Kaldor's Growth Facts and Stylized Stories: What Verdoorn and 'Stylized Facts' Can Teach Development Readers

The Person First: Kaldor as Pattern-Spotter, Not Axiom-Builder

Nicholas Kaldor (1908–1986) is often filed under “Cambridge controversies, prices of production, and arguments with everyone,” but a large part of his legacy is humbler: he looked at the world economy in the 1950s and 1960s and asked what seemed repeatedly true about growth and distribution across countries and time. The facts he highlighted were not laboratory-clean; he called them stylized facts—a phrase that, in the wrong hands, sounds like a polite synonym for making things up. In Kaldor’s hands, it was closer to: here is what a serious observer keeps bumping into, even though measurement is hard and every country is unique.

If you are new to the Keynesian and post-Keynesian world, a useful map is the contrast between a neoclassical growth model that deduces paths from a small set of production-function assumptions, and a Kaldorian style that is willing to start from patterns in national income accounts and then look for mechanisms—effective demand, returns to scale in some sectors, manufacturing as an engine, export-led dynamics, and so on. Our multiplier in plain terms is one cousin of this way of thinking: demand can matter, not only in the short run.

What People Mean by “Kaldor’s Growth Facts”

Kaldor’s “facts” were not a single list handed down in stone; later writers consolidated them, sometimes adding or trimming. A common modern classroom summary (after Charles Jones and other growth historians) is close to: relatively stable capital/output ratios; stable savings and investment rates; stable labor and capital shares of income in the long data for many countries; convergence is weak at best without controlling for a half-dozen other things; and rich countries keep growing rather than all settling to zero growth, even if growth rates are not a simple constant. Kaldor’s own mid-century emphasis included labor’s share moving within bounds and capital deepening not absorbing all of national income, among other items.

Jargon note: A stylized fact is a pattern strong enough in aggregate data to guide research, but not a universal law. Think “broadly true, sometimes messy, always needing country context.”

The point of collecting such facts is not to win a fact-check. It is to discipline the theory you allow yourself. If a model, no matter how elegant, always predicts a sharp collapse of labor’s share in exactly the year your data say it did not happen, the model is not yet social science. Kaldor’s temperament—shared with a generation who lived through the Depression and the Trente Glorieuses—was allergic to a priori micro foundations that required reality to pretend to be a frictionless Walrasian fable. That is not the same as refusing models; it is a preference for models that survive contact with history.

Verdoorn’s Law: Why Manufacturing Shows Up in the Story

One dynamic that post-Keynesians and structuralist development economists often call Verdoorn’s law (after the Dutch economist Petrus Verdoorn, 1911–1992) is the empirical regularity that in manufacturing, productivity growth is positively related to the growth of output itself, over reasonable horizons. A loose intuition: when plants run fuller, they learn, specialize, and adopt techniques that raise output per hour; agglomeration and supplier networks improve; increasing returns appear at a sectoral level, even if each firm is small.

Jargon note: Increasing returns at sector level can coexist with competition in product markets if the returns come from shared inputs, knowledge spillovers, or thick markets—classic endogenous-growth themes and old Young–Kaldor stories about cumulative causation.

A serious reader of development must hold two truths at once. First, a poor country is not a baby rich country: structure matters—Prebisch–Singer terms-of-trade arguments and land reform and state capacity (see state capacity in development) are not footnotes. Second, a country that never gets a self-reinforcing sector where productivity and demand rise together may get stuck, even with brave entrepreneurs. The Verdoorn flavor is a warning against treating labor as an undifferentiated blob and against assuming constant returns in every line of the national accounts. Our Sraffian production primer offers another angle on prices, distribution, and technical choice in multi-sector systems.

Kaldor sometimes pressed a policy moral from this: industrial policy that builds manufacturing depth can pay macro returns—controversial, because services-rich success stories and Dutch disease in resource booms complicate a simple manufacturing fetish. The reader’s job is not to pick a tribe; it is to ask what scale and linkages a sector offers for this country at this technology frontier.

The Manufacturing Share Debate, Briefly and Fairly

The share of manufacturing in employment has fallen in many now-rich countries, mostly because productivity rose faster in manufacturing than in services (a Baumol “cost disease” relative price point), and because consumer demand shifts. That does not automatically refute a Verdoorn-style engine for a late developer today, where a manufacturing push might still be the fast route to learning and export discipline. The historical sequence (industrialize, then later deindustrialize in employment terms) is not a proof that a poor country can skip the entire ladder while relying only on extractives and tourism—resource curse arguments matter here.

Jargon note: Dutch disease is the name for a currency appreciation and non-tradable price surge when resource exports boom, hollowing out manufacturing and farming that compete at world prices.

From Stylized Facts to Cumulative Causation: Myrdal, Young, and Kaldor in One Room

A related idea Kaldor championed, though not in formal DSGE, is cumulative causation—a Gunnar Myrdal–flavored point that success breeds success when demand, investment, and skills cluster together, while lagging regions can also get locked in, due to out-migration, weak tax bases, and self-fulfilling expectations. The policy stakes are not “market always self-corrects” versus “state always must plan,” but: which thresholds of coordination are missing, and who can pay the initial cost of a public good that unlocks a Verdoorn loop? Our Gerschenkron, late industrialization, and catch-up growth article helps connect the late industrializer angle.

This is where the Keynes lineage matters for readers schooled in rational representative agents. Keynes stressed uncertainty and the fragility of long-term expectations; Kaldor used history to show that paths branch. If that sounds closer to a narrative social science than a single ergodic steady state, you are not wrong. Post-Keynesian macro, at its best, is explicit about the epistemic limits that our Minsky on fragility article names in a financial register.

What Empirical Work in the 21st Century Did to Kaldor’s Menu

Economic history since Kaldor has refined each “fact” with new national accounts and Penn World Table–style work.

Kaldor, Cambridge, and the Capital Controversies: A One-Screen Map

No essay on Kaldor is complete if it pretends growth “facts” float free of distribution and pricing. In the Cambridge capital controversies of the 1960s, Kaldor was a combatant in arguments about capital aggregation and reswitching—whether smooth neoclassical parables of substitution between “capital” and labor could survive Sraffian criticism. You do not have to pick a side to see why he cared; our gentle entries on Sraffa’s Production of Commodities and the transformation problem supply the missing plumbing.

Plain takeaway: if the profit rate and the growth path are joint outcomes of markup pricing, class conflict, and technical choice—as many post-Keynesians argue—then a single representative agent with a Cobb–Douglas production function may miss the levers that moved British and Indian history in Kaldor’s applied essays. Stylized facts about labor’s share stop being mere description; they become battlefields over who sets prices and who bears adjustment when trade opens or austerity bites.

Jargon note: Cobb–Douglas is a workhorse function in which output is a simple power mix of labor and capital with constant returns; it delivers neat factor shares that real data often violate unless you massage what counts as “capital.” For a first pass at why reswitching terrifies a certain style of neoclassical parable, think: the same technique can become profitable again as the wage–profit configuration moves—so “more capital-intensity” is not a single, immutable ladder.

Capital shares, labor shares, and the role of housing in capital measurement all look different when intangibles and land are handled carefully. Convergence results depend on which conditional variables you include—human capital, institutions, and geography all matter, echoing Acemoglu and institutions themes.

None of that kills Kaldor’s method—look at the data, argue about mechanisms, stay suspicious of a single parable—it only suggests updating the list and keeping humility about policy levers when open economy financial cycles can undo a domestic industrial push in months. That is a reason post-Keynesian readers return to balance-of-payments and currency constraints; see Bretton Woods and orders of money and endogenous money in tandem for intuitions on how external finance interacts with “domestic” growth.

A Teaching Suggestion: The “Stylized Fact” as Writing Discipline

If you are writing a development brief, try Kaldor’s discipline as a writer: state three patterns you believe hold for your case country; then for each pattern, name one mechanism and one threat that would falsify your story. If you find yourself unable to state a falsifier, you may be in narrative rather than in evidence—a fault no school in economics has a monopoly on.

What Kaldor’s Lens Gets Wrong, Sometimes

• Data are national: stylized country facts can break inside regions and for genders—feminist economics and care remind us the macro totals hide work.
• Manufacturing is not a moral category; polluting, low-road factories are not a development destiny.
• State-led successes and state-led catastrophes both exist; facts do not select a political system without institutional detail.
• Kaldor himself had policy episodes readers judge harshly in hindsight; intellectual tools and political choices are not the same thing.

A Regression Intuition (One Paragraph) and Why It Is Not a Magic Trick

If you are comfortable with a single sentence of econometrics without a matrix: picture regressing productivity growth in manufacturing on output growth in manufacturing, plus a few controls for human capital and imported inputs. A positive partial correlation is consistent with a Verdoorn story; it is not proof of a deep law of nature, because reverse causation and omitted variable bias—variables you did not measure that push both series—abound. The post-Keynesian habit is to pair such results with input–output maps, case histories, and balance-of-payments constraints, not a single t-stat and a victory lap. That is the open economy, money, and trade world in a nutshell.

Jargon note: Omitted variable bias is what happens when a variable outside your regression secretly drives both the left- and right-hand sides; your computer may report precision that your theory does not earn.

Conclusion: Stories That Fit the World More Than They Fit the Chalkboard

Kaldor’s “growth facts” and the Verdoorn link between output and productivity discipline theory without replacing it: they embarrass models that only fit balanced growth in closed economies on blackboards. The Austrian capital structure story and a Kaldorian cumulative causation cluster disagree on mechanism and policy, but both insist that time and production structure stay in the frame when you talk about money and trade at macro scale.

Further Reading

• Charles I. Jones, “The Facts of Economic Growth” in the Handbook of Macroeconomics (2016) — a modern, careful update of the stylized-facts genre.
• Nicholas Kaldor, Causes of the Slow Rate of Economic Growth in the United Kingdom (Inaugural Lecture, 1966) — a famous policy-leaning text that helped shape the deindustrialization debate.
• John McCombie and Antonio Spasciani, Kaldor and Verdoorn’s Law (2017), working paper to book chapters — a technical entry with empirical surveys.
• Allyn Young, “Increasing Returns and Economic Progress” (1928) — the classic upstream to endogenous growth ideas before the phrase existed.
• Jesús Antonio de la Fuente, Mathematical Methods and Models for Economists (2000) — for readers who want the algebra behind growth accounting without the hype.

---

For education only; not policy or investment advice.