Factor mobility and trade

1
Factor mobility and trade

• Applied International Trade Analysis
• Lecture 8

2
Two forms of international factor mobility (IFM)

• LABOR worker migration (from low wage countries
  (developing countries) to high wage countries
  (developed countries)),
• CAPITAL capital investment tend to flow from low
  to high return-to-capital countries
• FDI (foreign direct investment) long term
  investment into firms with the aim of permanently
  benefiting the advantages of being present in the
  foreign markets (greenfield, mergers
  acquisitions),
• PI (portfolio investment) short term capital
  investment with the aim of maximizing the
  return-to-capital (financial investments))
• IMF classification
• foreign ownership should exceed 10 of firm's
  equity

3
TRADE AND FACTOR MOBILITY IN DIFFERENT TRADE
THEORIES

• NEOCLASSICAL MODEL (HOS MODEL)
• Free trade in goods leads to international
  equalization of product prices, which in turn
  leads to international (absolute and relative)
  factor price equalization.
• Hence, given the FPE theorem (factor price
  equalization) mobility of factors is not needed.
  Similarly, when there is a complete mobility of
  factors, no trade in goods is needed.
• In the HOS model, trade and factor mobility are
  SUBSTITUTES larger trade in goods leads to
  smaller factor mobility and vice versa.

4
Two implications of HOS model

• Protectionism volume of FDI and worker
  migrations should be very large (this thesis is
  only partly confirmed by huge migrations in the
  beginning of the 20th century and in the 1960s
  and 1970s,
• Liberalism volume of FDI and worker migrations
  should decline (this thesis does not hold as
  notwithstanding trade liberalization in the
  second half of the 20th century volume of world
  FDI has grown faster than volume of world trade.

5
FACTOR-PROPORTIONS MODEL (Brainard 1993)

• HOS model, amended with differentiated goods,
• imperfect competition and intra-industry trade.
• Implications of factor-proportions model (FP
  model)
• If two countries are similar in terms of FP
• There will be no inter-industry trade nor FDI
• Intra-industry trade pattern prevails
• If two countries are very different in terms of
  FP
• FDI flows are likely
• Exports will arise (inter-industry type) from
  host to the home country
• Trade and factor mobility are COMPLEMENTS.

6
PRODUCT LIFE CYCLE(Vernon 1960)

• 1. Phase Product introduction Production in the
  home country arises in order to satisfy home
  demand,
• 2. Phase Product maturity home country exports
  standardized products, there will some relocation
  of production via FDI abroad,
• 3. Phase Product standardization FDI (complete
  relocation of production to the most cost
  advantageous location abroad), huge imports of
  standardized products back to the home country.
• Implications for trade and factor mobility
• in the 2. phase partial SUBSTITUTION between
  trade and factor mobility
• in the 3. phase complete COMPLEMENTARITY between
  trade and factor mobility

7
INCOME EFFECTS LINDER THEORY(Linder 1961)

• Countries similar in factor proportions (and in
• terms of GDPpc) will have similar structure of
• preferences. As a consequence, intra-industry
• trade will arise.
• Implications of Linder theory
• Empirical studies show that the majority of FDI
  and trade that arises due to FDI is within the
  advanced countries group, i.e. between countries
  that are countries of origin as well as countries
  of destination at the same time.
• Similarity in income levels, hence, motivates
  trade as well as FDI at the same time, leading to
  COMPLEMENTARITY of both.

8
MOTIVES for FDI

• Table Geographical breakdown of sales of U.S.
  affiliates in
• Europe, 1966-93 ()

Source World Investment Report, 1998 108.
9

• in 1960s, serving the huge local markets in
  Europe seems to dominate (tariff and trade cost
  jumping motive for FDI),
• recently, market access motive seem to be partly
  amended by efficiency motives (sales to other
  markets increase up to 40).
• U.S. affiliates in Europe engage little in
  exports back to U.S. (only 4), indicating
  COMPLEMENTARITY of trade and FDI

10

• Table Export orientation of U.S. affiliates,
  1966-
• 1993 (v )

11
Dominant motives of U.S. firms for FDI

• In advanced countries and Latin America dominant
  market-seeking motives for FDI, i.e. access to
  local markets (in 1960s in 1970s export
  orientation of affiliates was below 20 and 10,
  respectively).
• In Asian countries U.S. affiliates serve as
  export platform (export orientation exceeds 65,
  in Malesia, Singapur and Hong Kong it exceeds
  80), indicating clear resource-seeking and
  efficiency-seeking motives for FDI.

12
EVOLUTION OF WORLD TRADE AND FDI

• Table Total world exports and FDI, 1961-1997

Source Bowen et al., 1998 465, World Investment
Report, 1998 2, average of inward and outward
FDI.
13

• After the WW2 and especially after the Kennedy
  round GATT (1964-1967), world trade has increased
  tremendously.
• In contradiction to HOS implications, after 1975
  FDI flows have increased even faster.

14
TEST OF FACTOR-PROPORTIONS MODEL(Brainard 1993a)

• Given the assumption of differences in factor
  proportions, FDI should lead to inter-industry
  flows from affiliates back to the home country.
• There should be no intra-industry trade flows
  neither between parent firm and affiliate nor
  between home and the foreign country.
• The empirical model
• IITij a1DGPLi (or DKLi) a2minBDPi a3maxBDPi
  a4FFij a5Dj.

15
Model is estimated in logs at the industry level,
where
16

• Two forms of intra-industry trade flows
• MIITij intra-industry trade flows of U.S.
  affiliates abroad and foreign affiliates in the
  U.S. (inter-affiliate trade),
• TIITij total intra-industry trade flows between
  country i and U.S.
• Model is estimated for trade between U.S. and 27
  partner countries using 3-digit BEA (64 product
  groups).

17
Table Estimates of characteristics of
inter-affiliate trade (MIIT) and total
intra-industry trade (TIIT) in factor-proportions
model
18

• Parameter estimates for affiliate IIT and total
  ITT are
• similar
• both proxy variables for differences in relative
  factor abundance are significantly negative (in
  line with prediction of the model for total
  trade), while MIIT is more responsive than TIIT,
• minGDP is in line with model predictions, but not
  maxGDP,
• in line with model predictions, transport cost
  negatively affect trade, while MIIT is more
  responsive than TIIT.
• Empirical tests give similar results for both
  forms of IIT,
• which is
• in line with predictions of factor-proportion
  model for total trade,
• opposite to predictions for inter-affiliate
  trade.
• Moreover, there is surprisingly high correlation
  between both dependent variables, which implies
  that factor-proportions model with differentiated
  goods is not appropriate model for explaining
  motives for FDI, where U.S. serve as target or
  source country.
• In contradiction to the factor-proportions model,
  U.S. FDI seem to be motivated by similarities
  (not differences) in relative factor abundances,
  where transport cost play an important role.

19
TEST OF PROXIMITY-CONCENTRATION TRADE-OFF
MODEL(Brainard 1993b)

• The model predicts that FDI increases in market
  access
• barriers (transport cost, tariffs) and decreases
  in plant
• economies of scale (horizontal FDI in order to
  serve local
• market).
• There is trade-off between increasing production
  cost
• (less efficient use of economies of scale if
  production is
• divided among many locations) and lower prices
  (trade
• cost are redundant).
• Implications of the model FDI SUBSTITUTES for
  trade
• the higher the trade barriers and the lower the
  plant
• economies of scale.

20

• Model
• Dependent variables
• Outward side share of affiliates exports in
  total U.S. exports (OUTSH) or export share in
  total U.S. sales in the target country (EXSH).
• Inward side share of foreign affiliates sales in
  U.S. in total U.S. sales of that country (INSH)
  or import share in total U.S. sales of that
  country (IMSH).
• Independent variables
• similarity of preferences (Linder) oz. similarity
  in factor abundance GDPpc in foreign country,
• trade barriers transport cost (FF), foreign
  tariffs (FAT) and U.S. tariffs (USAT), non-trade
  barriers in U.S. (NTB), openness for trade and
  FDI (TOPN, FOPN), corporate tax (TAX), extent of
  depreciation in 1985-89 (EXR),
• economies of scale at the plant level (PSCL).

21
(No Transcript)
22

• Results indicate
• Linder matters similarity in income
  (preferences) increase affiliate sales and
  decrease regular trade
• Trade barriers matter transport cost, tariffs as
  well as openness for FDI increase U.S. affiliate
  sales abroad and decrease regular exports, while
  trade openness decrease U.S. affiliate sales and
  increase exports
• Economies of scale matter significant plant
  economies of scale decrease affiliate sales and
  increase regular imports

23
MOTIVES for FDI