Transcript Supplementary Notes on the Specific

Specific Factors and
Income Distribution
© Prof. J. Peter Neary 2009
Economics Department, Oxford
1
Specific-Factors Model
Ricardian model has nothing to say about domestic income
distribution
BUT: In reality, trade has important effects on distribution
One major reason: Factors cannot move immediately or
costlessly between sectors
An interesting alternative model, which focuses clearly on this
issue, is the Specific-Factors model
Difference in assumptions: 3 factors instead of 1
• Labour (L) is mobile between sectors
• Capital (K) and Land (T) are immobile or “sector-specific”
Similar to 2-sector Ricardian model in other respects:
• 2 goods: Manufacturing and Food
• Full Employment; perfect competition
2
QM
QM = QM(K, LM)
Fig. 3-1: The Production Function
for Manufactures
LM
MPLM
Fig. 3-2: The Marginal Product
of Labour in Manufactures
•
•
•
Equals the slope of the prod. func.
Downward-sloping: Reflecting
Diminishing Returns
As more labour is added to fixed
amount of capital, its marginal
return falls
LM
3
QF
PPF: Production
Possibility
Frontier
Production Function
For Food
LF
QM
Production Function
For Manufactures
Full-Employment Locus:
LM+LF=L
Exercise: repeat this slide and last
for 2-good Ricardian model
LM
Fig. 3-3: The PPF in the Specific-Factors Model
4
Equilibrium Allocation of Labour
Profit maximisation in each sector =>
w = Value of marginal product of labour
i.e., w = P x MPL
For given price, VMPL curve is just the MPL curve
i.e., it is the demand curve for labour
e.g., in manufacturing:
VMPLM
w
PM x MPLM
LM
LM
5
Equilibrium Allocation of Labour (cont.)
VMPLF
Similarly in food:
w
LF
LF
Now, flip this diagram around:
VMPLF
w
LF
LF
6
Equilibrium Allocation of Labour (cont.)
Finally, combine the two labour demand curves:
VMPLM
VMPLF
w
LF
LM
LM
LF
L
Horizontal axis measures L: i.e., full employment
Intersection of two demand curves determines equilibrium wage
and allocation of labour between sectors
Note: w is exogenous in partial equilibrium, endogenous in general equilibrium: follow the arrows!
7
Effects of a Rise in the Relative Price of Manufactures
VMPLM
VMPLF
B
w
A
LF
LM
LM
LF
Curve shifts upwards: equilibrium shifts from A to B
•
•
•
•
Employment and hence output of manufactures increase (intuitive)
Wage rises BUT by less than the price increase
i.e., wage-earners gain in terms of food, lose in terms of manufactures
Owners of capital gain, owners of land lose (in terms of both goods)
8
Effects of a Rise in the Relative Price of Manufactures (cont.)
Effects of price rise in PPF diagram:
• Initial equilibrium at A
• Price = MC in each sector => slope of price line = slope of PPF
QF
A
B
QM
Higher world price of M => steeper price line
=> New equilibrium at B: QM rises, QF falls
9
Effects of a Rise in the Capital Stock
VMPLM
VMPLF
B
w
A
LF
LM
LM
LF
Curve shifts rightwards: equilibrium shifts from A to B
(Looks very like last VMPL diagram, though interpretation is different)
• Employment and hence output of manufactures increase (intuitive)
• Wage rises relative to both goods prices: i.e., wage-earners definitely gain
• Owners of capital lose per unit, owners of land lose
10
Implications for Trade Patterns
N.B. This diagram is similar to the 2-good Ricardian case; differences:
• Countries have identical technology, and differ in factor endowments
• The supply curves are smooth
p=pM/pF
p*A
pF
pA
Foreign has less capital:
So lower relative supply of M
World supply curve
an average of the two
Assume Home has more capital:
So higher relative supply of M
Relative demand curve
determines equilibrium
in both autarky and free trade
QM  QM*
QF  QF*
11
Distributional Conflict versus Aggregate Gains
QF
A
B
QM
Move from autarky at A to free trade at B:
•
•
Since some factors gain and some lose, what can we say?
At least, we can say that the economy as a whole could do better than its initial
total consumption
i.e., starting from B, economy could consume along red portion of new price line
•
•
•
As in Ricardian model, trade expands the economy’s consumption possibilities
So, losers could be compensated and still leave gainers better off
BUT: In practice, compensation is rarely carried out
12
Specific Factors: General Conclusions
• Differences in resources: a source of comparative advantage
[Topical, with oil at $120 a barrel! (April 2008)]
• Trade (and any other change) has both winners and losers
• Winners are factors specific to export sectors; losers are
factors specific to import-competing sectors
• Winners could compensate losers …
• BUT: In practice, such compensation is rarely carried out fully
• Though there are examples of partial compensation
e.g., adjustment assistance, retraining subsidies, temporary subsidies, etc.
• Case study: Repeal of the Corn Laws in 1846
See: http://en.wikipedia.org/wiki/Corn_Laws
• Overall: A very neat, simple model:
– Rationalises partial equilibrium intuition, in a fully specified GE model
– Highlights effects on distribution
– But: Naïve theory of trade [Samuelson: “tropical countries export tropical products
because of the abundance of tropical products there”!]
13
Reconciling the HOS and Specific-Factors Models
So far: Two interesting but very different models: Are they
necessarily in conflict?
Not necessarily - One way of reconciling them: Interpret
each as applying to a different time frame.
• SF: Describes short-run responses
• HOS: Describes longer-run responses (after capital stock
has had time to adjust)
Mechanism driving this adjustment can be seen by looking at
full effect of a price change in SF model:
pˆ M  pˆ F
ˆ  pˆ F  rˆF
 rˆM  pˆ M  w
– Note: There is a magnification effect on the rentals (though not on the wage)
– Compare this with the effects in the HOS model
14
Reconciling the HOS and Specific-Factors Models (cont.)
This rise in rM relative to rF encourages capital to flow
(reallocate) over time out of sector F into sector M
As a result:
• Long-run response to price change is greater than short-run
response: i.e., declining sector declines by more
• Economy converges in longer run towards the equilibrium
predicted by the HOS model
Finally: An alternative (though related) difference in
interpretation between the two models concerns the
different kinds of shocks they illustrate:
• SF: Describes response to unanticipated shocks
• HOS: Describes response to anticipated shocks: Capital
will have already begun to move
15