Theses Doctoral

Essays in Industrial Development

Guillouet, Louise

Firms are the unit cells of the economy. Understanding how they create value is key todesigning policies that promote sustainable growth. In this dissertation, I study how two major trends: globalization and rising inequality, affect the causes and consequences of firm growth. Chapters 1 and 2 focus on the interaction of multinational firms and domestic firms in developing countries, while chapter 3 looks at the unequal distribution of consumer gains from the expansion of a firm in the United States.

Specifically, in Chapter 1, I study how the presence of multinational firms affects how domestic firms grow. I investigate the hypothesis that uncertainty about product quality, a distinctive feature in developing countries, leads consumers to prefer products made by multinational firms headquartered in high-income countries, as opposed to domestic firms. Combining barcode-level consumption data from Mexico with information about the origin of the producers of the goods, I measure a precise foreign price premium of at least 16%. While the availability of foreign goods increases consumers’ welfare, the dominance of foreign firms may also hinder the growth of domestic firms. I then document the following novel facts about the consumer packaged goods industry in Mexico: 1) domestic firm sales growth is driven by older goods rather than new goods; 2) domestic goods have slower and longer life-cycles than foreign goods; 3) the extensive customer margin is key to growth for both types of firms; 4) domestic firms depend relatively more on the intensive margin for customer growth; and 5) new customers of older domestic goods are poorer than those of new goods. I estimate a demand model, showing that the price premium elicited in the raw data can be attributed to consumers’ relative preference for foreign goods. Importantly, this preference fades over time. I show that this is consistent with consumers learning about product quality, and provide consumer-level empirical evidence for this mechanism. Demand-side policies may be useful complements to classic industrial policy tools.

Chapter 2 looks inside multinational firms to understand how contextual factors may affect the probability of spillovers from multinationals to the domestic sector. A distinct feature of multinationals is a three-tier hierarchy: foreign managers (FMs) supervise domestic managers (DMs) who supervise production workers. Surveys suggest that language barriers impede interactions between FMs and DMs. An experimental protocol that offers DMs free English language courses confirms that lowering communication costs increases their interactions with FMs. A second experimental protocol that asks human-resource managers at domestic firms to rate hypothetical resumes reveals that multinational experience and, specifically, DM-FM interactions are valued in the domestic labor market. Taken together, the protocols suggest that reducing language barriers can improve transfers of management knowledge to domestic workers, and a longer-run survey indicates treatment DMs’ improvements in soft skills. We further examine why MNCs and DMs may under-invest in language training. Complementary policies such as language subsidies can increase the probability of positive spillovers from Foreign Direct Investment.

In Chapter 3, I study the expansion of a large, high-quality firm in the United States and itsimpact on the competitive landscape. The arrival of high-end grocery stores in neighborhoods is a harbinger of gentrification. However, economic theory generally predicts that the entry of firms is good for consumer welfare. This paper combines barcode-level retail data with a newly collected dataset on the opening dates of Whole Foods, a high-end grocery chain in the United States, in new neighborhoods, to estimate the effect of entry. I show that Whole Foods’ entry causes prices to rise by three percent for households in the bottom half of the income distribution, while prices don’t change for households in the top half of the income distribution. This finding is robust to changing the sample of stores and the set of control variables and to a falsification test using announcement dates instead of entry dates. Building on differentiated competition models, I show that this unexpected effect of entry can happen because incumbent stores catering to high-income households are closer to Whole Foods’ assortment and therefore behave pro-competitively when Whole Foods arrives, while incumbent stores catering to low-income households are quite differentiated and are able to raise their prices. Policies seeking to address gentrification should take the business side of this phenomenon into account.

Geographic Areas


  • thumnail for Guillouet_columbia_0054D_17189.pdf Guillouet_columbia_0054D_17189.pdf application/pdf 247 KB Download File

More About This Work

Academic Units
Thesis Advisors
Verhoogen, Eric A.
Ph.D., Columbia University
Published Here
April 27, 2022