Zero-sum and frontier trade-offs: an investigation on compromises and compatibilities amongst manufacturing capabilities

Abstract

This paper examines well-accepted methodologies and rationales used to assert the presence/absence of trade-offs and compatibilities between manufacturing capabilities, and comparisons are made with respect to more recent theoretical developments. By means of a detailed analysis of a representative framework, important limitations and inconsistencies are identified. The proposal for the existence of zero-sum and frontier trade-offs intends to resolve some of those issues. Generally speaking, zero-sum trade-offs are identified by a statistically significant and negative correlation. On the other hand, frontier trade-offs are predicted to be observed after some level of compatibility between different capabilities is achieved. This means that, apart from widely-used probabilistic methodologies such as linear regression/correlation analysis, subsequent deterministic approaches and rationales need to be applied in order to acknowledge the potential existence of frontier trade-offs. These and other implications are discussed in light of previous studies, and suggestions for future research are offered.

Keywords:

• trade-off model
• compatibilities
• cumulative capabilities
• zero-sum trade-offs
• frontier trade-offs
• theory of performance frontiers
• probabilistic analyses
• deterministic analyses
• top performers/extreme cases
• thought experiments

Notes

Notes

1. The law of diminishing returns entails a proportional (e.g., linear) relationship between two variables that becomes a disproportional (e.g., non-linear) one after a certain point in the relationship is reached. Thus, when Schmenner and Swink (1998) state that their framework rests on this premise, it can be derived plausibly that the ‘cumulative capabilities’ they speak of in their framework are represented by the proportional aspect of the relationship, whereas ‘trade-offs’ are exemplified by the non-proportional aspect of it (see for example the discussion on page 110 of Schmenner and Swink 1998).

2. Although using perceptual measures of performance is consistent with the trade-off model's aims and goals, it could also be possible to utilise objective measures of achievement (e.g., ‘98% on-time delivery rates’, ‘30% of defective product’), as long as the sample includes firms from the same industry and similar context, or comparable competitors so that reasonable comparisons with respect to the marketplace-leading performers can be made.

3. Figure 3 can also be understood as an ideal ‘picture’ of a single firm as it advances in the development of two capabilities that observe both ‘cumulative capabilities’ and (frontier) ‘trade-offs’, as implied by several authors.

4. One way in which past studies assert the absence of a negative and significant relationship (and hence the absence of a (zero-sum) trade-off) is by means of a positive and significant coefficient between two capabilities. Another way is by means of a non-significant coefficient.

Additional information

Notes on contributors

Roberto Sarmiento

The author is no longer at this institution.