Kondratiev wave¶

The contested hypothesis that capitalist economies exhibit roughly 40-60-year cycles of expansion and contraction, driven by clusters of general-purpose innovations whose diffusion generates a long upswing, saturates, and gives way to a depressive phase in which the next techno-economic paradigm gestates — a heuristic placement ladder more than a predictive model.

Core Idea¶

The Kondratiev wave — also called the K-wave or long wave — is the hypothesis that capitalist economies exhibit roughly 40–60-year cycles of expansion and contraction, driven by clusters of general-purpose technological innovations whose diffusion generates a long upswing, saturates, and gives way to a long depressive phase during which the next techno-economic paradigm gestates. Nikolai Kondratiev (1925) inferred the pattern from price, interest-rate, and output series across the industrialised economies of the 19^th and early 20^th centuries. Joseph Schumpeter subsequently embedded the K-wave in a three-cycle framework — Kitchin cycles of 3–5 years, Juglar cycles of 7–11 years, and Kondratiev cycles of 45–60 years — and linked each long upswing to a specific innovation cluster: steam and textiles, rail and steel, electrification, automobiles and petroleum, and (by later interpreters) information and communication technologies.

The mechanism proposed is that a cluster of co-evolving general-purpose technologies, supporting infrastructure, finance, and complementary skills defines a techno-economic paradigm. The installation phase sees capital rush into building out the new infrastructure ahead of full diffusion, producing finance-led booms and often speculative bubbles. As the paradigm matures, its marginal returns fall, capacity overshoots, and a prolonged depressive phase sets in during which the next paradigm forms beneath the dominant one — until a handover resets the cycle. Carlota Perez's elaboration of this structure (installation/deployment phasing, with a financial crash at the turning point) is the most developed contemporary version.

The hypothesis is controversial in mainstream macroeconomics: the sample is small (four to five putative cycles in industrial history), the cycle's existence depends sensitively on detrending choices, and there is no generative model precise enough to falsify the period claim. The wave is better established as a heuristic organiser of long-run economic history than as a quantitatively predictive model.

Structural Signature¶

Sig role-phrases:

the secular-trend substrate — long-run series of industrial output, prices, interest rates, and productivity, decomposed into trend plus a long-wave residual
the techno-economic paradigm cluster — a set of co-evolving general-purpose technologies plus supporting infrastructure, finance, and complementary skills that defines an upswing
the installation phase — capital and infrastructure built out ahead of full diffusion, finance-led and bubble-prone
the turning-point crash — the speculative crash falling at the handover between installation and deployment (Perez's phasing)
the saturation / exhaustion point — the maturation at which the paradigm's marginal returns fall and capacity overshoots
the depressive gestation phase — the long downswing during which the successor paradigm forms beneath the still-dominant old one
the handover — the replacement of the old paradigm by its successor, resetting the cycle (the ~40–60-year period)
the measurement controversy (its limitation) — the cycle's existence depends sensitively on detrending choices and rests on only four or five putative cases, so the period claim is contested; the schema places events on a heuristic ladder rather than forecasting them

What It Is Not¶

Not an established empirical law. The long wave is a contested hypothesis, not a confirmed regularity: the sample is only four or five putative cycles, the cycle's very existence depends sensitively on detrending choices, and mainstream macroeconomics rejects the period claim as an artifact. It is better established as a heuristic organiser of long-run history than as a settled fact.
Not a predictive model. Its register is placement-and-diagnosis, not forecasting: an analyst can locate an event within a cycle only after assuming the cycle is there. There is no generative model precise enough to run forward from the schema to a date for the next upswing, so the one thing it most wants to deliver — period prediction — is exactly what it cannot.
Not a fixed 50-year period. The "40–60-year" wavelength is a loose, disputed range inferred post hoc, not a constant the economy is observed to obey. The K-wave's distinctive addition — that diffusion, saturation, and paradigm replacement synchronize into a roughly fixed global period — is precisely the part that is neither well-established nor portable.
Not a deterministic clock. The schema does not assert that history must tick through phases on schedule regardless of policy, institutions, or contingency. It is a descriptive arc keyed to innovation clusters, not a mechanism that forces a handover at a set interval; the timing of saturation and succession is open, not fated.
Not a substrate-general cycle pattern. Cross-domain "K-wave" thinking (leadership cycles in geopolitics, strategic-foresight long cycles) is explicit analogy borrowing the rise-saturation-handover gestalt, not a rediscovery of the same mechanism. What genuinely travels is the parent set — periodicity, s_curve_diffusion, paradigm_shift, creative_destruction — not the Kondratiev wave, whose price-series, capitalist-phase-structure cargo stays home.

Scope of Application¶

The Kondratiev wave lives within one domain — the heterodox, evolutionary, and historical wing of macroeconomics studying long-run industrial capitalist economies — and even there it travels as a descriptive placement-and-diagnosis ladder rather than a predictive mechanism, so each habitat below carries the contested-status caveat that the fixed-period claim is rejected by the mainstream and rests on only four or five putative cases. The long-cycle "K-wave" gestalt borrowed in strategic-foresight and geopolitics writing (Modelski's leadership cycles) is explicit analogy, not a habitat; what those uses genuinely reach for travels via the parents (periodicity, s_curve_diffusion, paradigm_shift, creative_destruction, long_horizon).

Heterodox and evolutionary macroeconomics — the home niche; the trend-plus-long-wave decomposition and the cluster-plus-phase reading used to characterize secular movements in prices, output, interest rates, and productivity that do not look stationary at the half-century horizon.
Long-run economic history — organizing two centuries of industrial growth into successive innovation-cluster eras (steam and textiles, rail and steel, electrification, automobiles and petroleum, ICT) and placing episodes like the post-1973 slowdown or the 1990s ICT surge as positions on one arc.
Schumpeterian business-cycle theory — the wave embedded as the longest band in the three-cycle framework (Kitchin 3-5y, Juglar 7-11y, Kondratiev 45-60y), each long upswing tied to a major innovation cluster.
Perez's techno-economic-paradigm program — the most developed contemporary version, separating the innovation cluster from the finance that builds it out via installation-versus-deployment phasing, with the speculative crash read as the expected turning point between them.

Clarity¶

Within its home niche of heterodox and evolutionary macroeconomics and long-run economic history, naming the Kondratiev wave makes one fact about industrial growth hard to ignore: secular series of prices, output, and productivity do not look stationary at the half-century horizon, so the analyst cannot treat business-cycle fluctuations as departures from a flat or smoothly-trending baseline. The wave gives a vocabulary for the decades-long object that short-cycle macroeconomics leaves implicit — the diffusion timeline of a general-purpose technology, the multi-decade investment boom an infrastructure build-out (rail, electrification, broadband) generates, and the saturation lull that follows. It lets a historian ask where the present sits in such an arc ("late downswing of the fourth wave," "early upswing of the ICT paradigm") rather than treating each episode as unrelated.

The label also sharpens the distinction between the innovation cluster that drives a long upswing and the finance that builds it out ahead of full diffusion — Perez's installation-versus-deployment phasing, with the speculative crash falling at the turning point between them. That separation is what the bare notion of "technological progress" blurs. But the clarity is bounded, and the concept is most honest when it foregrounds its own limits: because the existence of the cycle depends sensitively on detrending choices and rests on only four or five putative cases, the wave organises long-run history as a heuristic ladder far better than it forecasts, and the period claim itself remains the thing it cannot make legible — it can place an event in a cycle only once one has assumed the cycle is there.

Manages Complexity¶

Two centuries of industrial economic history is a thicket — decades of price, output, interest-rate, and productivity series; a procession of distinct technologies (steam, textiles, rail, steel, electricity, automobiles, petroleum, information and communication); recurring booms, crashes, and prolonged slumps that on their face have nothing to do with one another. The Kondratiev wave compresses that thicket into a single repeating object: one upswing-downswing arc of 40-60 years keyed to one innovation cluster, with a fixed internal phase sequence — installation (finance rushes in to build the new infrastructure ahead of diffusion, often into a speculative bubble), a turning-point crash, maturation, saturation and overcapacity, and a depressive gestation in which the successor paradigm forms beneath the dominant one before a handover resets the cycle. Once the long-run series is read as trend-plus-long-wave rather than as a flat baseline with scattered disturbances, the historian stops treating each episode as sui generis and instead tracks a small set: which innovation cluster currently dominates, and which phase of the arc the economy occupies. From that pair the qualitative character of a period reads off — late-downswing stagnation, early-upswing investment boom, installation-phase bubble — and the post-1973 slowdown, the 1990s ICT surge, and a present "late fourth wave" or "early ICT paradigm" placement become positions on one ladder rather than disconnected events.

The compression is genuine but its register must be stated honestly, because that boundary is itself part of what the concept manages. The wave collapses long-run history into a teachable, placeable arc — a mnemonic ladder and an organising frame for evolutionary and historical macroeconomics — far more than it yields a forecast. The reduction buys a parsimonious narrative of which-paradigm-which-phase; it does not buy a generative, period-predicting model, because the cycle's very existence is sensitive to detrending choices and rests on only four or five putative cases. So the one parameter the schema most wants to fix — the period itself — is the one it cannot make legible: an analyst can locate an event within a cycle only after assuming the cycle is there. The honest summary of the compression, then, is that it tames the descriptive sprawl of long-run industrial history into a low-dimensional cluster-and-phase reading, while leaving the predictive question outside what the collapse can deliver.

Abstract Reasoning¶

The Kondratiev wave licenses reasoning that reads long-run industrial history as cluster-plus-phase, and its moves are predominantly placement and diagnostic rather than predictive — a register the concept itself insists on.

The foundational move is trend-plus-long-wave decomposition. Confronting secular series of prices, output, and productivity that do not look stationary at the half-century horizon, the analyst refuses to treat business-cycle fluctuations as departures from a flat or smoothly-trending baseline and instead reads the series as trend plus a 40-60-year wave. The reasoning runs from "these decades-long swings are not noise around a level" to "there is a long-wave object to be located," and it is what makes the diffusion timeline of a general-purpose technology and the multi-decade investment boom of an infrastructure build-out into tracked structures rather than implicit background.

The central move is positional placement on the arc. Given that a wave is assumed, the analyst locates the present on it by identifying two things — which innovation cluster currently dominates (steam and textiles, rail and steel, electrification, automobiles and petroleum, information and communication technologies) and which phase of the fixed internal sequence the economy occupies (installation, turning-point crash, maturation, saturation and overcapacity, depressive gestation). From that pair the qualitative character of the period reads off: a "late downswing of the fourth wave" implies stagnation and overcapacity, an "early upswing of the ICT paradigm" implies an investment boom, an installation phase implies bubble-prone finance. The reasoning is to infer the texture of an era from its coordinates on the ladder, so that the post-1973 slowdown and the 1990s ICT surge become positions on one arc rather than disconnected events.

A third move is phase-diagnostic separation of innovation from finance. The analyst distinguishes the innovation cluster that drives a long upswing from the finance that builds it out ahead of full diffusion — Perez's installation-versus-deployment phasing — and so reads a speculative bubble and crash not as an exogenous financial accident but as the expected turning point between installation and deployment. The reasoning predicts where in the arc a finance-led bubble belongs (the installation phase, ahead of diffusion) and when the crash falls (the handover between phases), a structural reading the bare notion of "technological progress" cannot supply.

A fourth move is order-of-paradigms succession reasoning. Because the schema posits that the next techno-economic paradigm gestates beneath the dominant one during the depressive phase before a handover resets the cycle, the analyst reasons about sequence: a maturing paradigm with falling marginal returns and overshooting capacity is read as the seedbed in which a successor is already forming, so the question "what comes after this era's technology?" is posed as a search for the gestating cluster rather than treated as unforeseeable. The move predicts the form of the transition (saturation, then handover) without committing to its date.

The honest boundary is that the move the schema most wants — period prediction — is not licensed. The analyst can place an event within a cycle only after assuming the cycle is there, and the existence of the wave is sensitive to detrending choices and rests on only four or five putative cases, so the reasoning cannot run forward from the model to a forecast of when the next upswing arrives. The concept supports analogical placement on a heuristic ladder and diagnostic reading of which-paradigm-which-phase; it does not support the generative, period-predicting inference that a quantitatively falsifiable cycle would, and an analyst reasoning with it stays on the placement-and-diagnosis side of that line.

Knowledge Transfer¶

The Kondratiev wave is a contested empirical hypothesis about one substrate — industrial capitalist economies at one timescale — and its register is placement-and-diagnosis rather than mechanism, so an honest account of its transfer must concede that even its within-domain travel is qualified. Within its home niche of heterodox, evolutionary, and historical macroeconomics the schema does transfer as a working heuristic: the trend-plus-long-wave decomposition, the cluster-plus-phase placement (which innovation cluster dominates; which phase of installation → turning-point crash → maturation → saturation → depressive gestation the economy occupies), and Perez's installation-versus-deployment separation of innovation from finance all carry across the canonical readings — steam and textiles, rail and steel, electrification, automobiles and petroleum, and the ICT paradigm — and across the post-1973 slowdown, the 1990s ICT surge, and a present "late fourth wave / early ICT" placement. But the transfer is of a descriptive ladder, not a predictive mechanism: the K-wave-specific commitment — that long diffusion, saturation, and paradigm replacement synchronize into a roughly fixed 40–60-year global period — is the one thing the schema cannot make portable even at home, because the cycle's existence depends sensitively on detrending choices and rests on only four or five putative cases, and mainstream macroeconomics rejects the period claim as an artifact. So within the domain the placement-and-diagnosis apparatus transfers; the fixed-period generative claim does not, and any rendered use should flag that contested status.

Beyond industrial macroeconomics the transfer is, candidly, analogy — and the entry is most honest when it says so. Uses of "K-wave" thinking in strategic-foresight and geopolitics writing (Modelski's leadership cycles) are explicit analogies that borrow the long-cycle gestalt, not independent rediscoveries of the same mechanism; they rename the components and reuse the rise-saturation-handover shape while dropping the price-series, capitalist-phase-structure content that is the wave's actual cargo. What genuinely travels — and what these analogies are really reaching for — is not the Kondratiev wave but the more general structural ingredients it bundles, each already a prime in its own right: periodicity (the generic recurring-cycle pattern, of which the K-wave is one disputed instance at one timescale), s_curve_diffusion (the long upswing and saturation, without committing to a fixed wavelength), paradigm_shift and creative_destruction (the Schumpeterian gestation-and-handover engine, without the cycle-length claim), and long_horizon (the decades-long object short-cycle macro leaves implicit). Strip the macroeconomic specifics and what remains is exactly that set of parents; the K-wave's distinctive addition — the synchronization into a roughly fixed period — is the part that is neither well-established nor portable. The honest move when reaching cross-domain is therefore to carry the parents (periodicity, S-curve diffusion, paradigm shift, creative destruction, long horizon) and to leave "Kondratiev wave," as named, at home as a heuristic organiser of long-run industrial history, flagged for its contested empirical standing. The full boundary — descriptive ladder versus generative model, parent mechanisms versus the home-bound period claim — is drawn in Structural Core vs. Domain Accent.

Examples¶

Canonical¶

Kondratiev's original 1925 analysis examined long-run series — wholesale prices, interest rates, wages, foreign trade, and coal and pig-iron output — for Britain, France, and the United States from the 1790s to the 1920s. Detrending the level series, he read the residuals as roughly two and a half long waves: an upswing peaking near 1815, then a downswing to about 1849; a second upswing peaking around 1873, then a downswing to about 1896; and a third upswing running into the 1910s-20s. Schumpeter later keyed each upswing to an innovation cluster — the first to steam and textiles, the second to railways and steel, the third to electricity, chemicals, and the internal-combustion engine — arguing that the diffusion of each general-purpose cluster drove the boom, saturated, and gave way to a depressive phase in which the successor gestated.

Mapped back: The price, output, and interest-rate series are the secular-trend substrate, and reading them as trend-plus-residual is the decomposition that exposes the wave. Each Schumpeterian innovation set (steam/textiles, rail/steel, electricity/chemicals) is the techno-economic paradigm cluster driving one upswing; each downswing turn (1815, 1873) marks the saturation / exhaustion point, and the ~50-year spacing between successive upswings is the handover period. Kondratiev's own caution about the small sample foreshadows the measurement controversy.

Applied / In Practice¶

Carlota Perez's Technological Revolutions and Financial Capital (2002) applied the long-wave structure to the information-technology surge she dates from Intel's 1971 microprocessor. In her installation-versus-deployment phasing, the 1990s saw finance rush into building out the new ICT infrastructure ahead of full diffusion, producing the dot-com bubble; she read the 2000-2001 NASDAQ crash not as an exogenous accident but as the expected turning-point crash between installation and deployment. On this placement the economy should then enter a "deployment" or golden-age phase in which the matured paradigm spreads productively across the wider economy. The reading organized a sprawling episode into cluster-and-phase coordinates rather than treating boom and crash as unrelated — while, per the concept's own caveat, forecasting nothing more precise than the form of the transition.

Mapped back: The ICT set launched by the microprocessor is the techno-economic paradigm cluster; the 1990s buildout is the installation phase and the dot-com bubble its finance-led, bubble-prone signature. The 2000-2001 NASDAQ collapse is the turning-point crash at the installation/deployment handover. That Perez places the crash structurally yet declines to date the golden age exemplifies the measurement controversy — placement, not prediction.

Structural Tensions¶

T1: Placement ladder versus period forecast (the register the concept insists on). The wave's power is that it turns unrelated episodes — the post-1973 slowdown, the 1990s ICT surge — into coordinates on one arc, letting a historian say "late fourth wave" or "early ICT paradigm" rather than treating each as sui generis. But that placement is available only once the cycle is assumed: the analyst locates an event within a wave after positing the wave, never deriving the wave from a forward-running model. The single inference the schema most wants — when the next upswing arrives — is exactly the one it cannot license, because the period claim is what placement takes for granted. So the concept is most useful precisely where it is least testable, and reaches for prediction only by smuggling in the assumption it was meant to earn. Diagnostic: Is the wave being used to place an event given an assumed cycle, or to forecast a date — the move the schema cannot support?

T2: Structural crash versus exogenous accident (installation/deployment as explanation or over-fit). Perez's separation of the innovation cluster from the finance that builds it out ahead of diffusion reads a speculative bubble and crash not as a random financial accident but as the expected turning point between installation and deployment — the dot-com bubble and the 2000-2001 NASDAQ collapse fall exactly where the schema says a crash belongs. This is genuine explanatory gain: it predicts where in the arc bubble-prone finance sits. But the same move is retrodictively cheap — any major crash can be relabeled a turning-point handover after the fact, and with only four or five cases there is no independent check that the crash was structural rather than coincidental. The reading that makes finance legible also makes the schema hard to disconfirm, since every bubble confirms it. Diagnostic: Was the crash's placement at the installation/deployment handover predicted before the fact, or fitted to it afterward?

T3: Signal versus artifact (the wave's existence hinges on detrending). A Kondratiev wave is not read off raw series; it appears only in the residual after a trend is removed, and the concept concedes that its very existence depends sensitively on which detrending choice is made. Different filters make the same 40-60-year swing appear or dissolve, and there is no principled, model-independent way to fix the choice — so the object under study is partly manufactured by the method used to expose it. With only four or five putative cycles across two centuries, there is too little data to adjudicate which detrending is right, so signal and artifact are not separable from the inside. The decomposition that lets the analyst see a long wave is the same operation that could be creating one. Diagnostic: Does the wave survive a range of reasonable detrending choices, or does it appear only under the one filter that produces it?

T4: Organising narrative versus falsifiable claim (parsimony bought at the cost of testability). The wave compresses two centuries of industrial history into a teachable cluster-and-phase arc — a mnemonic ladder that short-cycle macroeconomics cannot supply, real and pedagogically powerful. But there is no generative model precise enough to make the period claim falsifiable: the "40-60-year" range is loose, inferred post hoc, and elastic enough to accommodate almost any spacing, so the schema explains long-run history without risking refutation by it. The very looseness that lets the arc fit steam, rail, electricity, autos, and ICT alike is what strips it of predictive teeth. A frame that can absorb every case is an organiser, not a law — and the concept is at its most honest when it claims the former and disclaims the latter. Diagnostic: Could any observed sequence of booms and slumps falsify the wave, or does the loose period range let it absorb whatever history delivers?

T5: Fixed period versus open timing (a cycle that must not be a clock). To be a wave at all the schema needs a roughly fixed ~40-60-year period — the synchronization of diffusion, saturation, and paradigm replacement into a recurring wavelength is its distinctive claim. Yet the concept also disclaims being a deterministic clock: it insists the timing of saturation and handover is open to policy, institutions, and contingency, not fated to tick on schedule. These pull against each other. Pin the period tightly and the schema overclaims a regularity the data will not bear; loosen it to preserve contingency and the "cycle" degrades into the unremarkable observation that technologies eventually mature and are replaced. The wave must be periodic enough to be a wave and flexible enough to survive history, and it cannot maximize both. Diagnostic: Is the period being treated as a real recurring wavelength, or as an elastic label that bends to whatever interval each case happens to show?

T6: Autonomy versus reduction (its own named cycle or the domain instance of its parents). "Kondratiev wave" is a named schema with proprietary cargo — Kondratiev's price-series analysis, Schumpeter's three-cycle nesting, Perez's installation/deployment phasing, the specific steam→rail→electricity→autos→ICT ladder — and within heterodox and historical macroeconomics that apparatus travels as a working heuristic. But beyond industrial capitalism the named wave does not transfer as mechanism; what carries is the more general parents it bundles — periodicity, s_curve_diffusion, paradigm_shift, creative_destruction, and long_horizon — each portable without the cycle-length commitment. Tellingly, the K-wave's one distinctive addition over those parents — the synchronization into a roughly fixed global period — is exactly the part that is neither well-established nor portable; strip it and only the parents remain. Diagnostic: Resolve toward the parents (periodicity, S-curve diffusion, paradigm shift, creative destruction, long horizon) when reaching outside industrial macroeconomics; toward the named wave when placing an episode of industrial history in situ.

Structural–Framed Character¶

The Kondratiev wave is best placed mixed, sitting framed-of-isostasy in the manner of IS-LM: it is a theoretical schema whose subject matter is a human-institutional substrate (industrial capitalist economies), and — unusually — a contested one whose very object may be partly a measurement artifact. The five criteria distribute across both sides. Evaluative_weight is low and points structural: the wave is a descriptive placement-and-diagnosis ladder — it locates an episode as "late fourth wave" or "installation phase," rendering no verdict, praising and blaming nothing. But the rest pull framed. Human_practice_bound is high: techno-economic paradigms, innovation clusters, finance-led installation booms, and the capitalist phase structure are all constituted by human economic institutions, so strip the practice away and there is no long wave — nothing runs observer-free the way a lithosphere rebounds. Institutional_origin is framed in a double sense: the schema is an artifact of a heterodox-economics tradition (Kondratiev, Schumpeter, Perez), and the entry candidly concedes the cycle's very existence "depends sensitively on detrending choices," so the object is partly manufactured by the method that exposes it — a weaker structural credential than a mechanism nature performs regardless of observation. Vocab_travels is low for the distinctive layer: techno-economic paradigm, installation/deployment, the steam→rail→ICT ladder pin to industrial macroeconomics, and beyond it only the parents travel. And import_vs_recognize is heuristic recognition within the home niche but explicit analogy beyond it (Modelski's leadership cycles), with the parents doing the carrying.

Like kinetics, the Kondratiev wave is a composite whose portable content decomposes across several umbrella primes rather than one: a periodically recurring (periodicity) rise-saturation-handover diffusion arc (s_curve_diffusion) driven by a gestation-and-replacement engine (paradigm_shift, creative_destruction) over a decades-long object (long_horizon). Each of those skeletons is genuinely substrate-portable — but they are what the wave instantiates from its umbrellas, not what makes "Kondratiev wave" itself travel: the cross-domain reach belongs to those parents, while the price-series decomposition, the innovation-cluster ladder, and the capitalist installation/deployment phasing stay home. The decisive point is that the wave's one distinctive addition over its parents — the synchronization of diffusion, saturation, and paradigm replacement into a roughly fixed 40–60-year global period — is precisely the part that is neither well-established nor portable; the genuinely structural ingredients all belong to the parents, and the proprietary claim is the contested one. Its character: a contested, tradition-authored heuristic schema on a human-institutional substrate, structural only in the periodicity-diffusion-paradigm-shift skeletons it bundles from its umbrellas and framed in the fixed-period, techno-economic-paradigm content that gives "Kondratiev wave" its identity — mixed overall, and held toward the framed side by its contested empirical standing.

Structural Core vs. Domain Accent¶

This section decides why the Kondratiev wave is a domain-specific abstraction and not a prime, and it carries the case for its domain-specificity in one place. Like kinetics, the wave is a composite whose portable content decomposes across several umbrella primes rather than one.

What is skeletal (could lift toward a cross-domain prime). Strip the macroeconomics and several thin relational structures survive, each genuinely substrate-portable: a recurring cycle of expansion and contraction (periodicity); a rise-saturation diffusion arc — a new capability spreads, matures, and its marginal returns fall (s_curve_diffusion); a gestation-and-handover engine in which a successor forms beneath a maturing incumbent and eventually replaces it (paradigm_shift, creative_destruction); and a decades-long object that short-cycle analysis leaves implicit (long_horizon). Each of these is real structure that recurs far beyond industrial capitalism, which is why the entry instantiates that bundle of parents. But they are the cores the wave shares — decomposed across its parents — not what makes it distinctive.

What is domain-bound. Everything that makes the schema the Kondratiev wave in particular is industrial-macroeconomics furniture. The substrate is long-run series of prices, output, interest rates, and productivity decomposed into trend plus a long-wave residual; the driver is a techno-economic paradigm cluster of co-evolving general-purpose technologies plus finance and skills; the internal sequence is installation → turning-point crash → maturation → saturation → depressive gestation → handover (Perez's phasing); and the empirical anchor is the specific steam → rail → electrification → autos → ICT ladder. Crucially, the wave's one distinctive addition over its parents — the claim that diffusion, saturation, and paradigm replacement synchronize into a roughly fixed 40–60-year global period — is precisely the part that is neither well-established nor portable: it depends sensitively on detrending choices, rests on only four or five putative cases, and is rejected by mainstream macroeconomics as an artifact. The decisive test: strip the price-series decomposition and the capitalist installation/deployment phase structure, and what remains is just the parent skeletons; the proprietary, home-bound content is exactly the contested fixed-period claim.

Why this does not clear the prime bar. A prime is a relational structure whose vocabulary travels and whose transfer is recognition of the same mechanism, not analogy. The wave's transfer is qualified even at home. Within heterodox, evolutionary, and historical macroeconomics it travels as a descriptive placement-and-diagnosis ladder — the trend-plus-long-wave decomposition, the cluster-plus-phase placement, and the innovation-versus-finance separation carry across the canonical readings — but the fixed-period generative claim does not port even within the domain, so what transfers is a heuristic, not a predictive mechanism. Beyond industrial macroeconomics the named wave does not transfer as mechanism at all: strategic-foresight and geopolitical "K-wave" uses (Modelski's leadership cycles) are explicit analogy that borrow the rise-saturation-handover gestalt while dropping the price-series, capitalist-phase-structure cargo. And when the genuinely portable structure is wanted, it is already carried, in more general form, by the parents the wave bundles — periodicity, s_curve_diffusion, paradigm_shift, creative_destruction, long_horizon — each usable without the cycle-length commitment. The cross-domain reach belongs to those parents; "the Kondratiev wave," as named, packs the fixed-period synchronization claim and the techno-economic-paradigm apparatus that should stay home as a contested heuristic organiser of long-run industrial history.

Relationships to Other Abstractions¶

Current abstraction Kondratiev wave Domain-specific

Parents (5) — more general patterns this builds on

Kondratiev wave is part of Creative Destruction Prime

The live Kondratiev mechanism contains creative destruction as innovation- driven displacement of an exhausted techno-economic paradigm by its successor.
Kondratiev wave is part of Cultural Diffusion Prime

The Kondratiev mechanism contains cultural diffusion in its exact innovation-diffusion alias sense: a general-purpose technology spreads through adopters, accelerates, and saturates along an S-shaped adoption path.
Kondratiev wave is part of Cycle Prime

The Kondratiev schema contains a closed phase path from installation through crash, deployment, saturation, depressive gestation, and successor handover back to installation.
Kondratiev wave is part of, typical Speculative Bubble Prime

Perez-style Kondratiev waves typically contain a speculative bubble during finance-led installation, followed by the turning-point crash into deployment.
Kondratiev wave is a decomposition of Periodicity Prime

Removing industrial-history framing from the Kondratiev wave leaves the hypothesis of approximate repetition after a 40–60-year displacement, with phase drift and variable amplitude explicitly acknowledged.

Not to Be Confused With¶

Kitchin / Juglar (shorter business cycles). The shorter bands nested with the K-wave in Schumpeter's three-cycle framework: Kitchin inventory cycles (~3–5 years) and Juglar fixed-investment cycles (~7–11 years), against the Kondratiev's ~45–60 years. They differ in period, driver, and empirical standing — the shorter cycles are far better established than the contested long wave. Confusing them collapses distinct timescales. Tell: is the cycle a few-years inventory/investment fluctuation (Kitchin/Juglar), or a half-century innovation-cluster arc (Kondratiev)?
Techno-economic paradigm (Perez). The most-developed refinement of the long-wave idea, not a separate concept: Perez separates the innovation cluster from the finance that builds it out (installation vs deployment phasing, with the crash at the turning point). It is essentially the K-wave elaborated — sharper on finance, still carrying the contested fixed-period standing. Near-synonym / specialization relation. Tell: is the referent the base long-wave hypothesis (Kondratiev wave), or its installation/deployment financial-phasing elaboration (Perez's techno-economic paradigm)?
Secular stagnation. A contrast: the thesis that an economy can settle into persistently low growth and weak demand — a standing condition, not a phase of a recurring cycle. The Kondratiev wave's depressive phase is one trough of an oscillation expected to hand over to a new upswing; secular stagnation posits no such periodic recovery. Tell: is the low-growth state a phase of a cycle that will turn (Kondratiev downswing), or a persistent non-cyclical condition with no built-in recovery (secular stagnation)?
Modelski's leadership / hegemonic long cycles. The geopolitical analogy: recurring ~century-long cycles of great-power rise, dominance, and decline in world politics. It borrows the K-wave's long-cycle, rise-saturation-handover gestalt but runs on political-military hegemony, not innovation clusters and price series — explicit analogy, not the same mechanism. Tell: is the cycle about techno-economic paradigms in industrial capitalism (Kondratiev), or great-power political dominance in the international system (Modelski) — a borrowed gestalt carried by the shared parents, not the same mechanism?
S-curve diffusion (s_curve_diffusion). One parent the wave bundles, singled out because it is the most-confused-for: the long upswing-and-saturation of a spreading technology without any commitment to a fixed wavelength or synchronized global period. The K-wave's distinctive (and contested) addition is precisely that many such diffusions synchronize into a ~40–60-year cycle. Tell: is the claim just that a capability diffuses, matures, and saturates (S-curve, well-established), or that such arcs recur on a fixed global period (the Kondratiev synchronization claim, contested)?
The parent bundle (periodicity, s_curve_diffusion, paradigm_shift, creative_destruction, long_horizon). The Kondratiev wave is a composite of these substrate-portable structures — a recurring cycle, a diffusion arc, a gestation-and-handover engine, over a decades-long object. Cross-domain "long cycle" invocations genuinely reach for these, not for the price-series/capitalist-phase cargo. Tell: strip away the innovation-cluster ladder and the fixed-period claim and what remains — periodicity + diffusion + paradigm shift + creative destruction + long horizon — is the parent set, treated more fully elsewhere; carry those (not "Kondratiev wave") outside industrial macroeconomics.

Neighborhood in Abstraction Space¶

Kondratiev wave sits in a crowded region of the domain-specific corpus (15^th percentile for distinctiveness): several abstractions share nearly its structure, so a description that fits it tends to fit its neighbors too.

Family — Macroeconomic Cycles & Curves (16 abstractions)

Nearest neighbors

Computed from structural-signature embeddings · 2026-07-12