South Korea Corporate Lending Rate (Balance)

KRCORPLOAN is the raw monthly series recorded without transformation, namely the weighted-average effective interest rate carried on the outstanding stock of loans extended to the corporate sector, quoted in percent per annum.

An interest rate is the intertemporal price that equates present and future claims on income, and the rate of this series applies not to a single contract but to the entire surviving book of corporate debt (Fisher 1930).

The balance-basis construction measures the rate on a stock of credit rather than on newly originated lending. The underlying object is the corporate-loan liability of the producing sector, which appears in the flow-of-funds accounts and the integrated social-accounting framework (Copeland 1949; Stone 1966). That liability is a sectoral credit stock first compiled empirically, and the sectoral national balance sheets quantify it (Copeland 1952; Goldsmith, Lipsey, and Mendelson 1963). Its compilation follows a codified instrument-and-sector classification (Dawson 1996).

The corporate-loan aggregate this rate prices is the bank-credit-to-firms concept, whose long-run measurement is established and harmonized across economies (Schularick and Taylor 2012; Jordà, Schularick, and Taylor 2017). Because the rate averages over heterogeneous balances, it is in effect a quantity-weighted index of contract rates, and its weighting follows a user-cost logic and superlative-index reasoning (Barnett 1980; Diewert 1976).

The producing sector that holds this debt is one node of the saving-investment identity, so the series also describes the cost of the credit that funds domestic capital formation (Obstfeld and Rogoff 1995).

KRED applies no transformation to KRCORPLOAN, neither rebasing, deflation, seasonal adjustment, nor annualization beyond the per-annum convention in which the rate is already quoted, so the value is the measurement as recorded. What the methodology describes is therefore only the basis by which such a number comes to exist, namely the construction of a weighted-average effective interest rate on a stock of outstanding loans.

The effective rate is an intertemporal price (Fisher 1930). Aggregating many individual contract rates into a single figure is an index-number problem in which the appropriate weights are loan balances. The formal justification for a balance-weighted average rests on user-cost weighting and the exact and superlative index theory (Barnett 1980; Diewert 1976).

The stock being weighted is the corporate-loan liability measured within the flow-of-funds accounts, which were originated and then given their first full empirical compilation (Copeland 1949; Copeland 1952). The same liability is also measured within the integrated income-capital-financial accounting and the sectoral balance sheets (Stone 1966; Goldsmith, Lipsey, and Mendelson 1963).

The sectoring and instrument classification that decide which balances count as corporate loans follow a codified compilation practice (Dawson 1996). The resulting credit aggregate is the bank-loan-to-firms series whose construction is documented (Schularick and Taylor 2012; Jordà, Schularick, and Taylor 2017).

The balance basis means the figure reprices slowly as old contracts mature and new ones enter. This is the same stock-versus-flow distinction that separates a position from a transaction in the external accounting (Lane and Milesi-Ferretti 2001).

KRCORPLOAN is the realized cost of credit on the corporate sector's existing debt. It measures the financing burden embedded in firms' balance sheets rather than the marginal rate on new loans, and its movements show how the intertemporal price is actually paid on the surviving stock (Fisher 1930).

The corporate-loan stock this rate prices is the bank-credit aggregate, whose expansion connects to financial-cycle dynamics and whose long-run behavior bears on growth and crisis risk (Schularick and Taylor 2012; Jordà, Schularick, and Taylor 2017). The balance-basis rate is therefore the price counterpart of that credit quantity.

Read through the flow-of-funds framework, the series is the cost the producing sector pays on the liabilities recorded in the sectoral accounts and the national balance sheets (Copeland 1949; Copeland 1952; Stone 1966; Goldsmith, Lipsey, and Mendelson 1963). The instrument classification fixes which credit is corporate (Dawson 1996).

Because corporate borrowing funds domestic investment, the rate is one determinant of the saving-investment gap. In the intertemporal identity that gap maps into the current account and the net-foreign-asset position (Obstfeld and Rogoff 1995). The information content of that position has been emphasized (Obstfeld 2012), and its measurement has been standardized (Lane and Milesi-Ferretti 2007).

A balance-weighted cost of funds in the user-cost sense thus summarizes how dearly the corporate sector services the debt that links national saving to the external position (Barnett 1980).

For a lender the balance-basis corporate rate is the realized yield on the outstanding stock of corporate-loan assets. It is the asset-side counterpart to the cost-of-funds figure that drives net interest margin, so it is read as the return on a stock of financial claims rather than the coupon on a single new contract.

Valuing that return is the same measurement problem as estimating the yield on a portfolio of claims. The balance weighting follows user-cost aggregation and superlative-index logic (Barnett 1980; Diewert 1976). The gap between the rate on the stock and the rate on new flows is the repricing lag that the intertemporal price makes explicit (Fisher 1930).

The loan stock that earns this yield is the documented bank-credit aggregate (Schularick and Taylor 2012; Jordà, Schularick, and Taylor 2017), recorded as a sectoral liability in the financial accounts (Copeland 1952; Stone 1966; Dawson 1996).

Treating a measured rate as the return on a position rather than on a flow is exactly the distinction drawn in external-balance-sheet work. That literature decomposes the yield on gross external assets (Gourinchas and Rey 2007a), traces its valuation channel (Gourinchas and Rey 2007b), and exposes the bias in estimated return differentials on cross-border claims (Curcuru, Dvorak, and Warnock 2008). All of it warns that a balance-weighted yield blends vintages and must be read as an average over the whole book (Lane and Milesi-Ferretti 2018).

On the 296 monthly observations spanning 2001-09-01 to 2026-04-01, fit with a constant and trend, the unit-root battery agrees that the corporate lending rate is integrated of order one. The augmented Dickey-Fuller test of Dickey and Fuller (1979), in the ARMA-consistent lag-augmented form of Said and Dickey (1984) and with lag length set as in Ng and Perron (2001), does not reject a unit root with p = 0.5873, the nonparametric Phillips and Perron (1988) test concurs with p = 0.6875, and the Kwiatkowski et al. (1992) stationarity test rejects its trend-stationary null at p < 0.01, so the verdict is an unambiguous I(1). The GLS-detrended power escalation of Elliott, Rothenberg, and Stock (1996) is reserved for ambiguous outcomes under the house protocol and is not required on this clean reading.

Because the level is integrated, the mean-shift and serial-correlation diagnostics are run on the first difference, the stationary object those procedures require, since a break search or a portmanteau on an integrated level would spuriously segment and read near-unit autocorrelations (Bai and Perron 1998; Perron 1989). The multiple-break procedure of Bai and Perron (1998), computed by the dynamic-programming algorithm of Bai and Perron (2003), finds a single break in the mean of the differenced series, at 2021-08-01, read as parameter instability in the sense of Andrews (1993). The Ljung and Box (1978) portmanteau statistic, refining the original Box and Pierce (1970) form, is computed on the first difference and rejects the white-noise null at lags 12 and 24, with Q = 417.00 and Q = 441.48 and p = 0.000 and p = 0.000.

The corporate lending rate is a monthly interest-rate level with no posited low-integer seasonal component, so the seasonal-unit-root machinery of Hylleberg et al. (1990) and the Canova and Hansen (1995) seasonal-stationarity test carry no meaningful object and are deliberately not run (Beaulieu and Miron 1992; Ghysels and Osborn 2001).