DISTANCE AND THE PATTERN OF
INTRA-EUROPEAN TRADE
W. Beckerman
THE basic questionto which this paperis
addressed is "What is the importance of
distance in determining the pattern of Western
European trade?" The importance of distance
in the pattern of trade has, of course, always
been recognized. The assumption of "no transport costs," which has always been necessary
in expositions of theories of internationaltrade,
is a recognition of the fact that transport costs,
that is, the costs of covering distance, exist and
are significant; so that abstraction from them
has to be made quite explicitly in order to
analyze other elements such as factor endowments.'
First, it is necessary to abstract from the effect on the pattern of trade related simply to
differencesin the size of countries. Thus, Germany's imports from Greece are likely to be
smaller than Germany's imports from France
simply because the total exports of Greece are
small relative to the total exports of France.
Therefore, in order to explain the distribution
of Germany'simports in terms of factors other
than the relative size of the supplying countries, say, in terms of distance, relative prices,
relative incomes, the data have to be corrected
to eliminate the effect of differences in the size
of the trade of the various countries. This may
be done by adjusting every country to an
"(equal importer" or "equal exporter" basis.
The framework used for this adjustment is as
follows.
To facilitate statistical work, Western Europe is defined as the member countries of the
O.E.E.C. The intra-European trade of these
countries is arranged in the form of a trade
matrix. The data used are the f.o.b. export
figures as recorded by the exporting countries.
Then in order to examine first the distribution
I
A recent article by W. Isard and M. Peck contains
some empirical evidence of the importance of transport costs
in trade within the United States. This article also contains
reference to related studies, such as that carried out in the
1920's by the German National Bureau of Statistics. See
Isard and Peck, "Location Theory and International and
Interregional Trade," Quarterly Journal of Economics,
97-II5.
LXVIII (February 1954),
of each country's imports, all countries have
been put on an "equal exporter" basis. To do
this the elements in all the rows (exports) of
all the countries have been multiplied by the
coefficients (one for each row) required to
bring the row totals up to (or down to) ioo.
Similarly, in order to examine the distribution
of each country's exports, the data have to be
adjusted to put each country on an "equal importer" basis by adjusting the columns relating
to each country in such a way that they all total
ioo. The principal tables are in the form of
trade matrices; there are two tables for each
year examined, one in which the countries are
adjusted to "equal exporter" basis, and one in
which they are adjusted to "equal importer"
basis. Three years have been covered in this
study, namely, 1938, I948, and I953.
The
tables for I938 and I953 are shown in full in
the Appendix. The tables for I948 can be supplied by the writer upon request.
Concentration of trade. Before proceeding
to analyze statistically the basic tables, certain
features are immediately apparent. One of
these is the abnormality of the distribution of
the figures. Consider, for example, Appendix
Table ia and examine the column for, say,
Sweden. Fourteen other countries are listed as
exporters to Sweden, so that if Sweden's imports were distributed evenly among them,
each would account for 7.15%o of her imports.
(The discussion will proceed in terms of the
equal importer or equal exporter basis unless
otherwise specified.) If the distribution were
not perfectly even but varied more or less according to some normal distribution, one would
expect an accumulation of figures round the
7 per cent level, with few figures at either very
high or very low percentages. In fact, as can
be seen from the table, the reverse is the case.
The figures seem to accumulate at the extremes
-
at high percentages (i6.7%
i6.7 % from Norway, I2.0 %
or at low percentages (I.07%
0.07% from Ireland,
2.4
from Iceland,
from Germany)
from Greece,
% f rom' Turkey).
311
THE REVIEW OF ECONOMICS AND STATISTICS
32
Sweden is not exceptional in this respect.
The result is that, taking all the countries together, there is a relatively high number of
very high or very low percentages, that is to
say, high or low compared with the average of
7.I5 per cent.2 This can be seen in the frequency distribution in Table i. In short, after
correcting for differences in the sizes of all
countries' total imports and exports, imports
and exports do not tend to be distributed in a
statistically normal fashion among the other
countries.
TABLE I. - FREQUENCY DISTRIBUTION OF ADJUSTED
PERCENTAGE DISTRIBUTION
OF INTRA-EUROPEAN
TRADE, I938
Class
Imports Exports
0-i%
44
I-2%
2 -3%0
3-4%
4- 570
5-6%o
6-7%
I5
20
23
25
12
I0
Class
45
23
25
7- 8%
8- 9%
9-I0%
I5
I9
Io-II%o
II-I2%
I3
I4
I2-I3%
I3+%
Imports Exports
I0
2
7
II
II
9
6
3
5
8
0
37
35
5
SOURCE: Appendix Tables ia and ib.
These figures suggest three conclusions: (I)
there is a striking degree of concentration of
each country's trade on a small range of other
countries, (2) the degree of concentration has
declined somewhat between I938 and I953,
and (3) both these conclusions apply equally
to imports and exports. Table 2 is only a rough
guide, however, to the degree of concentration.
Table 3 gives the Gini coefficients for imports
and exports in I938 and I953, showing how far
from a Lorenz curve distribution the trade percentages are. (Zero would measure equal distribution; unity would measure complete concentration.)
Table 3 confirms the conclusions, already
reached on the basis of Table 2, concerning the
approximate equality of concentration of imports and exports and the decline in concentration between I938 and I953.3
TABLE 3. - GINI COEFFICIENT OF CONCENTRATION FOR
TRADE OF O.E.E.C.
MEMBER COUNTRIES
Imports
Exports
The concentration of each country's trade
may be represented by considering, for each
country, what percentage of imports is supplied by the two or three most important suppliers of its imports. The results of such a
calculation vary, of course, from country to
country. The averages for all Member countries are shown in Table 2.
TABLE 2. - CONCENTRATION OF TRADE ON PRINCIPAL
SUPPLIERS AND PRINCIPAL CUSTOMERS *
(In per cents)
1938
'953
43.5
39.4
38-3
39.8
54.9
55.5
49.6
49.8
2 Major Countries
Imports
Exports
3 Major Countries
Imports
Exports
* Average percentage, for all O.E.E.C. members, of total intraEuropean imports/exports covered by major suppliers/customers.
2The arithmetic mean is 7.I5% for imports and 6.68%
for exports in 1938. This is because in I938 there are no
Austrian trade figures, so that as the tables are based on
export data for all countries Austria appears only as a
destination of exports (as an importer) and not as an exporter.
1938
1953
.507
.522
.442
*443
Concentration and distance. We have seen
the strong tendency of countries to concentrate
their trade on a few other countries rather than
spread it evenly. An even distribution would,
in fact, be rather surprising, for various reasons. The object of this study, however, is
simply to investigate the relationship between
the manner in which the trade is distributed
and the relative distances between countries.
The concept of "economic distance" relates
to the cost of transversing distance rather than
the actual mileage involved. Unfortunately,
countries cannot be regardedas occupying only
points in space between which the distances to
be converted into economic distances are easily
measured. Countries occupy large areas, and
the concept of distance between areas depends
' Examination of the basic data (see Appendix)
shows that the reduction in concentration is partly
due to the changed relationship in the trade between Germany on the one hand and Greece and Turkey on the other.
But the decline in trade concentration between I938 and
I953
appears to be also partly due to a more general,
though less marked, decline among the other countries of
Europe. Only Norway and Portugal seem to be exceptions
to the general trend.
DISTANCE AND TRADE PATTERNS
on such considerations as whether we conceive
the distances between the areas to be the distances between the closest points, or between
the geometrical centers, or between their
"centers of gravity" as determined by some
system of weighting.
For purposes of measuring the cost of transporting a commodity from one country to another, the appropriateconcept would appear to
be a mixture of these alternative measures.
Products are likely to be shipped (or entrained)
between points as proximate as possible. But
the availability of harbor or railway facilities
determineswhat is possible. On the other hand,
somebody has to bear the cost of transporting
the goods from the source of production to the
export point, and from the import point to the
final destination. This means that though in
country I, port A may be nearer to country 2
than port B, the goods being exported to country 2 may be produced very near to port B and
a long way from port A. This qualification to
distance brings us closer to the "center of gravity" concept - of course it is the economic
center of gravity, rather than the physical center of gravity, that is relevant.4
Finally, even if there were some unique distance between countries in terms of some unit
of distance such as miles, there is the problem
of allowing for variation in the cost of alternative means of transport. The main problem
here is the differencebetween sea and rail transport. While France is adjacent to Belgium the
cost (per unit weight) of rail transport from
the source of production in France to the final
location of consumption in Belgium may be
greater than the cost (per unit weight) of transport of some product manufactured in London
and consumed in Oslo and which is therefore
transported almost entirely by sea.
Before proceedingto a more detailed anbalysis
I This modification has important consequences. For
example, measuring between "nearest points" France is
clearly as close to Italy as it is to Belgium, since she is
adjacent to both countries. Measuredin terms of the geometrical center, she is again about as close to Italy as to
Belgium (though not if we consider also the geometrical
centers of Italy and Belgium as well -Italy's would be
much further away). But measuredin terms of economic
center of gravity she is probably somewhat closer to Belgium than to Italy on account of the concentration of
French industrialactivity in the Paris region and the coal
mining areas of northernFrance.
33
of the relative distance between the countries
represented in the basic tables, an assessment
has been made, for each country, of which two
or three countries are "nearest" to it in the
sense described above. (The actual ordering of
countries on which this selection has been based
is shown in Table 5 below.) To do this an estimate has been made of the percentage of trade
of each country going to the two or three nearest
countries. The results for the average of
O.E.E.C. Member countries are shown in Table
4a.
TABLE 4a.
-
OF TRADE ON NEAREST
CONCENTRATION
COUNTRIES'
(In per cents)
2 Nearest Countries
Imports
Exports
3 Nearest Countries
Imports
Exports
1938
1953
27.9b
26.5
28.4
26.2
34.9 b
34.5
36.7
33.0
Average percentage, for all O.E.E.C. members, of total intraEuropean imports/exports of each country coming from/going to the
two or three nearest countries.
b See footnote 2 above.
a
As it stands, however, Table 4a cannot accurately show the difference in the degree to
which trade is concentrated on neighbors as between two and three countries, for even if there
were no correlation at all between relative distance and concentrationthe percentage of trade
covered by any three countries would be
greater than that covered by any two countries. This, and other incomparabilities,means
that the data have to be standardizedby dividing all the figures by the percentages of trade
that would be covered anyway by the number
TABLE 4b.
2
CONCENTRATIONOF TRADE ON NEAREST
DATA *
COUNTRIES,STANDARDIZED
(In per cents)
I938
1953
I.95
I.99
1.97
Nearest Countries
Imports
Exports
I.99
3 Nearest Countries
Imports
I.63
1.72
Exports
I.83
I.64
Average percentage, for all O.E.E.C. members, of total intraEuropean imports/exports of each country coming from/going to the
two or three nearest countries. Data standardized by procedure outlined in text.
34
THE REVIEW OF ECONOMICS AND STATISTICS
of countries specified if there were a perfectly
even distribution. The result of this operation
is given in Table 4b.
The above table suggests the following conclusions: ( i ) even allowing for statistical qualification there appears to be a strong tendency to concentrate trade on "near" countries;
(2)
this tendency is not so strong, however, if
the three nearest countries are considered
rather than the two nearest countries; (3) it is
just as strong for exports as for imports; and
(4) it does not appear to have declined between
I938 and I953.
Thus far we have not discussed the problem
of how to measure, in practice, the relative distances between countries in the sense outlined
above. One possible measure would be to examine the mark-up between (i) the f.o.b. average value of exports of certain specific goods as
they appear in the export statistics of countries
accompaniedby specification of the destination
of these exports and (2) the c.i.f. average value
of those quantities of the same goods, which
are specified as being imported from the corresponding countries, in the- import statistics of
the relevant importing countries. To take a
hypothetical example, suppose that in French
export statistics, exports of alcoholic beverages are shown by quantity and value and at
the same time by destination. One could
readily compute the average value (price)
f.o.b. of French exports of alcoholic beverages
to each main destination separately. Turning
to the import statistics of each of these countries in which similar data are available, one
could readily compute the average value c.i.f.
of their imports of alcoholic beverages from
France. The mark-up for each country will
vary, and the differencesin this mark-up should
indicate roughly the relative costs (including
insurance, etc.) of sending alcoholic beverages
to each of the countries concerned.5
In actual practice, this procedure is subject
to several limitations, owing to various defects
of international trade statistics. For example,
the time period concerned may not be identical
for both countries, since products may be recorded in French export statistics long before
' The source actually used in this study is O.E.E.C.
Statistical Bulletins, Foreign Trade, Series IV.
they are recorded in the importing statistics
of the recipient countries. The exact classification of products is also not identical in all
countries. There may also be substantial
changes in the weight of some products (for
example, tobacco) that rest in bond for a long
time between being officially exported and officially imported. There may be discrepancies
arising from transit trade-country A may
show, in its export statistics, that a certain
quantity of some product has been exported to
country B, but it is possible that a part of
this amount has not been officially imported
into country B at all, and that it was merely a
transit transaction. If the average value of this
part of the commodity in trade differs substantially from that of the rest then the procedure
outlined above will give an incorrect result.
In view of such difficultiesthe above method
of estimating distance has been used to obtain
only a rough ordinal comparison of distances
rather than a cardinal comparison. That is to
say the limitations on the data enable us to say
only that country A is further from country Z
than is country B, but we cannot say by how
much. Even the mere ordering of countries
can only be carried out with a rough accuracy
by this method. This is largely because, in many
cases it has not been possible to find enough
products giving sufficientpairs of statistics (on
the import side and on the export side) for a
reliable averaging of the mark-ups. A large
number of observations are needed because
otherwise there are inconsistencies in the results. For example, according to the data for
one product Germany is nearer to Greece than
is Denmark, but according to the data for another product the reverse result may be obtained.
For these and other reasons the sort of data
referred to above can only be used to give a
rough ordering of the relative distances between countries. In some cases obvious errors
can be legitimately corrected by simple knowledge of geography. For example, the following
results of percentage mark-ups is obtained
for Sweden on the basis of her trade in paper
and paperboard.6
8S.I.T.C. No.
641.
DISTANCE AND TRADE PATTERNS
Country
It will be obvious from inspection that in the
absence of fully reliable statistical indicators
there is a considerable element of uncertainty
PercentageMark-up
Denmark
Belgium
Norway
Netherlands
Germany
- I1
+ I%
? 5%
+ 6%
+ 6%o
UIKend
9%
+IO%
France
Italy
Portugal
Turkey
Greece
35
about the ordering of countries which are close
together. For example, in the column relating
to Belgium, the exact ordering of the Netherlands, France, and Germany is liable to be
erroneous. It is on account of the absence
of sufficiently reliable statistical data that not
all countries could be included in Table 5.
+I2%
+I4%
+23%
+28%
TABLE 5. RANKING OF COUNTRIESIN ORDEROF ESTIMATEDECONOMICDISTANCE
FROMSELECTEDCOUNTRIES'
Belgium
Denmark
France
Germany
Italy
Netherlands
France
Germany
Sweden
Germany
Norway
Belgium
Germany
Netherlands
Austria
Switzerland
France
Switzerland
Austria
France
U.K.
Norway
Denmark
Netherlands
U.K.
Belgium
U.K.
Portugal
Norway
Netherlands
Belgium
Italy
Germany
Greece
Turkey
Sweden
Portugal
Italy
France
Portugal
Italy
Sweden
Denmark
Italy
Denmark
Sweden
U.K.
Portugal
U.K.
Netherlands
Greece
Turkey
Greece
Turkey
Greece
Turkey
Norway
Portugal
Greece
Belgium
Norway
Sweden
Turkey
Denmark
Greece
Netherlands
Norway
Sweden
Turkey
U.K.
Turkey
Italy
Portugal
Belgium
Germany
France
Sweden
Denmark
U.K.
Denmark
Norway
Germany
Greece
Italy
Portugal
Ireland
Belgium
Netherlands
Germany
France
U.K.
U.K.
Norway
Sweden
Germany
Netherlands
Belgium
Netherlands
Belgium
U.K.
Germany
Switzerland
Austria
France
Norway
Sweden
Belgium
Netherlands
Norway
Denmark
Portugal
Italy
France
Portugal
Italy
France
Italy
Portugal
France
U.K.
Belgium
Denmark
Germany
Portugal
Sweden
Denmark
Greece
Turkey
Greece
Turkey
Greece
Turkey
Netherlands
Norway
Sweden
Italy
Greece
Turkey
Denmark
a Some countries have been given a longer list than others. This depends on how many countries it has been possible to order for each
selected country on the basis of the technique described in the text. It will be observed, for example, that Switzerland only appears in the
columns relating to Italy and Germany, and Austria in the columns of these two countries, plus Turkey.
If one is attempting only to order the countries,
one can probably improve the above ranking
by, say, transposing Greece and Turkey
(Turkey should be "further away" from Sweden than is Greece), or Norway and Belgium,
and so on. On the basis of evidence for other
products and including any adjustments that
seem to be reasonably certain, the countries
were ordered as shown in Table 5.
Such inaccuracies in the ordering are just as
likely to reduce the correlation coefficients
which have been calculated below as to increase
them.
Being limited to ranked data, the obvious
7 Switzerland does not publish any trade data of the
type required here and only appears when fairly good
judgments can be made (namely, in the colunmnsfor Germany and Italy).
36
THE REVIEW OF ECONOMICS AND STATISTICS
test of a relationship between distances and
importance in trade is the Spearman rank correlation coefficient. Applied here, this coefficient measures, roughly speaking, the strength
of the relationship between the ranking of
countries with respect to their relative distance
from a given country and the ranking of the
same countries with respect to their relative
importance in the trade of the given country.
The results obtained, for several given countries, are given in Table 6.
TABLE 6. -RANK
between distance and trade, and (2) this correlation appears to be about equal for imports
and exports.
In order to evaluate the significance of these
coefficients, however, an important point must
be now introduced. Since one country's imports
are another country's exports the actual distribution of (say) the first country's imports
will depend on a mixture of two distance elements: (i) the relative distance of every other
country to the given country, which will influ-
CORRELATION COEFFICIENTS, DISTANCE AND TRADE
Distance and imports
1938
1948
Belgium
.72
Denmark
.76
.82
.52
France
Germany b
.74
.X8
1953
.85
.8i
1938
1948
1953
.8i
.87
.62
.73
.52
.75
.68
.80
.95
(.36)
.65
.77
(.87)
.74
.I8
.43
(.68)
.78
.78
.26
J3
.20
(.57)
.65
(.50)
.79
Netherlands
.87
.83
.84
.91I77
Norway
Sweden
Turkey
U.K.
.90
.87
*43
.65
.85
.80
.37
J15
.95
.67
.6i
.I8
.63
.83
.59
.33
.83
.54
.45
.67
.55
.63
.72
.58
.67
.6o
Arithmetic mean
Including all trade
Excluding German trade with
Greece and Turkey
Excluding U.K. but including
German trade with Greece and
Turkey
Distance
& change
in imports
Distance and exports
49
.o6
(.45)
.83
Italy
'
.50
1938-53
.75
(-
.67
.90
.70
.JO
.78
.85
.47
*34
.84
.78
*44
.o8
.72
.62
.59
.54
.67
.66
.63
.62
.68
.77
.68
.62
.55
a Significance levels for the correlation coefficients are as follows:
b
Level of
.o I %
significance
.02%
.I %
.0%5
.66i
I938 imports
.6I2
.458
.532
All other figures .44I
.SI4
.592
.641
Figures in parentheses exclude trade with Greece and Turkey.
It should be noted that the right-handcolumn
of Table 6 constitutes a check on the possibility
of having made errors in the ranking of distance which fortuitously increase the correlation coefficients. For the same distance ranking is, of course, used for the calculations relating to the pattern of trade in the given years
and for the changes in the pattern over time.
A distance ranking which was fortuitous from
the point of view of the former would not
necessarily be so from the point of view of the
latter.
The last three rows in this table seem to confirm two of the conclusions reached above,
namely, (i) there is a fairly strong correlation
ence the import pattern of the given country
in one way; and (2) the relative distance of
the given country to each other country, which
will affect the export pattern of each other
country and will thereby also have an effect on
the import pattern of the given country.
Relative distances in the economic sense are
not symmetrical- the United Kingdom may
be the nearest country to Ireland, but Ireland is
not necessarily the nearest country to the
United Kingdom. Thus although the United
Kingdom may be highly placed in Irish imports on account of her proximity to Ireland,
another tendency may be at work sending
United Kingdom exports to countries such as
DISTANCE AND TRADE PATTERNS
France, which are "nearer" the United Kingdom than Ireland is.8
To sum up: since (i) relative distances will
affect both the distribution of exports and imports, (2) each country's exports are some
other country's imports, and vice versa, and
(3) relative distances are not necessarily symmetrical, the final pattern of trade will be a
mixture of the various pulls, the export pull
against the import pull. This means that it
would be most unlikely to obtain for all countries high rank correlation coefficients for both
imports and exports. For if, for example, import patterns of all countries were found to
have correlation coefficients of nearly unity
when compared with ranked relative distances,
this would suggest that the influence of distance on export patterns would be small unless
there were a high degree of symmetry in relative distances. That a considerable degree of
symmetry must exist can be seen from the fact
that in Table 6 both the import and the export
coefficientsare, on the whole, fairly high.
This does not mean, of course, that other
determinants such as changes in relative incomes and prices have not had any influence
on trade patterns. For the correction factor
used on the data (to an equal importer or
equal exporter basis) has eliminated such determinants as regards their influence on the
pattern of each country's trade in the given
years and hence as regards their influence on
changes in the pattern. To compare the influence of distance with that of the other determinants would be a much more difficult problem, for while the technique used above can
show the influence of distance other things
being equal,9 it is not possible to calculate the
influence of other things (such as prices and incomes), distances being equal, until a techs In a more formal manner one can envisage a trade
matrix transformed into a matrix in which coefficients allocate each country's imports on the basis of the relative
distance of each other country. Similarly the transformation could be made to one in which coefficients apply to
each country's exports. The relative strength of the resulting two coefficients applying to each box in the matrix will
determine the final value of the relevant element of the
matrix.
'Roughly speaking, the correction to equal exporter
basis wipes out differences in competitiveness, among other
things; and the correction to equal importer basis wipes
out differences in relative incomes, among other things.
37
nique is found for adjusting data to an "equal
distance" basis. For this, mere ranking is of
course insufficient.
Conclusions. Apart from the specific conclusions already mentioned above, the basic
tables show a tendency on the part of the lessdeveloped countries of Europe to concentrate
their trade more than the other countries. The
concentration is also very closely linked to
neighboringcountries. Now it is not surprising
that this should be the case, since the less-developed countries will, in general, have less
diversified economies; this will in practice
(though with no absolute theoretical necessity), lead to relative concentration countrywise of total trade. What is interesting is the
possibility of a causal connection between degree of development and distance. This requires the notion of "absolute distance," not
merely of relative distance, to and from other
countries- of being, in an absolute sense, far
away or near to other countries as a whole.
The fact that the countries on the periphery of
Europe tend to have less-developed economies
(Iceland, Portugal, southern Italy, Greece, and
Turkey) may be due partly to their being
simply "far away" from everybody that matters
in trade.10 The Scandinavian group of countries may constitute the exception that proves
the rule. For the countries nearest to them are
Germany and the United Kingdom, two very
important traders and two countries with which
they are connected by sea-passage-the
cheapest form of transport. The fact that the
great distance of Turkey and Greece from the
major trading countries of Europe is offset by
their proximity to Russia and Eastern Europe
has been poor compensation,for these countries
have been far less important traders than the
countries of northwest Europe.
This view can, of course, be developed along
many lines. The main implication in the field
of the economics of underdevelopedareas, for
example, is to confirm the fact that means of
transport usually constitute the most useful
form of overhead capital. What is important,
in other words, is to be "near" other countries
10It has been pointed out above that relative distances
are not necessarily symmetrical. Thus it is possible for a
country not to be relatively near any other country. Such
a country is "far away" in an absolute sense.
38
THE REVIEW OF ECONOMICS AND STATISTICS
preferably important countries. The rapid
development of the Canadian economy is no
doubt largely due to proximity to the United
States. It is probable that the favorable prospects for Latin America are also related to the
same cause.
As well as speculating on the future relative
rates of growth in terms of nearness, it would
be interesting to speculate on the effects of
changes in modes of transport and transport
costs. Apart from the general reduction in 'economic distances which would result from a reduction in the costs of air freight, for example,
a special problem is posed by the existence of
"psychic" distance. It is probable that the
manner in which the purchases of raw materials
by a firm are distributed geographically will
depend partly on the extent to which foreign
-
sources have been personally contacted and
cultivated. While the transport costs paid (directly or indirectly) by an Italian entrepreneur
on a raw material supplied by Turkey may be
no greater (as the material may come by sea)
than the same material supplied by Switzerland, he is more likely to have contacts with
Swiss suppliers, since Switzerland will be
"nearer"to him in a psychic evaluation (fewer
language difficulties, and so on), as well as in
the economic sense that air travel will absorb
less of his time. The growth of air travel and
freighting would not only tend to iron out discrepancies arising from sea versus rail transport (as the aircraft travel, in general, along
straight lines - or great circles) but would
have very interesting repercussions on psychic
distances.
0
00 99"o
0
t
en0
H
0
0
0
0
~C
~,
HO
m en
oo
"d0
-d
O
-1
o0 0CC0
0
~
%O
-4C
0 1-4
r4
40
4-.
0
6
-
6
P4~~~0
o
C7
0
\
~0 14%C0
6
Co6
-)
\-.
-
C)
o
.-o
0%0()
C\o
1-i4,-
.C%
o
t
.Ifw1-9c%Co*i'
o1-4 0-.
-o
-in
,A~
Z~~~~~~~~~~~~~~~~~~~c
-
+~~~0
0
0
inC
(
~
e4
%
1-4
00
o
)l%~
o
o
o
o
C
~o-~i-xin.c0ir
1d-C
,o0%
o o
Cr)041-404r)
0
00
C..
c
C1-..
6
CoC--0%14C-..(4
0
z-
H
Ho
0
\
in0C-.00%0
-
0
00
(4
\
in
i 0%OCoC
\ in
0-
c~~~~~~~~~~~ ~
-
04
00t
H
(4
0
1-4
(41o40
R)1
CO t1-.
i
\0t k
J)0%C
a
v
oenIt\
4e
%Inr
~~ ~ ~~~~~~~~~~~~~~~~
0
0
.
o0
in4%
..
..
-tI-
1-4i
1-4
o00
C)-Co(
ocio-'"o
O'-.i
-4
H~
4z
00'-
C..
0
la
i,
'04
0
400'0
-
-
.
U)
i,t \
0
0
08
1-41-4
%
00 nOt
-.-
u;
H~~(
~~
o
C7'
0
C
Co
'0'
--
0coi0
~
0%
0o.o~
C*
0
*IC~O)Cl
A4
1-4
0
1
10
6
4
6
0
~~~~~~0
06
\-.
~~~~~~~~~mi
0 ~~~0C-.Vo6o
0%0 OCo
-~~~~o
o
H~'
0
1
1-40
6
00
80
-
0\
C-\0
8
0
n t,d
1-)f4
14
0,0
-
Coin
80
80
t-
0
00Cco
4
-
't
00
40
0
C~00t
e4
0%%) 00.
~~
0
'A-.--
C7t-.C0
oco
~~~~i-e4en
'~~~~t-oo
.-C
o
I o0o
it
RtC
0
CO
04
e
0
"inCq
0
0
-i
~~et~~oo
0
0
~(0
0
0
U~4~
..
*
~o~
0C'0%oV)II)4%0'0'0-,
~~V)c4~~'0Cr)1-4 (4~~~~~O
-
4-~~~~~~ ~
)Cr%
9
q
~~.
OCo
)
1-4Co-
4
; o
4
c
0
(4-0
0
\
(4 (e4
oo
(')C--Cf)Co
~')
0%'
0m0
C)~~~~C
0r
0~~~~moo
c0)-.o'
0
rCI(
0jCf
t
0%4 lot
\"0
cI1I4(4Co
Cot- 1o)in-..
i
o C7%
Z
00
0
(41-4
4OCoO4(4Coo(4c1-40
.4 0
0Ci
0
~ ~ ~~~~~~~~~~~o
4440
-r
0oo
-
-
0"
"4
'0
4-4 4-e
~44
-
co-.14
0
(4
~
C)
0
0C
I1-
in1-40
6-
N0~~0N000~o
~~~'I1-4
00
to
)0
o0coo
oXR
0(4
o
o
-41414
).*
(4
4.4
It
0
c
0
~ ~ ~~
~ ~ ~ ~ ~ ~~~~~~~~~~~~~-
~
o4
Z
0
9
-4
ON
-enr-4
q
\6
o
100~
0
~
0
C)
o
o
o
~1-;
-,-
%-'.
(4V
40
;6o
"
r\0
0%(4C--CoCo
6
-0
Nr
-\
~~
64c
U
CdC
H
H
o
ON
"
en
N
ON 0 %O
C)
4~C.0in
in
--d0
0~~~~~~~~~
4.C
oo
-j
e
N
0I
-
--,0
O
O
~~'0
0~~0~g00414'041)
'004)
N~ in )
.
.
.
0~~o~~o-~-b
140'
00
~~t-
'IO
.
140
000i
~~~~?
14~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
('4
0
r)0
co
c~~6
u; o; c6
o6
c4
'0nOOO'It1414)14
4%
141
414
to
1400'tO14OO'00-.41)
C'4
14oH
)'100
'014
-
1)
0-1
1-4
I
It
8
')4 \
0-4
'14
14-
En
0440C'41
0'
H
0
in
1
oC
in
f)4co
in
4)
m
tt
0
\0
0 inoo
0
0
0
-It00
001
0
41)i
C'ht
0-4~~~~~>
~~~
'014~~~~~~~~~~~)0-.
00
0-
0
C'
4-14101
41)0
`-
1
C.
n
040-
0
0l
4
01
0 0141
014o1
in0
\0
0
00144
4
0
14t
1414it
,ir
H
0'1
0
(0
0
t"4141414)
14
Ull)
inC71,
'041)14
to
1)00--0\
;u
\01)
614u
S0i.
z
A1)
U
\6
o
U)
'4)4t-'
\
\t1'It
14)"o14004
It1441)014')14)41)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
;6
tC
101)400't1
-0
t
410000014
0
,
.
4o.
N
in)4in
0
'It
l)ce)0
)1)1410
\0
1
i
0 ci
go
H
t
en
4
64
i
4
!6
U001) mcis
r
44100
It
"
in"
0doin )
I1441
co14t
H
in ) "t
R
0
in0\r)C'o'100
'0
00 e 1)Ze)'t0
1~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
%O \0.41ON.In It0%
w
\8
T
0
I
0%0
o'o'0
o-
o
%
'Itm
14)I0'1414
0000-0140014141)C'
14 'o0014't
\4 tr
z.
0
'0re)
\q00-.
4)1
C'
'14-
.
0
14)4~~~~~\
014
01414in
'0
0-~'f
0
C'0
e)14)
144)
00
140
14
't)'4
\
11)
000
o14I)141%
144)C
o
o
4\
s
t00')C1401441)4
1400
000-.1400It'014
o'o0004914.
0(O4
(I)
\')
-
0-in"in
.0
1
4)0.I0-140
~~~I-H
14400000~
14O0'It001)O~~~~~~~~~~~~~~~~~~~0t0')
\0
0
m
\0
~ ~ -41)'
o
e4
41)--
;%6i;r
4)
4,l
e4)
t
0
C,,
0-'IO
14~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
1
C\414\
;411
4.
14
C
0--0
''It
Cs
4)1
in'4
0-14
I1-4
'It1)'t1
't-
14)0
1
1
I-.4
0
C'14)
in4)1
It
4
e;
00
14
1
4)
Ite' 0
C'00
14
in1
6
\14
0'0
.
bO
-Cin"0
l
0)0~
01)0in00,,0%-41)144\0
r041)C0
C0
140\
;)4I0--1001't44'C1
1;416
4
140 -0 in
1
(-00140
-
, in
41)C"0-
(-41)Itm \041)'0
14'T-
0
4)r-%
14al'0
q
;6-6c4
0
14
'O
OC
'?
m
H \
c
Ho
14-en
C
r
\
0
l
~~4
4~4
C'00
0414400114'\014044
%O.
C'40C100-.)
41'0
410
14
4)
44
-?
140-~o
0
$
14)
14
'0~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~'
14~~~~~~~~~~\'Iti
0
00
N0')41'0
%
1\4r
0
14 140-..
1e)'It'It
1~,6i4
i;)1
't-
0014
')itO
1~4
14')14'0
re)'041)41)1
\Cin0
4
H0
--C''ot-
00-n'
1;
14
6
4
I00
14
in
0--
4;
0,
1
0~~~~~~~~~
0
00
16~~~~~~~~~~~~~~~~~t
16
0
0"
0
~~'
0~~~~'10o~~
H
0
IV~~~~~
.
C
Nc