The Effects of Product Line Breadth: Evidence from the Automotive Industry

Abstract

Using a detailed dataset from the U.S. automotive industry, we empirically study the benefits and costs of maintaining a broader product line. Consistently with theoretical predictions, we find a positive association between product line breadth and both market share and unit production costs. We enrich the existing literature with new results that highlight previously unexplored operational aspects of the benefits and costs of product line breadth. Besides the effects on production costs through economies of scale, we find that product line breadth has a significant effect on mismatch costs arising from increased demand uncertainty, which are manifested through additional discounts and inventories. The effect of product line breadth on mismatch costs is comparable in magnitude to the effect on production costs: an additional product in the line is associated with an average increase of around $100 in discounts per vehicle and with carrying three additional days of supply in the average inventory of all the models of the line. This suggests that the operational benefits of inventory pooling that can be achieved by reducing product lines can be economically very substantial. Furthermore, we quantify the benefit of using a platform strategy to mitigate the effects of a broad product line on production costs - on average, for every 100,000 vehicles produced for other models based on the same platform, unit production costs are reduced by $55. Finally, we propose an additional, attribute-based measure of product line breadth and find that product line breadth can work as a hedge against changes in demand conditions. For example, automakers that offer a broader range of fuel economy levels increase their market share and reduce their average discounts as gas prices become more volatile.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 37 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.