Quantify the addressable market, competitive thickness, and entry barriers for a product or country pair

A market-entry assessment translates a single operational question, can we sell product Y into market X at acceptable risk, into a structured empirical answer. The work quantifies the size of the addressable trade flow, how crowded the destination is, what tariff treatment the shipment will actually face, how distance and adjacency pull the flow toward or away from substitute origins, and how substitutable the product is in the destination basket. Each piece is built from the same open, reproducible parquet bank that powers the rest of the TradeWeave workbench.

Problem

The typical brief arrives in one sentence. A client is considering entering market X with product Y and wants an objective read on whether the move is worth the commercial effort before any counterparty conversation begins. Behind the sentence sit a standard set of questions.

How large is the market? World imports of the HS6 line into the destination define the outer envelope, but the envelope shrinks fast once we condition on supplier country, contract terms, tariff schedule, and logistics corridor. How many other origins are already serving that destination, and how concentrated is their share? A market that is 85 percent served by two incumbents looks very different from a market where the leading exporter holds 20 percent and the long tail accounts for the rest. What tariff does the client actually pay at the border, including MFN, preferential treatment under any applicable agreement, and quantitative quotas where they bind? How does the client country compare on distance and adjacency with incumbents already shipping the product to that destination? How substitutable is the product inside the destination basket, which determines how quickly a new entrant can steal share versus how long incumbents can defend margin?

These are not guess-with-context questions. They are questions with primary data behind each one. The assessment assembles the data, runs the standard quantitative exercises, and writes the answer up in a form that a commercial team can act on.

Data

The foundation is CEPII BACI, the bilateral trade panel at HS6 that harmonizes the reporting asymmetries between exporter-declared and importer-declared flows. BACI gives the bilateral flow value and quantity for every origin-destination-HS6-year cell since 1995, which is what market size is measured against. Values are stored in thousands of USD and multiplied by one thousand for any display. The panel covers 238 reporters and 5,022 HS6 product codes.

Gravity-side covariates come from CEPII Gravity, version 202411. The file provides bilateral distance, contiguity, common language, common colonial history, regional trade agreement status, and the full set of country-year economic covariates (GDP, population) that every gravity specification uses. Tariffs come from the World Integrated Trade Solution pull of MFN applied rates and, where the client requires it, effectively applied rates including preferential regimes. Unit values are the CEPII unit-values release, which provides price per kilogram for every exporter-importer-HS6 cell and is the standard input for quality and price-positioning analysis. Product-level import demand elasticities use the Kee-Nicita-Olarreaga (2008, Review of Economics and Statistics) panel, which remains the most widely cited HS6 elasticity vector and is required for any meaningful substitutability read.

Method sketch

The empirical core is a gravity-predicted-versus-actual comparison. The exercise estimates a standard PPML gravity model on the full BACI-Gravity panel at HS6 for the product of interest, then uses the fitted model to produce a predicted flow for the client origin to the candidate destination. The ratio of actual to predicted flow separates destinations where the client country already trades close to its gravity-implied potential from destinations where a large gap exists. Counterfactual flows under tariff removal or FTA scenarios come from the same fitted coefficients by resetting the policy variables and re-simulating.

Destination concentration is read from a Lorenz curve and Gini coefficient over the exporter share distribution at the destination-HS6 cell, plus a Herfindahl-Hirschman index for the ten largest origins. The three numbers are complementary: HHI captures top-end concentration, Gini captures the shape of the full distribution, and the Lorenz curve makes the tail visible. For a product where HHI is high and Gini is high, the destination is dominated by a few incumbents and entry will require displacing one of them. For a product where HHI is low and Gini is moderate, the market is genuinely contestable.

Adjacency density follows Hausmann and Klinger (2006, CID Working Paper 128) and Hausmann-Hidalgo (2007). The density metric measures, for a candidate product, how close it sits to the existing export basket of the client country in the product space. A candidate product with high density to the current basket is likelier to be exported successfully because the underlying capabilities already exist. Low-density jumps are feasible but empirically less frequent and take longer. The workbench already hosts the product-space graph and the adjacency-density calculation at the origin level, which we reuse here.

Substitutability on the demand side is read through the Kee-Nicita-Olarreaga elasticity at HS6. The elasticity tells us how responsive destination import quantity is to relative price, which in turn tells us how much room there is to take share from incumbents by undercutting on price and how vulnerable a future entrant will be to the same playbook. Read together with unit-value positioning from the CEPII unit-values file, the client ends the engagement with a defensible view on whether the candidate play is a price war, a quality play, or a niche-occupancy move.

Deliverable

The standard deliverable is a fifteen-page written assessment plus a destination-rank Parquet file. The written document opens with an executive answer: go, do-not-go, or conditional-go with the specific conditions that would flip the recommendation. It continues through a market-size section with the BACI-grounded envelope, a competitive-structure section with the HHI, Gini and Lorenz read, a tariff-and-NTM section with MFN and preferential treatment plus any quota or sanitary constraint, a gravity-residual section with the predicted-versus-actual read, an adjacency-density section with the product-space position, a substitutability section with the elasticity read and unit-value positioning, and a risk section covering political, sanitary, and macro risks relevant to the corridor.

The Parquet file contains the full destination-rank table at origin-destination-HS6 for the product of interest, with the gravity residual, HHI, distance, tariff, and adjacency-density columns joined. It is designed to be loaded straight into the client's own analytics stack and re-sorted for alternative prioritization criteria.

Related workbench pages

Each assessment is accompanied by live links to the open workbench pages that underlie the analysis, so the client can verify the numbers against the source at any time and re-run the exercise for a different destination or product later without a second engagement.

• Market-entry scoring across country-product pairs
• Gravity model: distance, borders, and predicted trade
• Adjacent-product space and density
• Bilateral corridor profiles
• Per-product HS6 dashboards
• Tariff and NTM treatment

Timeline

The standard engagement runs two weeks per market-product pair. The first week is data assembly, gravity fit, and substantive checks against primary sources. The second week is writing, peer review, and final sanity-checking. Multi-market or multi-product briefs scale linearly and are scoped on a per-pair basis.