I.

INTRODUCTION

This paper provides a formal proof of the Factor Price Equalization Theorem within
the Heckscher Ohlin model derived by Ronald W. Jones in “The Structure of
Simple General Equilibrium Models” (1965), where formal proof is provided for the
Heckscher Ohlin, Stolper Samuelson and Rybczynski Theorems.

The basic structure of the model is:

a. Two countries: (A and B).

b. Two goods: Manufactures and food.
c. Two factors of production: Labor and land.

The assumptions of the model are:

a. Technologies of production are homogeneous of degree one, (i.e., they are

homothetic and have constant returns to scale) and are identical in both
countries.

b. There are different factor intensities in the two sectors (and there are no
factor intensity reversals for all ranges of factor prices). Manufactures is
labor intensive while food is land intensive.

c. Consumer preferences are identical in both countries.

d. All markets are competitive, there are no transportation costs within and
between countries and there is no government (and no trade policy).

3
e. Both countries are identical in everything except in their factor endowments,
which are different but sufficiently similar in order to avoid complete
specialization in production (factors lie within the diversification cone).

Notation can be expressed as follows:

L = Labor endowment.
T = Land endowment.
M = Manufactures sub-index.
F = Food sub-index.
Qj = Output of sector j, where j = M or F.
Lj = Labor used in sector j, where j = M or F.
Tj = Land used in sector j, where j = M or F.
w = Wages.
r = Rent.
PM = Unitary price of manufactures under free trade.
PF = Unitary price of food under free trade.
aij = Amount of factor i required to produce one unit of good j.
α = Positive technological parameter.
β = Positive technological parameter.

II. A FORMAL PROOF OF THE FACTOR PRICE EQUALIZATION THEOREM

The Factor Price Equalization Theorem states that under the assumptions, free
trade will make factor prices equal in the participating countries.

Since the unitary factor requirements are endogenous (they depend on factor
prices), it is important to provide production functions with functional forms that
comply with the basic Heckscher Ohlin assumptions, in order to prove the Factor
Price Equalization Theorem.

4
As noted earlier, technologies are identical in all countries and are assumed to be
homothetic and to have constant returns to scale. They are also different between
sectors (manufactures is labor intensive and food is land intensive). Thus, the
following Cobb Douglas production functions were assumed in this paper:

QM = (LM ) (TM )
α (1−α )

QF = (LF ) (TF )
β (1− β )

No factor intensity reversals imply that for any factor prices:

LM L F
>
TM TF

Homothetic production functions imply that given any factor prices:

LM a LM
=
TM aTM

LF a LF
=
TF aTF

The unitary production functions can be represented by:

1 = (a LM ) (aTM )
α (1−α )
(1)

1 = (a LF ) (aTF )
β (1− β )
(2)

5
and thus, for any factor prices2:

a LM a LF
>
aTM aTF

The zero profit conditions are:

a LM w + aTM r = C M = PM (3)

a LF w + aTF r = C F = PF (4)

Thus, unitary profits are:

Π M = PM (a LM ) (aTM ) − (a LM )w − (aTM )r
α (1−α )
(5)

Π F = PF (a LF ) (aTF ) − (a LF )w − (aTF )r
β (1− β )
(6)

Equations (1) to (6) give us a system of 6 equations and the following 6

endogenous variables:

w, r, aLM, aLF, aTM, aTF.

This system has four exogenous variables:

α, β, PM and PF.

α and β are constant technological parameters that are identical in both countries
by assumption, while PM and PF are also identical in the two countries under free

2 The appendix of this paper shows that for this condition to be fulfilled, all that is required is that
α > β.

6
trade because of competitive markets and perfect arbitrage, given that there are
no transport costs and no trade barriers.

In what follows, all 6 endogenous variables will be expressed in terms of the

exogenous variables.

Differentiating (5) for profit maximization in manufactures:

∂Π M
= αPM (a LM ) (aTM )
α −1 (1−α )
−w
∂a LM

∂Π M
= (1 − α )PM (a LM ) (aTM ) − r
α ( −α )
∂aTM

First order conditions state that:

αPM (a LM ) (aTM )(1−α ) = w

α −1

(1 − α )PM (a LM )α (aTM )(−α ) = r

Dividing the last two equations:

αPM (a LM )α −1 (aTM )(1−α ) w

( −α ) =
(1 − α )PM (a LM ) (aTM )
α
r

α (aTM ) w
= (7)
(1 − α )(a LM ) r

Differentiating (6) for profit maximization in food:

7
 ΠF
= βPF (a LF ) (aTF )(1− β ) − w
( β −1)
∂a LF

ΠF
= (1 − β )PF (a LF ) (aTF )
β (− β )
−r
∂aTF

First order conditions state that:

βPF (a LF ) (aTF )(1− β ) = w

( β −1)

(1 − β )PF (a LF )β (aTF )(− β ) = r

Dividing the last two equations:

βPF (a LF )( β −1) (aTF )(1− β ) w

(− β ) =
(1 − β )PF (a LF ) (aTF )
β
r

β (aTF ) w
= (8)
(1 − β )(a LF ) r

From (7), wages can be expressed as:

α (aTM )r
w= (9)
(1 − α )(a LM )

Replacing wages into (3):

 α (aTM )r 
a LM   + aTM r = PM
 (1 − α )(a )
LM 

8
  α (aTM )r 
  + aTM r = PM
 (1 − α ) 

 α  
r (aTM )  + 1 = PM
 (1 − α )  

 α + (1 − α ) 
r (aTM )  = PM
 (1 − α ) 

 (a ) 
r  TM  = PM
 (1 − α ) 

PM (1 − α )
r= (10)
(aTM )

Replacing (10) into (9):

α (aTM ) PM (1 − α )
w=
(1 − α )(a LM ) (aTM )

αPM
w= (11)
(a LM )

From (8):

(aTF ) = w(1 − β )(a LF ) (12)

rβ

Replacing (12) into (4):

9
  w(1 − β )(a LF )r 
a LF w +   = PF
 rβ 

 (1 − β ) 
w(a LF )1 + = PF
 β 

 β + (1 − β ) 
w(a LF )  = PF
 β 

w
a LF   = PF
β 

P β 
a LF =  F  (13)
 w 

Replacing (13) into (12):

(aTF ) = w(1 − β )PF β

rwβ

(aTF ) = (1 − β )PF (14)

r

Replacing (11) into (13):

 P β (a LM ) 
a LF =  F  (15)
 αPM 

Replacing (10) into (14):

10
  (1 − β )PF (aTM ) 
(aTF ) =   (16)
 PM (1 − α ) 

From (1):

α
 α 
  1−α
(aTM ) =  1 = (a LM )
−  1−α 
 


(17)
 a LM 

Replacing (15) and (16) into (2):

(1− β )
 P β (a LM )   (1 − β )PF (aTM ) 
β

1=  F    (18)
 αPM   PM (1 − α ) 

Replacing (17) in (18):

(1− β )
   α  

( )  ( )    
β  1 − β PF  a LM
−  1−α 

 P β (a LM )    
1=  F   
 αPM   PM (1 − α ) 
 
 

(1− β )
 β   (1 − β ) 
β
 α (1− β ) 
1 = (a LM ) (a LM )
β −  PF
 1−α 
 α   (1 − α ) 
PM    

(1− β )
PF  β   (1 − β ) 
β
β (1−α )−α (1− β )
1 = (a LM ) 1−α
PM  α   (1 − α ) 

(1− β )
PF  β   (1 − β ) 
β
β −αβ −α +αβ
1 = (a LM ) 1−α
PM  α   (1 − α ) 

11
 (1− β )
 β   (1 − β ) 
β
β −α
1 = (a LM )
PF
1−α
 α   (1 − α ) 
PM    

β −α
(a LM ) 1−α =
1
(1− β )
 β   (1 − β ) 
β
PF
 α   (1 − α ) 
PM  

1−α
  β −α
 
 1 
a LM = ( )  (19)
 PF  β   (1 − β ) 
β 1− β

 P  α   (1 − α )  
 M     

Defining:

 1−α   β (1−α )   (1−α )(1− β ) 

 (1 − β ) 
−  −   −  
P   β −α  β   β −α   β −α 
a LM =  F  α   (1 − α )  =µ (20)
 PM     

and replacing it into (11), (15) and (17):

αPM
w= (21)
µ

 P βµ 
a LF =  F  (22)
 αPM 

(aTM ) =  1 

1−α
(23)
µ

Defining:

12
  α   βα   α (1− β ) 

 β   β −α   (1 − β )  
     β −α 
P   β −α  
aTM =  F  α   (1 − α )  =η (24)
 PM     

and replacing it into (10) and (16):

PM (1 − α )
r= (25)
η

(aTF ) =  (1 − β )PF (η )
 
(26)
 PM (1 − α ) 

In the system defined by equations (1) to (6), all six endogenous variables have
now been expressed in terms of the exogenous variables. Factor prices and
unitary factor requirements are identical between countries since they depend on
parameters that are identical in both countries. Goods prices are equal because
of free trade, competitive markets and perfect arbitrage due to the lack of
transportation costs and trade barriers, while α and β are identical because of
equal technologies. Thus, wages and rent will be the same in both A and B as
shown in equations (21) and (25). QED.

III. SOME FURTHER RESULTS

In what follows, some of the results obtained so far are used to find expressions
for QM, QF, LM, LF, TM and TF, based on the exogenous parameters.

The use of factors in each sector is defined as:

QM a LM = LM (27)
QF a LF = LF (28)

13
 QM aTM = TM (29)

QF aTF = TF (30)

Competitive markets and full employment of factors imply that:

QM a LM + QF a LF = L (31)

QM aTM + QF aTF = T (32)

Equations (27) to (32) give us again a system of 6 equations and the following 6
endogenous variables:

LM, LF, TM, TF, M and F.

In this case, there are six exogenous variables which are:

L , T , PM, PF. α and β

where L and T are the factor endowments.

Solving for QM and QF in (31) and (32) by Cramer´s Rule:

L aTF − T a LF
QM = (35)
a LM aTF − aTM a LF

T aTM − L a LM
QF = (36)
a LM aTF − aTM a LF

14
Replacing (20), (22), (24) and (26) into (35) and (36):

 (1 − β )P F (η )   P F βµ 
L  −T  
QM =  P M (1 − α )   α PM 
 (1 − β )P F (η )   P F βµ 
µ  −η 
 P M (1 − α )   α PM 

 (1 − β )(η )   βµ 
L  −T  
(1 − α )   α 
QM =  (37)
 (1 − β )(η )   βµ 
µ  −η 
 (1 − α )  α 

Tη − Lµ
QF = (38)
 (1 − β )PF (η )   PF βµ 
µ  −η 
 PM (1 − α )   αPM 

LM is defined as QMaLM. Thus:

  (1 − β )(η )   βµ  
L  −T  
(1 − α )   α  
LM =   µ (39)
  (1 − β )(η )   βµ  
 µ  −η α  
  (1 − α )   

TM is defined as QMaTM. Thus:

  (1 − β )(η )   βµ  
L  −T  
(1 − α )   α  
TM =   η (40)
  (1 − β )(η )   βµ  
 µ  −η α  
  (1 − α )   

LF is defined as QFaLF. Thus:

15
  
 
LF =  Tη − Lµ   PF βµ 
  (1 − β )P (η )   
 PF βµ    αPM 
µ F
 − η 
  PM (1 − α )   αPM  

 
 
Tη − Lµ   βµ 
LF =    (41)
  (1 − β )(η )   βµ    α 
µ  −η α  
  (1 − α )   

TF is defined as QFaTF. Thus:

 
 
TF =  Tη − Lµ   (1 − β )PF (η ) 
 
  (1 − β )P (η )   PF βµ    PM (1 − α ) 
µ F
 −η 
  PM (1 − α )   αPM  

 
 
TF = 
Tη − Lµ   (1 − β )(η )  (42)
 
  (1 − β )(η )   βµ    (1 − α ) 
µ  −η α  
  (1 − α )   

Once more, all the endogenous variables have been expressed in terms of the
exogenous variables. The results obtained show that both the total employment of
each factor per sector and the output of food and manufactures may differ
between countries since these variables depend on factor endowments (which are
assumed to be different), as indicated by equations (37), (38), (39), (40), (41) and
(42). These results are in sharp contrast to factor prices and unitary factor
requirements, which have already been shown to be identical in both nations.

16
IV. CONCLUSIONS

This paper has shown that it is possible to prove the Factor Price Equalization
Theorem in the framework developed by Jones (1965) with endogenous unitary
factor requirements in production, by providing a functional form that meets the
Heckscher Ohlin assumptions. It has also shown that under the model´s
assumptions, unitary factor requirements per sector will also be identical between
countries, while total use of factors in each sector and total output of manufactures
and food will be different in each nation.

V. BIBLIOGRAPHY

Jones, Ronald W. (1965) “The Structure of Simple General Equilibrium Models”

The Journal of Political Economy Vol LXXIII December 1965 No. 6 pp. 557-
572.
Krugman P. R, y M. Obstfeld (2001) “Internacional Economics”, Fifth Edition,
McGraw-Hill
Appleyard D. R. y A. J. Field. (1997) “Economía Internacional”, McGraw-Hill,
Madird.

VI. APPENDIX: THE NO FACTOR INTENSITY REVERSAL CONDITION

For no factor intensity reversals, the required condition is that for any factor prices:

a LM a LF
>
aTM aTF

Replacing each term with the expressions (20), (22), (24) and (26):

17
 a LM µ
=
aTM η

 PF βµ 
 
a LF
=  αPM 
aTF  (1 − β )PF (η ) 
 
 PM (1 − α ) 

 βµ 
α 
a LF
=  
aTF  (1 − β )(η ) 
 (1 − α ) 
 

a LF (β − αβ )µ
=
aTF (α − αβ )η

Now,

a LM µ (β − αβ )µ a LF
= > =
aTM η (α − αβ )η aTF

if and only if α > β. QED.

18