What is next door to Viet Nam’s export basket?

Countries do not diversify at random. They jump into products that share capabilities with what they already make, tracing a walk on the Hausmann-Hidalgo product space (Hidalgo, Klinger, Barabási & Hausmann 2007). For Viet Nam in 2022, this page isolates the HS6 products the country does not yet export with revealed comparative advantage (RCA ≥ 1) but which sit closest to the basket it already has. These are the candidate entries for export strategy and targeted industrial policy.

The basket today

Revealed comparative advantage is Balassa’s (1965) classic: RCA(c, p) = (X_cp / X_c) / (X_wp / X_w). A country has RCA in a product when the product’s share of its export basket exceeds the product’s share of the world export basket. We flag every HS6 with RCA ≥ 1 in 2022 as in the country’s current competitive basket, and size each by current export value (BACI values are stored in thousands USD and multiplied by 1,000 for display). The top 20 lines carry most of the basket mass.

Adjacency density on the product space

Density ω(c, p) = Σ_p′ φ(p, p′) × M(c, p′) / Σ_p′ φ(p, p′), where M(c, p′) = 1 if RCA(c, p′) ≥ 1 and 0 otherwise (Hausmann & Hidalgo 2011, Atlas of Economic Complexity, chapter 2). Proximity φ is the minimum conditional probability that a country with RCA in one product also has RCA in the other (Hidalgo et al. 2007). Density runs from 0 to 1: high means the product’s capability neighbourhood is already largely occupied by Viet Nam’s basket. The 30 highest-density non-RCA products in 2022 are shown below.

Recommended entry candidates

The density ranking by itself says nothing about market size, complexity, or who the current competitors are. The table below adds the missing columns. Global market size is the sum of BACI world imports for the HS6 in 2022 (multiplied by 1,000 for USD). PCI is the product complexity index (Hidalgo & Hausmann 2009, PNAS 106: 10570), which is zero-centred and routinely negative; higher means more complex and more associated with high-income, high-capability exporters. Top exporters are the three countries with the largest export value on the HS6 in 2022, which is who a new entrant would be contesting the market against.

How does basket complexity move if the top-10 are adopted?

The Economic Complexity Index (ECI) of a country is closely tied to the weighted mean PCI of its RCA basket (Hidalgo & Hausmann 2009; Hausmann & Hidalgo 2011). A counterfactual re-weighting gives a simple order-of-magnitude read on how much basket complexity would shift if Viet Nam displaced its ten lowest-PCI RCA lines with the top-ten density-ranked adjacent products at constant aggregate basket weight.

Where on the product space are the candidates clustered?

The Hausmann-Klinger (2007, CID Working Paper 146) reading of the product space is that capabilities cluster by sector: countries successfully add products that share inputs, skills, and logistics with what they already make. Aggregating the top-30 density-ranked adjacent products by HS Section gives a one-glance read on which parts of the product space the next-door candidates sit in. Section counts below are taken from the HS 21-section classification (WCO, Harmonized System).

The density-complexity quadrant: where are the “best jumps”?

Hausmann & Klinger (2007, CID WP 146) and Hausmann & Hidalgo (2011, J. Econ. Growth 16: 309–342) argue that desirable diversification targets are high on two axes at once: high density (capability-feasible given the existing basket) AND high PCI (complex enough to pull basket complexity up). The scatter below places each non-RCA candidate product in the density × PCI plane. The top-right quadrant — high density, high PCI — is the “best jumps” zone: feasible upgrades. The top-left is feasible but complexity-flat; the bottom-right is aspirational but capability-distant; the bottom-left is neither. The top-10 density-ranked candidates from Figure 3 are marked in orange so the reader can see how the density-only ranking maps onto both axes.

Feasibility floor plus complexity upgrade: the top-10 shortlist

The density-only ranking in Figure 3 is capability-proximity biased: a product very close to the existing basket can sit on that list even when its PCI is below the basket’s own weighted mean, adding no complexity. Conversely, the density-PCI scatter in Figure 6 is visual-only. A cleaner industrial-policy shortlist answers a specific question: among products that clear a minimum density threshold (a feasibility floor), which ten have PCI furthest above the country’s current basket-weighted PCI? That is the list of complexity upgrades with existing capability. Hausmann, Hidalgo et al. (2014, Atlas of Economic Complexity, 2nd ed.) describe exactly this construct as the “strategic bets” quadrant, and note that the minimum density floor is an application choice: 0.20 is a conventional lower bound in the CID working-paper series (Hausmann & Klinger 2007, CID WP 146).

What did Viet Nam actually jump into since 2010?

The adjacency density in Figures 2-7 is a forward-looking feasibility signal. A backward-looking companion is the set of HS6 products in which the country genuinely crossed the RCA ≥ 1 threshold between 2010 and 2022: products that started below revealed comparative advantage and ended above it. Hausmann, Hidalgo et al. (2014, Atlas of Economic Complexity, 2nd ed.) call these realized product-space jumps, and show that they typically occur to HS6 lines that sat at high adjacency density in the base year — i.e. the density-predicted pattern repeats itself ex post. The bars below rank the largest realized RCA gains (ΔRCA = RCA₂₀₂₂ - RCA₂₀₁₀), filtered to products that were below RCA = 1 in 2010 and at or above RCA = 1 in 2022. These are the products where the country actually added revealed comparative advantage over the window.

Which sections are the most feasible to enter on average?

Figure 5 ranked sections by their share of the top-30 density-ranked candidates, which is a top-of-list view. The complementary read is the average adjacency density across every non-RCA candidate in a given section: this answers the section-level feasibility question, not the rank-cutoff question. A section with a high mean density is one where the country’s existing capabilities sit close to most products in that part of the WCO classification, even if no single product makes the top-30. Hausmann and Klinger (2007, CID WP 146) note that diversifying into a section with a uniformly high mean density is structurally easier than diversifying into a section where one or two outliers sit near the basket but the bulk of the section is far away.

How to read and not mis-read this

• Density is a capability-proximity statistic, not a guarantee. Hausmann and Klinger (2007, CID WP 146) and Hausmann & Hidalgo (2011) show that density predicts diversification probabilities in the aggregate; for any individual country-product pair, institutions, tariffs, logistics, and sunk costs can still block entry.
• PCI is zero-centred by construction. A negative PCI on a recommendation row is a reading, not a filter: the product is simply less complex than the HS6 mean. Many feasible industrial upgrades for lower-income countries sit at slightly-below-zero PCI, not far above.
• Global market size reads BACI world imports at HS6, which undercounts where bilateral trade is reported only by one side of the pair. For order-of-magnitude sizing it is correct; for litigation-grade numbers, cross-check against the importer’s own customs filing.

References

• Balassa, B. (1965). “Trade Liberalisation and Revealed Comparative Advantage.” Manchester School 33(2): 99–123.
• Hausmann, R., & Hidalgo, C. A. (2011). “The network structure of economic output.” Journal of Economic Growth 16: 309–342.
• Hausmann, R., Hidalgo, C. A., Bustos, S., Coscia, M., Simoes, A., & Yildirim, M. A. (2014). The Atlas of Economic Complexity: Mapping Paths to Prosperity. CID/MIT Press. First edition 2011.
• Hausmann, R., & Klinger, B. (2007). “The Structure of the Product Space and the Evolution of Comparative Advantage.” CID Working Paper No. 146.
• Hidalgo, C. A., & Hausmann, R. (2009). “The building blocks of economic complexity.” Proceedings of the National Academy of Sciences 106(26): 10570–10575.
• Hidalgo, C. A., Klinger, B., Barabási, A.-L., & Hausmann, R. (2007). “The product space conditions the development of nations.” Science 317: 482–487.