Korea’s July Import Prices Fell Again: Separating the Currency Effect From the Cost-of-Living Signal

원문 링크: WordPress 원문

공개 시장 해설용 글입니다. 투자 조언이 아닙니다.

언어 선택 / Choose a language: KO · 한국어 | EN · English

How the exchange rate creates the gap between the won basis and the contract-currency basis.

The conclusion: a large share of July’s decline is a currency story

Korea’s import price index measured in won fell 1.0 percent from June to July 2026. That is accurate, but reading it alone invites a conclusion the data does not support. In the same release, the import price index measured in contract currencies rose 0.9 percent over the same month. The gap between a fall in won terms and a rise in contract-currency terms is largely an exchange-rate effect: the monthly average won-dollar rate moved from 1,527.30 in June to 1,497.43 in July, a 2.0 percent decline. In other words, July did not primarily show foreign suppliers charging less. It showed the same foreign-currency invoices converting into fewer won. One more figure matters just as much. Compared with a year earlier, import prices in won terms are still 18.7 percent higher. Direction over one month and level over one year answer different questions, and household budgets respond mostly to level. The useful reading is not that cost pressure has ended, but that one specific channel eased for one month. Our earlier explainer on why prices do not fall immediately when oil gets cheaper sets out the same discipline for a different trigger.

The test is straightforward. If the decline had been driven mainly by cheaper foreign supply, the contract-currency index should have fallen as well. It rose. If instead an exchange-rate move dominated, the two bases should diverge in sign, which is precisely what July shows. The accurate short form of the headline is therefore narrower than the one usually printed: import costs converted into won declined for a month. That narrowing is not excessive caution; it follows the definition the statistics agency itself publishes, and the release presents both bases side by side specifically because neither substitutes for the other.

What the official release actually reported

The Bank of Korea published its July 2026 export and import price indices, together with trade indices, on 14 August 2026. The import price index in won terms fell from 161.71 in June to 160.09 in July on a 2020=100 base, which is a 1.0 percent monthly decline, a 12.2 percent rise from the end of the previous year, and an 18.7 percent rise from a year earlier. The export price index in won terms moved the other way, from 188.68 to 190.65, a 1.0 percent monthly gain and a 49.1 percent annual gain. Two reference figures published alongside the indices shape the interpretation. The monthly average won-dollar rate fell 2.0 percent on the month but remains 8.9 percent higher than a year earlier. Dubai crude averaged 76.75 dollars per barrel in July against 79.45 in June, a 3.4 percent monthly decline and an 8.3 percent annual increase. Both reference series share the same shape: lower than last month, higher than last year. Any summary that omits the comparison base can therefore produce the opposite impression from the same release, which is why the base belongs in the first sentence rather than a footnote.

Three axes are worth writing down before quoting any figure from this release. The first is the currency basis, won or contract. The second is the comparison period, month on month, year to date, or year on year. The third is the level of aggregation, headline index or product group. For July imports, the won basis reads minus 1.0 percent on the month, plus 12.2 percent year to date and plus 18.7 percent year on year, while the contract-currency basis reads plus 10.4 percent both year to date and year on year. All of those numbers are correct and none contradicts another. Most misreadings begin with an attempt to compress six valid figures into one sentence.

Won terms and contract-currency terms diverged this month

Korea publishes these indices on two bases. The contract-currency basis tracks prices in the currency in which the trade was actually invoiced. The won basis applies the exchange rate to those prices. In July 2026 the two bases pointed in opposite directions for imports: minus 1.0 percent in won terms against plus 0.9 percent in contract-currency terms. Exports show the same wedge in the other direction, rising 1.0 percent in won terms and 3.0 percent in contract-currency terms. The annual comparison shows the wedge even more clearly. Import prices are 18.7 percent higher year on year in won terms but 10.4 percent higher in contract-currency terms, while export prices are 49.1 percent higher in won terms and 37.8 percent higher in contract-currency terms. The difference is, broadly, what the exchange rate contributed. This distinction has practical value: relief created by a currency move can be reversed by a currency move, whereas a decline in the invoiced price itself is a somewhat more durable signal. Before drawing any conclusion from a headline, identify which basis the number uses.

The export side provides a useful mirror. July export prices rose 1.0 percent on the month in won terms but 3.0 percent in contract-currency terms. When the won appreciates, foreign-currency export receipts convert into fewer won, so the contract-currency gain can exceed the won-terms gain. The same currency move therefore widens the won-terms decline on the import side while trimming the won-terms gain on the export side. Recognising that one variable pushes two indices in arithmetically opposite directions makes it immediately clear why quoting only one of them leaves an explanation incomplete.

The transmission mechanism, stage by stage

An imported item’s final shelf price is assembled in stages. It begins with a supplier’s price in the invoicing currency, is converted at an exchange rate, and then accumulates freight, insurance, contract terms, customs duties and taxes, domestic processing and distribution costs, and finally the competitive conditions of the retail market. Each stage introduces its own delay. Crude oil reacts quickly to international benchmarks, but refining and distribution slow the pass-through. Goods bought under long-term contracts or drawn from inventory may not reflect this month’s price for several months. July’s detail illustrates the unevenness: raw materials rose 0.8 percent on the month while intermediate goods fell 2.2 percent, capital goods fell 1.8 percent and consumer goods fell 0.1 percent. Within energy itself, crude oil fell while liquefied natural gas rose, lifting the mining products group by 0.9 percent. A single monthly average therefore hides several different stories. Our discussion of how oil, rates and the currency create combined pressure maps these lags in more detail.

The stages an import-stage price passes through before reaching a household bill.

Where July’s decline was concentrated, and where it was not

The headline index conceals wide dispersion. Coal and petroleum products fell 3.9 percent on the month, the largest decline among the main manufactured groups, while chemical products fell 1.0 percent, textiles and leather 1.9 percent, and wood and paper products 2.0 percent. Mining products rose 0.9 percent, and within that group coal, crude oil and natural gas rose 1.8 percent. Agricultural, forestry and marine products fell 0.3 percent overall, but the components diverged sharply: agricultural products rose 2.2 percent and marine products rose 3.8 percent, while livestock fell 6.3 percent and forestry products fell 2.0 percent. Food and beverage inputs fell 3.2 percent. The annual comparison adds the missing context. Mining products remain 24.9 percent higher than a year earlier, coal and petroleum products 35.5 percent higher, and metal and non-metal minerals 32.3 percent higher. A monthly decline and a large annual increase are fully compatible, which is why a household can accurately report that costs still feel elevated in a month when an index headline points down.

Weights sharpen the point. Within the import index, mining products carry a weight of 249.6, of which coal, crude oil and natural gas alone account for 203.1. Agricultural, forestry and marine products together carry 43.7, split into 19.3 for agricultural products, 14.4 for livestock and 9.1 for marine products. The headline index is therefore far more sensitive to energy-related movements than a typical grocery basket is. A household whose spending is weighted toward food can accurately experience rising costs in a month when the national index declines. The index is built on a national average composition; a budget is built on an individual one.

The export side belongs on its own ledger

The same release covers exports. Export prices in won terms rose 1.0 percent on the month, led by computer, electronic and optical products together with coal and petroleum products, while agricultural, forestry and marine export prices fell 1.7 percent. On volumes, the export volume index rose 20.0 percent from a year earlier and the import volume index rose 14.7 percent. Terms of trade improved as well: the net barter terms of trade index rose 24.7 percent year on year and the income terms of trade index rose 49.7 percent. These are meaningful indicators of external sector strength, but they are not a statement about household budgets. Export prices respond to industry composition, invoicing currencies and overseas demand. Household spending responds to income, employment, interest costs and the prices of essentials. Compressing both into one sentence produces the familiar shortcut that strong exports must mean easier living costs, when the connection depends on timing and distribution. Our earlier analysis of why export gains and household costs need a separate read examines that gap directly.

How the signal reaches households, and at what speed

It helps to sort spending into three groups. The first contains fuel and transport, which sit closest to international energy prices. Even here, fuel taxes, refining margins and distribution structures prevent a one-to-one relationship. The second contains food and household goods, where imported inputs pass through processing, inventory and supply contracts before reaching a shelf, and where domestic harvests or livestock conditions can dominate the outcome entirely. The third contains housing costs, communications, education and regulated charges, where contracts and administrative decisions matter far more than import prices. July’s figures show a 1.8 percent monthly decline in capital goods and a 0.1 percent decline in consumer goods at the import stage. Those are facts about import invoices, not forecasts of any particular household’s next bill. Writing down the share of your own spending that falls into each of the three groups converts a national statistic into a personal question that can actually be answered.

A five-step check after this release

First, identify whether a quoted figure uses the won basis or the contract-currency basis, because in July they moved in opposite directions. Second, identify whether the comparison is monthly or annual, because import prices fell 1.0 percent on the month while remaining 18.7 percent above a year earlier. Third, move past the headline index to the product groups that match your own spending, since food inputs and livestock fell while agricultural and marine products rose. Fourth, check consumer-stage prices separately through the national statistical portal rather than assuming that a border-stage decline has already reached retail. Fifth, if the question concerns volumes or trade values rather than prices, use customs trade statistics instead of a price index. None of these steps forecasts anything. They exist to prevent one number from being asked to answer five different questions at once, which is the most common way a correct statistic produces an incorrect conclusion.

A one-month change and a one-year level answer different questions.

Using the checklist matters as much as having it. If fuel spending fell this month, note whether the cause was the international benchmark, fewer kilometres driven, or a seasonal pattern, because only the first is connected to the release. If food spending rose, compare it against the July detail, where agricultural products gained 2.2 percent and marine products 3.8 percent while livestock fell 6.3 percent. Matching personal spending against official product groups even once tends to end the habit of judging by headline alone. The purpose of the record is not prediction. It is comparability in the following month.

Uncertainty and the limits of a single month

These figures are provisional and may be revised. Exchange rates and oil benchmarks can reverse, and supply conditions, geopolitical events and domestic demand can all redirect the price path. What July established is narrow and can be stated in three sentences: import-stage pressure eased for one month in won terms, contract-currency prices rose slightly over the same month, and the level remains well above a year earlier. Compressing those three sentences into one always discards something material. Average indices also differ from the basket any individual household actually buys, so experience will vary within the same month. This article does not forecast exchange rates, energy prices or consumer inflation, and it does not recommend any spending or investment decision. The most durable habit is to set the next verification point rather than to conclude: if the next release shows both bases moving in the same direction, the evidence that this month’s relief was primarily a currency effect becomes considerably stronger.

Three recurring confusions in coverage of this release

The first confusion treats import prices and consumer prices as one measure. They sample different stages with different weights, so a decline in one does not guarantee a decline in the other. The second confusion mixes prices with volumes. July’s 20.0 percent annual rise in the export volume index is a statement about quantities, and it is separate information from any price change. The third confusion quotes provisional data as if it were final; these figures may be revised. Filtering for those three alone removes a large share of the overstated conclusions built on an otherwise accurate release. When verification is needed, the underlying series sit in the central bank’s statistics system, volume and value data sit in customs trade statistics, and consumer-stage prices sit with the national statistics office.

Three things worth watching in the next release

This release leaves a list of questions rather than a conclusion. The first is the exchange rate: the monthly average reached 1,497.43 won per dollar in July but remains 8.9 percent above a year earlier, so a reversal would remove the same factor that produced this month’s won-terms decline. The second is the direction of the contract-currency basis. If it rises again while only the won basis falls, the case that this was primarily a currency effect strengthens. If both fall together, there is a genuine argument about softening foreign prices. The third is the consumer-stage response, which arrives with a lag whose size depends on distribution structures and competitive conditions. Government monthly economic assessments can supply macro context, but they do not explain the retail price of any individual product. Fixing these three checkpoints in advance separates what is genuinely new in next month’s coverage from what is already known.

Sources

• Bank of Korea, Export and Import Price Indices and Trade Indices, July 2026 (provisional), released 14 August 2026

• Bank of Korea Economic Statistics System (ECOS) — price index time series

• Korea Customs Service trade statistics

• Statistics Korea national statistics portal — consumer price statistics

• Ministry of Economy and Finance — monthly economic assessment

Disclaimer: This article is for information only and is not investment advice. Figures follow the publishing institution’s official release, and provisional data may be revised. Individual spending, contract and asset decisions should be made separately, using official sources and professional advice.