Preliminary and Incomplete Abstract


Similarities between rankings and augmented distance



Download 1,67 Mb.
Page5/10
Date08.12.2018
Size1,67 Mb.
1   2   3   4   5   6   7   8   9   10

Similarities between rankings and augmented distance.

Despite the differences in the coefficients estimates for each type of flow, in the first section of this work it was shown the large flows between the USA and Canada in respect to the rest of the countries. To this regard, Table 3 (a ,b) shows the main USA partners in each type of flow and the cumulative sum of their shares in total USA flows. Looking at the top 5 countries within the USA partners (Table 3a), Canada is out of the top 5 USA partners in all kind of interactions. This can be expected given the amount of factors that have in common (short distance, sharing a border, trade agreement in force, large economies with similar income levels…), but it is not the only country that holds a high position in all the ranking out of the USA main partners. For example, United Kingdom is out of the top 5 countries with a largest share in the USA flows in 11 out of the 12 flows analyzed. Other countries that are commonly out of the main USA partners are Japan, Mexico and Germany.

The presence of a similar list of countries across the main partners in different types of flows has been observed for other countries8. For example, the USA is the most important partner of Canada in all kind of flows, except in tertiary students in which ranks the second. United Kingdom is the second main partner of Canada in 9 of the 12 types and the third in another one. France, Germany and Japan are in the top 5 of the Canadian partners in 9, 7 and 6 types of flows respectively. Germany, Italy, Spain, United Kingdom, Belgium, the Netherlands and the USA covers a large share of the countries that can be found in the top 5 of France main partners, except in tertiary students and in international patent activity.



An additional point to note is the high concentration of the flows (Table 3b) in a very limited number of partners. For example, in the case of the USA flows, about 30% of each flow is concentrated in its two main partners9, reaching the 50% in patents and close to the 60% in tourists’ arrivals and printed publications. But this pattern is not observed just in the USA. China’s shares are also quite concentrated in very few countries. For example, its FDI outward stocks and flows to Hong Kong account for more than the 80% of the total. In merchandise trade, the sum of the exports to the USA and Hong Kong account for more than one third of the total Chinese exports. Regarding the USA-Canada case, largely studied by the border effect literature, the USA accounts for at least the 50% of the Canadian flows in all kind of interactions, except of tertiary education (27%).

Table 3(a). Top 20 of the USA partners.

Rank

Goods exports

Service exports

FDI outward stocks

FDI outflows

Portfolio eq. assets

Portfolio long term debt

Tertiary students

Emigration

Tourists

Phone calls

Printed publicat. exports

Patents

1

CAN

GBR

GBR

NLD

GBR

GBR

CHN

MEX

CAN

MEX

CAN

JPN

2

MEX

JPN

NLD

GBR

JPN

CAN

IND

CAN

MEX

CAN

GBR

DEU

3

CHN

CAN

CAN

IRL

CAN

AUS

KOR

GBR

GBR

IND

MEX

KOR

4

JPN

DEU

LUX

LUX

CHE

NLD

JPN

PHL

JPN

GBR

AUS

CAN

5

GBR

MEX

IRL

CAN

FRA

DEU

CAN

DEU

DEU

COL

JPN

GBR

6

DEU

FRA

CHE

CHE

DEU

FRA

MEX

ITA

FRA

PHL

DEU

FRA

7

NLD

CHN

DEU

AUS

BRA

LUX

TUR

AUS

KOR

DOM

CHN

CHN

8

KOR

KOR

AUS

SGP

NLD

JPN

SAU

FRA

BRA

CHN

NLD

ISR

9

BRA

NLD

JPN

CHN

AUS

IRL

DEU

JPN

AUS

GTM

SGP

NLD

10

FRA

AUS

SGP

MEX

KOR

MEX

THA

SAU

ITA

DEU

FRA

ITA

11

SGP

BRA

MEX

JPN

HKG

BRA

GBR

PAK

ESP

PER

KOR

AUS

12

BEL

ITA

FRA

BEL

CHN

SWE

HKG

CHE

IND

BRA

BRA

CHE

13

HKG

IND

BEL

DEU

ESP

ESP

BRA

GRC

NLD

SLV

ITA

SWE

14

AUS

SGP

ESP

BRA

MEX

NOR

IDN

IRL

CHN

AUS

HKG

IND

15

CHE

HKG

BRA

SWE

ITA

ISR

FRA

NLD

VEN

FRA

IND

FIN

16

IND

CHE

HKG

HKG

IRL

ITA

COL

ESP

IRL

JPN

CHE

BEL

17

ITA

ESP

CHN

FRA

IND

KOR

NGA

IDN

COL

ITA

BEL

DNK

18

MYS

VEN

SWE

IND

SWE

DNK

KEN

CHN

SWE

JAM

PHL

AUT

19

ISR

BEL

ITA

ESP

SGP

CHL

MYS

COL

CHE

NGA

NZL

SGP

20

SAU

ISR

KOR

CHL

FIN

SGP

PAK

SWE

ARG

ARG

SWE

ESP

Table 3(b). Cumulative sum of the shares of the main USA partners.

Rank

Goods exports

Service exports

FDI outward stocks

FDI outflows

Portfolio eq. assets

Portfolio long term debt

Tertiary students

Emigration

Tourists

Phone calls

Printed publicat. exports

Patents

1

22%

17%

16%

18%

17%

20%

20%

19%

33%

19%

50%

40%

2

36%

30%

32%

30%

30%

38%

36%

35%

58%

32%

61%

52%

3

42%

42%

41%

39%

39%

46%

48%

44%

66%

39%

68%

63%

4

48%

50%

48%

48%

46%

53%

54%

53%

72%

44%

72%

68%

5

52%

57%

53%

54%

53%

59%

59%

61%

75%

48%

76%

73%

6

56%

61%

57%

58%

59%

65%

61%

64%

77%

52%

78%

77%

7

60%

65%

61%

62%

63%

69%

64%

67%

79%

56%

80%

80%

8

63%

69%

65%

65%

67%

72%

65%

70%

80%

59%

82%

82%

9

65%

73%

68%

68%

70%

75%

67%

72%

82%

63%

84%

84%

10

68%

76%

72%

70%

73%

77%

68%

74%

83%

65%

85%

86%

11

70%

79%

75%

73%

76%

79%

70%

75%

84%

67%

86%

88%

12

72%

82%

78%

75%

78%

81%

71%

76%

85%

69%

88%

90%

13

75%

84%

80%

77%

80%

83%

73%

78%

86%

70%

89%

91%

14

76%

87%

82%

79%

82%

85%

74%

79%

87%

72%

90%

93%

15

78%

88%

84%

80%

84%

86%

75%

80%

88%

73%

91%

94%

16

80%

90%

86%

82%

86%

88%

77%

81%

89%

74%

91%

95%

17

81%

91%

87%

83%

88%

89%

78%

82%

89%

75%

92%

95%

18

82%

92%

88%

85%

89%

90%

79%

83%

90%

76%

92%

96%

19

83%

94%

89%

86%

90%

90%

80%

84%

91%

77%

93%

97%

20

84%

95%

90%

87%

92%

91%

81%

85%

91%

78%

93%

97%

Source: Own calculations.

One can argue that the persistency of the main partners across flows can be linked with the size or with the level of openness of the partners. For example, 20% of the international tertiary students enrolled in the USA come from China. However, Chinese students are the more numerous international students in 12 countries, accounting for the 21% of total international tertiary students. So, Chinese tertiary students in the USA account for the same proportion than in the whole sample. If we take into account the share that the students from each country in the total international education flows, the country out of the USA partners with the largest deviation from the world’s proportion is India, which concentrates the 8% of the total students studying abroad, but accounts for the 17% of the foreign tertiary students in the USA.

For each country, we have computed the deviation (difference) between the shares that each partner represents in its total flows and the share that each partner accounts for in the total international flows. In this way, we have a relative measure on how important are the linkages between two countries in respect to the share of the partner in the flows of the rest of the countries. Then, Table 4 shows the pairwise correlation of these relative shares for different type of flows. All the correlations are positive and statistically significant.

The largest correlations are found for tourists’ arrivals and services exports. This can be explained by the fact that tourism is a part of services in the balance of payments. The deviations from the world’s shares of outgoing phone calls and emigration are also highly correlated, showing that information flows as phone calls are correlated with the pre-existence of social relations. Relative shares in minutes of outgoing phone calls are highly correlated with the relative shares of services exports. In addition, a high correlation is found for goods and services that can be interpreted as the presence of some complementarity between them or by some similarities on the factors affecting goods and services exports.



Table 4. Pairwise correlations.




(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

(11)

Merchandise exports (1)

-































Services exports (2)

0.783

-




























FDI outward stocks (3)

0.499

0.545

-

























FDI outflows (4)

0.451

0.545

0.769

-






















Portfolio equity assets (5)

0.373

0.422

0.473

0.324

-



















Portfolio long term debt (6)

0.436

0.532

0.347

0.302

0.536

-
















Education (7)

0.346

0.397

0.445

0.382

0.207

0.134

-













Emigration (8)

0.435

0.621

0.461

0.305

0.288

0.337

0.298

-










Tourists (9)

0.655

0.824

0.588

0.671

0.432

0.566

0.295

0.673

-







Outgoing phone calls (10)

0.656

0.764

0.602

0.506

0.469

0.474

0.417

0.694

0.811

-




Printed publications (11)

0.396

0.634

0.463

0.319

0.270

0.331

0.292

0.455

0.465

0.528

-

Patents (12)

0.381

0.480

0.250

0.194

0.347

0.344

0.220

0.307

0.452

0.482

0.308

Source: Own elaboration.

Now, we have computed a composite distance measure that considers the impact that each exogenous variable has in the standard gravity equation for each type of flow (according to the estimates shown in Table 1) and the values of the standard gravity variables that reflect any kind of distance. Then, it has been obtained a vector with 97*(96-1) = 9312 elements, containing for each country pair a composite distance measure implied by the coefficients estimates of each type of flow, according to the following calculation procedure10:



(2)

(3)

Where ; ; ; ; and .

Eq (2) describes an inverse composite distance obtained as the inverse of the product of the impact multiplier considering the coefficients estimates from the gravity equation and the similarities or differences of each country pair. Then, the composite distance has been rescaled multiplying the inverse of the product of all the multipliers () by the second term in eq (3), , making the composite distance a measure comparable to the geographical distance.



Share with your friends:
1   2   3   4   5   6   7   8   9   10


The database is protected by copyright ©sckool.org 2019
send message

    Main page