South Korea Coincident Composite Index (Cyclical)

KRCOINC is an index-number gauge of the current phase of aggregate economic activity, namely the detrended cyclical-fluctuation component of the coincident composite index. The series is seasonally adjusted, and KRED records it as published without any transformation.

The underlying coincident composite descends from an indicator tradition, one that defines the business cycle as the clustered comovement of many activity series and supplies the reference cycle against which a coincident measure is read (Burns and Mitchell 1946). Individual series are classified as roughly coincident with that cycle (Mitchell and Burns 1938). The composite-index and diffusion-index construction that combines the included coincident components into a single standardized aggregate was codified (Moore 1961), and the monthly composite was operationalized for real-time signaling (Shiskin 1961). That composite was later recast as the latent common factor shared by a small set of monthly coincident aggregates, giving the modern single-index foundation (Stock and Watson 1989; Stock and Watson 1991).

The cyclical-fluctuation value reported here is the deviation of that composite from its estimated trend, namely the specific cycle (Burns and Mitchell 1946). That specific cycle is rendered algorithmically datable (Bry and Boschan 1971).

The component aggregates include employment measured on the labor-force framework (Hussmanns, Mehran, and Verma 1990), whose status definitions are interpreted through labor-market flows (Clark and Summers 1979). The coincident composite is situated within the broader indicator system (Zarnowitz 1992), and because such an atheoretical comovement measure carries no structural identification, KRCOINC is read as a descriptive coincident gauge (Koopmans 1947).

KRED applies no transformation to KRCOINC, applying neither rescaling, deflating, smoothing, nor annualizing. Both the seasonal adjustment and the trend-cycle decomposition are already embedded in the published series at monthly cadence, so this account describes how such a composite cyclical indicator comes to exist rather than any KRED operation.

The underlying coincident composite is built in an indicator tradition that screens hundreds of series against historical revivals to select the roughly coincident components and to define the reference cycle (Mitchell and Burns 1938; Burns and Mitchell 1946). The standardized composite-index and diffusion-index arithmetic that combines those components was set out (Moore 1961), and its monthly real-time variant was supplied (Shiskin 1961). A dynamic-factor formalization in which the index is the estimated common factor of the coincident aggregates was subsequently provided (Stock and Watson 1989; Stock and Watson 1991).

The cyclical-fluctuation value extracts the deviation of the composite from a fitted long-run trend, isolating the specific cycle (Burns and Mitchell 1946). This turning-point determination is rendered reproducible through programming (Bry and Boschan 1971).

Because the series is seasonally adjusted at source, the qualitative response-balancing methods enter only where survey-based diffusion components feed the composite (Anderson 1952; Theil 1952; Carlson and Parkin 1975). These survey-based construction procedures have been reviewed in the literature (Nardo 2003). The whole apparatus is a measurement convention rather than a model (Zarnowitz 1992; Koopmans 1947), and KRED ingests the resulting monthly value unaltered.

As a coincident gauge, KRCOINC dates the current cyclical phase of aggregate activity rather than anticipating it, the role assigned to roughly coincident series within the reference-cycle scheme (Burns and Mitchell 1946; Mitchell and Burns 1938). The monthly composite is intended for real-time recognition of expansions and contractions (Shiskin 1961; Moore 1961), and the same single index is interpreted as an estimate of the unobserved state of the economy that summarizes employment, output, and income comovement (Stock and Watson 1989).

The cyclical-fluctuation form sharpens the turning-point reading by removing the trend. The coincident composite's record at marking turning points has been evaluated (Zarnowitz 1992), and its peaks and troughs are algorithmically datable (Bry and Boschan 1971).

Because employment is a central coincident component, the labor-force concepts and the flow interpretation of labor markets bear directly on how the index moves around recessions (Hussmanns, Mehran, and Verma 1990; Clark and Summers 1979).

The coincident reading is routinely cross-checked against attitudinal gauges, where KRCOINC functions as the contemporaneous activity benchmark in that comparison. Consumer sentiment has been established as a measurable determinant of spending (Katona 1951), and its independent cyclical content has been separately assessed (Ludvigson 2004).

In market use KRCOINC confirms rather than forecasts the cyclical regime, so it times the reference phase against which leading and lagging signals are judged and anchors business-cycle-aware positioning (Mitchell and Burns 1938; Burns and Mitchell 1946).

Investors read the cyclical-fluctuation value as a contemporaneous regime label and treat the single index as the state variable that conditions cross-asset returns (Stock and Watson 1989; Stock and Watson 1991). This reading draws on the composite-index construction and its monthly real-time variant (Moore 1961; Shiskin 1961).

The coincident composite's reliability at confirming turning points has been documented (Zarnowitz 1992), and its peaks and troughs are converted into datable form for backtesting regime-switching allocations (Bry and Boschan 1971).

Attitudinal series complement the hard-data reading, so KRCOINC serves as the activity yardstick against which sentiment-driven and rate-driven trades are calibrated. The spending-predictive content of consumer sentiment has been benchmarked (Carroll, Fuhrer, and Wilcox 1994), its independent signal has been weighed separately (Ludvigson 2004), and its behavioral basis has been supplied (Katona 1951).

KRCOINC is a seasonally-adjusted-at-source coincident composite index, and its published level is the statistical object. The sample is 436 monthly observations spanning 1990-01 to 2026-04, and the unit-root triplet is fitted with a constant and trend.

The augmented Dickey-Fuller regression of Dickey and Fuller (1979), in the lag-augmented form of Said and Dickey (1984), rejects the unit root at p = 0.000, the nonparametric test of Phillips and Perron (1988) rejects at p = 0.012, and the stationarity-null test does not reject at p > 0.10 (Kwiatkowski et al. 1992), so all three procedures concur on a clean trend-stationary I(0) classification without ambiguity. The GLS-detrended escalation of Elliott, Rothenberg, and Stock (1996) is reserved for ambiguous outcomes under the house protocol and is not needed on this clean reading.

Because the three procedures agree on a clean I(0) reading, the level is the stationary object, so the portmanteau and the break search are object-valid on the level itself (Perron 1989; Hamilton 2018). The portmanteau statistic of Ljung and Box (1978), refining the form of Box and Pierce (1970), rejects the white-noise null at p = 0.000, and the multiple-break procedure of Bai and Perron (1998), computed by the dynamic-programming algorithm of Bai and Perron (2003), locates a single break in the mean level at 1998-02, which reads as a shift in the trend level across the sample rather than a stochastic trend.

The seasonal battery is run with the interpretation that any absence of seasonality reflects the source adjustment rather than an inherent property. The seasonal-unit-root test of Hylleberg et al. (1990), with the monthly mechanics of Beaulieu and Miron (1993), rejects unit roots at the seasonal frequencies at p = 0.000, and the seasonal dummies are not jointly significant at p = 0.942, so no deterministic seasonality survives, which reflects the seasonal adjustment already applied at source and not an inherent absence of a seasonal pattern, consistent with the stationary-seasonality reading of Canova and Hansen (1995) and the seasonal-adjustment literature (Ghysels and Osborn 2001).