Firm heterogeneity and trade - Altervista · Comparison of models: Assumptions Ricardo HOS Krugman Production factors 1 2 - Within-country mobility of inputs Yes Yes Yes Between-country

Firm heterogeneity and trade

Giovanni Marin

Department of Economics, Society, Politics

Università degli Studi di Urbino ‘Carlo Bo’

References for this lecture

• BBGV Chapter 4

– Paragraphs 4.7, 4.8

Spring 2017 Global Political Economy 2

Comparison of models: Assumptions

Ricardo HOS Krugman

Production factors 1 2 -

Within-country mobility of inputs

Yes Yes Yes

Between-country mobility of inputs

No No No

Technology / productivity Heterogeneous Homogeneous Homogeneous

Relative factors’ endowment - Heterogeneous -

Trade frictions No No No

Returns to scale Constant Constant Increasing

Commodities Homogeneous Homogeneous Heterogeneous

(varieties)


Comparison of models Predictions of the model

Ricardo HOS Krugman

Inter-industry trade Yes Yes -

Intra-industry trade No No Yes

Full specialization Yes Not necessarily Yes (in varieties)

Commodity price equalization Yes Yes -

Factor(s) price equalization No Yes -

Trade is mutually beneficial Yes Yes Yes


Trade and firm dynamics

• The model of trade with monopolistic competition predicts a ‘competition effect’

• Some firms exit the market and the remaining firms gain market shares

• The basic model, however, is based on the assumption of identical firms (except for the variety they produce)


Firm heterogeneity in labour productivity (in log)


0.2

.4.6

.8

De

nsity

8 10 12 14log(Value added per employee)

Firms in the chemical sector in Italy Year 2011 Source: AIDA


• Only few firms import or export

• Also within sectors in which a country has comparative advantage (or is well endowed with the intensive factor), only a selection of firms actually exports


Figure 4.8 Export orientation of US manufacturing firms, 2002

Export orientation of US manufacturing firms, 2002

0 5 10 15 20 25 30 35 40

Miscellaneous Manufacturing

Printing and Related Support

Furniture and Related Product

Apparel Manufacturing

Wood Product Manufacturing

Nonmetallic Mineral Product

Food Manufacturing

Textile Product Mills

Fabricated Metal Product

Petroleum and Coal Products

Beverage and Tobacco Product

Leather and Allied Product

Paper Manufacturing

Textile Mills

Plastics and Rubber Products

Transportation Equipment

Primary Metal Manufacturing

Machinery Manufacturing

Chemical Manufacturing

Computer and Electronic Product

Electrical Equipment Appliance

percent of firms that export

Average for US

manufacturing

Source: van Marrewijk (2012), based on Bernard et al. (2007, Table 2).


• As only few firms actually trade, it is important to understand which are the characteristics of these firms

• Are these firms larger? Are they more ‘efficient’? Are they more productive? Are they more technologically-endowed?


Firm heterogeneity (Italy)

Exporter premia (%)

Size (employees) 69.5 Labour productivity (VA per employee) 11.4 Capital stock per employee 18.9 Share of graduates 23.5 Probability of doing R&D 17 Probability of adopting a product innovation 14.4 Probability of adopting a process innovation 9.7 Probability of applying for a patent 4.2 Probability of doing FDI 1.9


Italian manufacturing firms. Sector and year dummies included. Source: Mediocredito Centrale, years 1995-2007

Firm heterogeneity (US)


Table 4.3 Exporter premia in US manufacturing, 2002

Exporter premia (%)

Employment 164

Shipments 194

Value-added per worker 12

TFP – total factor productivity 3

Wage 6

Capital per worker 13

Skill per worker 12

Additional covariates Industry fixed effects

Source: based on Bernard et al. (2007, Table 3); all results are significant at the 1 percent level.

Firm heterogeneity

• Exporting firms are:

– Larger

– More capital intensive

– With a more skilled labour force

– More productive

– More innovative


Figure 4.9 Distribution by number of products and export destinations; USA, 2000

1 2 3 4 5+

12

34

5+

11.9

40.4

0

20

40

60

80

100

Sh

are

of e

xp

ort

ing

fir

ms

# countries

# products

a. Share of exporting firms

1 2 3 4 5+

12

34

5+

92.2

0

20

40

60

80

100

Sh

are

of e

xp

ort

va

lue

# countries

# products

b. Share of export value

Source: van Marrewijk (2012), based on data from Bernard et al. (2007, Table 4).

Heterogeneity between exporting firms

• Almost half (40.4) of US exporting firms just export one product to just one country

• Firms that export five or more products to five or more countries account for 92.2 percent of total export


Figure 4.10 Simultaneous exporting and importing; US manufacturing, 1997

US manufacturing; exporting and importing per sector

0

10

20

30

40

50

60

70

0 10 20 30 40 50 60 70

percent of firms that export

perc

ent

of

firm

s t

hat

import

Aggregate Manufacturing average

diagonal

regression line

Computer and

Electronic Product

Printing and Related Support

Leather and

Allied Product

Source: van Marrewijk (2012, based on data from Bernard et al. (2007, Table 7).

Import and export

• There exist a positive correlation between exporting and importing

• 41 percent of the exporting firms also import

• 79 percent of importers also export


Trade and selection

• For a given demand and market structure, firms with higher marginal costs make less profits than firms with lower marginal costs

• If profits are below zero (due to too high marginal costs), the firm exits the market


q3 profit

demand

mr

c1

c2

c3

p1

p2

q2 q1 quantity

price, mc, mr mc

p3 (p3-c3)q3

(p2-c2)q2

(p1-c1)q1

(p1-c1)q1

(p2-c2)q2

E1

E2

E3

Figure 4.11 Firm heterogeneity, prices, and profits

profit

demand before trade

quantity

price mc

firms exit

firms make lower profits

firms make higher profits

demand after trade

c3

c4

c5

profit before trade

profit after trade

Figure 4.12 Firm heterogeneity and trade

Trade and selection

• Trade induce a counterclock wise rotation of the demand curve demand becomes more elastic

• As we will see in the coming lectures, firms that engage in trade need to incur in a series of fixed costs of exporting only firms with a sufficiently high productivity can bear the fixed cost of exporting – Least productive firms will exit the market – Firms with intermediate productivity will remain on the market

but cannot bear the fixed cost of exporting – Firms with high productivity will increase their market share

and also export

• Doing FDI is even more ‘costly’ further selection


Trade and selection

• Trade induce a counterclock wise rotation of the demand curve demand becomes more elastic

• As we will see in the coming classes, firms that engage in trade need to incur in a series of fixed costs of exporting only firms with a sufficiently high productivity can bear the fixed cost of exporting – Least productive firms will exit the market – Firms with intermediate productivity remain on the market but

cannot bear the fixed cost of exporting – Firms with high productivity increase their market share and

also export

• Doing FDI is even more ‘costly’ further selection


Exporter premia (%)

FDI premia (%)

Size (employees) 69.5 164.0 Labour productivity (VA per employee) 11.4 11.1 Capital stock per employee 18.9 22.4 Share of graduates 23.5 23.4 Probability of doing R&D 17 29.5 Probability of adopting a product innovation 14.4 24.3 Probability of adopting a process innovation 9.7 16.3 Probability of applying for a patent 4.2 11.4 Probability of doing FDI 1.9 - Probability of exporting - 20.7


0.2

.4.6

.81

-2 -1 0 1 2

No FDI and no export Export

FDI

Italian manufacturing firms. Labour productivity partialled out of sector and year dummies. Source: Mediocredito Centrale, years 1995-2007

Documents

Firm heterogeneity and trade - Altervista · Comparison of models: Assumptions Ricardo HOS Krugman Production factors 1 2 - Within-country mobility of inputs Yes Yes Yes Between-country