Trade Impacts of the Terrorist Attacks of 11 September 2001: A Quantitative Assessment
Peter Walkenhorst and Nora Dihel *
May 2002
Paper prepared for the Workshop on
"The Economic Consequences of Global Terrorism"
DIW, Berlin, 14-15 June 2002
*) The authors are economists in the Trade Directorate, Organisation for Economic Co-operation and
Development, 2 rue André-Pascal, 75775 Paris Cedex 16, France. The views expressed in the paper
are not necessary those of the OECD or its Member countries.
TRADE IMPACTS OF THE TERRORIST ATTACKS OF 11 SEPTEMBER 2001:
A QUANTITATIVE ASSESSMENT
Abstract
This paper used estimates of changes in frictional trading costs and the Global Trade Analysis
Project (GTAP) computable general equilibrium model to assess the effects of the terrorist attacks of
11 September on international trade. Following the 11 September events, regulatory measures were
implemented to tighten security at air and seaports as well as land border crossings. As a result, trading
and transport companies have been confronted with additional frictional trading costs, in particular costs
relating to transport, handling, insurance, and customs. These additional costs make international trade
more expensive and will tend to reduce trade flows. The results from scenario analysis show that regions
with high trade to GDP ratios, such as South Asia or North Africa and the Middle East, suffer the largest
trade and welfare losses in relative terms. Across sectors, trade in agri-food products, textiles, non-metallic
minerals, and machinery is most severely affected by the higher frictional costs of trade.
Keywords
Terrorism, security concerns, non-tariff measures, frictional trading costs, CGE modeling.
JEL classification
National security and war (H560); Country and industry studies of trade (F140).
1. Background
The nature and scale of the terrorist attacks of 11 September 2001 on New York City and
Washington, D.C., meant that the political and economic impacts would not be locally or nationally
confined, but would affect governments, businesses, and people world-wide. Consumer and investor
confidence plummeted after the tragic events, and economic researchers downgraded their forecasts of
short and medium-term economic growth for virtually all countries around the globe. Moreover,
emergency measures were taken to tighten security at air and seaports as well as land border crossings.
Some disruption of trade flows during the immediate aftermath of the attacks seemed almost inevitable, yet
additional frictional trading costs due to tighter security, notably costs relating to transport, handling,
insurance, and customs, have the potential to continue to affect trade in the medium to long term.
Some have likened the higher frictional trading costs to additional taxes on business activity or
increases in border tariffs. Yet, a comparison with business spending on mandatory pollution abatement
equipment seems more appropriate, as the higher expenses for the private sector provide benefits to the
general public (higher environmental quality and lower risk of terrorist attacks, respectively), but are
generally not accompanied by additional tax or tariff revenues for governments. Some industry experts
have estimated that the additional security-related costs amount to up to three per cent of the value of
traded goods (Leonard, 2001), while other analysis suggests that the magnitude of the effects is more
modest (OECD, 2002a, 2002b). Nevertheless, in an era of large-scale economic interdependance and
vertical specialisation, even small changes in trading costs can have considerable economic effects.
Empirically, the impacts of the higher transport and trading costs on trade flows are to a large
extent masked through other economic developments. For example, following the 11 September events
world oil prices fell, thereby reducing the fuel costs of transport operators and at least partly offsetting the
security-related cost increases. Also, existing capacities of transport operators did not make it possible for
them in the short run to fully pass on the increases in frictional costs to consumers, so that transport fees
after the attacks did not necessarily reflect the changes in security-related costs.
This paper uses estimates of changes in frictional trading costs and the Global Trade Analysis
Project (GTAP) computable general equilibrium model to separate out and assess the effects of the terrorist
attacks of 11 September on international trade. Combining data on frictional cost changes with GTAP's
information on transaction costs and bilateral trade flows across a large number of sectors and countries
makes it possible to analyse and quantify the trade effects of the 11 September events. Moreover, the
general equilibrium framework enables an evaluation of the incidence of changes in the trading
environment and an assessment of economic welfare effects.
The remainder of the paper is organised in four parts. Section 2 reports on available empirical
evidence of the terrorist attacks of 11 September on the business environment for transport and trading
companies. Section 3 presents the quantitative approach to model the impact of frictional cost changes on
international trade flows and economic welfare. The results of the modelling analysis are then discussed in
section 4, and the final section 5 contains conclusions.
2. Impact of the terrorist attacks on frictional costs of trade
Because of the fear of future terrorist attacks, the 11 September events have triggered a variety of
regulatory measures and demands on transport companies that entail new costs and longer delays at border
crossings. The prevalence and thoroughness of inspections at airports, seaports, and land borders has been
increased, shipping companies have had to face expenditures for additional security equipment and
personnel, and transport insurance rates have been augmented. These developments have been most
pronounced in North America, but are similarly felt in other parts of the world.
2.1
Air transport
Tightening security on airports and aircraft has been one of the foremost priorities of public
authorities. Immediately after the attacks, all US airports were closed for three days to review and
strengthen security procedures. Around the world, access to airports and aircraft was tightened, training
programmes for ground and flight staff were initiated, and more sophisticated alert and airspace
management systems were launched. Concerning passenger traffic, more thorough controls of passengers
and their luggage have been implemented, the classification of objects in hand luggage that could
potentially be used as weapons has been modified, and in-flight protection has been stepped up by blocking
cockpit access and putting armed air marshals on flights.
Air freight transport has been similarly affected. Since the 11 September events, x-ray machines
to examine cargo boxes and check whether the contents match shipping labels have been used more
extensively. Other precautions taken include earlier drop-off deadlines at airports, bans on shipments from
unknown customers, and waiting periods before shipments are put on planes.
Following the 11 September attacks, insurers and re-insurers modified their aviation plans
considerably whenever a "war and allied perils clause" in existing contracts authorised such changes.
Coverage was reduced while premiums went up. Premia for air and sea cargo shipments were raised by
2
between 0.027 per cent and 0.05 per cent of total insured fleet value.1 Additional premia of $1.25 to $1.85
per passenger have been charged in passenger transport. Also, insurers decided to limit the cover for third
party damages caused by terrorist action to $50 million (instead of $1.5 billion). In order to avoid major
disruptions of air traffic due to lack of insurance coverage, regulators in several countries enforced shortterm measures to support third party insurance for airline companies in case of a new terrorist attack with
damages above $50 million.
In response to increased security provisions and insurance costs, many airlines began to apply
"security surcharges". Generally, fee increases amounted to up to $8 per passenger or fell within the range
of $0.10-$0.15 per kg of cargo (OECD, 2002a). In addition to increases in security fees, cargo shipments
have been affected by frequent hold-ups and detours. During the three-day shutdown of the US air
transport system, many transportation providers had to shift from air transport to road or rail transport to
maintain shipment commitments. Later on, many airlines in North America, Europe, and Asia dropped
flights from schedules, because of the weakening of passenger confidence and demand for air transport.
On a year on year basis, world-wide passenger traffic dropped by 23 per cent in October and freight traffic
by 9 per cent.2 This decrease was far more pronounced than the reduction in air traffic during the Gulf
War, for example. Yet, traffic volumes have been recovering towards normal levels by the end of 2001.
The marked decline in passenger traffic in autumn 2001 caused delays in the movement of goods
that are carried in the belly space of passenger planes. Also, several airlines closed some of their freight
handling facilities as a result of decreasing cargo volumes. For example, American Airlines temporarily
closed its freight handling operations at the international airports of Buffalo-Niagara, Greater Rochester,
and Salt Lake City. As a result, cargo had to be routed through other transport hubs.
2.2
Maritime transport
A large share of the world's cargo moves through seaports. Yet, in the past port facilities often
lacked security equipment, such as x-ray devices, and had relatively lax controls of access to docks and
ships. Only about 2 per cent of the 72 million containers that are moved internationally every year used to
be inspected.3 After 11 September, more extensive security checks and use of surveillance cameras and
cargo scanners have been a priority. The Port Authority of New York and New Jersey closed its operations
for two days, and more intensive screening of cargo loads in US and Canadian harbours has caused
substantial delays before cargo could be picked up. These hold-ups disturbed shipping timetables worldwide during the first weeks following the terrorist attacks. As ships to and from North America have
experienced delays, subsequent routes have been affected by delays as well.
Other precautionary measures in Canada and the USA have consisted of mandatory 96-hour
advance arrival notices and more frequent onboard Coast Guard inspections of crews and cargo. Also,
ships have had to travel at slow speeds inside US harbours, flanked on each side by a tugboat in order to
protect, for example, bridge supports against abrupt changes in direction. Shipping companies have been
charged $1 000 to $1 500 for the required tugboat escorts. 4
Over time, emergency measures taken immediately after the 11 September events have been
relaxed, transport operators have adjusted to the new procedures, and transport and trading activities
seemed to be running relatively smoothly again by the end of 2001. In the longer run, technological and
procedural progress might make it possible to further reduce the post-11 September delays and procedural
costs while providing the desired high level of security.
Similar to developments in the air transport industry, war-risk insurance premia for maritime
transport rose sharply after the terrorist attacks on New York City and Washington, D.C. In response, ship
3
owners and vessel operators have implemented war risk surcharges on ocean freight cargo transiting the
Middle East, Red Sea and Eastern Mediterranean. Beginning in early October, ocean carriers operating in
the Europe/Far East and certain Middle East trade lanes announced war risk surcharges for traffic to and
from specific ports as well as cargo transiting the Suez Canal, regardless of the origin and destination ports.
War risk surcharges range from $10 to $450 per Twenty Foot Equivalent Unit of full container loads, and
from $5 to $12 per cubic metre of less than full container loads.5 The application of the surcharges meant
that total insurance rates for shipments from India to Persian Gulf ports, for example, went up by about
50 per cent.6
2.3
Road and rail transport
The reinforcement of US customs vigilance during the processing of commercial vehicles led to
lengthy delays immediately after the terrorist attacks. The US-Mexican border was closed for a short
period, but hold-ups at the US-Canadian border were at least equally substantial. In Detroit, Port Huron,
and Buffalo waiting times amounted to 10 to 12 hours in mid-September. Later, delays were reduced as
additional customs personnel took up duty and logistics were improved by posting wait times on the US
customs internet-site. But by the end of September, truck operators still reported four-hour backups at the
US-Canadian border while loads were searched by customs agents. Prior to 11 September, delays at the
border had typically lasted only about 30 minutes (OECD, 2002a).
Road transportation companies themselves have been taking a variety of new measures against
terrorist attacks. Some are fencing unsecured freight yards and terminals, conducting background checks
on drivers, and issuing identity badges to employees, while others have installed satellite-tracking systems
to monitor the exact location of trucks and trailers or equipped their vehicles with sensors that can detect
whether a cargo container has been opened before reaching its destination. Similarly, train operators have
taken precautionary measures such as more frequent inspections of tracks, bridges and tunnels,
strengthening critical buildings and communication facilities, and installing fibre optic cables along tracks
to detect tampering.
2.4
Customs procedures
Prior to the terrorist attacks, estimates of the cost of time delays, paperwork, and compliance
related to border crossing ranged from 5 per cent to 13 per cent of the value of the goods involved,
depending on the types of goods traded (OECD, 2002a). The increased security at borders after
11 September 2001 has further augmented these customs-related costs. However, in the longer term the
attention and resources newly devoted to customs inspections might trigger efforts to develop more
efficient security procedures and improve border management.
Measures to facilitate trade could include the reduction of the number of in-transit cargo
inspections, the electronic collection of customs duties, and improved information sharing between
authorities. As a result, customs services could become more able to reconcile policing of borders with
smooth and open trade flows. The benefits of customs harmonisation have recently been illustrated
through research by Japan's Ministry of Economy, Trade, and Industry and the Mitsubishi Research
Institute that shows that the introduction of automated customs would lower the direct costs of customs
clearance by the equivalent of 0.2 per cent of the value of traded goods. If furthermore the indirect benefits
of a reduction in customs related delays are taken into account, additional cost reductions of up to 1 per
cent of merchandise value could be realised (Hertel, Walmsley and Ikatura, 2001).
Some concrete results of trade facilitation efforts following the 11 September events have already
materialised. For example, on 12 December 2001 Canada and the USA signed a "smart border
4
declaration" that outlines an 30-point action plan to collaborate in identifying and addressing security risks
while expending the flow of people and goods back and forth across the Canada-USA border. The action
plan aims to enhance security and strengthen cross-border commerce through improved technology, coordination, and information sharing.7 A similar initiative for the US-Mexico border was unveiled in
March 2002.
3. Implications for international trade
Most operators will try to pass along any new frictional trading costs to their clients, although
excess capacities and resulting price competition might make it difficult to raise prices in the short term.
For example, during the first weeks following the 11 September attacks, many airlines actually lowered
their fares for passenger transport as demand collapsed. One major reason for the demand reduction was
the substantial decline in travel for tourism purposes. People were concerned about potential further
attacks by terrorists and postponed or cancelled their travel plans. Destinations world-wide were affected.
Excess capacity has also been an issue in the air and maritime cargo industry, so that transport operators
have not been able in the short run to pass on higher security-related costs.
At the same time, developments for fuel prices have been favourable during autumn 2001,
lowering overall operating costs for transport companies. Transport activities are fuel-intensive, and
expenditure on propulsion fuels accounts for a large share of operating costs. Thus, increases in fuel costs
have a substantial impact on transport operators. In past incidents of war and terror in the Middle East, oil
prices tended to increase markedly, due to concern over a possible disruption of supplies. For example,
after Iraq's invasion of Kuwait in 1990, oil prices shot up from less than $20 to more than $35, triggering a
slowdown in global economic growth as well as a substantial increase in transport costs.
Oil price developments after the 11 September events have in comparison been rather atypical.
During the first month after the events, oil prices were on average about $4 per barrel lower than during the
preceding month (Figure 1). Throughout the subsequent military action in Afghanistan, oil prices stayed
below $20 per barrel. The corresponding decline in fuel costs facilitated the adjustments to new security
arrangements for transport operators. Indeed, several North American and European airlines, including
Northwest Airlines, British Airways, and Lufthansa, eliminated previously existing fuel surcharges that
had been introduced earlier in response to then rising expenditures on fuel.
(Figure 1)
Industry experts have estimated that the total costs of extra security measures implemented after
the 11 September events could amount to 1 to 3 per cent of the value of traded goods (Leonard, 2001).
However, these estimates were made soon after the 11 September events and seem to reflect the major
disruptions in the weeks following the terrorist attacks. Over time the costs of the new border security
measures have been declining and might continue to do so as technical and procedural progress is realised.
On the other hand, new terrorist attacks could give rise to renewed security concerns with implications for
the level of frictional costs to international trade.
Some researchers have empirically estimated the impact of changes in transport costs on trade.
For example, a recent study of bilateral trade between 103 countries reported an elasticity of trade flows
with respect to transport costs of about -3, implying that a transport cost increase of 1 per cent would lead
to a reduction in trade volume of 3 per cent (Limao and Venables, 2001). Also, concerning delays at
borders, researchers estimated that the daily cost of hold-ups equals on average a 0.5 per cent ad-valorem
tariff, with lower values for raw commodities and higher values for intermediate and consumer goods
(Hummels, 2001).
5
In the USA, transport and insurance costs amounted on average to 3.4 per cent of customs value
in 2000 (OECD, 2002a). Yet, the cost shares ranged from about 1 per cent for pharmaceuticals to more
than 23 per cent for crude fertilisers. Increases in transport and insurance costs will naturally have a
stronger impact on those commodities for which frictional costs are more important. Hence, trade in
commodities with a high ratio of weight and volume relative to value, like fertilisers, coal, and certain
fruits and vegetables, will tend to be more impeded by higher frictional costs than imports and exports of,
for example, transport equipment, pharmaceuticals, and natural gas.
A second cross-commodity consideration concerns the prevailing transport mode. Air transport
has been relatively most affected by the tightening of security after the 11 September events, so that
commodities that are to a large extent shipped by aircraft will likely be subject to higher relative increases
in transport costs than commodities that go normally by ship, train, or truck. Goods that are typically
transported by air tend to be light, high-value products that need to be delivered quickly, as they are
typically subject to frequent lifecycle or fashion changes. Examples include electronics equipment and
apparel. However, as indicated in the preceding paragraph, frictional costs generally represent only a small
percentage of the value of these products.
Besides higher transport and insurance costs, longer delays at air and seaports, as well as land
border crossings have represented another type of international trade impact. Such security-related delays
may make producers of perishable products, such as vegetables or fish, reluctant to ship overseas. The risk
of spoilage at airports may induce producers to limit their distribution to distances that can be served by
truck or train, for example. Moreover, about half of all trade between the USA, Canada, and Mexico is in
machinery and transportation equipment, in particular automotive products. Over the past years many
automotive companies have adopted just-in-time inventory practices that depend on frequent and
predictable shipments from suppliers, including those located in other countries. In this context,
unforeseen delays at border crossings can have immediate, disruptive effects on production runs, leading in
several cases to factory shutdowns after the 11 September events.8
The tightening of security measures and the impact on trade flows has been most pronounced in
the countries and regions directly affected by the terrorist attacks and the subsequent "war on terror", i.e.
North America, the Middle East, and Central Asia. Hence, trade with countries in these “sensitive” regions
has been subject to relatively steep increases in frictional costs following 11 September. Naturally, all US
trade was affected by the tightening of security at US borders, but as a large share of Canadian and
Mexican exports go to the USA, these NAFTA-partners were almost as strongly exposed to the new
measures as the USA itself.
Yet, changes in frictional trade costs were not limited to North America, but occurred worldwide. Indeed, North America is a region with a very low trade to GDP ratio, implying that changes in
trade costs tend to affect NAFTA countries to a lesser extent than other regions (Figure 2). In particular,
South Asia as well as North Africa and the Middle East, but also Western and Eastern Europe, show a
trade to GDP ratio that is more than twice as high as the North American one.
(Figure 2)
In order to determine the overall impact of higher frictional costs on different countries and
regions, the importance of trade for particular countries, the mix of traded commodity and their mode of
transport, as well as the exposure to "sensitive" locations have to be taken into account. Given the
complexity of international trade relations, applied economic models that make it possible to derive
estimates of the quantitative impacts of changes in commodity-specific trade costs are necessary tools for
this type of analysis. Quantification also enables comparisons of the relative magnitude of the economic
6
effects of terrorism risk occuring through trade vis-à-vis those coming about through other economic
activities, such as tourism, finance, and insurance.
4. Modelling the effects of increased security concerns on trade
Earlier research on the economic effects of terrorism has covered a number of issues, including
game-theoretic aspects of terrorist action (Lapan and Sandler, 1993), the impact of terrorism on investment
and economic growth (Enders and Sandler, 1996; Poirson, 1998; Abadie and Gardeazabal, 2001), and the
assessment of terrorism-related losses in the tourisms sector (Enders, Sandler, and Parise, 1992; Pizam and
Smith, 2000). Concerning the trade impacts of tighter security measures, limited research has been carried
out to date. Anderson and Marcouiller (1999) investigate the linkages between economic insecurity and
trade and find that inadequate institutions to constrain corruption and contract negligence reduce trade
flows to a larger extent than tariffs. Moreover, Bougheas, Demetriades, and Morgenroth (1999) report a
positive relationship between the level of trade cost-reducing infrastructure and the volume of trade, and
Hertel, Warmsley and Ikatura (2001) predict considerable gains from reducing frictional costs of trade
through improved customs administration.
The following analysis aims to complement existing research by deriving estimates of the post11 September impact of increased frictional costs on trade flows and economic welfare across geographical
regions and economic sectors. Half a year after the terrorist attacks, it remained unclear to what extent the
existing increases in frictional costs would persist, and how businesses and public authorities would adjust
to the new security environment in the longer term. Evidence to assess these issues will only become
available over time, so that a certain degree of uncertainty concerning the frictional cost effects of
11 September remains. Hence, any quantitative estimate should be seen as being primarily indicative.
4.1 Model and data
The analysis is carried out by using the well-established GTAP database and model. The latter is
a static, multi-region, computable general equilibrium model that operates under assumptions of perfect
competition and constant returns to scale. The model reflects bilateral trade flows, international transport
margins, and country and sector-specific rates of import protection. GTAP thereby makes it possible to
determine changes in production, consumption, trade, and economic welfare from particular trade-related
external shocks, such as changes in trade margins. A full description of the model can be found in Hertel
(1997).
The following investigation uses information from the GTAP-5 database to specify ten regions
and ten sectors. Nine of the ten regions refer to semi-continental or continental groupings, and the tenth to
an aggregate of small countries, for which no detailed information on economic structure is available in the
GTAP database. Of the ten sectors, nine represent different types of goods, while services are aggregated
into the tenth one. The correspondence of the regions and sectors modelled and their GTAP-5 components
is given in Annex Tables 1 & 2.
The GTAP trade data set contains information on transport margins, which are derived from
actual transport costs data at the detailed commodity level (SITC-4/5) in the United States. Based on this
US data, bilateral margins for other countries are estimated (Gehlhar and McDougall, 2002). Similarly, US
data on the commodity-specific modes of transport (ground, sea, air) is extrapolated to other countries.
As discussed earlier, the increase in frictional costs following the 11 September events results in
higher trading costs, which are, unlike tariffs, not accompanied by additional revenue for the government.
Instead, the benefits of the additional precautionary expenditures take the form of increased public
security. Yet, the costs and benefits of providing this public security service are not as such covered in the
7
GTAP database. Therefore, the analysis of non-tariff shocks and their impact on trade flows requires some
adjustments.
The approach taken in this study is analogous to the analysis of non-tariff liberalisation carried
out by Hertel et al. (2001). It rests on the notion of defining an “effective price” PMS*irs of commodity i,
imported from country r at domestic prices in destination market s. This price is related to the observed
price PMSirs as follows:
(1)
PMS*irs = PMSirs / AMSirs ;
where the technical coefficient AMS is unobserved and equal to one in the initial equilibrium. Changes in
the value of AMS capture the impact of changes in unobserved trading costs on the price of imports from a
particular exporter. Thus a decrease in AMSirs triggers an increase in the effective domestic price of good i
exported from r to s.
In order to maintain balanced trade flows in the data set, a compensating quantity adjustment is
required, so that the “effective quantity” of exports associated with this price is defined such as to ensure
that the product of observed price and quantity equals the product of effective price and quantity:
(2)
QXS*irs = QXSirs • AMSirs ;
Taking the total derivative of expressions (1) and (2), re-arranging terms, and expressing them in
percentage terms (denoted by lower case letters) gives the equations for import demand (equation (3)) and
composite import price (equation (4)):
(3)
qxsirs = - amsirs + qimis - σim • [pmsirs - amsirs - pimis] ;
(4)
pimis = Σkθiks • [pmsiks - amsiks] ;
where:
qxsirs: percentage change in bilateral imports of i of s from r;
qimis : percentage change in total imports of i into s;
σim : elasticity of substitution among imports of i;
pmsirs: percentage change in price of imports of i from r in s;
amsirs: percentage change in effective price of i from r in s due to change in unobserved trade costs;
pimis: percentage change in average import price of i in s;
θiks: share of imports of i from k in total imports of s;
With the existing trade elasticities in GTAP, it can be expected that an increase in trade costs will
result in a decrease of both observed expenditures on imports and the share of imports from the particular
trading partner to which the increase in trade costs applies.
8
4.2 Simulation scenarios
Several experiments are carried out to analyse the impact of post-11 September increases in
frictional costs on trade flows and economic welfare. There is ample anecdotal evidence that frictional
costs have risen since the terrorist attacks on New York and Washington. Yet, systematic statistical data
on the magnitude and structure of the changes is not available to date. Hence, the simulated changes in
trade cost have to be based on assumptions that generalise from the available empirical evidence.
Four scenarios were evaluated that use differing data on the structure of frictional cost changes.
The overall magnitude of the frictional cost change was taken to be one per cent of the value of transported
goods. Given the average magnitude of transport and insurance costs of 3.4 per cent ad valorem and
estimates of the average costs of time delays, paperwork, and compliance related to border crossing of
9 per cent, an increase in frictional trading costs by one per cent ad valorem appears to be an upper bound
for the long-term impacts of the increased security concerns following the terrorist attacks on the World
Trade Center and the Pentagon. Further terrorist action triggering additional tightening of border controls
could lead to renewed increases of trade costs, and impact predictions made immediately after the
11 September events (Leonhard, 2001) put the increases of frictional costs as high as three per cent of the
value of traded goods, but relatively flexible adjustments of businesses and customs authorities to the new
security environment during the six months following the attacks suggest that the cost increases in the
medium and long-term are much lower than during the days immediately after the tragic events.
The lower bound of trade impacts from the 11 September events might be seen as no change in
frictional costs and trade flows, implying that in the long run the increased attention devoted to
optimisation of customs services and technical progress will offset any burden from security measures put
into place after the attacks. This no-change situation is taken as the baseline against which four other
scenarios are evaluated. The latter are outlined in the following:
•
Uniform increase in frictional costs: In this scenario, all frictional costs are increased by one per
cent ad valorem.
•
Region-specific increase in frictional costs: In this scenario, frictional costs are on average
increased by one per cent ad valorem. However, the increases are not uniform across regions.
High risk regions (North America, North Africa & Middle East) and medium risk regions (Western
Europe & South Asia) are assumed to experience increases in frictional costs that are two and oneand-a-half times as high, respectively, as the cost increases in low risk regions (Eastern Europe,
Latin America, Sub-saharan Africa, Oceania, North Asia, Rest of World). Using expenditure on
frictional costs as weights, the frictional cost increases amount to 0.7 per cent in low-risk regions,
1.05 per cent in medium-risk regions, and 1.4 per cent in high-risk regions.
•
Sector-specific increase in frictional costs: In this scenario, frictional costs are on average
increased by one per cent ad valorem. However, the increases are not uniform across sectors. In
particular, it is assumed that the change in frictional costs for commodities that are transported by
air and sea are two and one-and-a-half times as high, respectively, as the cost increases in ground
transport. In particular, when using expenditure on frictional costs as weights, the mode-specific
frictional cost increases amount overall to 0.69 per cent for ground transport, 1.03 per cent for sea
transport, and 1.38 per cent for air transport. With the information on the sector-specific
composition of transport modes in the GTAP database, sector-specific estimates of frictional cost
changes can be derived.
•
Region and sector-specific increase in frictional costs: In this scenario, frictional costs are on
average increased by one per cent ad valorem, and a regional and sectoral differentiation as in the
9
"region-specific" and "sector-specific" scenarios is applied. The derived changes in frictional costs
range from 0.52 per cent (products in a low-risk region that are largely transported by road or rail)
to 1.78 per cent (products in a high-risk region that are largely transported by air).
The changes in frictional costs for the selected sectors and regions were introduced into GTAP,
and the model was solved using the Gregg 2-4-6 step solution procedure under standard closure.
4.3 Results
The results from the simulations suggest considerable trade and welfare impacts of a one per cent
ad valorem increase in frictional costs to trade. World welfare, measured in terms of equivalent variation,
would decline by about $75 billion per year. The biggest welfare losses in absolute terms would occur in
Western Europe, North America, and North Asia (Table 1).
(Table 1)
Yet, if the welfare losses are related to the size of the economies in the different regions, the
ranking of most affected regions changes considerably (Figure 3). In relative terms, South Asia as well as
North Africa and the Middle East are far more affected by higher frictional costs than, for example, North
America or North Asia. These results are largely driven by the economies’ strong dependence on
international trade (Figure 2).
(Figure 3)
Across sectors, the most marked impacts occur for agriculture and food products, textiles and
leather, non-metallic minerals, and machinery. Trade in these products is projected to decrease by one per
cent or more as a result of a one per cent ad valorem increase in frictional costs (Figure 4). In other words,
import demand for these commodities is elastic, while for mining products, chemicals, and basic metals
trade flows are expected to change only to a less than proportional extent.
(Figure 4)
Differences in the results across scenarios are apparent, but do not fundamentally change the
assessments. Only in a few cases are the scenario outcomes significantly different, as, for example, for the
region of North Africa and the Middle East that seems less affected than Western and Eastern Europe
under the assumption of uniform increases of frictional costs, but suffers more marked welfare losses than
the two European regions if it is taken to be a high-risk region that experiences higher than average
increases in frictional costs. Hence, exposure to international trade and the elasticities of import demand
appear to be more important parameters for determining the impact of higher frictional costs than regional
or sectoral differences in the magnitude of the frictional cost increases.
In general, the findings highlight the close economic relationships in the international economy.
Small changes in trade costs can have considerable impacts on trade flows and economic welfare.
Increases in frictional costs for imports into particular countries affect not only the importing and exporting
countries, but also other countries through the diversion of trade flows and substitution between products.
Hence, even countries that are not directly involved in a conflict or subject to terrorist attacks might
nevertheless suffer losses in trade and welfare as a result of increased security concerns and higher
frictional costs of trade.
10
5. Conclusions
This paper used estimates of changes in frictional trading costs and the GTAP model to assess the
effects of the terrorist attacks of 11 September on international trade. Following the 11 September events,
regulatory measures were implemented to tighten security at air and seaports as well as land border
crossings. As a result, trading and transport companies have been confronted with additional frictional
trading costs, in particular costs relating to transport, handling, insurance, and customs. These additional
costs make international trade more expensive and will tend to reduce imports and exports.
Four scenarios are evaluated that quantify the trade and welfare impacts of a one per cent ad
valorem increase in frictional costs of trade applying either uniformly across sectors and countries, or
being sectorally or regionally differentiated according to presumed exposure to terrorism risk following the
11 September events. Given that after the initial disturbances following the terrorist attacks on New York
and Washington international trade relations have returned towards normal again, a one per cent ad
valorem increase of trade costs likely presents an upper bound for the impacts of 11 September. The
global welfare losses of such an increase in frictional costs are estimated to amount to about $75 billion.
The analysis shows that even though North America would face substantial losses from the
increase in frictional costs, other regions that depend to a larger extent on international trade, such as South
Asia or North Africa and the Middle East, are more severely affected in relative terms. Many of the
countries in these regions are middle-income developing countries that have progressively opened their
economies towards international markets. The above-average losses in trade and welfare from increased
security measures following the 11 September events could raise popular concerns about further steps of
trade liberalisation and thereby compromise the long-term development prospects of these countries.
Agriculture and food products, as well as textiles and leather are among the product categories
for which above average impacts on trade flows are predicted. Trade in these products is often politically
sensitive and many countries retain substantial protection against imports of these products. In the
Uruguay Round of multilateral trade negotiations some progress was made in reducing trade barriers for
agri-food products and textiles, which would be partly offset through increases in frictional costs.
11
6. References
Abadie, A., and J. Gardeazabal, 2001. "The economic costs of conflict: a case-control study for the
Basque country." NBER Working Paper No. 8478, National Bureau of Economic Research,
Cambridge/Mass.
Anderson, J.E., and D. Marcouiller, 1999. "Trade, insecurity and home bias: an empirical investigation."
NBER Working Paper No. 7000, National Bureau of Economic Research, Cambridge/Mass.
Bourheas, S., P.O. Demetriades, and E.L.W. Morgenroth, 1999. "Infrastructure, transport costs, and
trade." Journal of International Economics 47: 169-189.
Enders, W., T. Sandler, and G.F. Parise, 1992. "An econometric analysis of the impact of terrorism on
tourism." Kyklos 45: 531-554.
Enders, W., and T. Sandler, 1996. "Terrorism and foreign direct investment in Spain and Greece."
Kyklos 49: 331-352.
Gehlhar, M., and R.A. McDougall, 2002. "Transport margins and modes." in Dimaranan, B.V. and R.A.
McDougall (editors): Global Trade, Assistance, and Production: The GTAP 5 Data Base. Center
for Global Trade Analysis, Purdue University, West Lafayette/Indiana.
Hertel, T. (editor), 1997. Global Trade Analysis: Modeling and Applications. New York and Melbourne:
Cambridge University Press.
Hertel, T., T. Walmsley, and K. Ikatura, 2001. "Dynamic effects of the 'new age' free trade agreement
between Japan and Singapore." Journal of Economic Integration 24: 1019-1049.
Hummels, D., 2001. "Time as a trade barrier." Mimeo, Department of Economics, Purdue University,
West Lafayette/Indiana.
Lapan, H.E., and T. Sandler, 1993. "Terrorism and signalling." European Journal of Political Economy 9:
383-397.
Leonard, J., 2001. "Impact of the September 11, 2001 terrorist attacks on North American trade flows."
Manufacturers Alliance e-Alert, Arlington/Virginia.
Limao, N., and A. Venables, 2001. "Infrastructure, geographical disadvantage, transport costs and trade."
World Bank Economic Review 15: 451-479.
OECD (Organisation for Economic Development and Co-operation), 2002a. "The impact of the terrorist
attacks of 11 September 2001 on international trading and transport activities." Unclassified
document TD/TC/WP(2002)9/final. Paris: OECD Publications.
OECD (Organisation for Economic Development and Co-operation), 2002b (forthcoming). "The
economic consequences of terrorism," in: OECD Economic Outlook #71. Paris: OECD Publications.
Pizam, A., and G. Smith, 2000. "Tourism and terrorism: a quantitative analysis of major terrorist acts and
their impact on tourism destinations." Tourism Economics 6: 123-138.
Poirson, H., 1998. "Economic security, private investment, and growth in developing countries." IMF
Working Paper WP/98/04, International Monetary Fund, Washington, D.C.
12
Notes
1.
See "Survey of members: national measures in respect of war risk insurance," International Air Transport
Association, 27 November 2001.
2
See "October traffic shows intensified decline." Press release no. 36, International Air Transport
Association, 30 November 2001.
3
See "When trade and security clash." The Economist, 6 April 2002.
4.
See "Freight-transportation system gets more expensive, slower." Wall Street Journal of 27 September
2001.
5.
See "War risk surcharge summary," Fritz Transportation International, December 2001.
6.
See "Ocean rates rising in Indian region." Journal of Commerce, 25 September 2001.
7.
See "Canada and the United States sign smart border declaration." Government of Canada News Release
No. 162, 12 December 2001.
8
See "The Canada-US border: an automotive case study." Center for Automation Research, Ann Arbor,
Michigan, January 2002.
13
Figure 1: World oil prices (weekly data), Jan. 2000 - Dec. 2001 (US Dollar per barrel)
30
25
20
15
11-Sep-01
10
01-Jan-00
01-Apr-00
01-Jul-00
01-Oct-00
01-Jan-01
01-Apr-01
01-Jul-01
01-Oct-01
01-Jan-02
Source: OECD (2002a).
Figure 2: Ratio of imports to gross domestic product in selected regions
50%
40%
30%
20%
10%
0%
Western
Europe
Eastern
Europe
N. Africa & Sub-saharan
Middle East
Africa
Oceania
Source: GTAP-5 database.
14
North Asia South Asia
North
America
Latin
America
Rest of the
World
Figure 3: Scenario results on welfare changes due to a 1% ad valorem increase in trade costs
Unifo rm
Regional diversity
Sectoral diversity
Regional & sectoral diversity
0.0%
-0.1%
-0.2%
-0.3%
-0.4%
-0.5%
-0.6%
-0.7%
W estern
Europe
E astern
Europe
N. Africa &
M iddle
East
Subsaharan
Africa
O ceania
North Asia South Asia
No rth
America
Latin
America
Rest of the
W orld
Note: Welfare changes are calculated as equivalent variation divided by gross domestic product.
Source: Authors.
Figure 4: Scenario results on changes in imports due to a 1% ad valorem increase in trade costs
U nifo rm
R egio nal d iversity
S ecto ral d iversity
R egio nal & secto ral diversity
0.00
-0.20
-0.40
-0.60
-0.80
-1.00
-1.20
-1.40
-1.60
A gricu ltu re
M ining
Fo o d
T extiles &
leather
Wood &
p aper
Source: Authors.
15
C hemicals
N o nm etallic
minerals
B asic
metals
M achinery
O ther
Mining
Food
Textiles &
leather
Wood &
paper
Chemicals
Nonmetallic
minerals
Basic
metals
Machinery
Other
Uniform increase
Western Europe
Eastern Europe
N. Africa & Middle East
Sub-saharan Africa
Oceania
North Asia
South Asia
North America
Latin America
Rest of the World
Total
Regional differentiation
Western Europe
Eastern Europe
N. Africa & Middle East
Sub-saharan Africa
Oceania
North Asia
South Asia
North America
Latin America
Rest of the World
Total
Sectoral differentiation
Western Europe
Eastern Europe
N. Africa & Middle East
Sub-saharan Africa
Oceania
North Asia
South Asia
North America
Latin America
Rest of the World
Total
Reg. & sect. differentiation
Western Europe
Eastern Europe
N. Africa & Middle East
Sub-saharan Africa
Oceania
North Asia
South Asia
North America
Latin America
Rest of the World
Total
Source: Authors.
Agriculture
Table 1: Change in imports and welfare as a result of a 1% ad valorem increase in trade costs
Imports (per cent)
-0.9
-1.0
-1.0
-1.5
-1.4
-0.6
-1.9
-1.0
-1.3
-1.4
-1.1
0.8
-0.8
-1.3
-0.6
-0.2
0.4
0.2
-0.3
-0.7
0.1
-0.6
-1.1
-0.9
-0.6
-1.1
-1.2
-0.7
-1.7
-1.0
-1.0
-1.0
-1.1
-1.0
-1.3
-1.0
-1.2
-1.1
-1.6
-2.0
-1.1
-1.8
-1.3
-1.4
-0.7
-0.7
-0.6
-0.7
-0.7
-0.6
-1.2
-0.9
-0.7
-0.6
-0.8
-0.2
-0.2
-0.2
-0.2
-0.1
-0.5
-0.3
-0.6
-0.3
-0.3
-0.3
-1.2
-1.0
-1.2
-1.0
-1.0
-1.2
-1.5
-1.1
-0.9
-0.8
-1.2
-0.3
-0.6
0.0
-0.4
-0.7
-0.9
-0.9
-0.9
-0.9
-0.5
-0.5
-1.2
-1.3
-0.8
-0.5
-0.8
-1.8
-0.6
-1.3
-0.8
-0.9
-1.3
-1.1
-1.1
-0.8
-0.7
-0.8
-0.7
-1.5
-0.6
-0.5
-0.7
-0.9
-31282
-3518
-2942
-1027
-972
-11845
-6380
-13434
-3105
-639
-75143
-0.9
-0.7
-1.5
-1.1
-1.1
-0.4
-2.0
-1.4
-1.0
-1.0
-1.1
0.8
-0.6
-1.8
-0.5
-0.1
0.2
0.2
-0.4
-0.5
0.1
-0.7
-1.1
-0.6
-1.0
-0.7
-0.8
-0.5
-1.8
-1.5
-0.7
-0.7
-1.1
-1.1
-1.0
-1.5
-0.9
-0.8
-1.1
-2.0
-1.5
-1.3
-1.0
-1.3
-0.7
-0.5
-0.9
-0.5
-0.5
-0.4
-1.3
-1.3
-0.5
-0.4
-0.9
-0.2
-0.2
-0.4
-0.2
-0.2
-0.4
-0.3
-0.8
-0.3
-0.3
-0.3
-1.2
-0.7
-1.8
-0.6
-0.6
-0.8
-1.6
-1.6
-0.6
-0.5
-1.2
-0.3
-0.4
0.0
-0.3
-0.4
-0.7
-0.9
-1.2
-0.6
-0.3
-0.5
-1.3
-0.9
-1.3
-0.3
-0.5
-1.3
-0.6
-1.8
-0.5
-0.6
-1.3
-1.2
-0.7
-1.3
-0.4
-0.4
-0.4
-1.5
-0.9
-0.2
-0.4
-1.0
-32798
-2413
-4564
-664
-684
-8470
-6672
-18602
-2166
-456
-77489
-0.7
-0.7
-1.0
-1.6
-1.5
-0.7
-2.0
-0.9
-1.2
-1.3
-1.0
0.7
-0.7
-1.3
-0.7
-0.2
0.4
0.3
-0.2
-0.7
0.0
-0.6
-0.9
-0.7
-0.6
-1.2
-1.2
-0.7
-1.7
-0.9
-0.9
-0.9
-1.0
-0.9
-1.0
-1.0
-1.3
-1.2
-1.6
-2.0
-1.0
-1.7
-1.3
-1.3
-0.6
-0.5
-0.6
-0.8
-0.8
-0.6
-1.3
-0.7
-0.7
-0.6
-0.7
-0.2
-0.1
-0.3
-0.3
-0.2
-0.5
-0.4
-0.6
-0.3
-0.3
-0.3
-1.0
-0.8
-1.3
-1.1
-1.1
-1.3
-1.6
-1.0
-1.0
-0.8
-1.1
-0.2
-0.4
0.0
-0.5
-0.7
-1.0
-1.1
-0.7
-0.8
-0.4
-0.5
-1.1
-1.1
-0.9
-0.6
-0.9
-2.0
-0.8
-1.3
-0.8
-1.0
-1.3
-1.1
-0.9
-0.9
-0.9
-0.9
-0.9
-1.8
-0.6
-0.6
-0.7
-1.0
-29494
-2993
-3144
-1096
-1063
-12772
-7008
-13451
-3214
-650
-74886
-0.8
-0.5
-1.6
-1.2
-1.1
-0.5
-2.1
-1.2
-0.9
-1.0
-1.1
0.8
-0.5
-1.8
-0.6
-0.1
0.2
0.3
-0.3
-0.5
0.1
-0.7
-0.9
-0.5
-1.0
-0.8
-0.9
-0.5
-1.8
-1.2
-0.7
-0.7
-1.0
-1.0
-0.7
-1.4
-0.9
-0.8
-1.2
-2.1
-1.4
-1.3
-1.0
-1.2
-0.7
-0.4
-1.0
-0.6
-0.5
-0.4
-1.4
-1.0
-0.5
-0.4
-0.7
-0.2
-0.1
-0.4
-0.3
-0.2
-0.4
-0.4
-0.7
-0.3
-0.3
-0.3
-1.1
-0.5
-1.9
-0.7
-0.7
-0.9
-1.7
-1.4
-0.7
-0.5
-1.1
-0.2
-0.3
0.1
-0.3
-0.5
-0.7
-1.2
-1.0
-0.6
-0.3
-0.5
-1.2
-0.7
-1.5
-0.4
-0.6
-1.5
-0.8
-1.7
-0.6
-0.7
-1.3
-1.1
-0.5
-1.5
-0.6
-0.6
-0.6
-1.9
-1.0
-0.3
-0.5
-1.0
-31160
-2010
-4835
-736
-763
-9390
-7394
-18401
-2302
-470
-77461
16
Welfare
(mill. USD)
Annex Table 1: Sectors defined by reference to the GTAP-5 Sectoral Classification
Sectors
Agriculture
Mining
Food
Textiles and leather
Wood and paper
Chemicals
Non-metallic minerals
Basic metals
Machinery
Other manufacturing and services
GTAP-number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
GTAP-sector
Paddy rice
Wheat
Cereal grains not elsewhere covered
Vegetables, fruit, nuts
Oil seeds
Sugar cane, sugar beet
Plant-based fibers
Crops not elsewhere covered
Bovine cattle, sheep and goats, horses
Animal products not elsewhere covered
Raw milk
Wool, silk-worm cocoons
Forestry
Fishing
Coal
Oil
Gas
Minerals not elsewhere covered
Bovine meat products
Meat products not elsewhere covered
Vegetable oils and fats
Dairy products
Processed rice
Sugar
Food products not elsewhere covered
Beverages and tobacco products
Textiles
Wearing apparel
Leather products
Wood products
Paper products, publishing
Petroleum, coal products
Chemical, rubber, plastic products
Mineral products not elsewhere covered
Ferrous metals
Metals not elsewhere covered
Metal products
Motor vehicles and parts
Transport equipment not elsewhere covered
Electronic equipment
Machinery and equipment not elsewhere covered
Manufactures not elsewhere covered
Electricity
Gas manufacture, distribution
Water
Construction
Trade
Transport not elsewhere covered
Water transport
Air transport
Communication
Financial services not elsewhere covered
Insurance
Business services not elsewhere covered
Recreational and other services
Public Administration, Defense, Education, Health
Dwellings
Source: Authors.
17
Annex table 2: Regions defined by reference to the GTAP-5 regions
Model region
GTAP-Code
AUT
Western Europe
BEL
DNK
FIN
FRA
DEU
GBR
GRC
IRL
ITA
LUX
NLD
PRT
ESP
SWE
CHE
XEF
HUN
Eastern Europe
POL
XCE
XSU
TUR
N. Africa & Middle East
XME
MAR
XNF
BWA
Sub-Saharan Africa
XSC
MWI
MOZ
TZA
ZMB
ZWE
XSF
UGA
XSS
GTAP-region
Austria
Belgium
Denmark
Finland
France
Germany
United Kingdom
Greece
Ireland
Italy
Luxembourg
Netherlands
Portugal
Spain
Sweden
Switzerland
Rest of EFTA
Hungary
Poland
Rest of Central Europe
Former Soviet Union
Turkey
Rest of Middle East
Morocco
Rest of North Africa
Botswana
Rest of SA Customs Union
Malawi
Mozambique
Tanzania
Zambia
Zimbabwe
Other Southern Africa
Uganda
Rest of Sub Saharan Africa
Source: Authors.
18
Model region
GTAP-Code
AUS
Oceania
NZL
CHN
North Asia
HKG
JPN
KOR
TWN
IDN
South Asia
MYS
PHL
SGP
THA
VNM
BGD
IND
LKA
XSA
CAN
North America
USA
MEX
XCM
Latin America
COL
PER
VEN
XAP
ARG
BRA
CHL
URY
XSM
XRW
Rest of World
GTAP-region
Australia
New Zealand
China
Hong Kong
Japan
Korea
Taiwan
Indonesia
Malaysia
Philippines
Singapore
Thailand
Vietnam
Bangladesh
India
Sri Lanka
Rest of South Asia
Canada
United States
Mexico
Central America
Colombia
Peru
Venezuela
Rest of Andean Pact
Argentina
Brazil
Chile
Uruguay
Rest of South America
Rest of World