Sarah Moshary

REVISION: How and When to Use the Political Cycle to Identify Advertising Effects
Date Posted:Mon, 23 Nov 2020 10:35:33 -0600
A central challenge in estimating the causal effect of TV advertising on demand is isolating quasi-random variation in advertising. Political advertising, which topped $14 billion in expenditures in 2016, has been proposed as a plausible source of such variation and thus a candidate for an instrumental variable. We provide a critical evaluation of how and where this instrument is valid and useful across categories. We characterize the conditions under which political cycles theoretically identify the causal effect of TV advertising on demand, highlight threats to the exclusion restriction and monotonicity condition, and suggest a specification to address the most serious concerns. We test the strength of the first stage category-by-category for 274 product categories. For most categories, weak-instrument robust inference is recommended, as first-stage F-statistics are less than 10 for at least 221 of 274 product categories in our benchmark specification. The largest first-stage ...

New: Beyond Consumer Switching: Supply Responses to Food Packaging and Advertising Regulations
Date Posted:Thu, 03 Sep 2020 04:21:23 -0500
This paper studies the effect of nutrition warning labels and advertising restrictions on the breakfast cereal market in Chile. In June 2016, the Ministry of Health required food products that exceed thresholds for sugar (22.5g) and calories (350kcal) to carry conspicuous front-of-package warning labels. Further, these products were barred from advertising on TV programs with high child viewership. Early evidence suggests that the regulation induce consumers to switch to products without warning labels; we show that this change in demand elicits a supply response. In particular, we present evidence of bunching just below the cutoffs. Using a structural model of cereal demand, we find that reformulation tends to reinforce the intent of the reform, in particular, by lowering the calorie content of cereal purchases.

REVISION: How and When to Use the Political Cycle to Identify Advertising Effects
Date Posted:Mon, 09 Mar 2020 11:22:51 -0500
A central challenge in estimating the causal effect of TV advertising on demand is isolating quasi-random variation in advertising. Political advertising, which topped $14 billion in expenditures in 2016, has been proposed as a plausible source of such variation and thus a candidate for an instrumental variable. We provide a critical evaluation of how and where this instrument is valid and useful across categories. We characterize the conditions under which political cycles theoretically identify the causal effect of TV advertising on demand, highlight threats to the exclusion restriction and monotonicity condition, and suggest a specification to address the most serious concerns. We test the strength of the first stage category-by-category for 274 product categories. For most categories, weak-instrument robust inference is recommended, as first-stage F-statistics are less than 10 for at least 221 of 274 product categories in our benchmark specification. The largest first-stage ...

REVISION: Politics, Persuasion and Choice
Date Posted:Wed, 20 Nov 2019 10:50:15 -0600
In the past decade, we have seen a certain convergence between politics and marketing, in sometimes unexpected ways. Political campaigns now routinely use marketing tools to persuade voters. Brands, on the other hand, are engaging with increasing likelihood with explicitly-political movements of interest to their customers. Moreover, both fields have long had an ambivalent relationship with the concept of truth and, as a result, we typically observe significant variation across individuals’ beliefs over optimal choices. In this paper, we report on, summarize, and extend our discussions at the 2019 Choice Symposium in which our session had diverse representation from marketing, economics and practice. We organize our ideas around three dimensions of political beliefs: 1) What are the characteristics of political beliefs?, 2) Who drives beliefs in the political setting and how? 3) What mechanisms facilitate truthful outcomes?

REVISION: Politics, Persuasion and Choice
Date Posted:Tue, 12 Nov 2019 12:41:21 -0600
In the past decade, we have seen a certain convergence between politics and marketing, in sometimes unexpected ways. Political campaigns now routinely use marketing tools to persuade voters. Brands, on the other hand, are engaging with increasing likelihood with explicitly-political movements of interest to their customers. Moreover, both fields have long had an ambivalent relationship with the concept of truth and, as a result, we typically observe significant variation across individuals’ beliefs over optimal choices. In this paper, we report on, summarize, and extend our discussions at the 2019 Choice Symposium in which our session had diverse representation from marketing, economics and practice. We organize our ideas around three dimensions of political beliefs: 1) What are the characteristics of political beliefs?, 2) Who drives beliefs in the political setting and how? 3) What mechanisms facilitate truthful outcomes?

REVISION: How and When to Use the Political Cycle to Identify Advertising Effects
Date Posted:Wed, 07 Aug 2019 10:31:33 -0500
A central challenge in estimating the causal effect of TV advertising on demand is finding quasi-random variation in advertising. Political advertising in the US has been proposed as a plausible instrumental variable because political spending on television has skyrocketed in recent elections, topping $4 billion in 2016. We take a multi-category approach to assessing how and where this instrument is valid and useful. We characterize the conditions under which political cycles theoretically identify the causal effect of TV advertising on demand, highlight potential threats to the exclusion restriction and monotonicity condition, and suggest a specification to address the most serious concerns. To characterize "where' the approach might be most useful, we test the strength of the first stage of the political ad instrument, using both a single instrument and also optimal instruments obtained by machine learning. For the majority of commercial advertising categories, our findings ...

REVISION: How and When to Use the Political Cycle to Identify Advertising Effects
Date Posted:Sun, 30 Jun 2019 06:31:29 -0500
A central challenge in estimating the causal effect of TV advertising on demand is finding quasi-random variation in advertising. Political advertising in the US has been proposed as a plausible instrumental variable because political spending on television has skyrocketed in recent elections, topping $4 billion in 2016. We characterize the conditions under which political cycles theoretically identify valid TV advertising effects on demand and highlight the potential areas of concern for the exclusion restriction and monotonicity condition. To characterize ``where' the approach might be most useful, we test the strength of the first stage of the political ad instrument using both a simple specification of the first stage and also optimal instruments obtained by machine learning. For the vast majority of commercial advertising categories, our findings suggest that researchers should consider using weak-instrument robust inference, as first stage F-statistics are less than 10 for at ...

REVISION: How and When to Use the Political Cycle to Identify Advertising Effects
Date Posted:Thu, 20 Jun 2019 12:36:12 -0500
A central challenge in estimating the causal effect of TV advertising on demand is finding quasi-random variation in advertising. Political advertising in the US has been proposed as a plausible instrumental variable because political spending on television has skyrocketed in recent elections, topping $4 billion in 2016. We characterize the conditions under which political cycles theoretically identify valid TV advertising effects on demand and highlight the potential areas of concern for the exclusion restriction and monotonicity condition. To characterize ``where' the approach might be most useful, we test the strength of the first stage of the political ad instrument using both a simple specification of the first stage and also optimal instruments obtained by machine learning. For the vast majority of commercial advertising categories, our findings suggest that researchers should consider using weak-instrument robust inference, as first stage F-statistics are less than 10 for at ...

New: Market Structure and Competition: Evidence from a Natural Experiment in Liquor Licensure
Date Posted:Wed, 24 Jan 2018 07:08:51 -0600
Washington state licensed private retailers to sell spirits for the first time in May 2012, but only if their premises exceeded 10,000ft. This restriction generates exogenous variation in the number of retailers across local liquor markets, which we leverage to estimate the causal effects of market structure on equilibrium outcomes. We find that spirits purchasing increases by 63% when moving from monopoly to duopoly markets, and that this increase is concentrated among the heaviest-drinking households. These results support the notion that local liquor availability can dramatically increase consumption. However, these effects dissipate quickly as the number of competitors increases, highlighting that the square-footage based licensure rule is a blunt policy tool. Surprisingly, price competition does not play an important role in the observed quantity increase. Instead, we find that firms compete in product assortment, tailoring product mix to the local competitive environment while ...

Update: Estimating the Effect of Potential Entry on Market Outcomes Using a Licensure Threshold
Date Posted:Sun, 28 Aug 2016 14:06:03 -0500
We study the effects of potential entry on market outcomes in the context of Washington state's 2012 switch from a state-run monopoly to private liquor sales. Concern about alcohol-related crime prompted regulators to institute a 10,000 square foot licensure requirement to curtail entry. This store size threshold generates plausibly exogenous variation in the number of eligible entrants in local liquor markets across the state. We find that widening the pool of potential entrants has a small effect on pricing, but a significant effect on product offerings. In particular, markets with more potential entrants see a compositional shift in product offerings towards cheaper goods. Further, we find that the size requirement changes the composition of entrant size, rather than the aggregate number of entrants.
New PDF Uploaded

New: Estimating the Effect of Potential Entry on Market Outcomes Using a Licensure Threshold
Date Posted:Mon, 07 Dec 2015 01:57:54 -0600
We study the effects of potential entry on market outcomes in the context of Washington state's 2012 switch from a state-run monopoly to private liquor sales. Concern about alcohol-related crime prompted regulators to institute a 10,000 square foot licensure requirement to curtail entry. This store size threshold generates plausibly exogenous variation in the number of eligible entrants in local liquor markets across the state. We find that widening the pool of potential entrants has a small effect on pricing, but a significant effect on product offerings. In particular, markets with more potential entrants see a compositional shift in product offerings towards cheaper goods. Further, we find that the size requirement changes the composition of entrant size, rather than the aggregate number of entrants.