## _wp0656

## Source details

**Canonical URL:** [_wp0656](https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf)

## Other formats

- [Markdown version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf.md)
- [Structured JSON version](/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf.json)

---

### Introduction and research question
- Central question: How does opening to international trade affect a country’s institutions (contract enforcement, property rights, rule of law)?
- Core insight: Trade can have two opposing effects on institutions:
  - Foreign competition (disciplining) effect: trade tends to raise preferences for better institutions because foreign competition reduces domestic rents.
  - Political power effect: trade increases firm-size dispersion (Melitz effect), shifting political weight to large exporters who, in the model, prefer worse institutions to limit entry and competition.
- Historical contextual examples cited: Caribbean sugar boom (18th century), coffee boom in Latin America (19th century), cotton and cattle booms in Central America (mid-20th century).

### Model overview and mechanisms
- Environment:
  - Monopolistic competition with heterogeneous producers (Melitz-style).
  - Institutions summarized by a common fixed cost f (higher f = worse institutions).
- Firm heterogeneity and trade:
  - Firms differ by marginal cost parameter a (productivity).
  - Only most productive firms export; trade increases inequality of firm size/profits: largest firms grow larger, smaller firms shrink or exit.
  - Political weight of an agent: wλ(w) = λ0 + λ1 w with λ0, λ1 > 0.
  - Political decision via weighted-median/pivotal voter; pivotal voter’s preferences determine equilibrium f.
- Preferences over f:
  - Each firm maximizes real wealth (real profits); more productive firms typically prefer higher f because higher f reduces entry and raises variable profits disproportionately for them.
  - Preference Curve r.fp gives, for each firm productivity a, the preferred level of f; it is nonincreasing and strictly decreasing for some values of a (Proposition 2, Corollary 3).

### Goods and factor market equilibrium (autarky and trade)
- Preferences and demand:
  - U = α ln z + β ∫v∈V ln xv dv, with ε = 1/(1-α) and A ≡ (∫ p(v)^{1-ε} dv)^{1/(1-ε)} defined as demand shift parameter.
- Firm profit expressions:
  - Autarky profit: πα(a) = (1/(1-α)) A E a^{1-ε} − f (equation (3) context).
- Autarky equilibrium:
  - Zero-profit cutoff A_a defined by equation (5): (1/(1-α)) A E a_A^{1-ε} − f = 0.
  - Goods market clearing: E L + A n [ (1/(1-α)) a_A^{1-ε} V_a − f ] = 0 (equation (6) context).
- Trade setup:
  - Two countries N and S with labor endowments N_L and S_L and masses of producers N_n and S_n.
  - Trade costs: per-unit iceberg τ and export fixed cost X_f. To export requires paying domestic fixed cost f and X_f.
  - Domestic and export profit expressions given by equations (8) and (9).
  - Cutoffs under trade: production cutoffs i_D a and export cutoffs i_X a characterized by equations (11)–(14).
- Comparative-static statements comparing trade vs autarky (text statements preserved):
  - i) i_AD a_D ≥ a_A: higher productivity required to produce domestically under trade.
  - ii) For producers operating under trade, domestic profits π_Da < π_Aa.
  - iii) There exists a cutoff a_X below which a firm earns higher total profits under trade than autarky (π_X + π_D > π_A). Distribution of profits becomes more unequal under trade.

### Political economy framework and equilibrium determination
- Political weight and pivotal voter:
  - Political weight: wλ(w) = λ0 + λ1 w with λ0, λ1 > 0.
  - Pivotal wealth w_p solves the weighted-median condition in equation (15); pivotal wealth increases in λ1 and decreases in λ0; w_p ≥ median wealth for λ1, λ0 > 0.
- Preference Curve properties:
  - r.fp is piecewise continuously differentiable, nonincreasing, and strictly decreasing for some ranges (Proposition 2).
  - For wealth = real profits (expressions (16) autarky and (17) trade), r.waf satisfies regularity conditions and yields downward sloping Preference Curves (Corollary 3).
- Political Curve:
  - r.pf defined implicitly by equation (19) mapping f to the pivotal voter's marginal cost a_p; Political Curve is well-defined and downward sloping almost everywhere (Proposition 4).
- Equilibrium definition and existence:
  - Equilibrium is fixed point (f, a_p) s.t. a_p is the pivotal firm given f and f is the preferred f of firm a_p.
  - Existence: at least one equilibrium exists (Proposition 6) and at least one stable equilibrium exists (Proposition 8).
  - Stability defined via iterative mappings Φ and Π (equations (20)–(21)) and formal condition (22)–(23).

### How trade changes political economy: political power vs foreign competition effects
- Political power effect (Proposition 10):
  - Opening to trade shifts the Political Curve inward: the pivotal voter moves toward more productive firms (TA p_f ≤ A p_f for all f in [L,H]); political power concentrates with larger exporters under trade.
- Foreign competition (disciplining) effect:
  - For most parameter values and for sufficiently high a, firms prefer better institutions under trade than autarky because domestic profits fall under trade; Preference Curve shifts inward.
  - Exception: most productive exporters may prefer worse institutions under trade because export profits can increase with f.
- Comparative outcome (Proposition 12):
  - If the political power effect dominates the foreign competition effect, trade can worsen institutions: existence of trade equilibrium T_f with f_T < f_A and autarky equilibrium f_A lies in its basin of attraction (formal condition involving integral Δ defined in text).
  - The sign of Δ captures whether political power shift outweighs preference shifts; Δ>0 implies institutions likely worsen.

### Numerical exploration and parameter regions (exact reported parameter values and qualitative findings)
- Reported parameter values and expressions (preserved exactly as presented):
  - β = 0.5; ε = 3; k = 0.1; b = 1000; S_L = 20; S_n = ... = 181; f_L = 0; f_H = 1.1; τ = 150; X_f = 0; λ0 = 1; λ1 = 0.875; λ = 48.
  - Text preserves ambiguity: "S_n = ... = 181; H_f = 1.1; τ = 150; X_f = 0; λ0 = 1; λ1 = 0.875; λ = 48. N_f= Details..."
- Numerical findings summarized (as described in text):
  - Institutions most likely to deteriorate post-opening when:
    - The economy is small in labor endowment relative to partner (N_L large relative to S_L in the grid sense used), and
    - The country captures a large share of world trade in the differentiated good (S_n large relative to N_n).
  - Intuition: small country with large export share ⇒ weak foreign competition effect and strong political power effect from large exporters.
  - When domestic firms are a small fraction of world producers (S_n small relative to N_n) or when the country is relatively large, the foreign competition (disciplining) effect tends to dominate and institutions improve or remain good.

### Historical case studies (illustrative evidence aligned with model)
- Caribbean sugar (1650–1850):
  - Rapid shift from smallholder to plantation economies; land concentration and political power concentrated with planters.
  - Caribbean produced between 80 and 90 percent of sugar consumed by Western Europeans in the 18th century.
  - Institutional outcomes: extreme slavery proportions (almost 9 out of 10 inhabitants slaves at sugar era height) and deliberate insecurity of peasant property rights to suppress alternatives to plantation labor.
- Coffee in Latin America (1850–1920):
  - Coffee exports: 320 metric tons in 1770; 90,000 in 1870; 1.6 million in 1920.
  - Puerto Rico land price: 3.41 pesos per cuerda in 1863 to 28.14 in 1877.
  - Share with land access fell from ~50% to 17% between 1850s and 1870s in a sample.
  - Political consequences: planters/processing elites consolidated political control, used land expropriation and policy measures to secure cheap labor and limit competition.
- Cotton and beef in Central America (1950–1980):
  - Cotton regional production: ~25,000 bales (1940s) → >1 million bales by mid-1960s; late 1970s Central America ranked third in world cotton output.
  - Guatemala example: farms <122 acres <2% of cotton land; farms >1100 acres produced 62%.
  - Beef exports: 30 million pounds in 1960 → 250 million pounds in 1978; >90% of exports to protected U.S. market.
  - Institutional consequences: large producers expanded by acquiring titled and untitled lands, evicting smallholders and weakening communal/ejidal rights; dispossession contributed to later social unrest.

### Pareto distribution assumption and closed-form solutions (appendix summary)
- Pareto assumption for marginal cost a:
  - Distribution expressed as Ga(a) with parameters b and k; parameter 0b> is minimum; k regulates dispersion.
  - Useful integral: V_y(a) ≡ ∫_0^y a^{1-ε} dGa.
  - Closed-form for Va: (1) (),( .2) (1) k k bk VaaA k ε ε −− = −− (as stated in source).
  - Regularity condition: 1.kε>− (when not satisfied total profits are infinite).
- Autarky closed-form:
  - Closed-form expression for cutoff aA given in equation (.3) (text exact expression preserved).
  - Aggregate price proportionality given in (.4).
- Trade closed-form (Southern cutoffs and aggregate price):
  - Closed-form expressions for cutoffs a_D and a_X given in source (text preserves exact compact forms).
  - Aggregate price proportionality compact form given in (.5).
  - A, B, C, D, F are positive constants (explicit expressions available upon request per source).
- Comparative statics implied by closed forms:
  - 0 S D S da df < 
  - 0 S X S da df > .

### Regularity conditions, mechanisms, and real-profit verification (appendix technical conditions)
- Regularity on real-wealth and profit functions:
  - r.waf(), r.waf() piecewise continuously differentiable; conditions (A.6)–(A.11) specify monotonicity, differentiability, boundary conditions, and parabolic approximation near zero (text preserves exact conditions).
- Mechanisms from autarky to trade:
  - Two driving effects identified:
    - (1) The most productive entrepreneur (0a→) is wealthier under trade than in autarky.
    - (2) Domestic producers experience a drop in profits as a consequence of trade (Melitz, 2003 effect).
  - Condition (A.12) corresponds to (1); condition (A.13) corresponds to (2) (complex inequalities preserved in source).
- Real profit functions — sufficiency and verification:
  - Autarky downward-sloping Preference Curve condition: 1 .(   .16) 12 A β ε < − .
  - Explicit autarky profit expression (A.3) preserved verbatim in source.
  - Trade profits expressions for non-exporting and exporting firms given in appendix (text preserves exact forms).
  - Sufficient parameter choice: making 1 β ε − small enough ensures admissibility conditions (A.6)–(A.11) for real profits under both regimes.

### Proofs and formal results (summary of main proof conclusions)
- Lemma 1: monotonicity properties of pivotal voter p(wλ) with respect to λ0, λ1; mp01, =wwλλ for any 0 0.λ> .
- Proposition 2: characterization of f via first-order conditions; r.fp nonincreasing and, under interior solutions, strictly decreasing on some range.
- Proposition 4: Political Curve r.pf existence, uniqueness, and differentiability via equation (19).
- Proposition 6: existence of intersection between Political and Preference curves (equilibrium existence).
- Proposition 8: existence and local stability of at least one equilibrium; no cycles and corner-solution stability arguments.
- Lemma 13 (used in proofs): properties of pivotal marginal cost with respect to λ0, λ1; integrability condition 1 .kλγ< for finiteness of vote aggregates.
- Proposition 10: opening to trade shifts the Political Curve inward (pivotal voter moves to more productive firms) when integral condition Δ>0 holds; parameter-region existence proven via scaling λ parameters and Pareto k threshold arguments.
- Proposition 12: multi-step dynamics and basin-of-attraction result showing when autarky equilibrium f_A lies in basin of attraction of a trade equilibrium with worse institutions (formal iterative argument using Φ and Π).

### Key conclusions
- Trade opening has ambiguous effects on institutional quality because of two countervailing channels:
  - Discipline via foreign competition (usually improves institutions).
  - Concentration of political power via Melitz-style firm reallocation (can worsen institutions if large exporters prefer worse institutions).
- Comparative statics: deterioration of institutions is more likely when a small country captures a large share of world trade in the rent-bearing good; improvement more likely when the country is relatively large or captures only a small share of world production.
- The model rationalizes historical episodes where trade coincided with institutional deterioration by concentrating political power among a small elite of exporters.

*Source: _wp0656 - 1.     Profits As a Function of Marginal Cost (PDF chapter/section), canonical URL: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf*

### 1.     Profits As a Function of Marginal Cost .....................................................................10

### 1.     Profits As a Function of Marginal Cost .....................................................................10

### Major sections and headings
- 1.     Profits As a Function of Marginal Cost .....................................................................10
- 2.     Profits and Firms’ Segmentation into Domestic and Exporting ................................11
- 3.     Profits As a Function of Marginal Cost for Two Different Values of f.....................18
- 4.     Profits As a Function of f...........................................................................................18
- 5.     Densities of Distributions of Profits and the Pivotal Firms .......................................21
- 6.     The Preference Curve, the Political Curve, and Possible Equilibria .........................21
- 7.     Comparing Institutions in Autarky and Trade ...........................................................25
- 8.     Ranges of Parameter Values Such That Institutions Deteriorate Under Trade .........27
- 9.     Institions Under Trade As a Function of Parameter Values ......................................27
- Appendix............................................................................................................................34

*Source: _wp0656 - 1.     Profits As a Function of Marginal Cost (PDF chapter/section), canonical URL: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf*

### References..............................................................................................................

### _wp0656 - References

### Introduction and research question
- Central question: How does opening to international trade affect a country’s institutions (contract enforcement, property rights, rule of law)?
- Core insight: Trade can have two opposing effects on institutions:
  - Foreign competition (disciplining) effect: trade tends to raise preferences for better institutions because foreign competition reduces domestic rents.
  - Political power effect: trade increases firm-size dispersion (Melitz effect), shifting political weight to large exporters who, in the model, prefer worse institutions to limit entry and competition.
- Key contextual examples noted: Caribbean sugar boom (18th century), coffee boom in Latin America (19th century), cotton and cattle booms in Central America (mid-20th century).

### Model overview and mechanisms
- Environment: Monopolistic competition with heterogeneous producers (Melitz-style). Institutions are summarized by a common fixed cost f (higher f = worse institutions).
- Role of heterogeneity:
  - Firms differ by marginal cost parameter a (productivity).
  - Only most productive firms export; trade increases inequality of firm size/profits: largest firms grow larger, smaller firms shrink or exit.
- Political power mapping:
  - Political weight of an agent wλ(w) = λ0 + λ1 w with λ0, λ1 > 0; wealthier agents obtain higher political weight.
  - Political decision modeled via a Benabou (2000) style weighted-median/pivotal voter; pivotal voter’s preferences determine equilibrium f.
- Preferences over f:
  - Each firm maximizes real wealth (real profits); more productive firms typically prefer higher f because higher f reduces entry and raises variable profits disproportionately for them.
  - Preference Curve r.fp gives, for each firm productivity a, the preferred level of f; it is nonincreasing and strictly decreasing for some values of a.

### Goods and factor market equilibrium (autarky and trade)
- Preferences: U = α ln z + β ∫v∈V ln xv dv, with ε = 1/(1-α) and A ≡ (∫ p(v)^{1-ε} dv)^{1/(1-ε)} defined as demand shift parameter.
- Firms:
  - Pay common fixed cost f to produce; marginal cost a (units of labor per output).
  - Profit (autarky): πα(a) = (1/(1-α)) A E a^{1-ε} − f, where expressions as in the text (equation (3)).
- Autarky equilibrium:
  - Zero-profit cutoff A_a defined by equation (5): (1/(1-α)) A E a_A^{1-ε} − f = 0.
  - Goods market clearing leads to equation (6): E L + A n [ (1/(1-α)) a_A^{1-ε} V_a − f ] = 0 (as in the text).
- Trade setup:
  - Two countries N and S with labor endowments N_L and S_L and masses of producers N_n and S_n.
  - Trade costs: per-unit iceberg τ and export fixed cost X_f. To export requires paying domestic fixed cost f and X_f.
  - Domestic and export profit expressions given by equations (8) and (9).
  - Cutoffs under trade: production cutoffs i_D a and export cutoffs i_X a characterized by equations (11)–(14).
- Comparative static features of trade vs autarky (as stated):
  - i) i_AD a_D ≥ a_A: higher productivity required to produce domestically under trade.
  - ii) For producers operating under trade, domestic profits π_Da < π_Aa.
  - iii) There exists a cutoff a_X below which a firm earns higher total profits under trade than autarky (π_X + π_D > π_A). Distribution of profits becomes more unequal under trade.

### Political economy framework and equilibrium determination
- Political weight and pivotal voter:
  - Political weight function wλ(w) = λ0 + λ1 w with λ0, λ1 > 0.
  - Pivotal wealth w_p solves the weighted-median condition in equation (15); pivotal wealth increases in λ1 and decreases in λ0; w_p ≥ median wealth for λ1, λ0 > 0.
- Preference Curve properties (Proposition 2 and Corollary 3):
  - Preference Curve r.fp is piecewise continuously differentiable, nonincreasing, and strictly decreasing for some ranges.
  - When wealth = real profits (expressions (16) autarky and (17) trade), r.waf satisfies regularity conditions and yields downward sloping Preference Curves.
- Political Curve:
  - Political Curve r.pf defined implicitly by equation (19) mapping f to the pivotal voter's marginal cost a_p; Political Curve is well-defined and downward sloping almost everywhere (Proposition 4).
- Equilibrium definition and existence:
  - Equilibrium is fixed point (f, a_p) s.t. a_p is the pivotal firm given f and f is the preferred f of firm a_p.
  - Existence: at least one equilibrium exists (Proposition 6) and at least one stable equilibrium exists (Proposition 8).
  - Stability defined via iterative mappings Φ and Π (equations (20)–(21)) and formal condition (22)–(23).

### How trade changes political economy: political power vs foreign competition effects
- Political power effect (Proposition 10):
  - Opening to trade shifts the Political Curve inward: the pivotal voter moves toward more productive firms (TA p_f ≤ A p_f for all f in [L,H]); i.e., political power concentrates with larger exporters under trade.
- Foreign competition effect:
  - For most parameter values and for sufficiently high a, firms prefer better institutions under trade than autarky because domestic profits fall under trade; Preference Curve shifts inward (disciplining effect).
  - Exception: most productive exporters may prefer worse institutions under trade because export profits can increase with f.
- Comparative outcome (Proposition 12):
  - If the political power effect dominates the foreign competition effect, trade can worsen institutions: specifically, if (TA p_f) < (A f^{-1}) in the sense of the inequality in Proposition 12, then there exists a trade equilibrium T_f with f_T < f_A and the autarky equilibrium f_A lies in its basin of attraction.
  - The sign of ∆ (difference integral defined in text) captures whether political power shift outweighs preference shifts; ∆>0 implies institutions likely worsen.

### Numerical exploration and parameter regions
- Numerical experiment grid: vary N_L and N_n relative to S_L and S_n to map where institutions improve vs deteriorate under trade.
- Parameter values used in numerical implementation (reported exactly):
  - β = 0.5; ε = 3; k = 0.1; b = 1000; S_L = 20; S_n = ((1/(1-α)) ((1/(1-α)) (k/(kL))?) — note: the text shows a complex inline expression; the explicit value S_n = 181 is reported.
  - f_L = 0; f_H = 1.1; τ = 150; X_f = 0; λ0 = 1; λ1 = 0.875; λ = 48; N_f unspecified here (N_f = ? in text left as N_f =).
  - The text states: "S_n = ... = 181; H_f = 1.1; τ = 150; X_f = 0; λ0 = 1; λ1 = 0.875; λ = 48. N_f= Details..." (preserve exactly as presented).
- Numerical findings summarized (figures discussed in text):
  - Institutions most likely to deteriorate post-opening when:
    - The economy is small in labor endowment relative to partner (N_L large relative to S_L in the grid sense used), and
    - The country captures a large share of world trade in the differentiated good (S_n large relative to N_n).
  - Intuition: small country with large export share ⇒ weak foreign competition effect and strong political power effect from large exporters.
  - When domestic firms are a small fraction of world producers (S_n small relative to N_n) or when the country is relatively large, the foreign competition (disciplining) effect tends to dominate and institutions improve or remain good.

### Historical case studies (illustrative evidence)
- Caribbean sugar (1650–1850):
  - Rapid shift from smallholder to plantation economies; land concentration and political power concentrated with planters.
  - Caribbean produced between 80 and 90 percent of sugar consumed by Western Europeans in the 18th century.
  - Political/institutional outcomes: extreme slavery proportions (almost 9 out of 10 inhabitants slaves at sugar era height) and deliberate insecurity of peasant property rights to suppress alternatives to plantation labor.
- Coffee in Latin America (1850–1920):
  - Coffee exports rose from 320 metric tons in 1770 to 90,000 in 1870 and 1.6 million in 1920.
  - Land price and ownership concentration: example Puerto Rico land price increase from 3.41 pesos per cuerda in 1863 to 28.14 in 1877; share with land access fell from ~50% to 17% between 1850s and 1870s in a sample.
  - Political consequences: established planters/processing elites consolidated political control, used land expropriation and other measures (tax on new planting in 1903, abolition of certain tenures) to secure cheap labor and limit competition.
- Cotton and beef in Central America (1950–1980):
  - Cotton: regional production rose from ~25,000 bales (1940s) to >1 million bales by mid-1960s; by late 1970s Central America ranked third in world cotton output.
  - Cotton industry highly concentrated: average plot sizes and heavy concentration (e.g., Guatemala: farms <122 acres <2% of cotton land; farms >1100 acres produced 62%).
  - Beef exports rose from 30 million pounds in 1960 to 250 million pounds in 1978; >90% of exports to protected U.S. market.
  - Institutional consequences: large producers expanded by acquiring titled and untitled lands (often by manipulating titling institutions), evicting smallholders and weakening communal/ejidal rights; dispossession contributed to social unrest in subsequent decades.

### Key conclusions
- Trade opening has ambiguous effects on institutional quality because of two countervailing channels:
  - Discipline via foreign competition (usually improves institutions).
  - Concentration of political power via Melitz-style firm reallocation (can worsen institutions if large exporters prefer worse institutions).
- Comparative statics: deterioration of institutions is more likely when a small country captures a large share of world trade in the rent-bearing good; improvement more likely when the country is relatively large or captures only a small share of world production.
- The model provides a framework able to rationalize historical episodes where trade coincided with institutional deterioration by concentrating political power among a small elite of exporters.

*Source: _wp0656 - References*

### 1. The Pareto Distribution and the Closed-Form Solutions

### 1. The Pareto Distribution and the Closed-Form Solutions to the autarky and Trade Equilibria

### Pareto distribution assumption and useful integrals
- The cumulative distribution function of a Pareto(  ,bk) random variable is given by the source expression (textual display omitted here).
- The parameter 0b> is the minimum value; k regulates dispersion.
- The paper assumes that 1/   ,a which is labor productivity, has the Pareto distribution.
- For marginal cost , ,a the cumulative distribution is ()  (  ),( .1) k GabaA= for 01/.ab<< .
- Define the integral Vya() ≡ ∫0 y ()() a dGa ε− . In the Dixit-Stiglitz CES framework, ()Vy is useful to write price indices and total profits when a ∼ Ga.
- Closed-form expression for Va given the Pareto Ga:
  - (1) (),( .2) (1) k k bk VaaA k ε ε −− = −−
  - Regularity condition imposed: 1.kε>− (when not satisfied total profits are infinite).

### A. Autarky closed-form solution — cutoff and aggregate price
- Closed-form expression for the cutoff aA:
  - A a () 1 1 1 (1) ((1) )1 (.3) 1(1   ) k k A k k k L aA nb  kff ε αβ ε αβ − − ⎛⎞ ⎛⎞ −−−Γ =≡ ⎜⎟ ⎜⎟ ⎜⎟ −− ⎝⎠ ⎝⎠
- Aggregate price is proportional to:
  - () () 1 1 (   .4) k k PfA ε β ε −− − ∝
- These expressions use (A.1) and (A.2) functional forms for Ga and Va.

### B. Trade closed-form solution — Southern cutoffs and aggregate price
- Using equations (11)-(14) and functional forms (A.1)-(A.2), closed-form solutions for Southern cutoffs:
  - 1 (1) 1 1 () k k S D S S a f f ε ε −− − ⎡⎤ = ⎢⎥ + ⎢⎥ ⎣⎦ A BC
  - 1 (1) 1 1 , () k k S X S X a f f ε ε −− − − ⎡⎤ ⎛⎞ ⎢⎥ =+ ⎜⎟ ⎜⎟ ⎢⎥ + ⎝⎠ ⎣⎦ D A F BC
- Aggregate price proportionality (compact form):
  - () () () 1 1 1 (1) 1 (1) 1 1 1 (1) 1 (.5) 1 () k k k k k k S S Sk DSN X S k SN S S X f PannA f f nn ff f β ε ε ε ε ε β ε ε ε ε ε ε β τ τ −− − − −− − −− − −− − −− − − − ⎡⎤ ⎛⎞ ⎢⎥ ∝+= ⎜⎟ ⎢⎥ ⎝⎠ ⎣⎦ ⎡⎤ ⎛⎞ ⎛⎞ ⎛⎞ ⎢⎥ ⎜⎟ =+ ⎜⎟ ⎜⎟ ⎢⎥ ⎜⎟ ⎜⎟ + ⎝⎠ ⎝⎠ ⎢⎥ ⎝⎠ ⎣⎦ A BC
- A, B, C, D, and F are positive constants (explicit expressions available upon request).
- Comparative statics implied:
  - 0 S D S da df < 
  - 0 S X S da df > .

### 2. Regularity conditions on admissible functions (real-wealth and profit functions)
- Admissibility and differentiability:
  - () , r waf () , r waf is piecewise continuously differentiable with respect to () , ; af
  - For some marginal entrepreneur 1 r b a≤ and any [] , , LH fff∈ :
    - () () ,0 if (.6) ,0 otherwise(   .7) rr r waf a aA a wafA a ∂ <≤ ∂ ∂ ≤ ∂
- Interpretation: wealth is everywhere weakly increasing in firm productivity and strictly increasing below marginal cost cutoff r a.
- Further smoothness and monotonicity:
  - () , r waf is twice piecewise continuously differentiable with respect to ; f uniformly continuous with respect to ; a and () () , is decreasing in (A.8) and , is decreasing in (A.9) r r waff f wafA f ∂ ∂ ∂ ∂
- Boundary conditions guaranteeing equilibrium existence:
  - ()() 1 0 lim,0 and lim,0(   .10) rr a a b wafwafA ff → → ∂∂ >< ∂∂
- Asymptotic/parabolic approximation near zero:
  - There exists 0, γ> and two continuously differentiable functions () () 12 ,0ffγγ> and () () 12 ,1(.11) r waf afofA γ γγ − =+ .
  - Implies wealth function approximable by a parabolic branch near 0.

### 3. Mechanisms: From autarky to trade
- Two driving effects:
  - (1) The most productive entrepreneur ( 0a→ ) is wealthier under trade than in autarky.
  - (2) Domestic producers experience a drop in profits as a consequence of trade (Melitz, 2003 effect).
- Condition (1) corresponds to:
  - () () () 1 11 1.(   .12) A DXX faf A faff af ε ε ε − − − < +
- Condition (2) holds when for any [] , : LH fff∈ the inequality (A.13) is satisfied (complex multi-term inequality involving k, ε, α, β, τ, n, f terms; see source).
- Additional model-interesting assumptions:
  - Lowest-productivity entrepreneur does not produce in autarky requires (A.14):
    - ()() () 1 11 .(   .14) 11 S L S k kL fA nk ε αβ ε αβ − − − ⎡⎤ ⎣⎦ > − − ⎡⎤ ⎣⎦
  - This condition is sufficient for the trade case when (A.13) holds.
- Pivotal voter uniqueness condition:
  - The pivotal voter is uniquely defined by marginal cost .a Necessary and sufficient condition: () ,0, rH a ap waf ∂ ∂ = < [] , . LH fff∀∈
  - Sufficient condition (median voter always produces): (1) 1 111 (.15) 2 k k H H A fb f ε ε − − − > + A BC, which suffices in autarky as (A.13) holds.

### 4. Real profit functions — conditions and verification
- Wealth convention: wealth defined by (16) and (17) in autarky and under trade respectively.
- Goal: determine sufficient conditions for real profits to satisfy (A.6) to (A.11).

A. Autarky
- Preference curve downward sloping as long as:
  - 1 .(   .16) 12 A β ε < −
- Expression for each firm’s profits (from eqs. (3), (5), (A.3), and price level (A.4)):
  - (1) 1 (1) 1 (, ) k kk k kk A af faf P f ε ε ε β ε π − − − − − Γ − =
- If (A.16) holds, then conditions (A.6) -- (A.11) are satisfied for autarky real profits.

B. Trade
- Profits expression for non-exporting firm:
  - 11 (, )(  ) ()() SSS TD SS aff aaf PfPf ε ε − − − =
- Profits expression for exporting firm:
  - 1111 (, )(  )(  ) ()() SSSS TDXXX SS affaaffaaf PfPf ε ε ε ε − − − − − + − =
- Verification summary:
  - (i) Conditions (A.6), (A.7), (A.10), and (A.11) straightforward to verify.
  - (ii) For (A.8): nominal profits from domestic production and exports are concave in f; by making exponent 1 β ε − small (i.e., making 1 0. β ε − → ), the derivative of price level wrt f can be made arbitrarily small, preserving concavity of real profits.
  - (iii) For (A.9): similar argument — responsiveness of S Pf to f can be made arbitrarily small as 1 β ε − decreases.
- Monotonicity of profit-maximizing fT(a):
  - Define arg max levels () ,()0 SS D SS D fa fa af π π ∗∗ > = and () ,() ()0 SSS D X SSS S DX faa faafaf ππ ππ ∗∗∗ + > =+ .
  - Gross profits concave in f so () fa ∗∗ and () fa ∗∗∗ are well-defined, positive, and nonincreasing in a.
  - Define trade-off φ a (difference in profits between domestic-only and domestic+exports for pivotal voter a): ()()()() SSS DDX aafa afa afa φ π π π ∗∗∗∗∗∗∗∗ ⎡⎤ = − + ⎣⎦ .
  - Pareto assumption implies φ(a) continuous and differentiable; envelope theorem yields φ′(a) > 0 when () fa ∗∗∗ < () fa ∗∗, so φ is continuous and increasing; monotonicity of T fa over a in some interval [1 0, b) holds or trivially otherwise.
- Conclusion: If 1 β ε − is small enough, real profit functions in autarky and under trade both satisfy (A.6) to (A.11).
  - Intuition: nominal profits satisfy (A.6) to (A.11); to transfer to real profits, aggregate prices must not be “too concave.” There are enough parameter degrees of freedom to make 1 β ε − sufficiently small while retaining concavity of nominal profits.

- Computational note: detailed compatibility checks of conditions (concavity, (A.12) to (A.15)) are involved; authors note sufficient degrees of freedom in parameter values for all restrictions to hold simultaneously. A comprehensive derivation of sufficient conditions is available upon request.

*Source: _wp0656 - 1. The Pareto Distribution and the Closed-Form Solutions*

### 5.         Proofs

### _wp0656 - 5.         Proofs

### Proof of Lemma 1 (pivotal voter and median properties)
- Defines the pivotal voter p(wλ) on (0,1) via double integrals over w and w′; compares values for λ11′> and λ00′> to establish monotonicity properties of p(wλ).
- Shows:
  - For λ11′>: integrals imply p01,11,01,0,p p w w wdFwwdFw λλ λλ λλ ′′ +∞ < ∫∫ and p01,11,01,00,0,p p w w wdFwwdFw λλ λλ λλ ′′ +∞ +=+ ∫∫, leading to p01,01,,.ppm wwwλλ ′ >≥.
  - For λ00′>: analogous integral decompositions yield p01,11,01,00,0,p p w w wdFwwdFw λλ λλ λλ ′′ +∞ ′′ +≥+ ∫∫ and p0101,,.pp wwλλ ′ ≤.
  - Observes mp01, =wwλλ for any 0 0.λ> .
  - Establishes (A.17): ( ) ( ) ( ) ( ) ( ) ( ) 0101 11 01 ,, ,0, 0 1 2.(.17) pp mp ww ww dF   ww  dF   ww  dF   wA λλλλ λλ λλ +∞ ⎡⎤ =− ⎢⎥ ⎣⎦ ∫∫∫, with right-hand side converging to zero as 0 λ grows large, hence lim 0 01 , .pm ww λ λλ →∞ = 0.

### Proof of Proposition 2 (characterization of f via first-order conditions)
- Interior solution f characterized by ( ),0, r f wpf ∂ ∂ =.
- When boundary, (A.8) implies:
  - rL fpf = if and only if ( ),0 rL f wpf ∂ ∂ ≥.
  - rH fpf = if and only if ( ),0 rH f wpf ∂ ∂ ≤.
- Defines:
  - ()() {} 1 1 sup0, ,,,0 rHr H bf ffpwpf − ∂ ∂ =∈≤ ⎡⎤ ⎣⎦
  - ()() {} 1 1 inf0, ,,,0 . rLr L bf ffpwpf − ∂ ∂ =∈≥ ⎡⎤ ⎣⎦
- Conditions (A.10) imply () () 11 . rH rL ffff −− <.
- Regularity: r fp is piecewise continuously differentiable in p; Preference Curve constant over ( ) 1 0, rH ff − ⎡⎤ ⎣⎦ and ( ) 1 1 , . rL b ff − ⎡⎤ ⎣⎦ .
- r fp is nonincreasing; when differentiable, r fp ′ ≤ 0.
- For small enough p_a, solution interior → Preference Curve strictly decreasing for some p range.

### Proof of Corollary 3
- Referenced: conditions on parameter values derived in section A.4.

### Proof of Proposition 4 (existence and differentiability of r p f)
- Existence: as r waf is nondecreasing, left-hand side of (19) increasing and continuous in p → unique r p f satisfying (19).
- Verifies that r p f corresponds to pivotal voter with wealth p_w defined in (15) via change of variables and uniqueness: ()() , . rrp wp f f w =.
- Differentiability: differentiating (19) w.r.t. f yields well-defined expression because r waf is piecewise continuously differentiable → r p f continuously differentiable in f.

- Lemma 13 (used in proof):
  - For marginal cost of production of the pivotal voter 0 |, r p f λλ :
    - 01 |, r p f λλ is decreasing in 1 λ and increasing in 0 λ.
    - 01 |, rm pf p λλ ≥ for any 01 0,0λλ>≥ and [] , LH fff∈.
    - 0 01 lim |, rm pf p λλ →∞ = for any 01 0,0λλ>≥ and [] , LH fff∈.
  - To ensure finiteness of total number of votes when wa f satisfies (A.11): ( )() 1 1 / 0 0 , , b r waf dGa λ λ ⎡⎤ + < ∞ ∫⎣⎦ requires 1 /.kλγ<.
  - Proof relies on one-to-one decreasing correspondence between wealth levels and marginal costs and Pareto distribution parameter k: integral ∫0 1 a dGa λ γ +∞ − converges iff 1 .kλγ<.

- Implicit differentiation of (19) gives r p f ′ with sign determined by Δ:
  - Δ ≡ ()() ()() 1 11 0 ,, ,, r r pf rr b rr pf waf waf waf dGa waf dGa ff λλ ′′ ∂ ∂ − ∂ ∂ ∫∫.
- Two cases based on relative positions of r pf and r f − 1; derive sufficient integral inequalities (A.18) and equivalent (A.19).
- Under (A.11) there exist bounds u>0, v>0 for integrals of ln, r waf f ∂ ∂; combine to yield sufficient condition (A.19).
- If political weight function sufficiently “convex” and parameters shifted (01 0,0, λλ ′′ >>), there exists threshold 1 , k γ λ < such that for any 11 , λλ ′ > (A.19) holds; integral ∫0 1 r pf a dGa λ γ ′ − diverges as 1 λ γ ′ → k.
- Concludes existence of parameter set for which the Political Curve is downward sloping on compact [] , LH fff∈.

### Proof of Proposition 6 (existence of intersection between Political and Preference curves)
- If ()() 1 , rHr H pfff − ≤ then ()(), HrH fpf is intersection; symmetrical statement for lower bound.
- Otherwise, continuity implies existence of () , LH fff∈ with ()() 1 , rr pfff − =, hence curves intersect at () , r f p f.

### Proof of Proposition 8 (existence and stability of equilibrium)
- Existence: shown.
- Lemma (no cycling): r Φ and r Π increasing; for any [] , LH fff∈ and any ( ) 1 0 , , b p ∈ sequences { }1 n r n f ≥ Φ and { }1 n r n f ≥ Π monotonic → no cycles.
- Lemma (corner solutions): intersection at boundary L f or H f is stable by monotonic convergence arguments; define ρ and show convergence to corner after one iteration.
- For interior equilibrium: absence of cycling and corner solutions implies intersection with Political Curve downward sloping; compare slopes to establish local stability. Uses limits of left and right derivatives in presence of kinks.

### Proof of Corollary 9
- See discussion on real profit functions in preceding section.

### Proof of Proposition 10 (pivotal voter shifts under trade)
- Defines difference Δ between aggregates under trade and autarky:
  - Δ ≡ ()() ()() ()() 1 0 ,,,, , A A pf b TTTT pf w afw afdGa w afw afdGa λλ ⎡⎤ ⎡⎤ − − − ⎣⎦ ⎣⎦ ∫∫.
- Pivotal voter shifts left ( ( ) ( ) TA pf pf < ) iff 0 . Δ >.
- Considers entrepreneur a_af with equal autarky and trade profits; two cases:
  - (i) A af p f < : rewrite Δ into three integrals; last term positive. Sufficient condition for 0 Δ > is an integral inequality comparing trade and autarky weightings.
  - (ii) A af p f > : analogous decomposition yields a sufficient integral condition for 0 Δ >.
- Defines ()() {} min, ( ) , AA pf pf af = % to get a unique sufficient condition:
  - ()() ()() ()() 0 ,,, , A A pfaf TTTT pf w afw afdGa w afw afdGa λλ ⎡⎤ ⎡⎤ − > − ⎣⎦ ⎣⎦ ∫∫ % .
- For symmetric political weights 1 0 .ww λ λ = + , 0 λ‘s cancel → ( )() ()() 1111 () 0 ,,, , A A pfaf TATA pf w af w af dGa w af w af dGa λλλλ ⎡⎤ ⎡⎤ − > − ⎣⎦ ⎣⎦ ∫∫% , rewritten as ( .20):
  - ()() ()() () () () () 1 111 () 0 , ,1,,( .20) , A A pfaf A TTA pf T waf wafdGa wafdGa A waf λ λ λ λ ⎡⎤ ⎛⎞ ⎢⎥ ⎡⎤ − > − ⎜⎟ ⎣⎦ ⎜⎟ ⎢⎥ ⎝⎠ ⎣⎦ ∫∫ % .
- If lim 0 , A T waf waf a → < then bracketed term on left does not go to zero as 0 . a → Using expansions 11 (), AA af a a f ε ε π − − = − , 11 (), DD af a a f ε ε π − − = − , 11 (). XXX X af a a f ε ε π − − = −; condition satisfied when triple-limit inequality (assumed in (A.12)) holds.
- As in Proposition 4, shifting political weight parameters (11 , λλ ′ >) and using (A.11) and Lemma 13, there exists threshold 1 , k γ λ < such that for 11 , λλ ′ > and suitable 01 0 λ ′ > and 001 , λλ λ ′′ >, (A.20) (and (A.21)) hold. Concludes existence of parameter set for which Pivotal Voter curves move inward as consequence of trade; uses compactness of [] , LH fff∈.

### Proof of Proposition 12 (multi-step dynamics and basin of attraction)
- For any entry barrier level f: ()() 1 , nn rr fp ff + ⎡⎤ Π = Φ ⎣⎦ and ()() 1 , nn rr pf pp + ⎡⎤ Φ = Π ⎣⎦; by continuity requirements (22) and (23) redundant; focus on (22).
- If ( )1 TA ff − is pivotal voter preferring A f under trade and ()() 1 , TA TA ffpf − > then ()() 1 TT AT TA ffpf − ⎡⎤ Π > Π ⎡⎤ ⎣⎦ ⎣⎦ and hence ()() . TAT TA pf pf > Π ⎡⎤ ⎣⎦ .
- Iterating () . T Π : for any n > 1, ()()()() 2 ... n TAT TAT TAT TA pf pf pf pf > Π > Π ≥ Π ⎡⎤ ⎡ ⎤ ⎡ ⎤ ⎣⎦ ⎣ ⎦ ⎣ ⎦.
- Taking limits defines lim n → ∞ n TnTTA p p f = Π ⎡⎤ ⎣⎦ and lim n → ∞ n TnA f f = Φ ; A f belongs to basin of attraction of () , TT fp and inequality implies A T f f < .

*Source: _wp0656 - 5.         Proofs (IMF working paper PDF).*

---


_Source: https://www.imf.org/-/media/websites/imf/imported-full-text-pdf/external/pubs/ft/wp/2006/_wp0656.pdf_
