New Trade Theory

New Trade Theory is the general-equilibrium framework of international trade pioneered by Paul R. Krugman in his foundational 1979 Journal of International Economics and 1980 American Economic Review papers—alongside contemporaneous work by Kelvin Lancaster (1980) and Elhanan Helpman (1981)—which explained why the majority of post-World War II global commerce occurred as two-way intra-industry trade in differentiated manufactured goods between industrialized countries with similar factor endowments and technology levels. Classical Ricardian theory and neoclassical Heckscher–Ohlin theory (including the Stolper–Samuelson theorem) grounded international trade in exogenous inter-country differences in comparative advantage, predicting that countries with identical capital-labor ratios and identical technologies would engage in zero trade, and that opening to trade would trigger painful inter-industry factor income redistribution. By integrating internal economies of scale with the Dixit–Stiglitz Model of monopolistic competition, Krugman demonstrated that even identical economies achieve substantial mutual welfare gains from trade through expanded product variety and pro-competitive firm scale expansion, and that positive transport costs interact with increasing returns to concentrate production in larger markets via the home-market effect.

In his 1979 paper, "Increasing Returns, Monopolistic Competition, and International Trade," Krugman formulated a general additively separable symmetric utility function across n differentiated varieties:

U = ∑_i=1ⁿ v(c_i), with v′(c_i) > 0, v″(c_i) < 0

where c_i denotes per-capita consumption of variety i. Crucially, Krugman (1979) assumed that the price elasticity of demand facing an individual firm, ε(c_i) = −v′(c_i) / [c_i v″(c_i)] > 1, decreases as per-capita consumption c_i rises (∂ε / ∂c_i < 0), an assumption known as Marshall's Second Law of Demand.

Each variety is produced by a single firm using labor as the sole input under internal economies of scale, characterized by a fixed overhead labor requirement α > 0 and a constant marginal labor requirement β > 0:

l_i = α + β x_i

where x_i = L c_i is total firm output in an economy of L workers receiving nominal wage w. Equilibrium in the Krugman (1979) economy is determined by two relationships between the real price p / w and per-capita consumption c in a symmetric equilibrium (c_i = c, x_i = L c):

1. Profit-Maximization (PP) Schedule: Setting marginal revenue equal to marginal cost βw yields the markup pricing condition:

p / w = β [ ε(c) / (ε(c) − 1) ] Because ∂ε / ∂c < 0, a higher per-capita consumption c lowers demand elasticity ε(c) and raises the firm's optimal markup, so the PP schedule slopes upward in (c, p/w) space.

1. Zero-Profit Free-Entry (ZZ) Schedule: Free entry and exit drive economic profits to zero, equating price to average cost (p = AC = (α + βx)w / x):

p / w = β + α / (L c) Because fixed overhead α is spread over a larger output as c increases, the ZZ schedule slopes downward in (c, p/w) space.

When two symmetric economies of size L open to frictionless trade (effectively doubling the integrated market population from L to 2L, exactly like an episode of labor force growth), the ZZ schedule shifts downward and to the left. Moving from the autarky intersection to the open-economy intersection along the upward-sloping PP curve reduces per-capita consumption of each individual variety (c falls), which increases the demand elasticity ε(c) and forces firms to cut their markups. Consequently, opening to trade in Krugman (1979) delivers three simultaneous adjustments:

• Pro-Competitive Scale Expansion: Lower markups combined with zero profits require each surviving firm to operate further down its average cost curve, expanding total firm output x = (L + L*) c and raising real wages w / p.
• Incomplete Variety Doubling: Because each firm produces a larger output x, the number of firms in each country n = L / (α + βx) falls slightly, so the total number of varieties available to consumers in the integrated world economy (n + n*) is greater than in autarky (n_autarky) but strictly less than 2n_autarky.
• Dual Welfare Gains: Consumers gain both from lower prices (higher real wages w/p) due to pro-competitive scale economies and from greater total product variety (n + n* > n_autarky).

In his 1980 American Economic Review paper, "Scale Economies, Product Differentiation, and the Pattern of Trade," Krugman specialized the utility function to the symmetric Constant Elasticity of Substitution (CES) specification of the Dixit–Stiglitz Model:

U = ( ∑_i=1^N c_i^θ )^{1 / θ}, 0 < θ < 1

where the elasticity of substitution between any two varieties is constant at σ = 1 / (1 − θ) > 1. Because the number of varieties N is large, each monopolistically competitive firm treats the aggregate price index as parametric and faces a constant price elasticity of demand σ regardless of market size L or per-capita consumption c_i.

Under constant elasticity σ, the profit-maximizing markup is invariant to market integration:

p = ( β w ) / θ = [ σ / (σ − 1) ] β w

Combining this fixed markup price with the zero-profit free-entry condition π_i = p x_i − (α + β x_i)w = 0 pins down a unique, constant equilibrium firm output x* and firm employment l* that depend only on technology parameters (α, β) and taste parameter θ:

x* = [ α θ ] / [ β (1 − θ) ] = [ α (σ − 1) ] / β and l* = α + β x* = α / (1 − θ) = α σ

Full employment in an economy of labor force L therefore determines the equilibrium number of domestically produced varieties n as a strict linear proportion of population size:

n = L / l* = [ L (1 − θ) ] / α = L / (α σ)

When two countries with labor forces L and L* open to costless trade in the Krugman (1980) CES model, firm scale x*, real price p / w, and the number of firms in each country (n and n*) remain unchanged, meaning there is zero pro-competitive firm scale expansion. Instead, all welfare gains from trade arise through the love-of-variety channel: every consumer gains access to all n + n* = (L + L*) / (α σ) global varieties, raising representative consumer utility from U_autarky = n^(1−θ)/θ (w/p) to U_trade = (n + n*)^(1−θ)/θ (w/p) = (n + n*)^1/(σ−1) (w/p). Even with zero comparative advantage and zero change in firm scale, every worker in both countries is strictly better off.

In Section II and III of Krugman (1980), Krugman introduced Samuelson iceberg transport costs between two countries: of every unit of a differentiated good shipped abroad, only a fraction g ∈ (0, 1) arrives at the destination (equivalent to an iceberg cost factor τ = 1 / g > 1). If a domestic firm charges an FOB mill price p, foreign consumers face an effective delivered price p / g = τ p. Because CES demand has constant price elasticity σ = 1 / (1 − θ) in both markets, firms do not absorb freight and set the same FOB markup p = βw / θ, leaving equilibrium firm size x* and variety counts n = L / (α σ) unchanged when countries have only one sector.

To analyze how market size shapes industrial specialization and wages, Krugman (1980) considered two countries with asymmetric labor endowments (L > L*) and two consumer groups or two sectors—or solved for the relative wage ω = w / w* required to balance bilateral trade when both countries produce differentiated manufactures. Defining the freeness of trade parameter ϕ = g^{σ − 1} = τ^{1 − σ} ∈ (0, 1), let a two-sector economy contain a costlessly traded homogeneous constant-returns agricultural sector (fixing w = w* = 1) alongside the differentiated manufacturing sector (consuming expenditure share μ of income). In equilibrium, manufacturing firms enter each country until zero profits hold, yielding a foundational spatial economics result known as the Home-Market Effect (HME):

s_n = n / (n + n*) = 1 / 2 + [ (1 + ϕ) / (1 − ϕ) ] ( s_L − 1 / 2 ), for s_L ∈ ( ϕ / (1 + ϕ), 1 / (1 + ϕ) )

where s_L = L / (L + L*) > 1 / 2 is the larger country's share of world expenditure and s_n is its share of world manufacturing varieties. Because (1 + ϕ) / (1 − ϕ) > 1 for any positive transport cost (ϕ > 0), a country with a larger domestic market (s_L > 1/2) attracts a more-than-proportionate share of the increasing-returns manufacturing industry (s_n > s_L) and becomes a net exporter of differentiated manufactured goods! Intuitively, because each firm incurs a fixed setup cost α at a single plant location, locating in the larger market allows the firm to serve the majority of its customers without incurring iceberg transport losses g, saving transport costs. If all varieties remained in the single-sector model without a homogeneous outside good, this home-market advantage manifests as a higher nominal and real wage (w > w*, w/P > w*/P*) in the larger economy.

Elhanan Helpman and Paul Krugman (1985) unified New Trade Theory with classical Heckscher–Ohlin factor-proportions theory in Market Structure and Foreign Trade. Consider two countries (Home and Foreign) endowed with two factors—capital (K) and labor (L)—producing a labor-intensive homogeneous good (food) under constant returns to scale and a capital-intensive differentiated good (manufactures) under Dixit–Stiglitz monopolistic competition.

Within the Factor Price Equalization (FPE) set, Helpman and Krugman demonstrated that total trade decomposes into two orthogonal components:

1. Inter-Industry Trade (Heckscher–Ohlin Net Trade): Driven by international differences in relative factor endowments (K/L vs. K*/L*). The capital-abundant country is a net exporter of manufactures and a net importer of homogeneous food.
2. Intra-Industry Trade (Chamberlinian Two-Way Trade): Driven by internal economies of scale and product differentiation. Even the labor-abundant country produces a positive measure of differentiated manufacturing varieties n* and exports them to the capital-abundant country, generating two-way trade within the manufacturing sector measured by the Grubel–Lloyd Intra-Industry Trade Index:

IIT_j = 1 − ( | X_j − M_j | ) / ( X_j + M_j ) ∈ [0, 1]

As two trading partners converge in their capital-labor ratios (K/L → K*/L*) and become symmetric in GDP size, net Heckscher–Ohlin inter-industry trade vanishes (IIT → 1) and bilateral trade volume is maximized for a given group GDP—explaining the empirical success of the gravity model of trade among OECD economies and laying the theoretical foundation for Krugman's (1991) Core–Periphery New Economic Geography model (for which Krugman was awarded the 2008 Nobel Memorial Prize in Economic Sciences) as well as the heterogeneous-firm Melitz Model (2003).

How does New Trade Theory explain trade between countries with identical technology and factor endowments?

In classical Ricardian and Heckscher–Ohlin models under constant returns to scale, identical countries have identical autarky relative prices and therefore zero incentive to trade. In New Trade Theory, internal economies of scale (fixed setup cost α > 0) prevent any single country from profitably producing the full spectrum of differentiated varieties on its own. Opening to trade creates an integrated world market where each firm specializes in a distinct variety (i ≠ j) and sells to consumers in both countries, generating pure two-way intra-industry trade (IIT = 1) driven by scale economies rather than comparative advantage.

What is the Home-Market Effect (HME) and how does it differ from comparative advantage?

In neoclassical constant-returns trade models, an exogenous increase in domestic demand for a good causes a country to become a net importer of that good. Krugman (1980) proved that when an industry exhibits internal economies of scale and faces positive transport costs, the opposite occurs: a country with a larger domestic demand for differentiated goods attracts a more-than-proportional share of firms (s_n > s_L) because firms locate in the larger market to economize on shipping costs while concentrating production in a single plant, making the larger country a net exporter of those goods.

Why does opening to trade mitigate income-distribution conflict in New Trade Theory compared to the Stolper–Samuelson theorem?

According to the Stolper–Samuelson theorem, inter-industry Heckscher–Ohlin trade raises the real return of a country's abundant factor while reducing the real return of its scarce factor. In Krugman (1981) and Helpman–Krugman (1985), when trade is predominantly intra-industry between similarly endowed countries, resources reallocate across varieties within the same industry rather than across industries with different factor intensities, allowing both capital and labor (including the scarce factor) to experience net real income gains via lower aggregate price indices P and expanded variety access.

1. Krugman, P. R. (1979). Increasing returns, monopolistic competition, and international trade. Journal of International Economics, 9(4), 469–479. https://doi.org/10.1016/0022-1996(79)90017-590017-5)
2. Krugman, P. R. (1980). Scale economies, product differentiation, and the pattern of trade. The American Economic Review, 70(5), 950–959. https://www.jstor.org/stable/1805774
3. Krugman, P. R. (1981). Intraindustry specialization and the gains from trade. Journal of Political Economy, 89(5), 959–973. https://doi.org/10.1086/261015
4. Helpman, E. (1981). International trade in the presence of product differentiation, economies of scale and monopolistic competition: A Chamberlin-Heckscher-Ohlin approach. Journal of International Economics, 11(3), 305–340. https://doi.org/10.1016/0022-1996(81)90001-590001-5)
5. Krugman, P. R. (1991). Increasing returns and economic geography. Journal of Political Economy, 99(3), 483–499. https://doi.org/10.1086/261763