The Fed says Chinese FDI firms took a larger share of Vietnam's U.S. exports, but its published dates do not reconcile

By Cumulant Research | July 18, 2026

According to the Federal Reserve's published figure, yes. After the researchers excluded exports caught by one screen for possible reroutingreroutingIn the working paper, rerouting means a matching product flow from China through Vietnam to the United States within a specified place and quarter, used as an indicator of possible relabeling rather than direct observation of it. [Working-paper methodology](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf), the share attributed to Chinese FDI firms rose from 11.2% in 2018-19 to 25.0% in 2020-23. The difference is 13.8 percentage points. [Federal Reserve accessible chart](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-accessible-20260717.htm)

A share is a slice of a changing pie. A smaller percentage can coexist with a larger dollar amount if the total pie grows quickly enough. That is why the fall in the domestic Vietnamese share from 32.6% to 22.7% does not, by itself, show that domestic firms exported fewer dollars. The Fed's accessible figure publishes percentages, not ownership-group export values. [Federal Reserve accessible chart](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-accessible-20260717.htm)

The scale is economically consequential. U.S. goods imports from Vietnam totaled $193.8 billion in 2025, and the U.S. goods deficit with Vietnam was $178.2 billion. Those bilateral totals record trade by country; they do not identify the ownership of the firms handling the goods. [USTR Vietnam trade summary](https://ustr.gov/countries-regions/southeast-asia-pacific/vietnam)

The Fed links transaction-level shipment records to the Vietnam Enterprise SurveyVietnam Enterprise SurveyThe Vietnam Enterprise Survey is an annual census of formally registered firms conducted by Vietnam's General Statistics Office. [Federal Reserve appendix](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html), an annual census of formally registered firms. The survey reports foreign investment and the countries supplying it. For classification, the Fed assigns each foreign-invested firmforeign-invested firmA foreign-invested firm is a business in Vietnam that reports investment from at least one foreign source. [Federal Reserve data description](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) to the country that supplied its largest amount of investment. [Federal Reserve appendix and ownership rule](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html)

That makes Chinese FDI firmChinese FDI firmIn the Fed analysis, this is a foreign-invested Vietnamese firm whose largest reported source of foreign investment is China, rather than necessarily a firm whose ultimate owner is known to be Chinese. [Federal Reserve ownership rule](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) more precise than simply calling every exporter located in Vietnam Vietnamese-owned. It is still not the same as tracing the ultimate beneficial owner through every corporate layer. The published rule identifies the largest reported source country of investment, not a complete ownership chain. [Federal Reserve ownership rule](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html)

An exported product can have a Vietnamese customs origin, leave from a firm whose largest foreign investor is in China, contain components from several countries and still include real production in Vietnam. These are different measurements: customs origin describes how the shipment is recorded, firm classification describes the exporter, and value addedvalue addedValue added is the value created through work and production after subtracting the cost of purchased inputs. [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf) estimates where economic value was created. The Fed focuses on firm classification, while the separate rerouting paper also estimates imported contentimported contentImported content is the portion of an exported product's value estimated to come from inputs purchased abroad. [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf) and domestic value addeddomestic value addedDomestic value added is the portion of an export's value estimated to have been created inside the exporting country. [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf). [Federal Reserve data description](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The import side of the Fed figure supplies another descriptive clue. Chinese FDI firms' share of all Vietnamese imports from China rose from 11.4% in 2018-19 to 22.1% in 2020-23. The statistic says who handled the imports. It does not say that 22.1% of those firms' own inputs came from China. [Federal Reserve accessible chart](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-accessible-20260717.htm)

The Fed treats the parallel rise in Chinese FDI firms' export and import shares as evidence consistent with production and supply-chain relocation. It also says its data cannot directly identify whether foreign firms shifted production from China to Vietnam, so the relocation conclusion is an interpretation of several patterns rather than a directly observed factory-by-factory movement. [Federal Reserve analysis](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html)

Before calculating the export shares, the Fed removed exports from firms that imported a product from China and exported the same HS eight-digit productHS eight-digit productAn HS eight-digit product is a detailed customs classification used to distinguish closely related traded goods. [Working-paper data description](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf) to the United States within the same quarter. This same-firm rule is the most restrictive version considered in the separate working paperworking paperA working paper is a research draft circulated before or outside final journal publication and may change as the authors revise it. [Harvard publication record](https://www.hks.harvard.edu/centers/cid/publications/exports-disguise-trade-re-routing-during-us-china-trade-war) on Vietnam's trade rerouting. [Federal Reserve methodology](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) [Working-paper methodology](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The advantage is that it is less likely to confuse unrelated imports and exports. The cost is that it can miss coordinated activity in which one firm imports, another processes or relabels, and another exports. The working paper says subsidiaries and nearby firms can coordinate and therefore treats its broader province-level matchprovince-level matchA province-level match compares imports and exports of the same detailed product within one Vietnamese province and quarter, even when different firms handle the shipments. [Working-paper methodology](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf) as the preferred compromise. [Working-paper sensitivity discussion](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

This matters for the ownership comparison because the omissions may not be evenly distributed. In the working paper, Chinese and Hong Kong-owned entrants represented 66.8% of the rerouting increase attributed to new entrants. That result does not invalidate the Fed's 25.0% share, but it raises the possibility that rerouting missed by a firm-level screen is concentrated in the same ownership group being measured. [Working-paper entrant analysis](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The difference between 8.8% and 21.1% is not a trivial technical adjustment. It shows that an estimate can change substantially depending on whether matching occurs within a province or across the entire country. The authors warn that broad matching can join unrelated flows, while narrow firm matching can miss cooperation across firms. [Working-paper accounting exercise](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The working paper asks a related but different question: how much of Vietnam's $52.8 billion increase in U.S.-bound exports from 2018 to 2021 can be assigned to rerouting, imported inputs and value created inside Vietnam? Using its preferred province-level matching method, it assigns 8.8% of the increase to rerouting. [Working-paper accounting exercise](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The same decomposition assigns 20.4% of the increase to Chinese imported content excluding rerouting, 31.0% to imported content from the rest of the world and 39.8% to domestic value added. The categories total 100.0%. [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The value-added estimate is also assumption-dependent. The authors say that interpreting their calculation as domestic value added requires the strong assumption that Vietnam conducted little back-and-forth trade with the countries supplying its inputs. The chart therefore reports their estimate, not a directly observed accounting total. [Working-paper decomposition and assumption](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The working paper is useful as a sensitivity check, not as a final audit of the Fed figure. Harvard lists an earlier version as Working Paper 24-072, while the currently linked author-hosted draft contains the estimates used here. [Harvard publication record](https://www.hks.harvard.edu/centers/cid/publications/exports-disguise-trade-re-routing-during-us-china-trade-war) [Current working-paper draft](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

The largest documentary problem is inside the Fed note itself. Figure 3 compares 2018-19 with 2020-23, but the appendix says the PanjivaPanjivaPanjiva is an S&P Global database containing transaction-level international shipment records. [Federal Reserve appendix](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) database supplies Vietnamese import and export transactions only between 2018 and 2021. The note does not explain how a shipment-linked trade figure extends through 2023 when the described shipment records stop in 2021. [Federal Reserve figure and appendix](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html)

There are benign possibilities: the appendix could contain the wrong endpoint, the chart label could be wrong, or later enterprise records could have been combined with a shorter shipment window. None is documented on the published page, so choosing among them would be speculation.

The Fed also says the Panjiva records begin in 2018, leaving no earlier observations with which to test whether the ownership shares were already moving before the tarifftariffA tariff is a tax charged when a product is imported. [Working-paper background](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf) increases. Without that pre-trendpre-trendA pre-trend is the direction a measure was already moving before the event being studied, which helps researchers judge whether the event plausibly changed the pattern. [Federal Reserve limitation](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html), the chart can establish a before-and-after difference in the available periods but cannot, by itself, assign the difference to the tariffs. [Federal Reserve limitation](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html)

The defensible conclusion is narrow: in the Fed's published, rerouting-adjusted figure, firms classified as Chinese FDI firms handled a substantially larger share of Vietnam's U.S.-bound exports in the later period. That is a meaningful change in who appears to sit behind the customs totals. [Federal Reserve accessible chart](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-accessible-20260717.htm)

The figure does not determine how much of Vietnam's export boom those firms generated, how much Chinese production relocated, how much income remained in Vietnam, or how much dependence on Chinese inputs persisted after production steps moved. Answering those questions requires dollar values, consistent time coverage, fuller ownership tracing and a direct measure of value added. [Federal Reserve limitations](https://www.federalreserve.gov/econres/notes/feds-notes/vietnams-export-boom-to-the-u-s-the-role-of-Chinese-firms-20260717.html) [Working-paper decomposition](https://sungjuwu.github.io/documents/Vietnam_rerouting_paper.pdf)

Nor does this evidence describe a market reaction. The cited analyses study trade flows, firm classifications and production-related measures, not changes in stock prices, exchange rates or investor expectations around the Fed note's publication. A market move and an economic effect are different claims, and neither should be inferred from the ownership-share chart.