Transcript
Advanced Topics in Trade Lecture 9 - Multinational Firms and Foreign Direct Investment
Heiwai Tang - SAIS
April 8, 2015
Today’s Agenda
I
Stylized facts about multinational firms (Antras and Yeaple, 2014).
I
Determinants of horizontal FDI by Brainard (1997).
I
A simple conceptual framework, based on Helpman (JEL 2006), to study different modes for firms to serve foreign markets.
I
A brief introduction of Antras and Helpman (2004) about incomplete contracts (imperfect contract enforcement) and vertical foreign direct investment (FDI).
Multinational firms have been largely overlooked in most trade models I
So far, in all models we studied in this course, the entire production process is assumed to be undertaken in the domestic economy. Thus, trade was always in final goods.
I
However, evidence suggests that firms frequently choose to service foreign markets through local subsidiaries, thus becoming multinational firms.
I
The literature refers to this arrangement as horizontal FDI.
I
Multinational corporations account for a significant fraction of world trade flows, with trade in intermediate inputs between divisions of the same firms constituting an important portion of these flows.
I
Outsourcing of production tasks to foreign affiliates are referred to as vertical FDI.
Why should we care about multinational firms? Answer: Because they are important in the global economy: I
Value added of all MNEs (including parent firms) was around 8 tr USD in 1997, which was roughly 25% of world GDP.
I
Excluding parent firms, about 10-12% of world GDP is accounted for by foreign affiliates (in 2007).
I
Also, 1/3 of the volume of world trade is intra-firm trade (transactions that happen within firm boundary.
I
In 2004, 36.1% of total U.S. imports were intra-firm.
I
About another third of the volume of world trade is accounted for by transactions in which MNEs are in one of the two sides of the exchange (71% of U.S. exports).
I
The 700 largest MNEs account for roughly 50% of world R&D spending and close to 70% of world business R&D spending.
Distribution of Worldwide Production of Multinational Motivation Enterprises (MNEs) Domestic Production versus Worldwide Sales •
Less than 50% of Ford’s and Toyota’s production occurs in their countries of origin
Share of Worldwide Production of Ford
Share of Worldwide Production of Toyota South America 3% Europe 8%
Asia Pacific Africa 15% North America 47%
Asia Pacific Africa 27%
Europe 29% South America 9%
North America 18%
Economics 1535: Lecture 18
I
Japan 44%
2
Source: Antras (2013) CREI talk Pol Antràs (Harvard)
Horizontal FDI (Part 1)
Spring 2014
3 /
Domestic Production versus Worldwide Distribution of Worldwide Production ofSales MNEs U.S. firms serve foreign markets mostly (75%) via foreign affiliate sales How American Firms Serve Foreign Markets, 2009
U.S. Exports $1.57 trillion
Foreign Affiliate Sales to 3rd Countries, $1.47 trillion
Local Sales of Foreign Affiliates $2.95 trillion
Economics 1535: Lecture 18
I
3
Source: Antras (2013) CREI talk Pol Antràs (Harvard)
Horizontal FDI (Part 1)
Spring 2014
4 / 46
Multinationals’ Presence in U.S. Trade Multinationals in U.S. Trade
I
Source: Antras (2013) CREI talk
Pol Antràs
(Harvard)
Horizontal FDI (Part 1)
Spring
Share of Intrafirm Imports from Top Destinations
Variation in the Share of Intra…rm Trade across Countri
Pol Antràs
I
(Harvard)
Horizontal FD I (Part 1)
Source: Antras (2013) CREI talk
Spring 2014
Variations in the Share of Intrafirm Trade across Industries
Variation in the Share of Intra…rm Trade across Industri
Pol Antràs
I
(Harvard)
Horizontal FDI (Part 1)
Source: Antras (2013) CREI talk
Spring 2014
Six Stylized Facts about Multinational Firms (Antras and Yeaple, Handbook of International Economics 2014)
Fact 1: MNEs are highly concentrated in developed countries. Stylized Facts About Multinational Activity: #1 Multinational activities are concentrated in developed countries where they are mostly two-way. Developing countries are concentrated more likely to the Fact One: Multinational activity is primarily in be developed recipient of multinational activities than the source. countries where it is mostly two-way. Developing countries are more likely to be the destination of multinational activity than the source 2
2
log (Outward FDI Stock / GDP)
0 ‐1 ‐2 ‐3 ‐4 ‐5 ‐6 ‐7 ‐8 ‐9
LBR
1 0 log (Inward FDI Stock / GDP)
HKG LUX IRLCHESGP BEL PAN NLD LBR CYP ISL GBR SWE DNK FRA EUU FIN AUT NOR MYS BHR ESPDEU CAN ESTPRT ISR AUSUSA RUS ITA CHL SYC ZAF LBN ZMB QAT ARE SVN HUNMLT JPN KWT KOR GRC NZL AZE KAZ SRB TTO MNE BRA ARG THA COL POLCZE HRV MEX UKR LTU AGO IND BRN MUS CHN VEN GAB SLB PHL OMN SAU BLZ LVA SVK TUR NER GIN BGR KHM VUT MAREGY BWA SEN NGA NIC MNG PNG CAF JAMPER JOR YEM SWZ PRY FJIGEO ALB MDA DZA HND MKDROM PAK BEN KEN CMR GTM ARM TCD LKA MRT SYR MLI TUN URY ECU NAM BIH MWI BLR GNB CRI IDN RWA WSM y = − 14.23 + 1.206 x BFA (1.004) (0.109) CIV BGD BDI LSO GUY BOL MDG R2 = 0.467 CPV SLV MOZ HTI KGZ GNQ
1
‐1
ZAR
‐2 ‐3
HKG
KNA SYC SGP LUX VCT LCA HND GRD COG MNE CYPBEL DMA MLT IRLCHE BHS BGR LBN BRN BLZJAM EST TTO ISLNLD STP SLB PAN GUY JOR MNG HUN NIC FJIGEO TUN PLW CPV CZE SWE BHR VUT CHLHRV GMB VNM MRT MOZ SVK KAZ SRB KHM ZMB LSO MAC MDG GNQ MDA MKD TCD PRTNZL GBR NAM MAR LVA ARM AUS DNK NOR ESP ROM EUU BIHZAF MYS NER GIN TZA UKRMDV AUT POL TKM THA FRA TMP SAU CRIURY LTU ALB SLV ISR CAN FIN BOL UGA EGY TON SVN MEX SDN NGA AGO TGO LAO CIV DOM PER MUS SLE GNB GHA CMR ARE QAT SWZ COL ARGRUS OMNDEUUSA BRA TUR ERI CAF KGZ MWI IDN PRY ECUAZE BLR ITA TJKPNG GTM ETH YEM MLI BENSENKIR PHL SYR PAK DZA IND UZB GRC GAB COM BTN BFA LKA AFG HTI BWA IRQWSM CHN VEN RWA KEN BGD KWT BDI JPN
‐4
NPL
KOR
‐5 ‐6 y = − 2.985 + 0.219 x (0.483) (0.054)
‐7
R2 = 0.087
‐8 ‐9
5.5
6.5
7.5
8.5
9.5
10.5
11.5
5.5
6.5
log (Real GDP Per Capita) Sources: UNCTAD and World Bank
7.5
8.5
9.5
log (Real GDP Per Capita) Sources: UNCTAD and World Bank
Figure: Aggregate FDI Stocks and Development
10.5
11.5
Fact 1: MNEs are highly concentrated in developed countries. Share of U.S. Intrafirm Imports and Capital Abundance of the Exporting Countries
Stylized Facts About Multinational Activity: #1
Share of Intrafirm Imports , Average 2000‐05
1 y = − 0.485 + 0.075 x (0.117) (0.011)
0.9
R2 = 0.214
IRL
IRQ
0.8 0.7
SVK MEX HUN GAB CRI MYS SUR
STP
JPN SGP LUX KWT
FINAUT GBRNLD LBY DNK CHE KOR NOR DOM BHS CAN SLVVCT ISL FRA YEM BEL COG SLB JAM NGA NZL BIHMAR CIV MDV VEN GMB CZE PRT SVN LCA THA BOL BRA AUS LBR ESP ITA POL TTO MRT TON BLZ MLI ISR GNB ZMB RUS IDN ESTTWN ZAF CHN TUN DJI DZA DMA CHL MWI COL ARG SLE WSM BHR GRC SYR ECU HRV NIC SEN BRB ALB NAM OMN UGA LKA TCDSDN HKG ARE FJI EGY GTM IND KEN QAT SYC TUR GUY TZA BGR PAN VNM COM CMR ZWE BFA HTI BEN SOM GNQ URY AGO BRN IRN JOR BTN GHA GRD CPVPER ETHMDG LBN GIN TGO ZAR BWA CYP RWA PAK MNG NPL CAF SWZ VUT NER MKDMUS MAC PNG MOZ KHM BGD LAO LSO UZB PRY CUB AFG
0.6
MLT SAU SWE
HND
PHL
0.5 0.4 0.3 0.2 0.1 BDI
0 6.5
7.5
8.5
9.5
10.5
11.5
12.5
Log Capital/Labor Ratio, Average 2000‐05 Sources: U.S. Census Related‐Party Trade Database and Penn World Tables (using perpetual inventory method of Caselli, 2005)
Figure: Share of U.S. Intra…rm Imports and Physical Capital Abundanc
Fact 2: Intrafirm trade is more prevalent in capital and Stylized Facts About Multinational Activity: #2 R&D intensive sectors.
1 y = − 0.142 + 0.108 x (0.122) (0.025)
0.9
3361
R2 = 0.182
0.8
3254
0.7 0.6 0.5 0.4 0.3 0.2
3159
0.1 3162
0 3
3.5
4
4.5
5
5.5
6
6.5
7
Share of Intrafirm Imports by NAICS 4, Average 2000‐05
Share of Intrafirm Imports by NAICS 4, Average 2000‐05
Fact Two: The relative importance of multinationals in economic activity is higher in capital intensive and R&D intensive goods, and a signi…cant share of two-way FDI ‡ows is intraindustry in nature 1 R2 = 0.345
0.8
3254
0.7 0.6 0.5 0.4 0.3 0.2
3159
0.1 3162
0
Log U.S. Capital/Employment by NAICS 4, Average 2000‐05 Sources: U.S. Census Related‐Party Trade Database and NBER‐CES Manufacturing Industry Database
y = 0.940 + 0.142 x (0.091) (0.022)
3361
0.9
‐5
‐4.5
‐4
‐3.5
‐3
‐2.5
‐2
‐1.5
Log (R&D Expenditure/Sales + 0.01) by NAICS 4, Average 2000‐05 Sources: U.S. Census Related‐Party Trade Database and Nunn and Trefler (2008)
Figure: The Share of Intra…rm Imports, Capital Intensity and R&D Intensity
Pol Antràs (Harvard)
Horizontal FDI (Part 1)
Spring 2014
19 / 4
Fact 3: MNEs’ activities are decreasing in distance. Stylized Facts About Multinational Activity: #3 The production of the foreign affiliates of multinationals is decreasing in distance (from the host country), butofatthea foreign slower a¢ rate than Fact Three: The production liates of aggregate multinationals falls trade (both andbut exports) o¤ imports in distance, at a slower rate than either aggregate exports or parent exports of inputs to their a¢ liates ‐26
y = − 2.877 + 0.318 x (0.636) (0.078)
3
R2 = 0.144
log(Affiliate Salesds/Exportsds)
log (Affiliate Salesds / GDPdGDPs)
4
y = − 28.16 − 0.573 x (0.658) (0.080)
‐28
‐30
‐32
‐34
‐36
R2 = 0.052
2 1 0 ‐1 ‐2 ‐3
‐38
‐4
‐40
‐5 5
6
Sources: Ramondo (2012) and World Development Indicators
7 8 log (Distance between s and d)
9
10
5
6
7 8 log (Distance between s and d)
9
10
Sources: Ramondo (2012), Feenstra's trade data and World Development Indicators
Figure: Gravity, FDI Sales and Trade Flows
Pol Antràs (Harvard)
Horizontal FDI (Part 1)
Spring 2014
22 / 4
Fact 4: MNEs’ headquarters tend to be associated with better performance.
Fact Four: Both the parents and the a¢ liates of multinational …rms te be larger, more productive, more R&D intensive and more export orient than non-multinational …rms Table 3. U.S. Parent Firms Percent Shares of US Activities
Enterprises Value Added Employment R&D Expenditure Exports
Manufacturing
All Industries
0.4 62.3 58.2 72.1 44.4
0.04 17.6 16.9 69.2 54.7
Source: Barefoot and Mataloni (2011), U.S. Census
Fact 4: MNEs’ headquarters tend to be associated with Stylized Facts About Multinational Activity: #4 better performance.
Table 4. A¢ liates Relative to Local Firms
Enterprises Employment Sales R&D Expenditure Exports
Finland
France
Ireland
Holland
Poland
Sweden
1.6 17.2 16.2 13.1 17.5
2.0 26.2 31.8 27.4 39.5
13.4 48.0 81.1 77.3 92.3
3.4 25.1 41.1 35.8 60.0
16.0 28.1 45.2 20.9 69.1
2.8 32.4 39.9 52.0 45.8
Source: OECD (2007).
Stylized Facts headquarters About Multinational Activity: #5 high Fact 5: MNEs’ focus on R&D and other value-added activities.
Within multinational enterprises, parents are relatively specialized in Factwhile Five:affiliates Withinaremultinational enterprises, relatively spec R&D primarily engaged in selling parents goods inare foreign in R&D while a¢ liates are primarily engaged in selling goods in foreign markets (both the host market and third markets).
markets, particularly in their host market. Table 5. Share of US Multinational Activity Due to Parent Firms 1999 Sales Value Added Employment R&D Expenditure Source: BEA
74 78 75 87
2009 65 68 68 84
Fact 6: MNEs’ headquarters focus on on R&D and other high value-added Stylized Factsactivities. About Multinational Activity: #6
Cross-Border Mergers and Acquisitions, compared to green-field investments, up a large Mergers proportion of FDI and areupa aparticularly Factmake Six: Cross-Border and Acquisitions make large fraction of importantFDI mode of entry into developed countries. and are a particularly important mode of entry into developed countries. Table 7. The Value of Mergers and Acquisitions Relative to FDI Flows
World Developed Developing Source: UNCTAD
2005
2006
2007
2008
2009
2010
0.47 0.65 0.19
0.43 0.54 0.21
0.52 0.68 0.18
0.41 0.60 0.16
0.21 0.34 0.08
0.27 0.42 0.14
Determinants of Horizontal FDI
Helpman (06): Trade, FDI, and the Organization of Firms I
A simplified version of Melitz (2003).
I
A firm’s profit from domestic sales πD (Θ) = ΘB − cfD , where Θ is some function of the firm’s productivity and B captures the level of industry demand at home, as well as factor prices (e.g. wages).
I
Profit from exports πX (Θ) = τ 1−ε ΘB ∗ − cfX , where B ∗ captures the level of industry demand in the foreign country, as well as factor prices (e.g. wages).
I
Profit from horizontal FDI πI (Θ) = ΘBI∗ − cfI ,
I
Intuitive assumptions of the ranking of fixed costs: fD < fX < fI
One killer graph that captures firms’ decisions
598
Journal of Economic Literature, Vol. XLIV (September2006)
Figure 3. Multinationals,Exporting, and Nonexporting Firms
technologyH or a traditional technologyL, as
Argentinianfirms. This raises the operatin
Brainard (AER 1997) ”An Empirical Assessment of the Proximity-Concentration Trade-off Between Multinational Sales and Trade” I
A pioneering paper in economics to empirically examine the determinants of horizontal FDI versus exporting, using sectoral data
I
Proposes an econometric model in which the share ln(Exp/(MNE affiliate sales + Exp)) is regressed on: I industry and country-specific measures of trade costs, and I industry measures of plant-specific and firm-specific economies of scale. I controls related to the importing country, such as GDP per capita, corporate tax rates, openness to trade, openness to FDI;
I
Specification: EXmj ln FDImj + EXmj
j = α + β1 Freightm + β2 Tariffmj + β3 GDP Diff j
+β4 Tax j + β5 Trade j + β6 FDI j + +β7 PlantSCm + β8 FirmSCm + jm , where m indexes country and j indexes industry.
Empirical Results (Brainard, AER 1997)
I
For each industry, measures plant-level economies of scale by the number of production employees in the median US plant (in terms of value added)
I
For each industry, measures firm-level economies of scale by the number of nonproduction workers in the average US-based firm.
I
Finds that these two variables have opposite effects on the ratio of exports to total sales abroad.
I
The coefficient on the difference in GDP per capita between the foreign country and the U.S. is positive and significant I
Perhaps Brainard is unintentionally testing both horizontal and vertical FDI at the same time?
Empirical ResultsEmpirical (Brainard,Results AER 1997) rainard (1997): 530
THE AMERICANECONOMICREVIEW TABLE 1-EXPORT
Independent variable FREIGHT TARIFF PWGDP TAX TRADE FDI
OLS (i)
SHARES (DEPENDENT VARIABLE =
Country random effects (ii)
Industryrandom effects (iii)
-0.2451 (-5.429) -0.274 (-6.239) 0.330 (4.272) -1.335 (-4.882) 1.9114 (7.416) -2.6163 (-9.264)
-0.2009 (-3.996) -0.2814 (-5.666) 0.3231 (2.371) -1.3566 (-2.809) 1.9395 (4.149) -2.6302 (-5.077)
-0.1264 (-2.672) -0.0872 (-2.038) 0.1922 (2.909) -0.9853 (-4.258) 2.1306 (9.887) -2.8126 (-11.944)
3.6903 (2.212)
3.9210 (1.281)
3.5633 (2.554)
1,159 0.118
1,159 0.040 1.8446 0.933
PSCALE CSCALE ADJ LANG EC COUP Constant Number of observations Adjusted R2 X2 P
OLS (iv) -0.2717 (-4.578) -0.3707 (-7.447) 0.2958 (3.747) -0.5695 (-1.795) 1.6558 (6.305) -0.8343 (-1.810) 0.1345 (2.735) -0.2726 (-4.656) -0.0313 (-0.156) -0.1767 (-1.803) -0.8107 (-5.933) 0.6247 (2.624) -4.7336 (-2.042)
1,159 0.080 23.425 0.001
1,035 0.233
SEPTEMBER1997 EXSH) Country random effects (v) -0.2852 (-4.813) -0.3895 (-7.259) 0.3050 (2.677) -0.5787 (-1.223) 1.5841 (4.035) -0.8502 (-1.219) 0.1331 (2.728) -0.2734 (-4.722) -0.0177 (-0.069) -0.1459 (-0.998) -0.7808 (-3.823) 0.6486 (1.805) -4.4334 (-1.270) 1,035 0.140 4.8503 0.963
Industryrandom effects (vi) -0.1228 (-1.767) -0.1644 (-3.412) 0.1461 (2.122) -0.2150 (-0.792) 1.8477 (8.262) -0.9120 (-2.334) 0.1087 (0.941) -0.2291 (-1.587) -0.0367 (-0.188) -0.2707 (-3.223) -0.8165 (-7.040) 0.5632 (2.788) -5.1163 (-2.535) 1,035 0.180 22.154 0.036
Notes: The table reportsestimates of equation (3); t values are reportedin parentheses.All variables are in logs. Samplesize differences reflect missing data.
Pol Antràs (Harvard)
Horizontal FDI (Part 1)
set of independent variables. The coefficients
addition, the country variables again interact
Spring 2014
4
between them. To assess this implication, we added a variable GDP , which is the logarithm of the absolute value of GDP between the United States and country i. Running this augmented regression
Empirical Results (Brainard, AER 1997) inard (1997): Updated Results More Recent D statistically signi…cant coe¢ cient on GDP suggests that marketwith size does matter: U.S. …rms are more on 2009 data, we obtain the coe¢ cient estimates shown in column 3. The positive and (moderately)
likely to export to smaller markets and to engage in foreign direct investment in larger countries. Table 3: Proximity-Concentration Empirics Dep. Var.: log
Xji Xji +Sji
(1)
(2)
(3)
(4)
(5)
(6)
Freight
-0.28**
-0.13**
-0.12**
-0.13**
-0.13*
0.01
[0.05]
[0.04]
[0.04]
[0.04]
[0.06]
[0.25]
Tari¤s
-0.23**
-0.28**
-0.27**
-0.29**
-0.38**
-0.04
[0.06]
[0.05]
[0.05]
[0.06]
[0.10]
[0.04]
GDP/POP
0.10
0.04
0.06
[0.07]
[0.08]
[0.08]
School
0.07 [0.09]
KL
0.08 [0.06]
GDP PlantSc
0.32
0.39*
[0.17]
[0.17]
0.09*
0.13*
0.13*
0.14*
0.18
[0.04]
[0.05]
[0.05]
[0.05]
[0.15]
-0.18**
-0.32**
-0.31**
-0.32**
-0.35*
[0.03]
[0.04]
[0.04]
[0.04]
[0.14]
Country Fixed E¤ects
No
No
No
No
Yes
Industry Fixed E¤ects
No
No
No
No
No
Yes
Year
1989
2009
2009
2009
2009
2009
Observations
1,762
2,315
2,315
2,315
2,482
2,482
0.15
0.09
0.09
0.09
0.16
0.40
CorpSc
R-square
Yes
Standard errors are in brackets (* signi…cant at 5%; ** at 1%). Pol Antràs (Harvard)
Horizontal FDI (Part 1)
Spring 2014
To further explore the role of relative factor endowment di¤erences, we next we replace GDP =P OP (GDP per worker di¤erences) with absolute di¤erences in years of schooling (School) and capital-labor
The Incomplete Contracts Approach to Study the Determinants of Vertical FDI
Helpman (JEL 2006)
I
Contracts are incomplete (from Lecture 3).
I
Firms’ optimal responses to contractual frictions: I
I
Where to produce? (Slicing up the value chain, such as R&D and product development, parts and components production, assembly, etc. in different countries) Extent of control that firms exert over different production stages: I
I
Should these production stages be kept within firm boundaries? Should they be contracted out to suppliers or assemblers?
Helpman (JEL 2006) Section 3 I
Based on Antras and Helpman (2004).
I
Ideas: Two-stage game. The final-good producer (in the North, e.g. U.S.) has to decide I
I
I
whether to produce the intermediate inputs in-house or outsource it to another firm If outsourcing turns out to be optimal, the firm decides whether to outsource it to a firm in the North or the South.
Different organizational modes are associated with I
I
Different bargaining power for the supplier and the final-good producer. Bargaining power is higher if the final-good producer owns and controls the supplier, through vertical integration. Different fixed costs: production in the South and vertical integration respectively are associated with higher fixed costs (e.g. overhead, management costs).
Helpman (JEL 2006) Section 3
I
The idea that control over another firm’s asset can enhance the bargaining power of the controlling firm is the key of the Property-rights Theory of the Firm.
I
What kind of final-good producer would prefer to vertically integrate with the supplier? and produce in the South?
Helpman (JEL 2006) Section 3 I
Cost structures aside, organizational mode of production depends on the (intrinsic) characteristics of the industry.
I
For example, in R&D intensive sector (e.g., pharmaceutical companies), the headquarter firm’s pre-production investment accounts for a big part of the production cost.
I
If the supplier deviates from ex-ante agreements or expectations, the final-good supplier could suffer a big loss (esp when investments are sunk and relationship-specific).
I
Antras and Helpman (2004) call this the ”hold-up” problem.
I
The party that has a larger share of ex-ante investment in the joint production unit is relatively more vulnerable to hold-up, and thus deserves a higher level of protection by having a larger bargaining power, through vertical integration.
Helpman (JEL 2006) Section 3 I
Antras and Helpman (2004) embed the two-stage bargaining game and differentiated sectors in the Melitz (2003) heterogeneous-firm model.
I
Production function in each sector takes the Cobb-Douglas form, and differs in headquarter-input intensity (like R&D, capital, skill intensity).
I
Predictions: Firms with different productivities will optimally sort themselves into one of the four production modes (conditional on survival): I I I
I
Outsourcing in the North (Domestic outsourcing) Outsourcing in the South (Offshoring) Vertical integration with suppliers in the North (Domestic M&A) Vertical integration with suppliers in the South (Vertical FDI)
Antras and Helpman (2004) Predictions in a Graph global sourcing
567
Fig. 4.—Equilibrium in the headquarter-intensive sector
depicted in figure 3—as the benchmark case. In this event the freeentry condition (11), together with (6) and (8), imply
Antras and Helpman (2004) Sorting of Firms based on Productivity
global sourcing
565
Fig. 2.—Organizational forms
high-productivity firms outsource components in the South, low-productivity firms outsource them in the North, and the least productive firms exit. On the other hand, integration takes place in headquarterintensive sectors (i.e., high h). The most productive firms integrate in the South and somewhat less productive firms outsource in the South. Firms with even lower productivity acquire components in the North,
