The Global Firm
Lecture 3
Methodology in Positive Economics (Part I) &
OLI Framework to MNE Analysis
Paul Deng
Feb. 15, 2011
1
Big Picture
2
Big Picture
3
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

Economic theories and assumptions

In part 2, we’ll discuss what is a good theory
4
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?
 Positive economics
 ”what is”
 Independent of any particular ethical position or value
judgements
 Normative economics
 ”what ought to be”
 Often with value judgement
 For example: majority of Karl Marx’s theories; fairness;
policy discussions on inequality issues.
5
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

An example of minimum wage – the goal is to raise living
standards of poor people or the low skilled, or prevent them
being exploited

Normative approach tends to argue the issue from morality
and ethics

Positive approach will look at (and analyze) the actual
effect of raising minimum wage on poor people’s living
standard
6
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

An example of minimum wage

Positive approach will look at (and analyze) the actual effect of raising
minimum wage on poor people’s living standard
wage
Labor
supply
E
Labor
demand
Q
7
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

An example of minimum wage

Positive approach will look at (and analyze) the actual effect of raising
minimum wage on poor people’s living standard
wage
Labor
supply
E
Minimum wage set at a
too-low level, won’t
acheive policy goal 
China’s case
Labor
demand
Q
8
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

An example of minimum wage

Positive approach will look at (and analyze) the actual effect of raising
minimum wage on poor people’s living standard
wage
Labor
supply
Dm
Minimum wage set at a toohigh level, creating structural
unemployment  the case in
most western European
countries
Sm
E
Labor
demand
Q
unemployment
9
Friedman on Positive Economics (Part I)

What is positive economics vis-a-vis normative economics?

An example of minimum wage

Positive approach will look at (and analyze) the actual effect of raising
minimum wage on poor people’s living standard
wage
Labor
supply
Dm
Minimum wage set at a toohigh level, creating structural
unemployment  the case in
most western European
countries
Sm
E
A perfect example of polices of
”good intentions but with
bad outcomes”!
Labor
demand
Q
unemployment
10
Friedman on Positive Economics (Part I)

Economic theory and its assumptions
 How realistic should the assumptions be?

Do unrealistic assumptions lead to bad theories? Or do
realistic assumptions necessarily lead to good theories?

How should a theory be judged ultimately? (to be
discussed next time)
11
Friedman on Positive Economics (Part I)

Economic theory and its assumptions
 How realistic should the assumptions be?

Theory, by definition, is abstract from reality, so it cannot be
completely realistic, including its assumptions.

The best theory is often the simpliest with the widest applications.
So its assumptions often cannot cover every aspect of reality.

Often times, ”unrealistic” assumptions help capture human
behavior as if people behave such a way. For example,
 the assumption of rationality in human behavior
 the assumption of profit-maximization of a firm
12
Friedman on Positive Economics (Part I)

The assumption of rationality in human behavior
 Rationality is one of the most questioned assumptions in economics

But most criticisms are not to the point

Indeed, psychological studies show that there can be many cases where human
behave irrationally, but

in most cases, human beings behave as if they were rational

More importantly, economic theories, mostly based on rationality assumptions,
have quite good prediction power in how people are going to behave.

For example,
 Denmark’s paternity-leave policies and Danish birth rate
 US cigarette tax and its effect on smoking
13
Friedman on Positive Economics (Part I)

The assumption of firm’s profit-maximization

Businessmen in classroom often label the assumption one of the ”craziest” . In their
own words: ”we never do such optimization, we never draw MR and MC, we don’t
even know our own firm’s supply curve, not to mention the demand curve…”

Oh, yes – all above is true, but again they are not to the point

The assumption of profit-maximization just states that firms behave as if they knew
the relevant cost and demand functions, calculated MC and MR, etc.

But more importantly, as we will discuss later, the realism of assumptions do not
really matter, as long as the theory can predict what’s going to happen, with fairly
good precision.

You may contrast what you observe in reality with the predictions you learned in firm
theory. They are fairly close --- for example, firm’s profit margin is going to be driven
down just as the theory would predict. In this case, firm behaves as if they were
under perfect competition. (for more detailed discussion on this, read Friedman,
p.21-22).
14
Friedman on Positive Economics (Part I)

Economic theory and its assumptions

Do unrealistic assumptions lead to bad theories? Or vice versa?

The answer is NO.

Most good theories have very unrealistic assumptions

If a theory has very realistic assumptions, i.e., assumptions trying to cover every
detail of reality, the theory often becomes too complex (the opposite of
simplicity) to be comprehended and often without much general use.

This leads Milton Friedman to conclude the following,
”the more significant the theory, the more unrealistic the
assumptions.”
 sounds like a pretty crazy idea, but think about it...

In other words, a theory or a hypothesis is important ”if it explains much by
little”!
15
OLI Framework in MNE Analysis


Introduction
A theoretical example: Ethier (1986)
16
Introduction to OLI Framework

What is OLI?




Ownership
Location
Internalization
A basic theoretical framework, proposed by John Dunning, to
explain




the incentives for MNEs to go overseas
the organizational forms of MNEs
MNE’s location choices
decision choice between FDI and its alternatives, such as
international trade, licensing and outsourcing
17
Introduction to OLI Framework

Before we get into the advantages of being a MNE, let’s
understand two basics:
1. FDI vs. portfolio capital investment


Portfolio investment seeks higher returns to capital
MNE is often formed to take advantage of specific business opportunities,
rather than secondary benefits of interest rate differential, or investment
returns
2. The disadvantages of being a MNE vs. domestic firm




Communication and transportation costs
Language and cultural differences
Adaptation costs to different regulations, government procedures,etc.
Additional risks: exchange rate, political stability, degree of property rights
protection
 Because of these inherent disadvantages, it’s natural to assume that MNEs
must be more productive (or they must enjoy some special advantage) over their
domestic counterparts.
18
Introduction to OLI Framework


According to Dunning (1977, 1981), for a firm to become multinational,
it must enjoy or attempts to get access to the following:

Ownership advantage --- product, design, patent, trade secret and resources --link this to Hart’s property rights approach

Location advantage --- cheap input factors, transport cost, trade barriers

Internalization advantage --- the additional benefits from establishing a foreign
subsidiary vs. joint venture, licensing, etc.
Dunning’s theory so far assumed away government intervention. In the
case of an interventionlist government, a less productive doemstic firm
may, however, become multinational, if it gets:


Special government (or bank) financing
Various subsidies, such as R&D, preferred tax rates.
 A recent example: China’s State-Owned-Enterprises (or SOEs) going abroad
(see Zilibotti, 2011).
19
Ethier (1986), A theory based on OLI framework

A model that explains and predicts the emergence of MNE due to
internalization advantage

In other words, this model aims to endogenize ”internalization” --- the ”I” in
OLI framework

This is a general equilibrium (or eqm) model
 Manufacturing eqm in three stages of production: research, upstream
production, and downstream production --- a vertical integrated firm

Labor market eqm between two countries through traded good (wheat in
the paper)
20
Ethier (1986), A theory based on OLI framework

The core idea of the model

The uncertainty in research outcome and the complexity of technology itself
produce uncertainty in its valuation between home producer and foreign user
(in downstrea operation).

This problem is further exacerbated by information asymmetry. To save
transaction costs (such as large flow of information exchange and potential
low-bid for technology), multinationals have every incentive to internalize the
whole production process.
21
Ethier (1986), A theory based on OLI framework
22
Ethier (1986), A theory based on OLI framework
23
Ethier (1986), A theory based on OLI framework
24
Ethier (1986), A theory based on OLI framework
25
Ethier (1986), A theory based on OLI framework
26
Ethier (1986), A theory based on OLI framework
27
Ethier (1986), A theory based on OLI framework
28
Ethier (1986), A theory based on OLI framework
General Equilibrium
I will pass the derivation of
labor market equilibrium, or
LE. For details, please refer
to p. 817 in Ethier’s paper.
As shown in left graph, the
general equilibrium point, E,
lies below 1/aH. This may not
be always the case, E can
also lie between 1/aL and 1/aH.
29
Ethier (1986), A theory based on OLI framework
30
Ethier (1986), A theory based on OLI framework
 The dotted lines
highlighed are the general
eqm under arm’s length
contract. The contract must
be designed (costly to do)
in a way that calls for stateinvariant quality. (see
discussion on p. 823).
31
Ethier (1986), A theory based on OLI framework
32
Ethier (1986), A theory based on OLI framework
33
Ethier (1986), A theory based on OLI framework
This is the case where M deviates
from A, when 1/aH<w<1/aL.
The optimal choice is M over A.
Note that in this case, w* under M
is greater than w* under A.
This is because under M, the
internalization gives the right
incentive for greater research
efforts and better quality, leading to
higher wages in both home and
foreign countries in manufacturing
sector.
34
Some Further Thoughts on Ethier (1986)

Ethier’s model predicts multinational activities take place between countries where
relative factor intensity (such as labor/land or labor/capital) is similar (see again on
p.827).

This is generally in line with what we observe --- direct investments in industries that
require high skills and more knowledge capital are mostly concentrated among
developed countries, which are similar in factor intensity.

But because there is only one wage rate across different sectors in the model, the
theory failed to capture the fact that some direct investments happen simply
because there are dissimilarities in factor intensity (see Markusen-Helpman model
in reference list).

Lately, we even observe that investments in knowledge capital can also happen
when factor intensity significantly differs ---case in example: MNE’s establishment of
R&D centers in China and India, where human capital is as productive but much
cheaper.
35
Potential Empirical Test on Ethier (1986)
To test Ethier’s core idea on the relation between uncertainty of research
outcome and incentives to internalize, one possibility is to look at whether
FDI is partly driven by industry’s skill intensity.
Skill intensity could be measured by share of research workers relative to
total number of workers in an industry, or could be measured by share of
R&D investments relative to industry’s total value-added.
This test could be done both for a single country, such as the U.S. or for
a group of countries, such as OECD countries.
36