South Korea Goods Imports (BOP Basis)

KREXPORTS and KRIMPORTS are raw monthly flow series recorded as reported without any transformation, the United States dollar value of goods that residents sell to and buy from the rest of the world measured on the balance-of-payments basis. Exports are the goods delivered to nonresidents and imports the goods acquired from them, both valued free-on-board, and because each is a strictly positive gross magnitude the two are read on their logarithms, with the amount by which exports exceed imports forming the goods-trade balance, the largest single component of the current account (Obstfeld and Rogoff 1995). As an expenditure flow the same magnitudes enter the national accounts through the external component of aggregate demand (Meade and Stone 1941).

Each flow is one side of the double-entry external accounts, where every cross-border delivery of goods is matched by an offsetting financial or income entry, in the lineage of moneyflows accounting (Copeland 1949). The goods entries sit in the rest-of-world account of the integrated social-accounting system, from which the trade totals are read (Stone 1966).

Because the flows accumulate over time, their running sum drives the net foreign asset position that measures a country's external wealth (Lane and Milesi-Ferretti 2001), the cumulated positions belong to the family of national balance-sheet measures (Goldsmith, Lipsey, and Mendelson 1963), and the gross flows carry information about external adjustment beyond what the net balance alone reveals (Obstfeld 2012).

What residents export and import responds to relative national price levels, so the flows are read against the purchasing-power benchmark that anchors the nominal exchange rate (Cassel 1918), and their commodity composition reflects demand for products distinguished by their country of origin (Armington 1969).

The trade flows are the quantity through which external imbalances are resolved by the trade channel of international adjustment (Gourinchas and Rey 2007b), and they belong to the family of national-income aggregates whose boundary and netting conventions fix what is counted (Kuznets 1941). The series are not seasonally adjusted, so their within-year pattern is recorded here intact.

KRED applies no transformation to KREXPORTS or KRIMPORTS, neither deflating, re-scaling, smoothing, seasonally adjusting, nor annualizing them, so each published number is the compiled flow exactly as measured, in millions of United States dollars and not seasonally adjusted. Any within-year seasonal shape therefore remains in the data as recorded rather than being removed.

The measurement basis is the double-entry external-accounts framework, first founded as moneyflows accounting (Copeland 1949) and later implemented as a full empirical system of sectoral payments and balances (Copeland 1952), under which each delivery of goods across the frontier is entered simultaneously against its financing or income counterpart. The definitional totals for such flows follow the social-accounting treatment of national income and its related aggregates (Stone 1947), whose complete double-entry construction of income, expenditure, saving, and investment fixes where exports and imports sit (Meade and Stone 1941), and the same quantities are read from the external account of the integrated social-accounting system (Stone 1966).

Because compiled flows are estimates rather than exact counts, they are understood through the framework for quantifying the precision of national-income figures (Stone, Champernowne, and Meade 1942), and their boundary and valuation conventions follow the national-income measurement framework (Kuznets 1941). The practitioner conventions for sectoring the units and classifying the instruments that record these external transactions follow the handbook treatment (Dawson 1996), and the sectoral balance-sheet tabulation into which the cumulated flows feed follows the national balance-sheet construction (Goldsmith, Lipsey, and Mendelson 1963).

The running sum of the flows is carried into the external asset and liability position by the estimation methods for foreign assets and liabilities (Lane and Milesi-Ferretti 2007), and in every case the series exists as an accounting-consistent flow read directly from the accounts rather than a modelled quantity, consistent with the intertemporal accounting identity that links the current account to the change in net foreign assets (Obstfeld and Rogoff 1995).

The goods-trade flows are among the most closely watched external indicators, because they trace the strength of foreign demand and the domestic absorption of imports against which growth and the external balance evolve. The balance between them is the goods portion of the current account, which the intertemporal approach reads as the economy's net lending to or borrowing from the rest of the world (Obstfeld and Rogoff 1995), the running sum of which measures the economy's external wealth (Lane and Milesi-Ferretti 2001), and the gross flows carry information about that position beyond the net figure alone (Obstfeld 2012).

External imbalances are worked off partly through the trade channel, so movements in exports and imports are how adjustment in the external accounts actually occurs (Gourinchas and Rey 2007b). The composition and level of the flows respond to relative national price levels, the purchasing-power benchmark against which competitiveness is judged (Cassel 1918), and to the persistent deviations from that benchmark documented in the real-exchange-rate record (Rogoff 1996).

Demand for traded goods distinguished by their origin gives the flows their elasticity to relative prices and activity abroad (Armington 1969), and the weighting of trade across partner markets rests on a general-equilibrium view of traded goods (Black 1976). The elevated and shifting external positions that these flows cumulate into condition the path of the subsequent cycle (Lane and Milesi-Ferretti 2018).

As an expenditure aggregate the flows sit inside the national-income framework, whose boundary conventions determine what activity is counted as exported or imported (Kuznets 1941), and whose complete accounting template locates net exports within aggregate demand (Meade and Stone 1941).

For market participants the goods-trade flows are a read on the external balance that ultimately funds and prices the currency, since the trade channel is one of the two routes through which external positions adjust (Gourinchas and Rey 2007b), and the current-account identity ties that balance to the economy's net external borrowing (Obstfeld and Rogoff 1995). A widening or narrowing goods balance feeds the net foreign asset dynamics that desks track when sizing external funding and currency risk (Gourinchas and Rey 2007a), and the cumulated positions must be read with attention to valuation and return measurement (Curcuru, Dvorak, and Warnock 2008).

Gross flows carry pricing-relevant information beyond the net balance, so the two sides are monitored separately rather than only in difference (Obstfeld 2012). The estimation of the external assets and liabilities into which the flows accumulate frames the sovereign and cross-border exposures that lenders and reserve managers weigh (Lane and Milesi-Ferretti 2007), and the post-crisis behaviour of those positions bears on how flows translate into balance-sheet risk (Lane and Milesi-Ferretti 2018).

The competitiveness that drives the flows is judged against relative price levels, the purchasing-power anchor for the exchange rate (Cassel 1918), with the well-documented slow reversion of real-rate deviations shaping the horizon over which trade responds (Rogoff 1996). Because trade shares set the weights in effective-exchange-rate construction, the flows also inform the trade-weighted currency measures used for positioning (Pauls 1987; Loretan 2005).

KRIMPORTS is a strictly positive gross flow, so its published log-level is the statistical object, expected to be integrated of order one, and the unit-root triplet is fitted on the log-level with a constant and trend over 556 monthly observations spanning 1980-01 to 2026-04.

The augmented Dickey-Fuller regression of Dickey and Fuller (1979), in the lag-augmented form of Said and Dickey (1984), fails to reject the unit root at p = 0.530, the nonparametric test of Phillips and Perron (1988) concurs at p = 0.092, and the stationarity-null test rejects trend stationarity at p < 0.01 (Kwiatkowski et al. 1992), so all three procedures agree on a clean I(1) classification, robust to the size-and-power corrections that guard the Phillips-Perron statistic against a near-unit moving-average root (Ng and Perron 2001). The confirmation-only GLS-detrended escalation of Elliott, Rothenberg, and Stock (1996) is reserved for ambiguous outcomes and is not invoked on this clean reading.

Because the log-level is I(1), the portmanteau and the break search are run on its first difference, since a level portmanteau reads the near-unit autocorrelation of a stochastic trend and a level mean-break search spuriously segments it (Perron 1989; Hamilton 2018; Bai and Perron 1998). The portmanteau statistic of Ljung and Box (1978), refining the form of Box and Pierce (1970), rejects the white-noise null on the differenced series, with Q = 211.20 at lag 12 and Q = 277.85 at lag 24, both at p = 0.000, and the multiple-break procedure computed by the dynamic-programming algorithm of Bai and Perron (2003) finds no break in the mean of the differenced series.

Because the series is not seasonally adjusted, the seasonal battery is expected to find genuine seasonality, and it does. The seasonal-unit-root test of Hylleberg et al. (1990), with the monthly mechanics of Beaulieu and Miron (1993) and the finite-sample refinements catalogued for monthly data (Franses 1991), rejects unit roots at the seasonal frequencies at p = 0.000, so the seasonality is deterministic rather than stochastic, and the seasonal dummies are jointly significant with F = 13.03 at p = 0.000, confirming a fixed within-year pattern the series never had adjusted away. The seasonal portmanteau likewise rejects at p = 0.000 on the differenced series, and this stationary-but-present seasonality reading is consistent with the seasonal-stationarity and seasonal-adjustment literature (Ghysels and Osborn 2001).