In a study my co-authors and I investigate the contingent role of product architecture on the relationship between buyer-supplier network structure and product quality. The hypotheses are tested on a combination of primary and secondary data from the North American automobile industry. The results suggest that correspondence between buyer-supplier networks and product architecture networks may not be conducive for product quality. While buyer network density (or tight coupling) is better suited for improving the product quality of modular systems, buyer network centrality (or loose coupling) is better suited for improving the product quality of integral systems. The implication is that greater information sharing between buyers and suppliers is more likely to improve product performance of modular products as compared to integral products.
To test our hypotheses, we assembled data using a combination of archival and primary sources. We focus on firm-system as the unit of analysis. This choice was guided by the observation that there is considerable variation in the composition of supplier networks for different systems of a given firm. We collected data on buyer-supplier links for twelve vehicle systems in 2010. The systems we examine are as follows: transmission, major engine system, engine cooling, suspension, drive systems, exhaust, electrical, power equipment, body hardware, braking system, climate system and fuel system. We focus on these 12 vehicle systems instead of more disaggregate subsystems or vehicle components to keep the functional interactions or dependencies tractable (Novak and Eppinger 2001, Gokpinar et al. 2010). Additionally, a practical consideration was that data on product quality are available at the vehicle system level but not at the subsystem level. We identify suppliers of thirteen automobile manufacturers with significant operations in North America (BMW, Chrysler, Ford, General Motors, Honda, Hyundai, Kia, Mazda, Mercedes-Benz, Mitsubishi, Nissan, Subaru and Toyota). The thirteen auto manufacturers accounted for approximately 85 percent of vehicles sold in North America. The data collection efforts yielded a total of 964 unique suppliers for 13 automobile firms and 12,667 unique B-S relationships. The following figure presents an illustrative supply network for the electrical system network comprising of 3,656 ties between 507 entities (13 automakers and 494 suppliers).
The data on quality of auto manufacturer’s vehicle systems for 2011 was assembled from Consumer Reports. There are two important issues to be considered in matching B-S relationships with quality outcomes. First, data on B-S relationships was available at the ‘automaker-vehicle system’ level whereas data on product quality is at the ‘model-vehicle system’ level. For example, consider the case of Subaru of America, the automaker. While we constructed the supplier network for Subaru, the quality data is available at the model level (e.g., Outback, Legacy) of Subaru. We aggregated model-level quality by weighting on model sales to construct the automaker’s quality for a vehicle system. Second, we only included suppliers of models that were manufactured in North America. For example, Subaru of America manufactured three of its five models (i.e., Outback, Legacy and Tribeca) in its North American assembly plants. We identified suppliers of these three models of Subaru and excluded suppliers of two manufactured outside North America. Likewise, we aggregated product quality for Subaru by using sales weights for the three models manufactured in North America.
We relied on primary data sources to operationalize the Strength of Design Interface (SDI) variable. We conducted 31 in-depth interviews with product and design engineers in automobile and supplier firms to gauge the dependencies between various systems in a vehicle. The in-depth interviews lasted approximately 90 minutes in duration. The experts participating in the survey were employed at General Motors (USA and India), Ford, Chrysler, Honda, Toyota, BMW, Porsche, Hyundai (USA and South Korea), Aisin (USA), Bosch (USA and South Korea), Cummins, Denso, Mando, SAIC Motors, Maruti (India) and Ashok Leyland (India) and had, on average, approximately 13 years of experience in automotive design. To assess SDI, we asked product and design engineers to evaluate the degree of information, spatial, structural, material and energy dependencies between vehicle systems (see ‘Measures Section’ for details on the questionnaire items). We also asked these experts whether the level of dependency criticality and absolute criticality of systems vary across major automakers. All but one expert indicated that dependencies do not vary across automakers. In other words, the strength of the design interface is system-specific and is not contingent on the automobile manufacturer.
We also collected data on several supplier and buyer characteristics. Data on the size of the supplier’s plants (in square feet), distance in miles between the supplier and auto manufacturer, the union status of suppliers, and quality awards of supplier were obtained from ELM Analytics. We also collected data on buyer characteristics such as sales from the Ward’s Automotive Yearbook and R&D expenditures from COMPUSTAT and annual company reports.
A test of our hypotheses requires a close alignment between the theory, measures and the empirical model. We model the impact of buyer-supplier network characteristics in time period ‘t-1’ on product quality in time period ‘t’. The temporal separation facilitates a more confident interpretation of the quality consequences of B-S network structure. Our final dataset for empirical analyses is comprised of 156 firm-system observations (13 makes x 12 systems). The B-S network characteristics and control variables are for 2010 and product quality data is for 2011.
Our findings show that buyers in dense networks experience better product quality outcomes compared to buyers that are centrally located. We find that integral systems have higher quality compared to modular systems. More importantly, our results reveal that buyer network density enhances the quality of modular systems to a greater extent compared to integral systems. On the other hand, a central position for buyers is beneficial for improving product quality for integral systems than for modular systems.
The findings of this study offer numerous valuable insights for managerial practice. In recent years, there has been increased emphasis on improving supply chain performance. The emphasis has been on using concepts such as modularization, flexible and lean manufacturing to boost performance. Despite such advances, the poor performance of supply chains, as evidenced in growing number of product recalls in the marketplace, continues to frustrate and disenchant managers. Our findings caution managers to refrain from viewing network density as beneficial and network centrality as detrimental to quality. Our findings, instead, point to the need to evaluate the normative B-S network structure in conjunction with the architecture of the product. We performed a univariate transfer function analysis to understand the economic value added by matching buyer-supplier network structure and product architecture. We computed the direct impact of changes in automakers’ network density and network centrality on product quality for high and low levels of SDI. The high and low levels for buyer network density, buyer network centrality and SDI are set at the 90th and 10th percentiles. To compute the dollar impact of change in product quality, we turn to previous research that has assessed the shareholder value impact of product quality. Tellis and Johnson (2007) note that increasing product quality from 10th percentile to 90th percentile value leads to a 9.35% increase in shareholder wealth. The average market capitalization of publicly traded automobile firms in our sample is $73.37 billion. In our sample, this translates to an increase of $6.86 billion. The difference in product quality between 90th and 10th percentile in our sample is 1.87 points. Thus, a one point increase in product quality corresponds to an increase in market capitalization of $3.67 billion.
Our post-hoc analyses reveal that if buyer network density is increased from the 10th percentile to the 90th percentile, there is significant increase in product quality for low SDI systems relative to high SDI systems (difference in quality = 0.11 points). That is, when buyer network density is increased, there is a greater increase in product quality for modular systems than for integral systems. The total economic value created when buyer network density is aligned with modular systems is $403.56 million. Although dense networks are costly to maintain because of intensive information sharing, our post-hoc analyses reveal that buyers could harness considerable economic value from dense ties by aligning it with the appropriate product architecture. Similarly, if buyer network centrality is increased from the 10th percentile to the 90th percentile, there is significantly greater increase in product quality for high SDI (system design interface) systems relative to low SDI systems (diff = 0.05 point change). Correspondingly, there is greater increase in shareholder wealth at high levels of SDI than at low levels of SDI. The total economic value created when buyer network centrality is aligned with integral systems is $183.43 million.
Source: Kalaignanam, K., Kushwaha, T. and Nair, A., 2015, January. Buyer-supplier network structure and product architecture. In Academy of Management Proceedings (Vol. 2015, No. 1, p. 13626). Academy of Management.