate planned export marketing strategies into actions was sig-nificant at p < 0.05 and > 0.01 significance level. This relationship in this model was captured in terms of hypothesis 2b and stated that the marketing planning and implementation capability of an exporting firm will have a positive relationship to the success of export expansion. Thus hypothesis 2b was not supported.