Emil Mathias Strøm Halseth

Research

My fields are industrial organization, auctions, and grocery retail – how firms compete with each other in markets, studied with quasi-experimental methods, structural econometrics, and surveys.

Publications

• “The Effect of Cross-Border Shopping on Commodity Tax Revenue: Results from Norway's COVID-19 Border Closings” with Richard Friberg, Frode Steen, and Simen A. Ulsaker The Scandinavian Journal of Economics, 126(4), 733–772, 2024 – awarded the journal's prize for the best article published in 2024 Publisher (open access)
  Abstract

  We use grocery data from Norway and COVID-19 border closings to gauge the effect of cross-border shopping on commodity tax revenue. Detailed store-category level data identify differential treatment effects that depend on distance to Swedish stores. Economically significant effects extend to up to two hours' drive from the border, and even further for prominent cross-border shopping products as beer, cigarettes and soda. Across all products, cross-border shopping decreases tax revenue from VAT by 3.6% at the national level. National commodity tax revenue from carbonated soft drinks (subject to a sugar tax) is reduced by 8.1% and from cigarettes by 11.9%.

Working Papers

• “The Competitive Effects of Store-Format Repositioning: Evidence from a Norwegian Grocery Acquisition” PDF Online appendix
  Abstract

  This paper evaluates Coop's 2015 acquisition of ICA, which raised national concentration in Norwegian grocery retail and reshaped local competition through store-format repositioning toward soft-discount. Store-level difference-in-differences estimates for a leading soft-discount incumbent imply that rival conversions to the acquirer's soft-discount format reduced sales by 4.4–12.9%, while closures increased them by 2.4–11.1%. Convenience-discount conversions show no sales response. Local product prices were unchanged, consistent with uniform national pricing, and assortments adjusted little. Nationally, food prices fell 3.8–5.2% relative to neighboring countries. These findings suggest the merger reallocated complementary firm-specific assets – ICA's locations and Coop's soft-discount concept – strengthening competition despite greater concentration.

• “Do Informed Consumers Pay Less? Evidence from a Survey with Linked Grocery Purchase Data” with Richard Friberg, Frode Steen, and Simen A. Ulsaker PDF
  Abstract

  We examine how consumer price knowledge affects prices paid in grocery markets, combining survey-based price recall data from over 2,000 Norwegian households with 18 months of linked transaction histories. Better-informed consumers pay lower prices by timing purchases to coincide with sales: a 10 percentage-point increase in price knowledge is associated with a 1.3 percent reduction in prices paid. This provides direct empirical support for the central mechanism in Varian's (1980) model of sales. Consumers who actively seek price information have higher price knowledge, suggesting that policies improving price accessibility may improve consumers' ability to identify and act on temporary discounts.

Ongoing Projects

• Local Competition and Market Structure in Norwegian Grocery Retail with Frode Steen and Simen A. Ulsaker · project manager Supported by a research grant from the Norwegian Competition Authority (Konkurransetilsynet), 2026–2028.
  Project outline

  How does the structure of local grocery markets – the number of stores and their distribution across chains – affect competition, cost efficiency, and store accessibility? The Norwegian grocery sector is highly concentrated at the national level, yet competition largely plays out in local markets, and the policy debate has centred on measures that shape local market structure: the 2024 ban on negative servitudes, proposals to cap the number of stores per chain in a local market, and the Competition Authority's new market-investigation powers, which open for structural remedies such as divestiture. This project develops and estimates a structural equilibrium model of store entry and local competition between grocery chains, building on recent frameworks in empirical industrial organization and detailed data on store locations, chain affiliation, and local market conditions. The estimated model is then used for counterfactual analyses that quantify how concrete policy interventions would affect entry patterns, market structure, competitive intensity, and consumer welfare – explicitly accounting for the chains' strategic responses.

• Consumer Responses to Round-Price Promotions: Evidence from “Kronemarked” with Richard Friberg, Frode Steen, and Simen A. Ulsaker Early-stage project with the FOOD Research Project, NHH, using consumer-level grocery transaction data made available through the FOOD partnership.
  Project outline

  “Kronemarked” is a recurring, store-wide promotional event in Norwegian grocery retail: a large set of products spanning most categories is temporarily repriced by rounding regular prices down to a small set of round price points – 10, 15, 20, 30, 40, and 50 kroner – and included products carry a dedicated in-store label showing the round price. The event is heavily advertised and functions as a traffic driver for the store as a whole. Because regular prices sit at varying distances above the nearest round number, the monetary discount differs widely across included products: an item regularly priced at 21 kroner and one at 29 kroner are both sold at 20. The promotion thus separates two features that normally move together in a sale: the salient signal that an item is on offer, and the depth of the discount. Some products recur across campaigns, while others appear only once.

  The project asks two questions. First, what is the relative importance of inclusion and labelling versus discount depth in consumers' responses – that is, how sensitive is demand to the size of the discount, conditional on the product carrying the label? Second, what role does the promotion play in introducing new products and variants to consumers: does the chain use it to push particular products, and do purchases induced by the campaign persist afterwards, shifting consumers' longer-run preferences?

  We use consumer-level transaction data from a major Norwegian grocery chain, provided through the FOOD partnership, to study how households' purchases of included products change relative to the same products outside the campaign period, to the same products in stores that do not run the promotion, to non-included products in the same store, and during versus after the campaign. Variation in discount depth across products sold at the same round price point, and across campaigns for the same product, identifies the effect of discount depth conditional on the label. Because the distance from a product's regular price to the nearest round number is plausibly as good as random, conditional on the price level, this variation offers a quasi-experimental route to the causal effect of discount depth on demand. The household panel further allows us to trace trial, repeat purchase, and substitution over time.

• “Pre-empting Competitive Sales: Private Offers, Early Resolution, and Seller Information” Working paper, early draft PDF Online appendix Numerical workbook
  Abstract

  A private offer can end a competitive sale before other buyers bid. We study costly offers by a buyer with private value and resolution benefits to a seller with private continuation prospects. Accepted trades must finance foregone competition and unsuccessful attempts. Some buyers offer prices that screen seller states; others pay more for certain acceptance. An effective ban preserves later competition and saves offer costs, but sacrifices resolution gains. Its welfare effect depends on the surplus denied to excluded buyers. We establish probing and knockout equilibria and give numerical examples with opposite welfare rankings.

Master's Thesis

• “How Efficient Is Vinmonopolet's Pricing Policy? A Structural Estimation of the Norwegian Wine Auctions” with Ola Kristoffer Nestvold Master's thesis, NHH, 2019 PDF

Affiliations

• Associated researcher, The FOOD Research Project

Software

• Anonymizer A skill for Codex, Claude Code and OpenCode that anonymizes interviews and open-ended survey answers before analysis. The model proposes auditable changes under project rules, and a human approves the result. Free and open source (Apache-2.0). Source on GitHub Worked example
• Systematic Document Analysis A document analysis plugin for Codex, Claude Code and OpenCode. Runs one independent reader per file and exports results with an audit trail. Free and open source (MIT). Download for Windows Source on GitHub
• PDF Scholar A reading-first PDF reader and annotator built for working with research articles, reports, and books. Free and open source (MIT). Download Source on GitHub

Contact

Work: esh@osloeconomics.no
Personal: emilmshalseth@gmail.com
Oslo Economics, Klingenberggata 7A, 0161 Oslo, Norway